Vision 2030 of Kingdom of Saudi Arabia
will discussed some of value based health competetion and some other issue.
let’s delve into major aspects of the healthcare transformation as part of Vision2030. Based on our reading, please respond to the below questions:
1-In your words, what is the value-based healthcare? In your response, differentiate it from other similar models of healthcare, such as pay-for service model and etc. (4 paragraphs).
2-Summarize the benefits of value-based healthcare? (3-4 paragraphs).
3-How can the New Model of Care contribute into having a value-based healthcare? (3-4 paragraphs).
4-In page 94 in the Health Sector Transformation Program Delivery Plan one of the initiatives is to group healthcare organizations (Accountable Care Organization). Discuss how can the grouping of healthcare organizations in clusters (e.g. E1 cluster, C1 cluster) can improve quality and efficiency of healthcare services. (4-5 paragraphs).
Discuss how DRG is linked to ACOs? (2 paragraphs).
Academic Medicine, Vol. 95, No. 5 / May 2020682
Invited Commentary
Value in health care is the measured
improvement in a patient’s health
outcomes for the cost of achieving that
improvement.1 The goal of value-based
care transformation is to enable the
health care system to create more value
for patients. Because value is created only
when a person’s health outcomes improve,
descriptions of value-based health care that
focus on cost reduction are incomplete.
Reducing costs is important but not
sufficient: If the real goal of value-based
health care were cost reduction, pain killers
and compassion would be sufficient.
Value-based health care is often
conflated with quality, a vague concept
that implies myriad virtues and in
health care often focuses on inputs
and process compliance. Quality
improvement efforts may not improve
patients’ health outcomes, however;
even given similar processes, different
teams’ results vary. In addition,
requirements to track and report
process compliance may distract
caregivers from the more significant
goal of improving health outcomes.
Diabetes care in Italy provides one
example of process compliance not
ensuring better outcomes: Analysis
of regional variations in process
compliance and in outcome indicators
showed better process compliance in the
north but better outcomes for patients
in the south.2
Certainly clinicians should practice with
the consistency demanded by scientific
methods and follow evidence-based care
guidelines. But results matter. The goal
of value-based health care is better health
outcomes.
Value and patient satisfaction are also
commonly confused. While the patient
satisfaction movement has brought a
much-needed emphasis on treating
people with dignity and respect, the
essential purpose of health care is
improving health. Value is about helping
patients. Satisfaction surveys ask
patients, “How were we?” Value-based
care providers ask, “How are you?”
Why Improving Value Matters
Improving a patient’s health outcomes
relative to the cost of care is an aspiration
embraced by stakeholders across the
health care system, including patients,
providers, health plans, employers, and
government organizations. Value-based
health care aligns these diverse parties’
goals so well that, shortly after the
concept was introduced in 2006, health
economist Uwe Reinhardt described it
as “a utopian vision.”3 While Reinhardt
expressed concern about the challenges
of moving to a value-based system,
he lauded the larger objectives of the
transformation.
By focusing on the outcomes that matter
most to patients, value aligns care with
how patients experience their health.
In this context, health outcomes can be
described in terms of capability, comfort,
and calm.4 Capability is the ability of
patients to do the things that define
them as individuals and enable them to
be themselves. It is often tracked with
functional measures. Comfort is relief
from physical and emotional suffering.
In addition to reducing pain, improving
patients’ comfort requires addressing
the distress and anxiety that frequently
accompany or exacerbate illness. Calm is
the ability to live normally while getting
care. It encompasses freedom from the
chaos that patients often experience in
the health care delivery system, and it
Abstract
Value in health care is the measured
improvement in a person’s health
outcomes for the cost of achieving
that improvement. While some
descriptions conflate value-based
health care and cost reduction, quality
improvement, or patient satisfaction,
those efforts—while important—are
not the same as value, which focuses
primarily on improving patient health
outcomes. A decade of research into
organizations that have achieved better
outcomes while often lowering costs
suggests a strategic framework for
value-based health care implementation
that starts with identifying and
understanding a segment of patients
whose health and related circumstances
create a consistent set of needs. An
interdisciplinary team of caregivers
then comes together to design and
deliver comprehensive solutions
to address those needs. The team
measures the health outcomes and
costs of its care for each patient and
uses that information to drive ongoing
improvements. Care provided in this
way aligns delivery with how patients
experience their health and reconnects
clinicians to their purpose as healers. It
also asks physicians to think differently
about their role within the larger care
team and about the services that team
provides. The authors suggest medical
schools should incorporate education
on the principles and implementation
of value-based health care throughout
the undergraduate medical curriculum
to prepare their graduates to lead
the transformation to value-based
health care as they enter the physician
workforce.
Defining and Implementing Value-Based
Health Care: A Strategic Framework
Elizabeth Teisberg, PhD, Scott Wallace, JD, MBA, and Sarah O’Hara, MPH
Acad Med. 2020;95:682–685.
First published online December 10, 2019
doi: 10.1097/ACM.0000000000003122
Please see the end of this article for information
about the authors.
Correspondence should be addressed to Scott
Wallace, Value Institute for Health and Care, Dell
Medical School, 1601 Trinity St., Bldg. B, Stop Z1600,
Austin, TX 78712; telephone: (512) 495-5202;
email: Scott.Wallace@Austin.UTexas.edu; Twitter:
@DellMedValue.
Copyright © 2019 The Author(s). Published by
Wolters Kluwer Health, Inc. on behalf of the
Association of American Medical Colleges. This is an
open-access article distributed under the terms of
the Creative Commons Attribution-Non Commercial-
No Derivatives License 4.0 (CCBY-NC-ND), where
it is permissible to download and share the work
provided it is properly cited. The work cannot be
changed in any way or used commercially without
permission from the journal.
mailto:Scott.Wallace@Austin.UTexas.edu
http://creativecommons.org/licenses/by-nc-nd/4.0/
http://creativecommons.org/licenses/by-nc-nd/4.0/
Invited Commentary
Academic Medicine, Vol. 95, No. 5 / May 2020 683
is especially important for people with
chronic and long-term conditions. Care
that improves outcomes in all 3 of these
dimensions creates a better experience for
patients. Moreover, capability, comfort,
and calm describe outcomes that result
from the efficacy and empathy of health
care, rather than its hospitality.
Value-based health care connects
clinicians to their purpose as healers,
supports their professionalism, and
can be a powerful mechanism to
counter clinician burnout. Critics who
characterize value-based health care as
underpinning a model of “industrial
health care”5 distort the meaning of the
term value, misinterpreting it as focused
on cost. Instead, value-based health
care’s focus on better health outcomes
aligns clinicians with their patients. That
alignment is the essence of empathy.
Measured health outcomes demonstrate
clinicians’ ability to achieve results
with patients and families and drive
improvement in the results that matter
most to both patients and clinicians. This
intrinsic motivation is often missing in
the health care system, where clinicians
are directed to spend countless hours on
tasks that do not impact their patients’
health.
Better outcomes also reduce spending
and decrease the need for ongoing care.
By improving patients’ health outcomes,
value-based health care reduces the
compounding complexity and disease
progression that drive the need for more
care. A patient whose diabetes does not
progress to kidney failure, blindness, and
neuropathy is, over time, dramatically less
expensive to care for than a patient whose
condition continually worsens.6
Value-based health care is a path to
achieving the aspirational goals of the
Institute for Healthcare Improvement’s
“triple aim”—improving the patient
experience of care, improving the health
of populations, and reducing the per
capita cost of health care—as well as
improving clinician experience, a fourth
aim that others have proposed.7,8 Patient
and clinician experience improvements
are described above. Population health
only improves when the health outcomes
of many individuals improve, which is the
focus of value-based health care. Costs
also cumulate from the care provided
for individuals. By organizing teams to
care for individuals with similar needs, a
value-based approach enables expertise
and efficiency, rather than rationing, to
drive costs down. This puts decisions
about how to deliver care in the hands
of the clinical team, rather than those of
an insurance administrator, supporting
the professionalism of clinicians and the
power of clinician–patient relationships
to deliver effective and appropriate care.
A Framework for Implementing
Value-Based Health Care
Improving value in health care is
not an unreachable utopian ideal.
Around the globe, health care delivery
organizations—in varied payment
settings, with an array of regulatory
structures and many different care
traditions—have demonstrated
dramatically better health outcomes
for patients, usually at lower overall
costs.1,4,9,10 More than a decade’s research
into these organizations by 2 of the
authors (E.T. and S.W.) elucidates a
clear framework for value-based care
transformation.11
That framework, shown in Figure 1
and described below, can guide
organizations in building value-based
health care systems. This transformation
starts when the organization identifies
and understands a segment of
patients whose health and related
circumstances create a consistent set
of needs. A dedicated, co-located,
multidisciplinary team of caregivers
designs and delivers a comprehensive
solution to those needs. This integrated
team measures meaningful health
outcomes of its care for each patient
and the costs of its services and then
learns from that information to drive
ongoing improvements in care and
efficiency. Finally, as health outcomes
improve, evidence of better care creates
opportunities for the team to serve more
patients through expanded partnerships.
Understand shared health needs of
patients
Throughout the economy, service
providers organize their offerings around
a defined set of customers whose needs
are similar. Think, for example, of
transportation providers. Transportation
is an incredibly broad economic sector.
Services range from jets that deliver tons
of time-sensitive cargo to drones that
deliver individual bags of blood, and
from buses to rented electric scooters. In
each case, the transportation company
matches its services to the needs of its
customer segment.
Health care is the outlier. In the
health care system, most services are
organized around the service providers.
Endocrinologists practice in groups
with other endocrinologists, as do
cardiologists, ophthalmologists, and
podiatrists. A patient with diabetes, who
likely needs the coordinated—or, even
better, integrated—services of these
various clinicians, has to be the organizer.
Health care’s general failure to structure
around patient needs accounts for its
inconvenience and lack of integrated
services. Failing to structure for what is
common and routine also increases the
burden on caregivers, who too often must
improvise to solve routine problems. This
structural mismatch is a root cause of
why health care is so expensive and does
not deliver better results for patients.1
To be effective and efficient, health care
should be organized around segments
of patients with a shared set of health
needs, such as “people with knee pain”
or “elderly people with multiple chronic
Figure 1 Strategic framework for value-based health care implementation to achieve better
patient outcomes.
Invited Commentary
Academic Medicine, Vol. 95, No. 5 / May 2020684
conditions.” Organizing care in this
way allows clinical teams to anticipate
consistent patient needs and provide
frequently needed services efficiently,
doing common things well. The efficiency
afforded by structuring care around
patient segments frees clinicians from
scrambling to coordinate services that are
needed routinely. The added bandwidth
allows them to personalize services
for individual patients who may have
somewhat different needs.
Design a comprehensive solution to
improve health outcomes
Starting by identifying the common
needs of a patient segment enables teams
to design and deliver care that provides
a comprehensive solution for patients
or families. When the goal of care shifts
from treating to solving patients’ needs,
care teams can both address the clinical
needs of patients and begin to address
the nonclinical needs that, when left
unmet, undermine patients’ health.
For example, a clinic for patients with
migraine headaches might provide not
only drug therapy but also psychological
counseling, physical therapy, and
relaxation training. Similarly, a clinic
for patients with cancer might include
transportation assistance as a service for
those who have difficulty getting to their
regular chemotherapy appointments.
Broadening and integrating the services
provided to patients achieves better
outcomes by identifying and addressing
gaps or obstacles that undermine
patients’ health results.4
Integrate learning teams
Implementing multifaceted solutions
requires a dedicated team drawn from
an array of disciplines, many of which
are not typically viewed as medical.
An effective team integrates services,
reducing or even eliminating the need for
coordinators. Team members are often
co-located, enabling frequent informal
communication that supplements the
formal channels of communication to
ensure effective and efficient care. What
is critical is thinking together to improve
and personalize care and learning
together so health outcomes improve
with experience. The team structure can
also expand across locations, extending
state-of-the-art knowledge to remote
clinicians and enabling world-class
care to be delivered locally rather than
requiring patients to travel.
Measure health outcomes and costs
It is a truism of business that
management requires measurement.
Recognizing that the essential purpose
of health care is improving the health of
patients, it is axiomatic that health care
teams must measure the health results
as well as the costs of delivering care
for each patient. Leaders cannot align
health care organizations with their
purpose without measurement of health
outcomes. In addition, the current dearth
of accurate health outcomes and cost data
impedes innovation.
Measurement of results allows teams to
know they are succeeding. Measuring
health outcomes also provides the data
needed to improve care and efficiency.
Although caregivers are burdened with
reporting reams of information, they
rarely consistently track the health
outcomes that matter most to patients
and thus to themselves as clinicians. Cost
and health outcomes data also enable
condition-based bundled payment
models, empowering teams of caregivers
to reclaim professional autonomy and
practice clinical judgment—two integral
elements of professional satisfaction and
powerful antidotes to the affliction of
burnout.1,12–14
Measuring health outcomes is not as
complex as it is often perceived to be.
Routine clinical practice does not dictate,
nor can it support, the voluminous health
outcome measure sets used in clinical
research. Instead, clinicians need to focus
on measuring the outcomes that define
health for their patients. Those outcomes
cluster by patient segment—the
outcomes that matter most to patients
with congestive heart failure are strikingly
consistent while also markedly different
from the outcomes that matter most to
women who are pregnant. Within any
given patient segment, though, patients
define health in terms of capability,
comfort, and calm, as described above,
and these dimensions can be usually
captured in 3 to 5 measures. For
example, men undergoing prostate
cancer surgery are most concerned about
the common impairments from that
procedure—incontinence, impotence,
and depression—as well as time away
from work for recovery.
In addition to health outcomes, teams
must measure the costs of their services
for every patient. Cost-grouping
methodologies like the one developed at
the University of Utah15 or applications
of time-driven activity-based costing16
can provide the data teams need both
to demonstrate the value of their care
and to identify areas for improving their
efficiency.
Expand partnerships
Organizing around patients with shared
needs and demonstrating better value
in care create opportunities to expand
partnerships and improve health
outcomes for more people. For example,
with evidence of care that has fewer
complications and allows employees to
return to work more quickly, employers
are increasingly willing to contract
directly with providers and even to
pay more per episode of care than
they had previously, because faster and
fuller recovery reduces other employer
costs such as those associated with
absenteeism.17 Partnerships among
clinical organizations may also expand
as teams gain expertise and the ability
to work across more stages of the care
cycle or more locations. Integrated teams
may work with partners for an array of
reasons, such as using new technology
to share information with patients,
supporting rural clinicians as they
provide patients with care close to home,
or offering services to support lifestyle
changes in a community. These are
natural partnerships because the shared
goals of creating high value and achieving
better health outcomes for patients align
the interests of patients, family members,
employers, health plans, and clinicians,
as well as medical technology suppliers
whose services may facilitate these
relationships.
Value-Based Health Care in
Medical Education
Moving to a system of value-based
health care requires that physicians and
physicians-in-training learn to think
differently about their role within the
larger care team, about what constitutes
an effective care solution, and about
the importance of measuring the health
outcomes that matter most to patients.
That learning should begin during
medical school.
The University of Texas at Austin’s Dell
Medical School (DMS) offers an example
Invited Commentary
Academic Medicine, Vol. 95, No. 5 / May 2020 685
of how education about value-based
health care can be incorporated into
undergraduate medical education.
Throughout the 4 years of medical
school, DMS students study the
principles of value-based care delivery
described above. During their clinical
rotations, they also see these principles in
practice in UT Health Austin’s affiliated
clinics, which are organized around
segments of patients with shared needs
and designed to provide comprehensive
solutions delivered by interdisciplinary,
outcomes-focused teams.
In the joint pain clinic, for instance,
DMS students observe interdisciplinary
care teams providing comprehensive
care to alleviate pain and improve
functioning. Treatment may entail joint
replacement surgery, but surgery is used
less frequently than in other orthopedic
care settings. Those who are not seen
as good candidates for surgery might
instead receive physical therapy, care
for depression or other mental health
conditions, and/or weight loss support.
The clinic tracks health outcomes for all
patients and has found that while the
rate of patients receiving lower extremity
surgery is 30% lower than that of patients
in conventional care settings, more
than 60% of patients report significant
reductions in pain and improvements
in function 6 months after the initial
appointment.18
DMS’ curriculum also allows third-
year medical students (and other
interested health professionals) to
complete a master’s degree in health care
transformation, focused on the principles
and implementation of value-based
health care. We encourage other medical
schools to incorporate similar training
throughout their curricula to prepare
their graduates to lead the transformation
to value-based health care as they enter
the physician workforce.
Funding/Support: None reported.
Other disclosures: None reported.
Ethical approval: Reported as not applicable.
E. Teisberg is executive director, Value Institute for
Health and Care, Cullen Trust for Higher Education
Distinguished University Chair in Value-Based Care,
Dell Medical School, and professor, McCombs School
of Business, University of Texas at Austin, Austin, Texas.
S. Wallace is managing director, Value Institute
for Health and Care, and associate professor,
Department of Education, Dell Medical School,
University of Texas at Austin, Austin, Texas.
S. O’Hara is a course and content specialist, Value
Institute for Health and Care, Dell Medical School,
University of Texas at Austin, Austin, Texas.
References
1 Porter ME, Teisberg EO. Redefining Health
Care: Creating Value-Based Competition
on Results. Boston, MA: Harvard Business
School Press; 2006.
2 Cimino A, Giorda C, Meloncelli I, et al.
Indicators of Quality of Diabetes Care in
Italy: The AMD Annals. Rome, Italy: AMD
Associazione Medici Diabetologi; 2006.
3 Reinhardt UE. Health Reform: Porter and
Teisberg’s utopian vision. Health Affairs.
https://www.healthaffairs.org/do/10.1377/
hblog20061010.000063/full. Published October
10, 2006. Accessed November 12, 2019.
4 Wallace S, Teisberg EO. Measuring
what matters: Connecting excellence,
professionalism, and empathy. Brain Inj Prof.
2016;12:12–15.
5 Montori VM. Turning away from industrial
health care toward careful and kind care.
Acad Med. 2019;94:768–770.
6 Deerberg-Wittram J, Ludtke L. Diabeter:
Value-Based Healthcare Delivery in Diabetes.
Boston, MA: Boston Consulting Group; 2016.
7 Berwick DM, Nolan TW, Whittington J. The
triple aim: Care, health, and cost. Health Aff
(Millwood). 2008;27:759–769.
8 Bodenheimer T, Sinsky C. From triple
to quadruple aim: Care of the patient
requires care of the provider. Ann Fam Med.
2014;12:573–576.
9 Morales DL, Carberry KE, Heinle JS,
McKenzie ED, Fraser CD Jr, Diaz LK.
Extubation in the operating room after
Fontan’s procedure: Effect on practice and
outcomes. Ann Thorac Surg. 2008;86:576–581.
10 Stowell C, Akerman C. Better value in health
care requires focusing on outcomes. Harvard
Business Review. https://hbr.org/2015/09/
better-value-in-health-care-requires-
focusing-on-outcomes. Published September
17, 2015. Accessed November 21, 2019.
11 Teisberg EO, Wallace S. Creating a high-value
delivery system for health care. Semin Thorac
Cardiovasc Surg. 2009;21:35–42.
12 Andrawis JP, McClellan M, Bozic KJ. Bundled
payments are moving upstream. NEJM
Catalyst. https://catalyst.nejm.org/bundled-
payments-upstream-musculoskeletal-
institute. Published February 26, 2019.
Accessed November 20, 2019.
13 Kaplan RS, Porter ME. The big idea: How to
solve the cost crisis in health care. Harvard
Business Review. https://hbr.org/2011/09/
how-to-solve-the-cost-crisis-in-health-
care. Published September 2011. Accessed
November 21, 2019.
14 Porter ME, Barron JF, Chacko JM, Tang
R. The UCLA Medical Center: Kidney
Transplantation. Boston, MA: Harvard
Business School Publishing; 2012.
15 Kawamoto K, Martin CJ, Williams K, et
al. Value Driven Outcomes (VDO): A
pragmatic, modular, and extensible software
framework for understanding and improving
health care costs and outcomes. J Am Med
Inform Assoc. 2015;22:223–235.
16 Kaplan RS, Anderson SR. Time-driven
activity-based costing. Harvard Business
Review. https://hbr.org/2004/11/time-driven-
activity-based-costing. Published November
2004. Accessed November 20, 2019.
17 Woods L, Slotkin JR, Coleman MR. How
employers are fixing health care. Harvard
Business Review. https://hbr.org/cover-
story/2019/03/how-employers-are-fixing-
health-care. Published March 2019. Accessed
November 12, 2019.
18 Koenig K, medical director, Musculoskeletal
Institute, UT Health Austin. Personal
communication with S. Wallace and S.
O’Hara, June 20, 2019.
https://www.healthaffairs.org/do/10.1377/hblog20061010.000063/full
https://www.healthaffairs.org/do/10.1377/hblog20061010.000063/full
https://hbr.org/2015/09/better-value-in-health-care-requires-focusing-on-outcomes
https://hbr.org/2015/09/better-value-in-health-care-requires-focusing-on-outcomes
https://hbr.org/2015/09/better-value-in-health-care-requires-focusing-on-outcomes
https://catalyst.nejm.org/bundled-payments-upstream-musculoskeletal-institute
https://catalyst.nejm.org/bundled-payments-upstream-musculoskeletal-institute
https://catalyst.nejm.org/bundled-payments-upstream-musculoskeletal-institute
https://hbr.org/2011/09/how-to-solve-the-cost-crisis-in-health-care
https://hbr.org/2011/09/how-to-solve-the-cost-crisis-in-health-care
https://hbr.org/2011/09/how-to-solve-the-cost-crisis-in-health-care
https://hbr.org/2004/11/time-driven-activity-based-costing
https://hbr.org/2004/11/time-driven-activity-based-costing
https://hbr.org/cover-story/2019/03/how-employers-are-fixing-health-care
https://hbr.org/cover-story/2019/03/how-employers-are-fixing-health-care
https://hbr.org/cover-story/2019/03/how-employers-are-fixing-health-care
Center for Improving Value in Health
Saudi Arabia
Global Innovation Hub for
Improving Value in Health
TRANSFORMATION TOWARDS
SUSTAINABLE HEALTH
SYSTEMS THROUGH VALUE
BASED HEALTH CARE
2
www.g20.org
|Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
This report collects key messages and insights f rom the G20 side event on Value
Based Health Care (VBHC) held in Riyadh between the 13th and 14th of January,
2020. The event, Accelerating Transformation Towards Sustainable Health Systems
Through Value Based Health Care (‘the event’) was attended by G20 member
and guest nations, representatives of relevant International Organizations (IOs),
international experts, and 200 delegates f rom more than 25 countries.
The Report is intended as a resource for policymakers and leaders of healthcare
institutions and provides practical learnings on the introduction of VBHC in health
systems.
The event was hosted by the Center for Improving Value in Health (‘the Center’)
and had contributions f rom 25 speakers f rom 21 countries for keynotes and panels.
It provided a platform for presentations and panel discussions for international
experts and more than 50 policymakers and senior professionals f rom the Saudi
Health System and civil society.
The event offered an opportunity to share experience and learning f rom both
public and private sectors and was embraced by the participants with high levels of
engagement, a sense of openness and a willingness to collaborate to improve the
lives of citizens.
Figure 1: The G20 Side Event on VBHC
was attended by international and local
stakeholders
“This event focuses
on sharing practical
knowledge on VBHC”
Dr. Reem Al Bunyan,
CEO, Center for Improving
Value in Health,
Saudi Arabia
3www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
Editors:
Prof. Rifat Atun, Harvard University
Dr. Reem Al Bunyan, The Center for
Improving Value in Health, Saudi Arabia
Dr. Sara Al Munif, The Center for
Improving Value in Health, Saudi Arabia
Ms. Israa Bargo, The Center for
Improving Value in Health, Saudi Arabia
Mr. Craig Barratt, The Center for
Improving Value in Health, Saudi Arabia
Dr. John McGhee, The Center for
Improving Value in Health, Saudi Arabia
Contributors:
In alphabetical order
Ms. Herta Adam, European Commission
Mr. Lav Agarwal, India
Dr. Ahmed AlJedai, Saudi Arabia
Mr. Lars Dahl Allerup, Denmark
Dr. Riyadh AlShamsan, Saudi Arabia
Dr. Omar AlShanqeety, Saudi Arabia
Dr. Pilar Aparicio, Spain
Mr. Sinan Atlig, Pfizer
Mr. Ramon Maspons Bosch, Spain
Mrs. Francesca Colombo, OECD
Mrs. Joanne Fitzgerald, Australia
Prof. Lyalya Gabbasova, Russia
Mr. Said Haddad, Johnson & Johnson
Dr. Derrick Heng, Singapore
Dr. David Herr, Germany
Mr. Christian Howell, Medtronic
Dr. Chintan Maru, Leapf rog to Value
Dr. Giuseppe Ruocco, Italy
Mr. Mohammed Saleh, Microsoft
Mr. Lucas Scherdel, WEF
Dr. Peter Schmeissner, United States
Dr. Agnès Soucat, WHO
Mr. Erik Jan Wilhelm, Zilveren Kruis
Ms. Rose Willis, United Kingdom
4www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
SUMMARY OF KEY FINDINGS
I. Value Based Health Care (VBHC) can be realized in diverse settings
A. VBHC approaches introduce innovative solutions to transform health
systems to achieve greater value and sustainability
B. VBHC involves development of new data tools and analytic capabilities
C. Non-financial incentives and factors impact on the success of VBHC
D. A key enabler for the successful delivery of VBHC is the ability to identify and
measure outcomes that matter most to patients and individuals
II. Collaborations and knowledge transfer between organizations and countries is
critical for scale-up of VBHC solutions
E. VBHC can be applied and have a positive impact at different scales
F. Reciprocity among stakeholders is key to successful collaborations
G. Large scale VBHC projects only succeed when roadblocks are removed to
implementation
H. Practical learning is transferable within and across countries
I. One can leverage learning f rom other sectors in introducing and scaling up
and new solutions
5www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
INTRODUCTION
Current trends in global health care systems
Globally, health systems face immense challenges due to growing demand for
healthcare services, rising costs and increasingly higher complexity of illnesses
and the solutions to address them. Despite higher expenditures in health systems,
health outcomes have not improved proportionately. From 2014 to 2040, the world’s
spend on health care will increase almost three-fold to reach approximately USD 25
trillion per year. This increase in spending is primarily driven by population ageing,
increase in chronic diseases, inefficiency and ineffectiveness of the current health
system models and rising costs f rom f new technologies. In most Organisation for
Economic Co-operation and Development (OECD) countries, health care spending
is increasing faster than economic growth and driven by ageing and excess cost
growth due to chronic illness and health system inefficiency.
The combined challenges of rising demand and growing healthcare expenditures
that exceed the rate of economic growth pose a real and substantial risk to
sustainability of health systems and their ability to provide better access to safe,
high quality health care to all citizens of the G20 member countries and beyond.
The conference participants recognized the critical importance of VBHC as an
instrumental part of achieving sustainable Universal Health Coverage (UHC) – a
target for the United Nations Sustainable Development Goal (SDG) 3 ‘Ensure healthy
lives and promote well-being for all at all ages’.
Accelerating the introduction and scale-up of VBHC to address pressing threats to
health systems
“Health systems are
organized to manage
acute events with single/
short interventions and
need moving towards
medium long term
health caring”
Dr. Giuseppe Ruocco,
Ministry of Health, Italy
Health systems need to transition to
VBHC models that achieve a better
balance of outcomes and resources used,
if the current threats are to be effectively
managed.
Traditional health care models use
payment models, such as ‘fee for service’
and ‘activity-based payment’, primarily
focus on paying for inputs and the volume
of services provided. These payment
models reward ever-increasing volumes
of activities and services, but without the
commensurate improvement in health
outcomes and outcomes that matter
to patients. By contrast, VBHC offers a
model that puts increasing value for
6www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
patients first. This includes increased accessibility and affordability of health care for
patients. From the pharmaceutical industry perspective, Mr. Sinan Atlig f rom Pfizer,
the Regional President Biopharma MEA, agreed that health care costs and overall
spending on medicines need to be kept in check to create a more sustainable health
system.
Furthermore, many panelists emphasized the importance of defining health care
broadly to include maintenance or improvement of health through promotion of
good health, and through prevention, diagnosis and treatment of illness, injury,
disease, and other physical or mental impairments.
The CEO of the Center for Improving Value in Health, Dr. Reem AlBunyan highlighted
in her opening remarks, that value in health care means better health and better
care at lower cost, with a focus on the person. Ms. Herta Adams, the Deputy Head
of Health Determinants and International Relations at the European Commission
proposed a similarly broad definition of value in the context of VBHC, based on four
pillars:
Personal value, accomplishing the personal goals of patients
Technical value, achieving highest outcome quality with limited resources
Population value, equitably delivering care and allocating resources to
patient populations
Societal value, fostering societal participation by delivering health care
The report consists of two sections that explore: (1) how VBHC can be realized through
common key enablers in diverse settings and (2) how collaborations and knowledge
transfer between organizations and countries help spread and establish successful
VBHC models.
“VBHC means placing
patients – both their
experience and
outcomes – at the heart
of decision-making”
Ms. Herta Adam,
European Commission
7www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
SUMMARY OF KEY FINDINGS
1. VBHC can be realized through common enablers in diverse settings
The first section of this report explores learnings f rom various sessions revolving
around the necessary building blocks and levers to set up VBHC models. Case
studies illustrate key learnings and best practices on VBHC approaches.
A. VBHC approaches introduce innovative solutions to transform health
systems to achieve greater value and sustainability
According to Prof. Rifat Atun, Professor of Global Health Systems at Harvard University,
current health systems are under pressure because of a coalescence of several
fundamental contextual drivers that pose threats, including rapid population ageing,
rising burden of chronic illnesses, higher patient expectations, fiscal constraints on
governments due to economic downturns. This is against a backdrop of an ever-
increasing innovation in new health technologies, diagnostics and medicines which
are not optimally harnessed to improve system performance.
Worldwide, health systems are struggling to achieve desired levels of effectiveness,
efficiency, equity and responsiveness that meet user expectations. This is because
health systems are failing to innovate because of:
(i) Innovation misalignment: There is robust ‘delivery of innovations’, such as
new diagnostics, health technologies and medicines, but there is
a failure in the ‘innovation in delivery’, with few fundamental
improvements in healthcare service delivery models which remain archaic.
(ii) Policy misalignment: Innovation is encouraged for by industrial policies but
their adoption and scale-up is constrained by health policies
8www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
In his presentation, Prof. Atun explained
how health systems have historically
transitioned f rom one model to another
over time: f rom a model with structural
focus, where payments followed
inf rastructures (e.g. hospitals), to a more
functional focus, in which payments
follow activities. However, we are currently
observing a new fundamental shift
towards value, and the many practical
examples shared during the event are
the best examples of this trend. In this
model, payments follow value, with
better health outcomes for individuals
or populations, as exemplified by VBHC
approaches.
“There is no shortage
of innovation, but they
are in pockets. We need
to scale innovation to
create impact”
Ms. Herta Adam,
European Commission
“Everything is new, and
we do not have off-the-
shelf solutions. There
is no such thing as
failure, it is a learning
experience”
Prof. Rifat Atun,
Harvard University
According to Prof. Atun, VBHC is underpinned by four major principles: transparency,
optimization of cost and outcomes, shared accountability, and shared risk and reward.
In practice VBHC models include several core critical components. The first of these
is ‘digital data systems’ and analytic capability that enable pooling and application
of data to measure costs, processes and outcomes. The second is ‘stratification’ or
‘categorization’ of population groups according to their characteristics to ensure
better understanding of needs and the development of targeted solutions. The third
is ‘risk-adjusted bundled care’, which brings together a targeted set of interventions
across the care continuum for populations with different levels of risk of illness.
The fourth component is the ‘integrated care’ across the care continuum, enabled by
‘integrated care pathways’ and integrated provider networks that ensure seamless
care. An integrated care pathway ideally contains the entire patient journey for a
specific disease, f rom prevention, detection, treatment, monitoring to rehabilitation
(Figure 2). The fifth component is
‘outcome based payment models’
that incentivize achievement of better
outcomes and lowering of costs. The sixth
and often overlooked crucial component
is ‘behavior change’ in policymakers,
payers, healthcare providers and health
service users that creates a conducive
environment to challenge the status quo
and encourage the development and
scale-up of VBHC solutions to address the
current weaknesses in health systems.
9www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
Figure 2: Integrated Care Pathways
Legacy Focus
VBHC Focus
Digital systems are a critical ingredient that connects these components to enable
the development of person centered integrated care delivery models aimed at
achieving better patient reported outcomes and creation of greater value, as
illustrated by case study [1]: DigiPROM at Charite Berlin.
Case study [1]: DigiPROM at Charite Berlin
Situation: The Federal Ministry of Health, Germany wanted to study the feasibility
of patient-reported outcome measures (PROMs) in German hospitals.
Ambition: Testing digitalized PROMs for back pain-related interventions and
exploring DigiPROM as a concept for standardization across Germany. Additionally,
this pilot tests the technical feasibility to combine clinical data with PROMs.
Outcome: Patients benefited f rom improved treatment due to quality transparency
and improvement.
Case study [2]: Ayushman Bharat
Situation: The Indian government strives towards UHC for its population to address
urging issues such as the increasing burden of non-communicable diseases and
rising healthcare costs.
Ambition: In 2017, the Ayushman Bharat program was launched. This program
rests on two main pillars, AB-HWCs and PMJAY. AB-HWCs, Ayushman Bharat
Health and Wellness Centers, provide primary care to all citizens for an initially
limited selection of health services. PMJAY, Pradhan Mantri Jan Arogya Yojana,
aims to prevent excessive health expenditures for the 40% poorest for secondary
and tertiary health care services.
Outcome: To date, 27,923 AB-HWCs have been transformed across the country and
provided treatment to 11.3m. people for hypertension, 5.7m. for diabetes and 9.2m.
women screened for breast cancer. To date, approx. 20,000 hospitals participate in
PMJAY and 7.5m. patients were treated under this cover worth USD 1.8 bn.
10www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
With a focus on outcomes, VBHC considers the whole
patient journey, including disease prevention and
monitoring of health. This contrasts with legacy systems
that mainly focus on diagnosis and treatments, driven
by their focus on reimbursement by activity (e.g. fee-for-
service, and activity-based payment models)
Focusing the care pathways only on the direct delivery of clinical care would be too
narrow an approach in achieving value. Around 80% of the factors that determine
health outcomes lie not within healthcare services but include social-determinants
of health, as Mr. Lav Agarwal, the Joint Secretary, Indian Ministry of Health & Family
Welfare, points out. These factors include nutrition, sanitation and water access, air
pollution, physical activity, and among others consumption of alcohol and tobacco.
That is why the Indian government has acted and initiated programs aimed at
educating its population and promoting healthier lifestyles. Such programs include
“Fit India Movement” and “Eat Right India.” Those programs are part of a larger
national initiative, Ayushman Bharat: “Bless India with long healthy life” in Hindi,
which aims to provide UHC in primary care, based on VBHC principles (case study
[2]: Ayushman Bharat).
The delegation f rom Australia shared a case study f rom their recent reforms on
activity-based funding (ABF) that aimed to achieve the following:
Improving access to services for patients
Increasing the efficiency of public hospitals
Improving the transparency and sustainability of public hospital funding
The development and implementation of funding and pricing approaches
for safety and quality.
As part of these reforms, the Independent Hospital Pricing Authority (IHPA) was
established to determine the National Efficient Price, and set up the underlying
inf rastructure to support ABF, such as classifications, data specifications, costing
and pricing. (Case study [3]: Pricing Framework for Australian Public Hospitals)
IHPA’s decisions on these matters were detailed in the Pricing Framework for
Australian Public Hospital Services 2017-18:
1. No funding for a public hospital episode including a sentinel event, applying
to all relevant episodes of care in all hospitals
2. Reduced funding level for all Hospital Acquired Complications, to reflect the
additional cost of a hospital admission with a hospital acquired complication
3. Undertake further public consultation to inform a future pricing and funding
approach in relation to avoidable hospital readmissions, based on a set of
definitions to be developed by the Australian Commission on Safety and
Quality in Health Care.
11www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
Case study [3]: Pricing Framework for Australian Public Hospitals
Ambition: to improve Australians’ health outcomes and decrease avoidable
demand for public hospital services.
Outcome: Whilst this reform is still in the early stages of implementation, there
is evidence that clinicians and hospital managers are using the pricing signal to
undertake significant work to improve the rate of occurrence of HACs across the
system.
B. An enabler for the successful delivery of VBHC is the ability to identify and
measure outcomes that matter most to individuals
During the event Ms. Francesca Colombo, Head of the OECD’s Health Division,
highlighted two noteworthy numbers: first, the average spending of 36 OECD
member states for health care amounts to around 8.8% of GDP, with the United
States leading the pack with 16.9%; second, some OECD countries (e.g. Switzerland
and Germany), undertake six times more knee replacement operations per 100,000
inhabitants than others (e.g. Poland). This raises the questions of whether this
spending and variance in elective surgery rates are due to over/undertreatment, or
differing health needs in different contexts and if a country is getting value f rom the
spending and health interventions.
Such questions can only be answered by measuring patients’ wellbeing before
and after the treatment. To measure the value of treatments in a comprehensive
and standardized way, the OECD launched the Patient-Reported Indicators
Surveys (PaRIS) initiative. This initiative aims to record and report the outcomes
and experiences of patients undergoing various health treatments. Initial data sets
collected by PaRIS include data regarding hip surgery and cancer, with data points
that include; the extent of pain and when they are mobile after surgery, what health-
“It is not about what is
being done in health
systems, it is about what
health systems deliver to
people and populations
at large”
Ms. Francesca Colombo, OECD
related quality of life do patients have
undergoing cancer therapy and to what
extent do they experience anxiety and
depression.
Ms. Colombo highlighted that when
people are at the center of health care,
we can really measure what matters to
them and therefore bring benefits to
them as well as healthcare providers and
policymakers. For example, healthcare
providers profit f rom a new source of
feedback to improve the quality they
12www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
provide. On the other hand, policymakers profit f rom focusing their attention on
areas that need legislative improvements to create a favorable environment for the
development of person-centered healthcare delivery models.
The OECD is not alone in investing in programs that record and report the outcomes
and experiences of patients undergoing various health treatments. Many countries
such as Germany and Spain are designing systems for recording and measuring
patient-reported outcomes in pilot projects. For example, Germany has had over
15 years of experience with patient-reported outcomes (Case study [1]: DigiPROM
at Charite Berlin; Case study [4]: Martini Hospital in Hamburg). The collection of
data on patient-reported outcomes has had important implications for the quality
of treatment, as Dr. David Herr f rom the German Ministry of Health explained. The
follow up of clinical outcomes provided valuable feedback to individual surgeons
that actively started to learn f rom each other by sharing best practices. The use of
patient-reported outcomes have enabled similar success in the case of prostate
cancer in Spain (case study [5]: Patient-centric care in Cruces Hospital)
Lastly, to facilitate measurement of outcomes and patient experience, new data tools
and analytic capability are currently being developed, which are discussed later in
this paper.
Case study [4]: Martini Hospital in Hamburg
Situation: The Martini hospital, specialized in high volume radical prostatectomy
introduced patient-reported outcomes collection.
Ambition: Recording systematically the clinical outcome of patients after one
week, six months, and then yearly.
Outcome: Patient-reported outcomes provided valuable feedback to individual
surgeons and allowed best practice sharing, raising the quality of clinical
intervention.
13www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
Case study [5]: Patient-centric care in Cruces Hospital
Situation: The Cruces Hospital in Spain deploys the guidelines of the International
Consortium for Health Outcomes Measurement (ICHOM) for a selection of
candidate diseases. One of them is prostate cancer.
Ambition: Creating a more patient-centric approach by following the guidelines
f rom ICHOM. These guidelines include the collection of standardized data sets
concerning clinical parameters as well as information reported by patients.
Outcome: In the case of prostate cancer, symptoms such as urinary incontinence
were significantly reduced after treatment, along with decreasing costs. The
initiative is interesting due to the long history of structured recording of the clinical
data and PROMs in the Basque health system.
C. Non-financial incentives and factors impacting the success of VBHC
The panel discussions revealed that money is only one lever in an inventory of
incentives to drive the transformation of health systems towards VBHC. Recalling that
change is driven by people, Dr. Omar Alshanqeety, CEO of the Program for Health
Assurance and Purchasing in Saudi Arabia, illustrated that purpose, fulfillment and
enjoyment are also incentives that people long for in their work. Addressing the
intrinsic motivation of people is key, summarized Dr. Derrick Heng, Group Director
of Public Health f rom the Singaporean Ministry of Health.
Recognition for achievements can also be a powerful incentive: the representative
f rom the Russian Ministry of Health, Prof. Lyalya Gabbasova, Assistant to the Minister
of Health, highlighted that best practice primary care facilities in Russia are awarded
special recognition and status, and to date more than 3,000 facilities have received
this status.
“Finance is necessary,
but not suff icient to
change behavior ”
Dr. Omar Alshanqeety, Program
for Health Assurance and
Purchasing, Saudi Arabia
People need to feel empowered and
enabled by providing them the right
tools. Additionally, health professionals
and patients should be empowered to
take ownership of their decision making.
This links back to motivating people by
allowing them to change the system, as
it was emphasized by Mr. Agarwal f rom
the Indian Ministry of Health & Family
Welfare. He also raised the important
point of educating consumers, not
only current patients, and taking them
along on the journey of health system
transformation.
14www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
D. VBHC involves development of new data tools and analytic capabilities
Data are available in vast quantities, and this holds true in the context of health care.
However, these data should not only be collected but also connected, as agreed
by virtually all the speakers who were present at the conference. Only by linkage of
available data, can one readily leverage advanced analytics to derive genuine and
novel insights for improving health care. Microsoft’s MEA Regional Industry Lead for
Health and Life Sciences, Mr. Mohammed Saleh identified five major categories of
health data:
Clinical (e.g. electronic medical records [EMRs], images, scans)
Pharma and life sciences (e.g. clinical trials, -omics)
Patient and citizen (e.g. purchasing data)
Claims and cost (e.g. claim requests, prices)
Geo/social/environmental (e.g. social service data)
Most of the data exist for every patient today but the data sources cannot talk to each
other. Today, we have a system of record, e.g. Electronic Medical Records (EMRs),
with little analytics and connections to other data sets. Pushing the connectivity and
exchange of data and adding analytical capabilities, systems of Artificial Intelligence
(AI) can be created. Such systems can cross-analyze large amounts of data in real-time
and perform prescriptive analytics (Case study [6]: Ochsner Health System).
To enable such technologies, it is important to share comprehensive high-quality
data sets among institutes, entities and across borders. In interviews with the
organizers of the event, both the representatives f rom the OECD and the World
Economic Forum (WEF) called for better global convergence and interoperability of
digital systems and outcome measurements.
“It’s not about
integrating data, it’s
about connecting data”
“From a technology
perspective the tools are
there. The open question
is governance, and this is
the big issue nowadays.
Who owns the data?”
Mr. Mohammed Saleh,
Microsoft
“It is not just good
enough to collect data,
there are oceans of data.
We need analytics that
generate intelligence ”
Prof. Rifat Atun,
Harvard University
15www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
Case study [6]: Ochsner Health System
Situation: Physicians supervise many patients and need to split their time among
those patients. Cardiac arrests in patients can occur and are life-threatening
without warning signs perceivable by treating physicians.
Ambition: Predict cardiac arrest before it happens by reading out vitals every
minute and having them analyzed and processed by machine learning algorithms.
If the software detects suspicious signals it will alert a physician preemptively.
Outcome: During a 90-day pilot, the number of patients suffering a cardiac or
respiratory arrest was reduced by 44%, mortality by cardiac arrest was eliminated
on the floor during the trial.
Mr. Agarwal f rom the Indian Ministry of Health & Family Welfare showcased that
creating a unique medical ID on a robust IT platform is pivotal to ensure interoperability
as:
–
“What we need to do is
to create a standard for
interoperability that is
applicable in different
realities. It needs to be
focused on outcomes.”
Mr. Lucas Scherdel, WEF
“ There is a strong
impetus to extract more
value f rom our health
care system to ensure
sustainability”
Dr. Derrick Heng,
Ministry of Health, Singapore
It will allow information and
feedback flow across primary,
secondary and tertiary healthcare
by mapping of facilities and
resources at every level of care
It enhances trust and accountabili
ty due to simplicity and ease of use
It allows robust f raud detection
by developing an integrated fraud
and abuse detection system
Dr. Heng f rom the Singaporean
Ministry of Health mentioned that by
collecting and analyzing clinical quality
and cost data, one cannot only identify
opportunities for improvements but also
determine best practices for enhancing
value.
A survey of the event participants revealed
that one third of the respondents thought
co-production of solutions by different
stakeholders the most important
enabler, but 44% thought consulting
widely, challenging existing practice, co-
production of solutions and the creation
of an ecosystem for innovation were all
equally important (Figure 3).
16www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
Figure 3: Event Audience Poll. Which is
the most important VBHC enabler?
II. Collaborations and knowledge transfer between organizations and countries
This section explores key learnings f rom insightful presentations and panel
discussions on the critical success factors and those hindering the establishment of
VBHC models.
E. VBHC can be applied and have positive impact at different scales
There is no ‘boilerplate’ solution that countries can simply pick up and apply
unchallenged locally. During the meeting a variety of pilots, initiatives and international
“In the past we talked a
lot about VBHC, but we
had no actions. We had
those actions in the last
years”
“When [VBHC] was
def ined it was more a
doctrine, now it is much
more practical because
we are working with
it. We understand now
what we can do with it
and what not”
Mr.Erik Jan Wilhelm, Zilveren
Kruis
collaborations were presented. All the
examples were guided by trust between
the involved parties, transparency and
alignment on collected data, focus
on solutions rather than procedures,
novel reimbursement models, a shift in
mindset of all involved stakeholders, and
patient centricity.
The presence of multiple examples across
different geographical scales provides a
proof-of-concept that VBHC approaches
can be adapted, applied and scaled-up
globally in different realities and scopes.
Panelists at the conference were positive
that the philosophy of VBHC is now
supported by concrete examples f rom
different geographies and sizes.
The acceleration of the introduction and
17www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
adoption of different VBHC approaches has happened thanks to public and private
initiatives and partnerships at varied scales, such as the partnerships between Roche
and The Capital Region in Denmark or Medtronic with the Lehigh Valley Health
Network. Or partnerships that require government commitment at the region or
country level, such as in Italy and Saudi Arabia, and investments in programs by
“You cannot scale
solutions ‘as is’, each
solution is different”
Mr.Said Haddad, Johnson &
Johnson
large international organizations at the
international scale.
As an example of a countrywide program,
involving many stakeholders, are UK
initiatives with the aim to measure
variability in the delivery of care. As
described by Ms. Rose Willis, Deputy
Head of Provider Efficiency at the UK
Department of Health and Social Care, a
large volume of inefficient spending in the
system can be prevented by measuring
and communicating variability in care
and clinical practice, e.g. in hospitals. The
measures of variability were developed
in collaboration with clinicians to whom
they will apply. £2.9bn has been saved so far by this initiative.
On the interface between the country and international levels, panelists presented
first international collaborations across country borders, for example the collaboration
between a Dutch payer and a clinic in Germany for prostate cancer surgeries (Zilveren
Kruis in the Netherlands and the Martini Clinic in Hamburg), or the innovative shared
value-driven procurement by Catalonia in Spain and in the UK for pacemakers.
18www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
Figure 4: Some of the initiatives, pilots, and case studies discussed during the G20
side event on VBHC, categorized by geography and type
F. Reciprocity is key to successful collaborations
Successful VBHC pilots which are subsequently extended for broader implementation
require strong ties and goal alignments with collaborators. Collaboration between
patients, physicians, and organizations need to be mutually beneficial to create
strong enough incentives for all concerned to follow through and generate a
necessary level of trust.
To create this trust and special relationship, it is important to ascertain what VBHC
means to each party. VBHC means different things to different people, for example
better outcomes, better efficiencies, or better financials. The knowledge of context
is instrumental to align on what the partners want to achieve by the use of VBHC
approaches.
“Partnership is not a
noun, it is a verb. You
have to work on it”
Prof. Rifat Atun,
Harvard University
Mr. Said Haddad, the VP of Ethicon
EMEA for Johnson & Johnson, called for
a value based model of procurement.
Traditionally, stakeholders oppose each
other when negotiating prices. The
purchaser wants to obtain the lowest
possible price whereas the vendor wants
the opposite. To overcome the fixation
on price as the key decision criteria,
requires a shift in the mindset – moving
away f rom price and towards value.
Interestingly, 70% of procurement in
19www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
Europe is done through tenders with a major focus on price. In the Middle East, Mr.
Haddad explained, it’s even higher at 80-85%.
Mr. Ramon Maspons, the Chief Innovation Officer f rom the Catalonian Ministry of
Health, shared a similar view. He described the need to transition to value based
procurement as more and more innovations and solutions that incorporate
innovative technologies, devices and medicines developed and more funds are
needed to procure these innovations.
There are 22 value based projects currently operational in Catalonia. In one of
the pilots, the Catalonian Ministry of Health procured an automated implantable
cardioverter defibrillator for Sant Pau hospital (Case study [7]: The Sant Pau case). This
defibrillator allowed remote monitoring and included cardiac re-synchronization,
adding another layer of care for patients. Patient outcome and wellbeing improved,
hospital productivity rose and their overall spending was reduced while the vendor
made money. This was achieved because procurement was not of the product
“We are not buying
products anymore, we
are buying solutions.
Using procurement as a
trigger for change”
Prof. Rifat Atun,
Harvard University
“Never before was there
so much innovation in
the funnel, never before
so little money for
adopting innovation”
Ramon Maspons, Catalonian
Ministry of Health
alone, but an entire solution; the remote
monitoring and maintenance of each
device was done by the manufacturer,
reducing hospital visits by the patients,
and both sides had the common aim of
lowering mortality as an incentive.
20www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
Case study [7]: The Sant Pau case
Situation: The Catalan Ministry of Health, together with the Hospital de la Santa
Creu I Sant Pau piloted a new approach to treat patients with arrhythmias
(abnormal heartbeats)
Ambition: Comprehensive treatment of patients with automated implantable
cardioverter defibrillator (AICD) including cardiac re-synchronization. This
approach was designed to reduce follow up visits, introduce remote monitoring
of patients, and reduce costs by including health outcomes.
Outcome: Quality of life for patients improved drastically. There was a reduction
in in-office visits by ca. 10% and reduction of inappropriate shocks per patient by
ca. 66%.
Mr. Lars Dahl Allerup, New Business Development Manager f rom The Capital Region
of Denmark, concurred that price should not be the most important criterion in
procurement. Moreover, he raised the question if providers of care really knew what
patients want. What really matters to the specific patient? The answer for Mr. Allerup
is obvious: ask the patient and then center health care around patients and their
needs and wellbeing (Case study [8]: Public-private partnership (PPP) for a new way
to treat renal cancer).
Case study [8]: Public-private partnership (PPP) for a new way to treat renal
cancer
Situation: The Capital Region of Denmark and Roche Pharma started a public-
private partnership for patients with non-clear celled renal cancer.
Ambition: Tailoring the treatment to the patient’s need and thereby increasing
patient value overall. This will be achieved by collecting data on patient-reported
outcomes and real-world treatment data.
Outcome: First patient treatment will start in February 2020.
21www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
Another example of adapting the mentality of achieving greater value for money in
health systems was provided by Singapore. The Singaporean Ministry of Health has
initiated a comprehensive health care transformation process that rests on three
pillars:
Value driven care
Bundled payments
Health technology assessment & value based pricing
The first pillar, value driven care, is implemented through a major initiative in
Singapore currently focusing on 17 conditions treated in an inpatient setting, with
the aim to improve outcomes and lowering costs. More details of the value driven
care initiative are provided in case study [9].
The second pillar, bundled payments (i.e. single price for all services required during
a patient’s entire episode of care), present an effective way to cap the costs of
treatment: payments are based on the diagnosed condition and not the volume of
procedures or the hours spent by a physician managing a case. Singapore plans to
introduce this bundled payment system by 2021.
For the third pillar, the Agency for Care Effectiveness (ACE) was established in
2015. The ACE conducts health technology assessments (HTA) to evaluate health
technologies in terms of clinical and cost effectiveness. HTA assessment agencies are
widespread in Europe, North America and Japan, and Singapore learned f rom those
countries. Moreover, ACE also engages in price negotiations with manufacturers to
ascertain value based pricing. As a striking example, Dr. Heng mentioned a 50%
price discount on a new medicine for treating Hepatitis C.
Case study [9]: Value driven care approach in Singapore
Situation: In 2018, the Singaporean Ministry of Health started a program for value
driven care. This program covers inpatient treatments in all public health care
institutions for a collection of 17 conditions. These conditions include Caesarean
section, total hip replacement and ischemic stroke.
Ambition: Maximizing the value of treatment by improving the outcomes while
lowering the costs. This can be achieved by generating learnings f rom patient
outcomes and cost data.
Outcome: So far, there are indications for improving value at the national level. One
of the next steps is to include patient-reported outcome measurements (PROMs).
22www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
Singapore’s bundled payment approach is increasingly used elsewhere. Dr. Agnès
Soucat, Director for Health System Governance and Financing f rom the World Health
Organization, explained that bundling and blending different payment methods is
a prudent way forward in pursuit of improving quality. Estonia, a pioneer of bundling
payments in primary health care, has had such a system since the 2000s. In 2011, their
mix consisted of a basic allowance, capitation (predefined payments per person and
a certain period), fee for service and performance based payments. A prudent mix
of payments leverages the advantages of each kind while minimizing their caveats.
Panelists agreed that only when stakeholders join forces and organizations f rom
the public, private, and third sector realign their capabilities can a larger scale VBHC
implementation be feasible.
G. Large scale VBHC projects only succeed when roadblocks are removed
Italy is currently rolling out a successful VBHC project to encompass the whole country.
The Italian government already provides UHC to its entire population, irrespective
of any individual and social condition, and it is financed by a solidarity system. To
further improve patient experience and reduce costs, they have successfully tested
VBHC in pilots (Case study [10]: Lombardy). Now, Italy is striving to implement the
concept of VBHC universally. For this to become a reality, Dr. Giuseppe Ruocco,
Secretary General and Chief Medical Officer at the Italian Ministry of Health stressed
that VBHC needs to be fully embedded into the Italian health system and society.
For Italy, this strategy rests on an array of initiatives, among them:
The creation of a monitoring system to assess the burden of disease in two
dimensions: health implications and required investments
The measurement of value created by organizational processes and resource
consumption
Concrete measures that were taken by the government are the creation of
An e-Health National Plan, which provides health to patients outside of the
hospital setting
The new Health Pact, coordinating the reorganization of territorial health
care by newly developed collaborative organization network and
establishing pharmacies as first point of contact in the territories
The National Prevention Plan, promoting better and healthier lifestyles and
improving food safety. The National Prevention Plan also promotes health
literacy across the population and better training for health professionals.
23www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
Case study [10]: Lombardy
Situation: In 2011, Lombardy was one of the first in regions in Europe to pilot the
concept of the VBHC for treating patients with chronic diseases.
Ambition: Replacing the hospital-centric and f ragmented system with a system
that leverages collaboration between the various health actors and establishing a
single take-over fee. Furthermore, the care of chronic patients should be shifted
more to a home-based setting with the support of telemedicine.
Outcome: Successful projects tested in the initial phase were added to an initiative
called Nuove Reti Sanitarie (NRS), which aims to make those innovative approaches
a common practice in care delivery.
As described earlier in this report, India launched a similarly comprehensive and
disruptive health care project to benefit its population (Case study [2]: Ayushman
Bharat). A cornerstone of their initiative is that payment incentives are aligned with
the outcomes of treatments, e.g. pay for performance.
These nation-wide improvements in health care would have not been possible
without the willingness of legislators to allow disruption in health care. Legislators
need to introduce reforms to support VBHC models and cut down on regulatory,
institutional and legal VBHC barriers, as pinpointed by Mr. Christian Howell, the
vice president of Medtronic VBHC partnerships within the Americas region. Equally
important are ways to cultivate innovations in an ecosystem of aligned stakeholders
(Case study [11]: Partnership Lehigh Valley Health Network and Medtronic). Such
an ecosystem, Mr. Howell explained, is an environment that is mutually beneficial
because all stakeholders agree on critical components. This requires stakeholders
to agree on a strategic f ramework, governance model and operating model. In
practice this means that the provider’s and manufacturer’s teams need to speak
the same language, be aligned on what is measured, and aligned on how conflicts
are resolved. Furthermore, the legal f ramework needs to be discussed, and data
and information system inf rastructure built. Once the cornerstones of agreement
and the model have been set, the clinical intervention process can be initiated. This
intervention process can be split into three stages:
Benchmarking and design the intervention: Def ining the opportunity of
intervention, the measurement of the current state, the development
of value models and the design and building of the clinical intervention
approach
Creation of compliance mechanisms and business model
Kick-off and improvements: Taking the new clinical intervention live and
establishing continuous measurements and monitoring of the impact of the
24www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
interventions allow for data-driven amendments of the intervention
For Mr. Howell, such an ecosystem between provider, payer and producer needs to
be in place before any clinical intervention starts in a specific location, e.g. a hospital
or provider network.
What was clear f rom the examples presented at the event was that VBHC models
require an open mind for new approaches to traditional problems and the
establishment of deeper relationships within healthcare stakeholders such as the
development of innovative solutions rather than sale of products by manufacturers,
which requires better alignment with the needs of providers and payers. Mr. Haddad
f rom Johnson & Johnson summarized this effectively; “To create viable VBHC systems,
we need:
Mindset change of all stakeholders including payers, patients, providers and
governments
Trust in the system between stakeholders, they should take risk and
experiment together
Flexibility in the system; where hospitals at the community level should be
given some autonomy to enter new kind of agreements, without having to
wait for the respective health care system as a whole to change”
Mr. Allerup f rom The Capital Region of Denmark emphasized that education and
training of stakeholders are instrumental and should not be neglected. In this
context, not only the education of citizens and patients but especially that of the
workforce needs to be improved. Dr. Omar Alshanqeety f rom the Program for
Case study [11]: Partnership Lehigh Valley Health Network and Medtronic
Situation: Lehigh Valley health network and Medtronic agreed on a 5-year strategic
partnership. This partnership serves to create a mutually beneficial environment
to deliver VBHC. Indications of interest include, among others, stroke, type I
diabetes, and heart failure.
Ambition: Improving health, care and satisfaction of patients while reducing
costs. This should be achieved by the co-creation of new applications, health
care platforms, beta-testing of new ideas and the sharing of data and an aligned
governance.
Outcome: It was a win-win for both parties. The health care system benefited f rom
a reduction in care variation while complying with the care pathway, better health
outcomes and lower cost of care. Medtronic profited f rom increased revenue, new
business opportunities and new expert knowledge in health care.
25www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
Health Assurance and Purchasing in Saudi Arabia agreed and pointed out that the
new way of delivering care is different f rom the old way care has been delivered. This
necessitates a different understanding, a new mindset and new skills.
According to Ms. Colombo f rom the
OECD we need to rethink fundamentally
the training of health workers. Ms.
Colombo raised the question of how to
teach skills that allow professionals to
accelerate in a rapidly changing health
environment. They will need to learn how
to coordinate in such environments and
demonstrate an understanding of how
patients make decisions.
“We need different
medical skills for the 21st
century.”
Ms. Francesca Colombo, OECD
H. Practical learning is transferable within and across countries
Throughout the event there was a strong agreement that this was the first global
event where practical learning on global applications of VBHC models was shared.
Previously, similar conferences focused mainly on the theoretical aspects of value
driven health care models, lacked the richness of examples and were not global in
scope.
The richness and diversity of the shared examples were apparent to all participants,
and there was a shared consensus that the dialogue between stakeholders across
geographies and organizations would need to continue to drive forward in health
systems the transformation towards VBHC.
Everybody agreed that practical examples
f rom countries cannot be transferred
‘as-is’ without an understanding of the
different contexts. The stakeholders
need an understanding of which
learnings are transferrable to different
contexts, and which are context agnostic.
Such understanding is necessary
to accelerate the implementation
of VBHC both in similar and diverse
situations. One example is the VBHC
model implemented by Medtronic in
collaboration with several providers. The
first implementation took 18 months as
it required in-depth experimentation
and learning across the Medtronic
“ This is a journey that
several countries are
undertaking, and there
is a huge opportunity in
co learning.”
Dr. Omar Alshanqeety,
Program for Health Assurance
and Purchasing, Saudi Arabia
26www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
organization. Thanks to this experience, the latest project was implemented in as
little as one month.
The need for international experimentation and knowledge sharing was reinforced
by Dr. Soucat, f rom the WHO, whom suggested to establish international
experimentation, in which new implementations of VBHC would introduce
controlled changes in the models chosen in order to comparatively test hypotheses.
Such experiments already exist; however, it is hard to currently draw conclusions due
to limited comparability.
Another example of an international organization promoting international learning
on VBHC is the WEF. To foster international learning, the WEF set up the Global
Coalition for Value in Healthcare to encourage Public-Private Partnerships on
VBHC. This platform partners with various stakeholders in health to co-design and
test innovative concepts promoting patient-centered health care. Partners include
universities, research institutes, management consultancies and many more.
According to Mr. Lucas Scherdel f rom WEF, such programs have shown signs of
success. In his view, what is needed now, is a global organization that will bundle all
the efforts and takes an active lead on global learning on and implementation of
VBHC, especially around the IT and data topic.
Dr. Chintan Maru, Founder of Leapf rog to Value at The Global Development
Incubator, emphasized that low-income and middle-income Countries (LMICs)
provide important VBHC lessons for all countries. This may come as a surprise, but
can be pinpointed on three aspects that promote innovation in LMICs:
Experimentations in those geographies are not restricted by legacy health
systems. They have little inf rastructure and do not suffer f rom structural
rigidness seen in developed countries. This rigidity is driven by regulatory
constraints and a medical education focused on knowledge acquisition
and not agility
LMICs, in contrast to high-income countries, have the political will and
necessity to increase health care spending to increase access
Innovation in LMICs is often driven by the ingenuity of the countries’
poor populations. As a powerful example f rom the tech sector, he
mentioned the rapid penetration of mobile payments in Af rica, driven
by the usage of prevalent mobile phones as a convenient payment vehicle.
To move forward with VBHC, LMICs need to further cultivate their own robust
ecosystem for experimentation. There is no fixed blueprint for transferring concepts
f rom high-income countries to LMIC. To further promote such experimentations,
those realities will need the necessary risk capital, and Dr. Maru suggests that some
bilateral development funds could be allocated for such purpose.
In line with Dr. Maru’s findings, the WHO offers support for country decision makers
27www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
in a new online repository, the WHO compendium of innovative health technologies
for low-resource settings launching in 2020. This repository will provide information
on recommended interventions for therapeutic areas, type of services, age groups
and so forth. Additionally, it will have guidelines for necessary investments and give
orientation on a country specific level about costs and cost-effectiveness.
Furthermore, Ms. Herta Adam f rom the European Commission highlighted the
Commission’s support for countries in building strong, good quality and resilient
health care systems. In 2019, they expanded its efforts by entering in a partnership
program with the WHO to further improve health systems and promote UHC in over
100 countries.
I. Leveraging other sector learnings to improve the health care sector
The Event gathered valuable knowledge on the current state of the field. It translated
theoretical f rameworks into real-world examples f rom across the globe. Further
knowledge sharing events can help foster the urgently required alignment across
nations and other stakeholders.
Attendees at the Event and panelists suggested that future events may dedicate
some time to welcome experts f rom other fields, industries and sectors that have
successfully tackled similar issues facing health care today.
One f requently used comparison was re-iterated by Ms. Colombo f rom the OECD on
the aspect of safety: in the aviation industry errors and incidents are transparently
shared across the entire industry to prevent similar errors f rom happening again.
According to Ms. Colombo, health care needs a different safety culture, where we do
not panic on error, but value transparency.
Parallels to the tech industry were drawn by Mr. Howell f rom Medtronic. In the
company’s view, a platform in VBHC should work in a similar way like Uber connects
service providers with the end-users. Both stakeholders are aligned on the same
values, e.g. safe driving. This economy of trust needs to be governed:
How we share data, e.g. Uber App, no sharing of cell phone numbers
How to resolve disputes, e.g. exclusion of drivers with low ratings, reimbursement
of fees to end-user in case of dissatisfaction
How to measure outcomes, e.g. 5 star rating for drivers and passengers
Similarly, Mr. Bosch f rom the Catalan Ministry of Health, offered to the audience the
parallelism between VBHC models and Spotify. Spotify disrupted the music industry.
Traditionally, consumers used to pay by volume, paying a fee for service or buying
a CD or single song f rom iTunes. This business model pivoted to a flat rate model
in which users pay for value/outcome: value based reimbursement or flat rate for
unlimited consumption of music.
28www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
The participants of the event agreed
that it was a huge success and had the
potential to lead to real change. Prof.
Atun stated that this has been the
first time that he has seen such rich
experiences on VBHC shared, indicating
that the field is really in the transition
f rom theoretical f rameworks toward
real-world applications.
“We have an opportunity
now to capture these
experiences and share
them globally”
Prof. Rifat Atun,
Harvard University
29www.g20.org |Accelerating Transformation Towards Sustainable Health Systems Through Value Based Health Care
Figure 5: Event Audience Poll. How
important is it to have a shared definition
of what value means?
Participants and panelists re-iterated the need to openly discuss and share
their experiences and learnings. Furthermore, they urged for convergence and
interoperability, especially on the IT and data & analytics dimensions. This discussion,
they stressed, should involve all stakeholders in the health and care ecosystem.
The Event participants felt that further convergence and agreement was needed on
the nomenclature and use of terms (Figure 5):
How do we define value?
How do we define platforms?
How we define outcomes?
Prof. Atun closed his final remarks with the call for greater solidarity and sustainability
in health systems to drive economic growth. VBHC provides an opportunity to build
health systems that are better prepared for the future.
CONCLUSION and LOOKING AHEAD
www.g20.org
PBL-038
October 2019
Design and development of the Diagnosis
Related Group (DRG)
Prospective payment rates based on Diagnosis Related Groups (DRGs) have been established as
the basis of Medicare’s hospital reimbursement system. The DRGs are a patient classification
scheme which provides a means of relating the type of patients a hospital treats (i.e., its case
mix) to the costs incurred by the hospital. The design and development of the DRGs began in the
late sixties at Yale University. The initial motivation for developing the DRGs was to create an
effective framework for monitoring the quality of care and the utilization of services in a hospital
setting. The first large-scale application of the DRGs was in the late seventies in the State of New
Jersey. The New Jersey State Department of Health used DRGs as the basis of a prospective
payment system in which hospitals were reimbursed a fixed DRG specific amount for each
patient treated. In 1982, the Tax Equity and Fiscal Responsibility Act modified the Section 223
Medicare hospital reimbursement limits to include a case mix adjustment based on DRGs. In
1983 Congress amended the Social Security Act to include a national DRG-based hospital
prospective payment system for all Medicare patients.
The evolution of the DRGs and their use as the basic unit of payment in Medicare’s hospital
reimbursement system represents a recognition of the fundamental role which a hospital’s case
mix plays in determining its costs. In the past, hospital characteristics such as teaching status and
bed size have been used to attempt to explain the substantial cost differences which exist across
hospitals. However, such characteristics failed to account adequately for the cost impact of a
hospital’s case mix. Individual hospitals have often attempted to justify higher cost by
contending that they treated a more “complex” mix of patients; the usual contention being that
the patients treated were “sicker.” Although there has been a consensus in the hospital industry
that a more complex case mix results in higher costs, the concept of case mix complexity had
historically lacked a precise definition. The development of the DRGs provided the first
operational means of defining and measuring a hospital’s case mix complexity.
The concept of case mix complexity
The concept of case mix complexity initially appears very straightforward. However, clinicians,
administrators and regulators have often attached different meanings to the concept of case mix
complexity depending on their backgrounds and purposes. The term case mix complexity has
been used to refer to an interrelated but distinct set of patient attributes which include severity
of illness, prognosis, treatment difficulty, need for intervention and resource intensity. Each of
these concepts has very precise meaning which describes a particular aspect of a hospital’s case
mix.
• Severity of illness. Refers to the relative levels of loss of function and mortality that may be
experienced by patients with a particular disease.
2
• Prognosis. Refers to the probable outcome of an illness including the likelihood of
improvement or deterioration in the severity of the illness, the likelihood for recurrence and
the probable life span.
• Treatment difficulty. Refers to the patient management problems which a particular illness
presents to the health care provider. Such management problems are associated with
illnesses without a clear pattern of symptoms, illnesses requiring sophisticated and
technically difficult procedures and illnesses requiring close monitoring and supervision.
• Need for intervention. Relates to the consequences in terms of severity of illness that lack of
immediate or continuing care would produce.
• Resource intensity. Refers to the relative volume and types of diagnostic, therapeutic and
bed services used in the management of a particular illness.
When clinicians use the notion of case mix complexity, they mean that the patients treated have
a greater severity of illness, present greater treatment difficulty, have poorer prognoses and
have a greater need for intervention. Thus, from a clinical perspective case mix complexity refers
to the condition of the patients treated and the treatment difficulty associated with providing
care. On the other hand, administrators and regulators usually use the concept of case mix
complexity to indicate that the patients treated require more resources which results in a higher
cost of providing care. Thus, from an administrative or regulatory perspective case mix
complexity refers to the resource intensity demands that patients place on an institution. While
the two interpretations of case mix complexity are often closely related, they can be very
different for certain kinds of patients. For example, while terminal cancer patients are very
severely ill and have a poor prognosis, they require few hospital resources beyond basic nursing
care.
In the past, there has sometimes been confusion regarding the use and interpretation of the
DRGs because the aspect of case mix complexity measured by the DRGs has not been clearly
understood. The purpose of the DRGs is to relate a hospital’s case mix to the resource demands
and associated costs experienced by the hospital. Therefore, a hospital having a more complex
case mix from a DRG perspective means that the hospital treats patients who require more
hospital resources but not necessarily that the hospital treats patients having a greater severity
of illness, a greater treatment difficulty, a poorer prognosis or a greater need for intervention.
Patient classification
Given that the purpose of the DRGs is to relate a hospital’s case mix to its resource intensity, it
was necessary to develop an operational means of determining the types of patients treated and
relating each patient type to the resources they consumed. While all patients are unique, groups
of patients have demographic, diagnostic and therapeutic attributes in common that determine
their level of resource intensity. By developing clinically similar groups of patients with similar
resource intensity, patients can be aggregated into meaningful patient classes. Moreover, if
these patient classes covered the entire range of patients seen in an inpatient setting, then
collectively they would constitute a patient classification scheme that would provide a means of
Design and development of the Diagnosis Related Group (DRG)
3
establishing and measuring hospital case mix complexity. The DRGs were therefore developed as
a patient classification scheme consisting of classes of patients who were similar clinically and in
terms of their consumption of hospital resources.
During the process of developing the DRG patient classification scheme, several alternative
approaches to constructing the patient classes were investigated. Initially, a normative approach
was used which involved having clinicians define the DRGs using the patient characteristics
which they felt were important for determining resource intensity. There was a tendency for
their definitions to include an extensive set of specifications, requiring information which might
not always be collected through a hospital’s medical information system. If the entire range of
patients were classified in this manner, it would ultimately lead to thousands of DRGs, most of
which described patients seen infrequently in a typical hospital. It, therefore, became evident
that the process of DRG definition would be facilitated if data from acute care hospitals could be
examined to determine the general characteristics and relative frequency of different patient
types. In addition, statistical algorithms applied to this data would be useful to suggest ways of
forming DRGs that were similar in terms of resource intensity. However, it was also discovered
that statistical algorithms applied to historical data in the absence of clinical input would not
yield a satisfactory set of DRGs. The DRGs resulting from such a statistical approach, while similar
in terms of resource intensity, would often contain patients with a diverse set of characteristics
which could not be interpreted from a clinical perspective. Thus, it became apparent that the
development of the DRG patient classification scheme required that physician judgment,
statistical analysis and verification with historical data be merged into a single process. It was
necessary to be able to examine large amounts of historical data with statistical algorithms
available for suggesting alternative ways of forming DRGs but to do so in such a way that
physicians could review the results at each step to insure that the DRGs formed were clinically
coherent.
Basic characteristics of the DRG patient classification
scheme
Given the limitations of previous patient classification schemes and the experience of
attempting to develop DRGs with physician panels and statistical analysis, it was concluded that
in order for the DRG patient classification scheme to be practical and meaningful it should have
the following characteristics:
1. The patient characteristics used in the definition of the DRGs should be limited to
information routinely collected on hospital abstract systems.
2. There should be a manageable number of DRGs which encompass all patients seen on an
in-patient basis.
3. Each DRG should contain patients with a similar pattern of resource intensity.
4. Each DRG should contain patients who are similar from a clinical perspective (i.e., each class
should be clinically coherent).
Design and development of the Diagnosis Related Group (DRG)
4
Restricting the patient characteristics used in the definition of the DRGs to those readily
available insured that the DRGs could be extensively applied. Currently, the patient information
routinely collected includes age, principal diagnosis, secondary diagnoses and the surgical
procedures performed. Creating DRGs based on information that is only collected in a few
settings or on information which is difficult to collect or measure would have resulted in a
patient classification scheme which could not be applied uniformly across hospitals. That is not
to say that information beyond that currently collected might not be useful for defining the
DRGs. As additional information becomes routinely available it must be evaluated to determine
if it might result in improvements in the ability to classify patients.
Limiting the number of DRGs to manageable numbers (i.e., hundreds of patient classes, not
thousands) insures that for most of the DRGs, a typical hospital will have enough experience to
allow meaningful comparative analysis to be performed. If there were only a few patients in
each DRG, it would be difficult to detect patterns in case mix complexity and cost performance
and to communicate the results to the physician staff.
The resource intensity of the patients in each DRG must be similar in order to establish a
relationship between the case mix of a hospital and the resources it consumes. Similar resource
intensity means that the resources used are relatively consistent across the patients in each
DRG. However, some variation in resource intensity will remain among the patients in each DRG.
In other words, the definition of the DRG will not be so specific that every patient is identical,
but the level of variation is known and predictable. Thus, while the precise resource intensity of
a particular patient cannot be predicted by knowing to which DRG he belongs, the average
pattern of resource intensity of a group of patients in a DRG can be accurately predicted.
Since one of the major applications of the DRGs is as a means of communicating with the
physician community, the patients in each DRG must be similar from a clinical perspective. In
other words, the definition of each DRG must be clinically coherent. The concept of clinical
coherence requires that the patient characteristics included in the definition of each DRG relate
to a common organ system or etiology and that a specific medical specialty should typically
provide care to the patients in the DRG. For example, patients who are admitted for a D&C or a
Tonsillectomy are similar in terms of most measures of resource intensity such as length of stay,
preoperative stay, operating room time and use of ancillary services. However, different organ
systems and different medical specialties are involved. Thus, the requirement that the DRGs be
clinically coherent precludes the possibility of these types of patients being in the same DRG.
A common organ system or etiology and a common clinical specialty is a necessary but not
sufficient requirement for a DRG to be clinically coherent. In addition, all available patient
characteristics which medically would be expected to consistently affect resource intensity
should be included in the definition of the DRG. Furthermore, a DRG should not be based on
patient characteristics which medically would not be expected to consistently affect resource
intensity. For example, patients with appendicitis may or may not have peritonitis. Although
these patients are the same from an organ system, etiology and medical specialist perspective,
the DRG definitions must form separate patient classes, since the presence of peritonitis would
be expected to consistently increase the resource intensity of the appendicitis patients. On the
other hand, sets of unrelated surgical procedures cannot be used to define a DRG since there
Design and development of the Diagnosis Related Group (DRG)
5
would not be a medical rationale to substantiate that the resource intensity would be expected
to be similar.
The definition of clinical coherence is, of course, dependent on the purpose for the formation of
the DRG classification. For the DRGs, the definition of clinical coherence relates to the medical
rationale for differences in resource intensity. If, on the other hand, the purpose of the DRGs
related to mortality, the patient characteristics which were clinically coherent and, therefore,
included in the DRG definitions might be different. Finally, it should be noted that the
requirement that the DRGs be clinically coherent caused more patient classes to be formed than
would be necessary for explaining resource intensity alone.
Formation of the DRGs
The process of forming the DRGs was begun by dividing all possible principal diagnoses into 23
mutually exclusive principal diagnosis areas referred to as Major Diagnostic Categories (MDC).
Two new MDCs were created in the eighth version of the DRGs. The 25 MDCs are listed in table
1.
The MDCs were formed by physician panels as the first step toward insuring that the DRGs
would be clinically coherent. The diagnoses in each MDC correspond to a single organ system or
etiology and in general are associated with a particular medical specialty. Thus, in order to
maintain the requirement of clinical coherence, no final DRG could contain patients in different
MDCs. In general, each MDC was constructed to correspond to a major organ system (e.g.,
Respiratory System, Circulatory System, Digestive System) rather than etiology (e.g.,
malignancies, infectious diseases). This approach was used since clinical care is generally
organized in accordance with the organ system affected, and not the etiology. Thus, diseases
involving both a particular organ system and a particular etiology (e.g., malignant neoplasm of
the kidney) were assigned to the MDC corresponding to the organ system involved. However,
not all diseases or disorders could be assigned to an organ system-based MDC and a number of
residual MDCs were created (e.g., Systemic Infectious Diseases, Myeloproliferative Diseases and
Poorly Differentiated Neoplasms). For example, the infectious diseases food poisoning and
Shigella dysenteriae are assigned to the Digestive System MDC while pulmonary tuberculosis is
assigned to the Respiratory System MDC. On the other hand, infectious diseases such as miliary
tuberculosis and septicemia which usually involve the entire body are assigned to the Systemic
Infectious Disease MDC.
Once the MDCs were defined each MDC was evaluated to identify those additional patient
characteristics which would have a consistent effect on the consumption of hospital resources.
Since the presence of a surgical procedure which required the use of the operating room would
have a significant effect on the type of hospital resources (e.g., operating room, recovery room,
anesthesia) used by a patient, most MDCs were initially divided into medical and surgical groups.
The medical-surgical distinction is also useful in further defining the clinical specialty involved.
Patients were considered surgical if they had a procedure performed which would require the
use of the operating room. Since the patient data generally available does not precisely indicate
whether a patient was taken to the operating room, surgical patients were identified based on
Design and development of the Diagnosis Related Group (DRG)
6
the procedures which were performed. Physician panels classified every possible procedure
code based on whether the procedure would in most hospitals be performed in the operating
room.
Table 1. Major Diagnostic Categories
MDC Description
1 Diseases and Disorders of the Nervous System
2 Diseases and Disorders of the Eye
3 Diseases and Disorders of the Ear, Nose, Mouth and Throat
4 Diseases and Disorders of the Respiratory System
5 Diseases and Disorders of the Circulatory System
6 Diseases and Disorders of the Digestive System
7 Diseases and Disorders of the Hepatobiliary System and Pancreas
8 Diseases and Disorders of the Musculoskeletal System and
Connective Tissue
9 Diseases and Disorders of the Skin, Subcutaneous Tissue and Breast
10 Endocrine, Nutritional and Metabolic Diseases and Disorders
11 Diseases and Disorders of the Kidney and Urinary Tract
12 Diseases and Disorders of the Male Reproductive System
13 Diseases and Disorders of the Female Reproductive System
14 Pregnancy, Childbirth and the Puerperium
15 Newborns and Other Neonates with Conditions Originating in the
Perinatal Period
16 Diseases and Disorders of the Blood and Blood Forming Organs and
Immunological Disorders
17 Myeloproliferative Diseases and Disorders, and Poorly Differentiated
Neoplasm
18 Infectious and Parasitic Diseases (Systemic or Unspecified Sites)
19 Mental Diseases and Disorders
20 Alcohol/Drug Use and Alcohol/Drug Induced Organic Mental
Disorders
21 Injuries, Poisonings and Toxic Effects of Drugs
22 Burns
Design and development of the Diagnosis Related Group (DRG)
7
MDC Description
23 Factors Influencing Health Status and Other Contacts with Health
Services
24 Multiple Significant Trauma
25 Human Immunodeficiency Virus Infections
1.
Thus, closed heart valvotomies, cerebral meninges biopsies and total cholecystectomies would
be expected to require the operating room while thoracentesis, bronchoscopy and skin sutures
would not. If a patient had any procedure performed which was expected to require the
operating room that patient would be classified as a surgical patient. A complete list of all the
procedures expected to require the operating room is contained in Appendix E.
Once each MDC was divided into medical and surgical categories, then, in general, the surgical
patients were further defined based on the precise surgical procedure performed while the
medical patients were further defined based on the precise principal diagnosis for which they
were admitted to the hospital. The general structure of a typical MDC is shown by the tree
diagram found at the end of this chapter (figure 1). In general, specific groups of surgical
procedures were defined to distinguish surgical patients according to the extent of the surgical
procedure performed. For example, the procedure classes defined for the Endocrine, Nutritional
and Metabolic MDC are amputations, procedures for obesity, skin grafts and wound
debridements, adrenal and pituitary procedures, parathyroid procedures, thyroid procedures,
thyroglossal procedures and other procedures relating to Endocrine, Nutritional or Metabolic
diseases.
Since a patient can have multiple procedures related to their principal diagnosis during a
particular hospital stay, and a patient can be assigned to only one surgical class, the surgical
classes in each MDC were defined in a hierarchical order. Patients with multiple procedures
would be assigned to the surgical class highest in the hierarchy.
Thus, if a patient received both an extraction of endometrium (D&C in ICD-9-CM) and a resection
of uterus (hysterectomy in ICD-9-CM), the patient would be assigned to the hysterectomy
surgical class. It should be noted that as a result of the surgical hierarchy the ordering of the
surgical procedures on the patient abstract has no influence on the assignment of the surgical
class and DRG. Appendix D lists the surgical hierarchy for each MDC.
In general, specific groups of principal diagnoses were defined for medical patients. Usually the
medical classes in each MDC would include a class for neoplasms, symptoms and specific
conditions relating to the organ system involved. For example, the medical classes for the
Respiratory System MDC are ventilator support, pulmonary embolism, infections and
inflammations, neoplasms, chest trauma, pleural effusion, pulmonary edema and respiratory
failure, chronic obstructive pulmonary disease, simple pneumonia and pleurisy, interstitial lung
disease, pneumothorax, bronchitis and asthma, respiratory signs and symptoms and other
respiratory diagnoses.
Design and development of the Diagnosis Related Group (DRG)
8
In each MDC there is usually a medical and a surgical class referred to as “other medical
diseases” and “other surgical procedures,” respectively. The “other” medical and surgical classes
are not as precisely defined from a clinical perspective. The other classes would include
diagnoses or procedures which were infrequently encountered or not well defined clinically. For
example, the “other” medical class for the Respiratory System MDC would contain the diagnoses
“other somatoform disorders” and “congenital malformation of the respiratory system,” while
the “other” surgical class for the female reproductive MDC would contain the surgical
procedures “excision of liver” (liver biopsy in ICD-9-CM) and “inspection of peritoneal cavity”
(exploratory laparotomy in ICD-9-CM).
The “other” surgical category contains surgical procedures which, while infrequent, could still
reasonably be expected to be performed for a patient in the particular MDC. There are,
however, also patients who receive surgical procedures which are completely unrelated to the
MDC to which the patient was assigned. An example of such a patient would be a patient with a
principal diagnosis of pneumonia whose only surgical procedure is a destruction of prostate
(transurethral prostatectomy in ICD-9-CM). Such patients are assigned to a surgical class referred
to as “unrelated operating room procedures.” These patients are ultimately never assigned to a
well-defined DRG.
The process of defining the surgical and medical classes in an MDC required that each surgical or
medical class be based on some organizing principle. Examples of organizing principles would be
anatomy, surgical approach, diagnostic approach, pathology, etiology or treatment process. In
order for a diagnosis or surgical procedure to be assigned to a particular class, it would be
required to correspond to the particular organizing principle for that class. For example, in the
Urinary System MDC a surgical group was formed for all patients with a procedure on the
urethra (i.e., organizing principle based on anatomy). This surgical group was then further
divided based on whether the procedure performed was transurethral (i.e., organizing principle
based on surgical approach).
Figure 1 displays the basic structure of the DRG. Until the eighth version, the first step in the
determination of the DRG had been the assignment of the appropriate MDC based on the
principal diagnosis. The eighth version of the DRGs contained the first departure from the use of
principal diagnosis as the initial variable in DRG assignment, when the initial step in DRG
assignment was based on procedure (PRE MDC). If a patient has a heart transplant or implant of
heart assist system, ECMO or tracheostomy, liver transplant and/or intestinal transplant, bone
marrow transplant, lung transplant simultaneous pancreas/kidney transplant, or pancreas
transplant, then the patient is assigned to these DRGs independent of the MDC of the principal
diagnosis. Heart, intestinal, liver, bone marrow, lung, pancreas/kidney and pancreas transplants
are very resource intensive and can be performed for diagnoses in many different MDCs.
Tracheostomies are performed primarily for patients on long term ventilator support and
therefore such patients are very resource intensive. The eighth version also created two new
MDCs for patients with multiple trauma (MDC 24) and patients with an HIV infection (MDC 25).
Assignment to MDC 24 and 25 is based on both principal and secondary diagnoses. An
assignment to MDC 24 is based on the presence of two or more significant traumas in different
body systems (e.g. a fractured skull and a fractured femur). Assignment to MDC 25 is based on a
principal diagnosis of an HIV infection or a principal diagnosis of an HIV related complication
Design and development of the Diagnosis Related Group (DRG)
9
combined with a secondary diagnosis of an HIV infection (e.g. principal diagnosis of
pneumocystosis and a secondary diagnosis of an HIV infection).
Once the medical and surgical classes for an MDC were formed, each class of patients was
evaluated to determine if complications, comorbidities, the patient’s age or discharge status
consistently affected the consumption of hospital resources. Physician panels classified each
diagnosis code based on whether the diagnosis, when present as a secondary condition, would
be considered a substantial complication or comorbidity. A substantial complication or
comorbidity was defined as a condition, that because of its presence with a specific principal
diagnosis would cause an increase in length of stay by at least one day in at least 75 percent of
the patients. For example, sarcoidosis of lung, chronic obstructive pulmonary disease and
pneumococcal pneumonia are considered substantial complications or comorbidities for certain
diseases, while nontoxic diffuse goiter and essential hypertension are not. Each medical and
surgical class within an MDC was tested to determine if the presence of any substantial
comorbidities or complications would consistently affect the consumption of hospital resources.
For example, the presence of complications or comorbidities was not significant for patients
receiving a median nerve release (carpal tunnel release in ICD-9-CM) but was very significant for
patients with arrhythmia and conduction disorders. The same basic list of complications and
comorbidities are used across most DRGs. However, depending on the principal diagnosis of the
patient, some diagnoses in the basic list of complications and comorbidities may be excluded if
they are closely related to the principal diagnosis. For example, urinary retention is a
complication or comorbidity for a patient admitted for congestive heart failure but not for a
patient admitted for enlarged prostate. In addition, in some cases such as newborns or acute
myocardial infarction patients, special complications and comorbidity definitions were used in
defining the DRGs.
The final variable used in the definition of the DRGs was the patient discharge status. Separate
DRGs were formed for newborns if the patients were transferred to another acute care facility.
In addition, separate DRGs were formed for patients with alcoholism or drug abuse who left
against medical advice and for acute myocardial infarction patients and newborns who died.
For versions 2-24 of the DRGs, the further subdivisions of some medical and surgical DRGs was
primarily based on the presence or absence of a CC or pediatric age (0-17). For example, in DRG
version 24 there were 115 pairs of DRGs subdivided based on the presence or absence of a CC
and 43 pediatric DRGs (age 0-17). Beginning with version 25 the use of CCs and patient age was
completely revised. The revisions were so extensive that the version 25 DRGs were renamed to
be the Medicare Severity DRGs (MS-DRGs).
Except for new diagnosis codes that were added to ICD-9-CM after FY1984 (e.g., HIV), the CC list
of diagnoses used in the DRGs remained virtually identical to the original CC list used in FY1984.
As a result of the changes that occurred in hospitals during the first 22 years of PPS, the CC list
had lost much of its power to discriminate hospital resource use. Better coding of secondary
diagnoses, stricter criteria for extended hospital stays, increased availability of post acute care
services and the shift to outpatient care resulted in most patients (nearly 80 percent) admitted
to hospitals having a CC. Therefore, in version 25 (MS-DRGs) the diagnoses comprising the CC list
were completely redefined. The revised CC list is primarily comprised of significant acute
disease, acute exacerbations of significant chronic diseases, advanced or end stage chronic
Design and development of the Diagnosis Related Group (DRG)
10
diseases and chronic diseases associated with extensive debility. In general, most chronic
diseases were not included on the revised CC list. For a patient with a chronic disease, a
significant acute manifestation of the chronic disease was required to be present and coded for
the patient to be assigned a CC. The revision of the CC list reduced the number of Medicare
patients with a CC from approximately 80 percent to 40 percent.
In addition, to the revision of the CC list, each CC was also categorized as a major CC or a CC (i.e.,
non major CC) based on relative resource use. Approximately, 12 percent of all diagnoses codes
were classified as a major CC, 24 percent as a CC and 64 percent as a non CC. Diagnoses closely
associated with mortality (ventricular fibrillation, cardiac arrest, shock and respiratory arrest)
were assigned as a major CC if the patient lived but as a non CC if the patient died.
The major CC, CC and non CC categorization was used to subdivide the surgical and medical
DRGs into up to three levels with a patient being assigned to the most extreme level (e.g., a
patient with an MCC and a CC is assigned to the MCC level). Before subdividing the medical and
surgical DRGs into CC levels all the pediatric age distinctions were removed from the DRGs. To
create the MS-DRGs, individual DRGs were subdivided into three, two or one level depending on
the CC impact on resources used for that patient. The two way subdivision either created a
separate level for just the major CC patients or a separate level for the non CC patients. The CC
levels relate to the relative severity of illness of the patient. In the MS-DRG version 25, 152 DRGs
had 3 CC levels, 107 DRGs had two CC levels and 76 DRGs had no CC levels resulting in 745
MS-DRGs which is a net increase of 207 DRGs over the 538 in version 24. The following table
provides the MS-DRG version 37.0 subdivisions:
Table 2. MS-DRG v37.0 subdivisions
Base MS-DRGs Split Type Total MS-DRGs
159 3-way 477
42 2-way MCC/CC and no CC 84
66 2-way MCC and CC/no CC 132
68 No split 68
335 Total Base [blank] 761 Total MS-DRGs
2.
In MS-DRG version 37.0 there are 72,184 diagnoses and 77,559 procedures.
The Deficit Reduction Act of 2005 (P.L.109-171) requires CMS to eliminate any increase in
payment due to the occurrence of selected post admission complications, known as Hospital
Acquired Conditions (HACs). HACs are harmful events (e.g. accidental laceration during a
procedure) or negative outcomes (e.g. decubitus ulcer) that result from the processes of care
and treatment rather than from a natural progression of underlying illness. Under the Medicare
inpatient prospective payment system, the occurrence of an HAC can result in a higher payment
because the presence of the HAC diagnosis may cause the patient to be assigned to a
higher-paying MS-DRG, in effect financially rewarding poor quality care.
Design and development of the Diagnosis Related Group (DRG)
11
When IPPS was implemented the standard claim form did not contain a specification of whether
a secondary diagnosis was present on admission (POA). The Deficit Reduction Act requires
hospitals to report a POA indicator for all diagnoses beginning in fiscal year 2008. The reporting
of the POA indicator allows complications that occur post admission to be identified. The Deficit
Reduction Act requires that the post admission complications selected as HACs be (1) high cost,
high volume, or both; (2) be a CC or Major CC in MS-DRGs and (3) be reasonably preventable
through the application of evidence-based guidelines. For Fiscal Year 2009, CMS designated 12
conditions as HACs:
1. Foreign object retained after surgery
2. Air embolism
3. Blood incompatibility
4. Stage III and IV pressure ulcers
5. Falls and trauma
6. Catheter-associated urinary tract infection (UTI)
7. Vascular catheter-associated infection
8. Surgical site infection – Mediastinitis following Coronary Artery Bypass Surgery (CABG)
9. Manifestations of poor glycemic control
10. Deep vein thrombosis (DVT) /pulmonary embolism (PE) following total knee replacement
or hip replacement
11. Surgical site infection following bariatric surgery
12. Surgical site infection following certain orthopedic procedures of spine, shoulder or
elbow
For Fiscal Year 2013, two additional HACs were added:
13. Surgical site infection following cardiac device procedures
14. Iatrogenic pneumothorax with venous catheterization.
If an HAC diagnosis is present at admission, it will continue to be classified as a CC or major CC
and allowed to affect the MS-DRG assignment. However, if the HAC diagnosis is not present at
admission, it will no longer be classified as a CC or major CC and will not affect MS-DRG
assignment. The exclusion of an HAC diagnosis from MS-DRG assignment does not necessarily
mean the MS-DRG will change. Some MS-DRGs are not differentiated by the presence of a CC or
Major CC. For such MS-DRGs the exclusion of an HAC diagnosis will have no impact on MS-DRG
assignment. Further, if in addition to an HAC diagnosis there are non-HAC diagnoses present that
are a CC or major CC, the exclusion of the HAC diagnosis may not change the MS-DRG. Beginning
in Fiscal Year 2009, HAC diagnoses are excluded from MS-DRG assignment.
The actual process of forming the DRGs was highly iterative, involving a combination of
statistical results from test data with clinical judgment. At any point during the definition of the
DRGs there would often be several patient characteristics which appeared important for
understanding the impact on hospital resources. The selection of the patient characteristics to
Design and development of the Diagnosis Related Group (DRG)
12
be used and the order in which they would be used was a complex task with many factors
examined and weighed simultaneously. A complete list of the MS-DRGs is contained in Appendix
A in the MS-DRG definitions manual.
There are several MS-DRGs which contain patients whose medical record abstracts contain
clinically inconsistent or invalid information. For example, there are MS-DRGs for patients for
whom all their operating room procedures performed are unrelated to the major diagnostic
category of the patient’s principal diagnosis. Typically, these are patients admitted for a
particular diagnosis requiring no surgery, who develop a complication unrelated to the principal
diagnosis and have an operating room procedure performed for the complication or have a
diagnostic procedure performed for another concurrent diagnosis. The unrelated operating
room procedures have been divided into two groups based on hospital resource use: extensive
and non-extensive. For example, a patient with a principal diagnosis of congestive heart failure
who develops acute cholecystitis and whose only procedure is a resection of gallbladder
(cholecystectomy in ICD-9-CM) will be assigned to the extensive unrelated procedure MS-DRG
since a cholecystectomy is considered an extensive procedure. However, if a patient has a
principal diagnosis of arrhythmia and has a diagnostic excision of breast (breast biopsy in
ICD-9-CM) discovered while in the hospital, the patient will be assigned to the non- extensive
unrelated MS-DRG since the biopsy is considered a non-extensive procedure. The complete
definition of unrelated operating room procedures is contained in Appendix F.
When a principal diagnosis is coded which, although it is a valid ICD-10-CM code, is not precise
enough to allow the patient to be assigned to a clinically coherent MS-DRG the patient is
assigned to a diagnosis invalid as principal diagnosis MS-DRG. For example, ICD-10-CM code
O0930 is an unspecified complication of pregnancy with the episode of care unspecified. Thus,
this diagnosis code does not indicate the type of complication nor whether the episode of care
was antepartum, postpartum or for delivery. Since the MS-DRG definitions assign patients to
different sets of MS-DRGs depending on whether the episode of care was antepartum,
postpartum or for delivery, a patient with a principal diagnosis of O0930 must be assigned to the
diagnosis invalid as principal diagnosis MS-DRG.
It should be noted that patients with a principal diagnosis not typically considered a reason for
hospitalization such as Z413 (ear piercing) are not assigned to the diagnosis invalid as principal
diagnosis MS-DRG but are assigned a MS-DRG in the MDC most related to the diagnosis.
Patients are assigned to an ungroupable MS-DRG if certain types of medical records errors which
may affect MS-DRG assignment are present. Patients with an invalid or non-existent ICD-10-CM
code as principal diagnosis will be assigned to the ungroupable MS-DRG. Patients will also be
assigned to the ungroupable MS-DRG if their sex, or discharge status is both invalid and
necessary for MS-DRG assignment. For example, if a patient has a non-numeric discharge status
and has a principal diagnosis of an acute myocardial infarction, the patient will be assigned to
the ungroupable MS-DRG since patients with acute myocardial infarction will be assigned to
different MS-DRGs depending on whether their discharge status is alive or died. On the other
hand, if the same patient had a principal diagnosis of hypertension, the assignment would not be
to the ungroupable MS-DRG since discharge status is not used in the determination of the
MS-DRG for hypertensive patients.
Design and development of the Diagnosis Related Group (DRG)
13
The DRGs were originally developed at the Yale University School of Organization and
Management during the 1970’s under contract to the Centers for Medicare and Medicaid
Services (formerly Health Care Financing Administration). The second version and all subsequent
versions of the DRG definitions have been updated by 3M Health Information Systems under
contract with CMS. All versions of the DRGs, since the inception of the Medicare Prospective
Payment System, are summarized in the following table.
Table 3. Grouper versions
Grouper version Effective time period
MS-DRG 37.0 10/01/2019 – 09/30/2020
MS-DRG 36.0 10/01/2018 – 09/30/2019
MS-DRG 35.0 10/01/2017 – 09/30/2018
MS-DRG 34.0 10/01/2016 – 09/30/2017
MS-DRG 33.0 10/01/2015 – 09/30/2016
MS-DRG 32.0 10/01/2014 – 09/30/2015
MS-DRG 31.0 10/01/2013 – 09/30/2014
MS-DRG 30.0 10/01/2012 – 09/30/2013
MS-DRG 29.0 10/01/2011 – 09/30/2012
MS-DRG 28.0 10/01/2010 – 09/30/2011
MS-DRG 27.0 10/01/2009 – 09/30/2010
MS-DRG 26.0 10/01/2008 – 09/30/2009
MS-DRG 25.0 10/01/2007 – 09/30/2008
CMS 24.0 10/01/2006 – 09/30/2007
CMS 23.0 10/01/2005 – 09/30/2006
CMS 22.0 10/01/2004 – 09/30/2005
CMS 21.0 10/01/2003 – 09/30/2004
CMS 20.0 10/01/2002 – 09/30/2003
CMS 19.0 10/01/2001 – 09/30/2002
CMS 18.0 10/01/2000 – 09/30/2001
CMS 17.0 10/01/1999 – 09/30/2000
CMS 16.0 10/01/1998 – 09/30/1999
CMS 15.0 10/01/1997 – 09/30/1998
Design and development of the Diagnosis Related Group (DRG)
14
Grouper version Effective time period
CMS 14.0 10/01/1996 – 09/30/1997
CMS 13.0 10/01/1995 – 09/30/1996
CMS 12.0 10/01/1994 – 09/30/1995
CMS 11.0 10/01/1993 – 09/30/1994
CMS 10.0 10/01/1992 – 09/30/1993
CMS 9.0 10/01/1991 – 09/30/1992
CMS 8.0 10/01/1990 – 09/30/1991
CMS 7.0 10/01/1989 – 09/30/1990
CMS 6.0 10/01/1988 – 09/30/1989
CMS 5.0 10/01/1987 – 09/30/1988
CMS 4.0 10/01/1986 – 09/30/1987
CMS 3.0 05/01/1986 – 09/30/1986
CMS 2.0 10/01/1983 – 04/30/1986
3.
Summary
The DRGs, as they are now defined, form a manageable, clinically coherent set of patient classes
that relate a hospital’s case mix to the resource demands and associated costs experienced by
the hospital. DRGs are defined based on the principal diagnosis, secondary diagnoses, surgical
procedures, age, sex and discharge status of the patients treated. Through DRGs, hospitals can
gain an understanding of the patients being treated, the costs incurred and within reasonable
limits, the services expected to be required. The classification of patients into DRGs is a
constantly evolving process. As coding schemes change, as more comprehensive data is
collected or as medical technology or practice changes, the DRG definitions will be reviewed and
revised.
DRG versions 2.0–32.0 were defined using the ICD-9-CM codeset. MS-DRG v37.0 was
implemented using the ICD-10-CM/PCS codeset effective October 1, 2019.
Design and development of the Diagnosis Related Group (DRG)
15
Type of Surgery
O.R.
Procedure
Principal Diagnosis
Major Surgery
Major
Diagnostic
Category
Minor Surgery
Other Surgery
Surgery
Unrelated to
Principal
Diagnosis
Specific
Conditions
Relating to the
Organ System
Symptoms
Other
Neoplasm
No
Yes
Figure 1: Typical DRG structure for a Major Diagnostic Category
2/2/22, 5:34 PM What Is Value-Based Healthcare?
https://catalyst.nejm.org/doi/full/10.1056/CAT.17.0558 1/6
B R I E F A R T I C L E
What Is Value-Based Healthcare?
Explore the definition, benefits, and examples of value-based healthcare. How does value-based healthcare
translate to new delivery models?
NEJM Catalyst
January �, ����
This article appeared in NEJM Catalyst prior to the launch of the NEJM Catalyst Innovations in Care Delivery journal. Learn
more.
Value-based healthcare is a healthcare delivery model in which providers, including hospitals and physicians, are paid
based on patient health outcomes. Under value-based care agreements, providers are rewarded for helping patients
improve their health, reduce the effects and incidence of chronic disease, and live healthier lives in an evidence-based
way.
Value-based care differs from a fee-for-service or capitated approach, in which providers are paid based on the amount
of healthcare services they deliver. The “value” in value-based healthcare is derived from measuring health outcomes
against the cost of delivering the outcomes.
What Are the Benefits of Value-Based Healthcare Delivery?
The benefits of a value-based healthcare system extend to patients, providers, payers, suppliers, and society as a whole.
Value-Based Healthcare Benefits: Lower Healthcare Costs, Higher Patient Satisfaction, Reduced Risks
Figure � .
Free Virtual Web Event | AI & Machine Learning for Health Care Delivery Register today > (X)
https://catalyst.nejm.org/topic/catalyst-article-type/brief-article
https://catalyst.nejm.org/about/journal/
https://events.catalyst.nejm.org/all-events?promo=OCFBARM1&query=swb&utm_source=catsite&utm_medium=sbar&utm_campaign=2022_events
https://catalyst.nejm.org/
2/2/22, 5:34 PM What Is Value-Based Healthcare?
https://catalyst.nejm.org/doi/full/10.1056/CAT.17.0558 2/6
1. Patients spend less money to achieve better health. Managing a chronic disease or condition like cancer,
diabetes, high blood pressure, COPD, or obesity can be costly and time-consuming for patients. Value-based care
models focus on helping patients recover from illnesses and injuries more quickly and avoid chronic disease in the
first place. As a result, patients face fewer doctor’s visits, medical tests, and procedures, and they spend less money
on prescription medication as both near-term and long-term health improve.
2. Providers achieve efficiencies and greater patient satisfaction. While providers may need to spend more
time on new, prevention-based patient services, they will spend less time on chronic disease management. Quality
and patient engagement measures increase when the focus is on value instead of volume. In addition, providers are
not placed at the financial risk that comes with capitated payment systems. Even for-profit providers, who can
generate higher value per episode of care, stand to be rewarded under a value-based care model.
3. Payers control costs and reduce risk. Risk is reduced by spreading it across a larger patient population. A
healthier population with fewer claims translates into less drain on payers’ premium pools and investments. Value-
based payment also allows payers to increase efficiency by bundling payments that cover the patient’s full care
cycle, or for chronic conditions, covering periods of a year or more.
4. Suppliers align prices with patient outcomes. Suppliers benefit from being able to align their products and
services with positive patient outcomes and reduced cost, an important selling proposition as national health
expenditures on prescription drugs continue to rise. Many healthcare industry stakeholders are calling for
manufacturers to tie the prices of drugs to their actual value to patients, a process that is likely to become easier with
the growth of individualized therapies.
5. Society becomes healthier while reducing overall healthcare spending. Less money is spent helping
people manage chronic diseases and costly hospitalizations and medical emergencies. In a country where
healthcare expenditures account for nearly 18% of Gross Domestic Product �GDP�, value-based care has the
promise to significantly reduce overall costs spent on healthcare.
How Does Value-Based Healthcare Translate to New Delivery Models?
The proliferation of value-based healthcare is changing the way physicians and hospitals provide care. New healthcare
delivery models stress a team-oriented approach to patient care and sharing of patient data so that care is coordinated
and outcomes can be measured easily. Two examples are reviewed here.
Value-Based Care Models: Medical Homes
In value-based healthcare models, medical care does not exist in silos. Instead, primary, specialty, and acute care are
integrated, often in a delivery model called a patient-centered medical home �PCMH�. A medical home isn’t a physical
location. Instead, it’s a coordinated approach to patient care, led by a patient’s primary physician who directs a patient’s
total clinical care team.
PCMHs rely on the sharing of electronic medical records �EMRs� among all providers on the coordinated care team.
The goal of EMRs is to put crucial patient information at each provider’s fingertips, allowing individual providers to see
results of tests and procedures performed by other clinicians on the team. This data sharing has the potential to reduce
redundant care and associated costs.
Value-Based Care Models: Accountable Care Organizations
Accountable care organizations �ACOs� were originally designed by the Centers for Medicare & Medicaid Services
�CMS� to provide high-quality medical care to Medicare patients. In an ACO, doctors, hospitals, and other healthcare
providers work as a networked team to deliver the best possible coordinated care at the lowest possible cost. Each
member of the team shares both risk and reward, with incentives to improve access to care, quality of care, and patient
health outcomes while reducing costs. This approach differs from fee-for-service healthcare, in which individual
providers are incentivized to order more tests and procedures and manage more patients in order to get paid more,
regardless of patient outcomes.
https://catalyst.nejm.org/why-value-in-health-care-is-the-target/
https://catalyst.nejm.org/new-marketplace-insights-report-value-based-payment-gains-traction-amid-active-m-and-a-environment/
https://catalyst.nejm.org/topic/bundled-payments/
https://catalyst.nejm.org/big-tent-value-based-care-room-big-pharma/
https://catalyst.nejm.org/is-value-based-drug-pricing-compatible-with-pharma-innovation/
https://catalyst.nejm.org/caring-for-older-adults-in-a-value-based-model/
https://www.cms.gov/research-statistics-data-and-systems/statistics-trends-and-reports/nationalhealthexpenddata/nhe-fact-sheet.html
https://catalyst.nejm.org/what-is-patient-centered-care/
https://catalyst.nejm.org/topic/patient-centered-care/
https://catalyst.nejm.org/videos/primary-care-providers-whole-care-coordinators/
https://catalyst.nejm.org/topic/accountable-care-organizations-aco/
2/2/22, 5:34 PM What Is Value-Based Healthcare?
https://catalyst.nejm.org/doi/full/10.1056/CAT.17.0558 3/6
F E B R U A RY �, ����
Like PCMHs, ACOs are patient-centered organizations in which the patient and providers are true partners in care
decisions. Also like PCMHs, ACOs stress coordination and data sharing among team members to help achieve these
goals among their entire patient population. Clinical and claims data are also shared with payers to demonstrate
improvements in outcomes such as hospital readmissions, adverse events, patient engagement, and population health.
Hospital Value-Based Purchasing
Under CMS’s Hospital Value-Based Purchasing Program �VBP�, acute care hospitals receive adjusted payments based
on the quality of care they deliver. According to the CMS website, the program encourages hospitals to improve the
quality and safety of acute inpatient care for all patients by:
Eliminating or reducing adverse events �healthcare errors resulting in patient harm�
Adopting evidence-based care standards and protocols that make the best outcomes for the most patients
Changing hospital processes to create better patient care experiences
Increasing care transparency for consumers
Recognizing hospitals that give high-quality care at a lower cost to Medicare
CMS is expected to continue to refine its VBP measurements, making it important for hospitals to continuously
improve their clinical outcomes so they can simultaneously improve reimbursement and their reputation among
healthcare consumers.
What Is the Future of Value-Based Healthcare?
Moving from a fee-for-service to a fee-for-value system will take time, and the transition has proved more difficult than
expected. As the healthcare landscape continues to evolve and providers increase their adoption of value-based care
models, they may see short-term financial hits before longer-term costs decline. However, the transition from fee-for-
service to fee-for-value has been embraced as the best method for lowering healthcare costs while increasing quality
care and helping people lead healthier lives.
NEJM Catalyst
Topics
Value-Based Care
Accountable Care Organizations (ACO)
P H YS I C I A N J O B S
Silver Spring, MarylandChiefs / Directors / Dept. Heads
Director, Center for Tobacco Products
New YorkChiefs / Directors / Dept. Heads
Northwell Health – Senior Director, Division of Hematopathology
https://catalyst.nejm.org/hospital-readmissions-reduction-program-hrrp/
https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/Value-Based-Programs/HVBP/Hospital-Value-Based-Purchasing.html
https://catalyst.nejm.org/turning-value-based-health-care-into-a-real-business-model/
https://catalyst.nejm.org/topic/topics/value-based-care
https://catalyst.nejm.org/topic/topics/accountable-care-organizations
https://www.nejmcareercenter.org/?query=fjw
https://www.nejmcareercenter.org/job/477025/director-center-for-tobacco-products/?query=fjwp&rid=327757
https://www.nejmcareercenter.org/job/455922/northwell-health-senior-director-division-of-hematopathology/?query=fjwp&rid=3059
2/2/22, 5:34 PM What Is Value-Based Healthcare?
https://catalyst.nejm.org/doi/full/10.1056/CAT.17.0558 4/6
Sleepy Hollow, New YorkChiefs / Directors / Dept. Heads
Associate Medical Director, Emergency Medicine – Phelps Hospital – Sleepy Hollow, New York
Salt Lake City, UtahChiefs / Directors / Dept. Heads
Medical Director, Neonatology Level II NICU/Special Care Nursery
Sayre, PennsylvaniaChiefs / Directors / Dept. Heads
Chair, Department of Medicine
New YorkChiefs / Directors / Dept. Heads
Director of Heart Failure – Westchester, Northern Westchester Hospital and Phelps Hospital
JAN ��
NOV ��
A R T I C L E VOL. � NO. �
ACO and Social Service Organization Partnerships: Payment, Challenges, and Perspectives
D. Velasquez and J.F. Figueroa
The intersection between health care and social services is increasingly relevant to care delivery, but the contractual
mechanisms in place have yet to produce evidence of the clinical or operational improvements that are needed to
support and expand the model.
A R T I C L E VOL. � NO. ��
Value-Based Health Care for Children with Congenital Heart Diseases: A Feasibility Study from a Low–
Middle Income Country
B.S. Hasan and Others
While the goals of a patient-focused, outcomes-driven system of care may have universal appeal, initiatives at a
hospital in Pakistan reveal insights about local obstacles and opportunities.
A R T I C L E VOL. � NO. ��
The Clinical, Operational, and Financial Components of a Successful Bundled Payment Program for
Lower Extremity Total Joint Replacement
C. MacLean and Others
Bundled payments for joint replacement can yield significant savings for payers and incentive payments for providers,
but effective execution needs detailed cost and quality information, as well as flexibility and cooperation from
M O S T R E C E N T I N A LT E R N AT I V E PAY M E N T M O D E L S
https://www.nejmcareercenter.org/job/237899/associate-medical-director-emergency-medicine-phelps-hospital-sleepy-hollow-new-york/?query=fjwp&rid=3059
https://www.nejmcareercenter.org/job/477126/medical-director-neonatology-level-ii-nicu-special-care-nursery/?query=fjwf&rid=751
https://www.nejmcareercenter.org/job/473330/chair-department-of-medicine/?query=fjwp&rid=899
https://www.nejmcareercenter.org/job/392534/director-of-heart-failure-westchester-northern-westchester-hospital-and-phelps-hospital/?query=fjwp&rid=3059
https://catalyst.nejm.org/topic/catalyst-article-type/article
https://catalyst.nejm.org/toc/catalyst/3/1
https://catalyst.nejm.org/doi/full/10.1056/CAT.21.0319
https://catalyst.nejm.org/topic/catalyst-article-type/article
https://catalyst.nejm.org/toc/catalyst/2/12
https://catalyst.nejm.org/doi/full/10.1056/CAT.21.0258
https://catalyst.nejm.org/topic/catalyst-article-type/article
https://catalyst.nejm.org/toc/catalyst/2/10
https://catalyst.nejm.org/doi/full/10.1056/CAT.21.0240
https://catalyst.nejm.org/topic/broad-topics/alternative-payment-models
2/2/22, 5:34 PM What Is Value-Based Healthcare?
https://catalyst.nejm.org/doi/full/10.1056/CAT.17.0558 5/6
SEP ��
admission through the postdischarge recovery period. New York’s Hospital for Special Surgery describes its successful
process.
J O U R N A L
E V E N T S
I N S I G H T S C O U N C I L
J O U R N A L
In Depth
Case Studies
Articles
Insights Reports
Insights Interviews
Survey Snapshots
Current Issue
Issue Index
Editorial Board
Editorial Policies
Author Center
About the Journal
C ATA LY S T E X T R A S
Commentaries
Conversations
Talks
Clips
Topics
A B O U T
About NEJM Catalyst
Permissions & Reprints
Sponsorship
Help
Contact Us
Newsletter
Careers
S U B S C R I P T I O N S
Subscribe
Renew
Create Account
Manage Account
Pay Bill
Subscription Agent Information
Resources for Institutions
Institution Administration Center
J O U R N A L S
The New England Journal of Medicine
NEJM Catalyst Innovations in Care Delivery
NEJM Evidence
https://catalyst.nejm.org/toc/catalyst/current
https://catalyst.nejm.org/events
https://catalyst.nejm.org/insights-council
https://catalyst.nejm.org/topic/catalyst-article-type/in-depth
https://catalyst.nejm.org/topic/catalyst-article-type/case-study
https://catalyst.nejm.org/topic/catalyst-article-type/article
https://catalyst.nejm.org/topic/catalyst-article-type/insights-report
https://catalyst.nejm.org/topic/catalyst-article-type/insights-interview
https://catalyst.nejm.org/topic/catalyst-article-type/survey-snapshot
https://catalyst.nejm.org/toc/catalyst/current
https://catalyst.nejm.org/toc/catalyst/current#past-issues
https://catalyst.nejm.org/about/editors-and-publishers
https://catalyst.nejm.org/about/editorial-policies
https://catalyst.nejm.org/author-center
https://catalyst.nejm.org/about/journal
https://catalyst.nejm.org/topic/catalyst-article-type/commentary
https://catalyst.nejm.org/topic/catalyst-article-type/conversation
https://catalyst.nejm.org/topic/catalyst-article-type/talk
https://catalyst.nejm.org/topic/catalyst-article-type/clip
https://catalyst.nejm.org/topics
https://catalyst.nejm.org/about
https://catalyst.nejm.org/about/permissions
https://catalyst.nejm.org/about/sponsorship
https://catalyst.nejm.org/about/help
https://catalyst.nejm.org/about/contact-us
https://www.nejmcareercenter.org/
https://catalyst.nejm.org/action/storeProxy?action=footer-subscribe&promo=OCILSCCC
https://catalyst.nejm.org/action/storeProxy?action=renew&product=catalyst&promo=OCIQRTA1
https://myaccount.nejm.org/
https://catalyst.nejm.org/action/storeProxy?action=paybill&product=catalyst&promo=BCIQBTA1
https://catalyst.nejm.org/pb-assets/images/Catalyst-Subscription-Agent-Info-1635876168653
https://libraryhub.nejm.org/site-license-info/
https://catalyst.nejm.org/action/institutionMultiIpShow
https://www.facebook.com/NEJMCatalyst/
https://www.twitter.com/NEJMCatalyst/
https://www.instagram.com/nejm.catalyst/
https://www.youtube.com/user/NEJMvideo
https://www.linkedin.com/showcase/nejm-catalyst
https://www.nejm.org/
https://catalyst.nejm.org/
https://evidence.nejm.org/
2/2/22, 5:34 PM What Is Value-Based Healthcare?
https://catalyst.nejm.org/doi/full/10.1056/CAT.17.0558 6/6
NEJM Catalyst is a product of NEJM Group, a division of the Massachusetts Medical Society.
Copyright © ���� Massachusetts Medical Society. All rights reserved.
Terms of Use Privacy Policy
https://www.nejmgroup.org/legal/terms-of-use.htm
https://www.nejmgroup.org/legal/privacy-policy.htm
https://www.nejmgroup.org/
Health Technology, Quality, Law,
and Ethics
Chapter 15
The New Public Health. http://dx.doi.org/10.1016/B978-0-12-415766-8.00015-X
Copyright © 2014 Elsevier Inc. All rights reserved.
Learning Objectives
Upon completion of this chapter, the student should be
able to:
1. Describe responsibility for and methods of assessing and
regulating technological developments in health care;
2. Describe methods of health facility accreditation and
peer review;
3. Describe the concept of total quality management;
4. Identify and discuss ethical and legal issues in national
health systems;
5. Apply ethical considerations to health issues in his or
her home setting.
INTRODUCTION
Management of a production or a service system requires
attention to the quality of personnel as much as to the system
in which they work. Their motivation and sense of participa-
tion, the scientific and technological level of the program,
and the legal and ethical standards of individual providers
and of the system as a whole, are all important to the quality
of care provided and equity of health status achieved.
Quality is the result of input and process, and is measured
by outcome or performance indicators as well as perception
of the service by the patients, the staff, and the community
as a whole. Input refers to the institutional and financial
resources for education, human resources, supplies, medica-
tions, vaccines, diagnostic capacity, and services available.
Process refers to the use of those resources, including peer
group expectations of professionalism. Outcomes generally
include measures of morbidity, mortality, and functional sta-
tus of the patient and the population. Defining and measuring
achievements of national health objectives and targets, the
methods of financing services, and the efficiency of organi-
zation help to determine quality. Training, supply, and dis-
tribution of health personnel are all determinants of access
to and quality of care. Continuous and adequate availability
of essential preventive, diagnostic, and treatment services,
as well as accountability and internal methods of promoting
standards, are all elements of the quality of a health service
771
for the individual, the population as a whole, and groups
within the population with special needs.
The content and standards of service are assessed
through organized review by professional peers within
an institution, and from outside. Peer review within an
institution and external evaluation by accreditation or
governmental inspection, based on cumulative evidence
and the recognized current “state of the art”, contribute
to accountability and improved quality of care. Continu-
ous quality improvement (CQI) among health care teams
and organizations includes regular practice assessments,
evidence gathering, remediation, and re-evaluation, which
will be discussed later in this chapter. The perception of
the services by the community, along with the knowledge,
attitudes, beliefs, and practices of health, are all vital to
improvement of health status.
Health-related technology is also in a continuing state
of change. Systematic review and absorption of new scien-
tific knowledge, technology, and innovations are essential
to promote and renew health care methods. Public health
serves in a regulatory role to assure high-quality care to the
individual and the community. New technology, whether in
the form of diagnostic procedures, new drugs, devices, or
vaccines, or new types of health personnel, requires evalu-
ation for effectiveness and appropriateness to the system.
Technology assessment also involves epidemiological and
economic aspects of effectiveness. Failure to continuously
monitor developments and to assimilate those that are demon-
strably successful is an ethical and management failure which
tragically costs many millions of lives from preventable dis-
eases yearly, such as in delayed adoption of well-proven vac-
cines or tobacco restriction legislation. This is due to political
failure even more than professional weakness, and constitutes
one of the saddest ethical dilemmas of public health: failure
to convince policy makers of the prime importance of health
promotion and disease prevention in the health sector.
Ethics and law in public health reflect the values of a
society. They inevitably evolve as they face dramatic social,
economic, demographic, and political changes; new health
challenges; and new technological and scientific possibili-
ties for improving health. Ethics are the foundation of the
value systems of a society and thus of its health concepts.
772
Biblical sources articulated values of the Ten Command-
ments, Sanctity of Human Life, Improve the World, along
with the Hippocratic Oath of physicians to “do good and do
no harm”. Modern definitions of public health and bioethics
emerged from lessons learned from the horrors of eugenics
and genocide in the twentieth century with humanistic pre-
cepts of “Universal Human Rights” and “Health for All” in
the recent era (see Chapters 1 and 2).
The law is both permissive and restrictive. It sets the basic
responsibilities, powers, and limitations of public health prac-
tice, with legislation and court decisions. Innovations in the
technology of medical care and public health are powerful
forces contributing to increased longevity, quality of life, and
economic growth, but they also bring challenges to imple-
mentation impeded by additional costs of the health system
and slow adaptation in countries with the greatest need.
These are challenges to national and international political,
The New Public Health
organizational, and economic systems to address health with
the full potential for saving lives. Determining standards of
“good practice” is a continuing process with the rapid devel-
opment of new knowledge, technology, and experience.
The law is a dynamic process involving old and new
legislation, court decisions, and new issues not previously
faced, often following rather than anticipating public health
issues. Public health has had both positive and negative ethi-
cal experiences and continues to face new issues with chang-
ing population needs, technology, science, and economics.
INNOVATION, REGULATION, AND
QUALITY CONTROL
Health care technology has advanced with an increasing
stream of innovation since the seventeenth-century epidemi-
ological discoveries of Lind on scurvy (1747) and smallpox
TABLE 15.1 Health Care Innovations from the Seventeenth to the Twenty-First Centuries
Period Selected Highlights of Scientific, Technological, and Organizational Innovations in Health
17th century Biological basis of disease (Descartes), circulation of blood (Harvey), microscope (Leeuwenhoek)
18th century Thermometer, lime juice supplements (Lind, 1756), vaccination (Jenner, 1796), surgical anatomy (Hunter), clinical
sciences (Sydenham)
19th century Miasma theory vs germ theory; inventions of stethoscope (1816), blood transfusion (1818), anesthesia (1842),
hypodermic syringe (1852), ophthalmoscope (1851), laryngoscope (1855), pasteurization of wine, beer, milk
(1860s), cholera vaccine (1879), X-ray (1895), blood pressure cuff (1896); sanitation, municipal health depart-
ments, chlorination and filtration of community water supplies, antisepsis, Braille printing, hygiene in obstetrics,
nursing, microscopic pathology, pathological chemistry, microbiology, vaccines, X-ray, national health insurance,
syringes, well-child care, aspirin (1899), Bismarkian social insurance (1881)
1900–1930 Electrocardiogram (1901), Flexner report on medical education, salvarsan, insulin (1922), blood groups, vitamins,
conquest of yellow fever, vitamin B, vaccine for diphtheria (1923), tetanus vaccine (1924), electroencephalogram
(1924), iron lung respirator (1927), Social Security Act (1935), cost–benefit analysis, food fortification (iodized salt,
flour with vitamin B complex), improved work safety
1931–1945 Mandatory fortification of milk, salt, and flour in USA (1941), Pap test (1942), penicillin (1928), streptomycin,
randomized clinical trials, antimalarial drugs, vector controls, dialysis machine (1945)
1946–1960 Contact lens (1948), DNA double helix (1953), heart–lung bypass machine (1953), ultrasound (1955), cardiac
pacemaker (1958), Salk polio vaccine (1955), kidney transplant (1959), advances in vaccines, antihypertensives,
psychotropic drugs, cancer chemotherapy, prepaid group practice, UK National Health Service (1948), Medicare
in Canada (1946–1971)
1961–1980 Oral polio vaccine (Sabin), hip replacement (1962), oral rehydration therapy, measles vaccine (1964), coronary
bypass (1964), Medicare, Medicaid (1965), mammography (1965), portable defibrillator (1965), measles–mumps–
rubella vaccine, cost-effectiveness analysis, open heart surgery, pacemakers, organ transplantation, computed
tomography (CT), eradication of smallpox (1972), health maintenance organizations (HMOs), diagnosis-related
groups (DRGs), district health systems
1981–2000 Health promotion (1987), magnetic resonance imaging (MRI), positron emission tomography (PET), endoscopic
surgery, Helicobacter pylori and chronic peptic ulcer disease (1982), managed care, Haemophilus influenzae b
(Hib) vaccine, statins (1987), poliomyelitis eradication campaign (1982), local eradication of beta-thalassemia,
pandemic of HIV (1981 onward), AZT antiretroviral approved (1987), robotic surgery (2000)
2001–2013 Millennium Development Goals (MDGS 2000) with substantial progress achieved, managing emergencies of mass
terrorism and natural disasters, new vaccines (HPV), managing epidemics of measles and influenza, new diag-
nostic technologies, flour fortification to prevent birth defects, HIV still deadly but effective treatment and control
measures, new treatments for hepatitis C, robotic surgery, nanotechnology, scientific advances with great potential
benefit, Affordable Care Act (2010), Accountable Care Organizations
Source: Adapted from Health United States 2009. Special Feature: Medical technology. Introduction and timeline. 2009. Available at: http://www.ncbi.nlm.
nih.gov/books/NBK44737/#specialfeature.sec1 [Accessed 15 December 2012]. See Historical Markers in Chapter 1.
http://www.ncbi.nlm.nih.gov/books/NBK44737/#specialfeature.sec1
http://www.ncbi.nlm.nih.gov/books/NBK44737/#specialfeature.sec1
Chapter 15 Health Technology, Quality, Law, and Ethics
vaccination by Jenner (1796), to the dramatic innovations of
the end of the twentieth century (Table 15.1). The pace of
innovation is rapid, creating the need for regulation, quality
control, and technology assessment.
National governments are responsible for assuring that
pharmaceuticals, biological products, food, and the environ-
ment are regulated to protect the public. In some countries,
these responsibilities are divided among ministries of trade,
industry, commerce, health, and environment. In a federal
system of government, there may be a division of responsi-
bility among federal, state, and local government, but with
the national government often providing national standards
and leadership in this area.
Government regulation and control are meant to protect
the public health. The US Food and Drug Administration
(FDA) is responsible for enforcing the Food, Drug and Cos-
metic Act, the Fair Packaging and Labeling Act, sections of
the Public Health Services Act relating to biological prod-
ucts for control of communicable diseases, and the Radiation
Control for Health and Safety Act. The FDA is a Division
of the Department of Health and Human Services (DHHS).
State governments have the authority to supervise pharma-
cies and their products, which may be marketed across dif-
ferent states. All national governments have departments
responsible for conducting supervision of food, drugs, and
medical devices, often relying on international standards.
Drugs and devices include all drugs, diagnostic prod-
ucts, blood and its derivatives, biologicals, veterinary
medicines, and medicated premixed animal products. All
manufacturers and distributors are required by law to reg-
ister these products with the national authority in order to
be allowed to market or import them. All countries need to
govern the food, drugs, vaccines, and cosmetics regulated
for production, importation, marketing, and use within
their jurisdiction. Organizations within each government
must be responsible for assuring the consumer that foods
are pure (unadulterated) and wholesome, safe to eat, and
produced under sanitary conditions; that drugs and medical
devices are safe and effective for their intended uses; that
cosmetics are safe and made from appropriate ingredients;
and that labeling is truthful, informative, and not deceptive.
National authorities such as the FDA, under legisla-
tion and regulations, govern both domestic and imported
products. They establish and enforce standards, or adopt
external agency standards as a “gold standard”, meaning
that products meet high standards of safety and efficacy.
The FDA also monitors and inspects contents manufactur-
ing standards under good manufacturing practices (GMPs),
which includes regular accreditation of a manufacturer’s
facilities, staffing, planning, and monitoring capacity. Test-
ing of products is carried out to assess safety, potency, and
toxicity using accepted reference laboratory procedures as
published in the compendium Official Methods of Analysis
of the Association of Official Analytical Chemists.
773
When federal, state, or local investigators, sometimes
known as consumer safety officers, detect through labora-
tory monitoring or observe conditions that may result in a
public health hazard, and violation of food and drug laws
and regulations, they issue a written report to the manufac-
turers with recommendations for correcting the conditions.
In more blatant cases, the authorities may issue urgent recall
or seizure orders for products in violation of standards con-
stituting a danger to public health, such as contaminated
products, lead-painted children’s toys, or contaminated
foods causing foodborne disease outbreaks, which occur
not infrequently in imported and domestically produced
foods in the USA. The Los Angeles County Department of
Health inspects restaurants regularly and places a promi-
nent placard in the window giving a grade A, B, or C to
the restaurant for sanitation and safety. Those given D rat-
ings may be closed until specified faults are eliminated, or
a restaurant may be closed permanently. State governments
require restaurants to list calorie and salt content of foods
on their menus as part of the public health efforts to reduce
obesity.
Supervision of food standards may also fail, as occurred
in Israel in 2004 when total absence of vitamin B1 in a soy-
based baby formula imported from Germany resulted in
three deaths and permanent brain damage to other infants
due to severe beriberi. This episode led to criminal charges
in 2008 of negligence resulting in death against the own-
ers of the company that imported or produced the foods and
staff members of the Ministry of Health. Animal foods in
2007 and infant milk products imported from China in 2008
were found to be contaminated with melamine, which was
meant to mimic protein content but was toxic in combination
with other chemicals used. The infant formula caused seri-
ous illness in some 300,000 Chinese babies and six deaths.
The FDA and its counterparts in each country are
responsible for regulation of:
l food – foodborne illness, nutritional content, labeling,
dietary supplements
l drugs – prescription drugs and generics, over-the-counter
products
l medical devices – pacemakers, stents, contact lenses,
hearing aids
l biologics – vaccines, blood products
l animal feed and drugs – for livestock, pets
l cosmetics – safety, labeling
l radiation-emitting products – cell phones, lasers, micro-
waves
l combination products.
New drugs and biological products for human use are
required to pass rigorous review before approval for mar-
keting is granted. Applications are submitted by the manu-
facturer or sponsor with acceptable scientific data including
test results to evaluate the safety and effectiveness of the
774
product for the conditions under which it is being offered.
All manufacturers of drugs are required to be registered
with the FDA and to meet its requirements for each drug
produced and marketed, including the reporting of adverse
reactions and labeling criteria. Manufacturers are required
to operate in conformity with current GMPs, which include
stringent control over manufacturing processes, personnel
training, computerized operations, and testing of finished
products. The FDA publishes guidelines to help manufac-
turers to familiarize themselves with current standards. The
United States Pharmacopoeia, National Formulary, and
WHO Model Formulary 2008 are the official listings of
approved products.
Medical devices are also regulated by the FDA. Thou-
sands of products for health care purposes require premarket
approval, ranging from basic articles such as thermometers,
tongue depressors, and intrauterine devices (IUDs), to more
complex devices such as cardiac monitors, pacemakers,
breast implants, and kidney dialysis machines. These prod-
ucts are subject to controls of GMPs, labeling, registration
of the manufacturer, and performance standards.
Monitoring for efficacy and potential hazards has
been strengthened since the 1970s as a result of find-
ings of long-term carcinogenic and mutagenic effects of
estrogens, and toxic effects of chloramphenicol on bone
marrow. The drug thalidomide, widely used as an antin-
auseant and sleeping pill for pregnant women in Europe,
Canada, and Australia in the 1960s, was not approved by
the US FDA. This drug was found to cause large numbers
of serious birth deformities leading to its being banned in
most countries. Controls of blood and blood products have
been strengthened since the transmission of human immu-
nodeficiency virus (HIV), hepatitis B, and hepatitis C by
contaminated blood products in the 1980s. The responsi-
bility of this regulatory function is well illustrated by the
1995 criminal conviction of several senior health officials
in France for failing to stop the use of blood products con-
taminated with HIV in the mid-1980s. Concern regarding
possible carcinogenic effects of silicone breast implants
led to legal action and greater controls of all implantable
products. A balance between safety and well-regulated
approval of new products requires a highly professional
and motivated regulatory agency, well-developed proce-
dures, and well-trained staff.
The concepts of standardization of GMPs for pharma-
ceutical products and written protocols for good medi-
cal practice or good public health practice are accepted
norms based on best available evidence of current scientific
knowledge and experience. Recommended immunization
schedules, water quality, ambient air standards, food fortifi-
cation, and screening programs for early stages of diabetes
are examples of accepted practice that have become rec-
ommended standards of public health practice, paralleling
qualitative measures developed in clinical care.
The New Public Health
APPROPRIATE HEALTH TECHNOLOGY
The concept of intermediate technology pioneered by Dr
Ernst Schumacher in the 1960s proposed the development
of simple and inexpensive technology for developing coun-
tries such as India to promote local economic development.
Environmentally sustainable development and sources of
energy, energy conservation, and reductions in toxic and
harmful emissions are encouraged. In recent years ideas
have included small loan systems for rural entrepreneurs in
developing countries, and the use of simple cell phones for
communication, farm produce marketing, cash transfers in
remote areas without banking services, and many others.
Now called appropriate technology, this topic has gained
adherence in the health field in the search for low-cost and
simple techniques for preventing and managing common
illnesses.
Appropriate technology is defined by the World Health
Organization (WHO) as the level of medical technology
needed to improve health conditions in keeping with the
epidemiological, demographic, and financial situation of
each country. All countries have limited resources and
so must select strategies of health care and appropriate
technology to use those resources effectively to achieve
health benefits. Improved water pumps, solar energy, rain-
water collection and water reservoirs, sanitary latrines,
fly traps, insecticide-impregnated bed nets, biogas from
animal waste, improved home cooking stoves, and many
other simple devices can make enormous differences in
local sustainable agriculture, economic growth, and living
conditions. Cell phones are now used to monitor health
conditions such as hypertension, diabetes control, weight
and body mass index, and other non-communicable con-
ditions, and to transmit imaging from remote areas to
specialists in medical centers who can provide test read-
ings online. Simple, affordable, portable information tech-
nology can effectively support public health programs,
even in resource-poor environments.
The topics discussed in the growing literature and meet-
ings of the International Society of Technology Assessment
in Health Care represent the dynamic field of technology
assessment. The issues range from economic evaluation
of pharmaceuticals to modeling approaches, measures of
quality of life, technology dissemination and impact, and
outcomes measurement. The range of issues also includes
finance and health insurance, health care in developing
countries, informatics, telemedicine, technologies for the
disabled, screening, and cost-effectiveness. Evaluations
in the scrutiny of both high- and low-technology services
based on a combination of clinical, epidemiological, and
economic factors are necessary. As health costs rise, dis-
abling conditions increase and populations age, medical
innovation proceeds at a rapid rate, and both client and
community expectations in health care continually rise.
Chapter 15 Health Technology, Quality, Law, and Ethics
In developing countries, the training and supervision of
traditional birth attendants (TBAs) for prenatal preparation
and normal deliveries are important ways to reduce maternal
mortality in rural areas, as discussed elsewhere, and an impor-
tant Millennium Development Goal (MDG) which will not
be met by 2015. Community health workers (CHWs) in well
supervised and supported programs are essential to provide
preventive care to underserved rural poor populations with a
defined package of services that can be tailored to meet spe-
cific local needs, such as immunization, child growth moni-
toring, nutrition counseling, and malaria and TB control.
A major example of appropriate technology has
been the WHO initiatives to promote national drug for-
mularies (NDFs) as a consensus list of essential drugs
that are sufficient for the major health needs of a coun-
try, eliminating unnecessary duplication and combined
products on the commercial market. The WHO calls on
all member states to ensure the availability and rational
use of drugs and vaccines, and supports states wishing
to select an essential list of drugs for economic procure-
ment. Assistance with drug regulatory agencies, legisla-
tion, quality control, information, supply, and training is
offered to help the member countries. Standard reference
laboratories, the International Pharmacopoeia, and the
WHO Drug Bulletin promote international standards and
provide guidance to member states. The WHO Model List
of Essential Drugs is a valuable tool to improve quality
and cost management in national health systems.
Cochlear implants are now routinely used for chil-
dren with congenital or other loss of hearing, as well as in
elderly people. In August 2013, a new cell phone applica-
tion was announced which photographs the eye and can be
used to diagnose cataracts, macular degeneration (AMD),
and other eye pathology, for interpretation by experts far
away and to enable arrangements to be made for appropri-
ate intervention to prevent blindness, which is common in
developing countries. Other applications allow for moni-
toring of blood sugar of diabetics, hypertension, exercise,
dietary management, and other aspects of health. In the
same month, a camera, computer, and auditory device
allowing blind people to “see and read” was demonstrated.
The costs of such devices are initially high but will fall with
advances in computing and other technical developments.
In both developing and industrialized countries major
causes of death include cardiovascular diseases (coronary
heart disease and stroke), along with respiratory diseases,
cancer and injuries, all amenable to preventive and cura-
tive medical care. The key preventive measures for these
are: healthful diet, reduced obesity, smoking cessation,
exercise and physical fitness, hypertension management,
aspirin, immunizations and other low-cost and highly effec-
tive medications such as statins. These are all low-cost self-
care measures that can be promoted by local, state, and
national governments, private advocacy organizations, and
775
individuals in their families and communities. The princi-
ples of low technology, cost-effectiveness, and sound health
policy converge in addressing these fundamental issues.
Priority Interventions in Low- and Medium-
Income Countries
Disease control priorities for low- and medium-income coun-
tries are an important challenge for public health. Selection has
often been based on individual initiatives due to strong advo-
cacy in international organizations by donor countries, orga-
nizations, or individuals. In 1993, two landmark documents
attempted to apply a logical system to such considerations:
one was the World Bank’s now classic World Development
Report: Investing in Health and the other was Disease Control
Priorities in Developing Countries. The World Development
Report defined cost-effective clinical and public health cluster
programs essential to improving health outcomes for low- and
middle-income developing countries. The programs focus on
those diseases that contribute heavily to the burden of disease
and are amenable to relatively inexpensive interventions. The
report defined interventions most able to reduce the burden
of disease in low- and middle-income countries using clinical
and public health interventions, as summarized in Table 15.2.
The 1993 World Development Report provided policy
makers and public health practitioners with a concept
and tools for assessing cost-effectiveness of available
interventions for the major health problems in the devel-
oping world. It also provided useful measuring tools in
the form of disability-adjusted life years (DALYs) to
calculate the burden of disease and the cost-effective-
ness of interventions to address them. This World Bank
report addressed clinical interventions that would reduce
DALYs lost by 24 percent in low-income countries and
8 percent in middle-income countries, including treat-
ment of TB, with directly observed therapy, short course
(DOTS); integrated management of the sick child; pre-
natal and delivery care; family planning; treatment of
STIs; and limited care for pain, infections, and trauma
as resources permit. It also addressed public health inter-
ventions, which would reduce DALYs lost by 8.2 percent
in low-income countries and 4 percent in middle-income
countries, expanded immunization with vitamin A sup-
plements; tobacco and alcohol control; AIDS prevention;
and school health including deworming. Together, the
total reductions would be 32 percent for low-income and
12 percent for middle-income countries (Table 15.2).
These estimates have been refined by numerous stud-
ies conducted over the subsequent two decades. The sec-
ond edition of the Disease Control Priorities in Developing
Countries (2006) incorporates important changes in the
technologies available. The concept of viewing priorities
with an economic epidemiology model is still applicable,
and has increased in importance (Box 15.1).
776
As the MDGs are reaching their endpoint in 2015,
follow-up global health targets will need to recognize the
vital importance of non-communicable diseases (NCDs)
in developing countries. The global consensus on MDGs,
set out by the United Nations (UN) in 2001, indicates
progress in the epidemiological understanding of realities
TABLE 15.2 World Bank Model for Priority Cost-
Effective Health Interventions in Low- and Middle-
Income Developing Countries
Burden of Disease Averted
(%)
Service Type Low-Income
Countries
Middle-Income
Countries
Public health interventions
EPI-plus immunization (DPT,
polio, measles, BCG, hepatitis
B, yellow fever, vitamin A)
6.0 1.0
Other public health programs
(family planning, health, and
nutrition education)
NA NA
Tobacco and alcohol control
programs
0.1 0.3
AIDS prevention program 2.0 2.3
School health program
(including deworming)
0.1 0.4
Subtotal (public health) 8.2 4.0
Clinical interventions
Treatment of tuberculosis
(short course)
1.0 1.0
Integrated management of the
sick child
14.0 4.0
Prenatal and delivery care 4.0 –
Family planning 3.0 1.0
Treatment of STIs 1.0 1.0
Limited care: pain, trauma,
infection plus as resources
permit
1.0 1.0
Subtotal (clinical care) 24.0 8.0
Total 32.2 12.0
Note: Low-income = < US$350 gross national product (GNP) per capita;
middle income = > US$2500 GNP per capita. Cost per immunized
child = US$14.60 (US$0.50 per capita) and US$27.20 (US$0.80 per
capita) in low- and middle-income countries, respectively.
DPT = diphtheria–pertussis–tetanus; BCG = bacille Calmette–Guérin;
AIDS = acquired immunodeficiency syndrome; STI = sexually transmitted
infection.
Note: The World Development Report was an innovative basis for follow-
up work, as reported in Jamison DT, Breman JG, Measham AR, Alleyne
G, Claeson M, Evans DB, et al., editors. Disease control priorities in
developing countries. 2nd ed. Disease Control Priorities Project. Wash-
ington, DC: World Bank; 2006.
Source: Adapted from World Bank. World development report. Investing
in health. New York: Oxford University Press; 1993.
The New Public Health
in low-income countries and the need for consensus over
common targets. Since then, attention has been directed
towards the epidemiological shift to NCDs, which are the
most common causes of death in low- and medium-income
countries. Thus there is a double burden of infectious, nutri-
tion, maternal, and child priorities, alongside the NCDs.
The increasing adoption of vaccines such as Haemophilus
influenzae type b (Hib) and rotavirus alongside the standard
BOX 15.1 Disease Control Priorities in Developing
Countries
l Average life expectancy in low- and middle-income
countries increased dramatically since the 1960s, while
cross-country health inequalities decreased.
l Improved health has contributed significantly to eco-
nomic welfare since the 1960s.
l Five critical challenges face developing countries (and
the world) at the beginning of the twenty-first century:
– Rapid demographic growth
– HIV pandemic improved but still rampant
– Persistent malaria, TB, diarrhea, pneumonia
– Micronutrient malnutrition for mothers and infants
– NCDs
– Possible pandemics.
l Cost-effective interventions include:
– Interventions to reduce neonatal mortality (50 per-
cent of total child deaths)
– Treatment of HIV-positive mothers, treatment of sex-
ually transmitted infections
– Controlling tobacco use, particularly through
taxation
– Lifelong medical management of risk factors in indi-
viduals at high risk for heart attacks or strokes, using
aspirin and other drugs, would benefit tens of millions
of individuals.
l Reform of health services and systems is needed,
including:
– Provider incentives
– Provider focus on selected intervention to gain
experience
– Strengthening surgical capacity at district hospitals
– Targeting limited resources to diseases affecting the
poor, e.g., TB in low-income countries
– In middle-income countries, public finance (or pub-
licly mandated finance) of a substantial package of
clinical care for all.
l Generation and diffusion of new knowledge and prod-
ucts underpinned the enormous improvements in health
in the twentieth century and need to be applied for the
control of NCDs, HIV, TB, and neglected populations.
Note: HIV = human immunodeficiency virus; TB = tuberculosis;
NCD = non-communicable disease.
Source: Adapted from Jamison DT, Breman JG, Measham AR, Alleyne
G, Claeson M, Evans DB, et al., editors. Disease control priorities in
developing countries. Chapter 1, Investing in health, Table 1.1. 2nd ed.
Washington, DC: World Bank; 2006.
Chapter 15 Health Technology, Quality, Law, and Ethics
diphtheria–pertussis–tetanus (DPT), poliomyelitis (polio),
and measles–mumps–rubella (MMR) vaccines provides
new possibilities to control the major infectious disease
killers of children. New technologies such as the advent of
antiretroviral treatment for HIV have led to startlingly suc-
cessful improvements in the quality of life and longevity of
HIV/AIDS patients, and the prevention of onward transmis-
sion of HIV from mothers to babies and sexual partners.
The WHO and many other global health stakeholders con-
tinue this work and produce analyses to contribute to policy
making based on economic epidemiological evidence. This
work affects policy, slowly but importantly.
In 2003, the Bellagio Study Group on Child Survival
estimated that the lives of 6 million children could be saved
each year if 23 proven interventions were universally avail-
able in the 42 countries in which 90 percent of child deaths
occurred in 2000. The MDGs set out in 2001 provided tar-
gets for economic, educational, and environmental improve-
ments, with three specifically focused on health: reducing
child mortality; reducing maternal mortality; and control of
HIV, TB, and other diseases. While important progress is
being made, some of these targets will not be achieved by
2015. The global public health infrastructure will need to be
expanded in content and strengthened in order to implement
lessons learned in childhood routine immunization, safe
maternity care, and nutritional security (see Chapter 16).
In medium- and low-income countries the difficulties are
much more severe because of limited resources for health and
the weak infrastructure of facilities and human resources in
many countries. The key issues relate to NCDs, as in developed
countries, so the interventions most needed address cardiovas-
cular diseases, cancer, and injury, as well as diarrheal diseases,
malnutrition, vaccine-preventable diseases, HIV, TB, malaria,
and neglected tropical diseases. Efforts should be focused on
low-cost interventions such as smoking reduction, vitamin and
mineral fortification of foods, HIV, TB, and malaria control,
along with maternal and child health protection.
Priority Selection in High-Income Countries
As discussed in Chapters 5 and 13, high- and middle-income
countries also face complex health challenges, including
aging populations, health costs, rapid development of new
drugs and technologies, high rates of NCDs, and the rising
prevalence of obesity and diabetes. Selection of priorities
for health care expenditure from public and private sources
has become a major focus of managing health systems.
In the industrialized countries, technological advances in
the medical and public health fields have been major contribu-
tors to increasing longevity but also rising health costs. This
situation has led to pressures for greater selectivity in adopting
costly innovations without adequate assessment of benefits
and costs. Many countries have adopted more cautious poli-
cies with regard to financing high levels of expansion of new
777
technology in the field of medical equipment, clinical proce-
dures, or medications. Organized assessment of technology is
now an essential feature of health management at the interna-
tional, national, and local levels of service delivery. The major
responsibility for technology assessment is at the national
level, even with decentralization of service management.
With available resources being limited, health systems
must choose interventions to be selected and how health
systems are to be organized for efficiency and effectiveness
while meeting public expectations. The US Patient Protec-
tion and Affordable Care Act (PPACA, more generally called
ACA or “Obamacare”) is undertaking reform measures to
promote efficiency and prevention to reduce per capita health
costs and to include more people in prepaid health care (see
Chapters 10 and 13). These include preventive measures as
recommended by Healthy People 2020 and implementation
committees for selection of cost-effective measures to reduce
morbidity and mortality to reduce health costs. The range
of services to be promoted includes smoking cessation,
increased physical activity, weight loss, healthy dietary prac-
tices, cancer screening, and many others that have not been
previously accessible to those living in poverty and with no
or limited health insurance. There is an emphasis on vaccina-
tion for children and adults.
The WHO promotes the widespread use of basic radio-
logical units (BRUs) to increase access to low-cost, effective,
diagnostic X-rays, especially in rural areas in developing
countries. BRUs are hardy, relatively inexpensive pieces of
radiological examination equipment that can be used in harsh
field conditions for simple diagnosis of fractures and respi-
ratory infections. The WHO estimates that 80 percent of all
diagnostic radiology can be performed adequately using sim-
ple, safe, and low-cost equipment, supported by training of
local people to operate and maintain the equipment. This is a
consensus view of leading radiologists and clinicians helping
the WHO to develop model equipment and training material.
The WHO World Health Report of 2009 focused on
health technology assessment, stating:
“Technology continues to transform the medical care system and
to improve length and quality of life – but at substantial cost. It
is almost inconceivable to think about providing health care in
today’s world without medical devices, machinery, tests, computers,
prosthetics, or drugs. Medical technology can be defined as the
application of science to develop solutions to health problems or
issues such as the prevention or delay of onset of diseases or the
promotion and monitoring of good health.”
Appropriate technology in the health field is becoming
increasingly complex, laden with economic, legal, and ethi-
cal issues. Professional and public opinion demands make
this a highly sensitive area of health policy, but responsible
management of resources requires decision making that
includes consideration of the effectiveness, costs, and alter-
natives of any new technology (Box 15.2). Failure to adopt
778
new innovations can result in obsolescence, while excessive
expenditures for hospitals and medical technology prevent
a health system from developing more cost-effective pre-
ventive approaches, such as improved ambulatory care, or
supportive care for the chronically ill.
HEALTH TECHNOLOGY ASSESSMENT
Technology adoption can be a highly emotional and con-
troversial issue, in advocacy of new cancer treatments or in
criticism of managed care or national regulatory agencies,
but spending limited national resources on some devices
or medications of unproven value or inappropriately long
hospital stays denies resources needed for other aspects of
health care. A society must be able and willing to pay for
medical innovation or improving quality of life by medical
and public health interventions. Underfunding of a health
system can deny these benefits just as misallocation of
resources does, and this is a political issue even more than
a professional one.
Medical and health technology assessment is the pro-
cess of determining the contribution of any form of care
to the health of the individual and community. It is a sys-
tematic analysis of the anticipated impact of a particular
BOX 15.2 Health Technology Assessment
Questions that form the basis of technology assessment for a
medical innovation include the following:
l Is it safe and cost-effective for the stated purpose?
l Is it a new service, or does it replace a less efficient inter-
vention which can be phased out of service?
l What is the need it addresses?
l Where is it in the order of priorities of development of
the facility?
l Does it duplicate a service already available in the
community?
l Does it make medical sense (i.e., does it help in diagnosis
and treatment for the patient’s benefit)?
l What are the alternatives?
l What are the resources needed in terms of supplies, staffing,
and upkeep?
l Can the facility afford it?
l What could otherwise be done with the resources it
requires?
Sources: Adapted from Kass N. Public health ethics: from foundations
and frameworks to justice and global public health. J Law Med Ethics
2004;32:232–42.
Sullivan SD, Watkins J, Sweet B, Ramsey SD. Health technology assess-
ment in health-care decisions in the United States. Value Health
2009;12:S39–44. Available at: http://www.ispor.org/htaspecialissue/
Sullivan [Accessed 14 December 2012].
Velasco-Garrido M, Busse R. Policy brief: Health technology assessment:
an introduction to objectives, role of evidence, and structure in Europe.
Geneva: WHO on behalf of the European Observatory on Health Systems
and Policies; 2005. Available at: http://www.euro.who.int/__data/assets/
pdf_file/0018/90432/E87866 [Accessed 22 October 2012].
The New Public Health
technology in regard to its safety and efficacy as well as its
social, political, economic, legal, and ethical consequences.
The technology may be a machine, a vaccine, an opera-
tion, or a form of organization and management of services.
Analysis should include cost–benefit and cost-effectiveness
studies (see Chapter 11) as well as clinical outcomes and
other performance indicators.
Pressures from medical professionals, manufacturers of
new medical equipment, and the public for adoption of new
methods can be intense and continuous. Care must be taken
that the specialists involved in committees for assessment
are not those who may directly or indirectly benefit from
the exploitation of technology, and who therefore may have
conflicts of interest. Assessment must be multidisciplinary,
involving policy analysts, physicians, public health special-
ists, economists, epidemiologists, sociologists, lawyers,
and ethicists. The available information needs to include
evidence from clinical trials, critical analysis of the litera-
ture, and the economic effect of adopting the technology on
allocation of resources.
Medical technology varies in complexity and cost, not
only to produce but in its utilization. Medical technology
that is inexpensive to supply and administer is known as
low technology or low-tech, while high technology or high-
tech refers to costly and complex diagnostic and treatment
devices or procedures.
At the low-tech end of the technology scale, oral rehy-
dration therapy (ORT) was developed in the 1960s for oral
replacement of fluids and electrolytes lost in diarrheal dis-
ease, particularly in children. It has been described as one
of the greatest medical breakthroughs of the twentieth cen-
tury. The introduction and wide-scale use of ORT for pre-
vention of dehydration from diarrheal diseases throughout
the world has saved hundreds of thousands of lives. Use
of insecticide impregnated bed nets and reintroduction of
DDT in household spraying along with vector control and
improving diagnostic tools are low-tech but effective and
key tools in malaria control.
Advances in endoscopic surgical techniques since the
1990s, and in robotic surgery since 2000, have greatly
improved patient care by reducing trauma, discomfort, and
length of hospital stay and endoscopy has become the surgi-
cal approach of choice for many procedures. Since reports
of the first 100 operations performed in France in 1990,
endoscopy has spread rapidly to all parts of the world. It is
now recognized by surgeons worldwide as a safer, less trau-
matic and more effective alternative to traditional invasive
surgery. Although the operating time is longer, patients are
discharged from the hospital within several days and return
to work shortly thereafter, compared to the long hospital
stays after more invasive surgical procedures in the past.
Following traditional abdominal surgery, a patient may
acquire infections and require intensive care initially and a
recovery period of many weeks.
http://www.ispor.org/htaspecialissue/Sullivan
http://www.ispor.org/htaspecialissue/Sullivan
http://www.euro.who.int/__data/assets/pdf_file/0018/90432/E87866
http://www.euro.who.int/__data/assets/pdf_file/0018/90432/E87866
Chapter 15 Health Technology, Quality, Law, and Ethics
Endoscopic surgery for cholecystectomy and esopha-
geal, colorectal, hernia repair, renal, orthopedic, and other
forms of surgery which previously were carried out with the
patient remaining in hospital for many days are now done
on a not-for-admission basis. Not-for-admission surgery has
become standard practice in hospitals, extending the range
of outpatient surgery and the comfort of patients who can
return to their own homes to recuperate and return to regu-
lar activities much sooner. Fewer complications arise and
patient comfort and economic implications are important.
As a result, fewer hospital beds are needed for postoperative
care than previously thought necessary, while surgical and
ambulatory care facilities may need expansion to accommo-
date the growing elderly populations needing surgical inter-
ventions but requiring shorter recovery. This innovation is
now accepted as the standard of much of modern surgical
care and shows that simple organizational changes can save
money and improve patient safety and comfort.
The bacterium Helicobacter pylori was first identified
as the cause of peptic ulcers of the stomach and duodenum
in 1982 (Robin Warren and Barry Marshall, Nobel Prize
2005). This discovery led to effective diagnosis and rapid,
inexpensive treatment of chronic peptic ulcer disease. This
has resulted in elimination of a major component of surgi-
cal procedures for chronic peptic ulcer diseases as well as a
reduction in gastric cancer (see Chapter 4). Surgery for gas-
trectomies, vagotomies, and other outdated forms of treat-
ment are now virtually gone, contributing to a decreased
need for hospital beds even for an aging population. This
and many other innovations in medical care have led to a
growth in the use of ambulatory care for many forms of
surgical, medical, and mental health care, along with much
shorter length of hospital stay than in previous times. All of
these factors have led to greater emphasis on ambulatory,
outpatient, and home care services.
The dissonance between high-tech and low-tech proce-
dures may lead to serious consequences in any health system.
Choices require well-informed analysis of benefits, costs,
alternatives, ethical considerations, and political consequences
before limited health care resources are allocated between
hospital-based high-tech medicine and low-tech primary care.
High-tech procedures are usually applied in hospital
settings in the context of other highly specialized care for
seriously ill, often terminal, patients. Computed tomogra-
phy (CT), invented in the 1960s, quickly proved to be an
extremely valuable diagnostic tool. Advances in CT, mag-
netic resonance imaging (MRI), and subsequent imaging
techniques have proven to be cost-effective and lifesaving,
replacing less efficient and more dangerous invasive proce-
dures. The CT and MRI scans allow the clinician to reach a
rapid diagnosis of many lesions before they can be detected
by other invasive and dangerous diagnostic techniques, at
stages where the lesions are subject to earlier and more
effective interventions. Imaging technology is advancing
779
rapidly and promising inexpensive new systems for long-
distance transmission of imaging to medical centers may
provide enormous benefits to people living in rural or
developing countries. Recent advances in low-intensity CT
screening of long-term heavy smokers for lung cancer have
recently been added to recommended and potentially effec-
tive and cost-saving practice and may change the outlook
for this disease in the coming decade (US Preventive Ser-
vices Task Force, 2012 Flahault and Martin Moreno, 2013).
Technology assessment also examines methods of pre-
venting and managing medical conditions. Treatment proto-
cols or clinical guidelines are based on decision analysis of
accumulated weight of evidence. Published clinical studies
are assessed in meta-analyses, using statistical methods to
combine the results of independent studies, where the stud-
ies selected meet predetermined criteria of quality. This pro-
vides an overview from pooling of data, but also implies an
evaluation of the studies and data used. Clinical guidelines
are part of raising standards of care, but also contribute to
cost containment. Many countries form professional study
groups to carry out meta-analyses on important health pol-
icy issues and new technologies.
Technology Assessment in Hospitals
There is considerable variance among countries, hospitals
consume between 40 and 70 percent of total national health
expenditures, with pressures for increased staffing and
novel medical technology being a continual inflationary
factor. Industrialized countries have all reduced their acute
care hospital bed supplies and length of stay so that their
expenditures for hospital care have fallen to between 30-40
percent of total health expenditures. Shorter stays and older
patients have resulted in a drift towards intensive care, espe-
cially for internal medicine patients. Medical innovation is
a continuing process with new diagnostic and treatment
modalities reaching the market.
Hospitals no longer live in splendid isolation in the
medical economy. A national or state government needs
regulatory procedures to rationalize distribution of medical
technology. The “certificate of need” is a form of technology
assessment that has been used in the USA since the 1960s
to assess and regulate the development of hospital services
to prevent oversupply and costly duplication of services. It
attempts to establish and implement the use of rational crite-
ria for diffusion of expensive new technology. Whether this
has had a lasting impact on restraining the excesses of high-
tech medicine is arguable. This regulatory approach was
limited to the hospital setting and failed to stop the develop-
ment of high-tech medical services such as ambulatory for-
profit CT, imaging, and in vitro fertilization centers.
Many countries have adopted national technology
assessment systems to review topics as far-ranging as
guidelines for acute cardiac interventions; liver, heart, and
7
80
lung transplantation; minimal access surgery; and beam and
isotope radiotherapy. Other technology assessment guide-
lines include diagnostic ultrasound, sleep apnea, molecu-
lar biology, prostate cancer, MRI, and new medications for
inclusion in a national health system’s approved basket of
services.
Despite the limitations of this approach, where govern-
ments do not directly operate health care services, gov-
ernmental regulation is necessary to prevent inequities in
services by excessive development in some geographic
areas at the expense of others, or by overexpansion of the
institutional sector of health care at the expense of primary
care. Regulatory mechanisms are essential in health care
planning to restrain excessive and inappropriate use of
high-tech services, but need augmentation by fiscal incen-
tives to promote other essential services.
Hospitals everywhere face serious problems of hospital-
acquired infections, which occur in about 5 percent of all
hospitalizations. Healthcare-associated infections (HAIs),
including multidrug-resistant bacterial infections, cause
long lengths of stay, high costs, and most importantly, unex-
pected deaths and serious disabilities. Prevention of hospital-
acquired infection requires ongoing training, staffing, and
organization. The Centers for Disease Control and Prevention
(CDC) defines HAIs as “infections caused by a wide variety
of common and unusual bacteria, fungi, and viruses during
the course of receiving medical care”. Some of the preventive
measures are simply promoting frequent hand washing by
caregivers and visitors, and immunization of staff members
against influenza and pneumonia, which can be problematic
if there is staff resistance to influenza vaccination.
Training and routine supervision of cleaning staff are
also vital, as are strict infection control measures for isola-
tion rooms, strict protocols for catheter care, surgical suite
sterility, surgical site infections, central line associated
bloodstream infections, ventilator-associated pneumonias,
catheter-associated urinary tract infections, and Clostridium
difficile-associated disease. Guidelines for their control in
surgical dialysis, pediatric, outpatient, and other vulnerable
departments are available from CDC. The benefits of pre-
ventive procedures for this problem include cost estimates
ranging from US$5.7–6.8 billion (20 percent of infections
preventable) to US$25.0–31.5 billion, yet 70 percent of HAIs
are preventable by well-known methods such as frequent and
careful hand washing by medical and nursing staff, catheter
and infusion care, and other similar measures (CDC, 2012).
Technology Assessment in Prevention and
Health Promotion
Technology assessment of preventive care programs
includes evaluation of the methodology itself, along with
the costs and measurable benefits, as in reduced burden
The New Public Health
of disease. DOTS is the standard management of sputum-
positive and sputum-negative TB, at low cost for DALYs
saved. The coexistence of HIV and other complications
has created multidrug-resistant tuberculosis (MDR-
TB), which is difficult and costly to treat and cases
constitute a source of continuing spread of the disease.
A 2012 meta-analysis of cost-effectiveness of MDR-TB
treatment in Estonia, Peru, the Philippines, and Russia
shows it to be cost-effective and best carried out on an
ambulatory basis (Fitzpatrick and Floyd, 2012).
Wide use of available and effective vaccines such as
Hemophilus influenza b (Hib), pneumococcal pneumo-
nia, influenza and rotavirus reduce hospitalizations and
mortality from respiratory and diarrhoeal diseases among
children, the elderly and other age groups. Vaccine prices
generally fall after their initial period of use as manu-
facturing costs are lessened by improved methods or by
bulk purchase contracts, as occurs in the public sector. For
example, in 2012 MMR vaccine cost US$19.33 per dose if
purchased through the CDC, but US$52.73 per dose if pur-
chased in the private sector in 10 packs of single-dose units
of the vaccine. A combined diphtheria, tetanus, acellular
pertussis (DTaP) vaccine cost US$15.00 when purchased
through CDC, while the same vaccine purchased with hep-
atitis B and inactivated polio vaccine (IPV) cost US$52.10 per
dose. But the combination saves repeated visits and loss
of compliance for that reason. The new human papilloma-
virus (HPV) cervical cancer vaccine cost US$130.27 per
dose for the series of three doses per person, while the vac-
cine against diarrhea-causing rotavirus, approved in 2006,
cost US$106.57 per dose for the recommended three doses
(CDC, 2012).
The WHO recommends the inclusion of rotavirus vac-
cination in a country’s immunization program, but the costs
of the current generations of rotavirus vaccines are high in
comparison to the budgets for vaccines for prevention of
childhood illnesses in many developing countries. Many
cost-effectiveness studies have shown this vaccine to be
highly beneficial and it could help to reduce the very high
global burden of disease of over 500,000 child deaths and 2
million hospitalizations occurring annually (Tu et al., 2011).
Vaccine programs must take into account transportation
and administrative costs and expenses of ordering, storing,
inventory control, cold chain, insurance, wastage, and spoil-
age. Multiple vaccines in one dose are less costly and less
inconvenient for all. Examples include DTaP plus polio and
Hib, or MMR (see Chapters 4 and 6). There is a need for
implementation of legal protection of manufacturers from
excessive litigation judgments while protecting the interests
of the public and individuals who may have reactions to
vaccines.
In 2012, the reappearance of pertussis and diphthe-
ria raised concerns about immunization coverage and
Chapter 15 Health Technology, Quality, Law, and Ethics
efficacy. Public opinions on vaccination may not be as
supportive as in previous years. Mothers who oppose per-
tussis immunization for their children, such as occurred in
the UK during the 1980’s, leave their children vulnerable
to a serious and often deadly disease, which has recurred
since 2010.
The WHO estimates the cost of all immunization activi-
ties in all 117 low- and middle-income countries for the
period 2006–2015 to be US$75 billion, while low-income
countries would need US$35 billion. The rate of adoption
of currently available and new vaccines will be determined
by governmental decisions in each country, although exter-
nal aid – such as that of the Global Alliance for Vaccines
and Immunization (GAVI), an international public–private
consortium to promote vaccination – is a valuable resource.
The United Nations Children’s Fund (UNICEF) is con-
cerned about supply problems as well as costs, but the key
issue relates to political decisions, funding, and capacities
of national immunization systems.
Despite an excellent vaccine having been available since
the 1960s, measles epidemics continue to occur in the indus-
trialized countries. In the 1900s global deaths from measles
were in the order of 1 million people per annum. Two major
epidemics of measles occurred in Canada in the early 1990s,
despite high rates of immunization coverage. Following this, a
1993 Delphi conference of experts from 31 countries reached
a consensus recommending a two-dose measles immuniza-
tion policy. Measles eradication has been set as a goal by the
WHO and 90 percent reduction in cases and fatalities has
been achieved since the 1990s. However, measles elimination
requires coverage of 95 percent of children and two doses of a
measles-containing vaccine (preferably MMR).
Measles reappeared as a widespread disease in Europe in
2010–2013 with tens of thousands of cases, many hospital-
izations, and some deaths. It spread to the Americas, brought
by travelers, and resulted in modest sized outbreaks, includ-
ing the UK in 2012–2013. Eradicating measles by 2020 is
projected to cost an additional discounted US$7.8 billion and
avert a discounted 346 million DALYs between 2010 and
2050. As new vaccines enter the field, it is important to evalu-
ate their effectiveness, costs, and the benefits to be derived.
The cost of the hepatitis B vaccine initially was over
US$100 for an immunization schedule of three doses but
has come down dramatically to less than US$1 per dose
in developing countries for bulk purchases. However, in
the USA, the price of vaccination per dose is estimated at
US$41 if given by a general practitioner, US$15 if adminis-
tered through an existing childhood immunization program,
and US$17 if given through the school medical system.
This is a standard vaccine covered by public and private
health insurance systems. The vaccine is a cost-effective
method to prevent liver cancer and the long-term effects of
chronic hepatitis.
781
Screening and education for thalassemia in high-prev-
alence areas have nearly eradicated the clinical disease but
not its carrier status in Cyprus, southern Greece, and other
countries. Newborn screening and case management for
phenylketonuria, congenital hypothyroidism, Tay–Sachs
disease, and many other genetic diseases have been shown
to be far less expensive than post-facto treatment of severely
developmentally delayed and dependent children born with
these diseases (see Chapter 6).
The success of Papanicolaou (Pap) smear screening in
reducing cancer of the cervix mortality since the 1960s has
been dramatic. The discovery of causation of cancer of the
cervix by HPV strains led to development of an effective
vaccine, which has been in use since 2006. Recent evidence
shows that male circumcision can reduce transmission of
HPV as well as HIV and other sexually transmitted dis-
eases, and it is being adopted as an effective intervention
in countries with high rates of both HIV and cancer of the
cervix, such as in sub-Saharan Africa.
The drastic reduction in cancer of the cervix provides
a powerful demonstration of the effectiveness of public
health screening and other measures to control this major
malignant cause of death in women. Screening for cervi-
cal cancer by Pap smears is recommended annually for
high-risk groups, and every 2 or 3 years for other adult
women (Box 15.3). Screening will remain vital for many
years to come as the HPV vaccine comes into general
use, and as its cost is reduced, but its protective effect for
individual and herd immunity will not replace the need
for ongoing screening for this very common cancer. HPV
vaccine is also being recommended for all boys to pre-
vent oral and anogenital cancers and HPV transmission
to girls.
Routine mammography screening for breast cancer
every 1–2 years is recommended by the US National Can-
cer Institute for women over the age of 40 and for younger
women with high-risk factors (e.g., previous cancer, fam-
ily history, genetic markers). Cost-effectiveness analysis is
now an essential part of decision making in health policy
and priorities. While there is controversy over the frequency
of routine testing, mammography remains a mainstay in
women’s health and contributes to early case finding and
falling mortality rates from breast cancer. Figure 15.1 dem-
onstrates differences in utilization of mammography among
US women in the age group 50–64 years within the previ-
ous 2 years, by insurance status. US women with private
insurance (mostly through place of employment) had over
70 percent compliance, those with public insurance (pri-
marily Medicaid) averaged about 60 percent compliance,
while those with health insurance had average compliance
rates of about 45 percent during the period 1993–2010.
The UK National Health Service (NHS) invites women
between the ages of 50 and 70 for screening every 3 years;
Cancer of the cervix is the second most common cancer among
women worldwide, with about 500,000 new cases and 250,000
deaths worldwide annually. Approximately 80 percent of cases
occur in low-income countries, where cervical cancer is the
second commonest cancer in women (WHO, 2012).
In the USA, and other industrialized countries, the incidence
and mortality of cancer of the cervix have been going down
steadily since the introduction of Papanicolaou (Pap) smear
testing. Cervical cancer incidence declined during the period
1999–2008 by 2.3 percent per year and mortality declined by
1.9 percent per year an estimated 12,170 cases of invasive cer-
vical cancer diagnosed in the USA with 4220 deaths in 2012.
Prevention of cancer of the cervix has until recently mainly
focused on Pap smears to detect the disease while still in a pre-
cancerous (cancer in situ) phase, and this procedure reduced
rates dramatically over the latter part of the twentieth century. The
newly developed and highly effective vaccines against key strains
of human papillomavirus (HPV) is now being used in routine
immunization of young girls and more recently boys as well. The
high cost of the vaccine precludes its rapid diffusion to most parts
of the world but its use is spreading and being included in immu-
nization programs funded by donor agencies in sub Saharan
Africa. The vaccine should, in principle, also be used by adult
women, in addition to continuation of routine Pap smear testing.
In the past decade, evidence of HPV as the cause of cancer
of the cervix and the presence of HPV in uncircumcised men has
brought circumcision back to professional and public debate.
Reports from Africa of reduced risk of acquiring HIV among
circumcised men have brought new attention to adult male cir-
cumcision, which is now actively promoted many sub-Saharan
African countries.
The technological breakthroughs of the Pap smear in the
1950s, HPV testing in the 1990s, and the HPV vaccine in the
2000s should also include prevention by male circumcision.
Visual inspection of the cervix and cryotherapy can treat pre-
cancerous cervical lesions in areas of developing countries
as part of community health worker programs.
Colorectal cancer, the 7th leading cause of death in high
income countries, is amenable to prevention by early screening
using colonoscopy and fecal occult blood (FOB) testing. Screening
is recommended for all persons over age 50 at 5 year intervals
along with annual FOB testing. Where there is a family history of
colorectal cancer or polyps, routine screening should begin ear-
lier. Increasing use of screening and improved medical care are
resulting in improving survival and declining mortality rates.
Stomach cancer is 10th leading cause of death in upper
middle income countries. Prevention relies on early treatment
of chronic peptic ulcer disease caused by Helicobacter pylori
infection. This is readily diagnosed by a simple breath test and
completely cured by low cost antibiotics. Increased awarenss
and access to these services would enhance long term trends of
reducing mortality from stomach cancer.
Liver cancer is 8th leading cause of death in upper middle
income countries due to the global prevalence of hepatitis B
and helaptitis C. Hepatitis B is now falling due to widespread
vaccination in childhood. Hepatitis C is now the major cause
of liver cancer affecting hundreds of millions of persons world-
wide. There is still no vaccine currently available, but screen-
ing and treatment is now used in the industrialized countries
and will become more widely used as simpler, less costly treat-
ments with less side effects are becoming available.
Early detection of lung cancer with spiral low dose
tomodensitometry for smokers is recently being recommended
by many professional bodies.
More basic cancer preventive measures such as smoking
cessation, healthy diets, regular exercise, and moderate alco-
hol use are discussed in chapter 5.
Sources: World Health Organization. Sexual and reproductive health.
Cancer of cervix. Available at: http://www.who.int/reproductivehealth/top-
ics/cancers/en/ World Health Organization. The top 10 leading causes of
death (2011). Available at: http://who.int/mediacentre/factsheets/fs310/en/
index1.html
Centers for Disease Control and Prevention. Cervical cancer trends 2012.
Available at: http://www.cdc.gov/cancer/cervical/statistics/trends.htm
[Accessed (13.12.2012)].
BOX 15.3 Technology for Prevention of Cervical, Colorectal, Liver, Stomach and Lung Cancers
1993
0
10
20
30
40
50
P
er
ce
nt
ag
e 60
70
80
90
100
1995 1997 1999 2001
Year
2003 2005 2007 2009
Private insurance
Public insurance only
No insurance
FIGURE 15.1 Self-reported percentage of women aged 50–64 years
receiving a mammogram in the past 2 years, by health insurance sta-
tus, USA, 1993–2010. Source: Centers for Disease Control and Prevention.
QuickStats from the National Center for Health Statistics: Percentage of women
aged 50–64 years who reported receiving a mammogram in the past 2 Years,
by health insurance status — National Health Interview Survey, United States,
1993–2010. MMWR Morbid. Mortal. Wkly. Rep. 2013;62:651. Available at:
http://www.cdc.gov/mmwr/pdf/wk/mm6232 (accessed 12.1.14).
with an average of 75 percent respond to the invitation, of
whom 4 percent are referred for further testing (UK Can-
cer Research, 2012) and as in most industrailized countries
breast cancer mortality rates are falling impressively (see
Chapter 5).
Health promotion in reducing exposure to HIV and
cigarette smoking has been shown to be very cost-effec-
tive despite its low-tech or non-technological meth-
odology, involving primarily group or mass education.
Hypertension screening and case management is low-
tech but highly effective in preventing strokes and blind-
ness.
Low-tech innovations have had an important impact
in reducing death and injury. These include mandatory
use of car seat belts (introduced since the 1970s and
1980s in many countries), children’s car seats, air bags,
and bicycle and motorcycle helmets. Iodization of salt,
vitamin A supplementation, and food fortification pre-
vent large numbers of clinical cases of severe retarda-
tion, death, and blindness at low cost per child protected.
http://www.who.int/reproductivehealth/topics/cancers/en/
http://www.who.int/reproductivehealth/topics/cancers/en/
http://who.int/mediacentre/factsheets/fs310/en/index1.html
http://who.int/mediacentre/factsheets/fs310/en/index1.html
http://www.cdc.gov/cancer/cervical/statistics/trends.htm
http://www.cdc.gov/mmwr/pdf/wk/mm6232
783Chapter 15 Health Technology, Quality, Law, and Ethics
TABLE 15.3 Examples of High-Tech and Low-Tech Health Problem Solving
Problem High Tech Low Tech
Birth defects Surgical repairs, rehabilitation Folic acid fortification of flour, vitamin supplementation
before and during pregnancy
Infectious diseases Treatment – antibiotics Vaccination, sanitation, handwashing, infection control in
hospitals, health facilities, and nursing homes
Breast cancer Screening – mammography Nutrition, self-examination, routine medical examination
Colon cancer screening Colonoscopy Nutrition, vitamin D supplements; fecal occult blood testing
Acute myocardial infarction,
primary, secondary prevention
Coronary angioplasty, stent, bypass
surgery, heat transplantation
Antiplatelet thrombosis treatment (e.g., aspirin, intravenous
streptokinase, beta-blocker); rehabilitation; diet, exercise,
smoking cessation
Gallstones Lithotripter, abdominal
cholecystectomy
Endoscopic surgical removal
Head injuries Intensive care Helmets for bicycle riders and motorcyclists, seat belts in
front and rear of motor vehicles
Thalassemia Transfusions, chelating agents;
prenatal diagnosis, amniocentesis,
chorionic villus biopsy
Screening, education, counseling
Dehydration Infusions Oral rehydration
Neural tube defects Surgery, pregnancy termination Folic acid fortification of flour and grain products, supple-
ments for women of fertile age
Liver cirrhosis, liver failure,
cancer
Liver transplant Hepatitis B vaccine, risk reduction activities among intrave-
nous drug users, screening blood donors
Cancer of stomach Surgery, chemotherapy Dietary change, cure of Helicobacter pylori-generated
gastric ulcers
Cancer of cervix Pap smear screening; visual inspection and cryotherapy;
human papillomavirus vaccine
Education for reducing risk factors for the cardiovascu-
lar disorders is far less costly than the premature deaths
and high medical costs of patients suffering stroke and
congestive heart failure. Health education, condom and
needle supply, and screening of blood donations are the
most important effective community health measures
against the spread of HIV. Table 15.3 shows a compari-
son of high-tech and low-tech approaches, which often
complement each other, to selected health problems.
Technology assessments represent the current con-
sensus derived from reviews of published studies and
exchange of views of highly qualified clinicians, epide-
miologists, and economists within a context of technol-
ogy assessment. They may change over time as new data
or innovations are reported, and this possibility should
be kept in mind in such discussions. Technology assess-
ment mobilizes information and critically analyzes many
aspects of medical technology to build a wide community
consensus to influence policy decisions. Public opinion,
political leadership, and administrative practice, as well
as the scientific merit of a case are all factors in develop-
ing a consensus.
Technology Assessment in National Health
Systems
Technology assessment requires an organization within
the framework of national regulatory agencies. The FDA
serves this purpose as a statutory body within the US Pub-
lic Health Service. Sweden, Canada, Australia, the UK, the
Netherlands, Spain, and other countries also have technol-
ogy assessment advisory or regulatory agencies established
by national governments to monitor and examine new tech-
nologies as they appear. Sweden has a widely representa-
tive national Swedish Council for Technology Assessment
in Health Care which has an advisory role to the national
health authorities.
The processes used in traditional systems to regulate
food and drugs for efficacy, safety, and cost are more
recently being applied to new medical devices and proce-
dures. The unrestricted proliferation of new procedures
presents serious dilemmas for national agencies con-
cerned with financing health care and controlling cost
increases. Non-governmental health insurance shares
this concern, as does industry, which bears much of the
784
cost of health insurance through negotiated, collective
bargaining, “voluntary” health insurance in the USA.
Most industrialized countries have national health ser-
vices or national health insurance and are thus vitally
interested in health costs and technology assessment.
Many industrialized countries maintain technology
assessment and cost-control activities. In the USA, the
Agency for Healthcare Research and Quality (AHRQ)
maintains oversight and studies related to clinical infor-
mation, including evidence-based practice, outcomes and
effectiveness, comparative clinical effectiveness, risks
and benefits, and preventive services.
In Canada, the Health Protection Branch of the Federal
Department of Health reviews medical devices and drugs
and, with consent of the provincial governments, approves
new medical procedures. Concern by governments over the
cost implications of new procedures led to this practice.
Since 1988, a network of government and professional
bodies has formed a non-profit agency for technology
assessment (Canadian Agency for Drugs and Technologies
in Health, CADTH). This supports the provincial adminis-
tration of health insurance in resisting professional, com-
mercial, or political pressures to add untested technology
or procedures to the health system as covered benefits. A
comparison of rates of procedure performance between
provinces shows very high discrepancies, as high as two-
fold, in procedures such as coronary artery bypass graft or
prostatectomy. Control of acquisition of high-tech equip-
ment by national or state authorities is essential to prevent
expenditures on high-cost equipment without adequate
assessment.
The New Public Health
DISSEMINATION OF TECHNOLOGY
The rapid spread of high-tech medical equipment has played
a substantial role in escalating health costs. A comparison of
the number of MRI scanners per million population in mem-
ber countries of the Organisation for Economic Co-operation
and Development (OECD) (Table 15.4) showed Japan and the
USA with the highest number at 43.1 in 2010 (23.2 in 2000–
01) and 31.6 (15.4 in 2000) per million, respectively, while
the median was 10.5. Canada ranked fourteenth among the 20
OECD countries with 8.2 MRI scanners per million. Compar-
ing CT scanners showed that Japan had the highest number
at 97.3 while the median was 15.6. Canada was in eighteenth
place among the 28 OECD countries reporting in 2010 with
14.2 CTs per million population. However, it should be noted
that countries with the most machines have lower productiv-
ity per machine. The USA has far fewer examinations per
machine than other countries (OECD, 2012).
The use of endoscopic surgery has spread worldwide
since the 1990s. Health professionals become almost
instantly aware of new developments from the news media
as well as professional diffusion of information at confer-
ences, in exchange visits, in published articles, and most
dramatically via the Internet. National policy to foster the
introduction of appropriate new technology requires a
careful program of regulatory and financial incentives and
disincentives to encourage or discourage diffusion of new
methods of prevention as well as of treatment and com-
munity health care. Kidney transplantation has become a
cost-effective and patient-friendly alternative to long-term
dialysis both in hospital and at home. The key limitation is
TABLE 15.4 High-Tech Medical Equipment Units, Selected Organisation for Economic Co-operation and
Development Countries and Years, 1986–2010 (Rate per Million Population)
CTs MRIs
Country 1986 2000 2005 2010 1986 2000 2005 2010
Japan 27.5 84.4a 92.6 97.3c 0.1 23.2a 40.1 43.1c
USA 12.8 25.1a 32.3 40.7d 0.5 15.4a 26.7 31.6
Israel – 5.7 6.6 9.2 – 1.4 1.7 2.0
Germany 6.9 12.2 15.4 17.7 0.7 4.9 7.1 10.3
UK 2.7 5.4 7.5 8.2 0.3 5.6 5.4 5.9
Canada – 9.8b 11.6 14.2 – 2.5 5.7 8.2
France 4.7 7.0 10.0 11.8 0.5 1.7 4.8 7.0
CTs = computed tomography scanners per million population; MRIs = magnetic resonance imaging units per million population.
Notes:
a1999
b2001
c2008
d2011.
Source: Organisation for Economic Co-operation and Development. Health policies and data: OECD health data – 2012 data; frequently requested data.
Available at: http://www.oecd.org/health/healthpoliciesanddata/oecdhealthdata2012-frequentlyrequesteddata.htm [Accessed 13 October 2012].
http://www.oecd.org/health/healthpoliciesanddata/oecdhealthdata2012-frequentlyrequesteddata.htm
Chapter 15 Health Technology, Quality, Law, and Ethics
the shortage of donors. The same can be said for liver trans-
plantation, which has also been shown to be cost-effective
in terms of DALYs saved from chronic liver diseases such
as cirrhosis and hepatitis B and C. Heart transplantation has
spread among major centers in western countries.
The black market in organs for transplantation has
become an international scandal of exploitation of poor
people in low-income countries, and is the subject of police
investigation in many countries, but it is difficult to control.
Bone marrow transplantation is now used widely and
is effective in saving the lives of many people with malig-
nant and non-malignant hematological disorders. Stem cell
therapy, by introducing new adult and embryonic stem cells
into damaged tissue to treat disease or injury, is becoming
feasible for a wide variety of conditions including cancer,
type 1 diabetes mellitus, Parkinson’s disease, Huntington’s
disease, celiac disease, cardiac failure, muscle damage, and
neurological disorders.
Limitation of new techniques or procedures to selected
medical centers allows the passage of time to fully assess
the merits and deficiencies of new technology before gen-
eral diffusion into the health care system. Such limitation,
however, is fraught with the danger of depriving the popu-
lation of benefits of new medical technology, and the pos-
sibility of restraint of trade to the economic advantage of
selected providers. Current advances in robotic-assisted sur-
gery will expand during the coming decade for brain, bone,
joint, prostate, and other surgery, and need to be assessed
with regard to patient care benefits, costs, and the econom-
ics of capitalization of such innovations. Stem cell therapy
is already widespread for some malignant conditions, and
with advances in genomics and molecular biological tech-
nology it is very likely to become a major therapeutic inter-
vention for many more conditions in the coming years. The
effects of new technology on insurance and managed care
systems are necessarily involved in decision making as to
inclusion of new procedures in their service plans.
Publication in the professional literature is an accepted
method of establishing the scientific merit of a treatment or
an intervention. Too rapid diffusion of a medical practice
can lead to disillusionment and confusion as to the mer-
its of a particular medical procedure, as happened during
the 1960s and 1970s with anticoagulant therapy for acute
myocardial infarction and gastric freezing for peptic ulcers.
Reviews of the literature should be critical and should
assess the scientific merits of published data, as well as the
sources of funding. Well-controlled large-scale clinical tri-
als are vital to establish the relative values of alternative
therapeutic approaches, as are meta-analyses of multiple
studies.
Dissemination of information about new medical inno-
vations in the popular media is almost immediate. Many
major newspapers and television networks have well-
informed medical reporters and commentators who have
785
access to electronic medical journals as quickly as do medi-
cal specialists in each field. News magazines may carry spe-
cial articles on new innovations, creating instant demand for
them as benefits in a health program. This ready access to
information has both benefits and dangers.
In the USA, health insurers have led the way in develop-
ing technology assessment and information synthesis, and
in evaluating the costs and benefits of new procedures. The
process is affected by public opinion, as well as by court
decisions. A landmark decision against a health mainte-
nance organization (HMO) in 1993 awarded US$29 million
in damages to the family of a terminal breast cancer patient
who died following refusal of the HMO to authorize a bone
marrow transplant, which was at the time an experimental
procedure. Denial of new technology may lead to increases
in malpractice suits. In countries with limited financial
resources, selection of technological innovations in health
care that can benefit patient care or the public health
requires a careful balance in order to use limited resources
well, and to gain from the application of appropriate new
health care technology.
Payment systems by national or private insurance
systems are crucial to introduce and control diffusion of
technology. Block budgets for hospitals have been more
effective in Canada than in the USA in restraining the pro-
liferation of high-tech equipment. This has led to criticism
of the limited access of Canadians to medical technology,
such as CT, MRI, and advanced cancer therapies. In the
USA, universal application of the diagnosis-related group
(DRG) payment system for Medicare, Medicaid, and most
private insurance had the effect of increasing ambulatory
surgery very dramatically, from 16 percent of all surgery
in 1980 to 80 percent in 2008 of all surgical procedures in
community hospitals (i.e., non-federal short-stay hospitals
or 85 percent of all hospitals in the country). Inpatient sur-
gical procedure rates declined from 85 percent in 1980 to
35 percent in 2008. Although the rate of visits to hospital-
based surgery centers remained largely unchanged in the
USA from 1996 to 2006, the rate of visits to private ambula-
tory surgery centers increased by about 300 percent.
HMOs and managed care organizations are paid on a
per capita basis and have a strong incentive for cost contain-
ment. They have developed procedures and medical guide-
lines for investigation and intervention that seek to reduce
unnecessary procedures. At the same time, HMOs are very
active in promoting preventive care and non-hospital care
insofar as this is compatible with good patient care.
Coronary bypass procedures decreased in frequency in
the USA between 2001–02 and 2007–08. In the USA, such
procedures are less frequently carried out in women and
African Americans, because of lesser access to health insur-
ance for African Americans and possibly because of biases
in terms of case assessment criteria in women. Cardiac
invasive procedures increased dramatically since the 1980s
786
in most industrialized countries, but with wide variation in
their use. The benefits of aggressive invasive management
of cardiovascular diseases remain controversial, but many
such procedures have proven beneficial in reducing mortal-
ity rates and improving quality of life.
Critical analysis of the need for surgery has resulted in
lower tonsillectomy and radical mastectomy rates along
with the increased use of outpatient procedures. Tonsillec-
tomy, a routine procedure until the 1960s, is now performed
infrequently since it was found to be of little medical value.
Cataract surgery is now largely done on an ambulatory
basis. The technology of home care has come to play an
important role in early discharge of patients from the hospi-
tal, as has the wide use of cancer chemotherapy and radia-
tion therapy on an outpatient basis.
DIFFUSION OF TECHNOLOGY
Innovations in health care through scientific and technologi-
cal advances are continuing, with exciting breakthroughs
being made in effective new treatments and public health
interventions, and this requires health authorities, practitio-
ners, and the public to maintain constant awareness of the
current state of the art. Diffusion of new technology or adap-
tations from basic science advances may begin slowly, and
then reach a “tipping point”, at which time a dramatic change
of trend occurs and it becomes the new standard or fashion.
Those with economic interests in the product try to
advertise and promote sales, while practitioners are ready
to try new methods to help their patients, but those who
must pay for services may ask for evidence of effectiveness,
safety, added value over present and known methods, and
benefit to the length or quality of life of the individual. This
can become a highly charged debate when those responsible
for adopting new measures in national health plans must
weigh one proposed addition against another, each with its
ardent professional, community, or business promoters. The
new HPV vaccine approved by the FDA in 2006 for preven-
tion of cancer of the cervix is an example.
The HPV vaccine is recommended for preteen girls at the
age of 11–12 years and also for females aged 13–26 to off-
set future sexual exposure to HPV-infected males and since
2012 recommended routinely for teenage boys as well. The
two competing main manufacturers of HPV vaccine are nat-
urally interested in increasing their market and market share,
and willing to reduce prices. The cost has been lowered
substantially for use in developing countries if purchased in
bulk, but costs are still prohibitive unless funded by interna-
tional donors. Competing low-cost manufacture in India has
encouraged the two main manufacturers to lower prices to
seek broader markets. In 2011, one manufacturer lowered its
price dramatically to US$5 per dose, a 67 percent reduction
in the current lowest public price. This has allowed GAVI to
adopt an HPV strategy for developing countries, where 88
The New Public Health
percent of cervical cancer deaths occur, with 275,000 deaths
of relatively young women each year (GAVI, 2011, 2012).
Pioneering projects promoting visual examination of the
cervix and local cauterization of abrasions by trained nurses
and community health workers are meant to increase access
to care in traditional villages remote from medical centers.
The duration of immunity and whether booster doses will
be required are still not known. Policy makers need to con-
sider whether the same money would have greater benefit
if used to provide pneumococcal pneumonia and rotavirus
vaccine for children in developing countries, which would
quickly save hundreds of thousands of lives. It is likely that
the wonderful new public health technology that is the HPV
vaccine will be absorbed quickly into public health prac-
tice at least in the industrialized countries, and is now being
introduced by international donor agencies in sub-Saharan
Africa.
QUALITY ASSURANCE
Quality assurance is an integral part of public health func-
tion and involves ensuring the quality of both health prac-
titioners and facilities. It is an approach that measures and
evaluates the proficiency or quality of services rendered.
Hospital accreditation is a long-standing method of qual-
ity assurance, providing many generations of health provid-
ers in North America with first hand experience of quality
assurance in community hospitals and long-term care facili-
ties, as well as ambulatory and mental health services.
Hospital accreditation has contributed to improvement in
standards of facilities and patient care throughout Canada
and the USA and has provided a working model for replica-
tion or adaptation internationally.
Adverse Events and Negligence
Iatrogenic diseases are adverse events that occur as a result
of medical management and result in measurable disability.
Negligent adverse events are those events caused by a failure
to meet standards of care reasonably expected of the aver-
age physician or other provider of care. Hospital-acquired
infections, anesthesia mishaps, falls, and drug errors are the
most common iatrogenic events.
Iatrogenic disease is a major cause of morbidity, pro-
longation of hospitalization, and even death. Hospital-
acquired (nosocomial) infections are estimated to occur
in 7–10 percent of hospital cases in Britain and the USA.
Primarily these are caused by urinary, respiratory tract, and
wound infections. It is becoming more common that infec-
tions involving organisms previously responsive to antibi-
otics are now resistant to many antibiotics and difficult to
treat. Infection control in hospitals is therefore an essential
part of hospital organization. Because hospitals are increas-
ingly being paid by DRGs, any secondary event prolonging
Chapter 15 Health Technology, Quality, Law, and Ethics
hospital stays may have adverse financial effects on the hos-
pital. In the USA, recent decreases in Medicare reimburse-
ments for nosocomial infections reflect this trend to provide
financial incentives to improve hospital infection control.
There is, therefore, a strong financial as well as professional
interest in reducing hospital-acquired infections.
A classic study of 32,000 hospitalizations in New York
State carried out by a Harvard University team showed
that 3.7 percent of hospitalized patients suffered adverse
events or injuries caused by medical mismanagement which
resulted in measurable disability. Of these, 28 percent were
due to negligence, so that 1.03 percent of all hospitalizations
involved medical negligence leading to measurable injury.
Of the total of some 100,000 adverse events in the study
group, 57 percent recovered within a month and 7 percent
had severe injury. Some 14 percent or 14,000 people with
adverse events died as a result; 51 percent of these deaths
were due to negligence. A 1999 report of the US National
Institute of Medicine estimated that between 44,000 and
98,000 people die annually in the USA from medical errors
occurring in hospitals, but these data are considered to be
overestimated in some studies. Adverse drug events (ADEs)
result in 700,000 emergency department visits and 120,000
hospitalizations annually, with US$3.5 billion spent on
extra medical costs. CDC estimates that at least 40 percent
of the costs of ambulatory (non-hospital setting) ADEs are
preventable.
A 2008 report by the Office of Inspector General of the
US DHHS reported that 13.5 percent of Medicare benefi-
ciaries experienced adverse events and that for 1.5 percent
of beneficiaries, these adverse events contributed to their
deaths. An additional 13.5 percent of beneficiaries in the
sample experienced temporary harm as a result of their
medical care, bringing the total percentage of beneficiaries
experiencing instances of care-related harm to 27 percent.
Nearly half (44 percent) of these adverse or temporary harm
events were preventable.
Hospital-acquired infections cause 99,000 patients
deaths in the USA every year (AHRQ, 2009). Higher rates
are seen among the elderly and the poor. Rates are lower in
teaching hospitals than in community hospitals. About 20
percent of the events were related to drug reactions or dos-
age errors. Less than 3 percent of those injured brought civil
litigation for the negligence. The search for “bad apples” –
that is, unethical, criminal, or incompetent health providers –
is necessary, but not sufficient to stem the problems created
by the health system itself. Prevention requires organized
activity. Investigation of adverse events helps to identify
methods of prevention and to protect the patient’s rights.
A program of measures to reduce hospital infection must be
based on epidemiological analysis of recorded events in the
search for common causes and preventable factors.
Organized surveillance and control requires a ratio of
one infection control practitioner per 250 acute care beds,
787
a trained hospital epidemiologist, and routine reporting of
wound infections to practicing surgeons (CDC, Hospital
Infection Program). Computer-aided medication dispens-
ing, as well as automated and other safety systems are
critical elements in minimizing morbidity and mortality
resulting from preventable human errors. In response to
the high frequency and cost of medical litigation, many
states in the USA have enacted legislation to restrict court
awards for medical negligence. Proposals for alternatives
to the tort system of medical malpractice compensation
include arbitration and mediation, an administrative sys-
tem similar to that used for workers’ compensation, and
a no-fault system of compensation, such as exists in New
Zealand, Sweden, and Finland. In a no-fault system the
complainant need not prove negligence on the part of the
provider, but only that he or she suffered an adverse event
which is compensable at standard rates depending on the
degree of disability. In the USA, federal legislation pro-
vides compensation for vaccine injuries, and three states
have enacted restricted no-fault systems for birth-related
neurological injuries.
In addition, there is greater emphasis on the adoption of
failsafe mechanisms, such as introducing warning systems
in anesthesia machines to alert the anesthetist if oxygen
flow in the patient’s tubing falls below a safe point. This
system was tested in Boston hospitals and found to reduce
adverse anesthetic events to zero cases over a 3-year period.
Vitamin K injection was made mandatory for all newborns
in New York State, as was already the case in some other
states, when a study showed deaths from hemorrhagic
disease of the newborn in cases where vitamin K was not
administered.
Inappropriate medical practice patterns are an equal, or
even larger problem for health systems. Comparisons of
surgical rates within the USA for coronary bypass proce-
dures, hysterectomies, and caesarean sections show wide
variation between different areas of the country. The costs
of excess surgery not only are economically wasteful but
also involve risks for the patient from the surgery itself
or anesthesia mishaps, infection, pain, and discomfort,
with legal and ethical questions of unwarranted interven-
tions not for the benefit of the patient. Health systems are
increasingly required to evaluate and control excess surgi-
cal, investigative, or other medical procedures, not only for
financial reasons but also for protection against litigation
and infringement of patients’ rights.
Licensure and Certification
The requirements that society establishes for allowing an
individual to practice medicine, and any health profession,
are vital to maintaining and improving the quality of care
(see Chapter 14). These standards require defining the train-
ing and experience needed by the individual, examination
788
procedures, and recognition for continued education and
maintenance of competence. This requires a statutory base
and national bodies operating under a national authority, sep-
arate from the agency operating the health system services.
Separation of licensing from operation of the health service
is essential in maintaining high professional standards.
The licensing authority is accountable to the state and
the public. In some cases, this function is delegated to self-
regulating professional bodies. In Canada, the licensing of
the medical profession and specialty recognition are carried
out by the medical profession with self-regulation. In the
UK, medical licensing is by a state-appointed board and in
the USA by state boards.
Medical schools, postgraduate training programs,
and fellowships are all subject to periodic comprehensive
assessments. Institutions that fail to meet the standard may
have funding or licensure suspended until they have per-
formed adequate remediation.
Health Facility Accreditation
Hospital accreditation in North America is by a voluntary
grouping of professional associations, including the Cana-
dian and American Colleges of Physicians and Surgeons,
the hospital associations, and the Colleges of Nurses. The
Joint Commission, originally operating in both Canada and
the USA, carries out regular inspections of hospitals. In Can-
ada, other organizations including the federal Department of
Health, provincial ministries of health, the Canadian Diabetes
Association, the Public Health Association, and the Standards
Council of Canada participate in the Joint Commission as
observers. Initially focusing on acute care hospitals, accredi-
tation has been gradually extended to cover special hospi-
tals, long-term facilities, home care programs, public health
departments, and ambulatory care services.
Health facility accreditation is a systematic, multidisci-
plinary inspection of the physical and organizational struc-
ture of the facility or program and the functioning of its
component parts. Factors measured include staff qualifica-
tions, facilities, organization, record keeping, and continu-
ing education of staff.
The process of accreditation requires a request for
accreditation from the board of governors of the hospital
or health facility, implying acceptance of the standards
of the commission. The accreditation process includes a
self-assessment, an on-site survey, and follow-up action
for correction of deficits and improvements. The com-
mission is invited to conduct a survey, and resurvey as it
sees fit. The hospital pays a fee and commits itself to pro-
vide all data requested and to cooperate with the site visit.
The commission issues a confidential report, giving the
accreditation rating and interim statement of deficiencies,
and requests progress reports in correcting deficiencies. It
is also empowered to carry out follow-up inspections and
The New Public Health
resurveys. Box 15.4 lists the areas of a large community
or teaching hospital, regional health authorities, hospitals,
and community-based programs and services, from both
private and public sectors, not only in Canada but around
the world.
The assessment survey examines the goals and objec-
tives of the organization and its administration, the direc-
tion and staffing of the facility, policies, and procedures.
Review includes medical staff organization, credentials and
review procedures, clinical privileges, selection of depart-
ment chairpersons and their responsibilities, standing com-
mittees, schedule of meetings, bylaws, and the role of the
governing board of the hospital. The presence and nature of
quality assurance organization, records review procedures,
and continuing educations are assessed. The quality of
clinical records is assessed by examination of charts for the
completeness of histories and documentation of the course
of the hospital stay including laboratory reports.
Each section of the program being accredited is assessed
in the following categories:
l statement of purposes, goals, and objectives
l organization and administration
l human and physical resources
l orientation, staff development, and continuing education
l patient care
l quality assurance.
These categories are also used in the programs covered
by the contracts between Accreditation Canada, formerly
the Canadian Council on Health Services Accreditation
(CCHSA), and other health and social service agencies.
Hospital accreditation was established in the UK and
Australia in the 1980s and is attracting interest in other
countries seeking ways to maintain and promote standards.
The procedure for accreditation of hospitals is still volun-
tary in Canada, but in effect has become universal for hos-
pitals of medium and large size (over 75 beds) and common
for smaller hospitals. It is seen as advantageous for the gov-
erning board and the community and also for the medical
staff in terms of medicolegal protection. In the USA, hospi-
tal accreditation has become virtually universal since pay-
ment for federally funded health insurance (Medicare and
Medicaid) beneficiaries is not allowed for non-accredited
hospitals, and many private insurers make this requirement
as well. In some states, accreditation is mandatory for all
hospitals.
Since the 1990s, CCHSA’s accreditation program has
expanded to cover a diversity of health care and service
areas, through contract arrangements with independent non-
hospital facilities such as highly specialized programs as well
as community health and social service organizations. In 2006,
CCHSA introduced standards for child welfare, hospice, pal-
liative and end-of-life care facilities, prison facilities, biomedi-
cal laboratories, and supplementary criteria for telehealth. In
789Chapter 15 Health Technology, Quality, Law, and Ethics
l System wide:
– Governance
– Infection prevention and control
– Leadership
– Leadership for aboriginal health services
– Leadership for assisted reproductive technology
– Leadership for primary care
– Managing medications
l Population based:
– Cancer populations
– Child and youth populations
– Maternal/child populations
– Mental health populations
– Populations with chronic conditions
– Public health services
l Service excellence:
– Acquired brain injury services
– Ambulatory care services
– Ambulatory systemic cancer therapy services
– Assisted reproductive technology
– Case management services
– Child welfare services
– Community health services
– Community-based mental health services and support
standards
– Critical care services
– Developmental disabilities services
– Diagnostic imaging services
– Emergency department services
– Health care staffing services
– Home care and support services
– Hospice palliative and end-of-life services
– Independent medical/surgical facilities
– Laboratory and blood services
– Long-term care services
– Medical imaging centers
– Medicine services
– Mental health services
– Obstetrics services
– Operating rooms
– Organ and tissue donation standards for deceased
donors
– Organ and tissue transplant
– Organ donation standards for living donors
– Point-of-care testing
– Primary care services
– Rehabilitation services
– Reprocessing and sterilization of reusable medical
devices
– Spinal cord injury acute services
– Spinal cord injury rehabilitation services
– Substance abuse and problem gambling services
– Surgical care services
– Telehealth services
l Service distinction:
– Acute stroke services
– Audit tool for reprocessing and sterilization of reusable
medical devices
– Inpatient stroke rehabilitation services
– Providing an integrated system of services to people
with stroke
Source: Accreditation Canada. Available at: http://www.accreditation.ca/
en/content.aspx?pageid=54 [Accessed 14 December 2012].
BOX 15.4 Accreditation Canada Standards
2008 CCHSA officially became Accreditation Canada, pro-
viding services to other countries. The ever-changing health
and social environment now accommodates specialized needs
in a diversity of service areas as an adjunct to the hospital
accreditation process. Examples are shown in Box 15.5.
Licensing and regulation of health facilities are a gov-
ernment responsibility, but an independent accreditation
authority has advantages. The national authority may fail to
monitor its own facilities with the diligence or objectivity
needed, and there may be a conflict of interest. Where there
is a national system of organization, distinct departmen-
talization of the operating and certification functions may
provide a greater measure of objectivity. Assistance from
countries experienced in voluntary accreditation can help to
establish accreditation mechanisms and provide technical
and professional support to countries wishing to establish
such programs.
In the current period of transition from central to decen-
tralized management of health services in many countries,
health facilities are being transferred from government
operation to independent operation as not-for-profit or
even for-profit facilities. Present methods of regulation by
national or state levels of government will require review
as decentralization and privatization take place. Regulation
by governmental authorities and non-governmental pro-
fessional bodies is mutually complementary in promoting
accountability, standards, and quality of services.
Peer Review
A large part of the work of clinical and departmental man-
agers in hospitals or other care settings relates to qual-
ity assurance. A major method of improving quality in a
health program is through peer review by which the staff
organizes systematic review of cases and records, using
statistics on performance indicators. In hospitals, this
includes review of deaths, maternal mortality and infant
mortality cases, surgical rates, complications following
surgery, and infection rates. Medical records and com-
puter information systems permit users to review records
http://www.accreditation.ca/en/content.aspx?pageid=54
http://www.accreditation.ca/en/content.aspx?pageid=54
790
BOX 15.5 Accreditation Canada International
Accreditation Program
The Canadian health services accreditation program began
in 1917 in conjunction with the American College of
Surgeons (ACS) with a hospital standardization program. The
first Minimum Standard for Hospitals developed require-
ments of just one page. In 1918, on-site inspections of hos-
pitals began, with 89 of 692 hospitals surveyed meeting the
requirements of the Minimum Standard. In 1926, the first
Standards Manual was issued.
In 1951, the American College of Physicians, the
American Hospital Association, the American Medical
Association, and the Canadian Medical Association joined
with the ACS to create the Joint Commission on Accreditation
of Hospitals (JCAH). It is an independent, not-for-profit orga-
nization whose purpose is to provide voluntary accredita-
tion. In 1953, the Canadian Hospital Association (now the
Canadian Healthcare Association), the Canadian Medical
Association, the Royal College of Physicians and Surgeons,
and l’Association des Médecins de Langue Française du
Canada established the Canadian Commission on Hospital
Accreditation. The Commission’s purpose was to create a
Canadian program for hospital accreditation, and in 1958
the Canadian Council on Hospital Accreditation (CCHSA)
was incorporated.
In 2008 CCHSA became Accreditation Canada
International. The accreditation program is used by all
types of health facilities, from large and complex hos-
pitals, to health systems, community health organiza-
tions, and residences providing long-term care. Its scope
includes a wide range of programs, including standards
on child welfare, hospice palliative and end-of-life care,
biomedical laboratory services, blood banks, and supple-
mentary criteria for Telehealth. The accreditation program
covers a diversity of health care and service areas, service
programs for brain injury, ambulatory care, assisted repro-
ductive technology – clinical and laboratory services,
Canadian Forces health services, cancer agencies, child
welfare organizations, First Nations and Inuit addictions
and community health services, the Federal Department
of Veterans’ Affairs, substance abuse and problem gam-
bling treatment services.
The accreditation service is on a contract basis with
specialized health programs, other federal government
departments, for-profit health facilities, and community
organizations across the provinces.
Accreditation Canada International works with other
countries to develop national accreditation programs for
their countries, and launched its first international program
for acute care, primary care, ambulatory care, and clinical
laboratories in 2010.
Source: Accreditation Canada International. Available at: http://www.
internationalaccreditation.ca/Accreditation/AccreditationProgram.aspx
[Accessed 12 September 2012].
The New Public Health
by diagnosis. These records can be utilized to assess other
events in hospitals, such as time from admission to sur-
gery, lengths of stay by diagnosis, response to abnormal
laboratory findings, and many other indicators of the pro-
cess of care. Obstetric departments can review the fre-
quency of and criteria for caesarean section deliveries.
Surgical departments review their appendectomy rates to
separate pathological findings from normal appendices.
Organized peer review has also been called medical audit
and essentially describes methods of self-policing and
education to learn from mistakes and experience and to
improve the quality of care.
In 1972, an amendment to the US Social Security Act
required hospitals and long-term care facilities to moni-
tor the quality of care given to Medicare and Medicaid
patients through professional standards review organiza-
tions (PSROs). These were medical audit committees with
specified tasks to conduct utilization review, medical care
evaluation, and profile analysis of physician or institutional
performance compared to accepted standards of the medical
community. In 1982, peer review organizations (PROs) were
created by federal statutes to replace PSROs. The PROs are
non-profit corporations, staffed by physicians and nurses,
to review medical necessity, quality, and appropriate level
of care under the Medicare and Medicaid programs. The
Centers for Medicaid and Medicare Services have an Office
of Clinical Standards to conduct surveys, provide certifica-
tion, and develop best practices guidelines, in a health care
quality improvement program (HCQIP).
Hospitals have departmental clinical meetings, adverse
incident or outcome committees, mortality rounds, and
clinical pathology conferences to help staff to evaluate and
learn from difficult cases. The presence of functioning peer
review mechanisms indicates that quality is of concern to
the professional and administrative network, raising the
consumer’s confidence in the system.
Maternal mortality committees have been widely used
to assess preventable factors in deaths related to maternity
and to point out areas of needed improvement in services.
Identification of high-risk pregnancies emerged from this
process and has become an important part of prenatal care.
Infant mortality reviews by professional groups can simi-
larly demonstrate areas of needed improvement in services.
Death rounds are held to review cases of death following
surgery or soon after admission, or “incidents”, such as
inappropriate medication given in error.
The successive waves of peer review initiatives in
the USA represent attempts by the federal government
to establish mandatory quality of care review by profes-
sional peers for facilities providing care to Medicare and
Medicaid patients. The concept of requiring standards of
care review has probably contributed to a greater aware-
ness of the accountability of hospital-based practice. Fre-
quent litigation may have contributed more to the sense that
http://www.internationalaccreditation.ca/Accreditation/AccreditationProgram.aspx
http://www.internationalaccreditation.ca/Accreditation/AccreditationProgram.aspx
Chapter 15 Health Technology, Quality, Law, and Ethics
the physician is accountable for services and outcomes of
care. PROs are a form of quality regulation that represent a
commitment by funding agencies to accountability in care
systems and to identification of organizational and admin-
istrative weaknesses in health care generally and not only
in hospitals. The generation of US physicians and health
systems managers trained since the 1970s accepts peer
review as an integral part of health services. Other coun-
tries use this kind of mechanism to maintain and promote
quality of care.
Tracer Conditions
Tracer conditions are common medical conditions (or pro-
cedures) for which diagnostic criteria are well established
and clear, there are effective preventions or treatments, and
a lack of treatment can cause significant harm to the patient.
Examples of tracer conditions include otitis media, appen-
dectomy, caesarean section, and hysterectomy. These con-
ditions, if evaluated in terms of incidence and actual chart
review, can provide useful insights into departmental medi-
cal standards. Incident reports by nursing staff and nosoco-
mial infections are examples of the functioning of the tracer
condition concept.
Incident reports in hospitals are designed to determine
the causes of errors, so that remedial action can be taken
and similar events prevented. Tracer condition studies have
become such an accepted part of modern health manage-
ment that the absence of an organized review system could
be considered a serious structural flaw in a health service,
requiring remedial action.
Setting Standards
Standards recommended by independent professional
organizations or by advisory committees appointed by
ministries of health can play important roles in defining
standards of care for specified conditions. In addition,
organized professional bodies can issue practice guidelines
or help governments or health care agencies to develop
standards or algorithms for management of specific topics
and conditions.
Specifying standards for preventive care, such as for
infants and adults, assists local health authorities in plan-
ning and evaluating services. The American Academy of
Pediatrics (AAP) has an extensive professional committee
structure that publishes periodic guidelines for pediatricians
on a wide variety of infant and child topics including nutri-
tion, immunization, prevention of anemia and lead toxicity,
child safety, and school health. Mandatory preventive care
for newborns includes eye care and vitamin K injection in
the USA (see Chapter 6). Mandatory immunization require-
ments for school entry and for health care personnel are dis-
cussed in Chapter 4.
791
The American Public Health Association (APHA) pub-
lishes the Control of Communicable Diseases Manual, now
in its nineteenth edition (2008). It is the authoritative US
manual on this topic. The AAP’s Red Book on infectious dis-
eases is used across North America by pediatricians in clini-
cal practice. These organizations and their counterparts in
obstetrics and many other clinical fields directly relevant to
public health continually update practitioners and policy per-
sonnel in the “state of the art” or “gold standard”, discussed
previously. This constitutes a professional self-guidance sys-
tem in standards. Managed care and other health provider
systems also issue guidelines for member practitioners that
serve to maintain standards of service.
The wide use of treatment protocols and scoring systems
in hospital medicine helps to define standards of care in a
measurable way. The Apgar score for rating newborn sta-
tus has been a standard in hospitals worldwide for decades,
helping to standardize infant assessment and care. The
APACHE (Acute Physiology And Chronic Health Evalua-
tion) scoring system is used widely to assess the chances
of survival of patients admitted to intensive care units and
to compare outcomes, for example, between teaching hos-
pitals and community hospitals. It is also used in assessing
patient outcomes with different modes of treatment. Scor-
ing systems are also used in community health care, as in
risk scoring for pregnancy care (see Chapter 6).
Algorithms and Clinical Guidelines
Algorithms are decision trees or a systematic series of deci-
sions based on the outcomes of previous decisions, tests, or
findings. Derived from operations research, this approach
applied to medicine identifies all available choices (e.g.,
exposed versus non-exposed) and follow-up decisions
based on findings from each previous option substantiated
by observation. It is often presented graphically like the
branches of a tree, showing the alternatives and subsequent
decisions to be made.
A clinical algorithm is a systematic process defining
a sequence of alternative, logical steps depending on out-
comes of previous ones, incorporating clinical, laboratory,
and epidemiological information, applied to maximize ben-
efits and minimize risks for the patient. It gives the provider
a review of the relevant literature and recommended stan-
dards of practice on a particular topic for preventive care
or case management. These guidelines are usually arrived
at by consensus of multidisciplinary working groups taking
into account published studies on the topic. The guidelines
may suggest that some procedures should not be carried out
routinely.
Clinical guidelines are meant to establish accepted stan-
dards of care and may have important economic implica-
tions. Medical Letter, published by the Consumers’ Union,
is a long-standing and useful publication that reviews
The New Public Health792
TABLE 15.5 Adult Health Maintenance Checklist by Age Group
Age (years)
Procedure 20–39 40–64 65+
Checkup visit Every 3 years Every 2 years Annually
Cholesterol With checkups With checkups With checkups
Fecal occult blood Age 40–49 if high risk Annually Annually
Clinical breast examination Every 1–3 years Annuallya Annuallya
Mammography Baseline age 35 Age 40–49, every 1–2 years Over 70, every 2 years
Pelvic examination Every 1–3 years Every 1–3 years Every 1–3 years
Pap smear From age 21-29 every 3 years;
from 30-65 every 5 years with
HPV DNA test
From age 30–65 every 5 years
with HPV DNA test
If previously negative,
may stop 3 years
Colonoscopy No From age 40 for those with family
history of colon cancer or polyps.
After age 50, every 3–5 years
After age 50, every 3–5 years
Prostate and PSA Immunizations No Annuallya Annuallya
Tetanus–diphtheria Every 10 years Every 10 years Every 10 years
Pneumococcal pneumonia For high risk For high risk Every 6 years
Influenza For high risk For high risk Annually
Skin cancer Annuallya Annuallya Annuallya
Bladder cancer Annual routine urinalysis Annual routine urinalysis Annual routine urinalysis
Lung cancer Routine examinationb Routine examinationb Routine examinationb
Testicular cancer Routine examinationb Routine examinationb Routine examinationb
Oral cancer Routine examinationb Routine examinationb Routine examinationb
Ovarian cancer Routine examinationb Routine examinationb Routine examinationb
Pancreatic cancer Routine examinationb Routine examinationb Routine examinationb
Routine vitamin supplements Routineb Routineb Routineb
Note:
PSA = prostate-specific antigen.
Agency for Healthcare Research and Quality. Rockville, MD: AHRQ. http://www.ahrq.gov [Accessed 13 September 2012].
aInconclusive
bnegative recommendation. The topics are under continuing review, and recommendations are in some cases left to the opinion of the provider as the current
cumulative evidence is not affirmative, e.g., clinical breast examination annually or breast self-examination.
Sources: US Preventive Services Task Force Ratings: Strength of recommendations and quality of evidence. guide to clinical preventive services. 3rd ed.
Periodic updates, 2000–2003. Available at: http://www.uspreventiveservicestaskforce.org/3rduspstf/ratings.htm [Accessed 13 September 2012].
therapeutic issues of everyday medical practice and the rele-
vant studies. It represents a balanced, updated view of medi-
cal practice and summaries of current literature, reviewed
by respected, experienced, and competent medical authori-
ties. Clinical practice guidelines are produced by hundreds
of professional, medical, and governmental agencies in
order to standardize and improve medical care.
Clinical and preventive care guidelines are helpful
in clinical practice and in preventive medicine. They are
increasingly used in managed care environments to assure
standards, quality of care, and cost-effectiveness as well as
legal protection. Guidelines for preventive medicine and
public health practice are also part of the process of pro-
moting the quality of individual and community health,
as discussed in Chapter 11. Annual revision of the infant
immunization program, discussed in Chapter 4, is a prime
example, as is the set of guidelines for preventive care for
adult health maintenance in Table 15.5.
The issue of application of current scientific knowledge
for population health is a continuing struggle for recognition
of the prime importance of health promotion and preventive
care for health of a population. The selection of priorities in
use of resources is vital especially in the many developing
countries that are in various stages of economic development,
or which have abundant income from natural resources such
as oil and minerals. Implementation of programs designed to
achieve the MDGs can help to serve this purpose.
Public health standards and clinical practice guidelines
are an increasing part of quality improvement. It is impor-
tant, however, that they are developed as best practices and
http://www.ahrq.gov
http://www.uspreventiveservicestaskforce.org/3rduspstf/ratings.htm
Chapter 15 Health Technology, Quality, Law, and Ethics
influenced as little as possible by commercial interests of
drug or vaccine manufacturers. The proliferation of such
guidelines by health authorities or professional associations
of the USA, the UK, Canada, Australia, and other countries
indicates a wide consensus on the importance of such writ-
ten standards, guidelines, or “best practice” statements. The
recommended childhood immunization program put for-
ward annually by the CDC in conjunction with the AAP and
other professional organizations is an example of such best
practices and is accepted by health insurers and providers as
the gold standard in this field. The concept of promotion of
quality in health care and the adoption of current scientific
standards are global issues and an integral part of the New
Public Health (Box 15.6).
BOX 15.6 EISC: The Excellence in Science Committee
of the Centers for Disease Control and Prevention
The Excellence in Science Committee (EISC) promotes the
Centers for Disease Control and Prevention’s (CDC’s) sci-
entific infrastructure and facilitates communication and
collaboration that enhance scientific areas and activities
needed for state-of-the-art conduct of science. EISC serves as
a consulting body for science-related issues and makes rec-
ommendations to the CDC to foster, support, and protect an
environment for the promotion of scientific integrity, quality
assurance, and the rapid dissemination of scientific innova-
tions, technology, and information, with the ultimate goal of
improving public health.
EISC’s specific functions include:
l promoting and protecting the scientific infrastructure
l providing a forum for information exchange among
administration, directors for science, and liaison work-
ing members/groups
l communicating science-related issues to the CDC and
related scientists
l promoting professional development and training
l recognizing and rewarding quality science
l acting as an advocate for scientific resources
l identifying and disseminating new information, e.g.,
new statistical/epidemiological techniques or new scien-
tific technologies
l developing, revising, and promoting the implementation
of cross-cutting scientific policies and procedures
l serving as a consulting body for science-related issues
and making recommendations to the CDC
l fostering the development of methods for assessing and
monitoring:
– the environment for quality science and qualitative
and quantitative scientific output within CDC and
related organizations
– the impact of CDC science on public health.
Source: Adapted from Centers for Disease Control and Prevention.
Science coordination and innovation [updated 3 November 2011].
Available at: http://www.cdc.gov/od/science/excellence/ [Accessed 13
September 2012].
793
The Canadian Province of Saskatchewan Health Ser-
vices Utilization and Research Commission publishes peri-
odic reports presenting consensus positions of panels of
medical faculty, clinical specialists in pathology and physi-
cal medicine, and public health specialists in nutrition, com-
munity health, and epidemiology. Its reports are circulated
widely and serve to update medical practitioners, reduce
unnecessary testing, promote appropriate use of laboratory
and other diagnostic procedures, and provide standards of
care for individual patients and community services, such
as long-term care facilities and home health agencies.
The Canadian Medical Association issued its Handbook
on Clinical Practice Guidelines in 2007, based on a system-
atic review of the literature, interviews of key professionals,
consensus conferences, and continuing evaluation of both
process and content of such guidelines. The Guideline Inter-
national Network (GIN) Fourth International Conference,
held in Toronto in 2007, involved experts in national and
international practice guidelines from 31 countries to share
experience and concepts in this ongoing field. The GIN
library contains more than 6600 (by October 2012) guide-
lines, evidence reports and related documents, developed or
endorsed by GIN member organizations (GIN, 2012).
An Institute for Clinical Evaluation (ICES) organization
at the University of Toronto, established in 1992 with core
funding provided by Ontario’s Ministry of Health and Long
Term Care, is mandated to conduct research that contributes
to the effectiveness, quality, equity, and efficiency of health
care and health services in Ontario. ICES uses an interdis-
ciplinary research approach to health care, health services,
and health policy.
The American College of Cardiology (ACC) provides a
framework of evidence-based clinical statements and guide-
lines developed by leaders in the field of cardiovascular med-
icine with continuing adoption of new scientific information
and experience in many aspects of this field (ACC, 2012).
Many professional organizations such as the AAP, American
Congress of Obstetricians and Gynecologists (ACOG), UK
Faculty of Public Health, and European Society of Cardiol-
ogy produce clinical guidelines which are updated regularly
to provide physicians and health systems managers with
current consensus on state-of-the-art standards, such as the
European Society of Hypertension Guidelines released in
2013 (i.e., less than 140 mm. systolic for all).
The US Health Care Financing Administration (HCFA),
Center for Medicare & Medicaid Services (CMS), and
National Institutes of Health (NIH) have consensus pro-
grams to develop guidelines that are widely disseminated
and set standards of practice. In 1977, the NIH issued its
first consensus paper on breast screening for cancer, and
this has been followed by many other topics each year
since. The AHRQ also produces research related to efficacy
of current and new practices and training material to pro-
mote their diffusion across the US health system. Cochrane
http://www.cdc.gov/od/science/excellence/
794
reviews and the Cochrane Library provide high standards of
literature reviews and meta-analysis on many topics which
serve to guide practitioners and policy makers in current
standards. The US Healthy People 2020 project provides
gold standards for preventive care which serve clinicians,
public health practitioners, and health planner standards
for their work. Evidence-based consensus guidelines were
issued on the following topics: breast cancer screening
for women aged 40–49, interventions to prevent HIV risk
behavior, management of hepatitis C, genetic testing for
cystic fibrosis, acupuncture, and effective medical treat-
ment for heroin addiction.
Clinical guidelines are increasingly being promoted
by professional, governmental, and managed care orga-
nizations with the purpose of promoting rational use of
health care resources and at the same time promoting
standards of care to incorporate good standards of clinical
practice. Clinical practice guidelines are now common in
the practice of primary care, mental health, and clinical
specialties. The University of Southern California’s list
of clinical guidelines website (http://medicine.ucsf.edu/)
provides access to hundreds of websites for such practice
guidelines.
Clinical guidelines provide practicing doctors, peer
review committees, health care managers, managed care
companies, governmental bodies, and professional organi-
zations with channels to set standards of practice and expec-
tations of care standards. Legal aspects of health care also
increasingly recognize the importance of clinical guidelines
where committees of appropriate medical professionals
convene and set out average or minimum standards of care
for defined clinical entities. Thus, peer-reviewed guide-
lines set an appropriate standard (a silver if not a gold stan-
dard) for judging malpractice or adequate practice. Clinical
guidelines should be under periodic review and subject to
critical discussion and updating using the Cochrane review
methods of literature review and analysis. Promotion by
advocacy or special interest groups can be constructive,
but the influence of drug companies can be insidious and
reduce the professional objectivity of such reviews and their
recommendations, a concern that must be carefully moni-
tored and continuously kept in mind as a potential compro-
mising bias.
The AAP produces policy statements, practice param-
eters, and model bills which have a wide distribution and
influence; they are published in the academy’s journal, Pedi-
atrics. The AAP clinical practice guidelines issued include
diagnosis and treatment of urinary tract infection in febrile
infants and young children, long-term treatment of the child
with simple febrile seizures, management of acute gastroen-
teritis in young children, management of otitis media with
effusion in young children, and others. The policy state-
ments of the AAP cover a wide range of topics including
use of bicycle helmets, 55 mile per hour maximum speed
The New Public Health
limits, folic acid for the prevention of neural tube defects,
and ethics in the care of critically ill infants and children.
AAP guidelines are valid for 5 years only and are reissued
or reconfirmed in order to keep up to date and to incorporate
new or revised knowledge into practice standards.
Empirically derived, peer-reviewed, regularly updated
guidelines have become an appropriate standard for prac-
tice and for judging malpractice, as well as balancing qual-
ity and cost-effectiveness. Clinical guidelines may become
restrictive, but they help to reduce practice by whim and
unsubstantiated belief to improve the quality of care over-
all. In large health care organizations they provide a basis
for continuing education for staff and advancement of stan-
dards of the organization.
The Community Guide produced by the CDC provides
an excellent source of evidence-based advice for community
programs. It serves the needs of public health professionals,
health care providers, legislators and policy makers, research-
ers, community-based organizations, employer–employee
groups, and other purchasers of health services. The guide
covers a wide range of health issues including alcohol, can-
cer, diabetes, mental health, motor vehicle safety, nutrition
and obesity, oral health, physical activity, pregnancy, sexual
behavior, social environment, substance abuse, tobacco, vac-
cines, violence, and workplace health issues.
In 1999, the UK National Health Service (NHS) estab-
lished the National Institute for Clinical Excellence (NICE)
as an independent organization to provide guidelines for
public health, health technologies, and clinical practice
guidelines for specific conditions. The Health Development
Agency of the NHS was included in the NICE organization
in 2005. Now called the National Institute for Health and
Care Excellence, NICE publishes guidelines that provide a
helpful basis for clinical practice and public health as well
as other areas in the NHS to update the services provided.
Topics for public health include smoking and tobacco con-
trol, diet and obesity, exercise and physical activity, sexual
and mental health, and alcohol.
ORGANIZATION OF CARE
Administrative and financing systems are essential ele-
ments of quality assurance. They can be designed to pro-
mote standards of care and to reduce fiscal incentives that
foster excess supply and overservicing. The organization
of financing health care has important implications for
quality, technology, and ethical issues in the New Public
Health.
Diagnosis-Related Groups
DRGs, discussed extensively in Chapter 11, were developed
in the 1960s as an alternative way of paying for hospital care
in order to encourage shortened lengths of stay. Experience
http://medicine.ucsf.edu/
Chapter 15 Health Technology, Quality, Law, and Ethics
with payment by days of care (per diem) showed that it pro-
moted unnecessary, lengthy, and potentially dangerous use
of hospital care, an important factor in the rapid escalation
of costs in the health system. DRGs were adopted for pay-
ment for Medicare beneficiaries in the USA in 1983 and
later became the standard method of payment for all insur-
ance systems.
In the DRG system the insurer pays the provider hos-
pital for a procedure or diagnosis rather than the number
of days of stay in hospital. This has led to a large reduc-
tion in hospital days of care and a remarkable growth in the
number of surgical procedures done on an outpatient basis.
Since the introduction of DRGs, outpatient surgical proce-
dures have grown from less than one-fifth to more than half
of inpatient surgical cases. Outpatient surgery is safer for
the patient and less costly to the insurer. DRGs have gradu-
ally been adopted as a case payment system for reimbursing
hospitals in most developed countries.
The DRG system is widely considered to promote qual-
ity of care as an active process focusing on quickly address-
ing the diagnosis and management of the patient with rapid
mobilization of treatment and return home. Critics of this
system allege that DRGs encourage inappropriate early
discharge of patients before optimal patient education and
follow-up care have been provided, but long length of hos-
pital stay has not been shown to improve patient outcomes.
Critics also suggest that this may promote altering diagno-
ses to higher cost units of service. Others think that DRGs,
by reducing length of stay, have turned hospitals into inten-
sive care units with ultra-sick patients. Despite these issues,
the trend towards short hospital stays and newer approaches
to active treatment seems to be compatible with better care
and improved outcomes, according to some measures. The
rapid decline in mortality rates from coronary heart disease
is thought to be due in large part to the activist treatment
approach, with lengths of stay of 1 week or less for acute
myocardial infarction compared to 6 weeks on average up
to the 1970s.
Managed Care
Managed care systems developed in the USA in response
to rapid cost escalation for health care and the successful
experience of HMOs. Managed care is based on the con-
cepts of resource management, and quality assurance with
rationalized use of technology. The system developed over
time with checks and balances to provide comprehensive
care at lower cost than traditional fee-for-service systems
by discouraging excessive utilization without compromis-
ing quality of service. Managed care systems include tradi-
tional HMOs and various other organizations which employ
physicians or are made up of independent physicians work-
ing together who own or contract for hospital services (see
Chapter 10).
795
HMOs, both for-profit and not-for-profit, and man-
aged care itself, have been widely criticized as excessively
limiting patient access to appropriate care in the interest
of cost containment. The 2010 PPACA (Obamacare) is
promoting development of newer innovations including
patient-centered medical homes (PCMH), accountable
care organizations (ACOs), and population health man-
agement systems (PHMSs), and early evidence shows that
these models are quality management approaches for inte-
grated primary and hospital patient care (see Chapter 11).
Obamacare is a highly politicized and much debated topic
in the USA; it seems likely to make a very big difference
in coverage and fair practices of insurance with lower
costs of private insurance.
District health systems in the UK, the Scandinavian
countries, and the post-Soviet model of health care incor-
porate organizational and financial linkage between care
systems and funding from tax sources. HMOs, sick benefit
funds, and district health systems provide both prepayment
and health services. Even in traditional private health insur-
ance systems, the insurer is increasingly taking on the role
of regulating reimbursement for medical services in order to
contain costs and curb abuses by providers. In this context,
emphasis is placed on maintaining health, preventive care,
and financial incentives to efficiency in overall care. Clini-
cal indications, utilization review, and organizational and
professional standards are now becoming accepted parts of
the health insurance milieu.
The competition between hospitals for referrals from
managed care plans in the USA has created a market
situation in which a high proportion of hospital beds are
empty, and in which mergers or closures of hospitals
are common. Closures or reductions in hospital bed sup-
ply are also occurring in the UK and in most industrial-
ized countries of Europe.
PERFORMANCE INDICATORS
Performance indicators are measures such as morbidity,
mortality, functional status, or immunization rates in a com-
munity, used to monitor the functioning of a health service.
Routinely collected statistics are analyzed to compare per-
formance against objectives, help monitor efficiency and
effectiveness, point out problem areas within the service,
and plan new health programs. This method is based on the
use of the concept of management-by-objectives in health
administration to promote achievement of national health
targets.
The UK has a strong tradition of mapping diseases as a
basis of epidemiological analysis and has applied this strat-
egy to mapping of performance indicators to assess health
care performance. The UK financing system is based on
capitation adjusted by standardized mortality rates on the
premise that mortality rates standardized and compared to
796
the national average serve as indicators of need. In this way,
the approach helps to promote equitable funding among
wealthy and poorer regions of the country, and thereby
improve services in areas of greater need.
Performance indicators were introduced into the NHS
during reforms of the late 1980s, providing a series of
outcome or performance measures that are used to adjust
payments allocated on a per capita basis to district health
authorities. These authorities can be penalized for low rates
of immunization, whereas general practitioners receive
incentive payments for full immunization coverage. The
result was a rapid improvement in immunization coverage
of infants and children compared to rates in the previous
decade. Incentive payments in many countries encourage
women to go to hospitals for delivery or to attend prenatal
care by making social maternity grants conditional on seek-
ing care.
Use of performance indicators requires the development
of health information systems with district health profiles
to provide ongoing monitoring of health indicators in a dis-
trict, compared to regional and national rates and targets.
Health profiles help to establish and monitor the prevalence
of chronic disease and measure the impact of health ser-
vices. This enables the study of the performance of preven-
tive and curative services, such as managing hypertension
to reduce the incidence of strokes and related conditions.
There are criticisms of performance indicators alleging a
potential for manipulation and abuse of health intervention
measures when the financial incentives are used for a spe-
cific activity. However, financial incentives are part of the
DRG system and have been successfully used in the UK to
improve vaccination coverage and implementation of other
preventive health practices by family physicians. In Israel in
2007, payments to hospitals provided a bonus for surgical
interventions for hip fracture within 48 hours of the event,
resulting in a marked rise in early intervention and a reduc-
tion in mortality from hip fractures.
CONSUMERISM AND QUALITY
With decentralization and the growth of managed care,
health systems must increase their attention to the attitudes
of the consumer. Quality is, in part, how the client per-
ceives the system, and how the system meets client needs
in an acceptable manner, where privacy, dignity, the right to
know, and the right to a defined set of services are protected.
However, the rights of the client are not unlimited. A public
or private health plan has the duty to manage the basket of
services responsibly, which includes limitations such as in
access to specialist services.
Patients’ rights and consumer protection in health care
often (but not always) include the right to select and change
a health care provider, as well as the right to receive high-
quality care for a designated range of services. The UK NHS
The New Public Health
issued a patient’s Charter of Rights during the 1990s, which
is perhaps idealistic and may not be actualized in practice,
but still outlines an ideal of value both for practical applica-
tion and for legal rights. The consumer’s formal protection
includes the right to complain and to seek redress of grievance
and compensation for injury suffered from neglect or incom-
petent care (Box 15.7). In North America and Europe, there
are at least four models of defining the rights of patients: the
paternalistic model, the informative model, the interpretive
model, and the deliberative model (WHO, 2012). Many new
charters have been established such as data protection, end-
of-life care, mental health, access to health services, quality
of care and care giving environment, nationally approved
treatments, drugs and programs, respect, consent and confi-
dentiality specific to the UK and in Canada, New Zealand
and other countries. The new US federal Affordable Care Act
of 2010 (PPACA) includes a large element of patient’s rights
protection, as discussed in Chapter 10.
The patient or consumer of health care needs to be
informed and conscious of health care costs if efforts to
restrain cost increases are to be effective. Public attitudes
are vital in terms of self-care, demands on the health service,
and limitations to the potential of health care and resources
for health care. The media and consumer organizations can
play important roles in advocacy for health, in raising pub-
lic consciousness of self-care, and as watchdogs on abuses.
Consumer acceptance is manifested through choice of
health plan and practitioner, or by seeking alternative care
BOX 15.7 Patients’ Rights, European Union, 2009
A review of patients’ rights in countries of the European
Union in 2009 focused on the following:
l Right to informed consent based on access to informa-
tion for care or participation in research
l Right to information concerning own health, diagnosis
l Right to medical records
l Right to confidentiality of personal and health informa-
tion and physical privacy during care
l Right to complain and compensation
l Right of free choice of provider and of treatment
l Respect of patient’s time
l Right to observance of quality standards access to high-
quality health services
l Right to safety and freedom from harm caused by the
poor functioning of health services, medical malpractice
and errors, and the right of access to health services and
treatments that meet high safety standards
l Right of access to innovative procedures, including diag-
nostic procedures, according to international standards and
independently of economic or financial considerations.
Source: European Patients’ Forum. Patients’ rights in the European Union.
Available at: http://www.eu-patient.eu/Documents/Projects/Valueplus/
Patients_Rights [Accessed 25 October 2012].
http://www.eu-patient.eu/Documents/Projects/Valueplus/Patients_Rights
http://www.eu-patient.eu/Documents/Projects/Valueplus/Patients_Rights
Chapter 15 Health Technology, Quality, Law, and Ethics
privately when service is unacceptable because of quality or
style. Erosion of confidence in a public system of care can
lead to a two-tier system with the public system serving the
poor and a private parallel system serving the middle and
wealthy classes. Such a division can seriously undermine a
public system unless it is addressed by improving the qual-
ity and manner of the service and by establishing supervi-
sion and limitations on public and private practice.
The growing inequality caused by the rise of private
practice outside a national health care system is a chronic
problem in the UK’s NHS, in Israel’s health system, and in
many countries developing their health systems through par-
allel public and private care. The issue is also surfacing in
the USA in the transition to managed care with its inherent
limitations of choice for people insured through their place of
work or covered under the Medicare and Medicaid programs.
The PPACA requires insurance companies to accept any-
one requesting cover without restrictions due to prior con-
ditions or high expenses for serious conditions, and without
other forms of discrimination common in the past. It also
includes provisions for coverage of preventive care services
and incentives for quality improvement. Extra billing, banned
in Canada’s national health insurance plan, is a recurring
issue with the medical profession in some provinces.
Consumer knowledge, attitudes, beliefs, and practices
are part of the health system, from health promotion to
tertiary care. Informed and health-conscious consumers
are stronger partners in the health system in achieving
improved health than an ill-informed and apathetic public,
so that health education and health promotion are funda-
mental to modern public health. The role of the consumer
in health care is unique in that there is a significant infor-
mation asymmetry between the consumer and provider.
Health education programs and wide use of the Internet
increase access to health and medical information, but this
gap can never completely be eliminated. Patients may use
their power as consumers to demand inappropriate care,
such as unnecessary surgery or antibiotics when clearly
not indicated, because of their preference for interven-
tion and action over watchful waiting. However, there is
an equal or perhaps greater danger of provider-induced
demand for repeated and possibly unnecessary interven-
tions that may be related to methods of paying the doctor
or the hospital. The traditional doctor–patient relationship
is still an important factor for the interests of patients and
their health. A still effective method of having an individ-
ual quit smoking is a brief but stern lecture by the family
physician.
THE PUBLIC INTEREST
Population-based interventions are often more effective
and less costly ways to reduce morbidity and mortal-
ity than individual prevention or treatment services. A
797
population-based preventive program may require behav-
ior change by the individual, such as in mandatory seat
belt and motorcycle helmet enforcement or banning smok-
ing in public places. Fortification of flour, milk, and salt
with essential micronutrients is a well-established public
health measure. There is an element of compulsion in this,
with the social gain usually considered to be sufficiently
important to outweigh individual rights. Immunization
is for the protection of individuals but also for the popu-
lation, so that refusals to immunize children and adults
can cause injury to others. Herd immunity is protective
of people who are at high risk. Mandatory immunization
for school entry in the USA has been effective in increas-
ing coverage to levels akin to the most advanced health
systems, over 95 percent coverage. Refusals and failure to
harmonize immunization policies in Europe have resulted
in mass epidemics of measles, rubella, and mumps in
recent years.
There is often a delicate balance between community
rights and individual rights which can lie at the heart of
many controversies in modern public health and health
care, ranging from chlorination or fluoridation of com-
munity water supplies to managed care systems for
health services. Women’s rights, gay rights and abortion
are highly controversial and politicized in the USA, and
in many other countries. The differences can become
extreme and the source of international strife, such as in
the movement to promote fundamentalist Sharia law in
many countries that are severely restrictive of women’s
and minority rights.
In public health, issues should be examined on their
merits, especially in terms of what is accepted as good pub-
lic health practice, based on evidence from clinical trials,
documented experience, and best practices in other coun-
tries. The evidence of successful public health measures in
improving individual and collective health status is power-
ful, yet must always be balanced within the context of indi-
vidual rights and the public interest. The ethical issues of
individual and community rights of public health are dis-
cussed later in this chapter.
TOTAL QUALITY MANAGEMENT
Total quality management (TQM), as discussed in Chap-
ter 12, was adapted from business management theory and
practice to health care in the 1990s and provides a basis for
promoting continuous improvement in health care systems.
TQM involves everyone in the system, from all levels of
management to production or service personnel and sup-
port staff, and thus helps to raise staff morale because of the
shared involvement. Health is provided through multidisci-
plinary groups which need to approach problems with open
and shared scientific inquiry and hypothesis formation, test-
ing, and revision to find operational solutions to problems.
798
Electronic health records and information technology pro-
vide many new opportunities to improve patient care and
data systems for monitoring the health status of population
groups for process and outcome measures, or health targets,
such as immunization coverage, or screening compliance
for colon, cervical, or breast cancer, as measures of perfor-
mance in primary care. Information technology adds a great
deal of capacity for quality monitoring and improvement
measures.
TQM incorporates statistical methods, comparing varia-
tions in patterns of service or use of resources. It employs
epidemiological methods to draw conclusions for policy
needs. It looks for continuous improvement, encouraging
cooperation, and motivation to achieve common goals of
service and client satisfaction. Psychological theory helps to
foster higher levels of motivation, with early identification
and resolution of conflict. Leadership is shared, and there is
a basic need for cooperation. Cost and quality are interre-
lated, as poor quality leads to waste, inefficiency, and dis-
satisfaction of both clients and staff. High-quality, humane,
and effective services are especially important in a competi-
tive environment where clients have the right to choose and
where costs and efficiency are factors in the well-being and
indeed the survival of institutions.
Medical care is increasingly practiced in larger health
care organizations. To provide technically competent
medicine is not by itself sufficient. The patient’s rights
and sense of personal worth are also of great importance.
Financial incentives can be effective in redirecting health
care priorities, such as in reducing hospital length of
stay and admissions, but may result in the patient or the
family feeling that they are not receiving the best care.
DRGs, HMOs, and other organizational and funding sys-
tems meant to increase efficiency of care may have the
effect of alienating patients from a health care system.
Staff attitudes towards patients are important for client
satisfaction. The service must include ready access to
a continuum of supportive services, such as home care
and counseling, so that the patient and family do not feel
abandoned by the system.
A byproduct of TQM is continuous quality improve-
ment (CQI), by which institutions wishing to improve qual-
ity train and empower the staff to work in teams to assess
their own performance and seek solutions to problems in
their operational unit. People of different ranks and profes-
sions work in a network organization as well as in a tradi-
tional hierarchical organization in which rank and seniority
provide authority. This community of practice is important
for staff morale and a shared sense of responsibility for the
patient and the institution.
CQI involves multidisciplinary approaches, not only
to review problems but also to seek better ways of func-
tioning and improving consumer satisfaction. The process
includes all those involved in providing care, support ser-
vices, and administration of a department, hospital, clinic,
The New Public Health
or community health program. This is not only professional
self-policing but a method to find better ways of meeting
needs and using resources. The involvement of all provid-
ers improves motivation and promotes a sense of common
purpose in the organization.
Applying these principles in a health care setting can
take many forms. Selection of topics by TQM/CQI com-
mittees in a hospital or another health facility may be
based on surveys or interviews with staff, patients, or
management. Satisfaction surveys among women follow-
ing delivery in an obstetrics unit could point out remedi-
able problems. An obstetrics department may be faced
with issues related to high or low volume of deliveries,
staff training, equipment and supplies, communica-
tion among staff, and among staff and patients and their
families, cleanliness, sterile technique, staff satisfaction,
client satisfaction, and many others. The team looking
at such a problem should be multidisciplinary, and the
emphasis should be on client attitudes and satisfaction.
Examination of the function of an emergency depart-
ment in a hospital would similarly look at many functional
and attitudinal aspects of the service including staff atti-
tudes, training needs, waiting times, consultation ser-
vices, and others. Addressing waiting times, for example,
can lead to ways to reduce these substantially, improving
both client satisfaction and the efficient management of
the emergency department. Any service is there to serve
patients and the community. A service is not primarily for
the benefit of the staff, but staff satisfaction and morale are
essential for successful service to clientele. CQI can also
be applied to assessing and improving compliance with
clinical guidelines or evidence. An example is assessing
the proportion of diabetics whose hemoglobin A1c (HbA1c)
is measured at least twice annually, who have eye and
feet examinations regularly, or whose blood pressure is
managed with an angiotensin-converting enzyme (ACE)
inhibitor.
The European Region of the WHO and the national
medical associations in Europe agreed in 1995 that medi-
cal associations should take leading roles in programs of
CQI to achieve better outcomes of health care in terms of
functional ability, patient well-being, consumer satisfac-
tion, and cost-effectiveness. This is in keeping with the
European Region’s Health for All targets: there should
be structures and processes in all member states to
ensure continuous improvement in the quality of care and
appropriate development and use of health technolo gies.
The introduction in the 1990s of general practitioner
fundholding for hospital care for patients on the general
practitioners’ roster in the UK encouraged the hospital to
maximize patient satisfaction with the care system. This
promotes application of CQI to improving the quality and
acceptability of care. Similarly, performance indicators
provide regional and district health authorities in the UK
with tools for CQI approaches. The UK NHS established
799Chapter 15 Health Technology, Quality, Law, and Ethics
The National Institute for Clinical Excellence (NICE), estab-
lished in 1999, has a mandate to review health service treat-
ments and effective therapies that should be commissioned
and made available within the National Health Service (NHS)
throughout England and Wales. The mission statement for
NICE is that it “contributes to better health around the world
through the more effective and equitable use of resources”.
In 2005 NICE was revised to include reviews of public health
interventions, and its mandate was expanded to include quality
standards for the English social care sector (English Health and
Social Care Act of 2011). Now renamed the National Institute
for Health and Care Excellence, NICE operates as a statutory
independent special health authority in England and Wales.
Commissioning bodies of the NHS are required to observe its
recommendations. Guidance can be used by the NHS, local
authorities, employers, voluntary groups, and anyone else
involved in delivering care or promoting well-being.
NICE recommendations are respected elsewhere in the
UK, but are not mandatory; in Scotland NICE recommen-
dations are published after further review by NHS Quality
Improvement Scotland (for health services issues) and by NHS
Health Scotland (for public health recommendations). NICE
recommendations are respected worldwide, including by the
European Commission and by national governments; NICE
International is a section of NICE established to meet non-UK
needs (e.g., evaluating rural health programs in China).
An independent committee including lay representation
advises on priorities for NICE consideration but final deci-
sions on topics referred to NICE are made by the Department
of Health. When making recommendations to the NHS on
which services (e.g., treatments) should be provided routinely,
it calculates the cost-effectiveness of treatment for each quality-
adjusted life year (QALY) of health gain purchased.
NICE publications include guidance on 374 interventional
procedures, 270 technology appraisals, 162 clinical guide-
lines, and 43 public health topics. From this latter group, some
examples include:
l Prevention of cardiovascular diseases (2010) – provides
evidence of effectiveness of population-based prevention
programs as more effective than programs aimed at high-
risk groups.
l Alcohol dependence and harmful alcohol use (2011) –
summarizes all NICE guidance; designed to inform mem-
bers of the public as well as health professionals.
l Preventing type 2 diabetes through population and com-
munity interventions (2011) – provides guidance to gov-
ernment departments, the commercial sector, health
service organizations, and non-governmental organiza-
tions on integration of public policy to prevent obesity, and
reduce diabetes prevalence and complications.
l Preventing uptake of smoking by children and young peo-
ple (2008) – document to advise local health service com-
missioners; identifies target populations, reviews campaign
messages, and provides recommendations for the mass
media and retailers.
l Promoting mental well-being at work (2009) – guidance
aimed at employers; reviews evidence in the field and rec-
ommends strategic approaches by firms, opportunities to
promote well-being and assess risk, and systems of flexible
working.
l Preventing unintentional injuries among under-15s in the
home (2010) – reviews evidence and makes recommenda-
tions to local authorities and related agencies on training
an appropriate workforce, advises government to fund cur-
ricula development, and indicates to the NHS appropriate
surveillance and treatment services.
NICE is often criticized (especially by the pharmaceuti-
cal industry) for the time taken to carry out investigations of
new treatments. It is also criticized by relatives of patients with
“glamorous” conditions (e.g., cancer) for not approving drugs
that might extend life by only 4–6 weeks, and perhaps approv-
ing instead new psychiatric therapies. The current government
has recently sought to overrule some of these NICE recommen-
dations in England. However, NICE methods and recommen-
dations are held in high repute, within the UK and beyond.
Sources: Christopher Birt FRCP FFPH, University of Liverpool, UK. Personal
communication.
National Institute for Health and Care Excellence. 2012. Available at: http://
guidance.nice.org.uk [Accessed 24 December 2012].
O’Flaherty M, Flores-Mateo G, Nnoaham K, Lloyd-Williams F, Rayner M,
Capewell S. Estimating potential cardiovascular mortality reductions with
different food policy options in the UK. Bull World Health Organ 2012;90:
522–31.
National Institute for Health and Clinical Excellence. Promoting mental
wellbeing through productive and healthy working conditions: guidance
for employers. NICE; 2009. Available at: http://www.nice.org.uk or http://
www.apho.org.uk/resource/item.aspx?RID=83868 [Accessed 18 August
2013].
Campbell B. Regulation and safe adoption of new medical devices and
procedures. Br Med Bull 2013;1–14 [Epub ahead of print]. http://dx.doi.
org/10.1093/bmb/ldt022.
BOX 15.8 The UK’s National Institute for Health and Care Excellence (NICE)
NICE as an independent body to promote “national
guidance on promoting good health and preventing and
treating ill-health”. NICE produces guidance in three
areas:
l public health – guidance for those working in the NHS,
local authorities and the wider public and voluntary sector
on promotion of good health and the prevention of disease
l health technologies – guidance on use of new and
existing medicines, treatments and procedures within
the NHS
l clinical practice – guidance on appropriate treatment
and care within the NHS of people with specific dis-
eases and conditions.
NICE guidelines are recommended practices with the
objective of reducing ineffective practices. During 2007,
guidelines were issued on topics including asthma, derma-
titis, caesarean section, chronic obstructive lung disease,
depression (in children and adults), eating disorders, fertil-
ity, contraception, multiple sclerosis, post-traumatic stress
disorder, and diabetic foot care (Box 15.8).
http://guidance.nice.org.uk
http://guidance.nice.org.uk
http://www.nice.org.uk
http://www.apho.org.uk/resource/item.aspx?RID=83868
http://www.apho.org.uk/resource/item.aspx?RID=83868
http://dx.doi.org/10.1093/bmb/ldt022
http://dx.doi.org/10.1093/bmb/ldt022
800
BOX 15.9 Organizations to Promote Quality in Health,
USA
l National Committee for Quality Assurance (NCQA) –
This non-profit organization, founded in 1979 by the
managed care industry, conducts surveys among man-
aged care plans to evaluate clinical standards, mem-
bers’ rights, and health service performance. It accredits
over 550 managed care plans in the USA, and in 2007
published rankings of the “best” health plans. Website:
Health Care Accreditation, Health Plan Accreditation Organization – NCQA
l Agency for Healthcare Research and Quality (AHRQ) –
This is part of the US Public Health Service. Founded in
1995, it was mandated to develop an evidence-based
practice program in 12 centers in the USA. It conducts
systematic reviews of the literature and publishes analy-
ses and findings of these reviews. Website: http://www.
ahrq.gov/
l Centers for Medicare & Medicaid Services (CMS) 2001 –
The CMS, previously the Health Care Financing
Administration (HCFA, 1977), is the federal agency of
the Department of Health and Human Services, respon-
sible for administering the Medicare and Medicaid and
the State Children’s Health Insurance Program (CHIP)
health plans. Its roles include quality assurance, the
requirements for managed care organizations, and qual-
ity improvement. Website: http://www.cms.gov/
l Institute for Healthcare Improvement (IHI) – Non profit
organization founded in 1991 as a global resource for
health care improvement knowledge to improve health
care by fostering collaboration among health care orga-
nizations. IHI examines office practices of physicians,
educational reform, and promotes interdisciplinary team
work in quality improvement. Website: http://www.ihi.
org/ihi/
l National Patient Safety Foundation (NPSF) – Sponsored
by the American Medical Association as a response to
findings of high rates of injury and death from iatrogenic
disease in the USA, the NPSF promotes research into
human error among health care providers, seeking ways
to reduce the frequency and effects of medical error,
such as misdiagnosis, medication errors, and mistakes
during procedures. Website: http://www.npsf.org/au/
l Joint Commission on Accreditation of Healthcare
Organizations (JCAHO) – Originating in 1917 by the
American College of Surgeons, it began accrediting
hospitals in 1918. It developed in 1953 as the JCAHO,
becoming a national voluntary accreditation organiza-
tion focusing mainly on hospitals. Its mandate was broad-
ened in 1987 and, as of 2007, had accredited more than
15,000 health care organizations. Accreditation is man-
datory for Medicare and Medicaid payment. The JCAHO
is changing its approach from standards-based assess-
ment every 3 years to one of reviewing performance data
quarterly as a continuous surveillance activity for risk
reduction. Website: http://www.jointcommission.org/
Source: Websites accessed 12 September 2012.
The New Public Health
The USA has a number of government and independent
organizations dedicated to improving quality in health care
systems. The CDC and the Institute of Medicine of the US
National Academies of Science play active roles in promot-
ing research quality and methods of CQI in the US health
care system. Canada is also very active in this regard, hav-
ing national and provincial institutes for the evaluation of
clinical effectiveness and clinical guidelines, and so too are
European countries (Box 15.9).
PUBLIC HEALTH LAW
Public health workers need knowledge of government
structure and public health legislation as basic to their pro-
fessional work to understand their responsibilities, powers
and liabilities. Law consists of a system of rules, regula-
tions, and orders that govern the behavior of individuals
and of society. Law represents the consensus of a society, as
enacted by an elected legislature, put into effect by the exec-
utive branch of government, and interpreted by the courts
as need be from time to time. The legislative and execu-
tive branches are separate under the US Constitution, but
the two are united in the parliamentary system (Box 15.10).
The authority, responsibility, and power to provide for and
protect the public health are basic functions of a sovereign
government, which may be delegated to another level of
government (higher or lower) or even a non-governmental
agency. The constitution of a sovereign government states
explicitly or implicitly that responsibility, but accepted
practice and court decisions (i.e., the common law) define
the powers of the national, state, or local government to
monitor and protect the health of its citizens.
In the USA, national legislation is enacted under the
powers of the federal government, namely to regulate inter-
state commerce and the power to tax and spend for the
general welfare. State legislation is enacted under the basic
power of the state to protect the health, welfare, and safety
of its citizens. Under these federal and state powers, a wide
range of health legislation and regulations is enacted affect-
ing public health, labor, and occupational health and safety,
environmental controls, public welfare, and the financing
of health services, agriculture, food, drugs, cosmetics, and
medical devices. Public health law relies on a wide range of
constitutional, statutory, administrative, and judicial deci-
sions in both civil and criminal actions. Appropriation of
funds is a legal act of legislative bodies to achieve objec-
tives directly or indirectly by financial incentives.
Categorical programs may be directed to specific issues
such as combating TB and promoting immunization or for
work to combat NCDs such as diabetes, or in improving
standards of facilities, and in providing health care services.
The regulatory, enforcement, policing, and punitive func-
tions of public health laws have evolved over many decades
and in many countries lack clear definition. In the USA,
Health Care Accreditation, Health Plan Accreditation Organization – NCQA
http://www.ahrq.gov/
http://www.ahrq.gov/
http://www.cms.gov/
http://www.ihi.org/ihi/
http://www.ihi.org/ihi/
http://www.npsf.org/au/
http://www.jointcommission.org/
801Chapter 15 Health Technology, Quality, Law, and Ethics
In federal nations, political authority is divided between two
autonomous sets of governments, one national and the other
subnational. Both operate directly with the people in their juris-
diction based on a constitutional division of power between
the national government, which exercises authority over the
whole national territory, and state or provincial governments
with independent authority within their own territories. The
constitution is the supreme law of a country. It sets out the
divisions of governmental powers including statutory authority,
administrative, natural resources, and taxation between federal
and state levels of government.
A federal legislature or congress makes the law of the land,
but is subject to rulings of a Supreme Court as are state and
local governments. State or provincial governments in a fed-
eral system have functions set out in the Constitution. They
also have elected legislatures, and executive branches with
taxing, regulatory, and punitive powers. Local governments
for county, municipal, or city governments also have dele-
gated taxing and regulatory powers including those of public
health.
Canada, the USA, Brazil, Australia, India, and Argentina
are organized on a federal basis. Federal countries also include
Austria, Germany, Malaysia, Mexico, Nigeria, Switzerland, and
Venezuela. Russia is called a federation. Usually there is some
overlapping or shared powers between national and state con-
stitutions, legislatures, and court systems, and public agencies,
taxing powers and regulatory functions, such as in interstate
commerce and emergency response to natural or other disasters.
In a unitary government system, most or all of the governing
power resides in a centralized government. This contrasts with
a federal system. In unitary systems the central government
commonly delegates authority to subnational units and chan-
nels policy decisions down to them for implementation.
A majority of nation-states are unitary systems. They vary greatly.
The UK includes England, Scotland, Wales, and Northern
Ireland, each with legislatures, but the Westminster Parliament
in London maintains national powers. In health, each of the four
member entities of the UK conducts a National Health Service
with autonomy but common features. The national government
may delegate certain powers to self-governing regions/local
authorities, and there is a growing tendency to devolve various
governmental functions such as health to regional authorities.
More than 150 countries are unitary states, including France,
Italy, Spain, China, and Japan.
In both forms of government, local authorities are estab-
lished under state law with governance by councils elected
by the people, with taxing and regulatory powers within the
state or provincial laws, with a high degree of autonomy but
within state regulation, standards, and financial support. Local
authorities have major responsibilities in public health such as
in sanitation, licensing, and regulation of businesses and zon-
ing, as well as many other areas, including social welfare.
Note: See also Chapter 10.
Source: Differences between federal and unitary forms of government.
Available at: http://www.preservearticles.com/201107139054/difference-
between-unitary-and-federal-forms-of-government.html [Accessed 15
December 2012].
Encyclopedia Britannica. Unitary government. Available at: http://www.
britannica.com/EBchecked/topic/615371/unitary-system [Accessed 15
December 2012].
Encyclopedia Britannica. Political systems. Available at: http://www.britan-
nica.com/EBchecked/topic/467746/political-system/36704/Federal-systems
[Accessed 15 December 2102].
BOX 15.10 Legal Structure of Federal and Unitary Countries
efforts are being made to update and reform laws in the pub-
lic health sector. In 1988, the Institute of Medicine (IOM) in
the USA (the Future of Public Health) called for codifica-
tion of public health law as essential for the public good,
while questioning the soundness of certain US public health
laws. More recently, the Model State Emergency Health
Powers Act in the USA, the Quarantine Act in Canada, and
the revised International Health Regulations (2007) have
sought to update century-old legislation. The revised inter-
national regulations provide for a global approach to con-
trol the spread of epidemics and public health emergencies
while minimizing disruption to international activities such
as travel, trade, and economics.
A combination of the regulatory, persuasive, and fund-
ing approaches is widely used in public health in control of
communicable and non-communicable diseases, in improv-
ing standards of facilities, and in providing health services.
The regulatory, enforcement, policing, and punitive func-
tions of public health are important in health promotion and
assurance of health care. The taxing power of government is
essential for public health to ensure that adequate facilities
and access to care are available to all members of the com-
munity, especially those in financial need and thus at greater
risk for disease.
Medical officers of health and their staff have legal
authority to issue formal orders for health protection of
the public. Situations which require court proceedings are
referred to the justice system. Situations that may require
enforcement by court proceedings are referred to the justice
system. Laws may be enacted to fund public health activi-
ties, whether provided by public health authorities or by
acting through official or non-official agencies or providers.
Public health authorities, namely medical officers of health,
have the legislative power to issue orders to individuals or
businesses where there is a threat to the health of the public
such as food establishments. Administrative resources are
needed to enforce laws, such as through the FDA and the
Environmental Protection Agency, which come under the
aegis of the Department of Health and Human Services.
Other departments such as Agriculture, Education, or inter-
departmental agencies (e.g., Homeland Security), also are
key to public health activities, such as in disaster situations.
http://www.preservearticles.com/201107139054/difference-between-unitary-and-federal-forms-of-government.html
http://www.preservearticles.com/201107139054/difference-between-unitary-and-federal-forms-of-government.html
http://www.britannica.com/EBchecked/topic/615371/unitary-system
http://www.britannica.com/EBchecked/topic/615371/unitary-system
http://www.britannica.com/EBchecked/topic/467746/political-system/36704/Federal-systems
http://www.britannica.com/EBchecked/topic/467746/political-system/36704/Federal-systems
802
Other intergovernmental activities may require special leg-
islation to empower, finance, and promote their cooperation,
such as in the case of establishing an authority to manage
long-term efforts to clean up a contaminated river or basin,
which involves the cooperation and coordination of many
local authorities.
Health protection of individuals and communities may
require legal action to detain a person in order to prevent
the spread of a reportable communicable disease, to protect
a mentally ill patient, or to restrain a violent person. Such
powers should be used as a last resort if voluntary compli-
ance and education fail, and where the danger to the commu-
nity or the individual is sufficient to convince a court of the
public need to override the personal liberty of an individual.
An example is a 2007 case of a person with MDR-TB who
was taken into custody on arrival for compulsory treatment
after traveling across the Atlantic Ocean on a commercial
airline, against the specific instructions of his physician, thus
endangering fellow passengers. Outbreaks of measles in the
UK (2006–2007) and in Israel via imported cases among
ultraorthodox Jews or conservative protestant groups in the
Netherlands, with transmission among religious people who
tend not to immunize their children, led to pressure by health
authorities to immunize those placed at risk by such contacts
at weddings or other large public events.
However, these measures are currently used less than
voluntary isolation or quarantine and placarding homes for
reportable infectious diseases such as measles. Powers are
essential in extreme cases where refusal to comply with
public health measures endangers others. Such powers
should have been used more vigorously in the early years
of the AIDS epidemic at a time when individual rights took
precedence over protection of the population, including
vulnerable high-risk groups. The severe acute respiratory
syndrome (SARS) epidemic of 2003 led to sequestering
hospital staff in Toronto, Canada, for lengthy periods to
prevent spread of the disease, and subsequent influenza
pandemic threats have raised questions as to whether hos-
pital personnel should be required to be immunized to pro-
tect patients and their families from onward transmission
of dangerous infections.
Recent cases in the USA, the UK, and Norway dem-
onstrate the responsibility of governments to protect the
public from incidents of violence by dangerous, mentally
disturbed individuals who carry out mass killings. In Nor-
way, 69 people, mostly teenagers, were killed by a radical
ideologue while many others sustained serious injuries; and
in the USA, Islamic terrorists at the Boston Marathon killed
three and seriously injured more than 200 others; a 20-year
old fatally shot his mother then killed 20 children and six
adult staff members at Sandy Hook elementary school in
Newtown, Connecticut, before killing himself; and an army
psychiatrist who had become an increasingly devout and
The New Public Health
radicalized Muslim psychiatrist shot and killed 13 people
and injured more than 30 others in a Texan army base.
Background checks and other restrictions on gun sales are
an important public health and political issue, especially in
the USA. The wide availability of guns, including military-
style assault weapons, presents a serious danger for impul-
sive or planned mass killings.
Public health has generally evolved with greater reli-
ance on health promotion through voluntary cooperation of
a patient or community than on compulsion. Enabling leg-
islation may permit a local authority to fluoridate its water
supply, but the enactment of local legislation and funding to
implement it may also require a public referendum. In some
states in the USA and in Israel, fluoridation of community
water supplies is mandatory, which is also part of the health
promotion approach to public health.
Appropriation of public funds to promote public health is
through approval by the legislature for a specified program.
Provision of public funds may take the form of categorical
grants for specified services, such as immunization, prenatal
care, school health, or specific disease management such as
TB control, cancer control, or AIDS education. Programs may
be designed to promote certain types and quality of services,
such as the Hill–Burton Act, which provided federal grants
for hospital construction in the 1950s to 1970s, conditioning
these grants on certain requirements concerning hospital licen-
sure and hospital planning. Such legislation has a “carrot and
stick” effect of attracting lower levels of government to seek
such funding but also requiring them to accept the conditions
and regulations that accompany the grants. The Canadian fed-
eral government’s cost sharing of provincial health (hospital
and medical) insurance programs is based on federal criteria
requiring public administration, portability between provinces,
accessibility without payment, comprehensiveness, and ban-
ning extra billing by physicians (see Chapter 13).
Public funds are also appropriated in the context of
legislated programs in which people are entitled to the ser-
vices defined in the appropriation legislation, such as in the
amendments to the Social Security Act providing Medicare
and Medicaid programs, or national health insurance legis-
lation in many countries. These and their regulations spell
out categories and specified entitlement benefits.
Legislation and court decisions to protect the rights of
the individual are part of public health. Public health law
is meant to protect individuals and communities from
potential abuse, of both individual and community human
rights, as in the US Bill of Rights. Enforcement of public
health law may infringe on individual rights by enforcing
sanitation, food and drug safety, and supervision of res-
taurants and catering firms. Laws may allow restriction of
civil rights, such as rarely used mandatory treatment of a
person with a dangerous contagious disease or mental ill-
ness. Freedom of religion may come into conflict with other
Chapter 15 Health Technology, Quality, Law, and Ethics
laws in public health where restrictive practices may deny
the use of publicly supported health facilities, as when a
religiously affiliated hospital may refuse an abortion proce-
dure in a case of rape. Religious practices or other personal
beliefs may endanger others in the community, such as in
the refusal to immunize children so that an imported infec-
tious disease may spread among non-immunized people
and even affect those who are immunized, as occurs with
imported measles cases even when domestic transmission
of the disease has previously been eradicated. General legis-
lative provisions applied to public health forbid misleading
or unethical advertising. Legislative provisions may also
ban advertising for products, such as tobacco, which are
legal but may be harmful to health. These laws affect public
health but are provisions in other statutes such as the regu-
lation of business enterprises. Legislation may also make
smoking in public places illegal, with fines for offenders
and operators of places such as public bars.
Since the 1973 US Supreme Court decision of Roe v.
Wade, the law has allowed women to seek safe and legal
abortion. This remains a highly controversial political issue
in the USA and several other countries. The potential con-
flict between community and individual interests and rights
is part of the dynamics of public health law and public health
practice. The issues involved are complex and highly politi-
cized, and often involve ethical distinctions where “the great-
est good for the greatest number” may limit the legitimate
rights of individuals and vice versa. The PPACA in 2010 is a
fundamental legislative initiative, upheld by the US Supreme
Court to become the law of the land. It will bring millions of
Americans into regulated health insurance with many protec-
tive elements to prevent abuse by private insurance company
through arbitrary exclusions or limitations.
The legal aspects of public health are vital to its opera-
tion and are increasingly complicated by ethical issues, and
by public and political debate. Health protective legislation
and regulation for sanitation of food, water, and air are fun-
damental to public health, as is the control of drugs, cosmet-
ics, vaccines, and biologicals, the manufacture of devices,
and the licensing of health personnel and facilities. Limi-
tations of legal suits (torts) against manufacturers of vac-
cines proved to be a successful measure in the USA with
the introduction of the National Vaccine Injury Compensa-
tion Program (NVICP) in 1988. This is funded by a modest
surcharge tax collected from vaccine manufacturers. It pro-
tects both public and private interests while providing a fair
compensation system to ensure patients’ rights but without
jeopardizing immunization to prevent widespread disease,
and also protects manufacturers from litigation with high
legal costs and excessive compensation awards by the jury
system. Promoting healthy behavior through the prudent
use of the legal system of regulation and taxation is increas-
ingly utilized to protect the health of the population. This is
803
widely applied in promoting road safety, in tobacco control
measures regarding banning of advertising, high taxes on
alcohol, and banning smoking in public places including
restaurants and bars.
Environmental Health
There is growing concern by the public and by governments
over climate change, global warming, air and water pollu-
tion, and other noxious and harmful industrial and com-
mercial processes. Environmental laws affecting the public
health include legislation on clean air, clean water, toxic
substances, solid waste control, and other noxious sub-
stances. Non-compliance with the legislative provisions can
result in prosecution in the civil or criminal courts or both.
Infringement of public health laws and regulations may
lead to criminal action as an increasingly common method
of sanction. While such violations may not be seen as “truly”
criminal and may be treated in the courts as misdemeanors,
they can lead to fines or even jail. Such cases are increas-
ingly being addressed seriously in the judicial system.
The CDC, in 1999, defined 10 great achievements
of public health of the twentieth century. These achieve-
ments are identified as control of infectious disease, motor
vehicle safety, fluoridation of drinking water, recognition
of tobacco use as a health hazard, immunization, decline
in deaths from coronary heart disease and stroke, safer
and healthier foods, healthier mothers and babies, family
planning, and safer workplaces (Goodman et al., 2006).
Of the 10 great achievements in the twenty-first century
(2001–2010) identified by CDC, seven of the 15 leading
causes of death (largely NCDs) resulted in a decline in the
age-adjusted death rate in the USA from 881.9 per 100,000
population in 1999 to 741.0 in 2009. This decline was a
result of a combination of supportive laws and legal tools
at the local, state, and federal levels. In other industrial-
ized countries similar legislation has led to equal or greater
achievements in public health over the past century.
Public Health Law Reform
Public health law is scattered through many legislative stat-
utes and administrative documents which developed his-
torically. Efforts to codify public health law may contribute
to greater understanding and enforceability of the many
separate pieces of legislation (Box 15.10). Such reform
will enhance understanding in the legislative, judicial, and
administrative branches of government as well as in busi-
ness, non-governmental organizations, and the community.
Box 15.11 suggests topics for model public health consoli-
dation or compendia for states. The principles of this formu-
lation may also apply to other countries at the national and
state or provincial levels.
804
ETHICAL ISSUES IN PUBLIC HEALTH
The field of public health includes a wide range of activities
and professional disciplines, ranging from health promo-
tion to disease protection, epidemiology to environmental
health, and financing to supervision or provision of clinical
care. Each of these disciplines works within systems that
face ethical dilemmas, and public health workers’ under-
standing and motivation within the ethical guidelines of
their professions and roles are important in their training
and practice conduct. Ethical frameworks have evolved in
part as the result of bitter experience with ethical failures
which were later recognized and affect public health stan-
dards of practice for future generations (Box 15.12).
Ethics in health are based on the fundamental religious
and humanistic values and concepts of a society. If the prin-
ciple of saving a life is valued above all other considerations
(i.e., Sanctity of Life or Pikuah Nefesh) (see Chapter 1),
then all measures available are to be used, irrespective of
the condition of the patient or the cost. If sickness and death
are seen as acts of God, possibly as punishment for sin, then
prevention and treatment may be considered to be interfer-
ing with the divine will, and the ethical obligation may be
limited to relief of suffering. Humanism balances these two
ethical imperatives: saving of life and relief of suffering.
Materialistic political philosophies may view health care as
primarily a function to preserve health for economic pros-
perity and social well-being. Secular humanism adopted
many of the religious precepts of the worth and rights of the
individual and these have become part of the standards of
law and ethics in modern secular societies.
The role of society in protecting the health of the popula-
tion grew during the nineteenth century with the sanitation
BOX 15.11 Public Health Law Program of the Centers
for Disease Control and Prevention
The Public Health Law Program (PHLP) is administered by
CDC’s Office for State, Tribal, Local and Territorial Support
(OSTLTS). PHLP develops law-related tools and provides
legal technical assistance to public health practitioners and
policy makers in state, tribal, local, and territorial (STLT)
jurisdictions.
The PHLP works with state/territorial health departments
and other partners to:
l identify public health law priorities
l research laws that impact the public’s health
l analyze public health legal preparedness
l conduct comparative analyses across jurisdictions;
prepare guidance, articles, reports, and toolkits; and
develop and disseminate public health law curricula.
Source: Centers for Disease Control and Prevention. Public health law.
Available at: http://www.cdc.gov/phlp/about.htm [Accessed 27 October
2012].
The New Public Health
movement, while medical care became an effective part of
public health during the twentieth century. The astonishing
successes of public health during the past century increased
life expectancy in the high-income countries by some 30
years, mostly through improved living conditions and
health protection, as well as societal and medical advances
to make care available to all. In the 1970s the Lalonde con-
cept that individual behavior was one of the key determi-
nants of health (see Chapter 2) placed much of the onus
of illness and its prevention on the individual, but fostered
health promotion as an essential component of public health
theory and practice. All these points of view are involved
in the ethical issues of the New Public Health (Box 15.13).
Resources for health care are limited even in indus-
trialized countries, so that priority setting and judicious
allocation of scarce resources are always issues. Money
spent on new technology with only marginal medical
advantages is often at the expense of well-tried and proven
lower cost techniques to prevent or treat disease. The
potential benefits gained by the patient from more and
more interventions are sometimes very limited in terms
of length or quality of life. These are difficult issues when
the physician’s commitment to do all to preserve the life
of the patient conflicts with the patient’s concept of qual-
ity of life and his or her right to decline or terminate heroic
measures of intervention. Many health systems use clini-
cal guidelines that are mandatory for a health facility or a
doctor in the clinic. Preparation for surgery requires a sig-
nature from the patient to consent to the procedure being
carried out, careful preoperative procedures to ensure that
the correct organ is addressed, antiseptic preparation of
the site, and checking that all instruments are accounted
BOX 15.12 Topics for a Model State Public Health Act
l Mission and functions
l Public health infrastructure
l Collaboration and relationships
l Public health authorities and powers
l Public health emergencies
l Public health information privacy
l Criminal/civil
l Enforcement
l Legislative response to the need to reform core public
health powers such as surveillance, reporting, epide-
miological investigations, partner notification, testing,
screening, quarantine, isolation, vaccination, and nui-
sance abatement
l Medicaid
l Affordable Care Act insurance agency
Source: Centers for Law and the Public’s Health. A Collaborative at Johns
Hopkins and Georgetown Universities (CDC Collaborating Center).
Available at: http://www.publichealthlaw.net/ [Accessed 23 September
2012], and Chapter 10 references.
http://www.cdc.gov/phlp/about.htm
Chapter 15 Health Technology, Quality, Law, and Ethics
for. The checklist approach is well established for care
in many settings and protects the patient from neglect or
faulty follow-up, such as in the management of hyperten-
sion and diabetes.
The suffering that a terminally ill patient may endure
during radical treatment, which may prolong life by only
hours or days, clashes with the physician’s ethical obli-
gation to do no harm to the patient. The ethical value
of sustaining the life of a terminally ill patient suffering
extensively is an increasing medical dilemma. The issue is
even more complex when economic values are included in
the equation. There are potential conflicts among the eco-
nomic issues, the role of the physician in preserving life,
the physician’s obligation to do no harm, the felt needs
of the patient and his or her family, and the needs of the
community as a whole. The complex issues involved in
the “right to die” and end-of-life care raise many ethical
and legal questions for the patient, the family, society, and
caregivers.
The state represents organized society and has, among
its responsibilities, a duty to promote healthful conditions
and to provide access to health care and public health
BOX 15.13 Study and Practice of Public Health Ethics
Ethics is a branch of philosophy that deals with distinctions
between right and wrong, with the moral consequences of
human actions. The ethical principles that arise in epidemio-
logical practice and research include:
l informed consent
l confidentiality
l respect for human rights
l scientific integrity.
“As a field of study, public health ethics seeks to understand
and clarify principles and values which guide public health
actions. Principles and values provide a framework for decision
making and a means of justifying decisions. Because public
health actions are often undertaken by governments and are
directed at the population level, the principles and values
which guide public health can differ from those which guide
actions in biology and clinical medicine (bioethics and medical
ethics) which are more patient or individual-centered.
As a field of practice, public health ethics is the application
of relevant principles and values to public health decision mak-
ing. Public health ethics inquiry carries out three core functions:
(1) identifying and clarifying the ethical dilemma posed,
(2) analyzing it in terms of alternative courses of action and their
consequences, and
(3) resolving the dilemma by deciding which course of action
best incorporates and balances the guiding principles and
values.” (CDC, 2001)
Sources: Last JM, editor. A dictionary of epidemiology. 4th ed. New York:
Oxford University Press; 2001.
Centers for Disease Control and Prevention. Science coordination and
innovation. Public health ethics; 2001. Available at: http://www.cdc.gov/
od/science/phec/ [Accessed 23 September 2012].
805
services. The conflict between individual rights and com-
munity needs is a continuous issue in public health. Appli-
cation of accepted public health measures for the benefit of
some people in society may require applying an interven-
tion to everyone in a community or a nation. The major-
ity thus are subject to a public health activity to protect a
minority, without designating which individual’s life may
be saved. Furthermore, a society may in special cases need
to restrict individual liberties to achieve the goal of reducing
disease or injury in the population. Raising taxes on alco-
hol and tobacco products, mandatory speed limits, driving
regulations, and seat belt usage laws are examples of pub-
lic health interventions that interfere with individual liberty
but protect individuals, and thereby the community at large,
from potential harm.
Many public health measures originally criticized as
interventions in private rights are generally accepted as
essential for health protection and promotion to reduce
the risk of disease in the population. Chlorination of com-
munity water supplies is a well-established, effective, and
safe intervention to protect the public health. Fluoridation
of drinking water to prevent tooth decay in children means
that other people are also drinking the same fluoridated
water, which is of less direct benefit to them. Fortification
of foods with vitamins and minerals is also a cost-effective
community health measure with advocates and opponents.
The addition of folic acid to food as the most effective way
to prevent neural tube defects in newborns is an intervention
mandated by the US FDA since 1998.
Confidentiality to assure the right of the individual to
privacy involves ethical issues in the use of health infor-
mation systems. Birth, death, reportable conditions (not
all reportable diseases are infectious), and hospitalization
data are basic tools of epidemiology and health manage-
ment. The use of detailed individual data is needed for
case-finding and follow-up activities which are vital to
good epidemiological management of diseases, includ-
ing STIs. However, caution is needed in data use to avoid
individual identification that could be used punitively,
for example, in denial of access to health insurance for
smokers, alcoholics, or AIDS patients because health
damage may be attributable to a self-inflicted risk factor.
Increasingly, however, reporting is also mandatory for
physical or sexual abuse and criminally linked injuries
as essential for the protection of individuals at risk or the
general public from serious harm.
Individual and Community Rights
The protection of the individual’s rights to privacy, and free-
dom from arbitrary and harmful medical treatments, proce-
dures, or experiments, may come up against the rights of the
community to protect itself against harmful health issues.
This conflict comes into much of what is done in public
http://www.cdc.gov/od/science/phec/
http://www.cdc.gov/od/science/phec/
806
health practice, which has both an enforcement basis in law
and practice and a humanitarian and protective aspect based
on education, persuasion, and incentives. Society permits
its governments to act for the common good, but sets limits
that are protected by the courts and administrative appeal
mechanisms.
Society has the right to legislate the side of the road
on which one is permitted to drive, the speed permitted,
the wearing of seat belts, and the non-use of alcohol or
drugs before driving or cell phones while driving. Offend-
ers may be punished by significant fines or jail and are
subject to strong educational efforts to persuade them to
comply. Similarly, the community must ensure sanitary
conditions and prevent hazards or nuisances from bother-
ing neighbors or the public. Society must act to protect the
environment against unlawful contamination or poisoning
of food, drugs, the atmosphere, the water supply, or the
ground.
Enforcement is thus a legitimate and necessary activ-
ity of the public health network to protect the community
from harm and danger to health. Table 15.6 shows topics
where individual rights and responsibilities predominate,
and a second set of rights that are the prerogative of the
community to protect its citizens against public health haz-
ards. Sometimes the issues overlap and sometimes come to
political, advocacy, or legal action, so that court decisions
are needed to adjudicate precedents for the future.
The AIDS epidemic in the 1980s and 1990s raised a
host of public health, ethical, and issues. Management of
the AIDS epidemic is in some respects in conflict with the
long-established role of society in contacting and quarantin-
ing people suffering from transmissible diseases. It is not
acceptable or feasible in modern society to isolate HIV car-
riers. But failure or delay of public health authorities even
in the late 1980s to close public bathhouses in New York
and other cities in the USA, where exposure to multiple
same-sex partners promoted transmission of the infection,
could be interpreted as negligence. During the 1980s, the
gay community in the USA centered its concern that HIV
testing would be used in a discriminatory manner. AIDS was
initially addressed as a civil liberties issue and not as a pub-
lic health problem. Screening, reporting, and case contact
follow-up were seen as an invasion of privacy and proved
counterproductive by increasing resistance to and avoidance
of testing. Protection of privacy and an educational approach
were adopted as most feasible and acceptable. International
opinion and national court decisions have emphasized the
right to privacy with decriminalization of non disclosure of
HIV status to sex partners (UNAIDS 2013).
The AIDS epidemic and public anxiety about contract-
ing AIDS through casual contact reinforced the need for
public education on safe sex. This has been raised as an
ethical issue because such education may be construed as
condoning teenage and extramarital relations. The issue of
The New Public Health
HIV screening of pregnant women in general or in high-risk
groups took on a new significance with the findings that
treatment of the pregnant woman reduces the risk of HIV
infection of the newborn, and that breastfeeding may be
contraindicated. This issue is arising anew in the context of
using the HPV vaccine for preteen girls to prevent the sexu-
ally transmitted infection, which is also controversial, and in
the USA this vaccination will be mandatory for school entry.
A pre-eminent ethical issue in public health is that of
assuring universal access to services, and/or the provision
of services according to need. An important ethical, politi-
cal, and social issue in the USA in the twenty-first century
is how to achieve universal access to health care. The soli-
darity principle of socially shared responsibility for fund-
ing universal access to health care is based on equitable
prepayment for health care for all by nationally regulated
mechanisms through place of work or general revenues of
government. A society may see universal access to health
care as a positive value, and at the same time utilize incen-
tives to promote the use of services of benefit to the indi-
vidual, such as hospital care, immunization, and screening
programs. Some services may be arbitrarily excluded from
health insurance, such as dental care, although this is to the
detriment of children and a financial hardship for many.
Strategies for program inclusion are often based on his-
torical precedent rather than cost-effectiveness or evidence.
While efforts are being made to include more children in the
program, the Medicaid system in the USA defines eligibil-
ity at income levels of 185 percent of the poverty line, thus
excluding a high percentage of the working poor. Health
is also a political issue in countries with universal health
systems where funding may be inadequate or patient dis-
satisfaction common.
Choices in health policy are often between one “good”
and another. Limitations in resources may make this issue
even more difficult in the future, with aging populations,
increasing population prevalence of physical disabilities,
and rapid increases in technology and its associated costs.
For example, the UK’s NHS at one point refused to provide
dialysis to people over the age of 65. When computed tomog-
raphy was first introduced, Medicare in the USA refused to
insure this service as an untested medical technique. Owing
to a lack of facility resources such as incubators and poor
prospects for the survivors, the Soviet health system con-
sidered newborns as living only if they weighed over 1000 g
and survived for more than 7 days. Those under 1000 g, who
would be considered living by other international definitions,
would be placed in a freezer to die. At the opposite extreme,
many western medical centers use extreme and costly mea-
sures to prolong life in terminally ill patients, preserving life
temporarily but often with much suffering for the person and
at great expense to the public system of financing health care.
In many countries, such as those in the former Soviet
system of health care, spending for hospital services, in
807Chapter 15 Health Technology, Quality, Law, and Ethics
TABLE 15.6 Individual and Community Rights and Responsibility in Health: Ethical/Legal Issues
Ethical/Legal Issues Individual Rights and Responsibilities Community Rights and Responsibilities
Sanctity of human life Right to health care; responsibility for
self-care and risk reduction
Responsible for providing feasible basket
of services, equitable access for all
Individual vs community
rights
Immunization for individual protection Immunization for herd immunity and community
protection; education; community may mandate
immunization
Right to health care All are entitled to needed emergency,
preventive, and curative care
Community right to care regardless of location,
age, gender, ethnicity, medical condition, and
economic status
Personal responsibility Individual responsible for health behavior,
diet, exercise, and non-smoking
Community education to health-promoting
lifestyles; avoid “blame the victim”
Corporate responsibility Management accountability to criminal
and civil action
Producer, purveyor of health hazard accountable
for individual and community damage
Provider responsibility Professional, ethical care and communication
with patient
Access to well-organized health care,
accredited to accepted standards
Personal safety Protection from individual, family, and
community violence
Public safety, law enforcement, protection of
women, children, and elderly; safety from
terrorism
Freedom of choice Choice of health provider; limitations of
gatekeeper functions; control costs while
function; right to second opinion; right of appeal
Confidentiality; informed consent; birth control
ensuring individual rights; limitations of
self-referrals to specialist
Euthanasia Individual’s right to die; limitations by
societal, ethical, and legal standards
Assure individual and community interests;
prevention of abuse by family or others with
conflict of interests
Confidentiality Individual’s right to privacy, limitation of
information
Mandatory reporting of specified diseases;
data for epidemiological analysis
Informed consent Right to know, risks vs benefits; agree or
disagree to treatment or participation i
n experiment
Helsinki Committee approval of research;
regulate fair practice in right to know; Patient’s
Bill of Rights
Birth control Right to information and access to birth
control and fertility treatment; woman’s
rights over her body
Political, religious promotion of fertility;
alternatives to abortion; protection of women’s
rights to choose
Access to health care Universal access, prepayment; individual
contribution through workplace or taxes
Solidarity principle and adequate funding; right
to cost containment, limitations on service
benefits
Regulation and incentives
to promote preventive care
Social security for hospital delivery,
attendance for prenatal care; primary care,
ambulatory care; home care
Incentive grants to assist communities for
programs of national interest; limit institutional
facilities
Global health Human rights and aspirations; economic
development, health, education, and jobs
Transfer of health risks; occupational hazards
and environmental damage
Rights of minorities Equality in universal access Special support for high-needs groups
Prisoners’ health Human rights Security and human rights; reduce inequalities in
sentencing convicts, harsh dangerous conditions
in prisons; prohibition of torture and execution
Allocation of resources Lobbying, advocacy for equity and innovation Equitable distribution of resources; targeting
high-risk groups; cost containment
some cases grossly in excess of need, is accompanied by
a lack of adequate funds for primary care or adding new
vaccines to the immunization program for children. The
majority of Americans have health insurance which increas-
ingly includes preventive care services, but a substantial
percentage lack such coverage which limits their access to
routine preventive care. The Affordable Care Act brings an
improvement in coverage and inclusion of preventive care
with incentives (see Chapters 10 and 13). In many coun-
tries, including in Europe, delay in updating immunization
808
programs may be due to a lack of funding or to delays
in professional or governmental acceptance of “new”
vaccines.
The closure or amalgamation of hospitals involves dif-
ficult decisions and is a source of friction between central
health authorities, the medical professions, and local com-
munities. Health reforms in many industrialized countries,
such as reducing hospital bed supplies and managed care
systems promoting cost containment and reallocation of
resources, raise ethical and political issues often based on
vested interests such as private insurance systems, hospi-
tals, and private medical practitioners.
Where there is a high level of cumulative evidence from
the professional literature and from public health practice
in “leading countries” with a strong scientific base and case
for action on a public health issue, when does it become bad
practice or even unethical public health practice to ignore and
fail to implement such an intervention? Such ethical failures
occur frequently and widely. For example, is it “unethical”
not to fortify grain products with folic acid, and salt with
iodine? Should there be a recommended European immu-
nization program; should milk be fortified with vitamin D;
should vitamin and mineral supplements be given to women
and children; should all newborns be given intramuscular
vitamin K routinely? Other examples include the issues of
fluoridation of water supplies and opposition to genetically
modified crops or generic drugs in African countries. These
issues are continuously debated and the responsibility of
the trained public health professional is to review the inter-
national literature on a topic and formulate a position based
on the cumulative weight of evidence. It is not possible to
wait for indisputable evidence because in epidemiology and
public health this rarely occurs. This is another reason for
guidelines established by respected agencies and profes-
sional bodies, which are free from financial obligations to
vested interest groups, being essential for review of the evi-
dence which continues to accumulate on many issues thought
to have been resolved or which reappear repeatedly despite
strong evidence of effectiveness and public health benefit.
Tragic Deviations in Public Health Ethics
In the nineteenth century the germ and miasma theories both
produced enormous gains in public health. The biomedical
paradigm addressed alleviation of disease risk or manifest
disease; the health paradigm addressed the improvement of
social and environmental conditions for reducing disease.
During the early part of the twentieth century, a segment of
the social hygiene movement promoted ideas of Social Dar-
winism or racial improvement by sterilization of mentally
ill, retarded, and other “undesirable” people.
The dominant biomedical model of public health and
medical professionals adopted policies of eugenics in
Sweden, the USA, and Canada, leading to policies and
The New Public Health
programs to force the sterilization of mentally handi-
capped or mentally ill patients. This distorted a socially
oriented concept of public health. This euthenasia policy
was adapted to a racially oriented policy with horrendous
policies of mass murder in the name of racial purity as a
public health policy in Nazi Germany with the near-total
support and participation of a highly Nazified medical
profession, and used in murder, by gassing or planned
starvation, of half a million “undesirables” under the
eugenics “T-4” program administered from Hitler’s head-
quarters. Although this program was stopped after parental
and Church protests in Germany, the methods used were
adopted in newly occupied countries and for concentra-
tion camps organized for the mass extermination of Jews,
Gypsies, and others in the Holocaust.
The eminent historian Sir Richard Evans (Regius Pro-
fessor of History at Cambridge University), in his classic
The Third Reich at War, wrote:
“At the heart of German history in the war years lies the mass
murder of millions of Jews in what the Nazis called ‘the final
solution to the Jewish question in Europe’. This book provides a full
narrative of the development and implementation of this policy of
genocide, while also setting it in the broader context of Nazi racial
policies toward the Slavs, and toward Gypsies, homosexuals, petty
criminals and ‘asocials’. … For many years, and not merely since
1933, the medical profession, particularly in the field of psychiatry,
had been convinced that it was legitimate to identify a minority of
handicapped as ‘a life unworthy of life’, and that it was necessary
to remove them from the chain of heredity if all the many measures
to improve the German race under the Third Reich were not to be
frustrated. Virtually the entire medical profession has been actively
involved in the sterilization programme, and from here it was but a
short step in the minds of man to involuntary euthanasia.”
The twentieth century was replete with mass murders,
executions, and genocide, with nationalistic, ideological,
and racist motives perpetrated by fascist, Stalinist, and
radical xenophobic political or religious movements when
gaining governmental power by election or by revolution,
in some cases applying common public health terminol-
ogy and concepts to uses of genocide and ethnic cleansing
(Box 15.14).
An outline of genocides of the past 100 years is seen
in Box 15.15. These include the Turkish genocide of the
Armenians in 1917 followed by horrific genocides in which
many millions of people were killed, carried out under the
communist regime of the Soviet USSR in the 1920s and
subsequently, in the People’s Republic of China under
Chairman Mao in the 1950s, and by the Khmer Rouge in
Cambodia in the 1980s, and in the wars resulting from the
breakup of the Yugoslav Republic in the 1990s.
The human and national cost of genocide lasts for gen-
erations. The hatred and fear may wane but the trauma goes
deep. It lasts with the victims and their descendants, but
Chapter 15 Health Technology, Quality, Law, and Ethics
also with the perpetrating country and its culture. The Nazi
Holocaust has had downstream effects in public health in
the German-speaking countries which last to the present
time, seven decades since the events took place. The long-
term damage done to public health in Germany and Austria
is described in Box 15.16.
The Nuremberg Doctors’ Trial in 1946–47 convicted
many leading Nazi physicians of crimes against humanity
and resulted in severe punishments including hanging or
long prison terms. This trial was a seminal event in estab-
lishing the ethical standards required for medical research
and human rights. However, many in the medical profession
aligned with these horrors remained leading figures in the
BOX 15.14 Values and Ethical Principles of Public
Health
l Sanctity of human life.
l Individual human rights – liberty, privacy, protection
from harm.
l Solidarity – sharing the burden of promoting and main-
taining health.
l Beneficence – reduce harm and burdens of disease and
suffering.
l Non-malfeasance – do no harm.
l Proportionality – restriction on civil liberties must be
legal, legitimate, necessary, and use the least restrictive
means available.
l Reciprocity principle – public responsibility to those who
face disproportionate health and social burden.
l Transparency principle – honest and truthfulness in the
manner and context in which decisions are made must
be clear and accountable.
l Precautionary principle – decision makers have a general
duty to take preventive action to avoid harm even before
scientific certainty has been established.
l Failure to act – public health officials and policy mak-
ers have a duty to act and implement preventive health
measures demonstrated to be effective, safe, and benefi-
cial to population health. Failure to enforce public health
regulations with resulting disease or deaths may consti-
tute negligence on the part of responsible officials with
civil or criminal penalties.
l Equity – reduce inequities.
l Cost and benefits
l Stewardship – responsibility of governance in a trustwor-
thy and ethical manner.
l Trust between the many stakeholders in health.
l Reasonableness – decisions should be evidence based
and revised based on new evidence.
l Responsive to needs and challenges as they may be
anticipated and appear with close monitoring of health
status.
Source: Modified from Lee LM. Guest editorial: Public health ethics
theory: review and path to convergence. Public Health Rev 2012;34(1).
Available at www.publichealthreviews.eu [Accessed 17 December 2012].
809
German medical community, one even being elected to head
the World Medical Association, then discussing the Helsinki
Declaration of Ethics in Biomedical Research, before being
forced to resign. The Nuremberg Trials and the subsequent
Helsinki Declaration laid the fundamentals of biomedical
ethics for the following generations, regulated by require-
ments of ethical procedures and institutional research board
approvals for funding, conducting, and publishing research
involving human subjects (Table 15.7).
The United Nations Convention on Prevention and Pun-
ishment of the Crime of Genocide (UNGC) of 1948 defines
acts committed with intent to destroy, in whole or in part,
members of a national, ethnical, racial, or religious group
as crimes against humanity. This convention specifies that
incitement to genocide is itself a crime against humanity.
Legal action should focus on state-sanctioned incitement
as a recognized early warning sign. The UNGC defines
genocidal acts to include the following as punishable under
international law:
l genocide
l conspiracy to commit genocide
l direct and public incitement to commit genocide
l attempt to commit genocide
l complicity in genocide.
The reappearance of genocide in the late twentieth cen-
tury in the Balkans and Rwanda, and in the twenty-first cen-
tury by Sudanese in Darfur, highlights genocide as a public
health concern and its prevention as a public health and
international political responsibility. Incitement to genocide
is a crime against humanity and was the basis for the trials
and convictions of leaders of the Rwandan Tutsi tribe, as
well as inciters to ethnic violence and the political leaders
and perpetrators of mass murders in the former Yugoslav
Republic. The threat and practice of genocide are still pres-
ent, whether in the murderous raids of Sudanese Janjaweed
militias in Darfur and South Sudan, the threats of genocide
by Iran and associated terrorist organizations against Israel
and Jews in general, or the killing of Christians in northern
Nigeria and Egypt, of Muslims in Burma/Myanmar, and
others. Incitement to genocide is now common as part of
international discourse.
Genocide represents the most extreme assault on the right
to life and respect for life. In the twentieth century, an esti-
mated 200 million people perished through genocide. Totali-
tarian dictatorships, past wars, and ideologies of exclusiveness,
ethnic purity, and religious fundamentalism increase the risks
for genocide. Perpetrators use dehumanizing, demonizing,
and delegitimizing hate language to desensitize or intimidate
bystanders and to mobilize, order, and instruct followers.
Genocide prevention requires international surveillance
networks for monitoring and reporting incitement and hate
language in the media, textbooks, places of worship, and the
Internet, which should monitor and identify their sources
The New Public Health810
Eugenics was a movement within the “social hygiene” con-
cept of the early part of the twentieth century. It was widely
promoted to reduce births among mentally ill and handi-
capped people in some states in the USA and was upheld
in decisions of the Supreme Court. It was also practiced
in Canada and Sweden. This idea was promoted by Hitler
in Mein Kampf and adopted by the Nazi Party, which was
legally elected to office in 1933 and began to implement it.
Organized massacres of mentally ill and handicapped chil-
dren and adults led to practices of organizing various modes
of killing, including gas chambers, which were applied in
concentration camps and in the Holocaust murder of 6 mil-
lion Jews and millions of others.
Genocide represents the most extreme assault right to life
and respect for life. In the twentieth century, an estimated 200
million have perished from genocide. Totalitarian dictator-
ships, past war and defeat, ideologies of exclusiveness, ethnic
purity, and religious fundamentalism increase risks for geno-
cide. Perpetrators use dehumanizing, demonizing, and delegit-
imizing hate language to desensitize or intimidate bystanders
and to mobilize, order, and instruct followers.
1915–1917 Armenian genocide by Ottoman Turkish
Empire – 1.2 million killed
1920s–1940s Eugenics movement in USA and Sweden
1920s Mass executions, deportations, and starvation as
policy in Soviet Union Stalinist regimes
1930s–1940s Mass sterilization of “defectives” in the USA
and Sweden
1930–1940s Mass murder of “defectives” in Nazi Germany
– 750,000 killed
1940s Quarantining as pretext for ghettos by Nazis
1940s Concentration camps, human experimentation
1940s Holocaust of 6 million Jews and genocide in Nazi
occupation of Poland and in Soviet Union
1947 Nuremberg Trials – convictions and capital punishment
for war crimes and genocide by Nazi leaders and doctors
1950s Mass starvation in Maoist China – estimated deaths
of 21 million people
1948 Convention on the Prevention and Punishment of the
Crime of Genocide
1975–1979 Cambodian genocide – 1.7 million killed
1988 Iraqi genocide of Kurds in town of Halabja by
poison gas
1988 Brazil genocide conviction of Tikuna people
1995 Serbian massacres in Srebrenica in Bosnia and
Herzegovina
2004 Rwandan genocides
2003–2012 Sudanese genocide in Darfur – over 400,000
killed
2011 Sudanese genocide of Nuba people
2012 Iran incitement to genocide of Israel
2012 Syria: civil war and genocide
2012 Democratic Republic of Congo massacres of Kivu
reported
Sources: Richter ED, Genocide Prevention Center, Braun School Public
Health, Hebrew University –Hadassah, Jerusalem, Israel. Personal commu-
nication.
United Nations. Convention on the prevention and punishment of the crime
of genocide. Available at: http://www.hrweb.org/legal/genocide.html and
www.un.org/millennium/law/iv-1.htm [Accessed 16 December 2012].
Richter ED. Commentary. Genocide: can we predict, prevent, and protect? J
Public Health Policy 2008;29:265–74.
Stanton G. The eight stages of genocide; 1998. Available at: www.genocide-
watch.org/aboutgenocide/8stagesofgenocide.html
Genocide Watch. http://www.genocidewatch.org/ [Accessed 25 December
2012].
BOX 15.15 Eugenics and Genocide: The Slippery Slope
and map their distribution and spread. Dehumanization,
demonization, delegitimization, disinformation, and denial
are the danger signs of potential genocidal actions. Geno-
cide results from human choice and bystander indifference.
One lesson of the Holocaust is that silence in response to
incitement to genocide makes one a complicit bystander.
Public health professionals and institutions have a responsi-
bility to speak out publicly on such dangerous early warn-
ing signs (Richter E, personal communication, 2012).
Human Experimentation
Human experimentation has been a subject of great concern
since the Nazi and Imperial Japanese armed forces’ experi-
ments on prisoners and concentration camp victims during
World War II. The Nuremberg Trials set forth standards of
professional responsibility to comply with internationally
accepted medical behavior (Table 15.7).
The Helsinki Declaration was first adopted by the
World Medical Assembly in 1964, and amended in 1975,
1983, 1989, and 1996. It delineates standards of medical
experimentation and requires informed consent from sub-
jects of medical research. These standards have become an
international norm for experiments, with national, state,
and hospital Helsinki committees regulating research pro-
posals within their jurisdiction. Funding agencies require
standard approval by the appropriate Helsinki committee
before considering any proposal, with informed consent on
any research project.
The Tuskegee experiment (Box 15.17) was a grave and
tragic violation of medical ethics, but in the context of the
1930s was consistent with widespread and institutionalized
racism. It provides an important case study which has reper-
cussions until the present time in suspicion of public health
endeavors, particularly among the African American com-
munity in the USA.
http://www.hrweb.org/legal/genocide.html
http://www.un.org/millennium/law/iv-1.htm
http://www.genocidewatch.org/aboutgenocide/8stagesofgenocide.html
http://www.genocidewatch.org/aboutgenocide/8stagesofgenocide.html
http://www.genocidewatch.org/
811Chapter 15 Health Technology, Quality, Law, and Ethics
In the German context the social–ecological health paradigm
can be traced back to the late eighteenth and early nineteenth
centuries when the country was a loose alliance of kingdoms
or princedoms lagging behind the economic, cultural, and
political developments in England and France. Ensuring popu-
lation health was seen as the obligation of the state, while the
family was responsible for caring for the health and well-being
of its members. Organized health care and health maintenance
was seen in the framework of Medizinische Polizey, as a model
of the health systems. Leading scholars in law and medicine
shared a normative perspective of promoting a healthy life-
style (known as dietetics), and provision of shelter, food, and
spiritual aid in asylums for the sick and disabled, or in private
homes the for wealthy.
In the second half of the nineteenth century evidence from
medical statistics and overwhelming practical experience indi-
cated that widespread poverty was the critical factor explain-
ing high rates of typhus or cholera epidemics in lower social
classes among children and industrial workers. A social health
movement fought for healthier living and working conditions,
education, and democracy. The movement’s prominent lead-
ers were Salomon Neumann, a physician pioneer in medical
statistics, and Rudolf Virchow, the renowned pathologist and
outspoken political activist.
Between 1890 and 1930 the conceptual framework of pub-
lic health was defined as “social hygiene” or “health science”,
an interdisciplinary field to conduct scientific research, aca-
demic teaching, and community-based activities aiming at the
promotion of individual and collective health and the preven-
tion of disease. In the 1920s the field was highly developed and
pioneering the modern academic public health.
Social hygiene was a general framework open to different
definitions. A group of academic teachers and publishing scien-
tists sharing the social–ecological paradigm, among them a high
proportion of German Jews, wanted to continue the social reform
strategy and to strengthen local communities to take an active role
in the formulation and implementation of health policies.
Public health activists sharing the biotechnological disease
paradigm favored a more focused approach aiming at the con-
trol of disease through medical care. Although there was no
supportive evidence, in the late nineteenth century a racial
eugenic movement emerged widely in Europe and the USA.
A conceptual model derived from the disease paradigm postu-
lated racial factors to explain disease. A healthy population was
assumed to be “free” of “racially contaminated” individuals and
inferior groups. Health-related public policy was supposed to
eliminate racially “unclean” members, e.g., by forced steriliza-
tion or murder. This was a central theme in Hitler’s Mein Kampf
and was enacted as basic policy by the Nazi Party in Germany
as a fundamental ideological basis of racial theory and public
health.
When the Nazis were legally elected in Germany in 1933,
and later seized power in Austria, this policy provided fertile
ground to open the door to euthanasia, leading to mass mur-
der. This was implemented in the well-organized, medically
directed execution of mentally and physically handicapped
Germans and others in psychiatric facilities. This provided a
working model for the industrialized murder of 6 million Jews
in the Holocaust and millions of gypsies, homosexuals, com-
munists, and others.
It took only 10 years to eradicate a 200-year tradition of
German socially oriented public health grounded largely in the
political philosophy of human rights and social justice. Most
of those advocates were exiled or murdered. Many of the aca-
demic medical leaders after World War II remained in key posi-
tions in the German public sector for decades.
In contrast to many other countries, the two wealthy
German-speaking countries, with over 90 million people,
have few academic public health resources. In there is only
one German School of Public Health, and a small number of
institutes, far fewer in Austria than in Germany. More than half
a century has passed since the Nazi period and the populations
of these two countries are slow to build a new socially oriented
public health system.
Sources: Horst Noack MD, PhD, Professor Emeritus, Medical University of
Graz, Austria. Personal communication; 24 December 2012.
Flügel A. Public Health und Geschichte. Weinheim: Beltz Juventa; 2012.
Heinzelmann W. Sozialhygiene als Gesundheitswissenschaft. Bielefeld:
Transcript Verlag; 2009.
Noack H. Governance and capacity building in German and Austrian public
health since the 1950s. Public Health Rev 2011;33:264–76.
BOX 15.16 The Rise, Fall, and Slow Recovery of German Public Health
Ethics in Public Health Research
The border between practice and research is not always
easy to define in public health, which has as one of its
major tasks the surveillance of population health. This
surveillance is mostly anonymous but relies on individu-
ally identifiable data needed for reportable and infec-
tious disease control as well as for causes of death, birth
defects, mass screening programs, and other special dis-
ease registries. It may also be necessary to monitor the
effects of chronic disease, for example, to ascertain repeat
hospitalizations of patients with congestive heart failure
to assess the long-term effects of treatment, and the effects
of strengthening ambulatory and outreach services to sus-
tain chronic patients at a safe and functional level in their
own homes.
Hospitalizations, immunizations, and preventive care
practices (e.g., Pap smears, mammography, and colonosco-
pies) are all part of the New Public Health. Impact assessment
of preventive programs may require special surveys and are
important to assess smoking and nutritional status and other
measures of health status and risk factors. Every effort must be
made to preserve the anonymity and privacy of the individual
but in some cases, where the disease is contagious, case contact
is crucial. This can entail identifying people who attended an
The New Public Health812
TABLE 15.7 Ethical Issues of Medical Research Derived from the Nuremberg Trials, the Universal Declaration of
Human Rights, and the Declaration of Helsinki
Nuremberg Doctors Trial,
1946–47
The voluntary consent of a human subject is absolutely essential, with the exercise of free power of
choice without force, fraud, deceit, duress, or coercion
Experiments should be such as to bear fruitful results, based on prior experimentation and the natural
history of the problem under study. They should avoid unnecessary physical and mental suffering
The degree of risk should not exceed the humanitarian importance of the experiment
Persons conducting experiments are responsible for adequate preparations and resources for even the
remote possibility of death or injury resulting from the experiment
The human subject should be able to end his participation at any time
The scientist in charge is responsible to terminate the experiment if continuation is likely to result in
injury, disability, or death
Universal Declaration
of Human Rights, 1948
Everyone has the right to a standard of living adequate for the health and well-being of himself and of his
family, including food, clothing, housing, and medical care and necessary social services
United Nations covenants for
protection of human rights
Covenant on Civil and Political Rights
Optional Protocol to the Covenant on Civil and Political Rights
Covenant on Economic, Social, and Cultural Rights
Convention Against Torture
Convention Against Genocide
The Geneva Conventions
Convention on the Rights of the Child
Convention on Elimination of Discrimination Against Women
Charter of the United Nations
Declaration of Helsinki, 1964 Research must be in keeping with accepted scientific principles, and should be approved by specially
appointed independent committees
Biomedical research should be carried out by scientifically qualified persons, only on topics where
potential benefits outweigh the risks, with careful assessment of risks, where the privacy and integrity
of the individual is protected, and where the hazards are predictable. Publication must preserve the
accuracy of research findings
Each human subject in an experiment should be adequately informed of the aims, methods,
anticipated benefits, and hazards of the study. Informed consent should be obtained, and a
statement of compliance with this code
Clinical research should allow the doctor to use new diagnostic or therapeutic measures if they
offer benefit as compared to current methods
In any study, the patient and the control group should be assured of the best available methods.
Refusal to participate should never interfere with the doctor–patient relationship. The well-being
of the subject takes precedence over the interests of science or society
Source: Summarized from the Nuremberg Trials (1948) and World Medical Association, Declaration of Helsinki.
Website sources include: World Medical Association. Available at: http://www.wma.net/
Australian Government Department of Health and Ageing. Available at: http://www.nhmrc.gov.au/health-ethics/human-research-ethics-committees-hrecs/
human-research-ethics-committees-hrecs/national
United Nations. A Summary of United Nations Agreements on Human Rights. Available at: http://www.hrweb.org/legal/undocs.html (accessed 10.1.14).
United Nations. Available at: http://www.un.org/en/events/humanrightsday/2007/hrphotos/declaration%20_eng (accessed 10.1.14).
US Food and Drug Administration. World Medical Association Declaration of Helsinki. Available at: http://www.fda.gov/ohrms/dockets/
dockets/06d0331/06D-0331-EC20-Attach-1 (accessed 10.1.14).
event or traveled on an airplane where an infected person may
have been, so as to take appropriate preventive measures.
The general distinction between research and practice
has to do with the intent of the activity. Clinical research uses
experimental methods to establish the efficacy and safety of
new interventions or unproved interventions; many drugs
and procedures in common use have never been subjected
to randomized controlled trials. In practice, many methods
are devised that are held to be effective and safe by expert
opinion and documented as such. Researchers comparing
HIV or hepatitis B transmission rates among intravenous
drug users not using needle-exchange programs would be
conducting unethical research, according to accepted cur-
rent standards, by giving needles to the experimental group
http://www.nhmrc.gov.au/health-ethics/human-research-ethics-committees-hrecs/human-research-ethics-committees-hrecs/national
http://www.nhmrc.gov.au/health-ethics/human-research-ethics-committees-hrecs/human-research-ethics-committees-hrecs/national
http://www.un.org/en/events/humanrightsday/2007/hrphotos/declaration%2520_eng
http://www.fda.gov/ohrms/dockets/dockets/06d0331/06D-0331-EC20-Attach-1
http://www.fda.gov/ohrms/dockets/dockets/06d0331/06D-0331-EC20-Attach-1
Chapter 15 Health Technology, Quality, Law, and Ethics
and withholding them from the control group. The scientific
justification of an experiment must be made explicit and
justifiable. Clinical equivalence is a necessary condition
of all clinical and public health research and provision of
standard of care treatment to control groups is a minimal
requirement for most research ethics boards. Determination
of the standard, and whether it should be place, time, and
community specific, is an area of ongoing controversy.
In 1996 a US Public Health Service study, supported by
the NIH and WHO, compared a short course of zidovudine
(AZT) to a placebo given late in pregnancy to HIV-positive
women in Thailand, measuring the rate of HIV infection
among the newborns. The experiment was terminated when
a protest editorial appeared in a prominent medical journal.
This study confirmed previous findings that AZT given dur-
ing late pregnancy and labor reduced maternal–fetal HIV
transmission by half. When a study shows clearly positive
results, it should be discontinued and reported so that the
findings can be applied generally. The findings indicated
that AZT should be used in developing countries, and the
BOX 15.17 The Tuskegee Experiment
The Tuskegee experiment was carried out by the US Public
Health Service between 1932 and 1972. It was meant to fol-
low the natural course of syphilis in 399 already infected
African American men in Alabama and 201 uninfected men.
The men were not told that they were being used as research
subjects. The experiment had been intended to show the
need for additional services for those infected with syphilis.
However, when penicillin became available, the research-
ers did not inform or offer the men treatment, even those
who were eligible when drafted into the army in 1942. The
experiment was stopped in 1972 as “ethically unjustified”
when the media exposed it to public scrutiny.
The case is considered unethical research practice
because, even at the time it was conducted, it did not pro-
vide the patients with available care and their well-being
was put aside in the interest of the descriptive study. A similar
experiment was conducted by the US Public Health Service
in cooperation with the Guatamala Ministry of Health dur-
ing the 1960s, in which syphilis was actually given to sol-
diers, prisoners, and others by sexual contact with prostitutes
known to have the disease, but the study was terminated
when it was discovered by a public health historian and
reached public attention in the USA.
In 1997, President Bill Clinton apologized to the survi-
vors and families of the men involved in the experiment on
behalf of the US government. The Tuskegee experiment is
the source of lingering widespread suspicion in the African
American community to the present time.
Sources: Lombardo PA, Dorr GM. Eugenics, medical education and pub-
lic health: another perspective on the Tuskegee syphilis experiment. Bull
Hist Med 2006;80:291–316.
Centers for Disease Control and Prevention. US Public Health Service
Syphilis Study at Tuskegee. Available at: http://www.cdc.gov/tuskegee/
timeline.htm [Accessed 13 December 2012].
813
manufacturers agreed to make it available at reduced costs.
The result has been a major success in helping with more
recent medications to reduce maternal–fetal transmission in
many places in Africa with help from GAVI, and a slowing
of the spread of HIV/AIDS-related deaths.
Public health may face the challenge of pandemic influ-
enza, such as avian flu, with decisions regarding the allocation
of vaccines, treatment of massive numbers of patients arriv-
ing at hospitals in acute respiratory distress with very limited
resources available, coping with sick or absent staff, and many
other issues requiring not only individual life and death situ-
ations, but mortality en masse. The ethical questions will be
replaced by struggles to cope with such situations. Preparation
for such potential catastrophic events will be a challenge to
public health organizations and the health system in general.
An outstanding case of a breach of ethics in public health
research occurred with the “Wakefield effect”, as described
in Chapter 4 and Box 15.18.
Ethics in Patient Care
Ethical issues between the individual patient and health
care provider are important in the New Public Health. A
doctor is expected to use diligence, care, knowledge, skill,
discretion, and caution in keeping with practice standards
accepted at the time by responsible medical opinion and
to maintain the basic medical imperative to do no harm
to the patient. Patients have the right to know their condi-
tion, available alternatives for treatment, and the risks and
benefits involved. They also have a right to seek alterna-
tive medical opinions, but this right is not unlimited, as any
insurance plan or health service may place restrictions on
payment for further opinions and consultation without the
agreement of a primary care provider.
Health care has a responsibility beyond that of the pay-
ment of health service bills and individual care by a physi-
cian, in institutions, or through services in the community
or the home. The contract for service is becoming less
between an individual physician and his or her patient, and
more among a health system, its staff, and the client. This
places a new onus on the physician to ensure that patients
receive the care they require. Conversely, the US provider
often faces the dilemma of knowing that a patient may not
access needed services because of a lack of adequate health
insurance.
Sanctity of Life Versus Euthanasia
The imperative to save a life is an important ethical and prac-
tical issue in health care. Advocates of physician-assisted
suicide (euthanasia) argue for the right of the patient to die
with dignity when the illness is terminal and the individual
is suffering excessively. This is not a medical decision alone,
and is an agonizing issue for society to address. The Nazi
euthanasia program and its human experiments provided
http://www.cdc.gov/tuskegee/timeline.htm
http://www.cdc.gov/tuskegee/timeline.htm
The New Public Health814
In 1998, The Lancet, published an article by a number of well-
known researchers headed by Dr Andrew Wakefield. The article
reported on 12 cases of autistic children and alleged to show a
connection to immunization with the MMR (measles–mumps–
rubella) vaccine.
The immediate effect of this “revelation” was widespread
alarm over the MMR vaccine and a fall off in immuniza-
tion coverage by measles-containing vaccines in the UK
and elsewhere with many mothers refusing to have their
child vaccinated due to a “risk of autism”. As a result,
measles epidemics occurred in the UK and in many other
countries, with measles again becoming endemic in many
parts of Europe, especially England and France.
After a long series of investigative journalism in the British
press, the article came under scientific scrutiny and withdrawal
of many of the coauthors but a consistent insistence by the lead
author of its authenticity.
Investigation by British medical authorities later found
Dr Wakefield guilty of medical negligence and the UK
General Medical Council withdrew his license to practice
medicine. The coauthors were found to have been credulous
and insufficiently vigilant in agreeing to coauthorship of the
paper. In 2000, 12 years after the original publication, The
Lancet formally withdrew the article.
The effect of this fraudulent scientific publication was a
serious loss of credibility of immunization in general and
especially regarding the MMR vaccine, one of the greatest life
savers in public health technology.
The return of measles in Europe to large scale epidemics
with frequent international transmission furthered the loss of
confidence of mothers in immunizations and public health.
Measles-containing vaccines were particularly strongly
affected owing to the publicity given to the Wakefield case.
The journal editors could be seen as irresponsible for failing
to ensure the scientific integrity of lead authors and coauthors,
and the journal for failing to retract a fraudulent article sooner
than 12 years after the first publication.
In other public health issues, single publications of findings
of small sample and poorly assessed studies published in haste
without adequate inquisitive review occur with great frequency.
The electronic media often include unscientific opinion blogs
which appear larger than life which provoke great anxiety over
accepted and successful public health interventions such as flu-
oridation or folic acid fortification of flour, with unsubstantiated
claims that they cause cancer, asthma, and other ill-effects.
The interface between ethics, law, and science in pub-
lic health requires continuous sensitivity to the downstream
effects of “shouting fire in the theater”.
Sources: Wakefield AJ, Murch SH, Anthony A, Linnell, Casson DM,
Malik M, et al. Ileal lymphoid nodular hyperplasia, non-specific colitis,
and pervasive developmental disorder in children [retracted]. Lancet
1998;351:637–41.
Office of Research Integrity. Definition of research misconduct. Available at:
http://ori.hhs.gov/misconduct/definition_misconduct.shtml
General Medical Council. Andrew Wakefield: determination of serious
professional misconduct 24 May 2010. Available at: www.gmc-uk.org/
Wakefield_SPM_and_SANCTION _32595267
Murch SH, Anthony A, Casson DH, Malik M, Berelowitz M, Dhillon AP, et al.
Retraction of an interpretation. Lancet 2004;363:750.
Godlee F, Jane Smith J, Harvey Marcovitch H. Editorial. Wakefield’s
article linking MMR vaccine and autism was fraudulent. BMJ
2011;342:c7452.
BOX 15.18 The Wakefield Effect
the direst of warnings to societies of what may follow when
the principle of the sanctity of the individual human life is
breached. The issue, however, returned to the public agenda
in the 1980s and 1990s as advances in medical science have
allowed the prolongation of human life beyond all hope of
recovery. Legislation in the Netherlands, the USA (“assisted
suicide” in the states of Washington, Oregon, and Montana),
and northern Australia has legally sanctioned euthanasia
with various safeguards in a variety of circumstances, such
as long-term comas or terminal illnesses.
Doctors, patients, relatives, and health care organiza-
tions need clear guidelines, orientation, procedures, legal
protection, and limitations where failure to take utmost
steps to “save” the patient by intubation, resuscitation, or
transplantation may cause legal jeopardy. Even though a
distinction can be drawn theoretically between permitting
and facilitating death, in practice, doctors in intensive care
units face such decisions regularly where the line is often
blurred. Hospital doctors routinely go to extreme mea-
sures to prolong the life of hopeless cases. Such decisions
should not be considered for economic reasons alone, but in
practice the costs of care of the terminally ill will be a driv-
ing force in debate of the issue. Living wills allow a patient
to refuse heroic measures such as resuscitation, with “do
not resuscitate” standing orders and assignment of power of
attorney to family members to make such decisions. Fam-
ily attitudes are important, but the social issue of redefining
the right of a patient to opt for legal termination of life by
medical means will be an increasingly important issue in
the twenty-first century.
The Imperative to Act or Not Act in Public
Health
As in other spheres of medicine and health, in public health
the decision whether to intervene on an issue is based on
identification and interpretation of the problem, the poten-
tial of the intervention to improve the situation, to do no
harm, and to convince the public and political levels of the
need for such intervention along with the resources to carry
it out. This process requires patience and a longer time-
frame than many other fields in health.
http://ori.hhs.gov/misconduct/definition_misconduct.shtml
http://www.gmc-uk.org/Wakefield_SPM_and_SANCTION _32595267
http://www.gmc-uk.org/Wakefield_SPM_and_SANCTION _32595267
Chapter 15 Health Technology, Quality, Law, and Ethics
Some interpretations of ethics in health consider that the
only purpose for which power can be rightfully exercised
over any member of a democratic community, against his
will, is to prevent harm to others. But this is not a dictum
that is applied to public health, which is obliged to act to
protect the public health in so many spheres such as food
and drug safety and environmental health, on a spectrum
that extends to banning smoking in public places, mandat-
ing food fortification, and many other areas of civil society.
Failure to act is an action, and when there is convincing
evidence of a problem that can be alleviated or prevented
entirely by an accepted and demonstrably successful inter-
vention, then the onus is on the public health worker to
advocate such action and to implement it as best as possible
under the existing conditions. Failure to do so is a breach of
“good standards of practice” and could be unethical. Iner-
tia of the public health system in the face of evidence of a
demonstrably effective modality such as adoption of state-
of-the-art vaccines or fortification of flour with folic acid
to prevent birth defects would come under this categoriza-
tion and may even constitute neglect and unethical practice.
This is not an easy categorization, because there is often
disagreement and even opposition to public health interven-
tions, as was the case with opposition to vaccination long
after Jenner’s crucial discovery of this procedure in the late
eighteenth century. It is also true today with opposition to
many proven measures such as fluoridation or fortification
of basic foods. Box 15.19 shows the ethical standards of the
APHA in 2006.
The use of ethical and high standards of practice in pub-
lic health (Box 15.20) requires an ideological commitment
to the advancement of health standards and use of best prac-
tices of international standards to the maximum extent pos-
sible under the local conditions in which the professional is
working. This is not an easy commitment as there is often
dispute and outright hostility to public health activities, in
part because of ethical distortions of great magnitude in the
past. But this is an optimistic field of activity because of the
great achievements it has brought to humankind. Prepara-
tion for disasters and unanticipated health emergencies in
addition to addressing current issues is a vital part of the
New Public Health and our ethical and professional com-
mitments.
SUMMARY
In order to maintain and improve standards of care, health
systems need quality assurance and technological assess-
ment as part of their ongoing operation. Poor-quality care
is costly in terms of iatrogenic diseases and prolonged or
repeated hospitalization. If innovations such as endoscopic
surgery are not introduced, then longer hospital stays are
needed for the same operation, wasting the patient’s time
and productivity, while utilizing expensive health care
815
resources, and incurring the risks associated with more
invasive surgery.
Health care is provided by people, as well as by insti-
tutions with a range of devices and equipment. The people
providing care, more than the technological facilities, set the
quality of care. Nevertheless, progress on the technological
side of medical care is vital to the continuing development
of the field. Modern medications, monitoring equipment,
laboratory services, and imaging devices have made enor-
mous contributions to advances in medical care. Appropriate
BOX 15.19 Principles of Ethical Public Health Practice:
American Public Health Association, 2006
l Public health should address principally the fundamen-
tal causes of disease and requirements for health, aiming
to prevent adverse health outcomes.
l Public health should achieve community health in a way
that respects the rights of individuals in the community.
l Public health policies, programs, and priorities should be
developed and evaluated through processes that ensure
an opportunity for input from community members.
l Public health should advocate and work for the empow-
erment of disenfranchised community members, aiming
to ensure that the basic resources and conditions neces-
sary for health are accessible to all.
l Public health should seek the information needed to
implement effective policies and programs that protect
and promote health.
l Public health institutions should provide communities
with the information they have that is needed for deci-
sions on policies or programs and should obtain the
community’s consent for their implementation.
l Public health institutions should act in a timely manner
on the information they have within the resources and
the mandate given to them by the public.
l Public health programs and policies should incorpo-
rate a variety of approaches that anticipate and respect
diverse values, beliefs, and cultures in the community.
l Public health programs and policies should be imple-
mented in a manner that most enhances the physical and
social environment.
l Public health institutions should protect the confidential-
ity of information that can bring harm to an individual or
community if made public. Exceptions must be justified
on the basis of the likelihood of significant harm to the
individual or others.
l Public health institutions should ensure the professional
competence of their employees.
l Public health institutions and their employees should
engage in collaborations and affiliations in ways that
build the public’s trust and the institution’s effectiveness.
Source: American Public Health Association. Public Health Leadership
Society. Principles of the ethical practice of public health. APHA; 2002.
Available at: http://www.apha.org/NR/rdonlyres/1CED3CEA-287E-4185-
9CBD-BD405FC60856/0/ethicsbrochure [Accessed 13 December
2012].
http://www.apha.org/NR/rdonlyres/1CED3CEA-287E-4185-9CBD-BD405FC60856/0/ethicsbrochure
http://www.apha.org/NR/rdonlyres/1CED3CEA-287E-4185-9CBD-BD405FC60856/0/ethicsbrochure
The New Public Health816
Publication in peer-reviewed journals is a key part of the
advancement in science and a vital part of the development
of the scientific basis for public health practice. The process of
publication should promote rigorous standards of high quality
ethical research and the wide dissemination of their findings.
Codes of practice for editors and publishers of peer-reviewed
journals have been developed by both the Committee on
Publication Ethics (COPE) (Rees, 2011) and the World
Association of World Editors (WAME).
Editors are subject to competitive pressures, and the over-
arching metric of success is seen to be the impact factor, a
measure of the frequency with which the “average article”
in a journal has been cited in a particular year or period.
Relevant, rigorous research of better quality will tend to be
cited more frequently, and thus editorial strategies that look
for quality and relevance in the given field will increase the
impact factor. However, there can also be potential distort-
ing factors. Publishing a highly controversial paper can result
in high citation levels. Publishing studies which demonstrate
negative findings may be less likely to attract large numbers
of citations.
Key issues relate to conflicts of interest, and the potential for
advertising and sponsorship to distort editorial decision mak-
ing (Gray, 2012). A particular concern has been the pernicious
influence of the tobacco industry in sponsoring, frequently
covertly, research which has aimed to confuse or obfuscate key
findings linking second hand exposure to tobacco to adverse
impacts on health. Similar tactics are used in other areas where
health and commercial interests collide. Clear statements of
potential conflicts of interest are essential. Journal owners must
not interfere in the evaluation, selection, or editing of individ-
ual articles, either directly or by creating an environment in
which editorial decisions are strongly influenced.
Other challenging areas are plagiarism and research mis-
conduct. The latter is extremely difficult both to detect and to
deal with, and requires close working between institutions and
editors who may suspect professional misconduct. In cases of
fraud, the publishing journal should withdraw the article in a
timely fashion (see Box 15.18: The Wakefield Effect).
There has been a rapid rise in open access publishing, in
part underpinned by an ethical belief that research is a public
good, and an increasing number of influential research funders
now require that there should be unrestricted access to the
published output of research. In addition, several publishers
make their journals free to those in selected low-income coun-
tries, promoting dissemination to those who might not other-
wise afford them.
In summary, publication in peer-reviewed journals remains
a key method for establishing and progressing the evidence
base for public health practice. The consequences of poor
or frankly fraudulent science can have a substantial adverse
impact both on health and on the use of resources. Editors must
adhere to high ethical and professional standards and remain
vigilant to avoid allowing external drivers to distort their deci-
sion-making processes. They must strive to maintain integrity
and high scientific standards to advance the field of public
health practice (Smith, 2007).
Sources: Selena Gray, BSc, MBCHB, MD, FFPH, FRCP, Professor, University
of West of England, Bristol, and Deputy Postgraduate Dean, Severn Deanery,
Bristol, UK. Personal communication.
Rees M. Code of conduct and best practice guidelines for journal editors.
Committee on Publication Ethics; 2011. Available at: http://publicationeth-
ics.org/ [Accessed 21 August 2012].
Gray S. The ethics of publication in public health. Public Health Rev 2012;34.
Epub ahead of print. Available at: www.publichealthreviews.eu [Accessed 20
December 2012].
Smith R. The trouble with medical journals. London: Royal Society of
Medicine Press; 2007.
BOX 15.20 The Ethics of Publication in Public Health
technology is a critical issue for international health, since
the most advanced technology may be completely inappro-
priate in a setting that cannot afford to maintain it or lacks
the trained personnel to operate it, or where it comes in
place of more vital basic primary care services. Technology
assessment needs to be seen in the context of the country and
its resources for health care.
Ethical issues in public health are no less demanding
than those related to individual clinical care. The rights of
the individual and those of the community are sometimes in
conflict. Technology, quality, the law, and ethics are closely
interrelated in public health. Well-informed and sensitive
analysis of all aspects of their development is a part of the
New Public Health. The balance between individual and
community rights is very sensitive and must be kept under
continuous surveillance.
The New Public Health is replete with technological
and ethical questions, especially in a time of cost restraint,
increasing technological potential, the public expectation
of universal access to health care, and the assumption that
everyone will live a healthy and long life. Health status has
always been linked with socioeconomic status and, despite
enormous gains, this remains true even in the most egalitar-
ian countries. Expansion of market mechanisms, such as
controlling the supply of hospital beds, doctors, and access
to referrals, competition and incentives/disincentives in
payment systems for hospital and managed care systems,
contribute to a need for dynamic health policy management
capacity. The New Public Health assumes a social responsi-
bility for health for all, using community and personal care
modalities as effectively as possible to achieve that overall
goal.
NOTE
For a complete bibliography and guidance for student
reviews and expected competencies please see companion
web site at http://booksite.elsevier.com/9780124157668
http://booksite.elsevier.com/9780124157668
http://publicationethics.org/
http://publicationethics.org/
Chapter 15 Health Technology, Quality, Law, and Ethics
BIBLIOGRAPHY
Health Technology
Agosti J.M., Goldie S.J. 2007. Introducing HPV vaccine in developing
countries–key challenges and issues. N. Engl. J. Med. 356:1908–1910.
Available at: http://www.nejm.org/doi/full/10.1056/NEJMp078053
(accessed 10.01.14).
Alliance, G.A.V.I., 2011. GAVI alliance secures lower price for rotavi-
rus vaccine. Available at: http://www.gavialliance.org/library/news/
press-releases/2012/gavi-secures-lower-price-rotavirus-vaccine/
(accessed 15.12.12).
American Cancer Society, 30 July 2013. US task force makes recommen-
dations for lung cancer screening. Available at: http://www.cancer.org/
cancer/news/us-task-force-makes-recommendations-for-lung-cancer-
screening (accessed 12.08.13).
American Congress of Obstetricians and Gynecologists, 2012. ACOG
practice bulletin no. 131: screening for cervical cancer. Obstet.
Gynecol. 120, 1222–1238. http://www.acog.org/About%20ACOG/
Announcements/New%20Cervical%20Cancer%20Screening%20
Recommendations.aspx (accessed 22.5.13).
Brenzel L., Wolfson L.J., Fox-Rushby J., Miller M., Halsey N.A. 2006.
Vaccine-preventable diseases. In: Jamison D.T., Breman J.G.,
Measham A.R., Alleyne G., Claeson M., Evans DB, et al., (Ed.) Dis-
ease control priorities in developing countries, second ed. World Bank
Washington, DC. Available at: http://www.ncbi.nlm.nih.gov/
pubmed/21250343 (accessed 7.10.2012).
Campbell, B., 2013. Regulation and safe adoption of new medical devices
and procedures. Br. Med. Bull. 1–14. Available at: http://www.ncbi.
nlm.nih.gov/pubmed/23896485 (accessed 10.01.14).
Canadian Council of Health Services Accreditation. http://www.accredita-
tion.ca/ (accessed 10.01.14).
Centers for Disease Control and Prevention. Healthcare associated infec-
tions (HAI) [updated 23 September 2013]. Available at: http://www.
cdc.gov/hai/ (accessed 10.01.14).
Centers for Disease Control and Prevention, 2009. The direct medical costs
of healthcare-associated infections in US hospitals and the benefits of
prevention. CDC, Atlanta, GA. Available at: http://www.cdc.gov/HAI/
pdfs/hai/Scott_CostPaper (accessed 14.12.12).
Centers for Disease Control and Prevention. Vaccine price list. Available
at: http://www.cdc.gov/vaccines/programs/vfc/awardees/vaccine-
management/price-list/index.html (accessed 19.09.12).
Centers for Disease Control and Prevention, 2013. Percentage of women
Aged 50–64 years who reported receiving a mammogram in the past
2 years, by health insurance status – National Health Interview Sur-
vey, United States, 1993–2010. MMWR. Morb. Mortal. Wkly. Rep.
62, 651. Available at: http://www.cdc.gov/mmwr/preview/mmwrhtml/
mm6232a5.htm?s_cid=mm6232a5_w (accessed 10.01.14).
Danish Centre for Health Technology Assessment, 2008. Health technology
assessment handbook, second ed. National Board of Health. Available
at: http://sundhedsstyrelsen.dk/~/media/C0ED080616D7410E8B-
6020B903AD0339.ashx (accessed 10.01.14).
Deyo, R.A., 2002. Cascade effects of medical technology. Annu. Rev.
Public Health 23, 23–44. Available at: http://www.annualreviews.
org/doi/abs/10.1146/annurev.publhealth.23.092101.134534 (accessed
10.01.14).
Fitzpatrick, C., Floyd, K., 2012. A systematic review of the cost and cost
effectiveness of treatment for multidrug-resistant tuberculosis. Phar-
macoeconomics 30, 63–80. Available at: http://www.ncbi.nlm.nih.
gov/pubmed/22070215 (accessed 13.12.12).
817
Flahault, A., Martin-Moreno, JM., 2013. Why do we choose to address health
2020? Public Health Reviews. 2013;35: epub ahead of print. Available
at: http://www.publichealthreviews.eu/upload/pdf_files/13/00_Edito-
rial (accessed 04.01.14).
Greenberg, M.R., 2006. The diffusion of public health innovations. Am.
J. Public Health 96, 209–210. Available at: http://www.ncbi.nlm.nih.
gov/pmc/articles/PMC1470490/ (accessed 10.11.14).
Harris, R.P., Helfand, M., Woolf, S.H., Lohr, K.N., Mulrow, C.D., Teutsch,
S.M., et al., 2001. Current methods of the US Preventive Services Task
Force. A review of the process. Am. J. Prev. Med. 20, 21–35. Avail-
able at: http://www.ncbi.nlm.nih.gov/pubmed/11306229 (accessed
10.01.14).
Health Services/Technology Assessment (HSTAT). Available at: http://
www.ncbi.nlm.nih.gov/books/NBK16710/ (accessed 17.08.12.).
Institute of Medicine, 1985. Assessing medical technologies. National
Academies Press, Washington, DC. Available at: http://www.nap.edu/
openbook.php?record_id=607 (accessed 10.01.14).
International Network of Agencies for Health Technology Assessment
(INAHTA). http://www.inahta.org/ (accessed 17.08.12).
Ikeda, N., Sapienza, D., Guerrero, R., Aekplakorn, W.,et al. 2914, Con-
trol of hypertension with medication: a comparative analysis of
national surveys in 20 countries. Bull. World Health Organ.; 92(1):
10–19. Available at: http://www.ncbi.nlm.nih.gov/pmc/articles/
PMC3865548/ (accessed 10.01.14).
Lancet, 2011. Financing HPV vaccination in developing countries. Lan-
cet 377, 1544. Available at: http://www.thelancet.com/journals/lancet/
article/PIIS0140-6736(11)60622-3/fulltext (accessed 10.01.14).
Larson, E.L., Quiros, D., Lin, S.X., 2007. Dissemination of the CDC’s
hand hygiene guideline and impact on infection rates. Am. J. Infect.
Control. 35, 666–675. Available at: http://www.ncbi.nlm.nih.gov/
pubmed/18063132 (accessed 10.01.14).
Lehoux, P., Tailliez, S., Denis, J.L., Hivon, M., 2004. Redefining health tech-
nology assessment in Canada: diversification of products and contextu-
alization of findings. Int. J. Technol. Assess 20, 325–336. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/15446762 (accessed 10.01.14).
Martelli, F., La Torre, G., Di Ghionno, E., Neroni, M., Cicchetti, A., Von
Bremen, K., et al., 2007. Health technology assessment agencies:
an international overview of organizational aspects. Int. J. Technol.
Assess 23, 414–424. Available at: http://www.ncbi.nlm.nih.gov/
pubmed/17937828 (accessed 10.01.14).
Murray, A., Lourenco, T., de Verteuil, R., Hernandez, R., Fraser, C.,
McKinley, A., et al., 2006. Clinical effectiveness and cost-effectiveness
of laparoscopic surgery for colo-rectal cancer: systematic reviews and
economic evaluation. Health Technol. Assess 10, 1–141. Available at:
http://www.ncbi.nlm.nih.gov/books/NBK62293/ (accessed 10.01.14).
National Library of Medicine, 2007. National information center on health
services research & health care technology. Etext on health technology
assessment (HTA) information resources. Available at: http://www.nlm.
nih.gov/archive//20060905/nichsr/ehta/ehta.ht (accessed 10.01.14).
Nixon, J., Stoykova, B., Glanville, J., Christie, J., Drummond, M., Klei-
jnen, J., 2000. The UK NHS economic evaluation database. Eco-
nomic issues in evaluations of health technology. Int. J. Technol.
Assess 16, 731–742. Available at: http://www.ncbi.nlm.nih.gov/
pubmed/11028129 (accessed 10.01.14).
Noorani, H.Z., Husereau, D.R., Boudreau, R., Skidmore, B., 2007. Pri-
ority setting for health technology assessments: a systematic review
of current practical approaches. Int. J. Technol. Assess 23, 310–315.
Available at: http://www.ncbi.nlm.nih.gov/pubmed/17579932 (accessed
10.01.14).
http://www.nejm.org/doi/full/10.1056/NEJMp078053
http://www.gavialliance.org/library/news/press-releases/2012/gavi-secures-lower-price-rotavirus-vaccine/
http://www.gavialliance.org/library/news/press-releases/2012/gavi-secures-lower-price-rotavirus-vaccine/
http://www.cancer.org/cancer/news/us-task-force-makes-recommendations-for-lung-cancer-screening
http://www.cancer.org/cancer/news/us-task-force-makes-recommendations-for-lung-cancer-screening
http://www.cancer.org/cancer/news/us-task-force-makes-recommendations-for-lung-cancer-screening
http://www.acog.org/About%20ACOG/Announcements/New%20Cervical%20Cancer%20Screening%20Recommendations.aspx
http://www.acog.org/About%20ACOG/Announcements/New%20Cervical%20Cancer%20Screening%20Recommendations.aspx
http://www.acog.org/About%20ACOG/Announcements/New%20Cervical%20Cancer%20Screening%20Recommendations.aspx
http://www.ncbi.nlm.nih.gov/pubmed/21250343
http://www.ncbi.nlm.nih.gov/pubmed/21250343
http://www.ncbi.nlm.nih.gov/pubmed/23896485
http://www.ncbi.nlm.nih.gov/pubmed/23896485
http://www.cdc.gov/hai/
http://www.cdc.gov/hai/
http://www.cdc.gov/HAI/pdfs/hai/Scott_CostPaper
http://www.cdc.gov/HAI/pdfs/hai/Scott_CostPaper
http://www.cdc.gov/vaccines/programs/vfc/awardees/vaccine-management/price-list/index.html
http://www.cdc.gov/vaccines/programs/vfc/awardees/vaccine-management/price-list/index.html
http://www.cdc.gov/mmwr/preview/mmwrhtml/mm6232a5.htm?s_cid=mm6232a5_w
http://www.cdc.gov/mmwr/preview/mmwrhtml/mm6232a5.htm?s_cid=mm6232a5_w
http://sundhedsstyrelsen.dk/%7e/media/C0ED080616D7410E8B6020B903AD0339.ashx
http://sundhedsstyrelsen.dk/%7e/media/C0ED080616D7410E8B6020B903AD0339.ashx
http://www.annualreviews.org/doi/abs/10.1146/annurev.publhealth.23.092101.134534
http://www.annualreviews.org/doi/abs/10.1146/annurev.publhealth.23.092101.134534
http://www.ncbi.nlm.nih.gov/pubmed/22070215
http://www.ncbi.nlm.nih.gov/pubmed/22070215
http://www.publichealthreviews.eu/upload/pdf_files/13/00_Editorial
http://www.publichealthreviews.eu/upload/pdf_files/13/00_Editorial
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1470490/
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1470490/
http://www.ncbi.nlm.nih.gov/pubmed/11306229
http://www.ncbi.nlm.nih.gov/books/NBK16710/
http://www.ncbi.nlm.nih.gov/books/NBK16710/
http://www.nap.edu/openbook.php?record_id=607
http://www.nap.edu/openbook.php?record_id=607
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3865548/
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3865548/
http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(11)60622-3/fulltext
http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(11)60622-3/fulltext
http://www.ncbi.nlm.nih.gov/pubmed/18063132
http://www.ncbi.nlm.nih.gov/pubmed/18063132
http://www.ncbi.nlm.nih.gov/pubmed/15446762
http://www.ncbi.nlm.nih.gov/pubmed/17937828
http://www.ncbi.nlm.nih.gov/pubmed/17937828
http://www.ncbi.nlm.nih.gov/books/NBK62293/
http://www.nlm.nih.gov/archive//20060905/nichsr/ehta/ehta.ht
http://www.nlm.nih.gov/archive//20060905/nichsr/ehta/ehta.ht
http://www.ncbi.nlm.nih.gov/pubmed/11028129
http://www.ncbi.nlm.nih.gov/pubmed/11028129
http://www.ncbi.nlm.nih.gov/pubmed/17579932
818
Sassi, F., 2003. Setting priorities for the evaluation of health interven-
tions: when theory does not meet practice. Health Policy 63, 141–
154. Available at: http://www.ncbi.nlm.nih.gov/pubmed/12543527
(accessed 10.01.14).
Soto, J., 2002. Health economic evaluations using decision analytic model-
ing. Principles and practices – utilization of a checklist to their devel-
opment and appraisal. Int. J. Technol. Assess 18, 94–111. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/11987445 (accessed 10.01.14).
Upshur, R.E.G., 2002. Principles for the justification of public health inter-
ventions. Can. J. Public Health 93, 101–103. Available at: http://journal.
cpha.ca/index.php/cjph/article/download/217/217 (accessed 10.01.14).
Williams, I., Bryan, S., McIver, S., 2007. How should cost-effectiveness
analysis be used in health technology coverage decisions? Evi-
dence from the National Institute for Health and Clinical Excellence
approach. J. Health Serv. Res. Policy 12, 73–79. Available at: http://
www.ncbi.nlm.nih.gov/pubmed/17407655 (accessed 10.01.14).
Woods, K., 2002. Health technology assessment for the NHS in England
and Wales. Int. J. Technol. Assess 18, 161–165. Available at: http://
www.ncbi.nlm.nih.gov/pubmed/12053415 (accessed 10.01.14).
Quality
Agency for Healthcare Research and Quality. http://www.ahrq.gov/
(accessed 14.12.12).
AHRQ. U.S. Preventive Services Task Force USPSTF), an introduction.
Available at: http://www.ahrq.gov/professionals/clinicians-providers/
guidelines-recommendations/uspstf/index.html.
American College of Cardiology (ACC), http://www.cardiosource.org/
science-and-quality.aspx (accessed 20.08.12).
Cancer Research UK, October 2012. Who is screened for breast cancer? Avail-
able at: http://www.cancerresearchuk.org/cancer-help/type/breast-cancer/
about/screening/who-is-screened-for-breast-cancer (accessed 10.01.14).
Centers for Disease Control and Prevention, 2013. Quick stats. MMWR.
Morb. Mortal. Wkly. Rep. 62, 651. Available at: http://www.cdc.gov/
mmwr/pdf/wk/mm6232 (accessed 17.08.13).
Reviews, Cochrane, http://www.cochrane.org/cochrane-reviews (accessed
14.12.12).
Dannenberg, A.L., Bhatia, R., Cole, B.L., Dora, C., Fielding, J.E., Kraft, K.,
et al., 2006. Growing the field of health impact assessment: an agenda
for research and practice. Am. J. Public Health 96, 262–270. Avail-
able at: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1470491/
(accessed 10.01.14).
Deeks, J.J., 2001. Systematic reviews in health care: systematic
reviews of evaluations of diagnostic and screening tests. BMJ 323,
157–162. Available at: http://www.ncbi.nlm.nih.gov/pmc/articles/
PMC1120791/ (accessed 10.01.14).
European Society of Cardiology, Guidelines. Available at: http://www.
escardio.org/GUIDELINES-SURVEYS/ESC-GUIDELINES/Pages/
GuidelinesList.aspx (accessed 17.08.12).
Epstein, A.J., Polsky, D., Yang, F., Yang, L., Groeneveld, P.W., 2011.
Coronary revascularization trends in the United States, 2001–2008.
JAMA 305, 1769–1776. Available at: http://www.ncbi.nlm.nih.gov/
pubmed/21540420 (accessed 13.08.13).
Fielding, J.E., Briss, P.A., 2006. Promoting evidence-based public
health policy: can we have better evidence and more action? Health
Aff. 25, 969–978. Available at: http://content.healthaffairs.org/con-
tent/25/4/969.full (accessed 10.01.14).
Institute of Medicine, 2007. State of quality improvement and imple-
mentation research: expert views. Workshop summary. Institute of
Medicine of the National Academies of Science. National Academies
Press, Washington, DC. Available at: http://www.nap.edu/catalog.
php?record_id=11986 (accessed 10.01.14).
The New Public Health
International Guideline Network Library. http://www.g-i-n.net/library/
international-guidelines-library/ (accessed 14.12.12).
Kohn, L.T., Corrigan, J.M., Donaldson, M. (Eds.), 1999. Institute of
Medicine. Committee on quality of health care in America. To err is
human: building a safer health system. National Academies Press,
Washington, DC. Available at: http://www.nap.edu/openbook.
php?isbn=0309068371 (accessed 10.01.14).
Krech, L.A., El-Hadri, L., Evans, L., Fouche, T., et al., 2014. The
medicines quality database: a free public resource. Available at:
h t t p : / / w w w. w h o . i n t / bu l l e t i n / vo l u m e s / 9 2 / 1 / 1 3 – 1 3 0 5 2 6 . p d f
(accessed 10.01.14).
National Council for Quality Assurance. http://www.ncqa.org/ (accessed
10.01.14).
Pilkington, H., Blondel, B., Drewniak, N., Jennifer Zeitlin, 2014.
Where does distance matter? Distance to the closest maternity unit
and risk of foetal and neonatal mortality in France. Eur. J. Public
Health, 1–6. Available at: http://eurpub.oxfordjournals.org/content/
early/2014/01/01/eurpub.ckt207.full (accessed 10.01.14).
Shortell, S.M., Gillies, R., Wu, F., 2010. United States innovations in
health care delivery. Public Health Rev. 32, 190–212. Available at:
http://www.publichealthreviews.eu/show/f/26 (accessed 26.10.12).
The National Audit Office, 2012. Healthcare across the UK: A compari-
son of the NHS in England. Wales and Northern Ireland, Scotland.
Available at: http://www.officialdocuments.gov.uk/document/hc1213/
hc01/0192/0192 (accessed 10.01.14).
UK National Institute for Health and Care Excellence (NICE). Available
at: http://www.nice.org.uk/ (accessed 17.08.12).
US Department of Health and Human Services, August 2008. Office of
Inspector General. Adverse events in hospitals: overview of key
issues. Available at: http://oig.hhs.gov/oei/reports/oei-06-07-00470.
pdf (accessed 17.08.13).
US Food and Drug Administration, 2007. Department of Health and
Human Services. Amendment to the current good manufacturing prac-
tice regulations for finished pharmaceuticals. Direct final rule. Federal
Register 72, 68064–68070. Available at: http://www.ihi.org/Pages/
default.aspx (accessed 17.08.12).
US National Library of Medicine. Available at: http://www.nlm.nih.gov/
(accessed 13.02.14).
US National Information Center on Health Services Research. https://www.
nlm.nih.gov/nichsr/ (accessed 13.02.14).
US Preventive Services Task Force. Guide to preventive services. Report
of the US Preventive Services Task Force (USPSTF). http://www.ahrq.
gov/professionals/clinicians-providers/guidelines-recommendations/
uspstf/index.htm. Available at: http://www.ahrq.gov/clinic/USpstfix.
htm (accessed 13.02.14).
Law
American Society of Law, Medicine and Ethics. The public’s health and
the law in the 21st century: Second annual partnership conference on
public health law. J. Law. Med. Ethics. Special Suppl. 31 (4). Avail-
able at: http://www.stacks.cdc.gov/view/cdc/6751/cdc_6751_DS1
(accessed 10.1.14).
Annas, G., 1998. Human rights and health – the universal declaration
of human rights at 50. N. Engl. J. Med. 339, 1778–1781. Available
at: http://www.nejm.org/doi/full/10.1056/NEJM199812103392411
(accessed 10.01.14).
Burris, S., Wagenaar, A.C., Mello, M.M., 2010. Making the case for laws
that improve health: A framework for public health law research. Mil-
bank. Q. 88 (2), 169–210. Available at; http://www.ncbi.nlm.nih.gov/
pmc/articles/PMC2980343/ (accessed 10.01.14).
http://www.ncbi.nlm.nih.gov/pubmed/12543527
http://www.ncbi.nlm.nih.gov/pubmed/11987445
http://journal.cpha.ca/index.php/cjph/article/download/217/217
http://journal.cpha.ca/index.php/cjph/article/download/217/217
http://www.ncbi.nlm.nih.gov/pubmed/17407655
http://www.ncbi.nlm.nih.gov/pubmed/17407655
http://www.ncbi.nlm.nih.gov/pubmed/12053415
http://www.ncbi.nlm.nih.gov/pubmed/12053415
http://www.ahrq.gov/
http://www.ahrq.gov/professionals/clinicians-providers/guidelines-recommendations/uspstf/index.html
http://www.ahrq.gov/professionals/clinicians-providers/guidelines-recommendations/uspstf/index.html
http://www.cardiosource.org/science-and-quality.aspx
http://www.cardiosource.org/science-and-quality.aspx
http://www.cancerresearchuk.org/cancer-help/type/breast-cancer/about/screening/who-is-screened-for-breast-cancer
http://www.cancerresearchuk.org/cancer-help/type/breast-cancer/about/screening/who-is-screened-for-breast-cancer
http://www.cdc.gov/mmwr/pdf/wk/mm6232
http://www.cdc.gov/mmwr/pdf/wk/mm6232
http://www.cochrane.org/cochrane-reviews
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1470491/
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1120791/
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1120791/
http://www.escardio.org/GUIDELINES-SURVEYS/ESC-GUIDELINES/Pages/GuidelinesList.aspx
http://www.escardio.org/GUIDELINES-SURVEYS/ESC-GUIDELINES/Pages/GuidelinesList.aspx
http://www.escardio.org/GUIDELINES-SURVEYS/ESC-GUIDELINES/Pages/GuidelinesList.aspx
http://www.ncbi.nlm.nih.gov/pubmed/21540420
http://www.ncbi.nlm.nih.gov/pubmed/21540420
http://content.healthaffairs.org/content/25/4/969.full
http://content.healthaffairs.org/content/25/4/969.full
http://www.nap.edu/catalog.php?record_id=11986
http://www.nap.edu/catalog.php?record_id=11986
http://www.g-i-n.net/library/international-guidelines-library/
http://www.g-i-n.net/library/international-guidelines-library/
http://www.nap.edu/openbook.php?isbn=0309068371
http://www.nap.edu/openbook.php?isbn=0309068371
http://www.who.int/bulletin/volumes/92/1/13-130526
http://web.ncqa.org/tabid/577/Default.aspx
http://eurpub.oxfordjournals.org/content/early/2014/01/01/eurpub.ckt207.full
http://eurpub.oxfordjournals.org/content/early/2014/01/01/eurpub.ckt207.full
http://www.publichealthreviews.eu/show/f/26
http://www.officialdocuments.gov.uk/document/hc1213/hc01/0192/0192
http://www.officialdocuments.gov.uk/document/hc1213/hc01/0192/0192
http://www.nice.org.uk/
http://oig.hhs.gov/oei/reports/oei-06-07-00470
http://oig.hhs.gov/oei/reports/oei-06-07-00470
http://www.ihi.org/Pages/default.aspx
http://www.ihi.org/Pages/default.aspx
http://www.nlm.nih.gov/
https://www.nlm.nih.gov/nichsr/
https://www.nlm.nih.gov/nichsr/
http://www.ahrq.gov/professionals/clinicians-providers/guidelines-recommendations/uspstf/index.htm
http://www.ahrq.gov/professionals/clinicians-providers/guidelines-recommendations/uspstf/index.htm
http://www.ahrq.gov/professionals/clinicians-providers/guidelines-recommendations/uspstf/index.htm
http://www.ahrq.gov/clinic/USpstfix.htm
http://www.ahrq.gov/clinic/USpstfix.htm
http://www.stacks.cdc.gov/view/cdc/6751/cdc_6751_DS1
http://www.nejm.org/doi/full/10.1056/NEJM199812103392411
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2980343/
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2980343/
Chapter 15 Health Technology, Quality, Law, and Ethics
Centers for Disease Control and Prevention, Collaborating Center for the
Law and the Public’s Health. Available at: http://www.publichealthlaw.
net/ (accessed 17.08.12).
Fox, D., Kramer, M., Standish, M., 2003. From public health to population
health: how can law redefine the playing field? J. Law Med. Ethics.
31, 21–29. Available at: http://www.stacks.cdc.gov/view/cdc/6751/
cdc_6751_DS1 (accessed 10.01.14).
Goodman, R.A., Moulton, A., Matthews, G., Shaw, F., Kocher, P., Mensah,
G., et al., 2006. Law and public health at CDC. MMWR. Morb. Mor-
tal. Wkly. Rep. 55, 29–33. Available at: http://www.cdc.gov/mmwr/
preview/mmwrhtml/su5502a11.htm (accessed 10.01.14).
Gostin, L.O., 2000. Public health law in a new century. Part II: Public
health powers and limits. JAMA 283, 2979–2984. Available at: http://
www.ncbi.nlm.nih.gov/pubmed/10865277 (accessed 10.01.14).
Gostin, L.O., 2000. Public health law in a new century: Part III: Public
health regulation: a systematic evaluation. JAMA 283, 3118–3122.
Available at: http://www.ncbi.nlm.nih.gov/pubmed/10865307
(accessed 10.01.13).
Ruhl, S., Stephens, M., Locke, P., 2005. The role of NGOs in public health
law. J. Law Med. Ethics. 31, 76–77. Available at: http://www. stacks.cdc.
gov/view/cdc/6751/cdc_6751_DS1 (page 76). (accessed 10.01.14).
Thiel, K.S., 2003. New developments in public health case law. J. Law
Med. Ethics. 31, 86–87.
Zasa, S., Clymer, J., Upmeyer, L., Thacker, S., 2003. Using science-based
guidelines to shape public health law. J. Law Med. Ethics. 31, 65–67.
Ethics
Aceijas, C., Brall, C., Schröder-Bäck, P., Otok, R., Maeckelberghe, E.,
Stjernberg, L., et al., 2012. Teaching ethics in schools of public
health in the European Region: findings from a screening survey.
Public Health Rev. 34. Epub ahead of print. Available at: www.publi-
chealthreviews.eu (accessed 05.12.12).
American College of Epidemiology, 2000. Ethics guidelines. Ann. Epidemiol.
10, 487–497. Available at: http://acepidemiology.org/content/ethics.
American Public Health Association, Principles of the ethical prac-
tice of public health. Available at: http://www.apha.org/NR/
rdonlyres/1CED3CEA-287E-4185-9CBD-BD405FC60856/0/ethics-
brochure (accessed 17.08.12).
Bachrach, S., 2004. In the name of public health – Nazi racial hygiene. N.
Engl. J. Med. 351, 417–420. Available at: http://content.nejm.org/cgi/
content/full/351/5/417?ijkey=8bee5b41cf11cadc3a826ad16fc2deef59
aa8f32&keytype2=tf_ipsecsha (accessed 05.12.12).
Bayer, R., Fairchild, A., 2004. The genesis of public health ethics. Bio-
ethics 18, 473–492. Available at: http://www.ncbi.nlm.nih.gov/
pubmed/15580720 (accessed 10.01.14).
Birn, A.E., Molina, N., 2005. In the name of public health. Am. J. Pub-
lic Health 95, 1095–1097. Available at: http://www.ncbi.nlm.nih.gov/
pmc/articles/PMC1449322/ (accessed 10.01.14).
Carter, S.M., Cribb, A., Allegrante, J.P., 2012. How to think about health
promotion ethics. Public Health Rev. 34. Epub ahead of print. Avail-
able at: www.publichealthreviews.eu (accessed 05.12.12).
Centers for Disease Control and Prevention, US Public Health Service
syphilis experiment at Tuskegee. Available at: http://www.cdc.gov/
tuskegee/timeline.htm (accessed 17.08.12).
Centers for Disease Control, Advancing excellence and integrity of CDC
science. Available at: http://www.cdc.gov/od/science/integrity/pheth-
ics/ (accessed 10.01.14).
Coleman, C.H., Bouësseau, M.-C., Reis, A., 2008. The contribution of
ethics to public health. Bull. World Health Org. 86, 578–579. Avail-
able at: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2649472/
(accessed 10.01.14).
819
Coughlin, S.S., Barker, A., Dawson, A., 2012. Ethics and scientific integ-
rity in public health, epidemiological and clinical research. Pub-
lic Health Rev. 34 (1). Available at: www.publichealthreviews.eu
(accessed 05.12.12).
Coughlin, S.S., 2006. Ethical issues in epidemiologic research and public
health practice. Emerg Themes Epidemiol 3, 3–16. Available at: http://
www.ncbi.nlm.nih.gov/pmc/articles/PMC1594564/ (accessed 10.01.14).
Dawson, A., Paul, Y., 2006. Mass public health programs and the obliga-
tions of sponsoring and participating organizations. J. Med. Ethics 32,
580–583. Available at: http://www.ncbi.nlm.nih.gov/pmc/articles/
PMC2563318/ (accessed 10.01.14).
Dawson, A., 2004. Vaccination and the prevention problem. Bioethics 18,
515–530. Available at: http://onlinelibrary.wiley.com/doi/10.1111/
j.1467-8519.2004.00414.x/abstract?deniedAccessCustomisedMessag
e=&userIsAuthenticated=false (accessed 10.01.14).
El Amin, A.N., Parra, M.T., Kim-Farley, R., Fielding, J.E., 2012. Ethical
issues concerning vaccination requirements. Public Health Rev. 34
(1). Available at: www.publichealthreviews.eu (accessed 05.12.12).
Kass, N., 2004. Public health ethics: from foundations and frameworks
to justice and global public health. J. Law Med. Ethics. 32, 232–
242. Available at: http://www.ncbi.nlm.nih.gov/pubmed/15301188
(accessed 10.01.14).
Lee, L.M., 2012. Guest editorial: Public health ethics theory: review and
path to convergence. Public Health Rev. 34 (1). Available at: www.
publichealthreviews.eu (accessed 05.12.12).
Lynch, H.F., 2012. Ethical evasion or happenstance and hubris? The US
Public Health Service STD Inoculation Study. Hastings Cent. Rep.
42, 30–38. Available at: http://onlinelibrary.wiley.com/doi/10.1002/
hast.17/abstract (accessed 10.01.14).
Nuremberg Code, 1949. Trials of war criminals before the Nurem-
berg military tribunals under control council law no. 10, Vol. 2. US
Government Printing Office, Washington, DC. pp. 181–182. Organ-
isation for Economic Co-operation and Development, 2007. Global
Science Forum. Best practices for ensuring scientific integrity and
preventing misconduct. OECD, Available at: http://www.oecd.org/
dataoecd/37/17/40188303 (accessed 19.5.2012).
Reverby, S.M., 2012. Ethical failures and history lessons: the US
Public Health Service research studies in Tuskegee and Guatemala.
Public Health Rev. 34. Epub ahead of print. Available at: http://
www.publichealthreviews.eu/upload/pdf_files/11/00_Reverby
(accessed 10.01.14).
Thieren, M., Mauron, A., 2007. Nuremberg Code turns 60. B World Health
Organ 85, 573. Available at: http://www.publichealthreviews.eu/
upload/pdf_files/11/00_Reverby (accessed 10.01.14).
Thomas, J.C., Sage, M., Dillenberg, J., Guillory, V.J., 2002. A code of
ethics for public health. Am. J. Public Health 92, 1057–1060. Avail-
able at: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1447186/
(accessed 10.01.14).
US Presidential Commission for the Study of Bioethical Issues, 2011.
Creation of the commissions. Available at: http://bioethics.gov/cms/
creation-of-the-commission (accessed 05.12.12).
United Nations Programme on HIV/AIDS (UNAIDS), 2013. Ending
overly broad criminalisation, non-disclosure, exposure and transmis-
sion: Critical scientific, medical and legal considerations. Available
at: http://www.unaids.org/en/media/unaids/contentassets/documents/
document/2013/05/20130530_Guidance_Ending_Criminalisation
(accessed 20.1.14).
Zusman, S.P., 2012. Water fluoridation in Israel, 1968–2012. Public Health
Rev. 34. Epub ahead of print. Available at: http://www.publichealthre-
views.eu/upload/pdf_files/11/00_Zusman (accessed 10.01.14).
http://www.stacks.cdc.gov/view/cdc/6751/cdc_6751_DS1
http://www.stacks.cdc.gov/view/cdc/6751/cdc_6751_DS1
http://www.cdc.gov/mmwr/preview/mmwrhtml/su5502a11.htm
http://www.cdc.gov/mmwr/preview/mmwrhtml/su5502a11.htm
http://www.ncbi.nlm.nih.gov/pubmed/10865277
http://www.ncbi.nlm.nih.gov/pubmed/10865277
http://www.ncbi.nlm.nih.gov/pubmed/10865307
http://www.stacks.cdc.gov/view/cdc/6751/cdc_6751_DS1
http://www.stacks.cdc.gov/view/cdc/6751/cdc_6751_DS1
http://refhub.elsevier.com/B978-0-12-415766-8.00015-X/ref0300
http://refhub.elsevier.com/B978-0-12-415766-8.00015-X/ref0300
http://refhub.elsevier.com/B978-0-12-415766-8.00015-X/ref0305
http://refhub.elsevier.com/B978-0-12-415766-8.00015-X/ref0305
http://acepidemiology.org/content/ethics
http://www.apha.org/NR/rdonlyres/1CED3CEA-287E-4185-9CBD-BD405FC60856/0/ethicsbrochure
http://www.apha.org/NR/rdonlyres/1CED3CEA-287E-4185-9CBD-BD405FC60856/0/ethicsbrochure
http://www.apha.org/NR/rdonlyres/1CED3CEA-287E-4185-9CBD-BD405FC60856/0/ethicsbrochure
http://content.nejm.org/cgi/content/full/351/5/417?ijkey=8bee5b41cf11cadc3a826ad16fc2deef59aa8f32%26keytype2=tf_ipsecsha
http://content.nejm.org/cgi/content/full/351/5/417?ijkey=8bee5b41cf11cadc3a826ad16fc2deef59aa8f32%26keytype2=tf_ipsecsha
http://content.nejm.org/cgi/content/full/351/5/417?ijkey=8bee5b41cf11cadc3a826ad16fc2deef59aa8f32%26keytype2=tf_ipsecsha
http://www.ncbi.nlm.nih.gov/pubmed/15580720
http://www.ncbi.nlm.nih.gov/pubmed/15580720
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1449322/
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1449322/
http://www.cdc.gov/tuskegee/timeline.htm
http://www.cdc.gov/tuskegee/timeline.htm
http://www.cdc.gov/od/science/phec/
http://www.cdc.gov/od/science/phec/
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2649472/
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1594564/
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1594564/
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1594564/
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1594564/
http://onlinelibrary.wiley.com/doi/10.1111/j.1467-8519.2004.00414.x/abstract?deniedAccessCustomisedMessage=%26userIsAuthenticated=false
http://onlinelibrary.wiley.com/doi/10.1111/j.1467-8519.2004.00414.x/abstract?deniedAccessCustomisedMessage=%26userIsAuthenticated=false
http://onlinelibrary.wiley.com/doi/10.1111/j.1467-8519.2004.00414.x/abstract?deniedAccessCustomisedMessage=%26userIsAuthenticated=false
http://www.ncbi.nlm.nih.gov/pubmed/15301188
http://onlinelibrary.wiley.com/doi/10.1002/hast.17/abstract
http://onlinelibrary.wiley.com/doi/10.1002/hast.17/abstract
http://www.oecd.org/dataoecd/37/17/40188303
http://www.oecd.org/dataoecd/37/17/40188303
http://www.publichealthreviews.eu/upload/pdf_files/11/00_Reverby
http://www.publichealthreviews.eu/upload/pdf_files/11/00_Reverby
http://www.publichealthreviews.eu/upload/pdf_files/11/00_Reverby
http://www.publichealthreviews.eu/upload/pdf_files/11/00_Reverby
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1447186/
http://bioethics.gov/cms/creation-of-the-commission
http://bioethics.gov/cms/creation-of-the-commission
http://www.unaids.org/en/media/unaids/contentassets/documents/document/2013/05/20130530_Guidance_Ending_Criminalisa%20
http://www.unaids.org/en/media/unaids/contentassets/documents/document/2013/05/20130530_Guidance_Ending_Criminalisa%20
http://www.publichealthreviews.eu/upload/pdf_files/11/00_Zusman
http://www.publichealthreviews.eu/upload/pdf_files/11/00_Zusman
Introduction
Innovation, Regulation, and Quality Control
Appropriate Health Technology
Priority Interventions in Low- and Medium-Income Countries
Priority Selection in High-Income Countries
Health Technology Assessment
Technology Assessment in Hospitals
Technology Assessment in Prevention and Health Promotion
Technology Assessment in National Health Systems
Dissemination of Technology
Diffusion of Technology
Quality Assurance
Adverse Events and Negligence
Licensure and Certification
Health Facility Accreditation
Peer Review
Tracer Conditions
Setting Standards
Algorithms and Clinical Guidelines
Organization of Care
Diagnosis-Related Groups
Managed Care
Performance Indicators
Consumerism and Quality
The Public Interest
Total Quality Management
Public Health Law
Environmental Health
Public Health Law Reform
Ethical Issues In Public Health
Individual and Community Rights
Tragic Deviations in Public Health Ethics
Human Experimentation
Ethics in Public Health Research
Ethics in Patient Care
Sanctity of Life Versus Euthanasia
The Imperative to Act or Not Act in Public Health
Summary
Note
Bibliography
Health Technology
Quality
Law
Ethics
Health Technology, Quality, Law,
and Ethics
Chapter 15
The New Public Health. http://dx.doi.org/10.1016/B978-0-12-415766-8.00015-X
Copyright © 2014 Elsevier Inc. All rights reserved.
Learning Objectives
Upon completion of this chapter, the student should be
able to:
1. Describe responsibility for and methods of assessing and
regulating technological developments in health care;
2. Describe methods of health facility accreditation and
peer review;
3. Describe the concept of total quality management;
4. Identify and discuss ethical and legal issues in national
health systems;
5. Apply ethical considerations to health issues in his or
her home setting.
INTRODUCTION
Management of a production or a service system requires
attention to the quality of personnel as much as to the system
in which they work. Their motivation and sense of participa-
tion, the scientific and technological level of the program,
and the legal and ethical standards of individual providers
and of the system as a whole, are all important to the quality
of care provided and equity of health status achieved.
Quality is the result of input and process, and is measured
by outcome or performance indicators as well as perception
of the service by the patients, the staff, and the community
as a whole. Input refers to the institutional and financial
resources for education, human resources, supplies, medica-
tions, vaccines, diagnostic capacity, and services available.
Process refers to the use of those resources, including peer
group expectations of professionalism. Outcomes generally
include measures of morbidity, mortality, and functional sta-
tus of the patient and the population. Defining and measuring
achievements of national health objectives and targets, the
methods of financing services, and the efficiency of organi-
zation help to determine quality. Training, supply, and dis-
tribution of health personnel are all determinants of access
to and quality of care. Continuous and adequate availability
of essential preventive, diagnostic, and treatment services,
as well as accountability and internal methods of promoting
standards, are all elements of the quality of a health service
771
for the individual, the population as a whole, and groups
within the population with special needs.
The content and standards of service are assessed
through organized review by professional peers within
an institution, and from outside. Peer review within an
institution and external evaluation by accreditation or
governmental inspection, based on cumulative evidence
and the recognized current “state of the art”, contribute
to accountability and improved quality of care. Continu-
ous quality improvement (CQI) among health care teams
and organizations includes regular practice assessments,
evidence gathering, remediation, and re-evaluation, which
will be discussed later in this chapter. The perception of
the services by the community, along with the knowledge,
attitudes, beliefs, and practices of health, are all vital to
improvement of health status.
Health-related technology is also in a continuing state
of change. Systematic review and absorption of new scien-
tific knowledge, technology, and innovations are essential
to promote and renew health care methods. Public health
serves in a regulatory role to assure high-quality care to the
individual and the community. New technology, whether in
the form of diagnostic procedures, new drugs, devices, or
vaccines, or new types of health personnel, requires evalu-
ation for effectiveness and appropriateness to the system.
Technology assessment also involves epidemiological and
economic aspects of effectiveness. Failure to continuously
monitor developments and to assimilate those that are demon-
strably successful is an ethical and management failure which
tragically costs many millions of lives from preventable dis-
eases yearly, such as in delayed adoption of well-proven vac-
cines or tobacco restriction legislation. This is due to political
failure even more than professional weakness, and constitutes
one of the saddest ethical dilemmas of public health: failure
to convince policy makers of the prime importance of health
promotion and disease prevention in the health sector.
Ethics and law in public health reflect the values of a
society. They inevitably evolve as they face dramatic social,
economic, demographic, and political changes; new health
challenges; and new technological and scientific possibili-
ties for improving health. Ethics are the foundation of the
value systems of a society and thus of its health concepts.
772
Biblical sources articulated values of the Ten Command-
ments, Sanctity of Human Life, Improve the World, along
with the Hippocratic Oath of physicians to “do good and do
no harm”. Modern definitions of public health and bioethics
emerged from lessons learned from the horrors of eugenics
and genocide in the twentieth century with humanistic pre-
cepts of “Universal Human Rights” and “Health for All” in
the recent era (see Chapters 1 and 2).
The law is both permissive and restrictive. It sets the basic
responsibilities, powers, and limitations of public health prac-
tice, with legislation and court decisions. Innovations in the
technology of medical care and public health are powerful
forces contributing to increased longevity, quality of life, and
economic growth, but they also bring challenges to imple-
mentation impeded by additional costs of the health system
and slow adaptation in countries with the greatest need.
These are challenges to national and international political,
The New Public Health
organizational, and economic systems to address health with
the full potential for saving lives. Determining standards of
“good practice” is a continuing process with the rapid devel-
opment of new knowledge, technology, and experience.
The law is a dynamic process involving old and new
legislation, court decisions, and new issues not previously
faced, often following rather than anticipating public health
issues. Public health has had both positive and negative ethi-
cal experiences and continues to face new issues with chang-
ing population needs, technology, science, and economics.
INNOVATION, REGULATION, AND
QUALITY CONTROL
Health care technology has advanced with an increasing
stream of innovation since the seventeenth-century epidemi-
ological discoveries of Lind on scurvy (1747) and smallpox
TABLE 15.1 Health Care Innovations from the Seventeenth to the Twenty-First Centuries
Period Selected Highlights of Scientific, Technological, and Organizational Innovations in Health
17th century Biological basis of disease (Descartes), circulation of blood (Harvey), microscope (Leeuwenhoek)
18th century Thermometer, lime juice supplements (Lind, 1756), vaccination (Jenner, 1796), surgical anatomy (Hunter), clinical
sciences (Sydenham)
19th century Miasma theory vs germ theory; inventions of stethoscope (1816), blood transfusion (1818), anesthesia (1842),
hypodermic syringe (1852), ophthalmoscope (1851), laryngoscope (1855), pasteurization of wine, beer, milk
(1860s), cholera vaccine (1879), X-ray (1895), blood pressure cuff (1896); sanitation, municipal health depart-
ments, chlorination and filtration of community water supplies, antisepsis, Braille printing, hygiene in obstetrics,
nursing, microscopic pathology, pathological chemistry, microbiology, vaccines, X-ray, national health insurance,
syringes, well-child care, aspirin (1899), Bismarkian social insurance (1881)
1900–1930 Electrocardiogram (1901), Flexner report on medical education, salvarsan, insulin (1922), blood groups, vitamins,
conquest of yellow fever, vitamin B, vaccine for diphtheria (1923), tetanus vaccine (1924), electroencephalogram
(1924), iron lung respirator (1927), Social Security Act (1935), cost–benefit analysis, food fortification (iodized salt,
flour with vitamin B complex), improved work safety
1931–1945 Mandatory fortification of milk, salt, and flour in USA (1941), Pap test (1942), penicillin (1928), streptomycin,
randomized clinical trials, antimalarial drugs, vector controls, dialysis machine (1945)
1946–1960 Contact lens (1948), DNA double helix (1953), heart–lung bypass machine (1953), ultrasound (1955), cardiac
pacemaker (1958), Salk polio vaccine (1955), kidney transplant (1959), advances in vaccines, antihypertensives,
psychotropic drugs, cancer chemotherapy, prepaid group practice, UK National Health Service (1948), Medicare
in Canada (1946–1971)
1961–1980 Oral polio vaccine (Sabin), hip replacement (1962), oral rehydration therapy, measles vaccine (1964), coronary
bypass (1964), Medicare, Medicaid (1965), mammography (1965), portable defibrillator (1965), measles–mumps–
rubella vaccine, cost-effectiveness analysis, open heart surgery, pacemakers, organ transplantation, computed
tomography (CT), eradication of smallpox (1972), health maintenance organizations (HMOs), diagnosis-related
groups (DRGs), district health systems
1981–2000 Health promotion (1987), magnetic resonance imaging (MRI), positron emission tomography (PET), endoscopic
surgery, Helicobacter pylori and chronic peptic ulcer disease (1982), managed care, Haemophilus influenzae b
(Hib) vaccine, statins (1987), poliomyelitis eradication campaign (1982), local eradication of beta-thalassemia,
pandemic of HIV (1981 onward), AZT antiretroviral approved (1987), robotic surgery (2000)
2001–2013 Millennium Development Goals (MDGS 2000) with substantial progress achieved, managing emergencies of mass
terrorism and natural disasters, new vaccines (HPV), managing epidemics of measles and influenza, new diag-
nostic technologies, flour fortification to prevent birth defects, HIV still deadly but effective treatment and control
measures, new treatments for hepatitis C, robotic surgery, nanotechnology, scientific advances with great potential
benefit, Affordable Care Act (2010), Accountable Care Organizations
Source: Adapted from Health United States 2009. Special Feature: Medical technology. Introduction and timeline. 2009. Available at: http://www.ncbi.nlm.
nih.gov/books/NBK44737/#specialfeature.sec1 [Accessed 15 December 2012]. See Historical Markers in Chapter 1.
http://www.ncbi.nlm.nih.gov/books/NBK44737/#specialfeature.sec1
http://www.ncbi.nlm.nih.gov/books/NBK44737/#specialfeature.sec1
Chapter 15 Health Technology, Quality, Law, and Ethics
vaccination by Jenner (1796), to the dramatic innovations of
the end of the twentieth century (Table 15.1). The pace of
innovation is rapid, creating the need for regulation, quality
control, and technology assessment.
National governments are responsible for assuring that
pharmaceuticals, biological products, food, and the environ-
ment are regulated to protect the public. In some countries,
these responsibilities are divided among ministries of trade,
industry, commerce, health, and environment. In a federal
system of government, there may be a division of responsi-
bility among federal, state, and local government, but with
the national government often providing national standards
and leadership in this area.
Government regulation and control are meant to protect
the public health. The US Food and Drug Administration
(FDA) is responsible for enforcing the Food, Drug and Cos-
metic Act, the Fair Packaging and Labeling Act, sections of
the Public Health Services Act relating to biological prod-
ucts for control of communicable diseases, and the Radiation
Control for Health and Safety Act. The FDA is a Division
of the Department of Health and Human Services (DHHS).
State governments have the authority to supervise pharma-
cies and their products, which may be marketed across dif-
ferent states. All national governments have departments
responsible for conducting supervision of food, drugs, and
medical devices, often relying on international standards.
Drugs and devices include all drugs, diagnostic prod-
ucts, blood and its derivatives, biologicals, veterinary
medicines, and medicated premixed animal products. All
manufacturers and distributors are required by law to reg-
ister these products with the national authority in order to
be allowed to market or import them. All countries need to
govern the food, drugs, vaccines, and cosmetics regulated
for production, importation, marketing, and use within
their jurisdiction. Organizations within each government
must be responsible for assuring the consumer that foods
are pure (unadulterated) and wholesome, safe to eat, and
produced under sanitary conditions; that drugs and medical
devices are safe and effective for their intended uses; that
cosmetics are safe and made from appropriate ingredients;
and that labeling is truthful, informative, and not deceptive.
National authorities such as the FDA, under legisla-
tion and regulations, govern both domestic and imported
products. They establish and enforce standards, or adopt
external agency standards as a “gold standard”, meaning
that products meet high standards of safety and efficacy.
The FDA also monitors and inspects contents manufactur-
ing standards under good manufacturing practices (GMPs),
which includes regular accreditation of a manufacturer’s
facilities, staffing, planning, and monitoring capacity. Test-
ing of products is carried out to assess safety, potency, and
toxicity using accepted reference laboratory procedures as
published in the compendium Official Methods of Analysis
of the Association of Official Analytical Chemists.
773
When federal, state, or local investigators, sometimes
known as consumer safety officers, detect through labora-
tory monitoring or observe conditions that may result in a
public health hazard, and violation of food and drug laws
and regulations, they issue a written report to the manufac-
turers with recommendations for correcting the conditions.
In more blatant cases, the authorities may issue urgent recall
or seizure orders for products in violation of standards con-
stituting a danger to public health, such as contaminated
products, lead-painted children’s toys, or contaminated
foods causing foodborne disease outbreaks, which occur
not infrequently in imported and domestically produced
foods in the USA. The Los Angeles County Department of
Health inspects restaurants regularly and places a promi-
nent placard in the window giving a grade A, B, or C to
the restaurant for sanitation and safety. Those given D rat-
ings may be closed until specified faults are eliminated, or
a restaurant may be closed permanently. State governments
require restaurants to list calorie and salt content of foods
on their menus as part of the public health efforts to reduce
obesity.
Supervision of food standards may also fail, as occurred
in Israel in 2004 when total absence of vitamin B1 in a soy-
based baby formula imported from Germany resulted in
three deaths and permanent brain damage to other infants
due to severe beriberi. This episode led to criminal charges
in 2008 of negligence resulting in death against the own-
ers of the company that imported or produced the foods and
staff members of the Ministry of Health. Animal foods in
2007 and infant milk products imported from China in 2008
were found to be contaminated with melamine, which was
meant to mimic protein content but was toxic in combination
with other chemicals used. The infant formula caused seri-
ous illness in some 300,000 Chinese babies and six deaths.
The FDA and its counterparts in each country are
responsible for regulation of:
l food – foodborne illness, nutritional content, labeling,
dietary supplements
l drugs – prescription drugs and generics, over-the-counter
products
l medical devices – pacemakers, stents, contact lenses,
hearing aids
l biologics – vaccines, blood products
l animal feed and drugs – for livestock, pets
l cosmetics – safety, labeling
l radiation-emitting products – cell phones, lasers, micro-
waves
l combination products.
New drugs and biological products for human use are
required to pass rigorous review before approval for mar-
keting is granted. Applications are submitted by the manu-
facturer or sponsor with acceptable scientific data including
test results to evaluate the safety and effectiveness of the
774
product for the conditions under which it is being offered.
All manufacturers of drugs are required to be registered
with the FDA and to meet its requirements for each drug
produced and marketed, including the reporting of adverse
reactions and labeling criteria. Manufacturers are required
to operate in conformity with current GMPs, which include
stringent control over manufacturing processes, personnel
training, computerized operations, and testing of finished
products. The FDA publishes guidelines to help manufac-
turers to familiarize themselves with current standards. The
United States Pharmacopoeia, National Formulary, and
WHO Model Formulary 2008 are the official listings of
approved products.
Medical devices are also regulated by the FDA. Thou-
sands of products for health care purposes require premarket
approval, ranging from basic articles such as thermometers,
tongue depressors, and intrauterine devices (IUDs), to more
complex devices such as cardiac monitors, pacemakers,
breast implants, and kidney dialysis machines. These prod-
ucts are subject to controls of GMPs, labeling, registration
of the manufacturer, and performance standards.
Monitoring for efficacy and potential hazards has
been strengthened since the 1970s as a result of find-
ings of long-term carcinogenic and mutagenic effects of
estrogens, and toxic effects of chloramphenicol on bone
marrow. The drug thalidomide, widely used as an antin-
auseant and sleeping pill for pregnant women in Europe,
Canada, and Australia in the 1960s, was not approved by
the US FDA. This drug was found to cause large numbers
of serious birth deformities leading to its being banned in
most countries. Controls of blood and blood products have
been strengthened since the transmission of human immu-
nodeficiency virus (HIV), hepatitis B, and hepatitis C by
contaminated blood products in the 1980s. The responsi-
bility of this regulatory function is well illustrated by the
1995 criminal conviction of several senior health officials
in France for failing to stop the use of blood products con-
taminated with HIV in the mid-1980s. Concern regarding
possible carcinogenic effects of silicone breast implants
led to legal action and greater controls of all implantable
products. A balance between safety and well-regulated
approval of new products requires a highly professional
and motivated regulatory agency, well-developed proce-
dures, and well-trained staff.
The concepts of standardization of GMPs for pharma-
ceutical products and written protocols for good medi-
cal practice or good public health practice are accepted
norms based on best available evidence of current scientific
knowledge and experience. Recommended immunization
schedules, water quality, ambient air standards, food fortifi-
cation, and screening programs for early stages of diabetes
are examples of accepted practice that have become rec-
ommended standards of public health practice, paralleling
qualitative measures developed in clinical care.
The New Public Health
APPROPRIATE HEALTH TECHNOLOGY
The concept of intermediate technology pioneered by Dr
Ernst Schumacher in the 1960s proposed the development
of simple and inexpensive technology for developing coun-
tries such as India to promote local economic development.
Environmentally sustainable development and sources of
energy, energy conservation, and reductions in toxic and
harmful emissions are encouraged. In recent years ideas
have included small loan systems for rural entrepreneurs in
developing countries, and the use of simple cell phones for
communication, farm produce marketing, cash transfers in
remote areas without banking services, and many others.
Now called appropriate technology, this topic has gained
adherence in the health field in the search for low-cost and
simple techniques for preventing and managing common
illnesses.
Appropriate technology is defined by the World Health
Organization (WHO) as the level of medical technology
needed to improve health conditions in keeping with the
epidemiological, demographic, and financial situation of
each country. All countries have limited resources and
so must select strategies of health care and appropriate
technology to use those resources effectively to achieve
health benefits. Improved water pumps, solar energy, rain-
water collection and water reservoirs, sanitary latrines,
fly traps, insecticide-impregnated bed nets, biogas from
animal waste, improved home cooking stoves, and many
other simple devices can make enormous differences in
local sustainable agriculture, economic growth, and living
conditions. Cell phones are now used to monitor health
conditions such as hypertension, diabetes control, weight
and body mass index, and other non-communicable con-
ditions, and to transmit imaging from remote areas to
specialists in medical centers who can provide test read-
ings online. Simple, affordable, portable information tech-
nology can effectively support public health programs,
even in resource-poor environments.
The topics discussed in the growing literature and meet-
ings of the International Society of Technology Assessment
in Health Care represent the dynamic field of technology
assessment. The issues range from economic evaluation
of pharmaceuticals to modeling approaches, measures of
quality of life, technology dissemination and impact, and
outcomes measurement. The range of issues also includes
finance and health insurance, health care in developing
countries, informatics, telemedicine, technologies for the
disabled, screening, and cost-effectiveness. Evaluations
in the scrutiny of both high- and low-technology services
based on a combination of clinical, epidemiological, and
economic factors are necessary. As health costs rise, dis-
abling conditions increase and populations age, medical
innovation proceeds at a rapid rate, and both client and
community expectations in health care continually rise.
Chapter 15 Health Technology, Quality, Law, and Ethics
In developing countries, the training and supervision of
traditional birth attendants (TBAs) for prenatal preparation
and normal deliveries are important ways to reduce maternal
mortality in rural areas, as discussed elsewhere, and an impor-
tant Millennium Development Goal (MDG) which will not
be met by 2015. Community health workers (CHWs) in well
supervised and supported programs are essential to provide
preventive care to underserved rural poor populations with a
defined package of services that can be tailored to meet spe-
cific local needs, such as immunization, child growth moni-
toring, nutrition counseling, and malaria and TB control.
A major example of appropriate technology has
been the WHO initiatives to promote national drug for-
mularies (NDFs) as a consensus list of essential drugs
that are sufficient for the major health needs of a coun-
try, eliminating unnecessary duplication and combined
products on the commercial market. The WHO calls on
all member states to ensure the availability and rational
use of drugs and vaccines, and supports states wishing
to select an essential list of drugs for economic procure-
ment. Assistance with drug regulatory agencies, legisla-
tion, quality control, information, supply, and training is
offered to help the member countries. Standard reference
laboratories, the International Pharmacopoeia, and the
WHO Drug Bulletin promote international standards and
provide guidance to member states. The WHO Model List
of Essential Drugs is a valuable tool to improve quality
and cost management in national health systems.
Cochlear implants are now routinely used for chil-
dren with congenital or other loss of hearing, as well as in
elderly people. In August 2013, a new cell phone applica-
tion was announced which photographs the eye and can be
used to diagnose cataracts, macular degeneration (AMD),
and other eye pathology, for interpretation by experts far
away and to enable arrangements to be made for appropri-
ate intervention to prevent blindness, which is common in
developing countries. Other applications allow for moni-
toring of blood sugar of diabetics, hypertension, exercise,
dietary management, and other aspects of health. In the
same month, a camera, computer, and auditory device
allowing blind people to “see and read” was demonstrated.
The costs of such devices are initially high but will fall with
advances in computing and other technical developments.
In both developing and industrialized countries major
causes of death include cardiovascular diseases (coronary
heart disease and stroke), along with respiratory diseases,
cancer and injuries, all amenable to preventive and cura-
tive medical care. The key preventive measures for these
are: healthful diet, reduced obesity, smoking cessation,
exercise and physical fitness, hypertension management,
aspirin, immunizations and other low-cost and highly effec-
tive medications such as statins. These are all low-cost self-
care measures that can be promoted by local, state, and
national governments, private advocacy organizations, and
775
individuals in their families and communities. The princi-
ples of low technology, cost-effectiveness, and sound health
policy converge in addressing these fundamental issues.
Priority Interventions in Low- and Medium-
Income Countries
Disease control priorities for low- and medium-income coun-
tries are an important challenge for public health. Selection has
often been based on individual initiatives due to strong advo-
cacy in international organizations by donor countries, orga-
nizations, or individuals. In 1993, two landmark documents
attempted to apply a logical system to such considerations:
one was the World Bank’s now classic World Development
Report: Investing in Health and the other was Disease Control
Priorities in Developing Countries. The World Development
Report defined cost-effective clinical and public health cluster
programs essential to improving health outcomes for low- and
middle-income developing countries. The programs focus on
those diseases that contribute heavily to the burden of disease
and are amenable to relatively inexpensive interventions. The
report defined interventions most able to reduce the burden
of disease in low- and middle-income countries using clinical
and public health interventions, as summarized in Table 15.2.
The 1993 World Development Report provided policy
makers and public health practitioners with a concept
and tools for assessing cost-effectiveness of available
interventions for the major health problems in the devel-
oping world. It also provided useful measuring tools in
the form of disability-adjusted life years (DALYs) to
calculate the burden of disease and the cost-effective-
ness of interventions to address them. This World Bank
report addressed clinical interventions that would reduce
DALYs lost by 24 percent in low-income countries and
8 percent in middle-income countries, including treat-
ment of TB, with directly observed therapy, short course
(DOTS); integrated management of the sick child; pre-
natal and delivery care; family planning; treatment of
STIs; and limited care for pain, infections, and trauma
as resources permit. It also addressed public health inter-
ventions, which would reduce DALYs lost by 8.2 percent
in low-income countries and 4 percent in middle-income
countries, expanded immunization with vitamin A sup-
plements; tobacco and alcohol control; AIDS prevention;
and school health including deworming. Together, the
total reductions would be 32 percent for low-income and
12 percent for middle-income countries (Table 15.2).
These estimates have been refined by numerous stud-
ies conducted over the subsequent two decades. The sec-
ond edition of the Disease Control Priorities in Developing
Countries (2006) incorporates important changes in the
technologies available. The concept of viewing priorities
with an economic epidemiology model is still applicable,
and has increased in importance (Box 15.1).
776
As the MDGs are reaching their endpoint in 2015,
follow-up global health targets will need to recognize the
vital importance of non-communicable diseases (NCDs)
in developing countries. The global consensus on MDGs,
set out by the United Nations (UN) in 2001, indicates
progress in the epidemiological understanding of realities
TABLE 15.2 World Bank Model for Priority Cost-
Effective Health Interventions in Low- and Middle-
Income Developing Countries
Burden of Disease Averted
(%)
Service Type Low-Income
Countries
Middle-Income
Countries
Public health interventions
EPI-plus immunization (DPT,
polio, measles, BCG, hepatitis
B, yellow fever, vitamin A)
6.0 1.0
Other public health programs
(family planning, health, and
nutrition education)
NA NA
Tobacco and alcohol control
programs
0.1 0.3
AIDS prevention program 2.0 2.3
School health program
(including deworming)
0.1 0.4
Subtotal (public health) 8.2 4.0
Clinical interventions
Treatment of tuberculosis
(short course)
1.0 1.0
Integrated management of the
sick child
14.0 4.0
Prenatal and delivery care 4.0 –
Family planning 3.0 1.0
Treatment of STIs 1.0 1.0
Limited care: pain, trauma,
infection plus as resources
permit
1.0 1.0
Subtotal (clinical care) 24.0 8.0
Total 32.2 12.0
Note: Low-income = < US$350 gross national product (GNP) per capita;
middle income = > US$2500 GNP per capita. Cost per immunized
child = US$14.60 (US$0.50 per capita) and US$27.20 (US$0.80 per
capita) in low- and middle-income countries, respectively.
DPT = diphtheria–pertussis–tetanus; BCG = bacille Calmette–Guérin;
AIDS = acquired immunodeficiency syndrome; STI = sexually transmitted
infection.
Note: The World Development Report was an innovative basis for follow-
up work, as reported in Jamison DT, Breman JG, Measham AR, Alleyne
G, Claeson M, Evans DB, et al., editors. Disease control priorities in
developing countries. 2nd ed. Disease Control Priorities Project. Wash-
ington, DC: World Bank; 2006.
Source: Adapted from World Bank. World development report. Investing
in health. New York: Oxford University Press; 1993.
The New Public Health
in low-income countries and the need for consensus over
common targets. Since then, attention has been directed
towards the epidemiological shift to NCDs, which are the
most common causes of death in low- and medium-income
countries. Thus there is a double burden of infectious, nutri-
tion, maternal, and child priorities, alongside the NCDs.
The increasing adoption of vaccines such as Haemophilus
influenzae type b (Hib) and rotavirus alongside the standard
BOX 15.1 Disease Control Priorities in Developing
Countries
l Average life expectancy in low- and middle-income
countries increased dramatically since the 1960s, while
cross-country health inequalities decreased.
l Improved health has contributed significantly to eco-
nomic welfare since the 1960s.
l Five critical challenges face developing countries (and
the world) at the beginning of the twenty-first century:
– Rapid demographic growth
– HIV pandemic improved but still rampant
– Persistent malaria, TB, diarrhea, pneumonia
– Micronutrient malnutrition for mothers and infants
– NCDs
– Possible pandemics.
l Cost-effective interventions include:
– Interventions to reduce neonatal mortality (50 per-
cent of total child deaths)
– Treatment of HIV-positive mothers, treatment of sex-
ually transmitted infections
– Controlling tobacco use, particularly through
taxation
– Lifelong medical management of risk factors in indi-
viduals at high risk for heart attacks or strokes, using
aspirin and other drugs, would benefit tens of millions
of individuals.
l Reform of health services and systems is needed,
including:
– Provider incentives
– Provider focus on selected intervention to gain
experience
– Strengthening surgical capacity at district hospitals
– Targeting limited resources to diseases affecting the
poor, e.g., TB in low-income countries
– In middle-income countries, public finance (or pub-
licly mandated finance) of a substantial package of
clinical care for all.
l Generation and diffusion of new knowledge and prod-
ucts underpinned the enormous improvements in health
in the twentieth century and need to be applied for the
control of NCDs, HIV, TB, and neglected populations.
Note: HIV = human immunodeficiency virus; TB = tuberculosis;
NCD = non-communicable disease.
Source: Adapted from Jamison DT, Breman JG, Measham AR, Alleyne
G, Claeson M, Evans DB, et al., editors. Disease control priorities in
developing countries. Chapter 1, Investing in health, Table 1.1. 2nd ed.
Washington, DC: World Bank; 2006.
Chapter 15 Health Technology, Quality, Law, and Ethics
diphtheria–pertussis–tetanus (DPT), poliomyelitis (polio),
and measles–mumps–rubella (MMR) vaccines provides
new possibilities to control the major infectious disease
killers of children. New technologies such as the advent of
antiretroviral treatment for HIV have led to startlingly suc-
cessful improvements in the quality of life and longevity of
HIV/AIDS patients, and the prevention of onward transmis-
sion of HIV from mothers to babies and sexual partners.
The WHO and many other global health stakeholders con-
tinue this work and produce analyses to contribute to policy
making based on economic epidemiological evidence. This
work affects policy, slowly but importantly.
In 2003, the Bellagio Study Group on Child Survival
estimated that the lives of 6 million children could be saved
each year if 23 proven interventions were universally avail-
able in the 42 countries in which 90 percent of child deaths
occurred in 2000. The MDGs set out in 2001 provided tar-
gets for economic, educational, and environmental improve-
ments, with three specifically focused on health: reducing
child mortality; reducing maternal mortality; and control of
HIV, TB, and other diseases. While important progress is
being made, some of these targets will not be achieved by
2015. The global public health infrastructure will need to be
expanded in content and strengthened in order to implement
lessons learned in childhood routine immunization, safe
maternity care, and nutritional security (see Chapter 16).
In medium- and low-income countries the difficulties are
much more severe because of limited resources for health and
the weak infrastructure of facilities and human resources in
many countries. The key issues relate to NCDs, as in developed
countries, so the interventions most needed address cardiovas-
cular diseases, cancer, and injury, as well as diarrheal diseases,
malnutrition, vaccine-preventable diseases, HIV, TB, malaria,
and neglected tropical diseases. Efforts should be focused on
low-cost interventions such as smoking reduction, vitamin and
mineral fortification of foods, HIV, TB, and malaria control,
along with maternal and child health protection.
Priority Selection in High-Income Countries
As discussed in Chapters 5 and 13, high- and middle-income
countries also face complex health challenges, including
aging populations, health costs, rapid development of new
drugs and technologies, high rates of NCDs, and the rising
prevalence of obesity and diabetes. Selection of priorities
for health care expenditure from public and private sources
has become a major focus of managing health systems.
In the industrialized countries, technological advances in
the medical and public health fields have been major contribu-
tors to increasing longevity but also rising health costs. This
situation has led to pressures for greater selectivity in adopting
costly innovations without adequate assessment of benefits
and costs. Many countries have adopted more cautious poli-
cies with regard to financing high levels of expansion of new
777
technology in the field of medical equipment, clinical proce-
dures, or medications. Organized assessment of technology is
now an essential feature of health management at the interna-
tional, national, and local levels of service delivery. The major
responsibility for technology assessment is at the national
level, even with decentralization of service management.
With available resources being limited, health systems
must choose interventions to be selected and how health
systems are to be organized for efficiency and effectiveness
while meeting public expectations. The US Patient Protec-
tion and Affordable Care Act (PPACA, more generally called
ACA or “Obamacare”) is undertaking reform measures to
promote efficiency and prevention to reduce per capita health
costs and to include more people in prepaid health care (see
Chapters 10 and 13). These include preventive measures as
recommended by Healthy People 2020 and implementation
committees for selection of cost-effective measures to reduce
morbidity and mortality to reduce health costs. The range
of services to be promoted includes smoking cessation,
increased physical activity, weight loss, healthy dietary prac-
tices, cancer screening, and many others that have not been
previously accessible to those living in poverty and with no
or limited health insurance. There is an emphasis on vaccina-
tion for children and adults.
The WHO promotes the widespread use of basic radio-
logical units (BRUs) to increase access to low-cost, effective,
diagnostic X-rays, especially in rural areas in developing
countries. BRUs are hardy, relatively inexpensive pieces of
radiological examination equipment that can be used in harsh
field conditions for simple diagnosis of fractures and respi-
ratory infections. The WHO estimates that 80 percent of all
diagnostic radiology can be performed adequately using sim-
ple, safe, and low-cost equipment, supported by training of
local people to operate and maintain the equipment. This is a
consensus view of leading radiologists and clinicians helping
the WHO to develop model equipment and training material.
The WHO World Health Report of 2009 focused on
health technology assessment, stating:
“Technology continues to transform the medical care system and
to improve length and quality of life – but at substantial cost. It
is almost inconceivable to think about providing health care in
today’s world without medical devices, machinery, tests, computers,
prosthetics, or drugs. Medical technology can be defined as the
application of science to develop solutions to health problems or
issues such as the prevention or delay of onset of diseases or the
promotion and monitoring of good health.”
Appropriate technology in the health field is becoming
increasingly complex, laden with economic, legal, and ethi-
cal issues. Professional and public opinion demands make
this a highly sensitive area of health policy, but responsible
management of resources requires decision making that
includes consideration of the effectiveness, costs, and alter-
natives of any new technology (Box 15.2). Failure to adopt
778
new innovations can result in obsolescence, while excessive
expenditures for hospitals and medical technology prevent
a health system from developing more cost-effective pre-
ventive approaches, such as improved ambulatory care, or
supportive care for the chronically ill.
HEALTH TECHNOLOGY ASSESSMENT
Technology adoption can be a highly emotional and con-
troversial issue, in advocacy of new cancer treatments or in
criticism of managed care or national regulatory agencies,
but spending limited national resources on some devices
or medications of unproven value or inappropriately long
hospital stays denies resources needed for other aspects of
health care. A society must be able and willing to pay for
medical innovation or improving quality of life by medical
and public health interventions. Underfunding of a health
system can deny these benefits just as misallocation of
resources does, and this is a political issue even more than
a professional one.
Medical and health technology assessment is the pro-
cess of determining the contribution of any form of care
to the health of the individual and community. It is a sys-
tematic analysis of the anticipated impact of a particular
BOX 15.2 Health Technology Assessment
Questions that form the basis of technology assessment for a
medical innovation include the following:
l Is it safe and cost-effective for the stated purpose?
l Is it a new service, or does it replace a less efficient inter-
vention which can be phased out of service?
l What is the need it addresses?
l Where is it in the order of priorities of development of
the facility?
l Does it duplicate a service already available in the
community?
l Does it make medical sense (i.e., does it help in diagnosis
and treatment for the patient’s benefit)?
l What are the alternatives?
l What are the resources needed in terms of supplies, staffing,
and upkeep?
l Can the facility afford it?
l What could otherwise be done with the resources it
requires?
Sources: Adapted from Kass N. Public health ethics: from foundations
and frameworks to justice and global public health. J Law Med Ethics
2004;32:232–42.
Sullivan SD, Watkins J, Sweet B, Ramsey SD. Health technology assess-
ment in health-care decisions in the United States. Value Health
2009;12:S39–44. Available at: http://www.ispor.org/htaspecialissue/
Sullivan [Accessed 14 December 2012].
Velasco-Garrido M, Busse R. Policy brief: Health technology assessment:
an introduction to objectives, role of evidence, and structure in Europe.
Geneva: WHO on behalf of the European Observatory on Health Systems
and Policies; 2005. Available at: http://www.euro.who.int/__data/assets/
pdf_file/0018/90432/E87866 [Accessed 22 October 2012].
The New Public Health
technology in regard to its safety and efficacy as well as its
social, political, economic, legal, and ethical consequences.
The technology may be a machine, a vaccine, an opera-
tion, or a form of organization and management of services.
Analysis should include cost–benefit and cost-effectiveness
studies (see Chapter 11) as well as clinical outcomes and
other performance indicators.
Pressures from medical professionals, manufacturers of
new medical equipment, and the public for adoption of new
methods can be intense and continuous. Care must be taken
that the specialists involved in committees for assessment
are not those who may directly or indirectly benefit from
the exploitation of technology, and who therefore may have
conflicts of interest. Assessment must be multidisciplinary,
involving policy analysts, physicians, public health special-
ists, economists, epidemiologists, sociologists, lawyers,
and ethicists. The available information needs to include
evidence from clinical trials, critical analysis of the litera-
ture, and the economic effect of adopting the technology on
allocation of resources.
Medical technology varies in complexity and cost, not
only to produce but in its utilization. Medical technology
that is inexpensive to supply and administer is known as
low technology or low-tech, while high technology or high-
tech refers to costly and complex diagnostic and treatment
devices or procedures.
At the low-tech end of the technology scale, oral rehy-
dration therapy (ORT) was developed in the 1960s for oral
replacement of fluids and electrolytes lost in diarrheal dis-
ease, particularly in children. It has been described as one
of the greatest medical breakthroughs of the twentieth cen-
tury. The introduction and wide-scale use of ORT for pre-
vention of dehydration from diarrheal diseases throughout
the world has saved hundreds of thousands of lives. Use
of insecticide impregnated bed nets and reintroduction of
DDT in household spraying along with vector control and
improving diagnostic tools are low-tech but effective and
key tools in malaria control.
Advances in endoscopic surgical techniques since the
1990s, and in robotic surgery since 2000, have greatly
improved patient care by reducing trauma, discomfort, and
length of hospital stay and endoscopy has become the surgi-
cal approach of choice for many procedures. Since reports
of the first 100 operations performed in France in 1990,
endoscopy has spread rapidly to all parts of the world. It is
now recognized by surgeons worldwide as a safer, less trau-
matic and more effective alternative to traditional invasive
surgery. Although the operating time is longer, patients are
discharged from the hospital within several days and return
to work shortly thereafter, compared to the long hospital
stays after more invasive surgical procedures in the past.
Following traditional abdominal surgery, a patient may
acquire infections and require intensive care initially and a
recovery period of many weeks.
http://www.ispor.org/htaspecialissue/Sullivan
http://www.ispor.org/htaspecialissue/Sullivan
http://www.euro.who.int/__data/assets/pdf_file/0018/90432/E87866
http://www.euro.who.int/__data/assets/pdf_file/0018/90432/E87866
Chapter 15 Health Technology, Quality, Law, and Ethics
Endoscopic surgery for cholecystectomy and esopha-
geal, colorectal, hernia repair, renal, orthopedic, and other
forms of surgery which previously were carried out with the
patient remaining in hospital for many days are now done
on a not-for-admission basis. Not-for-admission surgery has
become standard practice in hospitals, extending the range
of outpatient surgery and the comfort of patients who can
return to their own homes to recuperate and return to regu-
lar activities much sooner. Fewer complications arise and
patient comfort and economic implications are important.
As a result, fewer hospital beds are needed for postoperative
care than previously thought necessary, while surgical and
ambulatory care facilities may need expansion to accommo-
date the growing elderly populations needing surgical inter-
ventions but requiring shorter recovery. This innovation is
now accepted as the standard of much of modern surgical
care and shows that simple organizational changes can save
money and improve patient safety and comfort.
The bacterium Helicobacter pylori was first identified
as the cause of peptic ulcers of the stomach and duodenum
in 1982 (Robin Warren and Barry Marshall, Nobel Prize
2005). This discovery led to effective diagnosis and rapid,
inexpensive treatment of chronic peptic ulcer disease. This
has resulted in elimination of a major component of surgi-
cal procedures for chronic peptic ulcer diseases as well as a
reduction in gastric cancer (see Chapter 4). Surgery for gas-
trectomies, vagotomies, and other outdated forms of treat-
ment are now virtually gone, contributing to a decreased
need for hospital beds even for an aging population. This
and many other innovations in medical care have led to a
growth in the use of ambulatory care for many forms of
surgical, medical, and mental health care, along with much
shorter length of hospital stay than in previous times. All of
these factors have led to greater emphasis on ambulatory,
outpatient, and home care services.
The dissonance between high-tech and low-tech proce-
dures may lead to serious consequences in any health system.
Choices require well-informed analysis of benefits, costs,
alternatives, ethical considerations, and political consequences
before limited health care resources are allocated between
hospital-based high-tech medicine and low-tech primary care.
High-tech procedures are usually applied in hospital
settings in the context of other highly specialized care for
seriously ill, often terminal, patients. Computed tomogra-
phy (CT), invented in the 1960s, quickly proved to be an
extremely valuable diagnostic tool. Advances in CT, mag-
netic resonance imaging (MRI), and subsequent imaging
techniques have proven to be cost-effective and lifesaving,
replacing less efficient and more dangerous invasive proce-
dures. The CT and MRI scans allow the clinician to reach a
rapid diagnosis of many lesions before they can be detected
by other invasive and dangerous diagnostic techniques, at
stages where the lesions are subject to earlier and more
effective interventions. Imaging technology is advancing
779
rapidly and promising inexpensive new systems for long-
distance transmission of imaging to medical centers may
provide enormous benefits to people living in rural or
developing countries. Recent advances in low-intensity CT
screening of long-term heavy smokers for lung cancer have
recently been added to recommended and potentially effec-
tive and cost-saving practice and may change the outlook
for this disease in the coming decade (US Preventive Ser-
vices Task Force, 2012 Flahault and Martin Moreno, 2013).
Technology assessment also examines methods of pre-
venting and managing medical conditions. Treatment proto-
cols or clinical guidelines are based on decision analysis of
accumulated weight of evidence. Published clinical studies
are assessed in meta-analyses, using statistical methods to
combine the results of independent studies, where the stud-
ies selected meet predetermined criteria of quality. This pro-
vides an overview from pooling of data, but also implies an
evaluation of the studies and data used. Clinical guidelines
are part of raising standards of care, but also contribute to
cost containment. Many countries form professional study
groups to carry out meta-analyses on important health pol-
icy issues and new technologies.
Technology Assessment in Hospitals
There is considerable variance among countries, hospitals
consume between 40 and 70 percent of total national health
expenditures, with pressures for increased staffing and
novel medical technology being a continual inflationary
factor. Industrialized countries have all reduced their acute
care hospital bed supplies and length of stay so that their
expenditures for hospital care have fallen to between 30-40
percent of total health expenditures. Shorter stays and older
patients have resulted in a drift towards intensive care, espe-
cially for internal medicine patients. Medical innovation is
a continuing process with new diagnostic and treatment
modalities reaching the market.
Hospitals no longer live in splendid isolation in the
medical economy. A national or state government needs
regulatory procedures to rationalize distribution of medical
technology. The “certificate of need” is a form of technology
assessment that has been used in the USA since the 1960s
to assess and regulate the development of hospital services
to prevent oversupply and costly duplication of services. It
attempts to establish and implement the use of rational crite-
ria for diffusion of expensive new technology. Whether this
has had a lasting impact on restraining the excesses of high-
tech medicine is arguable. This regulatory approach was
limited to the hospital setting and failed to stop the develop-
ment of high-tech medical services such as ambulatory for-
profit CT, imaging, and in vitro fertilization centers.
Many countries have adopted national technology
assessment systems to review topics as far-ranging as
guidelines for acute cardiac interventions; liver, heart, and
7
80
lung transplantation; minimal access surgery; and beam and
isotope radiotherapy. Other technology assessment guide-
lines include diagnostic ultrasound, sleep apnea, molecu-
lar biology, prostate cancer, MRI, and new medications for
inclusion in a national health system’s approved basket of
services.
Despite the limitations of this approach, where govern-
ments do not directly operate health care services, gov-
ernmental regulation is necessary to prevent inequities in
services by excessive development in some geographic
areas at the expense of others, or by overexpansion of the
institutional sector of health care at the expense of primary
care. Regulatory mechanisms are essential in health care
planning to restrain excessive and inappropriate use of
high-tech services, but need augmentation by fiscal incen-
tives to promote other essential services.
Hospitals everywhere face serious problems of hospital-
acquired infections, which occur in about 5 percent of all
hospitalizations. Healthcare-associated infections (HAIs),
including multidrug-resistant bacterial infections, cause
long lengths of stay, high costs, and most importantly, unex-
pected deaths and serious disabilities. Prevention of hospital-
acquired infection requires ongoing training, staffing, and
organization. The Centers for Disease Control and Prevention
(CDC) defines HAIs as “infections caused by a wide variety
of common and unusual bacteria, fungi, and viruses during
the course of receiving medical care”. Some of the preventive
measures are simply promoting frequent hand washing by
caregivers and visitors, and immunization of staff members
against influenza and pneumonia, which can be problematic
if there is staff resistance to influenza vaccination.
Training and routine supervision of cleaning staff are
also vital, as are strict infection control measures for isola-
tion rooms, strict protocols for catheter care, surgical suite
sterility, surgical site infections, central line associated
bloodstream infections, ventilator-associated pneumonias,
catheter-associated urinary tract infections, and Clostridium
difficile-associated disease. Guidelines for their control in
surgical dialysis, pediatric, outpatient, and other vulnerable
departments are available from CDC. The benefits of pre-
ventive procedures for this problem include cost estimates
ranging from US$5.7–6.8 billion (20 percent of infections
preventable) to US$25.0–31.5 billion, yet 70 percent of HAIs
are preventable by well-known methods such as frequent and
careful hand washing by medical and nursing staff, catheter
and infusion care, and other similar measures (CDC, 2012).
Technology Assessment in Prevention and
Health Promotion
Technology assessment of preventive care programs
includes evaluation of the methodology itself, along with
the costs and measurable benefits, as in reduced burden
The New Public Health
of disease. DOTS is the standard management of sputum-
positive and sputum-negative TB, at low cost for DALYs
saved. The coexistence of HIV and other complications
has created multidrug-resistant tuberculosis (MDR-
TB), which is difficult and costly to treat and cases
constitute a source of continuing spread of the disease.
A 2012 meta-analysis of cost-effectiveness of MDR-TB
treatment in Estonia, Peru, the Philippines, and Russia
shows it to be cost-effective and best carried out on an
ambulatory basis (Fitzpatrick and Floyd, 2012).
Wide use of available and effective vaccines such as
Hemophilus influenza b (Hib), pneumococcal pneumo-
nia, influenza and rotavirus reduce hospitalizations and
mortality from respiratory and diarrhoeal diseases among
children, the elderly and other age groups. Vaccine prices
generally fall after their initial period of use as manu-
facturing costs are lessened by improved methods or by
bulk purchase contracts, as occurs in the public sector. For
example, in 2012 MMR vaccine cost US$19.33 per dose if
purchased through the CDC, but US$52.73 per dose if pur-
chased in the private sector in 10 packs of single-dose units
of the vaccine. A combined diphtheria, tetanus, acellular
pertussis (DTaP) vaccine cost US$15.00 when purchased
through CDC, while the same vaccine purchased with hep-
atitis B and inactivated polio vaccine (IPV) cost US$52.10 per
dose. But the combination saves repeated visits and loss
of compliance for that reason. The new human papilloma-
virus (HPV) cervical cancer vaccine cost US$130.27 per
dose for the series of three doses per person, while the vac-
cine against diarrhea-causing rotavirus, approved in 2006,
cost US$106.57 per dose for the recommended three doses
(CDC, 2012).
The WHO recommends the inclusion of rotavirus vac-
cination in a country’s immunization program, but the costs
of the current generations of rotavirus vaccines are high in
comparison to the budgets for vaccines for prevention of
childhood illnesses in many developing countries. Many
cost-effectiveness studies have shown this vaccine to be
highly beneficial and it could help to reduce the very high
global burden of disease of over 500,000 child deaths and 2
million hospitalizations occurring annually (Tu et al., 2011).
Vaccine programs must take into account transportation
and administrative costs and expenses of ordering, storing,
inventory control, cold chain, insurance, wastage, and spoil-
age. Multiple vaccines in one dose are less costly and less
inconvenient for all. Examples include DTaP plus polio and
Hib, or MMR (see Chapters 4 and 6). There is a need for
implementation of legal protection of manufacturers from
excessive litigation judgments while protecting the interests
of the public and individuals who may have reactions to
vaccines.
In 2012, the reappearance of pertussis and diphthe-
ria raised concerns about immunization coverage and
Chapter 15 Health Technology, Quality, Law, and Ethics
efficacy. Public opinions on vaccination may not be as
supportive as in previous years. Mothers who oppose per-
tussis immunization for their children, such as occurred in
the UK during the 1980’s, leave their children vulnerable
to a serious and often deadly disease, which has recurred
since 2010.
The WHO estimates the cost of all immunization activi-
ties in all 117 low- and middle-income countries for the
period 2006–2015 to be US$75 billion, while low-income
countries would need US$35 billion. The rate of adoption
of currently available and new vaccines will be determined
by governmental decisions in each country, although exter-
nal aid – such as that of the Global Alliance for Vaccines
and Immunization (GAVI), an international public–private
consortium to promote vaccination – is a valuable resource.
The United Nations Children’s Fund (UNICEF) is con-
cerned about supply problems as well as costs, but the key
issue relates to political decisions, funding, and capacities
of national immunization systems.
Despite an excellent vaccine having been available since
the 1960s, measles epidemics continue to occur in the indus-
trialized countries. In the 1900s global deaths from measles
were in the order of 1 million people per annum. Two major
epidemics of measles occurred in Canada in the early 1990s,
despite high rates of immunization coverage. Following this, a
1993 Delphi conference of experts from 31 countries reached
a consensus recommending a two-dose measles immuniza-
tion policy. Measles eradication has been set as a goal by the
WHO and 90 percent reduction in cases and fatalities has
been achieved since the 1990s. However, measles elimination
requires coverage of 95 percent of children and two doses of a
measles-containing vaccine (preferably MMR).
Measles reappeared as a widespread disease in Europe in
2010–2013 with tens of thousands of cases, many hospital-
izations, and some deaths. It spread to the Americas, brought
by travelers, and resulted in modest sized outbreaks, includ-
ing the UK in 2012–2013. Eradicating measles by 2020 is
projected to cost an additional discounted US$7.8 billion and
avert a discounted 346 million DALYs between 2010 and
2050. As new vaccines enter the field, it is important to evalu-
ate their effectiveness, costs, and the benefits to be derived.
The cost of the hepatitis B vaccine initially was over
US$100 for an immunization schedule of three doses but
has come down dramatically to less than US$1 per dose
in developing countries for bulk purchases. However, in
the USA, the price of vaccination per dose is estimated at
US$41 if given by a general practitioner, US$15 if adminis-
tered through an existing childhood immunization program,
and US$17 if given through the school medical system.
This is a standard vaccine covered by public and private
health insurance systems. The vaccine is a cost-effective
method to prevent liver cancer and the long-term effects of
chronic hepatitis.
781
Screening and education for thalassemia in high-prev-
alence areas have nearly eradicated the clinical disease but
not its carrier status in Cyprus, southern Greece, and other
countries. Newborn screening and case management for
phenylketonuria, congenital hypothyroidism, Tay–Sachs
disease, and many other genetic diseases have been shown
to be far less expensive than post-facto treatment of severely
developmentally delayed and dependent children born with
these diseases (see Chapter 6).
The success of Papanicolaou (Pap) smear screening in
reducing cancer of the cervix mortality since the 1960s has
been dramatic. The discovery of causation of cancer of the
cervix by HPV strains led to development of an effective
vaccine, which has been in use since 2006. Recent evidence
shows that male circumcision can reduce transmission of
HPV as well as HIV and other sexually transmitted dis-
eases, and it is being adopted as an effective intervention
in countries with high rates of both HIV and cancer of the
cervix, such as in sub-Saharan Africa.
The drastic reduction in cancer of the cervix provides
a powerful demonstration of the effectiveness of public
health screening and other measures to control this major
malignant cause of death in women. Screening for cervi-
cal cancer by Pap smears is recommended annually for
high-risk groups, and every 2 or 3 years for other adult
women (Box 15.3). Screening will remain vital for many
years to come as the HPV vaccine comes into general
use, and as its cost is reduced, but its protective effect for
individual and herd immunity will not replace the need
for ongoing screening for this very common cancer. HPV
vaccine is also being recommended for all boys to pre-
vent oral and anogenital cancers and HPV transmission
to girls.
Routine mammography screening for breast cancer
every 1–2 years is recommended by the US National Can-
cer Institute for women over the age of 40 and for younger
women with high-risk factors (e.g., previous cancer, fam-
ily history, genetic markers). Cost-effectiveness analysis is
now an essential part of decision making in health policy
and priorities. While there is controversy over the frequency
of routine testing, mammography remains a mainstay in
women’s health and contributes to early case finding and
falling mortality rates from breast cancer. Figure 15.1 dem-
onstrates differences in utilization of mammography among
US women in the age group 50–64 years within the previ-
ous 2 years, by insurance status. US women with private
insurance (mostly through place of employment) had over
70 percent compliance, those with public insurance (pri-
marily Medicaid) averaged about 60 percent compliance,
while those with health insurance had average compliance
rates of about 45 percent during the period 1993–2010.
The UK National Health Service (NHS) invites women
between the ages of 50 and 70 for screening every 3 years;
Cancer of the cervix is the second most common cancer among
women worldwide, with about 500,000 new cases and 250,000
deaths worldwide annually. Approximately 80 percent of cases
occur in low-income countries, where cervical cancer is the
second commonest cancer in women (WHO, 2012).
In the USA, and other industrialized countries, the incidence
and mortality of cancer of the cervix have been going down
steadily since the introduction of Papanicolaou (Pap) smear
testing. Cervical cancer incidence declined during the period
1999–2008 by 2.3 percent per year and mortality declined by
1.9 percent per year an estimated 12,170 cases of invasive cer-
vical cancer diagnosed in the USA with 4220 deaths in 2012.
Prevention of cancer of the cervix has until recently mainly
focused on Pap smears to detect the disease while still in a pre-
cancerous (cancer in situ) phase, and this procedure reduced
rates dramatically over the latter part of the twentieth century. The
newly developed and highly effective vaccines against key strains
of human papillomavirus (HPV) is now being used in routine
immunization of young girls and more recently boys as well. The
high cost of the vaccine precludes its rapid diffusion to most parts
of the world but its use is spreading and being included in immu-
nization programs funded by donor agencies in sub Saharan
Africa. The vaccine should, in principle, also be used by adult
women, in addition to continuation of routine Pap smear testing.
In the past decade, evidence of HPV as the cause of cancer
of the cervix and the presence of HPV in uncircumcised men has
brought circumcision back to professional and public debate.
Reports from Africa of reduced risk of acquiring HIV among
circumcised men have brought new attention to adult male cir-
cumcision, which is now actively promoted many sub-Saharan
African countries.
The technological breakthroughs of the Pap smear in the
1950s, HPV testing in the 1990s, and the HPV vaccine in the
2000s should also include prevention by male circumcision.
Visual inspection of the cervix and cryotherapy can treat pre-
cancerous cervical lesions in areas of developing countries
as part of community health worker programs.
Colorectal cancer, the 7th leading cause of death in high
income countries, is amenable to prevention by early screening
using colonoscopy and fecal occult blood (FOB) testing. Screening
is recommended for all persons over age 50 at 5 year intervals
along with annual FOB testing. Where there is a family history of
colorectal cancer or polyps, routine screening should begin ear-
lier. Increasing use of screening and improved medical care are
resulting in improving survival and declining mortality rates.
Stomach cancer is 10th leading cause of death in upper
middle income countries. Prevention relies on early treatment
of chronic peptic ulcer disease caused by Helicobacter pylori
infection. This is readily diagnosed by a simple breath test and
completely cured by low cost antibiotics. Increased awarenss
and access to these services would enhance long term trends of
reducing mortality from stomach cancer.
Liver cancer is 8th leading cause of death in upper middle
income countries due to the global prevalence of hepatitis B
and helaptitis C. Hepatitis B is now falling due to widespread
vaccination in childhood. Hepatitis C is now the major cause
of liver cancer affecting hundreds of millions of persons world-
wide. There is still no vaccine currently available, but screen-
ing and treatment is now used in the industrialized countries
and will become more widely used as simpler, less costly treat-
ments with less side effects are becoming available.
Early detection of lung cancer with spiral low dose
tomodensitometry for smokers is recently being recommended
by many professional bodies.
More basic cancer preventive measures such as smoking
cessation, healthy diets, regular exercise, and moderate alco-
hol use are discussed in chapter 5.
Sources: World Health Organization. Sexual and reproductive health.
Cancer of cervix. Available at: http://www.who.int/reproductivehealth/top-
ics/cancers/en/ World Health Organization. The top 10 leading causes of
death (2011). Available at: http://who.int/mediacentre/factsheets/fs310/en/
index1.html
Centers for Disease Control and Prevention. Cervical cancer trends 2012.
Available at: http://www.cdc.gov/cancer/cervical/statistics/trends.htm
[Accessed (13.12.2012)].
BOX 15.3 Technology for Prevention of Cervical, Colorectal, Liver, Stomach and Lung Cancers
1993
0
10
20
30
40
50
P
er
ce
nt
ag
e 60
70
80
90
100
1995 1997 1999 2001
Year
2003 2005 2007 2009
Private insurance
Public insurance only
No insurance
FIGURE 15.1 Self-reported percentage of women aged 50–64 years
receiving a mammogram in the past 2 years, by health insurance sta-
tus, USA, 1993–2010. Source: Centers for Disease Control and Prevention.
QuickStats from the National Center for Health Statistics: Percentage of women
aged 50–64 years who reported receiving a mammogram in the past 2 Years,
by health insurance status — National Health Interview Survey, United States,
1993–2010. MMWR Morbid. Mortal. Wkly. Rep. 2013;62:651. Available at:
http://www.cdc.gov/mmwr/pdf/wk/mm6232 (accessed 12.1.14).
with an average of 75 percent respond to the invitation, of
whom 4 percent are referred for further testing (UK Can-
cer Research, 2012) and as in most industrailized countries
breast cancer mortality rates are falling impressively (see
Chapter 5).
Health promotion in reducing exposure to HIV and
cigarette smoking has been shown to be very cost-effec-
tive despite its low-tech or non-technological meth-
odology, involving primarily group or mass education.
Hypertension screening and case management is low-
tech but highly effective in preventing strokes and blind-
ness.
Low-tech innovations have had an important impact
in reducing death and injury. These include mandatory
use of car seat belts (introduced since the 1970s and
1980s in many countries), children’s car seats, air bags,
and bicycle and motorcycle helmets. Iodization of salt,
vitamin A supplementation, and food fortification pre-
vent large numbers of clinical cases of severe retarda-
tion, death, and blindness at low cost per child protected.
http://www.who.int/reproductivehealth/topics/cancers/en/
http://www.who.int/reproductivehealth/topics/cancers/en/
http://who.int/mediacentre/factsheets/fs310/en/index1.html
http://who.int/mediacentre/factsheets/fs310/en/index1.html
http://www.cdc.gov/cancer/cervical/statistics/trends.htm
http://www.cdc.gov/mmwr/pdf/wk/mm6232
783Chapter 15 Health Technology, Quality, Law, and Ethics
TABLE 15.3 Examples of High-Tech and Low-Tech Health Problem Solving
Problem High Tech Low Tech
Birth defects Surgical repairs, rehabilitation Folic acid fortification of flour, vitamin supplementation
before and during pregnancy
Infectious diseases Treatment – antibiotics Vaccination, sanitation, handwashing, infection control in
hospitals, health facilities, and nursing homes
Breast cancer Screening – mammography Nutrition, self-examination, routine medical examination
Colon cancer screening Colonoscopy Nutrition, vitamin D supplements; fecal occult blood testing
Acute myocardial infarction,
primary, secondary prevention
Coronary angioplasty, stent, bypass
surgery, heat transplantation
Antiplatelet thrombosis treatment (e.g., aspirin, intravenous
streptokinase, beta-blocker); rehabilitation; diet, exercise,
smoking cessation
Gallstones Lithotripter, abdominal
cholecystectomy
Endoscopic surgical removal
Head injuries Intensive care Helmets for bicycle riders and motorcyclists, seat belts in
front and rear of motor vehicles
Thalassemia Transfusions, chelating agents;
prenatal diagnosis, amniocentesis,
chorionic villus biopsy
Screening, education, counseling
Dehydration Infusions Oral rehydration
Neural tube defects Surgery, pregnancy termination Folic acid fortification of flour and grain products, supple-
ments for women of fertile age
Liver cirrhosis, liver failure,
cancer
Liver transplant Hepatitis B vaccine, risk reduction activities among intrave-
nous drug users, screening blood donors
Cancer of stomach Surgery, chemotherapy Dietary change, cure of Helicobacter pylori-generated
gastric ulcers
Cancer of cervix Pap smear screening; visual inspection and cryotherapy;
human papillomavirus vaccine
Education for reducing risk factors for the cardiovascu-
lar disorders is far less costly than the premature deaths
and high medical costs of patients suffering stroke and
congestive heart failure. Health education, condom and
needle supply, and screening of blood donations are the
most important effective community health measures
against the spread of HIV. Table 15.3 shows a compari-
son of high-tech and low-tech approaches, which often
complement each other, to selected health problems.
Technology assessments represent the current con-
sensus derived from reviews of published studies and
exchange of views of highly qualified clinicians, epide-
miologists, and economists within a context of technol-
ogy assessment. They may change over time as new data
or innovations are reported, and this possibility should
be kept in mind in such discussions. Technology assess-
ment mobilizes information and critically analyzes many
aspects of medical technology to build a wide community
consensus to influence policy decisions. Public opinion,
political leadership, and administrative practice, as well
as the scientific merit of a case are all factors in develop-
ing a consensus.
Technology Assessment in National Health
Systems
Technology assessment requires an organization within
the framework of national regulatory agencies. The FDA
serves this purpose as a statutory body within the US Pub-
lic Health Service. Sweden, Canada, Australia, the UK, the
Netherlands, Spain, and other countries also have technol-
ogy assessment advisory or regulatory agencies established
by national governments to monitor and examine new tech-
nologies as they appear. Sweden has a widely representa-
tive national Swedish Council for Technology Assessment
in Health Care which has an advisory role to the national
health authorities.
The processes used in traditional systems to regulate
food and drugs for efficacy, safety, and cost are more
recently being applied to new medical devices and proce-
dures. The unrestricted proliferation of new procedures
presents serious dilemmas for national agencies con-
cerned with financing health care and controlling cost
increases. Non-governmental health insurance shares
this concern, as does industry, which bears much of the
784
cost of health insurance through negotiated, collective
bargaining, “voluntary” health insurance in the USA.
Most industrialized countries have national health ser-
vices or national health insurance and are thus vitally
interested in health costs and technology assessment.
Many industrialized countries maintain technology
assessment and cost-control activities. In the USA, the
Agency for Healthcare Research and Quality (AHRQ)
maintains oversight and studies related to clinical infor-
mation, including evidence-based practice, outcomes and
effectiveness, comparative clinical effectiveness, risks
and benefits, and preventive services.
In Canada, the Health Protection Branch of the Federal
Department of Health reviews medical devices and drugs
and, with consent of the provincial governments, approves
new medical procedures. Concern by governments over the
cost implications of new procedures led to this practice.
Since 1988, a network of government and professional
bodies has formed a non-profit agency for technology
assessment (Canadian Agency for Drugs and Technologies
in Health, CADTH). This supports the provincial adminis-
tration of health insurance in resisting professional, com-
mercial, or political pressures to add untested technology
or procedures to the health system as covered benefits. A
comparison of rates of procedure performance between
provinces shows very high discrepancies, as high as two-
fold, in procedures such as coronary artery bypass graft or
prostatectomy. Control of acquisition of high-tech equip-
ment by national or state authorities is essential to prevent
expenditures on high-cost equipment without adequate
assessment.
The New Public Health
DISSEMINATION OF TECHNOLOGY
The rapid spread of high-tech medical equipment has played
a substantial role in escalating health costs. A comparison of
the number of MRI scanners per million population in mem-
ber countries of the Organisation for Economic Co-operation
and Development (OECD) (Table 15.4) showed Japan and the
USA with the highest number at 43.1 in 2010 (23.2 in 2000–
01) and 31.6 (15.4 in 2000) per million, respectively, while
the median was 10.5. Canada ranked fourteenth among the 20
OECD countries with 8.2 MRI scanners per million. Compar-
ing CT scanners showed that Japan had the highest number
at 97.3 while the median was 15.6. Canada was in eighteenth
place among the 28 OECD countries reporting in 2010 with
14.2 CTs per million population. However, it should be noted
that countries with the most machines have lower productiv-
ity per machine. The USA has far fewer examinations per
machine than other countries (OECD, 2012).
The use of endoscopic surgery has spread worldwide
since the 1990s. Health professionals become almost
instantly aware of new developments from the news media
as well as professional diffusion of information at confer-
ences, in exchange visits, in published articles, and most
dramatically via the Internet. National policy to foster the
introduction of appropriate new technology requires a
careful program of regulatory and financial incentives and
disincentives to encourage or discourage diffusion of new
methods of prevention as well as of treatment and com-
munity health care. Kidney transplantation has become a
cost-effective and patient-friendly alternative to long-term
dialysis both in hospital and at home. The key limitation is
TABLE 15.4 High-Tech Medical Equipment Units, Selected Organisation for Economic Co-operation and
Development Countries and Years, 1986–2010 (Rate per Million Population)
CTs MRIs
Country 1986 2000 2005 2010 1986 2000 2005 2010
Japan 27.5 84.4a 92.6 97.3c 0.1 23.2a 40.1 43.1c
USA 12.8 25.1a 32.3 40.7d 0.5 15.4a 26.7 31.6
Israel – 5.7 6.6 9.2 – 1.4 1.7 2.0
Germany 6.9 12.2 15.4 17.7 0.7 4.9 7.1 10.3
UK 2.7 5.4 7.5 8.2 0.3 5.6 5.4 5.9
Canada – 9.8b 11.6 14.2 – 2.5 5.7 8.2
France 4.7 7.0 10.0 11.8 0.5 1.7 4.8 7.0
CTs = computed tomography scanners per million population; MRIs = magnetic resonance imaging units per million population.
Notes:
a1999
b2001
c2008
d2011.
Source: Organisation for Economic Co-operation and Development. Health policies and data: OECD health data – 2012 data; frequently requested data.
Available at: http://www.oecd.org/health/healthpoliciesanddata/oecdhealthdata2012-frequentlyrequesteddata.htm [Accessed 13 October 2012].
http://www.oecd.org/health/healthpoliciesanddata/oecdhealthdata2012-frequentlyrequesteddata.htm
Chapter 15 Health Technology, Quality, Law, and Ethics
the shortage of donors. The same can be said for liver trans-
plantation, which has also been shown to be cost-effective
in terms of DALYs saved from chronic liver diseases such
as cirrhosis and hepatitis B and C. Heart transplantation has
spread among major centers in western countries.
The black market in organs for transplantation has
become an international scandal of exploitation of poor
people in low-income countries, and is the subject of police
investigation in many countries, but it is difficult to control.
Bone marrow transplantation is now used widely and
is effective in saving the lives of many people with malig-
nant and non-malignant hematological disorders. Stem cell
therapy, by introducing new adult and embryonic stem cells
into damaged tissue to treat disease or injury, is becoming
feasible for a wide variety of conditions including cancer,
type 1 diabetes mellitus, Parkinson’s disease, Huntington’s
disease, celiac disease, cardiac failure, muscle damage, and
neurological disorders.
Limitation of new techniques or procedures to selected
medical centers allows the passage of time to fully assess
the merits and deficiencies of new technology before gen-
eral diffusion into the health care system. Such limitation,
however, is fraught with the danger of depriving the popu-
lation of benefits of new medical technology, and the pos-
sibility of restraint of trade to the economic advantage of
selected providers. Current advances in robotic-assisted sur-
gery will expand during the coming decade for brain, bone,
joint, prostate, and other surgery, and need to be assessed
with regard to patient care benefits, costs, and the econom-
ics of capitalization of such innovations. Stem cell therapy
is already widespread for some malignant conditions, and
with advances in genomics and molecular biological tech-
nology it is very likely to become a major therapeutic inter-
vention for many more conditions in the coming years. The
effects of new technology on insurance and managed care
systems are necessarily involved in decision making as to
inclusion of new procedures in their service plans.
Publication in the professional literature is an accepted
method of establishing the scientific merit of a treatment or
an intervention. Too rapid diffusion of a medical practice
can lead to disillusionment and confusion as to the mer-
its of a particular medical procedure, as happened during
the 1960s and 1970s with anticoagulant therapy for acute
myocardial infarction and gastric freezing for peptic ulcers.
Reviews of the literature should be critical and should
assess the scientific merits of published data, as well as the
sources of funding. Well-controlled large-scale clinical tri-
als are vital to establish the relative values of alternative
therapeutic approaches, as are meta-analyses of multiple
studies.
Dissemination of information about new medical inno-
vations in the popular media is almost immediate. Many
major newspapers and television networks have well-
informed medical reporters and commentators who have
785
access to electronic medical journals as quickly as do medi-
cal specialists in each field. News magazines may carry spe-
cial articles on new innovations, creating instant demand for
them as benefits in a health program. This ready access to
information has both benefits and dangers.
In the USA, health insurers have led the way in develop-
ing technology assessment and information synthesis, and
in evaluating the costs and benefits of new procedures. The
process is affected by public opinion, as well as by court
decisions. A landmark decision against a health mainte-
nance organization (HMO) in 1993 awarded US$29 million
in damages to the family of a terminal breast cancer patient
who died following refusal of the HMO to authorize a bone
marrow transplant, which was at the time an experimental
procedure. Denial of new technology may lead to increases
in malpractice suits. In countries with limited financial
resources, selection of technological innovations in health
care that can benefit patient care or the public health
requires a careful balance in order to use limited resources
well, and to gain from the application of appropriate new
health care technology.
Payment systems by national or private insurance
systems are crucial to introduce and control diffusion of
technology. Block budgets for hospitals have been more
effective in Canada than in the USA in restraining the pro-
liferation of high-tech equipment. This has led to criticism
of the limited access of Canadians to medical technology,
such as CT, MRI, and advanced cancer therapies. In the
USA, universal application of the diagnosis-related group
(DRG) payment system for Medicare, Medicaid, and most
private insurance had the effect of increasing ambulatory
surgery very dramatically, from 16 percent of all surgery
in 1980 to 80 percent in 2008 of all surgical procedures in
community hospitals (i.e., non-federal short-stay hospitals
or 85 percent of all hospitals in the country). Inpatient sur-
gical procedure rates declined from 85 percent in 1980 to
35 percent in 2008. Although the rate of visits to hospital-
based surgery centers remained largely unchanged in the
USA from 1996 to 2006, the rate of visits to private ambula-
tory surgery centers increased by about 300 percent.
HMOs and managed care organizations are paid on a
per capita basis and have a strong incentive for cost contain-
ment. They have developed procedures and medical guide-
lines for investigation and intervention that seek to reduce
unnecessary procedures. At the same time, HMOs are very
active in promoting preventive care and non-hospital care
insofar as this is compatible with good patient care.
Coronary bypass procedures decreased in frequency in
the USA between 2001–02 and 2007–08. In the USA, such
procedures are less frequently carried out in women and
African Americans, because of lesser access to health insur-
ance for African Americans and possibly because of biases
in terms of case assessment criteria in women. Cardiac
invasive procedures increased dramatically since the 1980s
786
in most industrialized countries, but with wide variation in
their use. The benefits of aggressive invasive management
of cardiovascular diseases remain controversial, but many
such procedures have proven beneficial in reducing mortal-
ity rates and improving quality of life.
Critical analysis of the need for surgery has resulted in
lower tonsillectomy and radical mastectomy rates along
with the increased use of outpatient procedures. Tonsillec-
tomy, a routine procedure until the 1960s, is now performed
infrequently since it was found to be of little medical value.
Cataract surgery is now largely done on an ambulatory
basis. The technology of home care has come to play an
important role in early discharge of patients from the hospi-
tal, as has the wide use of cancer chemotherapy and radia-
tion therapy on an outpatient basis.
DIFFUSION OF TECHNOLOGY
Innovations in health care through scientific and technologi-
cal advances are continuing, with exciting breakthroughs
being made in effective new treatments and public health
interventions, and this requires health authorities, practitio-
ners, and the public to maintain constant awareness of the
current state of the art. Diffusion of new technology or adap-
tations from basic science advances may begin slowly, and
then reach a “tipping point”, at which time a dramatic change
of trend occurs and it becomes the new standard or fashion.
Those with economic interests in the product try to
advertise and promote sales, while practitioners are ready
to try new methods to help their patients, but those who
must pay for services may ask for evidence of effectiveness,
safety, added value over present and known methods, and
benefit to the length or quality of life of the individual. This
can become a highly charged debate when those responsible
for adopting new measures in national health plans must
weigh one proposed addition against another, each with its
ardent professional, community, or business promoters. The
new HPV vaccine approved by the FDA in 2006 for preven-
tion of cancer of the cervix is an example.
The HPV vaccine is recommended for preteen girls at the
age of 11–12 years and also for females aged 13–26 to off-
set future sexual exposure to HPV-infected males and since
2012 recommended routinely for teenage boys as well. The
two competing main manufacturers of HPV vaccine are nat-
urally interested in increasing their market and market share,
and willing to reduce prices. The cost has been lowered
substantially for use in developing countries if purchased in
bulk, but costs are still prohibitive unless funded by interna-
tional donors. Competing low-cost manufacture in India has
encouraged the two main manufacturers to lower prices to
seek broader markets. In 2011, one manufacturer lowered its
price dramatically to US$5 per dose, a 67 percent reduction
in the current lowest public price. This has allowed GAVI to
adopt an HPV strategy for developing countries, where 88
The New Public Health
percent of cervical cancer deaths occur, with 275,000 deaths
of relatively young women each year (GAVI, 2011, 2012).
Pioneering projects promoting visual examination of the
cervix and local cauterization of abrasions by trained nurses
and community health workers are meant to increase access
to care in traditional villages remote from medical centers.
The duration of immunity and whether booster doses will
be required are still not known. Policy makers need to con-
sider whether the same money would have greater benefit
if used to provide pneumococcal pneumonia and rotavirus
vaccine for children in developing countries, which would
quickly save hundreds of thousands of lives. It is likely that
the wonderful new public health technology that is the HPV
vaccine will be absorbed quickly into public health prac-
tice at least in the industrialized countries, and is now being
introduced by international donor agencies in sub-Saharan
Africa.
QUALITY ASSURANCE
Quality assurance is an integral part of public health func-
tion and involves ensuring the quality of both health prac-
titioners and facilities. It is an approach that measures and
evaluates the proficiency or quality of services rendered.
Hospital accreditation is a long-standing method of qual-
ity assurance, providing many generations of health provid-
ers in North America with first hand experience of quality
assurance in community hospitals and long-term care facili-
ties, as well as ambulatory and mental health services.
Hospital accreditation has contributed to improvement in
standards of facilities and patient care throughout Canada
and the USA and has provided a working model for replica-
tion or adaptation internationally.
Adverse Events and Negligence
Iatrogenic diseases are adverse events that occur as a result
of medical management and result in measurable disability.
Negligent adverse events are those events caused by a failure
to meet standards of care reasonably expected of the aver-
age physician or other provider of care. Hospital-acquired
infections, anesthesia mishaps, falls, and drug errors are the
most common iatrogenic events.
Iatrogenic disease is a major cause of morbidity, pro-
longation of hospitalization, and even death. Hospital-
acquired (nosocomial) infections are estimated to occur
in 7–10 percent of hospital cases in Britain and the USA.
Primarily these are caused by urinary, respiratory tract, and
wound infections. It is becoming more common that infec-
tions involving organisms previously responsive to antibi-
otics are now resistant to many antibiotics and difficult to
treat. Infection control in hospitals is therefore an essential
part of hospital organization. Because hospitals are increas-
ingly being paid by DRGs, any secondary event prolonging
Chapter 15 Health Technology, Quality, Law, and Ethics
hospital stays may have adverse financial effects on the hos-
pital. In the USA, recent decreases in Medicare reimburse-
ments for nosocomial infections reflect this trend to provide
financial incentives to improve hospital infection control.
There is, therefore, a strong financial as well as professional
interest in reducing hospital-acquired infections.
A classic study of 32,000 hospitalizations in New York
State carried out by a Harvard University team showed
that 3.7 percent of hospitalized patients suffered adverse
events or injuries caused by medical mismanagement which
resulted in measurable disability. Of these, 28 percent were
due to negligence, so that 1.03 percent of all hospitalizations
involved medical negligence leading to measurable injury.
Of the total of some 100,000 adverse events in the study
group, 57 percent recovered within a month and 7 percent
had severe injury. Some 14 percent or 14,000 people with
adverse events died as a result; 51 percent of these deaths
were due to negligence. A 1999 report of the US National
Institute of Medicine estimated that between 44,000 and
98,000 people die annually in the USA from medical errors
occurring in hospitals, but these data are considered to be
overestimated in some studies. Adverse drug events (ADEs)
result in 700,000 emergency department visits and 120,000
hospitalizations annually, with US$3.5 billion spent on
extra medical costs. CDC estimates that at least 40 percent
of the costs of ambulatory (non-hospital setting) ADEs are
preventable.
A 2008 report by the Office of Inspector General of the
US DHHS reported that 13.5 percent of Medicare benefi-
ciaries experienced adverse events and that for 1.5 percent
of beneficiaries, these adverse events contributed to their
deaths. An additional 13.5 percent of beneficiaries in the
sample experienced temporary harm as a result of their
medical care, bringing the total percentage of beneficiaries
experiencing instances of care-related harm to 27 percent.
Nearly half (44 percent) of these adverse or temporary harm
events were preventable.
Hospital-acquired infections cause 99,000 patients
deaths in the USA every year (AHRQ, 2009). Higher rates
are seen among the elderly and the poor. Rates are lower in
teaching hospitals than in community hospitals. About 20
percent of the events were related to drug reactions or dos-
age errors. Less than 3 percent of those injured brought civil
litigation for the negligence. The search for “bad apples” –
that is, unethical, criminal, or incompetent health providers –
is necessary, but not sufficient to stem the problems created
by the health system itself. Prevention requires organized
activity. Investigation of adverse events helps to identify
methods of prevention and to protect the patient’s rights.
A program of measures to reduce hospital infection must be
based on epidemiological analysis of recorded events in the
search for common causes and preventable factors.
Organized surveillance and control requires a ratio of
one infection control practitioner per 250 acute care beds,
787
a trained hospital epidemiologist, and routine reporting of
wound infections to practicing surgeons (CDC, Hospital
Infection Program). Computer-aided medication dispens-
ing, as well as automated and other safety systems are
critical elements in minimizing morbidity and mortality
resulting from preventable human errors. In response to
the high frequency and cost of medical litigation, many
states in the USA have enacted legislation to restrict court
awards for medical negligence. Proposals for alternatives
to the tort system of medical malpractice compensation
include arbitration and mediation, an administrative sys-
tem similar to that used for workers’ compensation, and
a no-fault system of compensation, such as exists in New
Zealand, Sweden, and Finland. In a no-fault system the
complainant need not prove negligence on the part of the
provider, but only that he or she suffered an adverse event
which is compensable at standard rates depending on the
degree of disability. In the USA, federal legislation pro-
vides compensation for vaccine injuries, and three states
have enacted restricted no-fault systems for birth-related
neurological injuries.
In addition, there is greater emphasis on the adoption of
failsafe mechanisms, such as introducing warning systems
in anesthesia machines to alert the anesthetist if oxygen
flow in the patient’s tubing falls below a safe point. This
system was tested in Boston hospitals and found to reduce
adverse anesthetic events to zero cases over a 3-year period.
Vitamin K injection was made mandatory for all newborns
in New York State, as was already the case in some other
states, when a study showed deaths from hemorrhagic
disease of the newborn in cases where vitamin K was not
administered.
Inappropriate medical practice patterns are an equal, or
even larger problem for health systems. Comparisons of
surgical rates within the USA for coronary bypass proce-
dures, hysterectomies, and caesarean sections show wide
variation between different areas of the country. The costs
of excess surgery not only are economically wasteful but
also involve risks for the patient from the surgery itself
or anesthesia mishaps, infection, pain, and discomfort,
with legal and ethical questions of unwarranted interven-
tions not for the benefit of the patient. Health systems are
increasingly required to evaluate and control excess surgi-
cal, investigative, or other medical procedures, not only for
financial reasons but also for protection against litigation
and infringement of patients’ rights.
Licensure and Certification
The requirements that society establishes for allowing an
individual to practice medicine, and any health profession,
are vital to maintaining and improving the quality of care
(see Chapter 14). These standards require defining the train-
ing and experience needed by the individual, examination
788
procedures, and recognition for continued education and
maintenance of competence. This requires a statutory base
and national bodies operating under a national authority, sep-
arate from the agency operating the health system services.
Separation of licensing from operation of the health service
is essential in maintaining high professional standards.
The licensing authority is accountable to the state and
the public. In some cases, this function is delegated to self-
regulating professional bodies. In Canada, the licensing of
the medical profession and specialty recognition are carried
out by the medical profession with self-regulation. In the
UK, medical licensing is by a state-appointed board and in
the USA by state boards.
Medical schools, postgraduate training programs,
and fellowships are all subject to periodic comprehensive
assessments. Institutions that fail to meet the standard may
have funding or licensure suspended until they have per-
formed adequate remediation.
Health Facility Accreditation
Hospital accreditation in North America is by a voluntary
grouping of professional associations, including the Cana-
dian and American Colleges of Physicians and Surgeons,
the hospital associations, and the Colleges of Nurses. The
Joint Commission, originally operating in both Canada and
the USA, carries out regular inspections of hospitals. In Can-
ada, other organizations including the federal Department of
Health, provincial ministries of health, the Canadian Diabetes
Association, the Public Health Association, and the Standards
Council of Canada participate in the Joint Commission as
observers. Initially focusing on acute care hospitals, accredi-
tation has been gradually extended to cover special hospi-
tals, long-term facilities, home care programs, public health
departments, and ambulatory care services.
Health facility accreditation is a systematic, multidisci-
plinary inspection of the physical and organizational struc-
ture of the facility or program and the functioning of its
component parts. Factors measured include staff qualifica-
tions, facilities, organization, record keeping, and continu-
ing education of staff.
The process of accreditation requires a request for
accreditation from the board of governors of the hospital
or health facility, implying acceptance of the standards
of the commission. The accreditation process includes a
self-assessment, an on-site survey, and follow-up action
for correction of deficits and improvements. The com-
mission is invited to conduct a survey, and resurvey as it
sees fit. The hospital pays a fee and commits itself to pro-
vide all data requested and to cooperate with the site visit.
The commission issues a confidential report, giving the
accreditation rating and interim statement of deficiencies,
and requests progress reports in correcting deficiencies. It
is also empowered to carry out follow-up inspections and
The New Public Health
resurveys. Box 15.4 lists the areas of a large community
or teaching hospital, regional health authorities, hospitals,
and community-based programs and services, from both
private and public sectors, not only in Canada but around
the world.
The assessment survey examines the goals and objec-
tives of the organization and its administration, the direc-
tion and staffing of the facility, policies, and procedures.
Review includes medical staff organization, credentials and
review procedures, clinical privileges, selection of depart-
ment chairpersons and their responsibilities, standing com-
mittees, schedule of meetings, bylaws, and the role of the
governing board of the hospital. The presence and nature of
quality assurance organization, records review procedures,
and continuing educations are assessed. The quality of
clinical records is assessed by examination of charts for the
completeness of histories and documentation of the course
of the hospital stay including laboratory reports.
Each section of the program being accredited is assessed
in the following categories:
l statement of purposes, goals, and objectives
l organization and administration
l human and physical resources
l orientation, staff development, and continuing education
l patient care
l quality assurance.
These categories are also used in the programs covered
by the contracts between Accreditation Canada, formerly
the Canadian Council on Health Services Accreditation
(CCHSA), and other health and social service agencies.
Hospital accreditation was established in the UK and
Australia in the 1980s and is attracting interest in other
countries seeking ways to maintain and promote standards.
The procedure for accreditation of hospitals is still volun-
tary in Canada, but in effect has become universal for hos-
pitals of medium and large size (over 75 beds) and common
for smaller hospitals. It is seen as advantageous for the gov-
erning board and the community and also for the medical
staff in terms of medicolegal protection. In the USA, hospi-
tal accreditation has become virtually universal since pay-
ment for federally funded health insurance (Medicare and
Medicaid) beneficiaries is not allowed for non-accredited
hospitals, and many private insurers make this requirement
as well. In some states, accreditation is mandatory for all
hospitals.
Since the 1990s, CCHSA’s accreditation program has
expanded to cover a diversity of health care and service
areas, through contract arrangements with independent non-
hospital facilities such as highly specialized programs as well
as community health and social service organizations. In 2006,
CCHSA introduced standards for child welfare, hospice, pal-
liative and end-of-life care facilities, prison facilities, biomedi-
cal laboratories, and supplementary criteria for telehealth. In
789Chapter 15 Health Technology, Quality, Law, and Ethics
l System wide:
– Governance
– Infection prevention and control
– Leadership
– Leadership for aboriginal health services
– Leadership for assisted reproductive technology
– Leadership for primary care
– Managing medications
l Population based:
– Cancer populations
– Child and youth populations
– Maternal/child populations
– Mental health populations
– Populations with chronic conditions
– Public health services
l Service excellence:
– Acquired brain injury services
– Ambulatory care services
– Ambulatory systemic cancer therapy services
– Assisted reproductive technology
– Case management services
– Child welfare services
– Community health services
– Community-based mental health services and support
standards
– Critical care services
– Developmental disabilities services
– Diagnostic imaging services
– Emergency department services
– Health care staffing services
– Home care and support services
– Hospice palliative and end-of-life services
– Independent medical/surgical facilities
– Laboratory and blood services
– Long-term care services
– Medical imaging centers
– Medicine services
– Mental health services
– Obstetrics services
– Operating rooms
– Organ and tissue donation standards for deceased
donors
– Organ and tissue transplant
– Organ donation standards for living donors
– Point-of-care testing
– Primary care services
– Rehabilitation services
– Reprocessing and sterilization of reusable medical
devices
– Spinal cord injury acute services
– Spinal cord injury rehabilitation services
– Substance abuse and problem gambling services
– Surgical care services
– Telehealth services
l Service distinction:
– Acute stroke services
– Audit tool for reprocessing and sterilization of reusable
medical devices
– Inpatient stroke rehabilitation services
– Providing an integrated system of services to people
with stroke
Source: Accreditation Canada. Available at: http://www.accreditation.ca/
en/content.aspx?pageid=54 [Accessed 14 December 2012].
BOX 15.4 Accreditation Canada Standards
2008 CCHSA officially became Accreditation Canada, pro-
viding services to other countries. The ever-changing health
and social environment now accommodates specialized needs
in a diversity of service areas as an adjunct to the hospital
accreditation process. Examples are shown in Box 15.5.
Licensing and regulation of health facilities are a gov-
ernment responsibility, but an independent accreditation
authority has advantages. The national authority may fail to
monitor its own facilities with the diligence or objectivity
needed, and there may be a conflict of interest. Where there
is a national system of organization, distinct departmen-
talization of the operating and certification functions may
provide a greater measure of objectivity. Assistance from
countries experienced in voluntary accreditation can help to
establish accreditation mechanisms and provide technical
and professional support to countries wishing to establish
such programs.
In the current period of transition from central to decen-
tralized management of health services in many countries,
health facilities are being transferred from government
operation to independent operation as not-for-profit or
even for-profit facilities. Present methods of regulation by
national or state levels of government will require review
as decentralization and privatization take place. Regulation
by governmental authorities and non-governmental pro-
fessional bodies is mutually complementary in promoting
accountability, standards, and quality of services.
Peer Review
A large part of the work of clinical and departmental man-
agers in hospitals or other care settings relates to qual-
ity assurance. A major method of improving quality in a
health program is through peer review by which the staff
organizes systematic review of cases and records, using
statistics on performance indicators. In hospitals, this
includes review of deaths, maternal mortality and infant
mortality cases, surgical rates, complications following
surgery, and infection rates. Medical records and com-
puter information systems permit users to review records
http://www.accreditation.ca/en/content.aspx?pageid=54
http://www.accreditation.ca/en/content.aspx?pageid=54
790
BOX 15.5 Accreditation Canada International
Accreditation Program
The Canadian health services accreditation program began
in 1917 in conjunction with the American College of
Surgeons (ACS) with a hospital standardization program. The
first Minimum Standard for Hospitals developed require-
ments of just one page. In 1918, on-site inspections of hos-
pitals began, with 89 of 692 hospitals surveyed meeting the
requirements of the Minimum Standard. In 1926, the first
Standards Manual was issued.
In 1951, the American College of Physicians, the
American Hospital Association, the American Medical
Association, and the Canadian Medical Association joined
with the ACS to create the Joint Commission on Accreditation
of Hospitals (JCAH). It is an independent, not-for-profit orga-
nization whose purpose is to provide voluntary accredita-
tion. In 1953, the Canadian Hospital Association (now the
Canadian Healthcare Association), the Canadian Medical
Association, the Royal College of Physicians and Surgeons,
and l’Association des Médecins de Langue Française du
Canada established the Canadian Commission on Hospital
Accreditation. The Commission’s purpose was to create a
Canadian program for hospital accreditation, and in 1958
the Canadian Council on Hospital Accreditation (CCHSA)
was incorporated.
In 2008 CCHSA became Accreditation Canada
International. The accreditation program is used by all
types of health facilities, from large and complex hos-
pitals, to health systems, community health organiza-
tions, and residences providing long-term care. Its scope
includes a wide range of programs, including standards
on child welfare, hospice palliative and end-of-life care,
biomedical laboratory services, blood banks, and supple-
mentary criteria for Telehealth. The accreditation program
covers a diversity of health care and service areas, service
programs for brain injury, ambulatory care, assisted repro-
ductive technology – clinical and laboratory services,
Canadian Forces health services, cancer agencies, child
welfare organizations, First Nations and Inuit addictions
and community health services, the Federal Department
of Veterans’ Affairs, substance abuse and problem gam-
bling treatment services.
The accreditation service is on a contract basis with
specialized health programs, other federal government
departments, for-profit health facilities, and community
organizations across the provinces.
Accreditation Canada International works with other
countries to develop national accreditation programs for
their countries, and launched its first international program
for acute care, primary care, ambulatory care, and clinical
laboratories in 2010.
Source: Accreditation Canada International. Available at: http://www.
internationalaccreditation.ca/Accreditation/AccreditationProgram.aspx
[Accessed 12 September 2012].
The New Public Health
by diagnosis. These records can be utilized to assess other
events in hospitals, such as time from admission to sur-
gery, lengths of stay by diagnosis, response to abnormal
laboratory findings, and many other indicators of the pro-
cess of care. Obstetric departments can review the fre-
quency of and criteria for caesarean section deliveries.
Surgical departments review their appendectomy rates to
separate pathological findings from normal appendices.
Organized peer review has also been called medical audit
and essentially describes methods of self-policing and
education to learn from mistakes and experience and to
improve the quality of care.
In 1972, an amendment to the US Social Security Act
required hospitals and long-term care facilities to moni-
tor the quality of care given to Medicare and Medicaid
patients through professional standards review organiza-
tions (PSROs). These were medical audit committees with
specified tasks to conduct utilization review, medical care
evaluation, and profile analysis of physician or institutional
performance compared to accepted standards of the medical
community. In 1982, peer review organizations (PROs) were
created by federal statutes to replace PSROs. The PROs are
non-profit corporations, staffed by physicians and nurses,
to review medical necessity, quality, and appropriate level
of care under the Medicare and Medicaid programs. The
Centers for Medicaid and Medicare Services have an Office
of Clinical Standards to conduct surveys, provide certifica-
tion, and develop best practices guidelines, in a health care
quality improvement program (HCQIP).
Hospitals have departmental clinical meetings, adverse
incident or outcome committees, mortality rounds, and
clinical pathology conferences to help staff to evaluate and
learn from difficult cases. The presence of functioning peer
review mechanisms indicates that quality is of concern to
the professional and administrative network, raising the
consumer’s confidence in the system.
Maternal mortality committees have been widely used
to assess preventable factors in deaths related to maternity
and to point out areas of needed improvement in services.
Identification of high-risk pregnancies emerged from this
process and has become an important part of prenatal care.
Infant mortality reviews by professional groups can simi-
larly demonstrate areas of needed improvement in services.
Death rounds are held to review cases of death following
surgery or soon after admission, or “incidents”, such as
inappropriate medication given in error.
The successive waves of peer review initiatives in
the USA represent attempts by the federal government
to establish mandatory quality of care review by profes-
sional peers for facilities providing care to Medicare and
Medicaid patients. The concept of requiring standards of
care review has probably contributed to a greater aware-
ness of the accountability of hospital-based practice. Fre-
quent litigation may have contributed more to the sense that
http://www.internationalaccreditation.ca/Accreditation/AccreditationProgram.aspx
http://www.internationalaccreditation.ca/Accreditation/AccreditationProgram.aspx
Chapter 15 Health Technology, Quality, Law, and Ethics
the physician is accountable for services and outcomes of
care. PROs are a form of quality regulation that represent a
commitment by funding agencies to accountability in care
systems and to identification of organizational and admin-
istrative weaknesses in health care generally and not only
in hospitals. The generation of US physicians and health
systems managers trained since the 1970s accepts peer
review as an integral part of health services. Other coun-
tries use this kind of mechanism to maintain and promote
quality of care.
Tracer Conditions
Tracer conditions are common medical conditions (or pro-
cedures) for which diagnostic criteria are well established
and clear, there are effective preventions or treatments, and
a lack of treatment can cause significant harm to the patient.
Examples of tracer conditions include otitis media, appen-
dectomy, caesarean section, and hysterectomy. These con-
ditions, if evaluated in terms of incidence and actual chart
review, can provide useful insights into departmental medi-
cal standards. Incident reports by nursing staff and nosoco-
mial infections are examples of the functioning of the tracer
condition concept.
Incident reports in hospitals are designed to determine
the causes of errors, so that remedial action can be taken
and similar events prevented. Tracer condition studies have
become such an accepted part of modern health manage-
ment that the absence of an organized review system could
be considered a serious structural flaw in a health service,
requiring remedial action.
Setting Standards
Standards recommended by independent professional
organizations or by advisory committees appointed by
ministries of health can play important roles in defining
standards of care for specified conditions. In addition,
organized professional bodies can issue practice guidelines
or help governments or health care agencies to develop
standards or algorithms for management of specific topics
and conditions.
Specifying standards for preventive care, such as for
infants and adults, assists local health authorities in plan-
ning and evaluating services. The American Academy of
Pediatrics (AAP) has an extensive professional committee
structure that publishes periodic guidelines for pediatricians
on a wide variety of infant and child topics including nutri-
tion, immunization, prevention of anemia and lead toxicity,
child safety, and school health. Mandatory preventive care
for newborns includes eye care and vitamin K injection in
the USA (see Chapter 6). Mandatory immunization require-
ments for school entry and for health care personnel are dis-
cussed in Chapter 4.
791
The American Public Health Association (APHA) pub-
lishes the Control of Communicable Diseases Manual, now
in its nineteenth edition (2008). It is the authoritative US
manual on this topic. The AAP’s Red Book on infectious dis-
eases is used across North America by pediatricians in clini-
cal practice. These organizations and their counterparts in
obstetrics and many other clinical fields directly relevant to
public health continually update practitioners and policy per-
sonnel in the “state of the art” or “gold standard”, discussed
previously. This constitutes a professional self-guidance sys-
tem in standards. Managed care and other health provider
systems also issue guidelines for member practitioners that
serve to maintain standards of service.
The wide use of treatment protocols and scoring systems
in hospital medicine helps to define standards of care in a
measurable way. The Apgar score for rating newborn sta-
tus has been a standard in hospitals worldwide for decades,
helping to standardize infant assessment and care. The
APACHE (Acute Physiology And Chronic Health Evalua-
tion) scoring system is used widely to assess the chances
of survival of patients admitted to intensive care units and
to compare outcomes, for example, between teaching hos-
pitals and community hospitals. It is also used in assessing
patient outcomes with different modes of treatment. Scor-
ing systems are also used in community health care, as in
risk scoring for pregnancy care (see Chapter 6).
Algorithms and Clinical Guidelines
Algorithms are decision trees or a systematic series of deci-
sions based on the outcomes of previous decisions, tests, or
findings. Derived from operations research, this approach
applied to medicine identifies all available choices (e.g.,
exposed versus non-exposed) and follow-up decisions
based on findings from each previous option substantiated
by observation. It is often presented graphically like the
branches of a tree, showing the alternatives and subsequent
decisions to be made.
A clinical algorithm is a systematic process defining
a sequence of alternative, logical steps depending on out-
comes of previous ones, incorporating clinical, laboratory,
and epidemiological information, applied to maximize ben-
efits and minimize risks for the patient. It gives the provider
a review of the relevant literature and recommended stan-
dards of practice on a particular topic for preventive care
or case management. These guidelines are usually arrived
at by consensus of multidisciplinary working groups taking
into account published studies on the topic. The guidelines
may suggest that some procedures should not be carried out
routinely.
Clinical guidelines are meant to establish accepted stan-
dards of care and may have important economic implica-
tions. Medical Letter, published by the Consumers’ Union,
is a long-standing and useful publication that reviews
The New Public Health792
TABLE 15.5 Adult Health Maintenance Checklist by Age Group
Age (years)
Procedure 20–39 40–64 65+
Checkup visit Every 3 years Every 2 years Annually
Cholesterol With checkups With checkups With checkups
Fecal occult blood Age 40–49 if high risk Annually Annually
Clinical breast examination Every 1–3 years Annuallya Annuallya
Mammography Baseline age 35 Age 40–49, every 1–2 years Over 70, every 2 years
Pelvic examination Every 1–3 years Every 1–3 years Every 1–3 years
Pap smear From age 21-29 every 3 years;
from 30-65 every 5 years with
HPV DNA test
From age 30–65 every 5 years
with HPV DNA test
If previously negative,
may stop 3 years
Colonoscopy No From age 40 for those with family
history of colon cancer or polyps.
After age 50, every 3–5 years
After age 50, every 3–5 years
Prostate and PSA Immunizations No Annuallya Annuallya
Tetanus–diphtheria Every 10 years Every 10 years Every 10 years
Pneumococcal pneumonia For high risk For high risk Every 6 years
Influenza For high risk For high risk Annually
Skin cancer Annuallya Annuallya Annuallya
Bladder cancer Annual routine urinalysis Annual routine urinalysis Annual routine urinalysis
Lung cancer Routine examinationb Routine examinationb Routine examinationb
Testicular cancer Routine examinationb Routine examinationb Routine examinationb
Oral cancer Routine examinationb Routine examinationb Routine examinationb
Ovarian cancer Routine examinationb Routine examinationb Routine examinationb
Pancreatic cancer Routine examinationb Routine examinationb Routine examinationb
Routine vitamin supplements Routineb Routineb Routineb
Note:
PSA = prostate-specific antigen.
Agency for Healthcare Research and Quality. Rockville, MD: AHRQ. http://www.ahrq.gov [Accessed 13 September 2012].
aInconclusive
bnegative recommendation. The topics are under continuing review, and recommendations are in some cases left to the opinion of the provider as the current
cumulative evidence is not affirmative, e.g., clinical breast examination annually or breast self-examination.
Sources: US Preventive Services Task Force Ratings: Strength of recommendations and quality of evidence. guide to clinical preventive services. 3rd ed.
Periodic updates, 2000–2003. Available at: http://www.uspreventiveservicestaskforce.org/3rduspstf/ratings.htm [Accessed 13 September 2012].
therapeutic issues of everyday medical practice and the rele-
vant studies. It represents a balanced, updated view of medi-
cal practice and summaries of current literature, reviewed
by respected, experienced, and competent medical authori-
ties. Clinical practice guidelines are produced by hundreds
of professional, medical, and governmental agencies in
order to standardize and improve medical care.
Clinical and preventive care guidelines are helpful
in clinical practice and in preventive medicine. They are
increasingly used in managed care environments to assure
standards, quality of care, and cost-effectiveness as well as
legal protection. Guidelines for preventive medicine and
public health practice are also part of the process of pro-
moting the quality of individual and community health,
as discussed in Chapter 11. Annual revision of the infant
immunization program, discussed in Chapter 4, is a prime
example, as is the set of guidelines for preventive care for
adult health maintenance in Table 15.5.
The issue of application of current scientific knowledge
for population health is a continuing struggle for recognition
of the prime importance of health promotion and preventive
care for health of a population. The selection of priorities in
use of resources is vital especially in the many developing
countries that are in various stages of economic development,
or which have abundant income from natural resources such
as oil and minerals. Implementation of programs designed to
achieve the MDGs can help to serve this purpose.
Public health standards and clinical practice guidelines
are an increasing part of quality improvement. It is impor-
tant, however, that they are developed as best practices and
http://www.ahrq.gov
http://www.uspreventiveservicestaskforce.org/3rduspstf/ratings.htm
Chapter 15 Health Technology, Quality, Law, and Ethics
influenced as little as possible by commercial interests of
drug or vaccine manufacturers. The proliferation of such
guidelines by health authorities or professional associations
of the USA, the UK, Canada, Australia, and other countries
indicates a wide consensus on the importance of such writ-
ten standards, guidelines, or “best practice” statements. The
recommended childhood immunization program put for-
ward annually by the CDC in conjunction with the AAP and
other professional organizations is an example of such best
practices and is accepted by health insurers and providers as
the gold standard in this field. The concept of promotion of
quality in health care and the adoption of current scientific
standards are global issues and an integral part of the New
Public Health (Box 15.6).
BOX 15.6 EISC: The Excellence in Science Committee
of the Centers for Disease Control and Prevention
The Excellence in Science Committee (EISC) promotes the
Centers for Disease Control and Prevention’s (CDC’s) sci-
entific infrastructure and facilitates communication and
collaboration that enhance scientific areas and activities
needed for state-of-the-art conduct of science. EISC serves as
a consulting body for science-related issues and makes rec-
ommendations to the CDC to foster, support, and protect an
environment for the promotion of scientific integrity, quality
assurance, and the rapid dissemination of scientific innova-
tions, technology, and information, with the ultimate goal of
improving public health.
EISC’s specific functions include:
l promoting and protecting the scientific infrastructure
l providing a forum for information exchange among
administration, directors for science, and liaison work-
ing members/groups
l communicating science-related issues to the CDC and
related scientists
l promoting professional development and training
l recognizing and rewarding quality science
l acting as an advocate for scientific resources
l identifying and disseminating new information, e.g.,
new statistical/epidemiological techniques or new scien-
tific technologies
l developing, revising, and promoting the implementation
of cross-cutting scientific policies and procedures
l serving as a consulting body for science-related issues
and making recommendations to the CDC
l fostering the development of methods for assessing and
monitoring:
– the environment for quality science and qualitative
and quantitative scientific output within CDC and
related organizations
– the impact of CDC science on public health.
Source: Adapted from Centers for Disease Control and Prevention.
Science coordination and innovation [updated 3 November 2011].
Available at: http://www.cdc.gov/od/science/excellence/ [Accessed 13
September 2012].
793
The Canadian Province of Saskatchewan Health Ser-
vices Utilization and Research Commission publishes peri-
odic reports presenting consensus positions of panels of
medical faculty, clinical specialists in pathology and physi-
cal medicine, and public health specialists in nutrition, com-
munity health, and epidemiology. Its reports are circulated
widely and serve to update medical practitioners, reduce
unnecessary testing, promote appropriate use of laboratory
and other diagnostic procedures, and provide standards of
care for individual patients and community services, such
as long-term care facilities and home health agencies.
The Canadian Medical Association issued its Handbook
on Clinical Practice Guidelines in 2007, based on a system-
atic review of the literature, interviews of key professionals,
consensus conferences, and continuing evaluation of both
process and content of such guidelines. The Guideline Inter-
national Network (GIN) Fourth International Conference,
held in Toronto in 2007, involved experts in national and
international practice guidelines from 31 countries to share
experience and concepts in this ongoing field. The GIN
library contains more than 6600 (by October 2012) guide-
lines, evidence reports and related documents, developed or
endorsed by GIN member organizations (GIN, 2012).
An Institute for Clinical Evaluation (ICES) organization
at the University of Toronto, established in 1992 with core
funding provided by Ontario’s Ministry of Health and Long
Term Care, is mandated to conduct research that contributes
to the effectiveness, quality, equity, and efficiency of health
care and health services in Ontario. ICES uses an interdis-
ciplinary research approach to health care, health services,
and health policy.
The American College of Cardiology (ACC) provides a
framework of evidence-based clinical statements and guide-
lines developed by leaders in the field of cardiovascular med-
icine with continuing adoption of new scientific information
and experience in many aspects of this field (ACC, 2012).
Many professional organizations such as the AAP, American
Congress of Obstetricians and Gynecologists (ACOG), UK
Faculty of Public Health, and European Society of Cardiol-
ogy produce clinical guidelines which are updated regularly
to provide physicians and health systems managers with
current consensus on state-of-the-art standards, such as the
European Society of Hypertension Guidelines released in
2013 (i.e., less than 140 mm. systolic for all).
The US Health Care Financing Administration (HCFA),
Center for Medicare & Medicaid Services (CMS), and
National Institutes of Health (NIH) have consensus pro-
grams to develop guidelines that are widely disseminated
and set standards of practice. In 1977, the NIH issued its
first consensus paper on breast screening for cancer, and
this has been followed by many other topics each year
since. The AHRQ also produces research related to efficacy
of current and new practices and training material to pro-
mote their diffusion across the US health system. Cochrane
http://www.cdc.gov/od/science/excellence/
794
reviews and the Cochrane Library provide high standards of
literature reviews and meta-analysis on many topics which
serve to guide practitioners and policy makers in current
standards. The US Healthy People 2020 project provides
gold standards for preventive care which serve clinicians,
public health practitioners, and health planner standards
for their work. Evidence-based consensus guidelines were
issued on the following topics: breast cancer screening
for women aged 40–49, interventions to prevent HIV risk
behavior, management of hepatitis C, genetic testing for
cystic fibrosis, acupuncture, and effective medical treat-
ment for heroin addiction.
Clinical guidelines are increasingly being promoted
by professional, governmental, and managed care orga-
nizations with the purpose of promoting rational use of
health care resources and at the same time promoting
standards of care to incorporate good standards of clinical
practice. Clinical practice guidelines are now common in
the practice of primary care, mental health, and clinical
specialties. The University of Southern California’s list
of clinical guidelines website (http://medicine.ucsf.edu/)
provides access to hundreds of websites for such practice
guidelines.
Clinical guidelines provide practicing doctors, peer
review committees, health care managers, managed care
companies, governmental bodies, and professional organi-
zations with channels to set standards of practice and expec-
tations of care standards. Legal aspects of health care also
increasingly recognize the importance of clinical guidelines
where committees of appropriate medical professionals
convene and set out average or minimum standards of care
for defined clinical entities. Thus, peer-reviewed guide-
lines set an appropriate standard (a silver if not a gold stan-
dard) for judging malpractice or adequate practice. Clinical
guidelines should be under periodic review and subject to
critical discussion and updating using the Cochrane review
methods of literature review and analysis. Promotion by
advocacy or special interest groups can be constructive,
but the influence of drug companies can be insidious and
reduce the professional objectivity of such reviews and their
recommendations, a concern that must be carefully moni-
tored and continuously kept in mind as a potential compro-
mising bias.
The AAP produces policy statements, practice param-
eters, and model bills which have a wide distribution and
influence; they are published in the academy’s journal, Pedi-
atrics. The AAP clinical practice guidelines issued include
diagnosis and treatment of urinary tract infection in febrile
infants and young children, long-term treatment of the child
with simple febrile seizures, management of acute gastroen-
teritis in young children, management of otitis media with
effusion in young children, and others. The policy state-
ments of the AAP cover a wide range of topics including
use of bicycle helmets, 55 mile per hour maximum speed
The New Public Health
limits, folic acid for the prevention of neural tube defects,
and ethics in the care of critically ill infants and children.
AAP guidelines are valid for 5 years only and are reissued
or reconfirmed in order to keep up to date and to incorporate
new or revised knowledge into practice standards.
Empirically derived, peer-reviewed, regularly updated
guidelines have become an appropriate standard for prac-
tice and for judging malpractice, as well as balancing qual-
ity and cost-effectiveness. Clinical guidelines may become
restrictive, but they help to reduce practice by whim and
unsubstantiated belief to improve the quality of care over-
all. In large health care organizations they provide a basis
for continuing education for staff and advancement of stan-
dards of the organization.
The Community Guide produced by the CDC provides
an excellent source of evidence-based advice for community
programs. It serves the needs of public health professionals,
health care providers, legislators and policy makers, research-
ers, community-based organizations, employer–employee
groups, and other purchasers of health services. The guide
covers a wide range of health issues including alcohol, can-
cer, diabetes, mental health, motor vehicle safety, nutrition
and obesity, oral health, physical activity, pregnancy, sexual
behavior, social environment, substance abuse, tobacco, vac-
cines, violence, and workplace health issues.
In 1999, the UK National Health Service (NHS) estab-
lished the National Institute for Clinical Excellence (NICE)
as an independent organization to provide guidelines for
public health, health technologies, and clinical practice
guidelines for specific conditions. The Health Development
Agency of the NHS was included in the NICE organization
in 2005. Now called the National Institute for Health and
Care Excellence, NICE publishes guidelines that provide a
helpful basis for clinical practice and public health as well
as other areas in the NHS to update the services provided.
Topics for public health include smoking and tobacco con-
trol, diet and obesity, exercise and physical activity, sexual
and mental health, and alcohol.
ORGANIZATION OF CARE
Administrative and financing systems are essential ele-
ments of quality assurance. They can be designed to pro-
mote standards of care and to reduce fiscal incentives that
foster excess supply and overservicing. The organization
of financing health care has important implications for
quality, technology, and ethical issues in the New Public
Health.
Diagnosis-Related Groups
DRGs, discussed extensively in Chapter 11, were developed
in the 1960s as an alternative way of paying for hospital care
in order to encourage shortened lengths of stay. Experience
http://medicine.ucsf.edu/
Chapter 15 Health Technology, Quality, Law, and Ethics
with payment by days of care (per diem) showed that it pro-
moted unnecessary, lengthy, and potentially dangerous use
of hospital care, an important factor in the rapid escalation
of costs in the health system. DRGs were adopted for pay-
ment for Medicare beneficiaries in the USA in 1983 and
later became the standard method of payment for all insur-
ance systems.
In the DRG system the insurer pays the provider hos-
pital for a procedure or diagnosis rather than the number
of days of stay in hospital. This has led to a large reduc-
tion in hospital days of care and a remarkable growth in the
number of surgical procedures done on an outpatient basis.
Since the introduction of DRGs, outpatient surgical proce-
dures have grown from less than one-fifth to more than half
of inpatient surgical cases. Outpatient surgery is safer for
the patient and less costly to the insurer. DRGs have gradu-
ally been adopted as a case payment system for reimbursing
hospitals in most developed countries.
The DRG system is widely considered to promote qual-
ity of care as an active process focusing on quickly address-
ing the diagnosis and management of the patient with rapid
mobilization of treatment and return home. Critics of this
system allege that DRGs encourage inappropriate early
discharge of patients before optimal patient education and
follow-up care have been provided, but long length of hos-
pital stay has not been shown to improve patient outcomes.
Critics also suggest that this may promote altering diagno-
ses to higher cost units of service. Others think that DRGs,
by reducing length of stay, have turned hospitals into inten-
sive care units with ultra-sick patients. Despite these issues,
the trend towards short hospital stays and newer approaches
to active treatment seems to be compatible with better care
and improved outcomes, according to some measures. The
rapid decline in mortality rates from coronary heart disease
is thought to be due in large part to the activist treatment
approach, with lengths of stay of 1 week or less for acute
myocardial infarction compared to 6 weeks on average up
to the 1970s.
Managed Care
Managed care systems developed in the USA in response
to rapid cost escalation for health care and the successful
experience of HMOs. Managed care is based on the con-
cepts of resource management, and quality assurance with
rationalized use of technology. The system developed over
time with checks and balances to provide comprehensive
care at lower cost than traditional fee-for-service systems
by discouraging excessive utilization without compromis-
ing quality of service. Managed care systems include tradi-
tional HMOs and various other organizations which employ
physicians or are made up of independent physicians work-
ing together who own or contract for hospital services (see
Chapter 10).
795
HMOs, both for-profit and not-for-profit, and man-
aged care itself, have been widely criticized as excessively
limiting patient access to appropriate care in the interest
of cost containment. The 2010 PPACA (Obamacare) is
promoting development of newer innovations including
patient-centered medical homes (PCMH), accountable
care organizations (ACOs), and population health man-
agement systems (PHMSs), and early evidence shows that
these models are quality management approaches for inte-
grated primary and hospital patient care (see Chapter 11).
Obamacare is a highly politicized and much debated topic
in the USA; it seems likely to make a very big difference
in coverage and fair practices of insurance with lower
costs of private insurance.
District health systems in the UK, the Scandinavian
countries, and the post-Soviet model of health care incor-
porate organizational and financial linkage between care
systems and funding from tax sources. HMOs, sick benefit
funds, and district health systems provide both prepayment
and health services. Even in traditional private health insur-
ance systems, the insurer is increasingly taking on the role
of regulating reimbursement for medical services in order to
contain costs and curb abuses by providers. In this context,
emphasis is placed on maintaining health, preventive care,
and financial incentives to efficiency in overall care. Clini-
cal indications, utilization review, and organizational and
professional standards are now becoming accepted parts of
the health insurance milieu.
The competition between hospitals for referrals from
managed care plans in the USA has created a market
situation in which a high proportion of hospital beds are
empty, and in which mergers or closures of hospitals
are common. Closures or reductions in hospital bed sup-
ply are also occurring in the UK and in most industrial-
ized countries of Europe.
PERFORMANCE INDICATORS
Performance indicators are measures such as morbidity,
mortality, functional status, or immunization rates in a com-
munity, used to monitor the functioning of a health service.
Routinely collected statistics are analyzed to compare per-
formance against objectives, help monitor efficiency and
effectiveness, point out problem areas within the service,
and plan new health programs. This method is based on the
use of the concept of management-by-objectives in health
administration to promote achievement of national health
targets.
The UK has a strong tradition of mapping diseases as a
basis of epidemiological analysis and has applied this strat-
egy to mapping of performance indicators to assess health
care performance. The UK financing system is based on
capitation adjusted by standardized mortality rates on the
premise that mortality rates standardized and compared to
796
the national average serve as indicators of need. In this way,
the approach helps to promote equitable funding among
wealthy and poorer regions of the country, and thereby
improve services in areas of greater need.
Performance indicators were introduced into the NHS
during reforms of the late 1980s, providing a series of
outcome or performance measures that are used to adjust
payments allocated on a per capita basis to district health
authorities. These authorities can be penalized for low rates
of immunization, whereas general practitioners receive
incentive payments for full immunization coverage. The
result was a rapid improvement in immunization coverage
of infants and children compared to rates in the previous
decade. Incentive payments in many countries encourage
women to go to hospitals for delivery or to attend prenatal
care by making social maternity grants conditional on seek-
ing care.
Use of performance indicators requires the development
of health information systems with district health profiles
to provide ongoing monitoring of health indicators in a dis-
trict, compared to regional and national rates and targets.
Health profiles help to establish and monitor the prevalence
of chronic disease and measure the impact of health ser-
vices. This enables the study of the performance of preven-
tive and curative services, such as managing hypertension
to reduce the incidence of strokes and related conditions.
There are criticisms of performance indicators alleging a
potential for manipulation and abuse of health intervention
measures when the financial incentives are used for a spe-
cific activity. However, financial incentives are part of the
DRG system and have been successfully used in the UK to
improve vaccination coverage and implementation of other
preventive health practices by family physicians. In Israel in
2007, payments to hospitals provided a bonus for surgical
interventions for hip fracture within 48 hours of the event,
resulting in a marked rise in early intervention and a reduc-
tion in mortality from hip fractures.
CONSUMERISM AND QUALITY
With decentralization and the growth of managed care,
health systems must increase their attention to the attitudes
of the consumer. Quality is, in part, how the client per-
ceives the system, and how the system meets client needs
in an acceptable manner, where privacy, dignity, the right to
know, and the right to a defined set of services are protected.
However, the rights of the client are not unlimited. A public
or private health plan has the duty to manage the basket of
services responsibly, which includes limitations such as in
access to specialist services.
Patients’ rights and consumer protection in health care
often (but not always) include the right to select and change
a health care provider, as well as the right to receive high-
quality care for a designated range of services. The UK NHS
The New Public Health
issued a patient’s Charter of Rights during the 1990s, which
is perhaps idealistic and may not be actualized in practice,
but still outlines an ideal of value both for practical applica-
tion and for legal rights. The consumer’s formal protection
includes the right to complain and to seek redress of grievance
and compensation for injury suffered from neglect or incom-
petent care (Box 15.7). In North America and Europe, there
are at least four models of defining the rights of patients: the
paternalistic model, the informative model, the interpretive
model, and the deliberative model (WHO, 2012). Many new
charters have been established such as data protection, end-
of-life care, mental health, access to health services, quality
of care and care giving environment, nationally approved
treatments, drugs and programs, respect, consent and confi-
dentiality specific to the UK and in Canada, New Zealand
and other countries. The new US federal Affordable Care Act
of 2010 (PPACA) includes a large element of patient’s rights
protection, as discussed in Chapter 10.
The patient or consumer of health care needs to be
informed and conscious of health care costs if efforts to
restrain cost increases are to be effective. Public attitudes
are vital in terms of self-care, demands on the health service,
and limitations to the potential of health care and resources
for health care. The media and consumer organizations can
play important roles in advocacy for health, in raising pub-
lic consciousness of self-care, and as watchdogs on abuses.
Consumer acceptance is manifested through choice of
health plan and practitioner, or by seeking alternative care
BOX 15.7 Patients’ Rights, European Union, 2009
A review of patients’ rights in countries of the European
Union in 2009 focused on the following:
l Right to informed consent based on access to informa-
tion for care or participation in research
l Right to information concerning own health, diagnosis
l Right to medical records
l Right to confidentiality of personal and health informa-
tion and physical privacy during care
l Right to complain and compensation
l Right of free choice of provider and of treatment
l Respect of patient’s time
l Right to observance of quality standards access to high-
quality health services
l Right to safety and freedom from harm caused by the
poor functioning of health services, medical malpractice
and errors, and the right of access to health services and
treatments that meet high safety standards
l Right of access to innovative procedures, including diag-
nostic procedures, according to international standards and
independently of economic or financial considerations.
Source: European Patients’ Forum. Patients’ rights in the European Union.
Available at: http://www.eu-patient.eu/Documents/Projects/Valueplus/
Patients_Rights [Accessed 25 October 2012].
http://www.eu-patient.eu/Documents/Projects/Valueplus/Patients_Rights
http://www.eu-patient.eu/Documents/Projects/Valueplus/Patients_Rights
Chapter 15 Health Technology, Quality, Law, and Ethics
privately when service is unacceptable because of quality or
style. Erosion of confidence in a public system of care can
lead to a two-tier system with the public system serving the
poor and a private parallel system serving the middle and
wealthy classes. Such a division can seriously undermine a
public system unless it is addressed by improving the qual-
ity and manner of the service and by establishing supervi-
sion and limitations on public and private practice.
The growing inequality caused by the rise of private
practice outside a national health care system is a chronic
problem in the UK’s NHS, in Israel’s health system, and in
many countries developing their health systems through par-
allel public and private care. The issue is also surfacing in
the USA in the transition to managed care with its inherent
limitations of choice for people insured through their place of
work or covered under the Medicare and Medicaid programs.
The PPACA requires insurance companies to accept any-
one requesting cover without restrictions due to prior con-
ditions or high expenses for serious conditions, and without
other forms of discrimination common in the past. It also
includes provisions for coverage of preventive care services
and incentives for quality improvement. Extra billing, banned
in Canada’s national health insurance plan, is a recurring
issue with the medical profession in some provinces.
Consumer knowledge, attitudes, beliefs, and practices
are part of the health system, from health promotion to
tertiary care. Informed and health-conscious consumers
are stronger partners in the health system in achieving
improved health than an ill-informed and apathetic public,
so that health education and health promotion are funda-
mental to modern public health. The role of the consumer
in health care is unique in that there is a significant infor-
mation asymmetry between the consumer and provider.
Health education programs and wide use of the Internet
increase access to health and medical information, but this
gap can never completely be eliminated. Patients may use
their power as consumers to demand inappropriate care,
such as unnecessary surgery or antibiotics when clearly
not indicated, because of their preference for interven-
tion and action over watchful waiting. However, there is
an equal or perhaps greater danger of provider-induced
demand for repeated and possibly unnecessary interven-
tions that may be related to methods of paying the doctor
or the hospital. The traditional doctor–patient relationship
is still an important factor for the interests of patients and
their health. A still effective method of having an individ-
ual quit smoking is a brief but stern lecture by the family
physician.
THE PUBLIC INTEREST
Population-based interventions are often more effective
and less costly ways to reduce morbidity and mortal-
ity than individual prevention or treatment services. A
797
population-based preventive program may require behav-
ior change by the individual, such as in mandatory seat
belt and motorcycle helmet enforcement or banning smok-
ing in public places. Fortification of flour, milk, and salt
with essential micronutrients is a well-established public
health measure. There is an element of compulsion in this,
with the social gain usually considered to be sufficiently
important to outweigh individual rights. Immunization
is for the protection of individuals but also for the popu-
lation, so that refusals to immunize children and adults
can cause injury to others. Herd immunity is protective
of people who are at high risk. Mandatory immunization
for school entry in the USA has been effective in increas-
ing coverage to levels akin to the most advanced health
systems, over 95 percent coverage. Refusals and failure to
harmonize immunization policies in Europe have resulted
in mass epidemics of measles, rubella, and mumps in
recent years.
There is often a delicate balance between community
rights and individual rights which can lie at the heart of
many controversies in modern public health and health
care, ranging from chlorination or fluoridation of com-
munity water supplies to managed care systems for
health services. Women’s rights, gay rights and abortion
are highly controversial and politicized in the USA, and
in many other countries. The differences can become
extreme and the source of international strife, such as in
the movement to promote fundamentalist Sharia law in
many countries that are severely restrictive of women’s
and minority rights.
In public health, issues should be examined on their
merits, especially in terms of what is accepted as good pub-
lic health practice, based on evidence from clinical trials,
documented experience, and best practices in other coun-
tries. The evidence of successful public health measures in
improving individual and collective health status is power-
ful, yet must always be balanced within the context of indi-
vidual rights and the public interest. The ethical issues of
individual and community rights of public health are dis-
cussed later in this chapter.
TOTAL QUALITY MANAGEMENT
Total quality management (TQM), as discussed in Chap-
ter 12, was adapted from business management theory and
practice to health care in the 1990s and provides a basis for
promoting continuous improvement in health care systems.
TQM involves everyone in the system, from all levels of
management to production or service personnel and sup-
port staff, and thus helps to raise staff morale because of the
shared involvement. Health is provided through multidisci-
plinary groups which need to approach problems with open
and shared scientific inquiry and hypothesis formation, test-
ing, and revision to find operational solutions to problems.
798
Electronic health records and information technology pro-
vide many new opportunities to improve patient care and
data systems for monitoring the health status of population
groups for process and outcome measures, or health targets,
such as immunization coverage, or screening compliance
for colon, cervical, or breast cancer, as measures of perfor-
mance in primary care. Information technology adds a great
deal of capacity for quality monitoring and improvement
measures.
TQM incorporates statistical methods, comparing varia-
tions in patterns of service or use of resources. It employs
epidemiological methods to draw conclusions for policy
needs. It looks for continuous improvement, encouraging
cooperation, and motivation to achieve common goals of
service and client satisfaction. Psychological theory helps to
foster higher levels of motivation, with early identification
and resolution of conflict. Leadership is shared, and there is
a basic need for cooperation. Cost and quality are interre-
lated, as poor quality leads to waste, inefficiency, and dis-
satisfaction of both clients and staff. High-quality, humane,
and effective services are especially important in a competi-
tive environment where clients have the right to choose and
where costs and efficiency are factors in the well-being and
indeed the survival of institutions.
Medical care is increasingly practiced in larger health
care organizations. To provide technically competent
medicine is not by itself sufficient. The patient’s rights
and sense of personal worth are also of great importance.
Financial incentives can be effective in redirecting health
care priorities, such as in reducing hospital length of
stay and admissions, but may result in the patient or the
family feeling that they are not receiving the best care.
DRGs, HMOs, and other organizational and funding sys-
tems meant to increase efficiency of care may have the
effect of alienating patients from a health care system.
Staff attitudes towards patients are important for client
satisfaction. The service must include ready access to
a continuum of supportive services, such as home care
and counseling, so that the patient and family do not feel
abandoned by the system.
A byproduct of TQM is continuous quality improve-
ment (CQI), by which institutions wishing to improve qual-
ity train and empower the staff to work in teams to assess
their own performance and seek solutions to problems in
their operational unit. People of different ranks and profes-
sions work in a network organization as well as in a tradi-
tional hierarchical organization in which rank and seniority
provide authority. This community of practice is important
for staff morale and a shared sense of responsibility for the
patient and the institution.
CQI involves multidisciplinary approaches, not only
to review problems but also to seek better ways of func-
tioning and improving consumer satisfaction. The process
includes all those involved in providing care, support ser-
vices, and administration of a department, hospital, clinic,
The New Public Health
or community health program. This is not only professional
self-policing but a method to find better ways of meeting
needs and using resources. The involvement of all provid-
ers improves motivation and promotes a sense of common
purpose in the organization.
Applying these principles in a health care setting can
take many forms. Selection of topics by TQM/CQI com-
mittees in a hospital or another health facility may be
based on surveys or interviews with staff, patients, or
management. Satisfaction surveys among women follow-
ing delivery in an obstetrics unit could point out remedi-
able problems. An obstetrics department may be faced
with issues related to high or low volume of deliveries,
staff training, equipment and supplies, communica-
tion among staff, and among staff and patients and their
families, cleanliness, sterile technique, staff satisfaction,
client satisfaction, and many others. The team looking
at such a problem should be multidisciplinary, and the
emphasis should be on client attitudes and satisfaction.
Examination of the function of an emergency depart-
ment in a hospital would similarly look at many functional
and attitudinal aspects of the service including staff atti-
tudes, training needs, waiting times, consultation ser-
vices, and others. Addressing waiting times, for example,
can lead to ways to reduce these substantially, improving
both client satisfaction and the efficient management of
the emergency department. Any service is there to serve
patients and the community. A service is not primarily for
the benefit of the staff, but staff satisfaction and morale are
essential for successful service to clientele. CQI can also
be applied to assessing and improving compliance with
clinical guidelines or evidence. An example is assessing
the proportion of diabetics whose hemoglobin A1c (HbA1c)
is measured at least twice annually, who have eye and
feet examinations regularly, or whose blood pressure is
managed with an angiotensin-converting enzyme (ACE)
inhibitor.
The European Region of the WHO and the national
medical associations in Europe agreed in 1995 that medi-
cal associations should take leading roles in programs of
CQI to achieve better outcomes of health care in terms of
functional ability, patient well-being, consumer satisfac-
tion, and cost-effectiveness. This is in keeping with the
European Region’s Health for All targets: there should
be structures and processes in all member states to
ensure continuous improvement in the quality of care and
appropriate development and use of health technolo gies.
The introduction in the 1990s of general practitioner
fundholding for hospital care for patients on the general
practitioners’ roster in the UK encouraged the hospital to
maximize patient satisfaction with the care system. This
promotes application of CQI to improving the quality and
acceptability of care. Similarly, performance indicators
provide regional and district health authorities in the UK
with tools for CQI approaches. The UK NHS established
799Chapter 15 Health Technology, Quality, Law, and Ethics
The National Institute for Clinical Excellence (NICE), estab-
lished in 1999, has a mandate to review health service treat-
ments and effective therapies that should be commissioned
and made available within the National Health Service (NHS)
throughout England and Wales. The mission statement for
NICE is that it “contributes to better health around the world
through the more effective and equitable use of resources”.
In 2005 NICE was revised to include reviews of public health
interventions, and its mandate was expanded to include quality
standards for the English social care sector (English Health and
Social Care Act of 2011). Now renamed the National Institute
for Health and Care Excellence, NICE operates as a statutory
independent special health authority in England and Wales.
Commissioning bodies of the NHS are required to observe its
recommendations. Guidance can be used by the NHS, local
authorities, employers, voluntary groups, and anyone else
involved in delivering care or promoting well-being.
NICE recommendations are respected elsewhere in the
UK, but are not mandatory; in Scotland NICE recommen-
dations are published after further review by NHS Quality
Improvement Scotland (for health services issues) and by NHS
Health Scotland (for public health recommendations). NICE
recommendations are respected worldwide, including by the
European Commission and by national governments; NICE
International is a section of NICE established to meet non-UK
needs (e.g., evaluating rural health programs in China).
An independent committee including lay representation
advises on priorities for NICE consideration but final deci-
sions on topics referred to NICE are made by the Department
of Health. When making recommendations to the NHS on
which services (e.g., treatments) should be provided routinely,
it calculates the cost-effectiveness of treatment for each quality-
adjusted life year (QALY) of health gain purchased.
NICE publications include guidance on 374 interventional
procedures, 270 technology appraisals, 162 clinical guide-
lines, and 43 public health topics. From this latter group, some
examples include:
l Prevention of cardiovascular diseases (2010) – provides
evidence of effectiveness of population-based prevention
programs as more effective than programs aimed at high-
risk groups.
l Alcohol dependence and harmful alcohol use (2011) –
summarizes all NICE guidance; designed to inform mem-
bers of the public as well as health professionals.
l Preventing type 2 diabetes through population and com-
munity interventions (2011) – provides guidance to gov-
ernment departments, the commercial sector, health
service organizations, and non-governmental organiza-
tions on integration of public policy to prevent obesity, and
reduce diabetes prevalence and complications.
l Preventing uptake of smoking by children and young peo-
ple (2008) – document to advise local health service com-
missioners; identifies target populations, reviews campaign
messages, and provides recommendations for the mass
media and retailers.
l Promoting mental well-being at work (2009) – guidance
aimed at employers; reviews evidence in the field and rec-
ommends strategic approaches by firms, opportunities to
promote well-being and assess risk, and systems of flexible
working.
l Preventing unintentional injuries among under-15s in the
home (2010) – reviews evidence and makes recommenda-
tions to local authorities and related agencies on training
an appropriate workforce, advises government to fund cur-
ricula development, and indicates to the NHS appropriate
surveillance and treatment services.
NICE is often criticized (especially by the pharmaceuti-
cal industry) for the time taken to carry out investigations of
new treatments. It is also criticized by relatives of patients with
“glamorous” conditions (e.g., cancer) for not approving drugs
that might extend life by only 4–6 weeks, and perhaps approv-
ing instead new psychiatric therapies. The current government
has recently sought to overrule some of these NICE recommen-
dations in England. However, NICE methods and recommen-
dations are held in high repute, within the UK and beyond.
Sources: Christopher Birt FRCP FFPH, University of Liverpool, UK. Personal
communication.
National Institute for Health and Care Excellence. 2012. Available at: http://
guidance.nice.org.uk [Accessed 24 December 2012].
O’Flaherty M, Flores-Mateo G, Nnoaham K, Lloyd-Williams F, Rayner M,
Capewell S. Estimating potential cardiovascular mortality reductions with
different food policy options in the UK. Bull World Health Organ 2012;90:
522–31.
National Institute for Health and Clinical Excellence. Promoting mental
wellbeing through productive and healthy working conditions: guidance
for employers. NICE; 2009. Available at: http://www.nice.org.uk or http://
www.apho.org.uk/resource/item.aspx?RID=83868 [Accessed 18 August
2013].
Campbell B. Regulation and safe adoption of new medical devices and
procedures. Br Med Bull 2013;1–14 [Epub ahead of print]. http://dx.doi.
org/10.1093/bmb/ldt022.
BOX 15.8 The UK’s National Institute for Health and Care Excellence (NICE)
NICE as an independent body to promote “national
guidance on promoting good health and preventing and
treating ill-health”. NICE produces guidance in three
areas:
l public health – guidance for those working in the NHS,
local authorities and the wider public and voluntary sector
on promotion of good health and the prevention of disease
l health technologies – guidance on use of new and
existing medicines, treatments and procedures within
the NHS
l clinical practice – guidance on appropriate treatment
and care within the NHS of people with specific dis-
eases and conditions.
NICE guidelines are recommended practices with the
objective of reducing ineffective practices. During 2007,
guidelines were issued on topics including asthma, derma-
titis, caesarean section, chronic obstructive lung disease,
depression (in children and adults), eating disorders, fertil-
ity, contraception, multiple sclerosis, post-traumatic stress
disorder, and diabetic foot care (Box 15.8).
http://guidance.nice.org.uk
http://guidance.nice.org.uk
http://www.nice.org.uk
http://www.apho.org.uk/resource/item.aspx?RID=83868
http://www.apho.org.uk/resource/item.aspx?RID=83868
http://dx.doi.org/10.1093/bmb/ldt022
http://dx.doi.org/10.1093/bmb/ldt022
800
BOX 15.9 Organizations to Promote Quality in Health,
USA
l National Committee for Quality Assurance (NCQA) –
This non-profit organization, founded in 1979 by the
managed care industry, conducts surveys among man-
aged care plans to evaluate clinical standards, mem-
bers’ rights, and health service performance. It accredits
over 550 managed care plans in the USA, and in 2007
published rankings of the “best” health plans. Website:
Health Care Accreditation, Health Plan Accreditation Organization – NCQA
l Agency for Healthcare Research and Quality (AHRQ) –
This is part of the US Public Health Service. Founded in
1995, it was mandated to develop an evidence-based
practice program in 12 centers in the USA. It conducts
systematic reviews of the literature and publishes analy-
ses and findings of these reviews. Website: http://www.
ahrq.gov/
l Centers for Medicare & Medicaid Services (CMS) 2001 –
The CMS, previously the Health Care Financing
Administration (HCFA, 1977), is the federal agency of
the Department of Health and Human Services, respon-
sible for administering the Medicare and Medicaid and
the State Children’s Health Insurance Program (CHIP)
health plans. Its roles include quality assurance, the
requirements for managed care organizations, and qual-
ity improvement. Website: http://www.cms.gov/
l Institute for Healthcare Improvement (IHI) – Non profit
organization founded in 1991 as a global resource for
health care improvement knowledge to improve health
care by fostering collaboration among health care orga-
nizations. IHI examines office practices of physicians,
educational reform, and promotes interdisciplinary team
work in quality improvement. Website: http://www.ihi.
org/ihi/
l National Patient Safety Foundation (NPSF) – Sponsored
by the American Medical Association as a response to
findings of high rates of injury and death from iatrogenic
disease in the USA, the NPSF promotes research into
human error among health care providers, seeking ways
to reduce the frequency and effects of medical error,
such as misdiagnosis, medication errors, and mistakes
during procedures. Website: http://www.npsf.org/au/
l Joint Commission on Accreditation of Healthcare
Organizations (JCAHO) – Originating in 1917 by the
American College of Surgeons, it began accrediting
hospitals in 1918. It developed in 1953 as the JCAHO,
becoming a national voluntary accreditation organiza-
tion focusing mainly on hospitals. Its mandate was broad-
ened in 1987 and, as of 2007, had accredited more than
15,000 health care organizations. Accreditation is man-
datory for Medicare and Medicaid payment. The JCAHO
is changing its approach from standards-based assess-
ment every 3 years to one of reviewing performance data
quarterly as a continuous surveillance activity for risk
reduction. Website: http://www.jointcommission.org/
Source: Websites accessed 12 September 2012.
The New Public Health
The USA has a number of government and independent
organizations dedicated to improving quality in health care
systems. The CDC and the Institute of Medicine of the US
National Academies of Science play active roles in promot-
ing research quality and methods of CQI in the US health
care system. Canada is also very active in this regard, hav-
ing national and provincial institutes for the evaluation of
clinical effectiveness and clinical guidelines, and so too are
European countries (Box 15.9).
PUBLIC HEALTH LAW
Public health workers need knowledge of government
structure and public health legislation as basic to their pro-
fessional work to understand their responsibilities, powers
and liabilities. Law consists of a system of rules, regula-
tions, and orders that govern the behavior of individuals
and of society. Law represents the consensus of a society, as
enacted by an elected legislature, put into effect by the exec-
utive branch of government, and interpreted by the courts
as need be from time to time. The legislative and execu-
tive branches are separate under the US Constitution, but
the two are united in the parliamentary system (Box 15.10).
The authority, responsibility, and power to provide for and
protect the public health are basic functions of a sovereign
government, which may be delegated to another level of
government (higher or lower) or even a non-governmental
agency. The constitution of a sovereign government states
explicitly or implicitly that responsibility, but accepted
practice and court decisions (i.e., the common law) define
the powers of the national, state, or local government to
monitor and protect the health of its citizens.
In the USA, national legislation is enacted under the
powers of the federal government, namely to regulate inter-
state commerce and the power to tax and spend for the
general welfare. State legislation is enacted under the basic
power of the state to protect the health, welfare, and safety
of its citizens. Under these federal and state powers, a wide
range of health legislation and regulations is enacted affect-
ing public health, labor, and occupational health and safety,
environmental controls, public welfare, and the financing
of health services, agriculture, food, drugs, cosmetics, and
medical devices. Public health law relies on a wide range of
constitutional, statutory, administrative, and judicial deci-
sions in both civil and criminal actions. Appropriation of
funds is a legal act of legislative bodies to achieve objec-
tives directly or indirectly by financial incentives.
Categorical programs may be directed to specific issues
such as combating TB and promoting immunization or for
work to combat NCDs such as diabetes, or in improving
standards of facilities, and in providing health care services.
The regulatory, enforcement, policing, and punitive func-
tions of public health laws have evolved over many decades
and in many countries lack clear definition. In the USA,
Health Care Accreditation, Health Plan Accreditation Organization – NCQA
http://www.ahrq.gov/
http://www.ahrq.gov/
http://www.cms.gov/
http://www.ihi.org/ihi/
http://www.ihi.org/ihi/
http://www.npsf.org/au/
http://www.jointcommission.org/
801Chapter 15 Health Technology, Quality, Law, and Ethics
In federal nations, political authority is divided between two
autonomous sets of governments, one national and the other
subnational. Both operate directly with the people in their juris-
diction based on a constitutional division of power between
the national government, which exercises authority over the
whole national territory, and state or provincial governments
with independent authority within their own territories. The
constitution is the supreme law of a country. It sets out the
divisions of governmental powers including statutory authority,
administrative, natural resources, and taxation between federal
and state levels of government.
A federal legislature or congress makes the law of the land,
but is subject to rulings of a Supreme Court as are state and
local governments. State or provincial governments in a fed-
eral system have functions set out in the Constitution. They
also have elected legislatures, and executive branches with
taxing, regulatory, and punitive powers. Local governments
for county, municipal, or city governments also have dele-
gated taxing and regulatory powers including those of public
health.
Canada, the USA, Brazil, Australia, India, and Argentina
are organized on a federal basis. Federal countries also include
Austria, Germany, Malaysia, Mexico, Nigeria, Switzerland, and
Venezuela. Russia is called a federation. Usually there is some
overlapping or shared powers between national and state con-
stitutions, legislatures, and court systems, and public agencies,
taxing powers and regulatory functions, such as in interstate
commerce and emergency response to natural or other disasters.
In a unitary government system, most or all of the governing
power resides in a centralized government. This contrasts with
a federal system. In unitary systems the central government
commonly delegates authority to subnational units and chan-
nels policy decisions down to them for implementation.
A majority of nation-states are unitary systems. They vary greatly.
The UK includes England, Scotland, Wales, and Northern
Ireland, each with legislatures, but the Westminster Parliament
in London maintains national powers. In health, each of the four
member entities of the UK conducts a National Health Service
with autonomy but common features. The national government
may delegate certain powers to self-governing regions/local
authorities, and there is a growing tendency to devolve various
governmental functions such as health to regional authorities.
More than 150 countries are unitary states, including France,
Italy, Spain, China, and Japan.
In both forms of government, local authorities are estab-
lished under state law with governance by councils elected
by the people, with taxing and regulatory powers within the
state or provincial laws, with a high degree of autonomy but
within state regulation, standards, and financial support. Local
authorities have major responsibilities in public health such as
in sanitation, licensing, and regulation of businesses and zon-
ing, as well as many other areas, including social welfare.
Note: See also Chapter 10.
Source: Differences between federal and unitary forms of government.
Available at: http://www.preservearticles.com/201107139054/difference-
between-unitary-and-federal-forms-of-government.html [Accessed 15
December 2012].
Encyclopedia Britannica. Unitary government. Available at: http://www.
britannica.com/EBchecked/topic/615371/unitary-system [Accessed 15
December 2012].
Encyclopedia Britannica. Political systems. Available at: http://www.britan-
nica.com/EBchecked/topic/467746/political-system/36704/Federal-systems
[Accessed 15 December 2102].
BOX 15.10 Legal Structure of Federal and Unitary Countries
efforts are being made to update and reform laws in the pub-
lic health sector. In 1988, the Institute of Medicine (IOM) in
the USA (the Future of Public Health) called for codifica-
tion of public health law as essential for the public good,
while questioning the soundness of certain US public health
laws. More recently, the Model State Emergency Health
Powers Act in the USA, the Quarantine Act in Canada, and
the revised International Health Regulations (2007) have
sought to update century-old legislation. The revised inter-
national regulations provide for a global approach to con-
trol the spread of epidemics and public health emergencies
while minimizing disruption to international activities such
as travel, trade, and economics.
A combination of the regulatory, persuasive, and fund-
ing approaches is widely used in public health in control of
communicable and non-communicable diseases, in improv-
ing standards of facilities, and in providing health services.
The regulatory, enforcement, policing, and punitive func-
tions of public health are important in health promotion and
assurance of health care. The taxing power of government is
essential for public health to ensure that adequate facilities
and access to care are available to all members of the com-
munity, especially those in financial need and thus at greater
risk for disease.
Medical officers of health and their staff have legal
authority to issue formal orders for health protection of
the public. Situations which require court proceedings are
referred to the justice system. Situations that may require
enforcement by court proceedings are referred to the justice
system. Laws may be enacted to fund public health activi-
ties, whether provided by public health authorities or by
acting through official or non-official agencies or providers.
Public health authorities, namely medical officers of health,
have the legislative power to issue orders to individuals or
businesses where there is a threat to the health of the public
such as food establishments. Administrative resources are
needed to enforce laws, such as through the FDA and the
Environmental Protection Agency, which come under the
aegis of the Department of Health and Human Services.
Other departments such as Agriculture, Education, or inter-
departmental agencies (e.g., Homeland Security), also are
key to public health activities, such as in disaster situations.
http://www.preservearticles.com/201107139054/difference-between-unitary-and-federal-forms-of-government.html
http://www.preservearticles.com/201107139054/difference-between-unitary-and-federal-forms-of-government.html
http://www.britannica.com/EBchecked/topic/615371/unitary-system
http://www.britannica.com/EBchecked/topic/615371/unitary-system
http://www.britannica.com/EBchecked/topic/467746/political-system/36704/Federal-systems
http://www.britannica.com/EBchecked/topic/467746/political-system/36704/Federal-systems
802
Other intergovernmental activities may require special leg-
islation to empower, finance, and promote their cooperation,
such as in the case of establishing an authority to manage
long-term efforts to clean up a contaminated river or basin,
which involves the cooperation and coordination of many
local authorities.
Health protection of individuals and communities may
require legal action to detain a person in order to prevent
the spread of a reportable communicable disease, to protect
a mentally ill patient, or to restrain a violent person. Such
powers should be used as a last resort if voluntary compli-
ance and education fail, and where the danger to the commu-
nity or the individual is sufficient to convince a court of the
public need to override the personal liberty of an individual.
An example is a 2007 case of a person with MDR-TB who
was taken into custody on arrival for compulsory treatment
after traveling across the Atlantic Ocean on a commercial
airline, against the specific instructions of his physician, thus
endangering fellow passengers. Outbreaks of measles in the
UK (2006–2007) and in Israel via imported cases among
ultraorthodox Jews or conservative protestant groups in the
Netherlands, with transmission among religious people who
tend not to immunize their children, led to pressure by health
authorities to immunize those placed at risk by such contacts
at weddings or other large public events.
However, these measures are currently used less than
voluntary isolation or quarantine and placarding homes for
reportable infectious diseases such as measles. Powers are
essential in extreme cases where refusal to comply with
public health measures endangers others. Such powers
should have been used more vigorously in the early years
of the AIDS epidemic at a time when individual rights took
precedence over protection of the population, including
vulnerable high-risk groups. The severe acute respiratory
syndrome (SARS) epidemic of 2003 led to sequestering
hospital staff in Toronto, Canada, for lengthy periods to
prevent spread of the disease, and subsequent influenza
pandemic threats have raised questions as to whether hos-
pital personnel should be required to be immunized to pro-
tect patients and their families from onward transmission
of dangerous infections.
Recent cases in the USA, the UK, and Norway dem-
onstrate the responsibility of governments to protect the
public from incidents of violence by dangerous, mentally
disturbed individuals who carry out mass killings. In Nor-
way, 69 people, mostly teenagers, were killed by a radical
ideologue while many others sustained serious injuries; and
in the USA, Islamic terrorists at the Boston Marathon killed
three and seriously injured more than 200 others; a 20-year
old fatally shot his mother then killed 20 children and six
adult staff members at Sandy Hook elementary school in
Newtown, Connecticut, before killing himself; and an army
psychiatrist who had become an increasingly devout and
The New Public Health
radicalized Muslim psychiatrist shot and killed 13 people
and injured more than 30 others in a Texan army base.
Background checks and other restrictions on gun sales are
an important public health and political issue, especially in
the USA. The wide availability of guns, including military-
style assault weapons, presents a serious danger for impul-
sive or planned mass killings.
Public health has generally evolved with greater reli-
ance on health promotion through voluntary cooperation of
a patient or community than on compulsion. Enabling leg-
islation may permit a local authority to fluoridate its water
supply, but the enactment of local legislation and funding to
implement it may also require a public referendum. In some
states in the USA and in Israel, fluoridation of community
water supplies is mandatory, which is also part of the health
promotion approach to public health.
Appropriation of public funds to promote public health is
through approval by the legislature for a specified program.
Provision of public funds may take the form of categorical
grants for specified services, such as immunization, prenatal
care, school health, or specific disease management such as
TB control, cancer control, or AIDS education. Programs may
be designed to promote certain types and quality of services,
such as the Hill–Burton Act, which provided federal grants
for hospital construction in the 1950s to 1970s, conditioning
these grants on certain requirements concerning hospital licen-
sure and hospital planning. Such legislation has a “carrot and
stick” effect of attracting lower levels of government to seek
such funding but also requiring them to accept the conditions
and regulations that accompany the grants. The Canadian fed-
eral government’s cost sharing of provincial health (hospital
and medical) insurance programs is based on federal criteria
requiring public administration, portability between provinces,
accessibility without payment, comprehensiveness, and ban-
ning extra billing by physicians (see Chapter 13).
Public funds are also appropriated in the context of
legislated programs in which people are entitled to the ser-
vices defined in the appropriation legislation, such as in the
amendments to the Social Security Act providing Medicare
and Medicaid programs, or national health insurance legis-
lation in many countries. These and their regulations spell
out categories and specified entitlement benefits.
Legislation and court decisions to protect the rights of
the individual are part of public health. Public health law
is meant to protect individuals and communities from
potential abuse, of both individual and community human
rights, as in the US Bill of Rights. Enforcement of public
health law may infringe on individual rights by enforcing
sanitation, food and drug safety, and supervision of res-
taurants and catering firms. Laws may allow restriction of
civil rights, such as rarely used mandatory treatment of a
person with a dangerous contagious disease or mental ill-
ness. Freedom of religion may come into conflict with other
Chapter 15 Health Technology, Quality, Law, and Ethics
laws in public health where restrictive practices may deny
the use of publicly supported health facilities, as when a
religiously affiliated hospital may refuse an abortion proce-
dure in a case of rape. Religious practices or other personal
beliefs may endanger others in the community, such as in
the refusal to immunize children so that an imported infec-
tious disease may spread among non-immunized people
and even affect those who are immunized, as occurs with
imported measles cases even when domestic transmission
of the disease has previously been eradicated. General legis-
lative provisions applied to public health forbid misleading
or unethical advertising. Legislative provisions may also
ban advertising for products, such as tobacco, which are
legal but may be harmful to health. These laws affect public
health but are provisions in other statutes such as the regu-
lation of business enterprises. Legislation may also make
smoking in public places illegal, with fines for offenders
and operators of places such as public bars.
Since the 1973 US Supreme Court decision of Roe v.
Wade, the law has allowed women to seek safe and legal
abortion. This remains a highly controversial political issue
in the USA and several other countries. The potential con-
flict between community and individual interests and rights
is part of the dynamics of public health law and public health
practice. The issues involved are complex and highly politi-
cized, and often involve ethical distinctions where “the great-
est good for the greatest number” may limit the legitimate
rights of individuals and vice versa. The PPACA in 2010 is a
fundamental legislative initiative, upheld by the US Supreme
Court to become the law of the land. It will bring millions of
Americans into regulated health insurance with many protec-
tive elements to prevent abuse by private insurance company
through arbitrary exclusions or limitations.
The legal aspects of public health are vital to its opera-
tion and are increasingly complicated by ethical issues, and
by public and political debate. Health protective legislation
and regulation for sanitation of food, water, and air are fun-
damental to public health, as is the control of drugs, cosmet-
ics, vaccines, and biologicals, the manufacture of devices,
and the licensing of health personnel and facilities. Limi-
tations of legal suits (torts) against manufacturers of vac-
cines proved to be a successful measure in the USA with
the introduction of the National Vaccine Injury Compensa-
tion Program (NVICP) in 1988. This is funded by a modest
surcharge tax collected from vaccine manufacturers. It pro-
tects both public and private interests while providing a fair
compensation system to ensure patients’ rights but without
jeopardizing immunization to prevent widespread disease,
and also protects manufacturers from litigation with high
legal costs and excessive compensation awards by the jury
system. Promoting healthy behavior through the prudent
use of the legal system of regulation and taxation is increas-
ingly utilized to protect the health of the population. This is
803
widely applied in promoting road safety, in tobacco control
measures regarding banning of advertising, high taxes on
alcohol, and banning smoking in public places including
restaurants and bars.
Environmental Health
There is growing concern by the public and by governments
over climate change, global warming, air and water pollu-
tion, and other noxious and harmful industrial and com-
mercial processes. Environmental laws affecting the public
health include legislation on clean air, clean water, toxic
substances, solid waste control, and other noxious sub-
stances. Non-compliance with the legislative provisions can
result in prosecution in the civil or criminal courts or both.
Infringement of public health laws and regulations may
lead to criminal action as an increasingly common method
of sanction. While such violations may not be seen as “truly”
criminal and may be treated in the courts as misdemeanors,
they can lead to fines or even jail. Such cases are increas-
ingly being addressed seriously in the judicial system.
The CDC, in 1999, defined 10 great achievements
of public health of the twentieth century. These achieve-
ments are identified as control of infectious disease, motor
vehicle safety, fluoridation of drinking water, recognition
of tobacco use as a health hazard, immunization, decline
in deaths from coronary heart disease and stroke, safer
and healthier foods, healthier mothers and babies, family
planning, and safer workplaces (Goodman et al., 2006).
Of the 10 great achievements in the twenty-first century
(2001–2010) identified by CDC, seven of the 15 leading
causes of death (largely NCDs) resulted in a decline in the
age-adjusted death rate in the USA from 881.9 per 100,000
population in 1999 to 741.0 in 2009. This decline was a
result of a combination of supportive laws and legal tools
at the local, state, and federal levels. In other industrial-
ized countries similar legislation has led to equal or greater
achievements in public health over the past century.
Public Health Law Reform
Public health law is scattered through many legislative stat-
utes and administrative documents which developed his-
torically. Efforts to codify public health law may contribute
to greater understanding and enforceability of the many
separate pieces of legislation (Box 15.10). Such reform
will enhance understanding in the legislative, judicial, and
administrative branches of government as well as in busi-
ness, non-governmental organizations, and the community.
Box 15.11 suggests topics for model public health consoli-
dation or compendia for states. The principles of this formu-
lation may also apply to other countries at the national and
state or provincial levels.
804
ETHICAL ISSUES IN PUBLIC HEALTH
The field of public health includes a wide range of activities
and professional disciplines, ranging from health promo-
tion to disease protection, epidemiology to environmental
health, and financing to supervision or provision of clinical
care. Each of these disciplines works within systems that
face ethical dilemmas, and public health workers’ under-
standing and motivation within the ethical guidelines of
their professions and roles are important in their training
and practice conduct. Ethical frameworks have evolved in
part as the result of bitter experience with ethical failures
which were later recognized and affect public health stan-
dards of practice for future generations (Box 15.12).
Ethics in health are based on the fundamental religious
and humanistic values and concepts of a society. If the prin-
ciple of saving a life is valued above all other considerations
(i.e., Sanctity of Life or Pikuah Nefesh) (see Chapter 1),
then all measures available are to be used, irrespective of
the condition of the patient or the cost. If sickness and death
are seen as acts of God, possibly as punishment for sin, then
prevention and treatment may be considered to be interfer-
ing with the divine will, and the ethical obligation may be
limited to relief of suffering. Humanism balances these two
ethical imperatives: saving of life and relief of suffering.
Materialistic political philosophies may view health care as
primarily a function to preserve health for economic pros-
perity and social well-being. Secular humanism adopted
many of the religious precepts of the worth and rights of the
individual and these have become part of the standards of
law and ethics in modern secular societies.
The role of society in protecting the health of the popula-
tion grew during the nineteenth century with the sanitation
BOX 15.11 Public Health Law Program of the Centers
for Disease Control and Prevention
The Public Health Law Program (PHLP) is administered by
CDC’s Office for State, Tribal, Local and Territorial Support
(OSTLTS). PHLP develops law-related tools and provides
legal technical assistance to public health practitioners and
policy makers in state, tribal, local, and territorial (STLT)
jurisdictions.
The PHLP works with state/territorial health departments
and other partners to:
l identify public health law priorities
l research laws that impact the public’s health
l analyze public health legal preparedness
l conduct comparative analyses across jurisdictions;
prepare guidance, articles, reports, and toolkits; and
develop and disseminate public health law curricula.
Source: Centers for Disease Control and Prevention. Public health law.
Available at: http://www.cdc.gov/phlp/about.htm [Accessed 27 October
2012].
The New Public Health
movement, while medical care became an effective part of
public health during the twentieth century. The astonishing
successes of public health during the past century increased
life expectancy in the high-income countries by some 30
years, mostly through improved living conditions and
health protection, as well as societal and medical advances
to make care available to all. In the 1970s the Lalonde con-
cept that individual behavior was one of the key determi-
nants of health (see Chapter 2) placed much of the onus
of illness and its prevention on the individual, but fostered
health promotion as an essential component of public health
theory and practice. All these points of view are involved
in the ethical issues of the New Public Health (Box 15.13).
Resources for health care are limited even in indus-
trialized countries, so that priority setting and judicious
allocation of scarce resources are always issues. Money
spent on new technology with only marginal medical
advantages is often at the expense of well-tried and proven
lower cost techniques to prevent or treat disease. The
potential benefits gained by the patient from more and
more interventions are sometimes very limited in terms
of length or quality of life. These are difficult issues when
the physician’s commitment to do all to preserve the life
of the patient conflicts with the patient’s concept of qual-
ity of life and his or her right to decline or terminate heroic
measures of intervention. Many health systems use clini-
cal guidelines that are mandatory for a health facility or a
doctor in the clinic. Preparation for surgery requires a sig-
nature from the patient to consent to the procedure being
carried out, careful preoperative procedures to ensure that
the correct organ is addressed, antiseptic preparation of
the site, and checking that all instruments are accounted
BOX 15.12 Topics for a Model State Public Health Act
l Mission and functions
l Public health infrastructure
l Collaboration and relationships
l Public health authorities and powers
l Public health emergencies
l Public health information privacy
l Criminal/civil
l Enforcement
l Legislative response to the need to reform core public
health powers such as surveillance, reporting, epide-
miological investigations, partner notification, testing,
screening, quarantine, isolation, vaccination, and nui-
sance abatement
l Medicaid
l Affordable Care Act insurance agency
Source: Centers for Law and the Public’s Health. A Collaborative at Johns
Hopkins and Georgetown Universities (CDC Collaborating Center).
Available at: http://www.publichealthlaw.net/ [Accessed 23 September
2012], and Chapter 10 references.
http://www.cdc.gov/phlp/about.htm
Chapter 15 Health Technology, Quality, Law, and Ethics
for. The checklist approach is well established for care
in many settings and protects the patient from neglect or
faulty follow-up, such as in the management of hyperten-
sion and diabetes.
The suffering that a terminally ill patient may endure
during radical treatment, which may prolong life by only
hours or days, clashes with the physician’s ethical obli-
gation to do no harm to the patient. The ethical value
of sustaining the life of a terminally ill patient suffering
extensively is an increasing medical dilemma. The issue is
even more complex when economic values are included in
the equation. There are potential conflicts among the eco-
nomic issues, the role of the physician in preserving life,
the physician’s obligation to do no harm, the felt needs
of the patient and his or her family, and the needs of the
community as a whole. The complex issues involved in
the “right to die” and end-of-life care raise many ethical
and legal questions for the patient, the family, society, and
caregivers.
The state represents organized society and has, among
its responsibilities, a duty to promote healthful conditions
and to provide access to health care and public health
BOX 15.13 Study and Practice of Public Health Ethics
Ethics is a branch of philosophy that deals with distinctions
between right and wrong, with the moral consequences of
human actions. The ethical principles that arise in epidemio-
logical practice and research include:
l informed consent
l confidentiality
l respect for human rights
l scientific integrity.
“As a field of study, public health ethics seeks to understand
and clarify principles and values which guide public health
actions. Principles and values provide a framework for decision
making and a means of justifying decisions. Because public
health actions are often undertaken by governments and are
directed at the population level, the principles and values
which guide public health can differ from those which guide
actions in biology and clinical medicine (bioethics and medical
ethics) which are more patient or individual-centered.
As a field of practice, public health ethics is the application
of relevant principles and values to public health decision mak-
ing. Public health ethics inquiry carries out three core functions:
(1) identifying and clarifying the ethical dilemma posed,
(2) analyzing it in terms of alternative courses of action and their
consequences, and
(3) resolving the dilemma by deciding which course of action
best incorporates and balances the guiding principles and
values.” (CDC, 2001)
Sources: Last JM, editor. A dictionary of epidemiology. 4th ed. New York:
Oxford University Press; 2001.
Centers for Disease Control and Prevention. Science coordination and
innovation. Public health ethics; 2001. Available at: http://www.cdc.gov/
od/science/phec/ [Accessed 23 September 2012].
805
services. The conflict between individual rights and com-
munity needs is a continuous issue in public health. Appli-
cation of accepted public health measures for the benefit of
some people in society may require applying an interven-
tion to everyone in a community or a nation. The major-
ity thus are subject to a public health activity to protect a
minority, without designating which individual’s life may
be saved. Furthermore, a society may in special cases need
to restrict individual liberties to achieve the goal of reducing
disease or injury in the population. Raising taxes on alco-
hol and tobacco products, mandatory speed limits, driving
regulations, and seat belt usage laws are examples of pub-
lic health interventions that interfere with individual liberty
but protect individuals, and thereby the community at large,
from potential harm.
Many public health measures originally criticized as
interventions in private rights are generally accepted as
essential for health protection and promotion to reduce
the risk of disease in the population. Chlorination of com-
munity water supplies is a well-established, effective, and
safe intervention to protect the public health. Fluoridation
of drinking water to prevent tooth decay in children means
that other people are also drinking the same fluoridated
water, which is of less direct benefit to them. Fortification
of foods with vitamins and minerals is also a cost-effective
community health measure with advocates and opponents.
The addition of folic acid to food as the most effective way
to prevent neural tube defects in newborns is an intervention
mandated by the US FDA since 1998.
Confidentiality to assure the right of the individual to
privacy involves ethical issues in the use of health infor-
mation systems. Birth, death, reportable conditions (not
all reportable diseases are infectious), and hospitalization
data are basic tools of epidemiology and health manage-
ment. The use of detailed individual data is needed for
case-finding and follow-up activities which are vital to
good epidemiological management of diseases, includ-
ing STIs. However, caution is needed in data use to avoid
individual identification that could be used punitively,
for example, in denial of access to health insurance for
smokers, alcoholics, or AIDS patients because health
damage may be attributable to a self-inflicted risk factor.
Increasingly, however, reporting is also mandatory for
physical or sexual abuse and criminally linked injuries
as essential for the protection of individuals at risk or the
general public from serious harm.
Individual and Community Rights
The protection of the individual’s rights to privacy, and free-
dom from arbitrary and harmful medical treatments, proce-
dures, or experiments, may come up against the rights of the
community to protect itself against harmful health issues.
This conflict comes into much of what is done in public
http://www.cdc.gov/od/science/phec/
http://www.cdc.gov/od/science/phec/
806
health practice, which has both an enforcement basis in law
and practice and a humanitarian and protective aspect based
on education, persuasion, and incentives. Society permits
its governments to act for the common good, but sets limits
that are protected by the courts and administrative appeal
mechanisms.
Society has the right to legislate the side of the road
on which one is permitted to drive, the speed permitted,
the wearing of seat belts, and the non-use of alcohol or
drugs before driving or cell phones while driving. Offend-
ers may be punished by significant fines or jail and are
subject to strong educational efforts to persuade them to
comply. Similarly, the community must ensure sanitary
conditions and prevent hazards or nuisances from bother-
ing neighbors or the public. Society must act to protect the
environment against unlawful contamination or poisoning
of food, drugs, the atmosphere, the water supply, or the
ground.
Enforcement is thus a legitimate and necessary activ-
ity of the public health network to protect the community
from harm and danger to health. Table 15.6 shows topics
where individual rights and responsibilities predominate,
and a second set of rights that are the prerogative of the
community to protect its citizens against public health haz-
ards. Sometimes the issues overlap and sometimes come to
political, advocacy, or legal action, so that court decisions
are needed to adjudicate precedents for the future.
The AIDS epidemic in the 1980s and 1990s raised a
host of public health, ethical, and issues. Management of
the AIDS epidemic is in some respects in conflict with the
long-established role of society in contacting and quarantin-
ing people suffering from transmissible diseases. It is not
acceptable or feasible in modern society to isolate HIV car-
riers. But failure or delay of public health authorities even
in the late 1980s to close public bathhouses in New York
and other cities in the USA, where exposure to multiple
same-sex partners promoted transmission of the infection,
could be interpreted as negligence. During the 1980s, the
gay community in the USA centered its concern that HIV
testing would be used in a discriminatory manner. AIDS was
initially addressed as a civil liberties issue and not as a pub-
lic health problem. Screening, reporting, and case contact
follow-up were seen as an invasion of privacy and proved
counterproductive by increasing resistance to and avoidance
of testing. Protection of privacy and an educational approach
were adopted as most feasible and acceptable. International
opinion and national court decisions have emphasized the
right to privacy with decriminalization of non disclosure of
HIV status to sex partners (UNAIDS 2013).
The AIDS epidemic and public anxiety about contract-
ing AIDS through casual contact reinforced the need for
public education on safe sex. This has been raised as an
ethical issue because such education may be construed as
condoning teenage and extramarital relations. The issue of
The New Public Health
HIV screening of pregnant women in general or in high-risk
groups took on a new significance with the findings that
treatment of the pregnant woman reduces the risk of HIV
infection of the newborn, and that breastfeeding may be
contraindicated. This issue is arising anew in the context of
using the HPV vaccine for preteen girls to prevent the sexu-
ally transmitted infection, which is also controversial, and in
the USA this vaccination will be mandatory for school entry.
A pre-eminent ethical issue in public health is that of
assuring universal access to services, and/or the provision
of services according to need. An important ethical, politi-
cal, and social issue in the USA in the twenty-first century
is how to achieve universal access to health care. The soli-
darity principle of socially shared responsibility for fund-
ing universal access to health care is based on equitable
prepayment for health care for all by nationally regulated
mechanisms through place of work or general revenues of
government. A society may see universal access to health
care as a positive value, and at the same time utilize incen-
tives to promote the use of services of benefit to the indi-
vidual, such as hospital care, immunization, and screening
programs. Some services may be arbitrarily excluded from
health insurance, such as dental care, although this is to the
detriment of children and a financial hardship for many.
Strategies for program inclusion are often based on his-
torical precedent rather than cost-effectiveness or evidence.
While efforts are being made to include more children in the
program, the Medicaid system in the USA defines eligibil-
ity at income levels of 185 percent of the poverty line, thus
excluding a high percentage of the working poor. Health
is also a political issue in countries with universal health
systems where funding may be inadequate or patient dis-
satisfaction common.
Choices in health policy are often between one “good”
and another. Limitations in resources may make this issue
even more difficult in the future, with aging populations,
increasing population prevalence of physical disabilities,
and rapid increases in technology and its associated costs.
For example, the UK’s NHS at one point refused to provide
dialysis to people over the age of 65. When computed tomog-
raphy was first introduced, Medicare in the USA refused to
insure this service as an untested medical technique. Owing
to a lack of facility resources such as incubators and poor
prospects for the survivors, the Soviet health system con-
sidered newborns as living only if they weighed over 1000 g
and survived for more than 7 days. Those under 1000 g, who
would be considered living by other international definitions,
would be placed in a freezer to die. At the opposite extreme,
many western medical centers use extreme and costly mea-
sures to prolong life in terminally ill patients, preserving life
temporarily but often with much suffering for the person and
at great expense to the public system of financing health care.
In many countries, such as those in the former Soviet
system of health care, spending for hospital services, in
807Chapter 15 Health Technology, Quality, Law, and Ethics
TABLE 15.6 Individual and Community Rights and Responsibility in Health: Ethical/Legal Issues
Ethical/Legal Issues Individual Rights and Responsibilities Community Rights and Responsibilities
Sanctity of human life Right to health care; responsibility for
self-care and risk reduction
Responsible for providing feasible basket
of services, equitable access for all
Individual vs community
rights
Immunization for individual protection Immunization for herd immunity and community
protection; education; community may mandate
immunization
Right to health care All are entitled to needed emergency,
preventive, and curative care
Community right to care regardless of location,
age, gender, ethnicity, medical condition, and
economic status
Personal responsibility Individual responsible for health behavior,
diet, exercise, and non-smoking
Community education to health-promoting
lifestyles; avoid “blame the victim”
Corporate responsibility Management accountability to criminal
and civil action
Producer, purveyor of health hazard accountable
for individual and community damage
Provider responsibility Professional, ethical care and communication
with patient
Access to well-organized health care,
accredited to accepted standards
Personal safety Protection from individual, family, and
community violence
Public safety, law enforcement, protection of
women, children, and elderly; safety from
terrorism
Freedom of choice Choice of health provider; limitations of
gatekeeper functions; control costs while
function; right to second opinion; right of appeal
Confidentiality; informed consent; birth control
ensuring individual rights; limitations of
self-referrals to specialist
Euthanasia Individual’s right to die; limitations by
societal, ethical, and legal standards
Assure individual and community interests;
prevention of abuse by family or others with
conflict of interests
Confidentiality Individual’s right to privacy, limitation of
information
Mandatory reporting of specified diseases;
data for epidemiological analysis
Informed consent Right to know, risks vs benefits; agree or
disagree to treatment or participation i
n experiment
Helsinki Committee approval of research;
regulate fair practice in right to know; Patient’s
Bill of Rights
Birth control Right to information and access to birth
control and fertility treatment; woman’s
rights over her body
Political, religious promotion of fertility;
alternatives to abortion; protection of women’s
rights to choose
Access to health care Universal access, prepayment; individual
contribution through workplace or taxes
Solidarity principle and adequate funding; right
to cost containment, limitations on service
benefits
Regulation and incentives
to promote preventive care
Social security for hospital delivery,
attendance for prenatal care; primary care,
ambulatory care; home care
Incentive grants to assist communities for
programs of national interest; limit institutional
facilities
Global health Human rights and aspirations; economic
development, health, education, and jobs
Transfer of health risks; occupational hazards
and environmental damage
Rights of minorities Equality in universal access Special support for high-needs groups
Prisoners’ health Human rights Security and human rights; reduce inequalities in
sentencing convicts, harsh dangerous conditions
in prisons; prohibition of torture and execution
Allocation of resources Lobbying, advocacy for equity and innovation Equitable distribution of resources; targeting
high-risk groups; cost containment
some cases grossly in excess of need, is accompanied by
a lack of adequate funds for primary care or adding new
vaccines to the immunization program for children. The
majority of Americans have health insurance which increas-
ingly includes preventive care services, but a substantial
percentage lack such coverage which limits their access to
routine preventive care. The Affordable Care Act brings an
improvement in coverage and inclusion of preventive care
with incentives (see Chapters 10 and 13). In many coun-
tries, including in Europe, delay in updating immunization
808
programs may be due to a lack of funding or to delays
in professional or governmental acceptance of “new”
vaccines.
The closure or amalgamation of hospitals involves dif-
ficult decisions and is a source of friction between central
health authorities, the medical professions, and local com-
munities. Health reforms in many industrialized countries,
such as reducing hospital bed supplies and managed care
systems promoting cost containment and reallocation of
resources, raise ethical and political issues often based on
vested interests such as private insurance systems, hospi-
tals, and private medical practitioners.
Where there is a high level of cumulative evidence from
the professional literature and from public health practice
in “leading countries” with a strong scientific base and case
for action on a public health issue, when does it become bad
practice or even unethical public health practice to ignore and
fail to implement such an intervention? Such ethical failures
occur frequently and widely. For example, is it “unethical”
not to fortify grain products with folic acid, and salt with
iodine? Should there be a recommended European immu-
nization program; should milk be fortified with vitamin D;
should vitamin and mineral supplements be given to women
and children; should all newborns be given intramuscular
vitamin K routinely? Other examples include the issues of
fluoridation of water supplies and opposition to genetically
modified crops or generic drugs in African countries. These
issues are continuously debated and the responsibility of
the trained public health professional is to review the inter-
national literature on a topic and formulate a position based
on the cumulative weight of evidence. It is not possible to
wait for indisputable evidence because in epidemiology and
public health this rarely occurs. This is another reason for
guidelines established by respected agencies and profes-
sional bodies, which are free from financial obligations to
vested interest groups, being essential for review of the evi-
dence which continues to accumulate on many issues thought
to have been resolved or which reappear repeatedly despite
strong evidence of effectiveness and public health benefit.
Tragic Deviations in Public Health Ethics
In the nineteenth century the germ and miasma theories both
produced enormous gains in public health. The biomedical
paradigm addressed alleviation of disease risk or manifest
disease; the health paradigm addressed the improvement of
social and environmental conditions for reducing disease.
During the early part of the twentieth century, a segment of
the social hygiene movement promoted ideas of Social Dar-
winism or racial improvement by sterilization of mentally
ill, retarded, and other “undesirable” people.
The dominant biomedical model of public health and
medical professionals adopted policies of eugenics in
Sweden, the USA, and Canada, leading to policies and
The New Public Health
programs to force the sterilization of mentally handi-
capped or mentally ill patients. This distorted a socially
oriented concept of public health. This euthenasia policy
was adapted to a racially oriented policy with horrendous
policies of mass murder in the name of racial purity as a
public health policy in Nazi Germany with the near-total
support and participation of a highly Nazified medical
profession, and used in murder, by gassing or planned
starvation, of half a million “undesirables” under the
eugenics “T-4” program administered from Hitler’s head-
quarters. Although this program was stopped after parental
and Church protests in Germany, the methods used were
adopted in newly occupied countries and for concentra-
tion camps organized for the mass extermination of Jews,
Gypsies, and others in the Holocaust.
The eminent historian Sir Richard Evans (Regius Pro-
fessor of History at Cambridge University), in his classic
The Third Reich at War, wrote:
“At the heart of German history in the war years lies the mass
murder of millions of Jews in what the Nazis called ‘the final
solution to the Jewish question in Europe’. This book provides a full
narrative of the development and implementation of this policy of
genocide, while also setting it in the broader context of Nazi racial
policies toward the Slavs, and toward Gypsies, homosexuals, petty
criminals and ‘asocials’. … For many years, and not merely since
1933, the medical profession, particularly in the field of psychiatry,
had been convinced that it was legitimate to identify a minority of
handicapped as ‘a life unworthy of life’, and that it was necessary
to remove them from the chain of heredity if all the many measures
to improve the German race under the Third Reich were not to be
frustrated. Virtually the entire medical profession has been actively
involved in the sterilization programme, and from here it was but a
short step in the minds of man to involuntary euthanasia.”
The twentieth century was replete with mass murders,
executions, and genocide, with nationalistic, ideological,
and racist motives perpetrated by fascist, Stalinist, and
radical xenophobic political or religious movements when
gaining governmental power by election or by revolution,
in some cases applying common public health terminol-
ogy and concepts to uses of genocide and ethnic cleansing
(Box 15.14).
An outline of genocides of the past 100 years is seen
in Box 15.15. These include the Turkish genocide of the
Armenians in 1917 followed by horrific genocides in which
many millions of people were killed, carried out under the
communist regime of the Soviet USSR in the 1920s and
subsequently, in the People’s Republic of China under
Chairman Mao in the 1950s, and by the Khmer Rouge in
Cambodia in the 1980s, and in the wars resulting from the
breakup of the Yugoslav Republic in the 1990s.
The human and national cost of genocide lasts for gen-
erations. The hatred and fear may wane but the trauma goes
deep. It lasts with the victims and their descendants, but
Chapter 15 Health Technology, Quality, Law, and Ethics
also with the perpetrating country and its culture. The Nazi
Holocaust has had downstream effects in public health in
the German-speaking countries which last to the present
time, seven decades since the events took place. The long-
term damage done to public health in Germany and Austria
is described in Box 15.16.
The Nuremberg Doctors’ Trial in 1946–47 convicted
many leading Nazi physicians of crimes against humanity
and resulted in severe punishments including hanging or
long prison terms. This trial was a seminal event in estab-
lishing the ethical standards required for medical research
and human rights. However, many in the medical profession
aligned with these horrors remained leading figures in the
BOX 15.14 Values and Ethical Principles of Public
Health
l Sanctity of human life.
l Individual human rights – liberty, privacy, protection
from harm.
l Solidarity – sharing the burden of promoting and main-
taining health.
l Beneficence – reduce harm and burdens of disease and
suffering.
l Non-malfeasance – do no harm.
l Proportionality – restriction on civil liberties must be
legal, legitimate, necessary, and use the least restrictive
means available.
l Reciprocity principle – public responsibility to those who
face disproportionate health and social burden.
l Transparency principle – honest and truthfulness in the
manner and context in which decisions are made must
be clear and accountable.
l Precautionary principle – decision makers have a general
duty to take preventive action to avoid harm even before
scientific certainty has been established.
l Failure to act – public health officials and policy mak-
ers have a duty to act and implement preventive health
measures demonstrated to be effective, safe, and benefi-
cial to population health. Failure to enforce public health
regulations with resulting disease or deaths may consti-
tute negligence on the part of responsible officials with
civil or criminal penalties.
l Equity – reduce inequities.
l Cost and benefits
l Stewardship – responsibility of governance in a trustwor-
thy and ethical manner.
l Trust between the many stakeholders in health.
l Reasonableness – decisions should be evidence based
and revised based on new evidence.
l Responsive to needs and challenges as they may be
anticipated and appear with close monitoring of health
status.
Source: Modified from Lee LM. Guest editorial: Public health ethics
theory: review and path to convergence. Public Health Rev 2012;34(1).
Available at www.publichealthreviews.eu [Accessed 17 December 2012].
809
German medical community, one even being elected to head
the World Medical Association, then discussing the Helsinki
Declaration of Ethics in Biomedical Research, before being
forced to resign. The Nuremberg Trials and the subsequent
Helsinki Declaration laid the fundamentals of biomedical
ethics for the following generations, regulated by require-
ments of ethical procedures and institutional research board
approvals for funding, conducting, and publishing research
involving human subjects (Table 15.7).
The United Nations Convention on Prevention and Pun-
ishment of the Crime of Genocide (UNGC) of 1948 defines
acts committed with intent to destroy, in whole or in part,
members of a national, ethnical, racial, or religious group
as crimes against humanity. This convention specifies that
incitement to genocide is itself a crime against humanity.
Legal action should focus on state-sanctioned incitement
as a recognized early warning sign. The UNGC defines
genocidal acts to include the following as punishable under
international law:
l genocide
l conspiracy to commit genocide
l direct and public incitement to commit genocide
l attempt to commit genocide
l complicity in genocide.
The reappearance of genocide in the late twentieth cen-
tury in the Balkans and Rwanda, and in the twenty-first cen-
tury by Sudanese in Darfur, highlights genocide as a public
health concern and its prevention as a public health and
international political responsibility. Incitement to genocide
is a crime against humanity and was the basis for the trials
and convictions of leaders of the Rwandan Tutsi tribe, as
well as inciters to ethnic violence and the political leaders
and perpetrators of mass murders in the former Yugoslav
Republic. The threat and practice of genocide are still pres-
ent, whether in the murderous raids of Sudanese Janjaweed
militias in Darfur and South Sudan, the threats of genocide
by Iran and associated terrorist organizations against Israel
and Jews in general, or the killing of Christians in northern
Nigeria and Egypt, of Muslims in Burma/Myanmar, and
others. Incitement to genocide is now common as part of
international discourse.
Genocide represents the most extreme assault on the right
to life and respect for life. In the twentieth century, an esti-
mated 200 million people perished through genocide. Totali-
tarian dictatorships, past wars, and ideologies of exclusiveness,
ethnic purity, and religious fundamentalism increase the risks
for genocide. Perpetrators use dehumanizing, demonizing,
and delegitimizing hate language to desensitize or intimidate
bystanders and to mobilize, order, and instruct followers.
Genocide prevention requires international surveillance
networks for monitoring and reporting incitement and hate
language in the media, textbooks, places of worship, and the
Internet, which should monitor and identify their sources
The New Public Health810
Eugenics was a movement within the “social hygiene” con-
cept of the early part of the twentieth century. It was widely
promoted to reduce births among mentally ill and handi-
capped people in some states in the USA and was upheld
in decisions of the Supreme Court. It was also practiced
in Canada and Sweden. This idea was promoted by Hitler
in Mein Kampf and adopted by the Nazi Party, which was
legally elected to office in 1933 and began to implement it.
Organized massacres of mentally ill and handicapped chil-
dren and adults led to practices of organizing various modes
of killing, including gas chambers, which were applied in
concentration camps and in the Holocaust murder of 6 mil-
lion Jews and millions of others.
Genocide represents the most extreme assault right to life
and respect for life. In the twentieth century, an estimated 200
million have perished from genocide. Totalitarian dictator-
ships, past war and defeat, ideologies of exclusiveness, ethnic
purity, and religious fundamentalism increase risks for geno-
cide. Perpetrators use dehumanizing, demonizing, and delegit-
imizing hate language to desensitize or intimidate bystanders
and to mobilize, order, and instruct followers.
1915–1917 Armenian genocide by Ottoman Turkish
Empire – 1.2 million killed
1920s–1940s Eugenics movement in USA and Sweden
1920s Mass executions, deportations, and starvation as
policy in Soviet Union Stalinist regimes
1930s–1940s Mass sterilization of “defectives” in the USA
and Sweden
1930–1940s Mass murder of “defectives” in Nazi Germany
– 750,000 killed
1940s Quarantining as pretext for ghettos by Nazis
1940s Concentration camps, human experimentation
1940s Holocaust of 6 million Jews and genocide in Nazi
occupation of Poland and in Soviet Union
1947 Nuremberg Trials – convictions and capital punishment
for war crimes and genocide by Nazi leaders and doctors
1950s Mass starvation in Maoist China – estimated deaths
of 21 million people
1948 Convention on the Prevention and Punishment of the
Crime of Genocide
1975–1979 Cambodian genocide – 1.7 million killed
1988 Iraqi genocide of Kurds in town of Halabja by
poison gas
1988 Brazil genocide conviction of Tikuna people
1995 Serbian massacres in Srebrenica in Bosnia and
Herzegovina
2004 Rwandan genocides
2003–2012 Sudanese genocide in Darfur – over 400,000
killed
2011 Sudanese genocide of Nuba people
2012 Iran incitement to genocide of Israel
2012 Syria: civil war and genocide
2012 Democratic Republic of Congo massacres of Kivu
reported
Sources: Richter ED, Genocide Prevention Center, Braun School Public
Health, Hebrew University –Hadassah, Jerusalem, Israel. Personal commu-
nication.
United Nations. Convention on the prevention and punishment of the crime
of genocide. Available at: http://www.hrweb.org/legal/genocide.html and
www.un.org/millennium/law/iv-1.htm [Accessed 16 December 2012].
Richter ED. Commentary. Genocide: can we predict, prevent, and protect? J
Public Health Policy 2008;29:265–74.
Stanton G. The eight stages of genocide; 1998. Available at: www.genocide-
watch.org/aboutgenocide/8stagesofgenocide.html
Genocide Watch. http://www.genocidewatch.org/ [Accessed 25 December
2012].
BOX 15.15 Eugenics and Genocide: The Slippery Slope
and map their distribution and spread. Dehumanization,
demonization, delegitimization, disinformation, and denial
are the danger signs of potential genocidal actions. Geno-
cide results from human choice and bystander indifference.
One lesson of the Holocaust is that silence in response to
incitement to genocide makes one a complicit bystander.
Public health professionals and institutions have a responsi-
bility to speak out publicly on such dangerous early warn-
ing signs (Richter E, personal communication, 2012).
Human Experimentation
Human experimentation has been a subject of great concern
since the Nazi and Imperial Japanese armed forces’ experi-
ments on prisoners and concentration camp victims during
World War II. The Nuremberg Trials set forth standards of
professional responsibility to comply with internationally
accepted medical behavior (Table 15.7).
The Helsinki Declaration was first adopted by the
World Medical Assembly in 1964, and amended in 1975,
1983, 1989, and 1996. It delineates standards of medical
experimentation and requires informed consent from sub-
jects of medical research. These standards have become an
international norm for experiments, with national, state,
and hospital Helsinki committees regulating research pro-
posals within their jurisdiction. Funding agencies require
standard approval by the appropriate Helsinki committee
before considering any proposal, with informed consent on
any research project.
The Tuskegee experiment (Box 15.17) was a grave and
tragic violation of medical ethics, but in the context of the
1930s was consistent with widespread and institutionalized
racism. It provides an important case study which has reper-
cussions until the present time in suspicion of public health
endeavors, particularly among the African American com-
munity in the USA.
http://www.hrweb.org/legal/genocide.html
http://www.un.org/millennium/law/iv-1.htm
http://www.genocidewatch.org/aboutgenocide/8stagesofgenocide.html
http://www.genocidewatch.org/aboutgenocide/8stagesofgenocide.html
http://www.genocidewatch.org/
811Chapter 15 Health Technology, Quality, Law, and Ethics
In the German context the social–ecological health paradigm
can be traced back to the late eighteenth and early nineteenth
centuries when the country was a loose alliance of kingdoms
or princedoms lagging behind the economic, cultural, and
political developments in England and France. Ensuring popu-
lation health was seen as the obligation of the state, while the
family was responsible for caring for the health and well-being
of its members. Organized health care and health maintenance
was seen in the framework of Medizinische Polizey, as a model
of the health systems. Leading scholars in law and medicine
shared a normative perspective of promoting a healthy life-
style (known as dietetics), and provision of shelter, food, and
spiritual aid in asylums for the sick and disabled, or in private
homes the for wealthy.
In the second half of the nineteenth century evidence from
medical statistics and overwhelming practical experience indi-
cated that widespread poverty was the critical factor explain-
ing high rates of typhus or cholera epidemics in lower social
classes among children and industrial workers. A social health
movement fought for healthier living and working conditions,
education, and democracy. The movement’s prominent lead-
ers were Salomon Neumann, a physician pioneer in medical
statistics, and Rudolf Virchow, the renowned pathologist and
outspoken political activist.
Between 1890 and 1930 the conceptual framework of pub-
lic health was defined as “social hygiene” or “health science”,
an interdisciplinary field to conduct scientific research, aca-
demic teaching, and community-based activities aiming at the
promotion of individual and collective health and the preven-
tion of disease. In the 1920s the field was highly developed and
pioneering the modern academic public health.
Social hygiene was a general framework open to different
definitions. A group of academic teachers and publishing scien-
tists sharing the social–ecological paradigm, among them a high
proportion of German Jews, wanted to continue the social reform
strategy and to strengthen local communities to take an active role
in the formulation and implementation of health policies.
Public health activists sharing the biotechnological disease
paradigm favored a more focused approach aiming at the con-
trol of disease through medical care. Although there was no
supportive evidence, in the late nineteenth century a racial
eugenic movement emerged widely in Europe and the USA.
A conceptual model derived from the disease paradigm postu-
lated racial factors to explain disease. A healthy population was
assumed to be “free” of “racially contaminated” individuals and
inferior groups. Health-related public policy was supposed to
eliminate racially “unclean” members, e.g., by forced steriliza-
tion or murder. This was a central theme in Hitler’s Mein Kampf
and was enacted as basic policy by the Nazi Party in Germany
as a fundamental ideological basis of racial theory and public
health.
When the Nazis were legally elected in Germany in 1933,
and later seized power in Austria, this policy provided fertile
ground to open the door to euthanasia, leading to mass mur-
der. This was implemented in the well-organized, medically
directed execution of mentally and physically handicapped
Germans and others in psychiatric facilities. This provided a
working model for the industrialized murder of 6 million Jews
in the Holocaust and millions of gypsies, homosexuals, com-
munists, and others.
It took only 10 years to eradicate a 200-year tradition of
German socially oriented public health grounded largely in the
political philosophy of human rights and social justice. Most
of those advocates were exiled or murdered. Many of the aca-
demic medical leaders after World War II remained in key posi-
tions in the German public sector for decades.
In contrast to many other countries, the two wealthy
German-speaking countries, with over 90 million people,
have few academic public health resources. In there is only
one German School of Public Health, and a small number of
institutes, far fewer in Austria than in Germany. More than half
a century has passed since the Nazi period and the populations
of these two countries are slow to build a new socially oriented
public health system.
Sources: Horst Noack MD, PhD, Professor Emeritus, Medical University of
Graz, Austria. Personal communication; 24 December 2012.
Flügel A. Public Health und Geschichte. Weinheim: Beltz Juventa; 2012.
Heinzelmann W. Sozialhygiene als Gesundheitswissenschaft. Bielefeld:
Transcript Verlag; 2009.
Noack H. Governance and capacity building in German and Austrian public
health since the 1950s. Public Health Rev 2011;33:264–76.
BOX 15.16 The Rise, Fall, and Slow Recovery of German Public Health
Ethics in Public Health Research
The border between practice and research is not always
easy to define in public health, which has as one of its
major tasks the surveillance of population health. This
surveillance is mostly anonymous but relies on individu-
ally identifiable data needed for reportable and infec-
tious disease control as well as for causes of death, birth
defects, mass screening programs, and other special dis-
ease registries. It may also be necessary to monitor the
effects of chronic disease, for example, to ascertain repeat
hospitalizations of patients with congestive heart failure
to assess the long-term effects of treatment, and the effects
of strengthening ambulatory and outreach services to sus-
tain chronic patients at a safe and functional level in their
own homes.
Hospitalizations, immunizations, and preventive care
practices (e.g., Pap smears, mammography, and colonosco-
pies) are all part of the New Public Health. Impact assessment
of preventive programs may require special surveys and are
important to assess smoking and nutritional status and other
measures of health status and risk factors. Every effort must be
made to preserve the anonymity and privacy of the individual
but in some cases, where the disease is contagious, case contact
is crucial. This can entail identifying people who attended an
The New Public Health812
TABLE 15.7 Ethical Issues of Medical Research Derived from the Nuremberg Trials, the Universal Declaration of
Human Rights, and the Declaration of Helsinki
Nuremberg Doctors Trial,
1946–47
The voluntary consent of a human subject is absolutely essential, with the exercise of free power of
choice without force, fraud, deceit, duress, or coercion
Experiments should be such as to bear fruitful results, based on prior experimentation and the natural
history of the problem under study. They should avoid unnecessary physical and mental suffering
The degree of risk should not exceed the humanitarian importance of the experiment
Persons conducting experiments are responsible for adequate preparations and resources for even the
remote possibility of death or injury resulting from the experiment
The human subject should be able to end his participation at any time
The scientist in charge is responsible to terminate the experiment if continuation is likely to result in
injury, disability, or death
Universal Declaration
of Human Rights, 1948
Everyone has the right to a standard of living adequate for the health and well-being of himself and of his
family, including food, clothing, housing, and medical care and necessary social services
United Nations covenants for
protection of human rights
Covenant on Civil and Political Rights
Optional Protocol to the Covenant on Civil and Political Rights
Covenant on Economic, Social, and Cultural Rights
Convention Against Torture
Convention Against Genocide
The Geneva Conventions
Convention on the Rights of the Child
Convention on Elimination of Discrimination Against Women
Charter of the United Nations
Declaration of Helsinki, 1964 Research must be in keeping with accepted scientific principles, and should be approved by specially
appointed independent committees
Biomedical research should be carried out by scientifically qualified persons, only on topics where
potential benefits outweigh the risks, with careful assessment of risks, where the privacy and integrity
of the individual is protected, and where the hazards are predictable. Publication must preserve the
accuracy of research findings
Each human subject in an experiment should be adequately informed of the aims, methods,
anticipated benefits, and hazards of the study. Informed consent should be obtained, and a
statement of compliance with this code
Clinical research should allow the doctor to use new diagnostic or therapeutic measures if they
offer benefit as compared to current methods
In any study, the patient and the control group should be assured of the best available methods.
Refusal to participate should never interfere with the doctor–patient relationship. The well-being
of the subject takes precedence over the interests of science or society
Source: Summarized from the Nuremberg Trials (1948) and World Medical Association, Declaration of Helsinki.
Website sources include: World Medical Association. Available at: http://www.wma.net/
Australian Government Department of Health and Ageing. Available at: http://www.nhmrc.gov.au/health-ethics/human-research-ethics-committees-hrecs/
human-research-ethics-committees-hrecs/national
United Nations. A Summary of United Nations Agreements on Human Rights. Available at: http://www.hrweb.org/legal/undocs.html (accessed 10.1.14).
United Nations. Available at: http://www.un.org/en/events/humanrightsday/2007/hrphotos/declaration%20_eng (accessed 10.1.14).
US Food and Drug Administration. World Medical Association Declaration of Helsinki. Available at: http://www.fda.gov/ohrms/dockets/
dockets/06d0331/06D-0331-EC20-Attach-1 (accessed 10.1.14).
event or traveled on an airplane where an infected person may
have been, so as to take appropriate preventive measures.
The general distinction between research and practice
has to do with the intent of the activity. Clinical research uses
experimental methods to establish the efficacy and safety of
new interventions or unproved interventions; many drugs
and procedures in common use have never been subjected
to randomized controlled trials. In practice, many methods
are devised that are held to be effective and safe by expert
opinion and documented as such. Researchers comparing
HIV or hepatitis B transmission rates among intravenous
drug users not using needle-exchange programs would be
conducting unethical research, according to accepted cur-
rent standards, by giving needles to the experimental group
http://www.nhmrc.gov.au/health-ethics/human-research-ethics-committees-hrecs/human-research-ethics-committees-hrecs/national
http://www.nhmrc.gov.au/health-ethics/human-research-ethics-committees-hrecs/human-research-ethics-committees-hrecs/national
http://www.un.org/en/events/humanrightsday/2007/hrphotos/declaration%2520_eng
http://www.fda.gov/ohrms/dockets/dockets/06d0331/06D-0331-EC20-Attach-1
http://www.fda.gov/ohrms/dockets/dockets/06d0331/06D-0331-EC20-Attach-1
Chapter 15 Health Technology, Quality, Law, and Ethics
and withholding them from the control group. The scientific
justification of an experiment must be made explicit and
justifiable. Clinical equivalence is a necessary condition
of all clinical and public health research and provision of
standard of care treatment to control groups is a minimal
requirement for most research ethics boards. Determination
of the standard, and whether it should be place, time, and
community specific, is an area of ongoing controversy.
In 1996 a US Public Health Service study, supported by
the NIH and WHO, compared a short course of zidovudine
(AZT) to a placebo given late in pregnancy to HIV-positive
women in Thailand, measuring the rate of HIV infection
among the newborns. The experiment was terminated when
a protest editorial appeared in a prominent medical journal.
This study confirmed previous findings that AZT given dur-
ing late pregnancy and labor reduced maternal–fetal HIV
transmission by half. When a study shows clearly positive
results, it should be discontinued and reported so that the
findings can be applied generally. The findings indicated
that AZT should be used in developing countries, and the
BOX 15.17 The Tuskegee Experiment
The Tuskegee experiment was carried out by the US Public
Health Service between 1932 and 1972. It was meant to fol-
low the natural course of syphilis in 399 already infected
African American men in Alabama and 201 uninfected men.
The men were not told that they were being used as research
subjects. The experiment had been intended to show the
need for additional services for those infected with syphilis.
However, when penicillin became available, the research-
ers did not inform or offer the men treatment, even those
who were eligible when drafted into the army in 1942. The
experiment was stopped in 1972 as “ethically unjustified”
when the media exposed it to public scrutiny.
The case is considered unethical research practice
because, even at the time it was conducted, it did not pro-
vide the patients with available care and their well-being
was put aside in the interest of the descriptive study. A similar
experiment was conducted by the US Public Health Service
in cooperation with the Guatamala Ministry of Health dur-
ing the 1960s, in which syphilis was actually given to sol-
diers, prisoners, and others by sexual contact with prostitutes
known to have the disease, but the study was terminated
when it was discovered by a public health historian and
reached public attention in the USA.
In 1997, President Bill Clinton apologized to the survi-
vors and families of the men involved in the experiment on
behalf of the US government. The Tuskegee experiment is
the source of lingering widespread suspicion in the African
American community to the present time.
Sources: Lombardo PA, Dorr GM. Eugenics, medical education and pub-
lic health: another perspective on the Tuskegee syphilis experiment. Bull
Hist Med 2006;80:291–316.
Centers for Disease Control and Prevention. US Public Health Service
Syphilis Study at Tuskegee. Available at: http://www.cdc.gov/tuskegee/
timeline.htm [Accessed 13 December 2012].
813
manufacturers agreed to make it available at reduced costs.
The result has been a major success in helping with more
recent medications to reduce maternal–fetal transmission in
many places in Africa with help from GAVI, and a slowing
of the spread of HIV/AIDS-related deaths.
Public health may face the challenge of pandemic influ-
enza, such as avian flu, with decisions regarding the allocation
of vaccines, treatment of massive numbers of patients arriv-
ing at hospitals in acute respiratory distress with very limited
resources available, coping with sick or absent staff, and many
other issues requiring not only individual life and death situ-
ations, but mortality en masse. The ethical questions will be
replaced by struggles to cope with such situations. Preparation
for such potential catastrophic events will be a challenge to
public health organizations and the health system in general.
An outstanding case of a breach of ethics in public health
research occurred with the “Wakefield effect”, as described
in Chapter 4 and Box 15.18.
Ethics in Patient Care
Ethical issues between the individual patient and health
care provider are important in the New Public Health. A
doctor is expected to use diligence, care, knowledge, skill,
discretion, and caution in keeping with practice standards
accepted at the time by responsible medical opinion and
to maintain the basic medical imperative to do no harm
to the patient. Patients have the right to know their condi-
tion, available alternatives for treatment, and the risks and
benefits involved. They also have a right to seek alterna-
tive medical opinions, but this right is not unlimited, as any
insurance plan or health service may place restrictions on
payment for further opinions and consultation without the
agreement of a primary care provider.
Health care has a responsibility beyond that of the pay-
ment of health service bills and individual care by a physi-
cian, in institutions, or through services in the community
or the home. The contract for service is becoming less
between an individual physician and his or her patient, and
more among a health system, its staff, and the client. This
places a new onus on the physician to ensure that patients
receive the care they require. Conversely, the US provider
often faces the dilemma of knowing that a patient may not
access needed services because of a lack of adequate health
insurance.
Sanctity of Life Versus Euthanasia
The imperative to save a life is an important ethical and prac-
tical issue in health care. Advocates of physician-assisted
suicide (euthanasia) argue for the right of the patient to die
with dignity when the illness is terminal and the individual
is suffering excessively. This is not a medical decision alone,
and is an agonizing issue for society to address. The Nazi
euthanasia program and its human experiments provided
http://www.cdc.gov/tuskegee/timeline.htm
http://www.cdc.gov/tuskegee/timeline.htm
The New Public Health814
In 1998, The Lancet, published an article by a number of well-
known researchers headed by Dr Andrew Wakefield. The article
reported on 12 cases of autistic children and alleged to show a
connection to immunization with the MMR (measles–mumps–
rubella) vaccine.
The immediate effect of this “revelation” was widespread
alarm over the MMR vaccine and a fall off in immuniza-
tion coverage by measles-containing vaccines in the UK
and elsewhere with many mothers refusing to have their
child vaccinated due to a “risk of autism”. As a result,
measles epidemics occurred in the UK and in many other
countries, with measles again becoming endemic in many
parts of Europe, especially England and France.
After a long series of investigative journalism in the British
press, the article came under scientific scrutiny and withdrawal
of many of the coauthors but a consistent insistence by the lead
author of its authenticity.
Investigation by British medical authorities later found
Dr Wakefield guilty of medical negligence and the UK
General Medical Council withdrew his license to practice
medicine. The coauthors were found to have been credulous
and insufficiently vigilant in agreeing to coauthorship of the
paper. In 2000, 12 years after the original publication, The
Lancet formally withdrew the article.
The effect of this fraudulent scientific publication was a
serious loss of credibility of immunization in general and
especially regarding the MMR vaccine, one of the greatest life
savers in public health technology.
The return of measles in Europe to large scale epidemics
with frequent international transmission furthered the loss of
confidence of mothers in immunizations and public health.
Measles-containing vaccines were particularly strongly
affected owing to the publicity given to the Wakefield case.
The journal editors could be seen as irresponsible for failing
to ensure the scientific integrity of lead authors and coauthors,
and the journal for failing to retract a fraudulent article sooner
than 12 years after the first publication.
In other public health issues, single publications of findings
of small sample and poorly assessed studies published in haste
without adequate inquisitive review occur with great frequency.
The electronic media often include unscientific opinion blogs
which appear larger than life which provoke great anxiety over
accepted and successful public health interventions such as flu-
oridation or folic acid fortification of flour, with unsubstantiated
claims that they cause cancer, asthma, and other ill-effects.
The interface between ethics, law, and science in pub-
lic health requires continuous sensitivity to the downstream
effects of “shouting fire in the theater”.
Sources: Wakefield AJ, Murch SH, Anthony A, Linnell, Casson DM,
Malik M, et al. Ileal lymphoid nodular hyperplasia, non-specific colitis,
and pervasive developmental disorder in children [retracted]. Lancet
1998;351:637–41.
Office of Research Integrity. Definition of research misconduct. Available at:
http://ori.hhs.gov/misconduct/definition_misconduct.shtml
General Medical Council. Andrew Wakefield: determination of serious
professional misconduct 24 May 2010. Available at: www.gmc-uk.org/
Wakefield_SPM_and_SANCTION _32595267
Murch SH, Anthony A, Casson DH, Malik M, Berelowitz M, Dhillon AP, et al.
Retraction of an interpretation. Lancet 2004;363:750.
Godlee F, Jane Smith J, Harvey Marcovitch H. Editorial. Wakefield’s
article linking MMR vaccine and autism was fraudulent. BMJ
2011;342:c7452.
BOX 15.18 The Wakefield Effect
the direst of warnings to societies of what may follow when
the principle of the sanctity of the individual human life is
breached. The issue, however, returned to the public agenda
in the 1980s and 1990s as advances in medical science have
allowed the prolongation of human life beyond all hope of
recovery. Legislation in the Netherlands, the USA (“assisted
suicide” in the states of Washington, Oregon, and Montana),
and northern Australia has legally sanctioned euthanasia
with various safeguards in a variety of circumstances, such
as long-term comas or terminal illnesses.
Doctors, patients, relatives, and health care organiza-
tions need clear guidelines, orientation, procedures, legal
protection, and limitations where failure to take utmost
steps to “save” the patient by intubation, resuscitation, or
transplantation may cause legal jeopardy. Even though a
distinction can be drawn theoretically between permitting
and facilitating death, in practice, doctors in intensive care
units face such decisions regularly where the line is often
blurred. Hospital doctors routinely go to extreme mea-
sures to prolong the life of hopeless cases. Such decisions
should not be considered for economic reasons alone, but in
practice the costs of care of the terminally ill will be a driv-
ing force in debate of the issue. Living wills allow a patient
to refuse heroic measures such as resuscitation, with “do
not resuscitate” standing orders and assignment of power of
attorney to family members to make such decisions. Fam-
ily attitudes are important, but the social issue of redefining
the right of a patient to opt for legal termination of life by
medical means will be an increasingly important issue in
the twenty-first century.
The Imperative to Act or Not Act in Public
Health
As in other spheres of medicine and health, in public health
the decision whether to intervene on an issue is based on
identification and interpretation of the problem, the poten-
tial of the intervention to improve the situation, to do no
harm, and to convince the public and political levels of the
need for such intervention along with the resources to carry
it out. This process requires patience and a longer time-
frame than many other fields in health.
http://ori.hhs.gov/misconduct/definition_misconduct.shtml
http://www.gmc-uk.org/Wakefield_SPM_and_SANCTION _32595267
http://www.gmc-uk.org/Wakefield_SPM_and_SANCTION _32595267
Chapter 15 Health Technology, Quality, Law, and Ethics
Some interpretations of ethics in health consider that the
only purpose for which power can be rightfully exercised
over any member of a democratic community, against his
will, is to prevent harm to others. But this is not a dictum
that is applied to public health, which is obliged to act to
protect the public health in so many spheres such as food
and drug safety and environmental health, on a spectrum
that extends to banning smoking in public places, mandat-
ing food fortification, and many other areas of civil society.
Failure to act is an action, and when there is convincing
evidence of a problem that can be alleviated or prevented
entirely by an accepted and demonstrably successful inter-
vention, then the onus is on the public health worker to
advocate such action and to implement it as best as possible
under the existing conditions. Failure to do so is a breach of
“good standards of practice” and could be unethical. Iner-
tia of the public health system in the face of evidence of a
demonstrably effective modality such as adoption of state-
of-the-art vaccines or fortification of flour with folic acid
to prevent birth defects would come under this categoriza-
tion and may even constitute neglect and unethical practice.
This is not an easy categorization, because there is often
disagreement and even opposition to public health interven-
tions, as was the case with opposition to vaccination long
after Jenner’s crucial discovery of this procedure in the late
eighteenth century. It is also true today with opposition to
many proven measures such as fluoridation or fortification
of basic foods. Box 15.19 shows the ethical standards of the
APHA in 2006.
The use of ethical and high standards of practice in pub-
lic health (Box 15.20) requires an ideological commitment
to the advancement of health standards and use of best prac-
tices of international standards to the maximum extent pos-
sible under the local conditions in which the professional is
working. This is not an easy commitment as there is often
dispute and outright hostility to public health activities, in
part because of ethical distortions of great magnitude in the
past. But this is an optimistic field of activity because of the
great achievements it has brought to humankind. Prepara-
tion for disasters and unanticipated health emergencies in
addition to addressing current issues is a vital part of the
New Public Health and our ethical and professional com-
mitments.
SUMMARY
In order to maintain and improve standards of care, health
systems need quality assurance and technological assess-
ment as part of their ongoing operation. Poor-quality care
is costly in terms of iatrogenic diseases and prolonged or
repeated hospitalization. If innovations such as endoscopic
surgery are not introduced, then longer hospital stays are
needed for the same operation, wasting the patient’s time
and productivity, while utilizing expensive health care
815
resources, and incurring the risks associated with more
invasive surgery.
Health care is provided by people, as well as by insti-
tutions with a range of devices and equipment. The people
providing care, more than the technological facilities, set the
quality of care. Nevertheless, progress on the technological
side of medical care is vital to the continuing development
of the field. Modern medications, monitoring equipment,
laboratory services, and imaging devices have made enor-
mous contributions to advances in medical care. Appropriate
BOX 15.19 Principles of Ethical Public Health Practice:
American Public Health Association, 2006
l Public health should address principally the fundamen-
tal causes of disease and requirements for health, aiming
to prevent adverse health outcomes.
l Public health should achieve community health in a way
that respects the rights of individuals in the community.
l Public health policies, programs, and priorities should be
developed and evaluated through processes that ensure
an opportunity for input from community members.
l Public health should advocate and work for the empow-
erment of disenfranchised community members, aiming
to ensure that the basic resources and conditions neces-
sary for health are accessible to all.
l Public health should seek the information needed to
implement effective policies and programs that protect
and promote health.
l Public health institutions should provide communities
with the information they have that is needed for deci-
sions on policies or programs and should obtain the
community’s consent for their implementation.
l Public health institutions should act in a timely manner
on the information they have within the resources and
the mandate given to them by the public.
l Public health programs and policies should incorpo-
rate a variety of approaches that anticipate and respect
diverse values, beliefs, and cultures in the community.
l Public health programs and policies should be imple-
mented in a manner that most enhances the physical and
social environment.
l Public health institutions should protect the confidential-
ity of information that can bring harm to an individual or
community if made public. Exceptions must be justified
on the basis of the likelihood of significant harm to the
individual or others.
l Public health institutions should ensure the professional
competence of their employees.
l Public health institutions and their employees should
engage in collaborations and affiliations in ways that
build the public’s trust and the institution’s effectiveness.
Source: American Public Health Association. Public Health Leadership
Society. Principles of the ethical practice of public health. APHA; 2002.
Available at: http://www.apha.org/NR/rdonlyres/1CED3CEA-287E-4185-
9CBD-BD405FC60856/0/ethicsbrochure [Accessed 13 December
2012].
http://www.apha.org/NR/rdonlyres/1CED3CEA-287E-4185-9CBD-BD405FC60856/0/ethicsbrochure
http://www.apha.org/NR/rdonlyres/1CED3CEA-287E-4185-9CBD-BD405FC60856/0/ethicsbrochure
The New Public Health816
Publication in peer-reviewed journals is a key part of the
advancement in science and a vital part of the development
of the scientific basis for public health practice. The process of
publication should promote rigorous standards of high quality
ethical research and the wide dissemination of their findings.
Codes of practice for editors and publishers of peer-reviewed
journals have been developed by both the Committee on
Publication Ethics (COPE) (Rees, 2011) and the World
Association of World Editors (WAME).
Editors are subject to competitive pressures, and the over-
arching metric of success is seen to be the impact factor, a
measure of the frequency with which the “average article”
in a journal has been cited in a particular year or period.
Relevant, rigorous research of better quality will tend to be
cited more frequently, and thus editorial strategies that look
for quality and relevance in the given field will increase the
impact factor. However, there can also be potential distort-
ing factors. Publishing a highly controversial paper can result
in high citation levels. Publishing studies which demonstrate
negative findings may be less likely to attract large numbers
of citations.
Key issues relate to conflicts of interest, and the potential for
advertising and sponsorship to distort editorial decision mak-
ing (Gray, 2012). A particular concern has been the pernicious
influence of the tobacco industry in sponsoring, frequently
covertly, research which has aimed to confuse or obfuscate key
findings linking second hand exposure to tobacco to adverse
impacts on health. Similar tactics are used in other areas where
health and commercial interests collide. Clear statements of
potential conflicts of interest are essential. Journal owners must
not interfere in the evaluation, selection, or editing of individ-
ual articles, either directly or by creating an environment in
which editorial decisions are strongly influenced.
Other challenging areas are plagiarism and research mis-
conduct. The latter is extremely difficult both to detect and to
deal with, and requires close working between institutions and
editors who may suspect professional misconduct. In cases of
fraud, the publishing journal should withdraw the article in a
timely fashion (see Box 15.18: The Wakefield Effect).
There has been a rapid rise in open access publishing, in
part underpinned by an ethical belief that research is a public
good, and an increasing number of influential research funders
now require that there should be unrestricted access to the
published output of research. In addition, several publishers
make their journals free to those in selected low-income coun-
tries, promoting dissemination to those who might not other-
wise afford them.
In summary, publication in peer-reviewed journals remains
a key method for establishing and progressing the evidence
base for public health practice. The consequences of poor
or frankly fraudulent science can have a substantial adverse
impact both on health and on the use of resources. Editors must
adhere to high ethical and professional standards and remain
vigilant to avoid allowing external drivers to distort their deci-
sion-making processes. They must strive to maintain integrity
and high scientific standards to advance the field of public
health practice (Smith, 2007).
Sources: Selena Gray, BSc, MBCHB, MD, FFPH, FRCP, Professor, University
of West of England, Bristol, and Deputy Postgraduate Dean, Severn Deanery,
Bristol, UK. Personal communication.
Rees M. Code of conduct and best practice guidelines for journal editors.
Committee on Publication Ethics; 2011. Available at: http://publicationeth-
ics.org/ [Accessed 21 August 2012].
Gray S. The ethics of publication in public health. Public Health Rev 2012;34.
Epub ahead of print. Available at: www.publichealthreviews.eu [Accessed 20
December 2012].
Smith R. The trouble with medical journals. London: Royal Society of
Medicine Press; 2007.
BOX 15.20 The Ethics of Publication in Public Health
technology is a critical issue for international health, since
the most advanced technology may be completely inappro-
priate in a setting that cannot afford to maintain it or lacks
the trained personnel to operate it, or where it comes in
place of more vital basic primary care services. Technology
assessment needs to be seen in the context of the country and
its resources for health care.
Ethical issues in public health are no less demanding
than those related to individual clinical care. The rights of
the individual and those of the community are sometimes in
conflict. Technology, quality, the law, and ethics are closely
interrelated in public health. Well-informed and sensitive
analysis of all aspects of their development is a part of the
New Public Health. The balance between individual and
community rights is very sensitive and must be kept under
continuous surveillance.
The New Public Health is replete with technological
and ethical questions, especially in a time of cost restraint,
increasing technological potential, the public expectation
of universal access to health care, and the assumption that
everyone will live a healthy and long life. Health status has
always been linked with socioeconomic status and, despite
enormous gains, this remains true even in the most egalitar-
ian countries. Expansion of market mechanisms, such as
controlling the supply of hospital beds, doctors, and access
to referrals, competition and incentives/disincentives in
payment systems for hospital and managed care systems,
contribute to a need for dynamic health policy management
capacity. The New Public Health assumes a social responsi-
bility for health for all, using community and personal care
modalities as effectively as possible to achieve that overall
goal.
NOTE
For a complete bibliography and guidance for student
reviews and expected competencies please see companion
web site at http://booksite.elsevier.com/9780124157668
http://booksite.elsevier.com/9780124157668
http://publicationethics.org/
http://publicationethics.org/
Chapter 15 Health Technology, Quality, Law, and Ethics
BIBLIOGRAPHY
Health Technology
Agosti J.M., Goldie S.J. 2007. Introducing HPV vaccine in developing
countries–key challenges and issues. N. Engl. J. Med. 356:1908–1910.
Available at: http://www.nejm.org/doi/full/10.1056/NEJMp078053
(accessed 10.01.14).
Alliance, G.A.V.I., 2011. GAVI alliance secures lower price for rotavi-
rus vaccine. Available at: http://www.gavialliance.org/library/news/
press-releases/2012/gavi-secures-lower-price-rotavirus-vaccine/
(accessed 15.12.12).
American Cancer Society, 30 July 2013. US task force makes recommen-
dations for lung cancer screening. Available at: http://www.cancer.org/
cancer/news/us-task-force-makes-recommendations-for-lung-cancer-
screening (accessed 12.08.13).
American Congress of Obstetricians and Gynecologists, 2012. ACOG
practice bulletin no. 131: screening for cervical cancer. Obstet.
Gynecol. 120, 1222–1238. http://www.acog.org/About%20ACOG/
Announcements/New%20Cervical%20Cancer%20Screening%20
Recommendations.aspx (accessed 22.5.13).
Brenzel L., Wolfson L.J., Fox-Rushby J., Miller M., Halsey N.A. 2006.
Vaccine-preventable diseases. In: Jamison D.T., Breman J.G.,
Measham A.R., Alleyne G., Claeson M., Evans DB, et al., (Ed.) Dis-
ease control priorities in developing countries, second ed. World Bank
Washington, DC. Available at: http://www.ncbi.nlm.nih.gov/
pubmed/21250343 (accessed 7.10.2012).
Campbell, B., 2013. Regulation and safe adoption of new medical devices
and procedures. Br. Med. Bull. 1–14. Available at: http://www.ncbi.
nlm.nih.gov/pubmed/23896485 (accessed 10.01.14).
Canadian Council of Health Services Accreditation. http://www.accredita-
tion.ca/ (accessed 10.01.14).
Centers for Disease Control and Prevention. Healthcare associated infec-
tions (HAI) [updated 23 September 2013]. Available at: http://www.
cdc.gov/hai/ (accessed 10.01.14).
Centers for Disease Control and Prevention, 2009. The direct medical costs
of healthcare-associated infections in US hospitals and the benefits of
prevention. CDC, Atlanta, GA. Available at: http://www.cdc.gov/HAI/
pdfs/hai/Scott_CostPaper (accessed 14.12.12).
Centers for Disease Control and Prevention. Vaccine price list. Available
at: http://www.cdc.gov/vaccines/programs/vfc/awardees/vaccine-
management/price-list/index.html (accessed 19.09.12).
Centers for Disease Control and Prevention, 2013. Percentage of women
Aged 50–64 years who reported receiving a mammogram in the past
2 years, by health insurance status – National Health Interview Sur-
vey, United States, 1993–2010. MMWR. Morb. Mortal. Wkly. Rep.
62, 651. Available at: http://www.cdc.gov/mmwr/preview/mmwrhtml/
mm6232a5.htm?s_cid=mm6232a5_w (accessed 10.01.14).
Danish Centre for Health Technology Assessment, 2008. Health technology
assessment handbook, second ed. National Board of Health. Available
at: http://sundhedsstyrelsen.dk/~/media/C0ED080616D7410E8B-
6020B903AD0339.ashx (accessed 10.01.14).
Deyo, R.A., 2002. Cascade effects of medical technology. Annu. Rev.
Public Health 23, 23–44. Available at: http://www.annualreviews.
org/doi/abs/10.1146/annurev.publhealth.23.092101.134534 (accessed
10.01.14).
Fitzpatrick, C., Floyd, K., 2012. A systematic review of the cost and cost
effectiveness of treatment for multidrug-resistant tuberculosis. Phar-
macoeconomics 30, 63–80. Available at: http://www.ncbi.nlm.nih.
gov/pubmed/22070215 (accessed 13.12.12).
817
Flahault, A., Martin-Moreno, JM., 2013. Why do we choose to address health
2020? Public Health Reviews. 2013;35: epub ahead of print. Available
at: http://www.publichealthreviews.eu/upload/pdf_files/13/00_Edito-
rial (accessed 04.01.14).
Greenberg, M.R., 2006. The diffusion of public health innovations. Am.
J. Public Health 96, 209–210. Available at: http://www.ncbi.nlm.nih.
gov/pmc/articles/PMC1470490/ (accessed 10.11.14).
Harris, R.P., Helfand, M., Woolf, S.H., Lohr, K.N., Mulrow, C.D., Teutsch,
S.M., et al., 2001. Current methods of the US Preventive Services Task
Force. A review of the process. Am. J. Prev. Med. 20, 21–35. Avail-
able at: http://www.ncbi.nlm.nih.gov/pubmed/11306229 (accessed
10.01.14).
Health Services/Technology Assessment (HSTAT). Available at: http://
www.ncbi.nlm.nih.gov/books/NBK16710/ (accessed 17.08.12.).
Institute of Medicine, 1985. Assessing medical technologies. National
Academies Press, Washington, DC. Available at: http://www.nap.edu/
openbook.php?record_id=607 (accessed 10.01.14).
International Network of Agencies for Health Technology Assessment
(INAHTA). http://www.inahta.org/ (accessed 17.08.12).
Ikeda, N., Sapienza, D., Guerrero, R., Aekplakorn, W.,et al. 2914, Con-
trol of hypertension with medication: a comparative analysis of
national surveys in 20 countries. Bull. World Health Organ.; 92(1):
10–19. Available at: http://www.ncbi.nlm.nih.gov/pmc/articles/
PMC3865548/ (accessed 10.01.14).
Lancet, 2011. Financing HPV vaccination in developing countries. Lan-
cet 377, 1544. Available at: http://www.thelancet.com/journals/lancet/
article/PIIS0140-6736(11)60622-3/fulltext (accessed 10.01.14).
Larson, E.L., Quiros, D., Lin, S.X., 2007. Dissemination of the CDC’s
hand hygiene guideline and impact on infection rates. Am. J. Infect.
Control. 35, 666–675. Available at: http://www.ncbi.nlm.nih.gov/
pubmed/18063132 (accessed 10.01.14).
Lehoux, P., Tailliez, S., Denis, J.L., Hivon, M., 2004. Redefining health tech-
nology assessment in Canada: diversification of products and contextu-
alization of findings. Int. J. Technol. Assess 20, 325–336. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/15446762 (accessed 10.01.14).
Martelli, F., La Torre, G., Di Ghionno, E., Neroni, M., Cicchetti, A., Von
Bremen, K., et al., 2007. Health technology assessment agencies:
an international overview of organizational aspects. Int. J. Technol.
Assess 23, 414–424. Available at: http://www.ncbi.nlm.nih.gov/
pubmed/17937828 (accessed 10.01.14).
Murray, A., Lourenco, T., de Verteuil, R., Hernandez, R., Fraser, C.,
McKinley, A., et al., 2006. Clinical effectiveness and cost-effectiveness
of laparoscopic surgery for colo-rectal cancer: systematic reviews and
economic evaluation. Health Technol. Assess 10, 1–141. Available at:
http://www.ncbi.nlm.nih.gov/books/NBK62293/ (accessed 10.01.14).
National Library of Medicine, 2007. National information center on health
services research & health care technology. Etext on health technology
assessment (HTA) information resources. Available at: http://www.nlm.
nih.gov/archive//20060905/nichsr/ehta/ehta.ht (accessed 10.01.14).
Nixon, J., Stoykova, B., Glanville, J., Christie, J., Drummond, M., Klei-
jnen, J., 2000. The UK NHS economic evaluation database. Eco-
nomic issues in evaluations of health technology. Int. J. Technol.
Assess 16, 731–742. Available at: http://www.ncbi.nlm.nih.gov/
pubmed/11028129 (accessed 10.01.14).
Noorani, H.Z., Husereau, D.R., Boudreau, R., Skidmore, B., 2007. Pri-
ority setting for health technology assessments: a systematic review
of current practical approaches. Int. J. Technol. Assess 23, 310–315.
Available at: http://www.ncbi.nlm.nih.gov/pubmed/17579932 (accessed
10.01.14).
http://www.nejm.org/doi/full/10.1056/NEJMp078053
http://www.gavialliance.org/library/news/press-releases/2012/gavi-secures-lower-price-rotavirus-vaccine/
http://www.gavialliance.org/library/news/press-releases/2012/gavi-secures-lower-price-rotavirus-vaccine/
http://www.cancer.org/cancer/news/us-task-force-makes-recommendations-for-lung-cancer-screening
http://www.cancer.org/cancer/news/us-task-force-makes-recommendations-for-lung-cancer-screening
http://www.cancer.org/cancer/news/us-task-force-makes-recommendations-for-lung-cancer-screening
http://www.acog.org/About%20ACOG/Announcements/New%20Cervical%20Cancer%20Screening%20Recommendations.aspx
http://www.acog.org/About%20ACOG/Announcements/New%20Cervical%20Cancer%20Screening%20Recommendations.aspx
http://www.acog.org/About%20ACOG/Announcements/New%20Cervical%20Cancer%20Screening%20Recommendations.aspx
http://www.ncbi.nlm.nih.gov/pubmed/21250343
http://www.ncbi.nlm.nih.gov/pubmed/21250343
http://www.ncbi.nlm.nih.gov/pubmed/23896485
http://www.ncbi.nlm.nih.gov/pubmed/23896485
http://www.cdc.gov/hai/
http://www.cdc.gov/hai/
http://www.cdc.gov/HAI/pdfs/hai/Scott_CostPaper
http://www.cdc.gov/HAI/pdfs/hai/Scott_CostPaper
http://www.cdc.gov/vaccines/programs/vfc/awardees/vaccine-management/price-list/index.html
http://www.cdc.gov/vaccines/programs/vfc/awardees/vaccine-management/price-list/index.html
http://www.cdc.gov/mmwr/preview/mmwrhtml/mm6232a5.htm?s_cid=mm6232a5_w
http://www.cdc.gov/mmwr/preview/mmwrhtml/mm6232a5.htm?s_cid=mm6232a5_w
http://sundhedsstyrelsen.dk/%7e/media/C0ED080616D7410E8B6020B903AD0339.ashx
http://sundhedsstyrelsen.dk/%7e/media/C0ED080616D7410E8B6020B903AD0339.ashx
http://www.annualreviews.org/doi/abs/10.1146/annurev.publhealth.23.092101.134534
http://www.annualreviews.org/doi/abs/10.1146/annurev.publhealth.23.092101.134534
http://www.ncbi.nlm.nih.gov/pubmed/22070215
http://www.ncbi.nlm.nih.gov/pubmed/22070215
http://www.publichealthreviews.eu/upload/pdf_files/13/00_Editorial
http://www.publichealthreviews.eu/upload/pdf_files/13/00_Editorial
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1470490/
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1470490/
http://www.ncbi.nlm.nih.gov/pubmed/11306229
http://www.ncbi.nlm.nih.gov/books/NBK16710/
http://www.ncbi.nlm.nih.gov/books/NBK16710/
http://www.nap.edu/openbook.php?record_id=607
http://www.nap.edu/openbook.php?record_id=607
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3865548/
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3865548/
http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(11)60622-3/fulltext
http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(11)60622-3/fulltext
http://www.ncbi.nlm.nih.gov/pubmed/18063132
http://www.ncbi.nlm.nih.gov/pubmed/18063132
http://www.ncbi.nlm.nih.gov/pubmed/15446762
http://www.ncbi.nlm.nih.gov/pubmed/17937828
http://www.ncbi.nlm.nih.gov/pubmed/17937828
http://www.ncbi.nlm.nih.gov/books/NBK62293/
http://www.nlm.nih.gov/archive//20060905/nichsr/ehta/ehta.ht
http://www.nlm.nih.gov/archive//20060905/nichsr/ehta/ehta.ht
http://www.ncbi.nlm.nih.gov/pubmed/11028129
http://www.ncbi.nlm.nih.gov/pubmed/11028129
http://www.ncbi.nlm.nih.gov/pubmed/17579932
818
Sassi, F., 2003. Setting priorities for the evaluation of health interven-
tions: when theory does not meet practice. Health Policy 63, 141–
154. Available at: http://www.ncbi.nlm.nih.gov/pubmed/12543527
(accessed 10.01.14).
Soto, J., 2002. Health economic evaluations using decision analytic model-
ing. Principles and practices – utilization of a checklist to their devel-
opment and appraisal. Int. J. Technol. Assess 18, 94–111. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/11987445 (accessed 10.01.14).
Upshur, R.E.G., 2002. Principles for the justification of public health inter-
ventions. Can. J. Public Health 93, 101–103. Available at: http://journal.
cpha.ca/index.php/cjph/article/download/217/217 (accessed 10.01.14).
Williams, I., Bryan, S., McIver, S., 2007. How should cost-effectiveness
analysis be used in health technology coverage decisions? Evi-
dence from the National Institute for Health and Clinical Excellence
approach. J. Health Serv. Res. Policy 12, 73–79. Available at: http://
www.ncbi.nlm.nih.gov/pubmed/17407655 (accessed 10.01.14).
Woods, K., 2002. Health technology assessment for the NHS in England
and Wales. Int. J. Technol. Assess 18, 161–165. Available at: http://
www.ncbi.nlm.nih.gov/pubmed/12053415 (accessed 10.01.14).
Quality
Agency for Healthcare Research and Quality. http://www.ahrq.gov/
(accessed 14.12.12).
AHRQ. U.S. Preventive Services Task Force USPSTF), an introduction.
Available at: http://www.ahrq.gov/professionals/clinicians-providers/
guidelines-recommendations/uspstf/index.html.
American College of Cardiology (ACC), http://www.cardiosource.org/
science-and-quality.aspx (accessed 20.08.12).
Cancer Research UK, October 2012. Who is screened for breast cancer? Avail-
able at: http://www.cancerresearchuk.org/cancer-help/type/breast-cancer/
about/screening/who-is-screened-for-breast-cancer (accessed 10.01.14).
Centers for Disease Control and Prevention, 2013. Quick stats. MMWR.
Morb. Mortal. Wkly. Rep. 62, 651. Available at: http://www.cdc.gov/
mmwr/pdf/wk/mm6232 (accessed 17.08.13).
Reviews, Cochrane, http://www.cochrane.org/cochrane-reviews (accessed
14.12.12).
Dannenberg, A.L., Bhatia, R., Cole, B.L., Dora, C., Fielding, J.E., Kraft, K.,
et al., 2006. Growing the field of health impact assessment: an agenda
for research and practice. Am. J. Public Health 96, 262–270. Avail-
able at: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1470491/
(accessed 10.01.14).
Deeks, J.J., 2001. Systematic reviews in health care: systematic
reviews of evaluations of diagnostic and screening tests. BMJ 323,
157–162. Available at: http://www.ncbi.nlm.nih.gov/pmc/articles/
PMC1120791/ (accessed 10.01.14).
European Society of Cardiology, Guidelines. Available at: http://www.
escardio.org/GUIDELINES-SURVEYS/ESC-GUIDELINES/Pages/
GuidelinesList.aspx (accessed 17.08.12).
Epstein, A.J., Polsky, D., Yang, F., Yang, L., Groeneveld, P.W., 2011.
Coronary revascularization trends in the United States, 2001–2008.
JAMA 305, 1769–1776. Available at: http://www.ncbi.nlm.nih.gov/
pubmed/21540420 (accessed 13.08.13).
Fielding, J.E., Briss, P.A., 2006. Promoting evidence-based public
health policy: can we have better evidence and more action? Health
Aff. 25, 969–978. Available at: http://content.healthaffairs.org/con-
tent/25/4/969.full (accessed 10.01.14).
Institute of Medicine, 2007. State of quality improvement and imple-
mentation research: expert views. Workshop summary. Institute of
Medicine of the National Academies of Science. National Academies
Press, Washington, DC. Available at: http://www.nap.edu/catalog.
php?record_id=11986 (accessed 10.01.14).
The New Public Health
International Guideline Network Library. http://www.g-i-n.net/library/
international-guidelines-library/ (accessed 14.12.12).
Kohn, L.T., Corrigan, J.M., Donaldson, M. (Eds.), 1999. Institute of
Medicine. Committee on quality of health care in America. To err is
human: building a safer health system. National Academies Press,
Washington, DC. Available at: http://www.nap.edu/openbook.
php?isbn=0309068371 (accessed 10.01.14).
Krech, L.A., El-Hadri, L., Evans, L., Fouche, T., et al., 2014. The
medicines quality database: a free public resource. Available at:
h t t p : / / w w w. w h o . i n t / bu l l e t i n / vo l u m e s / 9 2 / 1 / 1 3 – 1 3 0 5 2 6 . p d f
(accessed 10.01.14).
National Council for Quality Assurance. http://www.ncqa.org/ (accessed
10.01.14).
Pilkington, H., Blondel, B., Drewniak, N., Jennifer Zeitlin, 2014.
Where does distance matter? Distance to the closest maternity unit
and risk of foetal and neonatal mortality in France. Eur. J. Public
Health, 1–6. Available at: http://eurpub.oxfordjournals.org/content/
early/2014/01/01/eurpub.ckt207.full (accessed 10.01.14).
Shortell, S.M., Gillies, R., Wu, F., 2010. United States innovations in
health care delivery. Public Health Rev. 32, 190–212. Available at:
http://www.publichealthreviews.eu/show/f/26 (accessed 26.10.12).
The National Audit Office, 2012. Healthcare across the UK: A compari-
son of the NHS in England. Wales and Northern Ireland, Scotland.
Available at: http://www.officialdocuments.gov.uk/document/hc1213/
hc01/0192/0192 (accessed 10.01.14).
UK National Institute for Health and Care Excellence (NICE). Available
at: http://www.nice.org.uk/ (accessed 17.08.12).
US Department of Health and Human Services, August 2008. Office of
Inspector General. Adverse events in hospitals: overview of key
issues. Available at: http://oig.hhs.gov/oei/reports/oei-06-07-00470.
pdf (accessed 17.08.13).
US Food and Drug Administration, 2007. Department of Health and
Human Services. Amendment to the current good manufacturing prac-
tice regulations for finished pharmaceuticals. Direct final rule. Federal
Register 72, 68064–68070. Available at: http://www.ihi.org/Pages/
default.aspx (accessed 17.08.12).
US National Library of Medicine. Available at: http://www.nlm.nih.gov/
(accessed 13.02.14).
US National Information Center on Health Services Research. https://www.
nlm.nih.gov/nichsr/ (accessed 13.02.14).
US Preventive Services Task Force. Guide to preventive services. Report
of the US Preventive Services Task Force (USPSTF). http://www.ahrq.
gov/professionals/clinicians-providers/guidelines-recommendations/
uspstf/index.htm. Available at: http://www.ahrq.gov/clinic/USpstfix.
htm (accessed 13.02.14).
Law
American Society of Law, Medicine and Ethics. The public’s health and
the law in the 21st century: Second annual partnership conference on
public health law. J. Law. Med. Ethics. Special Suppl. 31 (4). Avail-
able at: http://www.stacks.cdc.gov/view/cdc/6751/cdc_6751_DS1
(accessed 10.1.14).
Annas, G., 1998. Human rights and health – the universal declaration
of human rights at 50. N. Engl. J. Med. 339, 1778–1781. Available
at: http://www.nejm.org/doi/full/10.1056/NEJM199812103392411
(accessed 10.01.14).
Burris, S., Wagenaar, A.C., Mello, M.M., 2010. Making the case for laws
that improve health: A framework for public health law research. Mil-
bank. Q. 88 (2), 169–210. Available at; http://www.ncbi.nlm.nih.gov/
pmc/articles/PMC2980343/ (accessed 10.01.14).
http://www.ncbi.nlm.nih.gov/pubmed/12543527
http://www.ncbi.nlm.nih.gov/pubmed/11987445
http://journal.cpha.ca/index.php/cjph/article/download/217/217
http://journal.cpha.ca/index.php/cjph/article/download/217/217
http://www.ncbi.nlm.nih.gov/pubmed/17407655
http://www.ncbi.nlm.nih.gov/pubmed/17407655
http://www.ncbi.nlm.nih.gov/pubmed/12053415
http://www.ncbi.nlm.nih.gov/pubmed/12053415
http://www.ahrq.gov/
http://www.ahrq.gov/professionals/clinicians-providers/guidelines-recommendations/uspstf/index.html
http://www.ahrq.gov/professionals/clinicians-providers/guidelines-recommendations/uspstf/index.html
http://www.cardiosource.org/science-and-quality.aspx
http://www.cardiosource.org/science-and-quality.aspx
http://www.cancerresearchuk.org/cancer-help/type/breast-cancer/about/screening/who-is-screened-for-breast-cancer
http://www.cancerresearchuk.org/cancer-help/type/breast-cancer/about/screening/who-is-screened-for-breast-cancer
http://www.cdc.gov/mmwr/pdf/wk/mm6232
http://www.cdc.gov/mmwr/pdf/wk/mm6232
http://www.cochrane.org/cochrane-reviews
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1470491/
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1120791/
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1120791/
http://www.escardio.org/GUIDELINES-SURVEYS/ESC-GUIDELINES/Pages/GuidelinesList.aspx
http://www.escardio.org/GUIDELINES-SURVEYS/ESC-GUIDELINES/Pages/GuidelinesList.aspx
http://www.escardio.org/GUIDELINES-SURVEYS/ESC-GUIDELINES/Pages/GuidelinesList.aspx
http://www.ncbi.nlm.nih.gov/pubmed/21540420
http://www.ncbi.nlm.nih.gov/pubmed/21540420
http://content.healthaffairs.org/content/25/4/969.full
http://content.healthaffairs.org/content/25/4/969.full
http://www.nap.edu/catalog.php?record_id=11986
http://www.nap.edu/catalog.php?record_id=11986
http://www.g-i-n.net/library/international-guidelines-library/
http://www.g-i-n.net/library/international-guidelines-library/
http://www.nap.edu/openbook.php?isbn=0309068371
http://www.nap.edu/openbook.php?isbn=0309068371
http://www.who.int/bulletin/volumes/92/1/13-130526
http://web.ncqa.org/tabid/577/Default.aspx
http://eurpub.oxfordjournals.org/content/early/2014/01/01/eurpub.ckt207.full
http://eurpub.oxfordjournals.org/content/early/2014/01/01/eurpub.ckt207.full
http://www.publichealthreviews.eu/show/f/26
http://www.officialdocuments.gov.uk/document/hc1213/hc01/0192/0192
http://www.officialdocuments.gov.uk/document/hc1213/hc01/0192/0192
http://www.nice.org.uk/
http://oig.hhs.gov/oei/reports/oei-06-07-00470
http://oig.hhs.gov/oei/reports/oei-06-07-00470
http://www.ihi.org/Pages/default.aspx
http://www.ihi.org/Pages/default.aspx
http://www.nlm.nih.gov/
https://www.nlm.nih.gov/nichsr/
https://www.nlm.nih.gov/nichsr/
http://www.ahrq.gov/professionals/clinicians-providers/guidelines-recommendations/uspstf/index.htm
http://www.ahrq.gov/professionals/clinicians-providers/guidelines-recommendations/uspstf/index.htm
http://www.ahrq.gov/professionals/clinicians-providers/guidelines-recommendations/uspstf/index.htm
http://www.ahrq.gov/clinic/USpstfix.htm
http://www.ahrq.gov/clinic/USpstfix.htm
http://www.stacks.cdc.gov/view/cdc/6751/cdc_6751_DS1
http://www.nejm.org/doi/full/10.1056/NEJM199812103392411
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2980343/
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2980343/
Chapter 15 Health Technology, Quality, Law, and Ethics
Centers for Disease Control and Prevention, Collaborating Center for the
Law and the Public’s Health. Available at: http://www.publichealthlaw.
net/ (accessed 17.08.12).
Fox, D., Kramer, M., Standish, M., 2003. From public health to population
health: how can law redefine the playing field? J. Law Med. Ethics.
31, 21–29. Available at: http://www.stacks.cdc.gov/view/cdc/6751/
cdc_6751_DS1 (accessed 10.01.14).
Goodman, R.A., Moulton, A., Matthews, G., Shaw, F., Kocher, P., Mensah,
G., et al., 2006. Law and public health at CDC. MMWR. Morb. Mor-
tal. Wkly. Rep. 55, 29–33. Available at: http://www.cdc.gov/mmwr/
preview/mmwrhtml/su5502a11.htm (accessed 10.01.14).
Gostin, L.O., 2000. Public health law in a new century. Part II: Public
health powers and limits. JAMA 283, 2979–2984. Available at: http://
www.ncbi.nlm.nih.gov/pubmed/10865277 (accessed 10.01.14).
Gostin, L.O., 2000. Public health law in a new century: Part III: Public
health regulation: a systematic evaluation. JAMA 283, 3118–3122.
Available at: http://www.ncbi.nlm.nih.gov/pubmed/10865307
(accessed 10.01.13).
Ruhl, S., Stephens, M., Locke, P., 2005. The role of NGOs in public health
law. J. Law Med. Ethics. 31, 76–77. Available at: http://www. stacks.cdc.
gov/view/cdc/6751/cdc_6751_DS1 (page 76). (accessed 10.01.14).
Thiel, K.S., 2003. New developments in public health case law. J. Law
Med. Ethics. 31, 86–87.
Zasa, S., Clymer, J., Upmeyer, L., Thacker, S., 2003. Using science-based
guidelines to shape public health law. J. Law Med. Ethics. 31, 65–67.
Ethics
Aceijas, C., Brall, C., Schröder-Bäck, P., Otok, R., Maeckelberghe, E.,
Stjernberg, L., et al., 2012. Teaching ethics in schools of public
health in the European Region: findings from a screening survey.
Public Health Rev. 34. Epub ahead of print. Available at: www.publi-
chealthreviews.eu (accessed 05.12.12).
American College of Epidemiology, 2000. Ethics guidelines. Ann. Epidemiol.
10, 487–497. Available at: http://acepidemiology.org/content/ethics.
American Public Health Association, Principles of the ethical prac-
tice of public health. Available at: http://www.apha.org/NR/
rdonlyres/1CED3CEA-287E-4185-9CBD-BD405FC60856/0/ethics-
brochure (accessed 17.08.12).
Bachrach, S., 2004. In the name of public health – Nazi racial hygiene. N.
Engl. J. Med. 351, 417–420. Available at: http://content.nejm.org/cgi/
content/full/351/5/417?ijkey=8bee5b41cf11cadc3a826ad16fc2deef59
aa8f32&keytype2=tf_ipsecsha (accessed 05.12.12).
Bayer, R., Fairchild, A., 2004. The genesis of public health ethics. Bio-
ethics 18, 473–492. Available at: http://www.ncbi.nlm.nih.gov/
pubmed/15580720 (accessed 10.01.14).
Birn, A.E., Molina, N., 2005. In the name of public health. Am. J. Pub-
lic Health 95, 1095–1097. Available at: http://www.ncbi.nlm.nih.gov/
pmc/articles/PMC1449322/ (accessed 10.01.14).
Carter, S.M., Cribb, A., Allegrante, J.P., 2012. How to think about health
promotion ethics. Public Health Rev. 34. Epub ahead of print. Avail-
able at: www.publichealthreviews.eu (accessed 05.12.12).
Centers for Disease Control and Prevention, US Public Health Service
syphilis experiment at Tuskegee. Available at: http://www.cdc.gov/
tuskegee/timeline.htm (accessed 17.08.12).
Centers for Disease Control, Advancing excellence and integrity of CDC
science. Available at: http://www.cdc.gov/od/science/integrity/pheth-
ics/ (accessed 10.01.14).
Coleman, C.H., Bouësseau, M.-C., Reis, A., 2008. The contribution of
ethics to public health. Bull. World Health Org. 86, 578–579. Avail-
able at: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2649472/
(accessed 10.01.14).
819
Coughlin, S.S., Barker, A., Dawson, A., 2012. Ethics and scientific integ-
rity in public health, epidemiological and clinical research. Pub-
lic Health Rev. 34 (1). Available at: www.publichealthreviews.eu
(accessed 05.12.12).
Coughlin, S.S., 2006. Ethical issues in epidemiologic research and public
health practice. Emerg Themes Epidemiol 3, 3–16. Available at: http://
www.ncbi.nlm.nih.gov/pmc/articles/PMC1594564/ (accessed 10.01.14).
Dawson, A., Paul, Y., 2006. Mass public health programs and the obliga-
tions of sponsoring and participating organizations. J. Med. Ethics 32,
580–583. Available at: http://www.ncbi.nlm.nih.gov/pmc/articles/
PMC2563318/ (accessed 10.01.14).
Dawson, A., 2004. Vaccination and the prevention problem. Bioethics 18,
515–530. Available at: http://onlinelibrary.wiley.com/doi/10.1111/
j.1467-8519.2004.00414.x/abstract?deniedAccessCustomisedMessag
e=&userIsAuthenticated=false (accessed 10.01.14).
El Amin, A.N., Parra, M.T., Kim-Farley, R., Fielding, J.E., 2012. Ethical
issues concerning vaccination requirements. Public Health Rev. 34
(1). Available at: www.publichealthreviews.eu (accessed 05.12.12).
Kass, N., 2004. Public health ethics: from foundations and frameworks
to justice and global public health. J. Law Med. Ethics. 32, 232–
242. Available at: http://www.ncbi.nlm.nih.gov/pubmed/15301188
(accessed 10.01.14).
Lee, L.M., 2012. Guest editorial: Public health ethics theory: review and
path to convergence. Public Health Rev. 34 (1). Available at: www.
publichealthreviews.eu (accessed 05.12.12).
Lynch, H.F., 2012. Ethical evasion or happenstance and hubris? The US
Public Health Service STD Inoculation Study. Hastings Cent. Rep.
42, 30–38. Available at: http://onlinelibrary.wiley.com/doi/10.1002/
hast.17/abstract (accessed 10.01.14).
Nuremberg Code, 1949. Trials of war criminals before the Nurem-
berg military tribunals under control council law no. 10, Vol. 2. US
Government Printing Office, Washington, DC. pp. 181–182. Organ-
isation for Economic Co-operation and Development, 2007. Global
Science Forum. Best practices for ensuring scientific integrity and
preventing misconduct. OECD, Available at: http://www.oecd.org/
dataoecd/37/17/40188303 (accessed 19.5.2012).
Reverby, S.M., 2012. Ethical failures and history lessons: the US
Public Health Service research studies in Tuskegee and Guatemala.
Public Health Rev. 34. Epub ahead of print. Available at: http://
www.publichealthreviews.eu/upload/pdf_files/11/00_Reverby
(accessed 10.01.14).
Thieren, M., Mauron, A., 2007. Nuremberg Code turns 60. B World Health
Organ 85, 573. Available at: http://www.publichealthreviews.eu/
upload/pdf_files/11/00_Reverby (accessed 10.01.14).
Thomas, J.C., Sage, M., Dillenberg, J., Guillory, V.J., 2002. A code of
ethics for public health. Am. J. Public Health 92, 1057–1060. Avail-
able at: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1447186/
(accessed 10.01.14).
US Presidential Commission for the Study of Bioethical Issues, 2011.
Creation of the commissions. Available at: http://bioethics.gov/cms/
creation-of-the-commission (accessed 05.12.12).
United Nations Programme on HIV/AIDS (UNAIDS), 2013. Ending
overly broad criminalisation, non-disclosure, exposure and transmis-
sion: Critical scientific, medical and legal considerations. Available
at: http://www.unaids.org/en/media/unaids/contentassets/documents/
document/2013/05/20130530_Guidance_Ending_Criminalisation
(accessed 20.1.14).
Zusman, S.P., 2012. Water fluoridation in Israel, 1968–2012. Public Health
Rev. 34. Epub ahead of print. Available at: http://www.publichealthre-
views.eu/upload/pdf_files/11/00_Zusman (accessed 10.01.14).
http://www.stacks.cdc.gov/view/cdc/6751/cdc_6751_DS1
http://www.stacks.cdc.gov/view/cdc/6751/cdc_6751_DS1
http://www.cdc.gov/mmwr/preview/mmwrhtml/su5502a11.htm
http://www.cdc.gov/mmwr/preview/mmwrhtml/su5502a11.htm
http://www.ncbi.nlm.nih.gov/pubmed/10865277
http://www.ncbi.nlm.nih.gov/pubmed/10865277
http://www.ncbi.nlm.nih.gov/pubmed/10865307
http://www.stacks.cdc.gov/view/cdc/6751/cdc_6751_DS1
http://www.stacks.cdc.gov/view/cdc/6751/cdc_6751_DS1
http://refhub.elsevier.com/B978-0-12-415766-8.00015-X/ref0300
http://refhub.elsevier.com/B978-0-12-415766-8.00015-X/ref0300
http://refhub.elsevier.com/B978-0-12-415766-8.00015-X/ref0305
http://refhub.elsevier.com/B978-0-12-415766-8.00015-X/ref0305
http://acepidemiology.org/content/ethics
http://www.apha.org/NR/rdonlyres/1CED3CEA-287E-4185-9CBD-BD405FC60856/0/ethicsbrochure
http://www.apha.org/NR/rdonlyres/1CED3CEA-287E-4185-9CBD-BD405FC60856/0/ethicsbrochure
http://www.apha.org/NR/rdonlyres/1CED3CEA-287E-4185-9CBD-BD405FC60856/0/ethicsbrochure
http://content.nejm.org/cgi/content/full/351/5/417?ijkey=8bee5b41cf11cadc3a826ad16fc2deef59aa8f32%26keytype2=tf_ipsecsha
http://content.nejm.org/cgi/content/full/351/5/417?ijkey=8bee5b41cf11cadc3a826ad16fc2deef59aa8f32%26keytype2=tf_ipsecsha
http://content.nejm.org/cgi/content/full/351/5/417?ijkey=8bee5b41cf11cadc3a826ad16fc2deef59aa8f32%26keytype2=tf_ipsecsha
http://www.ncbi.nlm.nih.gov/pubmed/15580720
http://www.ncbi.nlm.nih.gov/pubmed/15580720
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1449322/
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1449322/
http://www.cdc.gov/tuskegee/timeline.htm
http://www.cdc.gov/tuskegee/timeline.htm
http://www.cdc.gov/od/science/phec/
http://www.cdc.gov/od/science/phec/
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2649472/
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1594564/
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1594564/
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1594564/
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1594564/
http://onlinelibrary.wiley.com/doi/10.1111/j.1467-8519.2004.00414.x/abstract?deniedAccessCustomisedMessage=%26userIsAuthenticated=false
http://onlinelibrary.wiley.com/doi/10.1111/j.1467-8519.2004.00414.x/abstract?deniedAccessCustomisedMessage=%26userIsAuthenticated=false
http://onlinelibrary.wiley.com/doi/10.1111/j.1467-8519.2004.00414.x/abstract?deniedAccessCustomisedMessage=%26userIsAuthenticated=false
http://www.ncbi.nlm.nih.gov/pubmed/15301188
http://onlinelibrary.wiley.com/doi/10.1002/hast.17/abstract
http://onlinelibrary.wiley.com/doi/10.1002/hast.17/abstract
http://www.oecd.org/dataoecd/37/17/40188303
http://www.oecd.org/dataoecd/37/17/40188303
http://www.publichealthreviews.eu/upload/pdf_files/11/00_Reverby
http://www.publichealthreviews.eu/upload/pdf_files/11/00_Reverby
http://www.publichealthreviews.eu/upload/pdf_files/11/00_Reverby
http://www.publichealthreviews.eu/upload/pdf_files/11/00_Reverby
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1447186/
http://bioethics.gov/cms/creation-of-the-commission
http://bioethics.gov/cms/creation-of-the-commission
http://www.unaids.org/en/media/unaids/contentassets/documents/document/2013/05/20130530_Guidance_Ending_Criminalisa%20
http://www.unaids.org/en/media/unaids/contentassets/documents/document/2013/05/20130530_Guidance_Ending_Criminalisa%20
http://www.publichealthreviews.eu/upload/pdf_files/11/00_Zusman
http://www.publichealthreviews.eu/upload/pdf_files/11/00_Zusman
Introduction
Innovation, Regulation, and Quality Control
Appropriate Health Technology
Priority Interventions in Low- and Medium-Income Countries
Priority Selection in High-Income Countries
Health Technology Assessment
Technology Assessment in Hospitals
Technology Assessment in Prevention and Health Promotion
Technology Assessment in National Health Systems
Dissemination of Technology
Diffusion of Technology
Quality Assurance
Adverse Events and Negligence
Licensure and Certification
Health Facility Accreditation
Peer Review
Tracer Conditions
Setting Standards
Algorithms and Clinical Guidelines
Organization of Care
Diagnosis-Related Groups
Managed Care
Performance Indicators
Consumerism and Quality
The Public Interest
Total Quality Management
Public Health Law
Environmental Health
Public Health Law Reform
Ethical Issues In Public Health
Individual and Community Rights
Tragic Deviations in Public Health Ethics
Human Experimentation
Ethics in Public Health Research
Ethics in Patient Care
Sanctity of Life Versus Euthanasia
The Imperative to Act or Not Act in Public Health
Summary
Note
Bibliography
Health Technology
Quality
Law
Ethics
HE
A
LTH FINANCING GUIDANCE NO.
10
DIAGNOSIS-RELATED GROUPS (DRG):
A Question & Answer guide on case-based
classifi cation and payment systems
FINA
L
DR
AFT
© Copyright World Health Organiza on 2020
Diagnosis-related groups (DRG): A Ques on & Answer guide on case-based classifi ca on and payment systems.
/ Aurelie Klein, Inke Mathauer, Karin Stenberg and
Triin Habicht
WHO/UHC/HGF/Guidance/20.10
Some rights reserved. This work is available under
the Crea ve Commons A ribu on-NonCommercial-
ShareAlike 3.0 IGO licence (CC BY-NC-SA 3.0 IGO;
h ps://crea vecommons.org/licenses/by-nc-sa/3.0/
igo).
Under the terms of this licence, you may copy,
redistribute and adapt the work for non-commercial
purposes, provided the work is appropriately cited, as
indicated below. In any use of this work, there should
be no sugges on that WHO endorses any specifi c
organiza on, products or services. The use of the WHO
logo is not permi ed. If you adapt the work, then you
must license your work under the same or equivalent
Crea ve Commons licence. If you create a transla on
of this work, you should add the following disclaimer
along with the suggested cita on: “This transla on was
not created by the World Health Organiza on (WHO).
WHO is not responsible for the content or accuracy of
this transla on. The original English edi on shall be
the binding and authen c edi on”.
Any media on rela ng to disputes arising under the
licence shall be conducted in accordance with the
media on rules of the World Intellectual Property
Organiza on.
Suggested cita on. Klein A, Mathauer I, Stenberg K,
Habicht T. Diagnosis-related groups (DRG): A Ques on
& Answer guide on case-based classifi ca on and
payment systems. Geneva: World Health Organiza on;
2020 (WHO/UHC/HGF/Guidance/20.10). Licence: CC
BY-NC-SA 3.0 IGO.
Cataloguing-in-Publica on (CIP) data. CIP data are
available at h p://apps.who.int/iris.
Sales, rights and licensing. To purchase WHO
publica ons, see h p://apps.who.int/bookorders. To
submit requests for commercial use and queries on
rights and licensing, see h p://www.who.int/about/
licensing.
Third-party materials. If you wish to reuse material
from this work that is a ributed to a third party, such
as tables, fi gures or images, it is your responsibility
to determine whether permission is needed for that
reuse and to obtain permission from the copyright
holder. The risk of claims resul ng from infringement
of any third-party-owned component in the work rests
solely with the user.
General disclaimers. The designa ons employed and
the presenta on of the material in this publica on do
not imply the expression of any opinion whatsoever
on the part of WHO concerning the legal status of any
country, territory, city or area or of its authori es,
or concerning the delimita on of its fron ers or
boundaries. Do ed and dashed lines on maps
represent approximate border lines for which there
may not yet be full agreement.
The men on of specifi c companies or of certain
manufacturers’ products does not imply that they are
endorsed or recommended by WHO in preference
to others of a similar nature that are not men oned.
Errors and omissions excepted, the names of
proprietary products are dis nguished by ini al capital
le ers.
All reasonable precau ons have been taken by WHO
to verify the informa on contained in this publica on.
However, the published material is being distributed
without warranty of any kind, either expressed or
implied. The responsibility for the interpreta on and
use of the material lies with the reader. In no event
shall WHO be liable for damages arising from its use.
The named authors alone are responsible for the views
expressed in this publica on.
Printed in Switzerland.
FINAL DR
A
FT
List of fi gures, tables and country boxes
Acknowledgements
Acronyms and abbreviations
Key messages and summary
Introduction and purpose of this guide
1. DEFINITIONS
1.1. What is a case-based groups system? What is a diagnosis-related groups system?
1.2. What are case group weights, the base rate, case mix and case mix index?
1.3. How is the payment determined in a case-based payment system?
2. ASSESSMENT PHASE
2.1. Can a CBG payment system contribute to achieving UHC objectives?
2.2. What are the conducive factors and institutional requirements for a CBG system?
2.3. How long does it take to develop and implement a CBG system? What are the related
costs?
2.4. Which stakeholders are likely to infl uence the design and implementation of the CBG
system and how should the different stakeholders be involved?
3. PREPARATION AND DESIGN PHASE
3.1. Which types of health services, facilities and providers should be included in CBG
payment?
3.2. Which variables can be used for classifi cation and what are the implications?
3.3. Which cost items can be included in the CBG payment?
3.4. When is it best to buy an existing DRG system and when should a country develop its
own country-specifi c CBG system?
3.5. How are case-group weights and the base rate calculated?
3.6. Which data are needed to calculate case-group weights and how can these be
generated?
3.7. Which adjustment factors can be considered to modulate the CBG payment?
3.8. Should CBG payments be linked to a volume or budget cap and, if so, how?
4. IMPLEMENTATION, MONITORING AND CONTINUED REVISION
4.1. What changes are required at provider level when introducing a CBG system?
4.2. How should the CBG system be piloted and phased in?
4.3. How can one counter inherent provider incentives created by CBGs?
4.4. How can a CBG system improve quality in service provision?
4.5. How and how often does a CBG system need to be revised?
4.6. How should one inform the public about changes in payment methods and billing
practices?
CONCLUDING REMARKS
GLOSSARY
REFERENCES
ANNEX 1: EXAMPLES OF CLASSIFICATION SYSTEMS WITH DIFFERENT LEVELS OF
COMPLEXITY
ANNEX 2: MAIN PROVIDER PAYMENT METHODS AND THE INCENTIVES THEY CREATE
ANNEX 3: LIST OF MAJOR DIAGNOSTIC CATEGORIES (MDC) OF TWO DRG VARIANTS
iv
vi
vii
viii
ix
1
1
3
3
5
5
8
10
13
14
14
16
19
20
26
29
33
34
36
36
38
40
42
43
44
46
4
7
50
56
58
59
TABLE OF CONTENTS
FINA
L DR
AFTT LEVELS OFLEVELS OF
ND THE INCENTIVESD THE INCENTIVE
ORIES (MDC) OORIES (MDC)
FT
L DR
A
iv HEALTH FINANCING GUIDANCE NO. 10
FIGURES
TABLES
Table 1. Poten al eff ects of CBG introduc on on cost containment, eff ec ve use of
resources, informa on collec on and administra ve effi ciency in comparison
to the previous payment method
Table 2. Main DRG variants which have been imported and/or further developed by
other countries
Table 3. Issues to consider when deciding whether to make or buy a DRG system
Table 4. Cost informa on required and sources of informa on for diff erent case
grouping systems
Table 5. Unintended consequences of CBG payment systems and measures to address
them
Figure 1. Overview of the level of complexity of defi ni on of cases, number of groups
and informa on requirements for case-based groups systems and diagnosis-
related groups systems
Figure 2. Formula for calcula ng the rate of a CBG or DRG case group
Figure 3. Year in which a country started moving towards DRGs and year in which the
system was used for payment, for selected countries
Figure 4. Aspects to consider when deciding which types of services, facili es and
providers should be included in a CBG system
Figure 5. Link between volume of services and provider revenue for diff erent
arrangements regarding budget caps
7
21
26
30
41
3
4
11
15
34
LIST OF FIGURES, TABLES AND
COUNTRY BOXES
FINA
L DR
AFT
vLIST OF FIGURES, TABLES AND COUNTRY BOXES
COUNTRY BOXES
10
12
18
19
22
23
25
27
28
30
37
45
Box 1. The importance of provider autonomy to accompany CBG introduc on in
Mongolia
Box 2. DRGs in Chile: a case of progressive implementa on
Box 3. Development of Ghana’s pa ent classifi ca on system
Box 4. Process for iden fi ca on of relevant DRG classifi ca on variables in Viet Nam
Box 5 Buy and make in Iran: adop ng an imported grouper to develop its own DRG
system
Box 6. Gradual development of a DRG system in Kyrgyzstan
Box 7. Matching classifi ca on systems and electronic claims data format to facilitate
the adop on of the NordDRG variant in Georgia
Box 8. Introduc on of a pa ent-level cos ng system to develop country-specifi c case-
group weights in Lithuania
Box 9. A mix of imported and locally-developed case-group weights in
Estonia
Box 10. PhilHealth’s plan to ins tu onalize the collec on of cost informa on
Box 11. Introduc on of ICD-10 in the Philippines and training measures
Box 12. Informa on provision to the public on the Universal Coverage Scheme, its
benefi ts and payment system in Thailand
FINA
L DR
AFT
ACKNOWLEDGEMENTS
Special thanks go to the authors of the country case boxes:
Chile: Camilo Cid, Pan American Health Organiza on and Gabriel Bas as, Department of
Public Health, Pon fi cia Universidad Católica de Chile.
Iran:
Mahdi Naderi
, Offi ce of HTA, Healthcare Standardiza on and Tariff Se ng, Ministry of
Health, Tehran.
Kyrgyzstan: Triin Habicht, Interna onal health fi nancing expert.
Mongolia: Tsolmongerel Tsilaajav, WHO SEARO consultant.
Philippines: Ronald Paguirigan and Melanie Coronel San llan, Philippine Health Insurance
Corpora on.
Thailand: Na adhanai Rajatanavin, Hathairat Kosiyaporn, Nithiwat Saengruang, Walaiporn
Patcharanarumol and Viroj Tangcharoensathien, Interna onal Health Policy Program,
Ministry of Public Health, Thailand.
Viet Nam: Dr. Tham Chi Dung, Department of Planning and Finance, Ministry of Health,
Hanoi.
Valuable comments from Cheryl Cashin, Elina Dale, Fahdi Dkhimi, Christopher Fitzpatrick,
Celina Gacias, Robert Jakob, Grace Kabaniha, Nenad Kostansjek, Kenneth Munge Kabubei,
Kris ina Kahur, Ronald Paguirigan, Viktoria Rabovskaja, Tomas Roubal, Mel Coronel San llan,
and Lluis Vinals Torres are gratefully acknowledged. Special thanks are due to Jong Hye
Rha for her valuable contribu ons to the literature research and the shaping of the fi rst
dra at the onset of this project. We also thank the colleagues from the Department of
Health Systems Governance and Financing for their useful sugges ons and feedback during
a departmental review mee ng. We thank everyone who shared her/his experience and
insights on implemen ng DRG or case-based systems during two webinar consulta ons and
a consulta on sessions at the side of the Liverpool Health Systems Research Symposium in
2018. Last but not least, we thank Agnes Soucat, Joe Kutzin and Tessa Tan-Torres for overall
inspira on, guidance and comments.
WHO gratefully acknowledge fi nancial support received from the United Kingdom
Department for Interna onal Development and the EU-Luxembourg-WHO UHC Partnership
Programme.
vi
HEALTH FINANCING GUIDANCE NO. 10
FINA
L DR
AFT
ACRONYMS AND ABBREVIATIONS
ACHI Australian Classifi ca on of Medical Interven ons
AR-DRG Australian Refi ned DRG system
BR Base rate
CBG Case-based groups
CGW Case-group weight
CMI Case mix index
DRG Diagnosis-related groups
FONASA Fondo Nacional de Salud, Chile
G-DRG Ghana DRG system
HCFA American Health Care Financing Administra on
ICD Interna onal Classifi ca on of Diseases and Related Health Problems
ICHI Interna onal Classifi ca on of Health Interven ons
MDC Major diagnos c category
NHIF Na onal Health Insurance Fund (Lithuania)
NHSO Na onal Health Security Offi ce (Thailand)
NOMESCO Nordic Medico-Sta s cal Commi ee
UCS Universal Coverage Scheme (Thailand)
UHC Universal health coverage
VSS Viet Nam Social Security
viiACRONYMS AND ABBREVIATIONS
FINA
L DR
AFT
KEY MESSAGES AND SUMMARY
viii
• A system of case-based groups (CBG) is a classifi ca on system that groups pa ent cases,
including services received, into standardized groups according to several variables – most
commonly diagnosis, treatment or procedure received, and pa ent characteris cs.
• The introduc on of a CBG system is a long-term endeavour which needs con nuous
investment and upda ng. The introduc on of a CBG system should have clear objec ves
and should be accompanied by a clear implementa on plan that includes intermediate
steps, responsibili es for the actors and involved, and melines.
• In most countries the CBG system is used to collect informa on on hospital ac vity, to inform
budget formula on/nego a on and o en, to serve as a payment system. Its largest impact
is that it clearly shows for which services and to which providers funds are allocated.
• The design and meframe for the introduc on of the CBG system have to fi t with the country
context and the capaci es of purchasers and providers. Available capaci es for coding, claim
processing, fi nancial management, management of purchaser-provider rela ons (including
contrac ng) and informa on management should be assessed.
• Providers require a degree of autonomy in the management of their funding in order to
respond to the incen ves set by a CBG payment system (e.g. by changing mix of staff or
other inputs).
• The introduc on of a new payment method is not only a technical process. The poli cal
economy around the introduc on of the system and the expected shi ing of resources will
create supporters and opponents. The Ministry of Health and purchasers should consult
relevant stakeholders, especially providers, during the development of the system.
• If it is possible to launch payment system using an imported diagnosis-related groups (DRG)
system within a few years, this is only feasible with strong capaci es of purchasers and
providers, and electronic medical records based on standardized coding systems already in
place. In most countries, adapta ons to the health management informa on system, the
purchasing modali es and adjustments to the imported DRG system will take several years.
• The development of a classifi ca on algorithm is the technically most complex task. Most
countries will start with an imported DRG grouping algorithm, or at least get some inspira on
from exis ng grouping algorithm, that will be adjusted to the country context.
• In other cases a gradual approach, star ng with a simple CBG system based on a simple
classifi ca on algorithm which is then con nuously developed, will be more realis c. Such
an approach can s ll achieve strategic purchasing objec ves while giving both fi nancial and
informa on management systems me to mature.
• Data quality is key for a DRG system but a simpler CBG system can incen vize and help
improve data collec on. Weaknesses in data collec on or data quality should not prevent a
country from introducing a CBG system.
• A CBG system is a tool to contribute to the objec ves of universal health coverage (UHC).
Incen ves set by the system should be carefully considered to ensure that integrated people-
centred care, and not economic considera ons, remains the overall objec ve. Monitoring
of impact on treatment quality is needed in addi on to claims monitoring. It is important to
note that gaming and coding creep is going to happen.
HEALTH FINANCING GUIDANCE NO. 10
FINA
L DR
AFT
HC
ted people-d peopl
jec ve. Monitoringec ve. Monitorin
monitoring. It is importaoring. It is import
INTRODUCTION AND PURPOSE OF THIS GUIDE
ixINTRODUCTION AND PURPOSE OF THIS GUIDE
More and more low- and middle-income countries are seeking to strengthen strategic
purchasing arrangements in the health sector by using evidence-based processes to defi ne
which specifi c health services should be purchased from which providers, how the services
should be paid for and at what rate they should be paid (1, 2). A purposively aligned mixed
provider payment system is one of the core policy instruments leading to more strategic
purchasing in order that diff erent payment methods will create a coherent set of incen ves to
infl uence and guide provider behaviour towards health system objec ves (3). Related thereto,
payments based on case-based groups (CBG) or, more specifi cally, diagnosis-related groups
(DRG) have gained increased interest from policy-makers across the globe. This document
focuses specifi cally on CBG and DRG. The diff erences between CBG and DRG are explained
below. In general, this publica on uses CBG as the wider umbrella term for a case-based
classifi ca on system and refers to DRG when these are meant explicitly.
When considering the introduc on of a CBG system, many policy choices and technical
ques ons arise. Policy-makers have to judge the pros and cons of a CBG system in rela on
to their country context and the health system challenges they wish to address. They will
have to explore whether, and if so how, to introduce a CBG system, and should assess
which ins tu onal arrangements and resources are available and which would be required.
Importantly, this guide on CBG does not suggest or imply that these payment methods are
a magic bullet or the only op on to consider when seeking to improve the payment system.
There is a large body of literature on CBG, and in par cular on DRG, but it focuses primarily
on high-income countries. The purpose of this guide is to explore specifi cally the CBG-related
policy ques ons and issues relevant to the context of low- and middle-income countries. This
document seeks to provide specifi c guidance and evidence to policy-makers and prac oners
on core ques ons regarding design and implementa on that they will be faced with when
introducing a CBG system. The guide is wri en in a Ques on & Answer (Q&A) format. It
provides literature references for further reading and for more technical details and presents
various country cases to illustrate specifi c issues.
The document consists of four parts:
Part 1 outlines defi ni ons, terminology and the main conceptual aspects related to CBG and
DRG.
Part 2 covers the assessment phase and highlights ques ons and issues that policy-makers
should consider before taking the decision to introduce a CBG system.
Part 3 delves into the prepara on phase by exploring policy and design aspects once a country
has decided to introduce a CBG system.
Part 4 is concerned with the implementa on phase and discusses implementa on ques ons,
requirements for system adjustments and the need for monitoring and revision in order to
iden fy and address unintended impacts of a CBG system.
Technical terms are included in a glossary at the end of the document for easy reference.
FINA
L DR
AFT
asy reference.asy reference
FINA
L DR
AFT
The following documents provide more details on CBG design and implementa on. This guide
builds on these references.
Langenbrunner J, Cashin C, O’Dougherty S, editors. Designing and implementing health care provider
payment systems: how-to manuals. Washington (DC): The World Bank; 2009 (https://openknowledge.
worldbank.org/bitstream/handle/10986/13806/48599 , accessed 20 September 2019).
Busse R, Geissler A, Quentin W, Wiley M, editors. Diagnosis-related groups in Europe: moving towards
transparency, efficiency and quality in hospitals. Maidenhead and New York (NY): Open University
Press – McGraw-Hill Education; 2011 (http://www.euro.who.int/__data/assets/pdf_file/0004/162265/
e96538 , accessed 20 September 2019).
Cashin C, editor. Assessing health provider payment systems. A practical guide for countries working
towards universal health coverage. Washington (DC): Joint Learning Network for Universal Health
Coverage; 2015 (http://www.jointlearningnetwork.org/resources/assessing-health-provider-payment-
systems-analytical-team-workbook, accessed 20 September 2019).
Annear P, Huntington D, editors. Case-based payment systems for hospital funding in Asia: an
investigation of current status and future directions. Manila: World Health Organization Regional
Office for the Western Pacific; 2015 (http://www.who.int/iris/handle/10665/208246, accessed 20
September 2019).
Bredenkamp C, Bales S, and Kahur K editors. Transition to DRG Payments for Health: Lessons from
Case Studies. International Development in Focus. Washington, D.C: World Bank; 2019.
x HEALTH FINANCING GUIDANCE NO. 10
FINA
L DR
AFT
A case-based groups (CBG) system is a
pa ent classifi ca on system that groups
pa ent cases, including services received,
into standardized case groups according
to diagnosis and treatment or procedure
received. It combines a clinical logic with an
economic logic. A CBG system can be used
to collect more detailed and/or standardized
informa on about the services provided to
pa ents, as well as in addi on as a provider
payment method (4, 5).
CBG classifi ca on systems usually have the
following features in common (5-7):
• Each case group contains cases with similar
pa erns of resource use (“resource-
homogeneous”).
• Cases within a group share common
features from a clinical perspec ve
(“clinically meaningful”).
• Each case can be classifi ed into one group
(i.e. the classifi ca on is exhaus ve).
• The use of variables to defi ne a group and
to assign a case is based on informa on
collected rou nely in medical records.
In its simplest version, there would be only
one case group – i.e. each pa ent falls
into the same group. If used for payment,
providers would receive the same amount
for every pa ent discharged. With one case
group and one payment rate, the payment
system would, however, not consider
diff erences in severity and costs across
cases (4) and would not create the desired
incen ves intended by a CBG system.
When a pa ent classifi ca on system is based
on medical special es or the chapters of
the Interna onal Classifi ca on of Diseases
(ICD) with no other variables, it would
usually consist of 25–27 groups. Addi onal
classifi ca on factors to refl ect the main
cost drivers – such as a whether surgery
or medical imaging is provided, added to
those chapters of the ICD for which they
are relevant – will ensure that groups are
economically more homogenous and lead
to pa ent classifi ca on systems with around
50 case groups in total.
As more variables are added into the
classifi ca on logic, the number of cases
groups will increase. This will also depend
on the informa on and level of granularity
available in medical records which can be
used for coding. However, the resul ng
number of groups needs to be manageable.
However, for case groups to remain
economically homogenous, a meaningful
average treatment cost has to be allocated
to each case group. This requires a certain
number of cases per group, a condi on which
would not be fulfi lled by a classifi ca on
system in which every diagnosis would be
considered as a dis nct case group (5).
A more complex and specifi c form of CBGs
are diagnosis-related groups (DRGs) in that
addi onal and more detailed informa on
on mul ple variables such as pa ent
characteris cs, procedures undertaken,
severity of the case, primary and secondary
diagnosis, comorbidi es and complica ons,
and/or type of admission are used for the
grouping.
1. DEFINITIONS
1.1. WHAT IS A CASE-BASED GROUPS SYSTEM?
WHAT IS A DIAGNOSIS-RELATED GROUPS SYSTEM?
1DEFINITIONS
2 HEALTH FINANCING GUIDANCE NO. 10
In general, a DRG (payment) system
comprises:
– a pa ent classifi ca on system based
on the set of several variables to defi ne
groups,
– a classifi ca on algorithm, which is the set
of instruc ons for assigning a par cular
case to a specifi c group according to the
pa ent classifi ca on logic, and
– usually a specialized so ware – called
“grouper so ware” – for the digi zed
case assignment.
Based on the number of variables used for
classifying pa ent cases, the number of case
groups range from several hundreds to over
2000 cases (4, 5, 8).
DRG systems vary primarily according to
the classifi ca on variables they use and
the grouping rules applied, such as which
diagnoses are classifi ed into the same
group or how the classifi ca on steps are
sequenced (9-11). Sec on 3.2 outlines which
classifi ca on variables are frequently used.
The diff erent DRG systems are referred to
as DRG variants or models and have been
given names (e.g. “Australian DRG”). A DRG
variant represents a specifi c classifi ca on
system, including a related algorithm and,
if applicable, a grouper so ware. Most DRG
variants exist in several versions that refl ect
the regular updates of the classifi ca on
algorithm. As countries regularly update
their DRG systems, the consecu ve versions
of a country variant can diff er signifi cantly
depending on the scope of the revision.
Sec on 3.4 presents the main DRG variants
in place around the globe.
The dis nc on between CBG and DRG
systems is not clear cut, rather exis ng
systems show a range of diff erent levels of
complexity. Figure 1 shows the con nuum
ranging from a simpler CBG system to a
DRG system. It stretches from a (theore cal)
system with only one case to more complex
systems with over 2000 case groups defi ned
through several variables (e.g. major
diagnosis, surgery provided or not, severity,
medical procedures, pa ent characteris cs
and comorbidi es). Such more complex
systems are manageable only by using
electronic pa ent records and grouping
so ware. These more complex systems are
referred to as DRG systems.
Fig. 1. Overview of the level of complexity of defi ni on of cases, number of groups and informa on
requirements for case-based group systems and diagnosis-related group systems
FINA
L DR
AFT
Case-based groups systems
3DEFINITIONS
FINA
L DR
AFT
A CBG system assigns a rela ve case-group
weight to each case group. This case-group
weight refl ects the resources needed to
diagnose and treat a case in a respec ve
case group compared to the average
cost of diagnosing and trea ng across all
cases. Alterna vely, as is the case in some
countries, a direct monetary value is given
to each case group.
The base rate is the average rate paid per
case. The payment for a case is the product
(mul plica on) of the base rate and the
respec ve rela ve case-group weight.
The base rate is the same for all cases and
usually links the CBG payments with the
overall available budget for the providers
and services paid by the CBG system (4).
The case mix is the sum of the rela ve case-
group weights of all cases treated in a health
facility during a given period.
The case mix index (CMI) is the case mix
for a given health facility during a given
period divided by the total number of cases
(classifi ed under the CBG system) treated
by this health facility during the given
period. It thus is the weighted average of
the case-group weights of all cases treated
in that facility. The CMI enables ac vity and
the level of its complexity to be compared
across health facili es. It can also be used
to determine global budget alloca ons to
health facili es (4, 12).
In general, the CMI of lower-level facili es
is lower than that of higher-level facili es,
such as ter ary care and university hospitals.
University or ter ary hospitals providing a
higher share of complex services would have
a CMI well above 1 (12, 13).
In a case-based payment system, providers
are paid on the basis of case groups (“case-
based groups”) with a fi xed fi nancial amount
per case discharged. This means that
payments for providers are linked to ac vity
(4, 14).
The payment for the cases is, in principle,
determined by a formula consis ng of:
– rela ve case-group weights which
indicate the rela ve cost-intensity of
cases of diff erent groups,
mul plied by
– a base rate,
mul plied by
– adjustment factors which allow one to
take into considera on the diff erences
in the economic or hospital context
(e.g. remoteness, teaching facility)
in which providers operate or the
sociodemographic or socioeconomic
profi les of the pa ents (6, 15).
1.2. WHAT ARE CASE GROUP WEIGHTS, THE BASE RATE,
CASE MIX AND CASE MIX INDEX?
1.3. HOW IS THE PAYMENT DETERMINED IN A CASE-BASED
PAYMENT SYSTEM?
Annex 1 outlines country examples with various classifi ca on systems, showing how diff erent
cases are allocated to groups and thus indica ng diff erent levels of complexity.
4 HEALTH FINANCING GUIDANCE NO. 10
Some countries started to set up their case-
based payment systems without using case-
group weights, base rates or adjustment
factors. Instead, they allocated an amount
to each defi ned case. This has been the case,
for instance, in Ghana, Morocco and the
Philippines in the beginning (16-18).
Figure 2 presents the core elements for
determining a DRG-based payment rate.
Figure 2. Formula for calcula ng the rate of a CBG or DRG case group
Source: Reproduced from reference (9).
FINA
L DR
AFT
The ul mate objec ves of UHC encompass
u liza on in line with need (i.e.
equitable access), quality in health care
and fair fi nancing, while intermediate
UHC objec ves relate to effi ciency,
equitable distribu on of resources, and
transparency. However, many countries
face challenges due to, inter alia, ineffi cient
use of funds (e.g. over-provision, under-
provision, inappropriate size of facili es or
insuffi ciently priori zed health services),
inadequate quality of service provision,
and a lack of transparency.
Each payment method, and the combined
mix of payment methods, sets incen ves
for providers and in turn have a posi ve
or nega ve impact on these objec ves
(19, 20). A summary of diff erent payment
methods and their characteris cs is
included in Annex 2. This sec on examines
the incen ves that a CBG payment system
creates and ways in which it can contribute
to the fi nal and intermediate objec ves of
UHC.
In general, CBG payments set two main
incen ves: 1) providers may reduce the
inputs (and hence costs) per case treated,
and 2) they may try to increase the number
of cases, both of which helps them to
increase their income.
With regard to the fi rst incen ve, a CBG
payment system can improve effi ciency
in the use of funds by reducing, or even
avoiding, overtreatment and unnecessary
services while maintaining quality, since
healthcare providers have an incen ve to
reduce the amount of inputs per case. A
certain degree of autonomy of providers
will be needed to incen vize them to
manage their resources more effi ciently
and to allow them to reallocate savings.
On the other hand, this incen ve makes
providers reduce treatment inputs, thus
poten ally also leading to under-provision
per case or to reduced quality of care
(skimping). Moreover, providers may
engage in cream skimming of pa ents with
less severe symptoms, refer complex cases,
or discharges pa ents too early in order to
reduce inputs needed for treatment and
hence costs (4, 20, 21).
With regard to the second incen ve above,
CBG payments can induce providers to
increase the number of treated cases (12,
22), for example by readmi ng pa ents or
unbundling cases into individual services
to in order to claim for more cases.
Nonetheless, an increase in case numbers
can be desirable when the hospital
u lisa on rates for certain services are
very low or when there are wai ng lists.
2. ASSESSMENT PHASE
2.1. CAN A CBG PAYMENT SYSTEM CONTRIBUTE TO
ACHIEVING UHC OBJECTIVES?
5ASSESSMENT PHASE
FINA
L DR
AFT
6 HEALTH FINANCING GUIDANCE NO. 10
Moreover, through upward adjustment
of case-group weights for selected cases,
resul ng in rela vely higher payment rates
for such cases, a CBG payment system
can be designed to incen vize providers
to priori ze certain health services and
to shi a en on to the needs of specifi c
popula on groups or geographical areas.
By paying for outputs and alloca ng funds
to ac vi es, a CBG payment system may
reduce the funds of providers whose
ac vity is too low to be economically
viable. As a result, the payment system
can result in a restructuring of the provider
sector as providers with insuffi cient ac vity
may have to close down, merge with other
providers, or adjust their service delivery
structures (4).
Furthermore, a CBG system collects
standardized informa on on hospital ac vity
which can improve transparency on the
types and volume of health services – such as
diagnoses, procedures, severity indicators
and pa ent characteris cs – delivered by
each hospital. When aggregated across
all hospitals, a CBG system can provide
data on the epidemiological profi le of a
popula on, can indicate which popula on
groups and health services are the focus
of hospital ac vity and can show how the
budget is distributed across hospitals (21,
23). The informa on collected through a
CBG system can also serve to measure and
compare performance across providers (e.g.
in rela on to cost of treatment or average
length of stay). This informa on can be used
strategically to encourage improvements
in quality of care or to review pa ent care
pathways systema cally (21, 24).
In sum, it is important to note that a CBG
system, whether for informa on collec on
and monitoring and/or for payment, is
not a magical solu on to health system
performance challenges. A careful
assessment is needed to determine whether
the introduc on of a CBG system is most
appropriate or whether another provider
payment reform might be more suitable to
address a country’s performance issues in a
given context.
However, when well designed and
implemented, a CBG (payment) system
can contribute to the UHC objec ves (2).
Nevertheless, monitoring and control
mechanisms are needed so as to avoid that
the incen ves set by a CBG payment system
lead to provider behaviour and hence
consequences that are not conducive to
UHC, i.e. increased overall expenditure,
increased service provision beyond the
desired level and reduced quality (4,
20, 21). More informa on on how the
incen ves set by a CBG payment system
and its consequences can be addressed is
provided in Sec on 4.3. Table 1 summaries
the poten al eff ects of CBG introduc on
on cost-containment, effi cient use of
resources, be er informa on collec on
and administra ve costs, depending on
the payment system in place prior to CBG
introduc on. Evidence suggests that a mix
of provider payment methods is needed
to balance mul ple policy objec ves (e.g.
quality of services and cost containment
(20).
FINA
L DR
AFT
7ASSESSMENT PHASE
Impact of CBG
introduc on when
shi ing from this
payment method
Cost containment
(Sec on 3.8
provides more
details)
Effi cient use of
resources
Informa on
collec on
requirements
(Sec on 3.2
provides more
details)
Administra ve
requirements
(Sec on 2.2 and
2.3 provide more
details)
Fee-for-service
Without a volume
or budget cap, CBG
will not lead to cost
containment
Fixed rate of CBG
payment will
reduce unnecessary
services which leads
to more effi cient
use of resources
CBG requires only
informa on to
iden fy the case,
no informa on
on every service
provided is required
Unit of payment
changes from
every service
to treatment
episode, simplifying
payments
Global budget (not
based on ac vity
– e.g. historical
budget alloca ons,
popula on-based)
Depending on the
budge ng formula,
CBG payments
might change
the incen ves for
effi cient use of
funding/ change to
more performance-
based budgetsCBG
payments
would allow
standardiza on
of payment rates
per case across all
providers
CBG system
will require
more detailed
informa on on
services provided
CBG system
increases
administra ve
burden, requiring
management and
verifi ca on of
claims
Line-item budget/
payment for inputs
CBG payments
may encourage
the reduc on of
unnecessary inputs,
especially if line-
item budgets were
based on u liza on
in the previous year
CBG system
will require
more detailed
informa on on
services provided
but might lead to
less informa on on
inputs used
Table 1. Poten al eff ects of CBG introduc on on cost containment, eff ec ve use of resources,
informa on collec on and administra ve effi ciency in comparison to the previous payment method
Source: Authors’ compila on based on references (8, 20, 25).
For further reading, please see:
Cashin C, editor. Assessing health provider payment systems. A practical guide for countries working
towards universal health coverage. Washington (DC): Joint Learning Network for Universal Health
Coverage; 2015 (http://www.jointlearningnetwork.org/resources/assessing-health-provider-payment-
systems-analytical-team-workbook, accessed 20 September 2019).
FINA
L DR
AFT
8 HEALTH FINANCING GUIDANCE NO. 10
The introduc on of a CBG system is o en
part of a larger health-fi nancing reform or
a package of measures to make purchasing
more strategic. Country experience shows
that clear and realis c objec ves are needed
for CBG introduc on, as well as a detailed
implementa on plan with clear melines.
Evidence also suggests that a health system
that is fi nanced predominantly from
public sources facilitates the launching of
such payment reforms (4, 5) Moreover, a
single-payer model or a harmonized and
standardized CBG payment system among
mul ple purchasers augments purchasing
power and will allow for a stronger leverage
of the CBG system in order to shape
provider incen ves and behaviour (7).
Beyond these conducive factors, there are
cri cal ins tu onal requirements that need
to be in place, as outlined below.
A CBG system requires detailed pa ent-
level informa on collected through an
informa on management system as well as
ins tu onal capacity to use such a system.
At best, a rou ne data collec on system,
including medical records, should exist; if
not, it needs to be developed (see Sec on
3.2 on what data need to be collected).
In fact, collec ng more detailed and be er
data on pa ent cases and the treatments
provided, or the introduc on of pa ent
records, can be one objec ve of introducing
a CBG system. In this case, the CBG system
and the envisaged, more comprehensive
health informa on management system can
be developed jointly, growing more complex
over me and mutually suppor ng their
further development (6, 14). Importantly,
an informa on management system
involves more than data collec on; it also
encompasses data analysis, verifi ca on,
valida on and audi ng (26). Data audi ng
is cri cal because there are likely to be
both inten onal and uninten onal coding
mistakes in most CBG systems (27). The
purchaser needs to be able to trust the
quality of data coding for otherwise it may
be reluctant to use the system for payment.
A thorough assessment of the exis ng
informa on management system will
reveal the current set-up and func oning
as well as poten al gaps. This is the basis
on which to further develop and adapt the
informa on management system to the
data needs of a CBG system.
Foremost, the health informa on
management system needs to be based on
standardized coding of pa ent cases (27,
28). For example, for a CBG system relying
on diagnosis as a criterion for grouping,
the Interna onal Classifi ca on of Diseases
(ICD) – and/or country-specifi c adapta ons
of it – can serve to provide a standardized
coding format.
2.2. WHAT ARE THE CONDUCIVE FACTORS AND
INSTITUTIONAL REQUIREMENTS FOR A CBG SYSTEM?
An eff ec ve informa on management system for hospital data
9ASSESSMENT PHASE
Health staff – i.e. doctors and nurses –
should have the capacity for rou ne coding
of pa ent cases, including diagnoses and
procedures. In many countries with a
CBG system, providers therefore employ
specifi c CBG or DRG coders. These staff are
trained in this specifi c competence and are
responsible for transla ng medical records
into the coding format required by the
CBG system (29). Moreover, the capacity
of health facility managers needs to be in
line with the (increased) level of autonomy
they are granted, which should accompany
the shi from input-based to output-based
payment. Moreover, the administra ve
staff need the capacity to collect, analyse
and make decisions about cost informa on
related to service delivery in order to
allocate resources across input categories
and medical departments (21).
At the purchaser level, capaci es for claim
management and verifi ca on are essen al
in order to ensure that claims comply with
relevant regula ons and that providers
receive payments on me. A purchaser
also needs the skills to design and adjust
the incen ves of the CBG payment system
in order to align provider behaviour with
health-system objec ves. Together with the
regulators, the purchaser also needs to be
able to monitor any poten ally unintended
consequences of CBG payments. In addi on,
the classifi ca on and coding procedures
need to be standardized, regulated and
enforced. This should be accompanied
by monitoring measures to encourage
improvements in the quality of coding and
regular updates. This is frequently under
the responsibility of the Ministry of Health,
which needs capacity to maintain the health
informa on management system and to
use data strategically. Ministries of health
or purchasers have o en created new units
with exper se in CBG coding, classifi ca on
and cos ng (26, 27). Those units will play
a role in con nuously upda ng the system,
as noted under Sec on 4.5.
When public funds are allocated to providers
via a CBG payment system, it is important
that public fi nancial management rules
allow output-based payment methods to be
used. If this is not the case, modifi ca ons in
exis ng budget formula on and execu on
rules may be needed (30). Providers also
need suffi cient fi nancial and managerial
autonomy and fl exibility in the use of funds
to allow them to respond to incen ves
set by the payment method. Provider
autonomy should go along with strong
accoun ng and audit systems to ensure
accountability in the use of funds (4, 6).
An account of the importance of provider
autonomy in Mongolia’s introduc on of a
CBG system is contained in Box 1.
Adjustments in public fi nancial management rules
Strong technical capacity of purchasers, providers and ministries
of health or other relevant regula ng agencies to manage the CBG
system
FINA
L DR
AFT
Box 1. The importance of provider autonomy to accompany the introduction of a
CBG system in Mongolia
Since it was put in place the Mongolian health insurance scheme has used a fl at case
payment rate for all inpa ent cases diff eren ated by the levels of health facili es. In
2006 CBG was introduced payments using 22 case groups and extended to 115 groups
in 2010. Ini ally, this switch to more refi ned output-based payment method was not
accompanied by any change in budge ng or in budget execu on rules. Both the Ministry
of Finance and the Ministry of Health have supervised providers through 1) appointment
of their execu ve managers; 2) annual budget cap based on historical line-item budget
planning for a part of provider’s revenue; 3) quotas on hospital staff ; 4) line item-based
budget execu on plans with limited fl exibility to shi funds across items; and 5) other
bureaucra c procedures on residual claimant – i.e. responsibility for covering poten al
losses and using poten al surpluses (31). Since providers lack managerial and fi nancial
autonomy to use or reallocate effi ciency gains through CBG payments, there is no strong
incen ve to reduce inputs per case.
In 2013, Mongolia’s new Integrated Budget Law signifi cantly changed public fi nancial
management regula ons. In par cular, the number of major line items in the budget
used for planning and repor ng was reduced from 38 to 5. Providers were given more
fl exibility to make realloca ons across budget line items and the maximum amounts that
they can use fl exibly without seeking prior approval have been increased. Since 2016,
hospital boards, which were set up in 2011, have had the authority to approve spending
plans and to appoint the execu ve management of public hospitals. It is expected that
these reforms in hospital autonomy, combined with gradual capacity-strengthening of
hospital managers in developing, implemen ng and repor ng on output-based budgets,
will create the space for the CBG payment system to set effi ciency incen ves and also
lead to stronger buy-in by providers into the new payment system.
Tsolmongerel Tsilaajav
FINA
L DR
AFT
10 HEALTH FINANCING GUIDANCE NO. 6
The fi rst country to develop and introduce
a DRG system was the United States of
America, with the process taking more
than 10 years before the system was ready
to be used for payments. Therea er, and
based on this experience, the rollout took
5–10 years in several European countries.
Importantly, ICD coding was already in
place both in the USA and Europe. Figure
3 gives an overview of how long it took to
start using CBGs for payment in various
countries (28). In most countries, a CBG
system has been introduced in phases with
gradual extensions, as further detailed
in Sec on 4.2. Even under the most
conducive condi ons, the introduc on
is likely to take several years (5). Box 2
describes the progressive development
and implementa on of DRGs over several
years in Chile.
2.3. HOW LONG DOES IT TAKE TO DEVELOP AND IMPLEMENT
A CBG SYSTEM? WHAT ARE THE RELATED COSTS?
FINA
L DR
AFT
11ASSESSMENT PHASE
Moreover, a CBG classifi ca on system and
related cost weights or payment rates need
con nuous adjustments and updates to
ensure that the grouping remains clinically
meaningful and economically homogenous
and that it refl ects new health technologies.
The introduc on of a CBG system is not a
one-off project (32). The issue of regular
upda ng is further discussed in Sec on 4.5.
The cost of the introduc on and design
process of a CBG system depends on many
factors and is diffi cult to es mate. No such
informa on was found in publicly available
sources. However, the cost implica ons of
the following aspects need to be considered,
apart from resources needed for the policy
dialogue and high-level planning:
When buying (impor ng) a DRG system:
– costs of buying a classifi ca on system,
including the grouping algorithm and
the grouping so ware in the case of a
DRG system;
– cost for adjus ng the imported DRG
system to the country context;
– ini al and ongoing licensing cost, where
required.
When developing a CBG system:
– costs of developing an own classifi ca on
system and the corresponding grouping
algorithm and grouping so ware.
In addi on, there are:
– costs for introducing/adjus ng the
facili es’ informa on management
systems;
– costs for introducing/adjus ng the
informa on management systems of
purchasers;
– costs for data collec on and cos ng
studies to inform payment rates and
rela ve cost weights under the CBG
system;
– costs for training staff of providers and
purchasers;
– costs of fi nancial incen ves that may be
needed to obtain ini al support from
providers or to encourage complete
repor ng (especially when the CBG
system is not used for payment);
– costs for communica on and advocacy
to providers and the wider public.
Figure 3. Year in which a country started moving towards DRGs and year in which the system was
used for payment, for selected countries
Source: adapted from reference (28).
FINA
L DR
AFT
12 HEALTH FINANCING GUIDANCE NO. 10
Box 2. DRGs in Chile: a case of progressive implementation
In Chile a fi rst DRG pilot, focusing on informa on collec on and analysis of case mix,
started in 2002 in four major hospitals with technical support from the Catholic University
of Chile. The pilot highlighted the relevance of having a strong hospital informa on
management system, the need to agree on and collect the necessary minimum set of
data and the importance of designa ng and training coding teams for data entry.
From 2007 onwards, the Ministry of Health and FONASA (the na onal health fund,
opera ng as a health insurance type purchaser) were evalua ng the possibility to use
DRGs as hospital payment method. Star ng in 2010, the Interna onal Refi ned (IR)-
DRG variant was gradually expanded to 25 public hospitals for informa on collec on.
Hospitals developed their coding capacity, implemented DRG so ware and applied a
standardized system for collec on of cost informa on. Case weights and calcula on
of the base rate have been developed and con nuously revised. In 2017, accoun ng
systems in public hospitals had matured suffi ciently to provide necessary informa on for
a detailed cos ng study to re-evaluate both rela ve case group weights and the base
rate.
Star ng in 2015, FONASA uses fl exible contrac ng mechanisms to pilot DRG payments
for private providers. FONASA decided to buy bed days from the private sector for a set
of pathologies for which there are long wai ng lists in the public sector. It expressed
those pathologies in DRGs. As cost structures diff er between public and private providers,
there are diff erences in DRG rates between the public hospitals, with their rates being
based on cos ng studies, and the private sector, with rates being based on a tender
process. However, funding channelled through DRG to private providers accounts for
less than 1% of FONASA’s total budget.
The implementa on of DRG payments in Chile has been slow and the DRG system is s ll
mainly used for informa on collec on. FONASA has announced its plan to extend DRG
payments in 2020 to public hospitals.
Camilo Cid and Gabriel Bas as
FINA
L DR
AFT
13ASSESSMENT PHASE
A range of diff erent stakeholders will be
involved and/or infl uencing the design and
implementa on of a CBG system. First, the
parliament and the whole of government
will be involved if changes to legisla on are
needed. The Ministry of Finance defi nes
public fi nancial management rules, which
also apply to the CBG payment system,
and has an interest in protec ng fi nancial
sustainability. Interest groups such as
provider associa ons, pa ent associa ons
or medical associa ons are o en consulted
formally on health policies but may also
try to lobby informally to ensure that their
interests are refl ected (33, 34). The technical
nature of the introduc on of a CBG system
usually limits ac ve involvement in the
development process to the Ministry of
Health, the purchasing agencies, provider
associa ons and medical associa ons.
The Ministry of Health and/or the
purchasing agency are usually the drivers
for the development of a CBG system.
The Ministry of Health, as steward of the
health system, is interested in improving
data collec on and in ensuring that CBG
payments set the right incen ves for
providers to align their behaviour with
health-sector objec ves and UHC goals.
For the purchasing agency, a CBG payment
system serves to increase effi ciency in the
use of funds, to contain costs and to gather
be er informa on about the services and
providers on which the funds are spent
(35). Addi onally, high-level leadership by
the Ministry of Health will be required in
order to overcome poten al opposi on
and to support the development of a CBG
system that is acceptable to all stakeholders
(35). Country experience shows that a
specifi c unit, or at least a designated team,
should be set up to take charge of CBG
development. Over me a mandated unit
that is able to maintain the system will be
equally important (28).
The introduc on of a CBG payment system
implies (re-)alloca on of funds and/or
changes to how decisions on fund alloca on
are taken. Some facili es may receive more
funds than before while others may receive
less. As a result, some providers may be
cri cal of the introduc on of a CBG payment
system. However, diff erent providers will
have diff erent interests. Hospitals with a
complex case mix may be more suppor ve
of a CBG system than rural hospitals
providing a low volume of services. It is
important that providers are consulted
and involved during the design process and
that the classifi ca on logic is transparent
(29, 35). To encourage provider buy-in, the
purchaser and the Ministry of Health should
ensure that provider perspec ves are taken
into account, provider concerns about
poten al losses of revenue are discussed
and transi onal measures are considered
(see more details in Sec on 4.2 on phasing
in the CBG payment system). Provider trust
in the payment system and a feeling that it
leads to a transparent and fair distribu on
of resources are equally important (14).
External experts, agencies providing
technical assistance and private sector
developers of CBG systems or so ware
might also contribute to the development
process and it is important that their eff orts
are aligned with the government agenda.
2.4. WHICH STAKEHOLDERS ARE LIKELY TO INFLUENCE THE
DESIGN AND IMPLEMENTATION OF THE CBG SYSTEM
AND HOW SHOULD THE DIFFERENT STAKEHOLDERS BE
INVOLVED?
14 HEALTH FINANCING GUIDANCE NO. 10
CBG systems are based on the classifi ca on
of health services into clinically and
economically homogenous groups. For that
ma er, most CBG systems ini ally apply
to acute inpa ent care in hospitals. Acute
inpa ent care services are rela vely easier
to classify as cases can be defi ned as all
diagnoses, tests and treatments provided
between admission and discharge. Also,
inpa ent services usually account for a
large share of health sector spending (6,
36).
In contrast, it is diffi cult to apply a CBG
system to health services for which
diagnosis or other grouping criteria are
not good predictors of treatment costs.
This is the case, for instance, for long-term
psychiatric or rehabilita on care or when
the number of cases is too small to calculate
a meaningful average cost for cases within
a specifi c case group (e.g. for organ
transplants). Likewise, for intensive care
or cases requiring expensive therapeu cs
and medicines, it is not the diagnosis
that drives the cost but the intensive care
stay or the price of medicines. Most CBG
systems therefore do not apply to these
types of services or they provide addi onal
payments for such outliers, as noted under
Sec on 4.3 (22, 37).
Several high-income countries have
extended CBG payments to some outpa ent
care services. However, outpa ent health
care services are in most cases paid through
fee-for-service, line-item budgets, global
budgets or capita on payments. This is
because cost diff erences between diff erent
primary health care and/or outpa ent
services are less stark, thus less in line with
the underlying logic of CBGs. Using CBGs
for high-volume and low-cost services
could also imply dispropor onately high
administra ve costs (20, 38).
FINA
L DR
AFT
A CBG payment system that aims at cost
control or restructuring of the hospital
sector should include those health facili es
that consume a signifi cant share of health
service cost. Inpa ent care, and especially
surgical cases, usually consume a large
part of the budget. Primary care facili es
provide health services that are less
resource-intense. As a result, most CBG
payment systems apply to secondary and
ter ary care facili es (4, 20).
3. PREPARATION AND DESIGN PHASE
3.1. WHICH TYPES OF HEALTH SERVICES, FACILITIES
AND PROVIDERS SHOULD BE INCLUDED IN THE CBG
PAYMENTS?
Which types of health services should be part of the CBG payments?
Which types of facili es should be included in the CBG payment?
15PREPARATION AND DESIGN PHASE
In a number of low- and middle-income
countries, public fi nancing for health
services can be spent only on government
health-care providers. However, the
transparent alloca on of CBG funds to
providers could create the necessary trust
and accountability to make it possible to
pay private providers using public funds.
Addi onally, the inclusion of private
providers in a CBG payment system will allow
for the systema c collec on of informa on
on their services. It may be necessary
to adjust CBG payment rates for private
providers, especially when government
providers con nue to receive some budget
alloca ons (e.g. for health worker salaries)
along with the CBG payments (21, 29). In
addi on, through selec ve contrac ng,
purchasers may choose to pay only selected
public or private providers through CBG
payments.
The types of facili es in a hospital market
can be very heterogeneous, ranging from
local hospitals in remote areas to rather
urban-based secondary and ter ary
hospitals in the public and private sector
that provide high-technology services.
Decisions on the design of the CBG system,
including which providers to cover, should
be examined cri cally with regard to the
poten al impact on, and alignment with,
overall priori es in equitable access to
services and fi nancial coverage.
Figure 4 gives an overview of the range of
aspects to further consider when deciding
which types of services, which types of
facili es and which types of providers
(private/public) to include in the CBG
system.
FINA
L DR
AFT
Which types of providers should be included in the CBG payment?
Figure 4. Aspects to consider when deciding which types of services, facili es and providers should
be included in a CBG system
Aspects to consider regarding…
Source: compila on by authors based on references (5, 20, 37)
Which types of facili es should be covered?
Ter ary facili es Secondary care facili es Primary health care facili es
– Can research and teaching ac vi es
be adequately refl ected in the CBG
payment formula?
– Can high cost treatment be
adequately covered through CBG?
– Will CBG only cover general
inpa ent treatment while high cost
treatment and/ or teaching ac vi es
will be paid by diff erent methods?
– Are secondary care facili es
providing services which are major
cost drivers?
– Are secondary care facili es the
main providers of inpa ent care?
– Does the system go beyond pilo ng
and tries to cover the largest share
of inpa ent treatment provided?
– Is the administra ve burden
jus fi able and manageable for
facili es which provide high volume
and low cost services?
– Are suffi cient staff and capacity to
code available at primary health
care facili es?
Which types of services should be covered?
– Can services be grouped into clinically and
economically homogenous groups?
– Can cost of services be reasonably predicted by (a
combina on of) diagnosis, procedure and/ or other
pa ent characteris cs?
– Can a payment rate be (rela vely easily) established?
Which types providers should be included?
– Can the provider deliver priority health services?
– Is equitable access enhanced?
– Are providers subject to regula ons and monitoring
that allow to address concerns around pa ent safety?
– Is capacity to code and report along a CBG payment
system available?
FINA
L DR
AFT
16 HEALTH FINANCING GUIDANCE NO. 10
The star ng point for building a CBG
classifi ca on system, par cularly when
it is to be used for payment, are the key
determinants of resource consump on –
generally diagnoses and procedures – as
classifi ca on variables (15). These and
other variables that are commonly used
for classifi ca on are explained below. The
actual order of sequence depends on the
CBG system and its specifi c classifi ca on
algorithm.
Diagnosis: Most CBG systems start by
grouping cases on the basis of principal
diagnosis into major diagnos c categories
that correspond to organ system or
cause/e ology and that are generally
associated with medical special es. In
general, classifi ca on systems dis nguish
fewer than 30 major diagnos c categories
(5). Major diagnos c categories mostly
follow ICD chapters. Annex 3 provides two
examples of DRG variants with their major
diagnos c categories.
Coding of principal diagnoses (as well as
of secondary diagnoses, see below) can
be based on ICD-10, an earlier ICD version,
or ICD-11 (which was adopted by the
World Health Assembly in May 2019) or a
country-specifi c adapta on. The diagnosis
will provide the most accurate informa on
on a popula on’s disease burden. However,
diagnoses are not always good predictors of
cost because resource needs for treatment
tend to depend primarily on the procedures
used rather than on the diagnoses (5).
Some CBG systems start their classifi ca on
process by iden fying cases that are both
highly specialized and high-cost and assign
them to a so-called “pre-MDC group”
(where “MDC” refers to “major diagnos c
categories” into which the other cases will
then be categorized). The pre-MDC cases
are o en paid by other payment methods
since it is diffi cult to classify them in
economically homogeneous groups (5).
Secondary diagnoses: Secondary diagnoses
refl ect comorbidi es and complica ons
that are used in an addi onal grouping
step because they also have a signifi cant
eff ect on treatment cost. In addi on, some
classifi ca on systems dis nguish between
levels of severity within a diagnosis. DRG
systems include comorbidi es for coding,
but this adds complexity and requires
providers to have stronger coding capaci es
and purchasers to be able to check coding
accuracy (5, 14). Several high-income
countries have also added the func onal
status of pa ents as a classifi ca on variable1
(39).
Procedures: Most CBG systems also
include procedure codes because several
procedure or treatment op ons may exist
for a diagnosis. Procedure and treatment
op ons are equally decisive for treatment
costs. The principal diagnosis combined
with a procedure can be a good start for
the development of a simple CBG system
which can then be gradually developed
further. The simplest and most common
3.2. WHICH VARIABLES CAN BE USED FOR CLASSIFICATION
AND WHAT ARE THE IMPLICATIONS?
1 For coding of func onal status the Interna onal Classifi ca on of Func oning, Disability and Health (ICF) provides a
standardized coding format. The WHO Disability Assessment Schedule 2.0 (WHO-DAS 2.0) provides an ICF-based, generic
instrument for func onal status assessment. The WHO-DAS 2.0 has also been incorporated into the ICD-11 sec on on
func oning.
FINA
L DR
AFT
17PREPARATION AND DESIGN PHASE
diff eren a on is between medical and
surgical cases (5). Current procedure coding
is o en country-specifi c (5). The Beta-2
dra of the Interna onal Classifi ca on of
Health Interven ons (ICHI) was launched
by WHO in 2019. It will provide guidance
and orienta on for coding interven ons in
an interna onally comparable manner (40).
Pa ent characteris cs: Pa ent
characteris cs can serve as addi onal
classifi ca on variables. Pa ent age is the
most important pa ent characteris c to
aff ect treatment cost. Case-group weights
or payment rates can be diff eren ated for
children and adults undergoing procedures
for which age has diff erent cost implica ons.
Sex as a pa ent characteris c is primarily
used as a quality check to ensure that
sex-specifi c diagnoses and procedures
are reimbursed only for pa ents of the
respec ve sex. Moreover, with regard
to promo ng equitable access to health
services, cost weights or payment rates
could be diff eren ated for pa ents from
poor or marginalized popula on groups (5).
Administra ve variables related to
admission and discharge: Cases can
further be diff eren ated by admission
type, such as inpa ent versus day case/
outpa ent or elec ve versus non-elec ve
surgery. Another op on is to diff eren ate
by discharge type, such as discharge,
referral or death (5).
Variables related to resource consump on:
The case classifi ca on may consider
variables that indicate a higher resource
consump on within a case. Examples
include length of stay, use of mechanical
ven la on, or a stay in a specialist
department (e.g. intensive care unit,
psychiatric ward) or use of high cost drugs
(5).
Case classifi ca on in a DRG system is
generally based on the following fi ve main
steps (5):
1. Before the classifi ca on begins, the
data are checked to exclude cases with
incorrect or missing informa on.
2. Very high-cost and highly specialized
cases (e.g. organ transplants) are
iden fi ed and put into a special “pre-
MDC” category.
3. Cases are allocated to mutually exclusive
MDCs according to the principal
diagnosis. A few systems use other
variables such as age (e.g. to assign cases
to a neonatal major diagnos c category).
4. The grouping algorithm classifi es
pa ents into a “surgical” or “medical”
par on. Some systems include
addi onal variables with other relevant
non-surgical procedures, such as medical
imaging.
5. All DRG systems check for further
characteris cs of the case, such
as complexity of the principal and
some mes secondary diagnoses, type
of procedures, combina ons of both,
and some mes age, length of stay or
treatment se ng, or type of discharge.
This serves to assign the pa ent to the
fi nal case group.
However, sequencing of variables varies
across DRG variants. Annex 1 provides
simplifi ed overviews of diff erent
classifi ca on systems. When selec ng
the variables for the case classifi ca on
system, it is important to consider what
poten al incen ves these variables create
for providers. If a certain diagnosis can
be treated by more than one procedure,
providers will have an incen ve to use
the procedure with the highest return for
the hospital. A descrip on of how Ghana
developed its pa ent classifi ca on system
is contained in Box 3. Box 4 illuminates the
process of iden fying DRG classifi ca on
variables in Viet Nam.
18 HEALTH FINANCING GUIDANCE NO. 10
Box 3. Development of Ghana’s patient classification system
The Ghana Na onal Health Insurance Authority (NHIA) developed a CBG system –
called the “Ghana DRG system” or G-DRG – to be used for payment, star ng in January
2007. A team of experts from the Ghana Health Service and health workers – including
doctors, nurses, midwives and pharmacists – took approximately six months to develop
the classifi ca on algorithm. The G-DRG has been applied for payment in all facili es
receiving payments from the NHIA since April 2008.
The development of the classifi ca on system was based on the following steps:
1. The list of the principal diagnoses to be covered by the new payment system were
grouped into major diagnos c categories (MDCs). Special es refl ected in the MDCs
were: adult medicine, paediatrics, adult surgery, paediatric surgery, ear nose and
throat, obstetrics and gynaecology, den stry and ophthalmology.
2. Within these MDCs, diagnoses were further grouped according to whether a surgical
procedure is required or not.
3. For diagnoses needing a surgical procedure, further grouping steps were introduced,
namely:
• grouping according to the major organ system if applicable and
• grouping according to the complexity and type of procedures to be performed.
4. Diagnoses which did not need a surgical procedure were grouped by principal
diagnoses based on major organ systems.
5. The obtained case groups were further fi netuned by considering similari es of resource
use in providing care to the pa ents in order to create economically homogenous
groups.
The result was the G-DRG system composed of a manageable number of 546 inpa ent
DRGs which capture most inpa ent cases treated in Ghanaian health facili es.
Source: reference (41)
FINA
L DR
AFT
19PREPARATION AND DESIGN PHASE
Box 4. Process for identification of relevant DRG classification variables in Viet Nam
The Viet Nam Social Security (VSS) pays providers mainly through fee-for-service. The
defi cit of the health insurance fund increased by more than 15 mes between 2005 and
2009. In addi on, fee-for-service payments created high administra ve costs for both
the VSS and health care providers.
Therefore, in 2012, the Ministry of Health decided to launch a pilot project in all provincial
and district hospitals – which were all secondary care hospitals – in Ninh Binh province
to assess how the VSS can pay providers using DRG payments. The pilot DRG system was
based on the Thai DRG variant. The pilot served to fi netune this classifi ca on system to
the Vietnamese context, to determine which classifi ca on variables should be used for
grouping cases, and to develop related case-group weights. The result was a Viet Nam-
specifi c classifi ca on system and grouping algorithm with around 400 DRGs. These
are defi ned through principal diagnosis, procedures, comorbidity and pa ent clinical
complexity level, and are further specifi ed for some case groups by age, sex, length
of stay, body weight and type of hospital discharge. Pilo ng of this new system is s ll
ongoing and will expand to 34 selected hospitals in fi ve provinces.
The pilot phase to develop the Vietnamese DRG system took much longer than expected.
One reason was the need to link the defi ni on of case groups and the services they
include to clinical pathways. These had yet to be developed in line with the Ministry’s
plan for people’s health protec on, care and promo on. This process took several years
and is expected to be fi nalized in 2020.
Dr. Tham Chi Dung
FINA
L DR
AFT
Costs for providing health services can
be diff eren ated into several cost items,
such as salaries, medicines, consumables,
equipment or capital. These cost items can
be paid for through various fi nancial fl ows.
Some CBG systems do not pay for and do not
include salary costs, which o en con nue
to be provided by the Ministry of Health
through supply-side fi nancing. Capital
costs are excluded from most CBG systems
because decisions on capital investment
and expensive or high-tech equipment
o en require a long-term perspec ve and
are usually part of a na onal investment
plan (37). For example, in Austria, Czechia,
Finland, Germany and Ireland, capital costs
of providers are covered through addi onal
alloca ons and are therefore not included
in the DRG payments (37). However, CBG
payments to private providers might have
to refl ect capital cost as well.
The larger the share of costs covered
under the CBG payments, the more room
there is for the provider to rellocate these
resources internally. Moreover, fi nancial
3.3. WHICH COST ITEMS CAN BE INCLUDED IN THE CBG
PAYMENT?
20 HEALTH FINANCING GUIDANCE NO. 10
incen ves created by the CBG payments
might not be strong enough to change
provider behaviour if only a small share of
the total cost of health services is paid by
the CBG system; and hence the impact of
a CBG payment system and its contribu on
to making purchasing more strategic would
be reduced (4).
Including only selected cost items in the
CBG payment system makes it possible
to share the expenditure risk of unknown
health service use between the provider
and the purchaser. On the one hand, lower
u liza on rates would not necessarily lead
to a facility’s closure if staff and equipment
costs are funded from other sources. This
can allow providers to con nue opera ng
in remote areas with low u liza on and
with limited ac vity-based payments. On
the other hand, par al cost inclusion in
CBG payment limits the poten al of the
CBG system to restructure and op mize
the provider market – i.e. aligning hospital
bed capacity with the size and needs of the
popula on in its catchment area (4).
Countries usually import DRG systems
from another country, including the related
grouping algorithm and, if applicable,
related so ware. There are no documented
cases of a country buying a (simpler) CBG
system from elsewhere. However, buying
or making a DRG grouping algorithms and
grouper so ware is not an either-or choice.
A country can buy and apply without
further adjustment; 2) buy and adapt to
the context, leading to the development
of a country-specifi c DRG variant; 3) copy
(without buying, if this op on is available)
and adjust to the country’s context; or 4)
develop its own DRG system for its own
se ng. Even though some countries
have ini ally bought a DRG system, they
have later developed their own system or
changed to another DRG variant if those
have been a be er fi t for the (evolved)
country context.
The USA began developing a DRG system
in the 1970s and introduced it as a
payment method in 1983. Other systems
developed from new were those of the
United Kingdom (1992), Austria (1997) and
Netherlands (2005). However, most other
countries developed their DRG systems on
the basis of an exis ng DRG variant (42).
Busse et al. (2011) provide an overview of
which European countries developed their
own DRG system and which ones chose
to import an exis ng variant (Table 2) (5).
Since a DRG system requires constant
adjustment, most of the countries that
ini ally imported a DRG system will develop
their own country-specifi c variant over
me (28). Box 5 describes how the Islamic
Republic of Iran adopted an imported
grouper to develop its own DRG system.
An instruc ve descrip on of the
development and the current design of the
Health Care Financing Administra on DRG
variant which is used by the Centres for
Medicare and Medicaid Services is provided
in the reference (43) next page.
3.4. WHEN IS IT BEST TO BUY AN EXISTING DRG SYSTEM
AND WHEN SHOULD A COUNTRY DEVELOP ITS OWN
COUNTRY-SPECIFIC CBG SYSTEM?
FINA
L DR
AFT
is providedprovide
t page.page
21PREPARATION AND DESIGN PHASE
Table 2. Main DRG variants which have been imported and/or further developed by other countries
DRG variant Country of origin Adopted/ further
developed by
(not an exclusive
list)
Classifi ca on
system used
Link to more
informa on
Health Care
Financing
Administra on
(HCFA) DRG/
Medicare Severity
(MS) DRG
USA NordDRG, France,
APR DRG
Diagnoses: ICD-10-
Clinical Modifi ca on
procedure: ICD-10-
Procedure Coding
System
h ps://www.cms.
gov/Medicare/
Medicare-Fee-for-
Service-Payment/
AcuteInpa entPPS/
MS-DRG-
Classifi ca ons-and-
So ware.html
All pa ent refi ned
(APR) DRG
USA (used by over
25 state Medicaid
programs in 2019)
Spain, Portugal,
Australia
Diagnoses: ICD-10
Procedures: ICD-10-
Procedure Coding
System
h ps://www.3m.
com/3M/en_US/
health-informa on-
systems-us/
providers/grouping-
and-classifi ca on/
apr-drgs/
Australian Refi ned
(AR) DRG
Australia Ireland, Germany,
Iran, Lithuania,
Thailand
Diagnoses: ICD-
10-Australian
Modifi ca on
Procedures:
Australian
Classifi ca on
of Health
Interven ons
h ps://www.ihpa.
gov.au/what-we-do/
ar-drg-classifi ca on
NordDRG NordDRG owner
countries: Denmark,
Finland, Iceland,
Norway, Sweden
Estonia
Latvia
Diagnoses: ICD-10
Procedures: Nordic
Medico-Sta s cal
Commi ee
Classifi ca on of
Surgical Procedures
h p://www.
nordcase.org/eng/
Health Care
Resource Groups
(HRG)
United Kingdom Poland Diagnoses: ICD-10
Procedures: Offi ce
of Popula on
Censuses and
Surveys health care
procedure and
interven on coding
h ps://digital.
nhs.uk/services/
na onal-casemix-
offi ce
Thai DRG Thailand Philippines,
Vietnam, Indonesia
Diagnoses: ICD-10
Thai Modifi ca ons
Procedures:
ICD-9 Clinical
Modifi ca ons
h p://www.tcmc.
or.th/main/
Source: Compila on by authors based on reference (5).
For further reading, please see:
Centres for Medicare and Medicaid Services. Design and development of the Diagnosis Related Group
(DRG). Baltimore: CMS/ Department for Health and Human Services; 2013 (https://www.cms.gov/
ICD10Manual/version34-fullcode-cms/fullcode_cms/Design_and_development_of_the_Diagnosis_
Related_Group_(DRGs)_PBL-038 , accessed 31 Oct
ober 2019).
FIFINA
L
FIN
FT
NAL
DRA
FTFTT
f the Diagnosis Related
ervices; 2013 (https:/
esign_and_developmen
ober 2019).
FINA
L DR
AFT
22 HEALTH FINANCING GUIDANCE NO. 10
Box 5. Buy and make in Iran: adopting an imported grouper to develop its own DRG
system
Iran has been using fee-for-service payments and case-based payments for 90 procedures
to pay hospitals since the 1990s. In order to improve effi ciency and produc vity, u liza on
management and introduce performance-based budge ng, the Iranian Ministry of
Health decided in 2015 to explore the introduc on of a DRG payment system for case-
based hospital payment and budget alloca on. As a start, diff erent DRG variants were
assessed for how they would fi t with the Iranian health system and context. Following
a consulta on process with na onal and interna onal DRG experts in 2016 and a
qualita ve research project carried out by Iranian health economists, the Ministry of
Health iden fi ed the Australian Refi ned DRG variant (AR-DRG) as the most appropriate
op on. The reasons for this decision were the high quality and availability of AR-DRG
documenta on, the availability of technical assistance and support from the Australian
Independent Hospital Pricing Authority (IHPA) and grouper suppliers, the widespread
use of the AR-DRG system in many countries and a long history of refi nement and regular
updates.
Iran signed a confi den ality deed (a commercial contract and agreement which specifi es
if, how and for which purpose informa on can be used or passed on) with the IHPA
to evaluate the AR-DRG system in rela on to the Iranian context. According to this
contract, Iran has access to the AR-DRG version 9.0, which is based on ICD-10 Australian
Modifi ca on and the Australian Classifi ca on of Medical Interven ons (ACHI), and the
related Australian Coding Standards.
The development and implementa on of the DRG system is taking place in three phases.
During the fi rst phase of tes ng, data compa bility between the AR-DRG requirements
and data currently included in medical records was assessed in a number of public
hospitals. Currently, the second phase is focusing on system adapta ons needed to run
a DRG system. Those include adapta ons to classifi ca ons and clinical coding, provider
payment policies, performance and quality management procedures, price and tariff –
se ng as well as IT and health informa on management systems. Three universi es
and two AR-DRG grouper suppliers are suppor ng some 50 hospitals in managing
DRG-based budget alloca ons and in monitoring performance and clinical prac ces
at physician and hospital levels. The third phase entails the development of an Iran-
specifi c DRG variant by adjus ng and changing the AR-DRG case-group weights and the
classifi ca on algorithm to the needs of Iran.
Iran aims in the long term to develop its own payment system based on an integrated
care approach encouraging providers and purchasers to deliver and buy comprehensive
healthcare packages in order to improve the health-care delivery process.
Mahdi Naderi
FINA
L DR
AFT
23PREPARATION AND DESIGN PHASE
Box 6. Gradual development of a CBG system in Kyrgyzstan
The Mandatory Health Insurance Fund of Kyrgyzstan introduced a case-based payment
system for public providers in 1997 as part of broader health fi nancing reform. At the
me, CBG payments were addi onal on top of line-item budget alloca ons. The CBG
payments were accompanied by performance-based staff bonuses. Higher CBG base
rates are paid for pa ents who are exempted from formal co-payments.
The ini al classifi ca on algorithm was simple. There were 28 case groups mostly
following the hospital department structure or clinical sub-speciali es, which were
further divided according to whether a pa ent had been in the intensive care unit or not.
Thus, the CBG system consisted of 56 case groups in total. This ini al set of case groups
was used for about 20 months.
A fi rst revision increased the number of case groups to 144 and eliminated the direct
connec on between an intensive care unit stay and a higher payment rate. In 2003,
the system was revised a second me to replace ICD-9 diagnosis coding with ICD-
10. Nowadays, only principal diagnosis and the main (i.e. most resource-intensive)
procedures are used for the CBG assignment. The only addi onal characteris c taken
into account is the pa ent’s age (< 15 years, ≥15 years). In total there are now 284 CBGs.
Another revision is currently ongoing and has three main objec ves: 1) to update the
classifi ca on of surgical procedures, 2) to review the classifi ca on algorithm to improve
clinical homogeneity by now using main diagnos c categories as the fi rst classifi ca on
step, and 3) to reduce age-related splits in case groups.
Triin Habicht
With regard to simpler CBG systems,
countries usually develop their own
version. Examples include Ghana, Morocco,
Philippines and Tunisia (16, 17, 41, 44).
Likewise, Kyrgyzstan decided to develop its
own CBG system by combining diagnosis for
medical cases and procedures for surgical
cases and then further spli ng them by
age (Box 6). In such a system the number of
groups will be lower and the development
of a classifi ca on algorithm simpler, as will
the cos ng for rela ve case weights. For
such less complex CBG systems, grouping
so ware is not necessarily required (4).
The most complex part of a DRG system
is the development of the algorithm for
alloca ng cases to case groups. Impor ng
a classifi ca on algorithm from another
country rather than developing one’s
own can reduce me and costs. Buying a
classifi ca on algorithm usually includes
acquiring an accompanying grouping
so ware with a licence agreement. Some
classifi ca on algorithms are open source,
are freely available/downloadable and
can thus be adjusted to the country
context as needed. Other algorithms have
more restric ons. Most DRG systems
are developed by public or semi-public
organiza ons, but private companies also
develop and sell classifi ca on algorithms
and so ware. If a DRG variant, including
a related algorithm and/or so ware, is
bought, it is important that the classifi ca on
24 HEALTH FINANCING GUIDANCE NO. 10
algorithm is also shared openly and is
adjustable, with the developers being
responsive to the needs of the buyer. Some
country experiences have shown that
merely receiving the grouping so ware
without insights into the classifi ca on
algorithm and without op ons to adjust it
will create signifi cant problems and may
lead to a huge loss in me and spending.
Consequently, when buying a DRG system
it is very important to have insight into the
classifi ca on algorithm and to be able to
adjust the algorithm as the system evolves.
Further factors to consider when deciding
whether to buy a DRG system or create one’s
own are outlined in the following sec ons.
An exis ng classifi ca on algorithm can be
adopted if there are suffi cient similari es
in the classifi ca on systems, cost structures
and the treatment prac ces between
the DRG system to be imported and the
envisaged classifi ca on system. The larger
the diff erences, the more diffi cult and
expensive it will be to adjust the imported
system. The envisaged DRG variant should
cover the same levels (types) of health
facili es and the same categories of
services as the imported DRG system.
With respect to classifi ca on systems, one
important point is the underlying coding
for diagnoses and procedures. If those are
not the same, data has to be mapped to the
imported system before it can be used to
run the imported grouping algorithm. As
most DRG variants are based on ICD coding
for diagnoses, grouping by diagnoses
makes exis ng classifi ca on algorithms
more easily applicable. Comparison of
several country DRG variants has shown
that the major diagnos c categories tend
to be similar and that they are aligned with
medical special es (36).
O en, however, there are larger diff erences
between countries in the treatment
prac ces and how they are coded as well
as in the cost structures (5). The share of
cost on human resources, for instance,
tends to be larger in high-income countries
than in middle-income countries. To
assess diff erences in cost structures, the
classifi ca on algorithm can be tested
with exis ng informa on. Alterna vely,
clinical experts can advise as to how far
their own treatment prac ces diff er from
the treatment prac ces that underlie the
grouping logic of the imported system.
The adequacy of a DRG variant for the
context of an impor ng country can
be evaluated by using data that are
available (such as from hospital discharge
summaries). The required informa on
should include diagnosis (primary and
secondary), procedures and pa ent
characteris cs that can be used to generate
coded informa on. The coded informa on
can be fed into the classifi ca on algorithm
to analyse the extent to which cases are
allocated to the appropriate case groups.
However, such evalua ons are costly if
data are not available electronically. These
evalua ons are o en a fi rst step towards a
pilot phase (36). An account of Georgia’s
experience in this is contained in Box 7.
Similari es between the imported and the envisaged case
classifi ca on system
FINA
L DR
AFT
25PREPARATION AND DESIGN PHASE
A DRG system relies on pa ent data
collected by a well-func oning informa on
management system. To use an imported
DRG system, health facili es must
rou nely collect pa ent data at a similar
level of detail. If no strong pa ent
informa on management system exists,
the development of a CBG system that uses
exis ng data is a more viable op on (4, 29).
Impor ng a DRG system and its classifi ca on
algorithm can be useful if cost data are
lacking, and if the imported system
includes rela ve weights that can be used
as cost assump ons and then be gradually
adjusted. Using these cost assump ons
can reduce the need for addi onal cos ng
studies.
Table 3 compares the op on of developing
a country-specifi c CBG system or DRG
system to the transi onal strategy of fi rst
impor ng an exis ng system and then
gradually adjus ng it.
Availability of rou nely collected pa ent informa on
Box 7. Matching classification systems and electronic claims data format to
facilitate the adoption of the NordDRG variant in Georgia
Georgia applied a country-specifi c CBG payment system with a high number of case
groups defi ned through a combina on of ICD-10 and the Nordic Medico-Sta s cal
Commi ee (NOMESCO) classifi ca on of surgical procedures. Over the years this locally
built system has been fi netuned to limit the number of case groups (e.g. some areas of
cardiology) but there has been no comprehensive solu on for all clinical areas.
In mid-2017 the Ministry of Internally Displaced Persons from the Occupied Territories,
Labour, Health. and Social Aff airs of Georgia decided to introduce a DRG system for
payment. A feasibility study revealed that the NordDRG system is suitable for Georgia as
health providers already use the same surgical procedures classifi ca on system as the
Nordic countries. Similarly, Georgia applies ICD-10 which is also used by the NordDRG
system. Thus, there was no need to introduce a new primary classifi ca on system.
Consequently, the pragma c op on for Georgia is to adopt the NordDRG grouper, while
keeping the possibility of adding country-specifi c adjustments in the future.
Another important suppor ve factor for adop ng the NordDRG system is the availability
of digital pa ent-level claims data which include all informa on required by the
NordDRG classifi ca on algorithm. However, some eff orts are needed to further improve
data quality in parallel to implemen ng the new system. By the end of 2018, the Nordic
Casemix Center had developed the Georgian version of the NordDRG classifi ca on
algorithm. Transi on to the new system began in mid-2019 (45).
Source: reference (45)
FINA
L DR
AFT
26 HEALTH FINANCING GUIDANCE NO. 10
Table 3. Issues to consider when deciding whether to make or buy a DRG system
Impor ng and gradually adjus ng an
exis ng DRG system
Adopted/ further developed by (not
an exclusive list) Developing a new
DRG (CBG) system
Acceptability and
ownership
There is a risk of insuffi cient ownership.
Stakeholders must be engaged and the
system appropriately contextualized.
The process of developing a new system
will strengthen ownership among key
stakeholders and build capacity.
Cost Costs may be more modest, but there
may be licensing cost upfront and on a
con nuing basis (risk of vendor lock-in).
Addi onal cost for country-specifi c data
collec on and adjustments has to be
factored in.
Development cost is generally higher but
it will depend on the complexity of the
system.
Time period
required for the
introduc on
Impor ng may take rela vely shorter
me, being feasible within a few years,
but the me needed also depends on the
extent of necessary adjustments.
Developing takes rela vely longer but it
might support capacity-building for key
stakeholders.
Data needs Impor ng requires well-func oning
pa ent data collec on systems.
It requires less data collec on if
cost assump ons are imported, but
adapta ons will be needed.
Developing requires more data collec on
for the ini al development.
A locally developed system can be
designed in line with the exis ng capacity
of the pa ent data collec on system.
Suitability for a
country’s health
system
Impor ng a system requires adapta ons
to refl ect cost structures and pa erns of
clinical prac ce.
A locally developed system may refl ect a
country’s health system well, facilita ng
acceptance by providers.
The system might be perceived as fairer.
Maintenance and
regular revisions
An imported system may require external
support for maintenance unless suffi cient
capacity is available or built locally.
Suffi cient capacity is needed to build up
the system and maintain it locally It may
thus also require external support.
Source: Compila on based on references (4, 14, 29).
The rela ve weight of a case group refl ects
the average treatment cost of an average
case in this specifi c group in rela on to the
average treatment cost of all cases. Case-
group weights are computed by dividing
the average costs of cases falling within a
CBG through the average treatment costs
of all cases in a country, as outlined in
the equa on below. A case-group weight
of 1.2 for a specifi c group A would mean
that the average treatment cost for a case
in group A would be 20% higher than
3.5. HOW ARE CASE-GROUP WEIGHTS AND THE BASE RATE
CALCULATED?
Calcula on of case-group weights
FINA
L DR
AFT
er thaner tha
27PREPARATION AND DESIGN PHASE
the average treatment cost across all
groups. Consequently, a CBG with a case-
group weight of 1.0 implies that average
treatment costs of pa ents falling into that
CBG are equal to the average treatment
costs of all cases within the country (4, 12).
The formula for calcula ng the case group
weight is as follows (based on (4)):
Box 8 shows how Lithuania used pa ent
cos ngs to develop its own case-group
weights, while Box 9 describes how Estonia
has applied a mix of imported and locally-
developed case-group weights.
Box 8. Introduction of a patient-level costing system to develop country-specific
case-group weights in Lithuania
Lithuania started to use Australian Refi ned DRG system (AR-DRG) in 2012 a er a feasibility
study in 2009 and a preparatory phase from 2009 to 2011. During the prepara on
phase, the Australian Classifi ca on of Medical Interven ons (ACHI) was introduced
and the ICD-10 version in use was updated. Lithuania has aimed to develop its own
case-group weights from the very beginning. Un l now all acute care hospitals submit
annually aggregated cost data reports by 16 main cost categories to the Na onal Health
Insurance Fund (NHIF). These na onally collected cost data are combined with the cost
structure of each DRG group taken from the Australian rela ve case-group weights.
On this basis, the NHIF constructs country-specifi c case-group weights. These follow
the cost structure of the Australian DRG system, but the cost related to diff erent cost
categories is calculated on the basis of data from Lithuanian hospitals. The weakness
of this methodology is that it follows the cost structure of the Australian health system
which, in some cases, may diff er signifi cantly from the Lithuanian reality. Eventually a
more systema c approach to validate case group weights was needed.
Therefore, in 2015, the NHIF ini ated the pa ent-level cos ng project in 15 hospitals
that represent the whole scope of acute inpa ent care. The project has a triple aim:
1. to develop Lithuanian DRG case-group weights refl ec ng, the cost structure of
Lithuanian hospitals;
2. to analyse hospital sector performance; and
3. to give hospitals the possibility to analyse their own cost data to fi nd areas for
improvement.
The NHIF has developed a standardized cost accoun ng so ware and provides hands-on
technical guidance to support par cipa ng hospitals. The cos ng project is expected to
provide results in 2020. First experiences suggest that the complexity of the pa ent-level
cos ng system requires more me for implementa on at hospital level than ini ally
planned.
Source: reference (46)
FINA
L DR
AFT
DRA
FT
T
DRA
Feference (4 )
28 HEALTH FINANCING GUIDANCE NO. 10
Box 9. A mix of imported and locally-developed case-group weights in Estonia
In 2001, the Estonian Health Insurance Fund (EHIF) decided to implement a DRG payment
system and to adopt the NordDRG system.
In order to develop and calculate case-group weights that are country-specifi c, two
approaches were pursued. The fi rst one was to calculate case-group weights based on
exis ng fee-for-service tariff s. Claims with fee-for-service informa on were available in
digitalized form for all hospitals since 2000. DRG case-group weights were calculated
on the basis of data from January to June 2003. Since tariff s include all cost categories
(personnel, consumables, overheads and capital costs) and providers are allowed
to charge only the offi cial co-payments from the pa ents, these data gave a good
approxima on of rela ve costs. The second approach was to use the case-group weights
from the American Health Care Financing Administra on (HCFA) DRG variant.
Simula ons with data sets for both methods were undertaken and discussed with the
DRG Advisory Commi ee in 2003. This recommended to use Estonia’s own case-group
weights, except for DRGs with a low case number (less than 30) for which the HCFA
weights should apply.
DRG payments were then par ally introduced in 2004. The DRG base rate was calculated
using the overall available budget, thus assuming budget neutrality. DRG payment rates
are not diff eren ated by hospital type. Teaching hospitals receive the same base rate as
general hospitals, although their case mix index is higher since they treat rela vely more
complex cases. The unifi ed base rate refl ects the assump on that pa ents in the same
DRG group should have on average the same cost, regardless of where they receive their
treatment.
Un l now, Estonia has con nued to apply a similar methodology for DRG case-group
weight calcula on. This is possible because DRG payments form 70% of total claims
cost, while the other 30% is s ll reimbursed on the basis of fee-for-service. Since 2006,
fee-for-service tariff s have been calculated by using selected hospitals’ cost data and
applying an ac vity-based cos ng methodology.
Source: reference (47)
The base rate is calculated on the basis of
the overall available budget for the health
services remunerated through CBG and
the overall expected case mix for a specifi c
period (generally the budget year). As the
precise volume of services to be provided is
Calcula on of the base rate
FINA
L DR
AFT
29PREPARATION AND DESIGN PHASE
where BRt = base rate for the me period t
Budgett = available overall budget for services to be paid by CBG in me period t
Casesh, t-1 = total number of cases at provider h during the me period t-1 (previous
year).
not known prospec vely, a simple es mate
is an expecta on of the same volume as
provided in the previous year. The formula
used for calcula ng the base rate is as
follows (4):
A detailed description with examples of how to calculate relative case-group weights and base rates
can be found in:
Langenbrunner J, Cashin C, O’Dougherty S. Designing and implementing health care provider payment
systems: how-to manuals. Washington (DC): The World Bank; 2009: 159–70
(https://openknowledge.worldbank.org/bitstream/handle/10986/13806/48599 , accessed 28
September 2019).
To calculate case-group weights, the
average cost of trea ng a case within every
case group must be determined. While all
cost informa on is useful, it is specifi cally
required for the cost categories that are
included in and paid by the CBG system.
Table 4 provides an overview of data
requirements and possible data sources
for diff erent types of case grouping –
i.e. complexity levels of CBG systems.
Box 10 outlines how the collec on of
cost informa on is being gradually
ins tu onalized in the Philippines.
3.6. WHICH DATA ARE NEEDED TO CALCULATE CASE-
GROUP WEIGHTS AND HOW CAN THESE DATA BE
GENERATED?
Data needs
FINA
L DR
AFT
30 HEALTH FINANCING GUIDANCE NO. 10
Table 4. Cost informa on required and sources of informa on for diff erent case grouping systems
Type of case
grouping
Data requirements Data sources
No case grouping
(fl at case payment)
Average cost per case across all providers
covered by the CBG system
Historical provider budgets, other
provider expenditure and u liza on data
Case-based
grouping which
dis nguishes only
between diff erent
departments
or clinical
specializa ons
Department average cost per bed-day;
department length of stay
Provider budgets and cost-accoun ng
system/analysis; other provider
expenditure and u liza on data
Case-based
grouping which
includes subgroups
for diff erent
departments
or clinical
specializa ons, but
which is s ll less
complex than a
DRG system
Department average cost per bed-day;
pa ent-level treatment cost disaggregated
by diagnosis and/or procedure; length of
stay and other characteris cs of the case/
pa ent depending on those variables used
for defi ning the case groups
Provider budgets and cost-accoun ng
system/analysis; pa ent-level data on age,
sex, ICD code for primary diagnosis, length
of stay, surgery and other characteris cs
of the case (such as intensive care,
procedure, type of bed, type of discharge)
Diagnosis-based
grouping
Department average cost per bed-day;
disaggregated by diagnosis; length of stay
and other characteris cs of the case/
pa ent
Provider budgets and cost-accoun ng
system/analysis; pa ent-level data on age,
sex, ICD code for primary diagnosis, length
of stay, surgery and other characteris cs
of the case (such as intensive care,
procedure, type of bed, type of
discharge).
Source: Reference (4).
Box 10. PhilHealth’s plan to institutionalize the collection of cost information
The Philippine Health Insurance Corpora on (PhilHealth) started its refl ec on and
development of a DRG payment system in 2009 in order to move away from fee-
for-service payments. However, one major challenge was the lack of detailed and
disaggregated cost informa on to inform rela ve case-group weights. Collec on of cost
data started in 2009 using a top–down cos ng tool. However, these cos ng studies were
not systema cally conducted. Data collec on was further impeded by a lack of electronic
data collec on, the absence of na onal standards for hospital accoun ng, reluctance
of providers to provide transparent informa on and limited capacity and experience in
conduc ng cos ng studies.
Ten years later, in 2019, the PhilHealth Board approved a long-term comprehensive
plan with the objec ve of ins tu onalizing the collec on, analysis and policy use of cost
informa on. The plan includes three major pillars, namely:
FINA
L DR
AFTRAF
TT
DRA
FT
31PREPARATION AND DESIGN PHASE
Box 10. (Contd.)
First, PhilHealth has compiled a cos ng toolkit which includes data collec on forms,
training manuals, communica on templates to facilitate informa on exchange between
local PhilHealth offi ces and providers, and evalua on forms. These standardized tools
will allow for collec ng comparable data across health-care providers.
The second pillar consists in the ins tu onaliza on of the cost informa on collec on
and valida on process. Providers have to submit cost informa on regularly and this
requirement is part of accredita on standards and the provider contract. The submi ed
data will be analysed by PhilHealth in collabora on with academic or research
ins tu ons so as to develop and update cost weights. It will also be fed into a dialogue
between PhilHealth, the Department of Health, providers and pa ent representa ves.
This can contribute to refl ec ons on the aff ordability of the benefi ts as well as provide
informa on about the quality of health services. Moreover, it can serve as feedback
to providers and can enable communica on regarding the results of the data analysis
for mutual accountability between PhilHealth and providers, and the joint defi ni on of
priori es.
Finally, the third pillar focuses on strengthening IT systems and capaci es to automate
and reduce as far as possible the repor ng burden.
The implementa on of the cos ng informa on collec on plan is an cipated for 2020.
This will allow PhilHealth to collect cost informa on from providers rou nely in order to
develop case-group weights as a crucial element of a DRG payment system.
Melanie Coronel San llan
Data are ideally collected from the cost-
accoun ng system at hospital level which
enables the treatment cost per pa ent
to be iden fi ed and, if needed, facilitates
the alloca on of indirect costs incurred
by hospitals to individual cases. Such
detailed informa on may o en not be
collected rou nely, especially prior to
the roll-out of a CBG system. Some form
of cost es ma on or modelling may
therefore be needed (29). The quality and
granularity of available cost data are likely
to increase over me while capaci es
and methodologies for the collec on of
cost data become ins tu onalized. The
use of cost informa on for payments will
incen vize providers and purchasers to
improve data quality. The approaches
below outline diff erent methods for
genera ng cos ng data. A combina on of
these approaches can be an eff ec ve way
of closing data gaps.
Introduc on of cost-accoun ng systems
in a sample of providers. Although it
allows for cost informa on to be collected
rou nely, introducing a cost-accoun ng
system that enables pa ent data to be
linked to cost data at the necessary level
of disaggrega on can be costly. Some
Genera ng cost data
FINA
L DR
AFT
ry levey leve
costly. Some ostly. Som
countries chose to implement such a system
only in a sample of providers which receive
a compensa on payment for the addi onal
administra ve burden. The sample should
be representa ve so that the data collected
will provide adequate informa on for
designing and upda ng payment rates (37).
Cos ng studies. Cos ng studies can
be conducted if data are lacking or if
the informa on available from rou ne
data collec on is incomplete. Again, the
provider sample for the cos ng study must
be representa ve of all providers that will
be included in the payment system.
Use of fee-for-service payments. If fee-for-
service payments are replaced by a CBG
system, the exis ng fee schedules can be
used to es mate the costs of diff erent case
groups. This avoids costly collec on of cost
data through specifi c studies. However, it
should be considered carefully because fee-
for-service payments may include historical
ineffi ciencies. In a fi rst phase this might
limit the cost-containment poten al of the
CBG system, but CBG payments that are
similar to previous rates can also facilitate
acceptance of the new system by providers.
For example, Estonia, and Thailand used
fee-for-service rates as a star ng point for
developing rela ve cost weights (28, 47).
Impor ng cost assump ons from an
exis ng grouper. This op on is available
when a country imports a grouper that
includes rela ve case-group weights.
These assump ons can be used when
introducing the system and can then be
gradually fi netuned. The impor ng country
should then make adjustments with a
focus on the most frequent cases or the
main drivers of treatment cost (12). It is
important to undertake impact assessment
to explore the impact and iden fy if further
adjustments are needed.
Regardless of which method or mix of
methods is used to generate cost data
for the design of the CBG system, cost
assump ons and calcula ons must be
constantly updated to refl ect changes
in cost structures, innova ons and the
introduc on of new technologies.
32 HEALTH FINANCING GUIDANCE NO. 10
Recommendations on how to do costing studies can be found in:
Langenbrunner J, Cashin C, O’Dougherty S. Designing and implementing health care provider payment
systems: how-to manuals. Washington (DC): The World Bank; 2009: 159–70 (https://openknowledge.
worldbank.org/bitstream/handle/10986/13806/48599 , accessed 28 September 2019).
Özaltın A, Cashin C, editors. Costing of health services for provider payment: a practical manual based
on country costing challenges, trade-offs, and solutions. Arlington (VA): Joint Learning Network for
Universal Health Coverage; 2014 (http://www.jointlearningnetwork.org/uploads/files/resources/JLN_
Costing_Toolkit_Interactive_FINAL , accessed 28 September 2019).
33PREPARATION AND DESIGN PHASE
The basic idea of a CBG system is to pay
the same for each case within the same
group. However, adjustment factors can
be used to raise or lower CBG payments to
compensate for factors which cannot be
captured by the classifi ca on system itself
(8, 12), such as the remoteness of a health
facility or other specifi cs. The adjustment
factors can be considered as addi onal
coeffi cients in the formula to determine
the case payment (22), and must align with
policy objec ves. The following adjustment
factors are found in prac ce:
Geographical criteria: Regional diff erences
within countries (e.g. diff erences in prices
or transport distances) result in diff erent
input costs for providers. A purchaser may
also want to encourage service provision
in remote areas or compensate for lower
provider income resul ng from lower
u liza on rates in sparsely populated areas
(22). These diff erences can be addressed
with a geographical adjuster.
Equity criteria: Providing treatment to
pa ents who belong to defi ned vulnerable
popula on groups can be more resource-
intensive. To improve equity in access to
health services, regulators or purchasers
might want to off er fi nancial incen ves
to providers to treat such pa ents. For
instance, in the Australian hospital payment
system an adjustment factor is added
to provide higher payments for services
provided to indigenous Australians (49).
Provider level adjustments: Some
systems might apply diff erent payment
rates for diff erent providers. For instance,
there could be higher rates for hospitals
using more specialized technologies or
equipment. Adjustment factors could also
serve to compensate teaching hospitals
for their teaching ac vi es, which are
diffi cult to refl ect in a CBG payment system
classifi ca on. However, such adjustment
coeffi cients, if not carefully designed, may
create the wrong incen ves. If secondary
or ter ary facili es receive higher
payments for simple treatments which
can be provided at lower cost at lower-
level facili es, there is less incen ve for
providers to direct pa ents to lower-level
facili es (7). Such adjustment factors might
also encourage providers to invest in high-
tech equipment to obtain higher payments.
Hospital-specifi c adjustments: When a
CBG payment system is introduced, some
hospitals might face signifi cant reduc ons in
their revenues; for instance, their previous
overall budget might have been higher than
the CBG payment system’s remunera on
for their ac vity level. Specifi c hospital
adjustments can be used for an interim
period (with a clear phasing-out plan) to
ensure that the hospital’s income under
the new CBG system does not diverge too
much from its previous level of funding.
However, the hospital-specifi c adjustments
should not become permanent as this
would undermine the objec ve of the CBG
system (22).
3.7. WHICH ADJUSTMENT FACTORS CAN BE CONSIDERED
TO MODULATE THE CBG PAYMENT?
FINA
L DR
AFT
34 HEALTH FINANCING GUIDANCE NO. 10
The shi to a CBG payment system creates
incen ves to increase the number of cases
and can therefore escalate expenditure.
In this regard, several approaches exist
to share the fi nancial risk of increased
case numbers between purchasers and
providers.
The fi rst approach is to set payment rates
per case prospec vely and to apply a
volume cap or a budget cap. This cap is
based on how much can be covered with
the available budget. Purchasers and
providers need to nego ate this cap in
advance during the contrac ng process
(24, 29).
Figure 5 illustrates how total revenue per
provider develops under prospec ve se ng
of payment rates and in various scenarios
with and without a volume or budget cap.
When there is no volume/budget cap, the
provider revenue increases without limit
with the number of cases treated.
Under a hard budget or volume cap, any
cases that exceed the agreed budget cap
per provider within a budget period will
not be reimbursed. Providers have an
incen ve once the cap is reached to limit
cases or ra on services by crea ng wai ng
lists or making referrals. The fi nancial risk
associated with higher-than-expected
volume lies with the provider (24).
Under a so budget cap, addi onal services
will be reimbursed but at a lower payment
rate; the base rate is also commonly
reduced in the case of DRGs. The payment
rate can be set so that it is slightly below,
equal to or above the marginal costs for
trea ng a case. To encourage providers to
limit the number of cases they treat, the
rate would have to be equal to or below
the marginal cost for trea ng the service.
The fi nancial risk of higher-than-expected
volume is shared between provider and
purchaser (24).
3.8. SHOULD CBG PAYMENTS BE LINKED TO A VOLUME OR
BUDGET CAP AND, IF SO, HOW?
Figure 5. Link between volume of services and provider revenue for diff erent arrangements regarding
budget caps
Source: Authors’ compila on based on references (14, 24).
FINA
L DR
AFT
35PREPARATION AND DESIGN PHASE
The second approach is to set payment
rates retrospec vely and to adjust payment
rates in order to cover with the available
budget the en re volume of services that
have been provided. Providers collect
points that refl ect the rela ve severity
of the cases treated. At the end of the
budge ng period, usually the budget year,
the available budget is divided by the total
of points accumulated by all providers
covered by the payment system in order
to determine the base rate (24). This point
system is used, for instance, by the Universal
Coverage Scheme in Thailand (51).
A third op on is to use the CBG payment
rates and expected volume or the case mix
index as the basis for se ng a prospec ve
fi xed budget covering a defi ned period.
While this is a way to inform the budget-
se ng process, it in fact becomes a cap.
Importantly, the budget will not be adjusted
in line with the actual volume or case mix
of services provided. The provider will bear
the fi nancial risk of higher ac vity levels
but will also keep any savings if the ac vity
is lower than expected. When the budget is
set, case volumes from previous years are
taken into account in order to weaken the
incen ve to reduce the number of cases
(e.g. by rejec ng or referring pa ents,
or cream-skimming for pa ents with less
severe cases) (24).
Whichever budget or volume cap approach
is chosen, the predictability of the budget
is important for both providers and
purchasers. Tight budgets may make
providers look for other income such as
balance billing or informal payments. Strong
administra ve controls and monitoring
are therefore needed to minimize such
prac ces, as noted in Sec on 4.3 and
Sec on 4.5.
More information on setting soft budget caps can be found in:
De Lagasnerie G, Paris V, Mueller M, Kumar A. Tapering payments in hospitals – experiences in OECD
countries. Paris: Organisation for Economic Co-operation and Development; 2015 (OECD Health Working
Paper, No. 78; https://read.oecd-ilibrary.org/social-issues-migration-health/tapering-payments-in-
hospitals_5js4h5rlkxxr-en#page6, accessed 28 September 2019).
More country examples on budget and volume caps can be found in:
Busse R, Geissler A, Quentin W, Wiley R, editors. Diagnosis-related groups in Europe moving towards
transparency, efficiency and quality in hospitals. Maidenhead and New York (NY): Open University
Press – McGraw-Hill Education; 2011 (http://www.euro.who.int/__data/assets/pdf_file/0004/162265/
e96538 , accessed 28 September 2019).
FINA
L DR
AFT
4. IMPLEMENTATION,
MONITORING AND CONTINUED
REVISION
36 HEALTH FINANCING GUIDANCE NO. 10
This sec on highlights ques ons about implementa on that must be addressed during the
pilot and implementa on phases.
Introducing a CBG system usually requires
signifi cant adjustments to fi nancial
management systems and procedures at
provider level, including in informa on
collec on, claims management and human
resource management. Specifi c training
may also be needed to support providers.
4.1. WHAT CHANGES ARE REQUIRED AT PROVIDER LEVEL
WHEN INTRODUCING A CBG SYSTEM?
Changes in fi nancial fl ows to providers may
be signifi cant, especially if CBG payments
mean that providers are to be paid by
outputs. Where the CBG payment system
replaces input-based budget alloca ons,
providers will have to modify their internal
budge ng and fi nancial procedures.
However, providers may s ll have to develop
budget plans and report expenditure
following input-based budget structures. In
contrast, a shi from fee-for-service to CBG
payments might require fewer adapta ons
in fi nancial management. With more
detailed informa on on hospital ac vity,
the facility management might s ll be able
to reshuffl e resources to be er refl ect
the workload of each department. Facility
managers have to be able to lead such
transforma ons without risking disrup ons
in service provision (30).
The CBG classifi ca on algorithm may
require data that are not yet collected by
providers or data at levels of disaggrega on
not yet available. However, clinical
and pa ent data elements of suffi cient
quality are needed for many purposes in
Adapta ons in fi nancial management
Modifi ca ons to the provider’s data collec on and informa on
management systems
FINA
L DR
AFT
ciencien
purposes inpurposes
37IMPLEMENTATION, MONITORING AND CONTINUED REVISION
The introduc on of new repor ng
procedures requires changes to
administra ve processes. Purchasers and
providers will have to provide training to
ensure that staff are familiar with the new
system. Training topics include DRG system
design and grouping algorithms, the use of
primary classifi ca ons, coding standards
and rules, coding quality, repor ng, and
DRG-specifi c performance monitoring (28).
Providers may also hire specialized coders
whose role is to extract the necessary
informa on from medical records and to
code it as required by the classifi ca on
system. Coders will relieve doctors and
nurses of these administra ve tasks.
Likewise, regular training, with feedback, is
needed to improve the quality and accuracy
of coding (7). Box 11 the measures taken
for introducing ICD-10 in the Philippines
and the need to train and retain coders.
Changes in human resource management and the need for training
addi on to implemen ng a CBG system.
The introduc on of a CBG payment
system should not lead to a situa on
where providers focus only on collec ng
informa on on those variables or codes
that infl uence payment rates at the expense
of a more comprehensively documented
pa ent history. The introduc on of CBG
payments can incen vize data collec on
(4).
When a grouper so ware is used, data
collected through hospital informa on
management systems must be interoperable
with the so ware. The adapta on may
entail signifi cant cost. If not already used
by providers, computerized (informa on)
management systems can be developed
and introduced in parallel to ensure that
data are standardized and respond to
the needs for medical documenta on
on pa ent cases, repor ng on health
sta s cs, hospital management and claims
processing. Developing a separate system
that serves only the CBG increases the
repor ng burden of providers and should
be avoided (4).
Box 11. Introduction of ICD-10 in the Philippines and training measures
The Philippines Department of Health decided in 1998 to introduce ICD-10 as a pa ent
classifi ca on system in hospitals and registries in order to improve the quality of
morbidity and mortality sta s cs from health-care providers. Various measures were
introduced to accompany this process:
1. A na onal ICD-10 Technical Working Group was created in 1999 consis ng of
representa ves from the Department of Health, the Philippine Sta s cs Authority,
PhilHealth, the Associa on of Philippine Medical Colleges, the Associa on of
Municipal Health Offi ces of the Philippines and selected hospitals. The composi on of
this working group has since been revised to include other offi ces that are involved in
genera on of health sta s cs.
FINA
L DR
AFTAFTAFT
FINA
L DR
AFT
38 HEALTH FINANCING GUIDANCE NO. 10
Box 11. (Contd.)
2. Various training courses for diff erent target groups were developed to build up capacity
to implement ICD-10. The ICD-10 Training of Trainers is a two-week course designed
for poten al trainers from the Department of Health and its a ached agencies. There
is also a fi ve-day ICD-10 training course for coders that is off ered to medical records
staff , government regulatory offi cers, sta s cians and local civil registrars who use
morbidity and mortality data, in addi on to hospital coding staff . A separate three-
day course was specifi cally designed for medical staff who are responsible for the
documenta on related to pa ent records.
3. Through Administra ve Order No. 47 s. 2000, the Department of Health mandated
the use of ICD-10 for recording all diagnoses of pa ents. The licensing requirements
for hospitals were updated to include the full implementa on of ICD-10, including
procurement of ICD-10 books, use of ICD-10 in all medical records and hospital sta s cs
and repor ng, and the 5-day training course of medical records offi cers. PhilHealth
complemented the department’s Administra ve Order with its own informa on on
the use of ICD-10.
These eff orts resulted in be er-quality health data. However, there remain several concerns
– such as a fast turnover of trained coders (due to high demand in the interna onal
labour market) and inadequate IT infrastructure in providers’ and regulatory offi ces. It
will be important to enhance understanding among senior managers in PhilHealth and
the Department of Health of, among other issues, the importance of good data quality,
the need to improve monitoring and evalua on and the importance of strengthening
penal es and sanc ons for noncompliance with ICD-10 in order to further improve
coding quality in the Philippines.
Ronald Paguirigan
The introduc on of a CBG system can be
facilitated by an implementa on plan that
clearly defi nes the strategy and melines
for transi oning to the new system so that
purchasers and providers can prepare for
changes in fi nancial alloca ons.
To ensure provider buy-in and to avoid
disrup ons in service delivery, most
countries have chosen to phase in CBG
payments gradually. This helps in managing
fi nancial risks to providers by avoiding
signifi cant changes in provider income
from one year to the next. It also allows for
gradual standardiza on of payment rates
if signifi cant varia ons existed prior to the
introduc on of CBG payments (14).
4.2. HOW SHOULD THE CBG SYSTEM BE PILOTED AND
PHASED IN?
FINA
L DR
AFT
39IMPLEMENTATION, MONITORING AND CONTINUED REVISION
Various ways of phasing in and pilo ng can
be combined.
Shadow billing or shadow budgets: If
a well-func oning computerized claim
system is in place and can be used for CBG
claims, immediate countrywide pilo ng
through shadow payments is possible,
especially in smaller countries. Hospitals
start case repor ng according to the CBG
system and the respec ve payment is
calculated. During an agreed transi on
period, the actual payment s ll follows
the previous provider payment method.
This allows suffi cient me to adjust the
repor ng system and to allow providers
to prepare for the upcoming changes in
resource alloca ons (6).
Gradual increase of the share of hospital
income coming from CBG payments:
One op on is for hospital revenue to be
calculated according to both the case-
based payments and the previous payment
method. Ini ally a limited part of the case-
based hospital budget (e.g. 20%) will be
paid, while the remaining payment (e.g.
80%) is allocated according to the previous
payment method. The propor on of the
budget based on CBG payments is then
gradually increased (14). A second op on
is to pay diff erent base rates per hospital or
group of hospitals in order to avoid abrupt
changes in hospital income. The base rates
will then converge over me (28). This is
diff erent from covering only certain cost
items using CBG payments as discussed
below. Here, payment calcula on and
fi nancial management rules apply equally
to the whole budget and there is no
dis nc on between cost items.
Extension of geographical coverage:
The CBG system is piloted in one region
only or in a limited number of hospitals.
This geographical sample should be large
enough to draw general lessons and/or
to adjust the system to make it applicable
to the whole country at a later stage. This
can be based on voluntary par cipa on,
as was the case in the Republic of Korea,
but it should be noted that voluntary
par cipa on has the risk that self-selec on
may lead to a biased, nonrepresenta ve
sample (14).
Pilo ng CBG payment for one level of
hospital only: In this op on the CBG system
applies only to providers at a certain level of
care, for example secondary health services
only, or ter ary care only. This approach
makes it more diffi cult to include other
hospital levels at a later stage because cost
structures tend to diff er between levels of
care. In addi on, it may lead to diff erent
payments for diff erent levels of facili es
which is not in line with the CBG principle
of paying the same amount for all cases in
the same case group. This approach also
carries the risk that providers may try to
avoid trea ng pa ents by referring them
without suffi cient jus fi ca on.
Extension of hospital ac vi es covered:
CBG payments can also be applied to
a limited number of diagnoses and/or
procedures – e.g. those with long wai ng
mes or priority health services, as these
would be services for which the purchaser
wants to encourage increased provision.
However, under this op on, providers have
an incen ve to shi pa ents from services
not covered by CBG to services covered by
CBG if the la er are be er remunerated
(14).
Extension of cost items covered: Some
CBG systems ini ally cover only the cost
of consumables and medicines, and/or
opera onal costs. Addi onal cost items –
such as salaries, hospital administra on
costs or equipment – can be gradually
included over a defi ned me period.
This approach helps to strengthen the
management capacity of providers over
me. It is a useful approach in se ngs
FINA
L DR
AFT
40 HEALTH FINANCING GUIDANCE NO. 10
where providers have limited management
capacity and where an easy-to-administer
system is needed. However, the approach
limits the eff ects by which a CBG system
can trigger a restructuring of the provider
market, although this may be an objec ve
of the payment reform (14).
Extension of popula on groups covered
or pilo ng with one scheme only: The
CBG payment system can be piloted for
the treatment of pa ents covered under
a specifi c scheme (as was the case in
Thailand (35)) or for pa ents with specifi c
characteris cs (i.e. pa ents over 70 years).
The design of the pilot should avoid crea ng
incen ves for cream-skimming of pa ents
by providers (14). This op on should be
considered carefully as it carries a high risk
of discrimina on against pa ents whose
treatment is covered by a less a rac ve
payment system.
CBG payments create incen ves for
providers to increase the number of cases.
This can lead to cost escala on or to the
provision of high volumes of non-priority
or unnecessary services. CBG payments can
also incen vize under-provision of cases
(52).
Table 5 provides an overview of
administra ve measures to address these
problems. Three types of administra ve
measures can be iden fi ed: budget or
volume caps (for more informa on see
Sec on 3.8), regular revisions of the CBG
system (for more informa on see Sec on
4.5), and direct control mechanisms (i.e.
monitoring and audi ng) which are further
outlined below.
4.3. HOW CAN ONE COUNTER INHERENT PROVIDER
INCENTIVES CREATED BY CBGS?
When providers have incen ves to reduce
the number of inputs per case, the quality
of care can be compromised. Consequently,
there will be a need to monitor quality-
related indicators which commonly include
length of stay, survival rates a er 3, 6 or
12 months, and/or readmission for the
same diagnosis. Performance in rela on to
these indicators can be linked to payments
where the informa on management
system allows. In several countries,
frequent monitoring was established at the
beginning of CBG introduc on, including
making key indicators publicly available on
the purchaser’s website (28).
Monitoring of key indicators
FINA
L DR
AFT
41IMPLEMENTATION, MONITORING AND CONTINUED REVISION
Payments can be adjusted for outliers
regarding length of stay. As an example,
addi onal payments can be allocated
to complex cases with a length of stay
signifi cantly above the average, or reduced
rates could apply to inpa ent stays that are
shorter than a certain threshold (12, 52).
Addi onal payments could also be allocated
for cost-outliers. Some CBG systems group
those cases into a pre-MDC group in the
fi rst step of their classifi ca on algorithm.
Addi onal payments for cost outliers can
also encourage the use of technological
innova ons which tend to be more costly.
Adjusted payments for outliers
Table 5. Unintended consequences of CBG payment systems and measures to address them
CBG incen ve Control measures
Upcoding Use a suffi cient number of case groups to allow for economically homogenous
grouping and ensure appropriate payment through regular updates of case
classifi ca on, payment rates and rela ve case weights.
Introduce a global budget for a specifi c case mix and volume.
Undertake coding audits.
Increase volume Introduce a hard or so volume cap.
Apply a reduced case rate for payments over a certain threshold.
Introduce a global budget at regional level to allow for retroac ve calcula on of the
base rate on the basis of the case mix of services provided by all hospitals in the region
and the total budget available for the region.
Undertake regular updates of the CBG case classifi ca on and rela ve weights.
Monitor referral prac ce and wai ng lists.
Reducing quality Introduce clinical audi ng to assess whether diagnosis and treatment of a sample of
cases was performed according to applicable guidelines and pa ent pathways.
Monitor key variables, such as length of stay, rate of readmission, mortality rates.
Introduce specifi c payments for outliers, including signifi cantly longer or shorter
inpa ent stays than the average length of stay for a specifi c treatment, and cost
outliers.
Apply a reduced payment or no payment for readmissions, comorbidi es that were
not encoded at admission but need very mature and well-func oning informa on
and monitoring systems and carry the risk that comorbidi es will not be treated, or
pa ents not readmi ed.
Regularly update the CBG case classifi ca on system and rela ve weights.
Under-provision/
avoiding costly
pa ents
In addi on to the measures listed under reducing quality:
Monitor referrals and wai ng lists.
Focus on procedure instead of diagnosis in the classifi ca on algorithm to ensure that
payment is aligned with treatment.
Source: Compila on based on references (4, 12, 52).
FINA
L DR
AFT
42 HEALTH FINANCING GUIDANCE NO. 10
Checking of audi ng claims, coding
prac ce and data quality is important
to reduce incen ves for upcoding. For
example, audits of DRG payments in France
iden fi ed errors in up to 60% of medical
records analysed. This level of errors, for
which there are various causes, can have
signifi cant fi nancial implica ons. Unclear
defi ni ons and insuffi cient training in
coding can be reasons for disagreements
on how diagnoses should be coded (12).
Some coding mistakes can be considered
as fraudulent. Consequently, procedures to
check data quality systema cally should be
in place and should include:
• automa c checks of compliance with
data standards and format during data
collec on;
• electronic and/or manual checks to
ensure data and coding coherence before
payments are processed;
• retroac ve evalua on and analysis of
data to detect varia ons and outliers
that may refer to coding issues.
Audit and control mechanisms require
intensive capacity-building and a suffi cient
budget on the side of the purchaser. A
sound sampling methodology is important
when selec ng claims to be audited since
the audi ng of claims, including the coding
of medical records, is very expensive.
Electronic and system-wide checks of
claims can to a certain extent reduce the
need for detailed external audits (12).
Audi ng of claims, coding prac ce and data quality
A CBG system alone will not improve
the quality of health services provided.
However, countries have tried to integrate
variables into their CBG system in order to
incen vize providers to improve quality
in service provision (23) at various entry
points, namely:
At the hospital level, the payment amount
per case or the adjustment factors can be
linked to the quality criteria that apply to
all services delivered by the provider. These
payments can be made con ngent upon
the fulfi lment of selected quality criteria
of an accredita on scheme or a quality
assessment framework of the Ministry of
Health (23).
At the level of a CBG, the payment amount
for a specifi c case group can be condi onal
on respec ng specifi c quality or treatment
guidelines for respec ve CBGs, with a
focus on the procedural level. These could
include standards related to availability of
specialized medical equipment, suppor ve
services to provide comprehensive care
(e.g. onco-psychological care) or specialists.
This would require specifi c treatment
guidelines and/or iden fi ca on of best
prac ces. It would also be more diffi cult
to monitor as the purchaser would have
to rely primarily on self-repor ng by the
provider. As a result, such measures are
not applied frequently. One example is
from Germany, where a health insurance
4.4. HOW CAN A CBG SYSTEM IMPROVE QUALITY IN
SERVICE PROVISION?
FINA
L DR
AFT
43IMPLEMENTATION, MONITORING AND CONTINUED REVISION
A CBG system needs con nuous review
and adjustment to take account of changes
in clinical processes and consump on of
resources. This review process also allows
for further fi netuning of the classifi ca on
system and classifi ca on algorithm, and
for further adjustment of case group
weights and tariff s, including the base rate.
Moreover, a review can also help decrease
the scope for upcoding. This may lead to
changes in the number and defi ni on of
case groups and the rela ve cost weights.
In addi on, changes may be needed
that go beyond the CBG system, such as
adapta ons to the data collec on system,
the repor ng system or regula ons related
to the implementa on of the CBG system
(12, 32).
It is helpful to establish a specifi c body or
commi ee to take charge of the revision
process and to propose and adopt
adjustments. This commi ee should
include at the minimum purchaser and
provider representa ves. It is important
to have clear rules and procedures for
4.5. HOW AND HOW OFTEN DOES A CBG SYSTEM NEED TO
BE REVISED?
fund pays higher rates for coronary bypass
surgery if the treatment scores higher than
average on quality indicators that are part
of the German external quality assurance
system (23).
At the individual pa ent level, the
payment amount linked to a (pa ent)
case will be adjusted according to defi ned
quality criteria. The most common example
is a reduced payment – or even none – for
readmission under the same CBG within
a certain me frame or nonpayment for
comorbidi es that were not recorded at
admission and are therefore assumed
to result from treatment (23). These
quality criteria are thus primarily linked to
outcomes.
Including quality criteria in the design of the
CBG system requires addi onal data to be
collected. This can be easier at the provider
level, but more detailed informa on is
needed to introduce such a mechanism for
a specifi c CBG or at pa ent level. Again,
strong informa on management systems
will be needed. CBG payments have the
poten al to incen vize this addi onal data
collec on but monitoring and audi ng
mechanisms must be in place to ensure
data accuracy (23).
An important ques on is how to pay
for treatments based on technological
innova ons that improve quality and how
to induce their adop on. CBG payments
will only encourage providers to adopt
technological innova ons in treatment if
these reduce the cost per case. However,
if providers wish to adopt technological
innova ons that have a benefi cial impact on
treatment quality but are more costly than
conven onal treatment, supplementary
or separate payments or adjustments in
the CBG system may be applied to cover
the higher cost. If a CBG system is updated
frequently, new diagnos c and treatment
op ons can be refl ected through addi onal
case groups or adjusted rela ve weights
(30).
44 HEALTH FINANCING GUIDANCE NO. 10
adjus ng case classifi ca ons, crea ng new
case groups, or removing groups if they are
not appropriate. Clear rules help ensure
transparency in the design and revision
of the CBG system and avoid the process
being taken over by interest groups.
Most countries have created a “case mix
centre” under the Ministry of Health or the
purchasing agency, or as an independent
agency, to be in charge of upda ng the
classifi ca on algorithm. Thailand, for
instance, has a case mix centre under the
Ministry of Health which regularly proposes
updates to the country’s DRG system (28).
A case mix centre can also be involved in
cos ng studies.
If there is no rou ne data collec on process
to generate informa on on changes in the
average cost per case within a given group
or changes in cost structure, regular cos ng
studies will be needed. Given the me it
takes to collect, compile and analyse data,
most revisions of rela ve case weights are
based on data which are between one and
three years old (32).
In most cases, pa ent classifi ca on systems
are updated when necessary. Revisions in
classifi ca on systems may include updates
to the diagnosis classifi ca ons such as ICD-
10 country-specifi c varia ons or procedure
classifi ca ons. These changes must also be
refl ected in the classifi ca on algorithms
and grouper. Several CBG systems have a
process of annual upda ng, such as those
of Thailand or the Nordic Casemix Centre
which maintains the classifi ca on system
and algorithm used in several northern and
eastern European countries. In the la er,
member countries can submit proposals
for the crea on, merging or removal of
case groups (53).
In general, there is pressure from providers
to increase the number of groups over
me and to introduce specifi c mechanisms
to ensure that high-cost services are paid
at higher rates. However, an increased
number of case groups does not necessarily
improve the CBG system. Indeed, a CBG
system with many more case groups can be
more diffi cult to manage, especially from
a purchaser’s perspec ve, as verifi ca on
of the right classifi ca on of cases tends to
become more cumbersome. Consequently,
various countries also remove case groups
that are not used or may merge cases in
order to reduce the number of overall case
groups (5).
Informa on to the general public
should focus on the issues that aff ect
pa ents. Pa ents must be informed
about and understand their en tlements
and obliga ons, including co-payment
requirements, whether balance billing is
allowed and which referral and gate-keeping
rules apply. Pa ents who understand their
rights and obliga ons can also contribute
to checking provider compliance (e.g. they
can report providers who engage in balance
billing through complaints mechanisms
where they are in place).
Agencies in charge of CBG development
might also invite public comments on dra
documents which they make available on
their website, as is the case in Australia and
the United States. Publishing performance
and monitoring indicators can also facilitate
4.6. HOW SHOULD ONE INFORM THE PUBLIC ABOUT
CHANGES IN PAYMENT METHODS AND BILLING
PRACTICES?
FINA
L DR
AFT
n dran dra
ke available on available o
he case in Australia ane case in Australia an
States. Publishing perftates. Publishing pe
monitoring indicators monitoring indicators
45IMPLEMENTATION, MONITORING AND CONTINUED REVISION
engagement by the general public (28).
Purchasers can distribute informa on on
the new payment method through exis ng
channels, including websites, telephone
hotlines, na onal and local media, as
well as distribu ng informa on materials
to providers and local authori es. The
measures taken in Thailand to inform the
public about the country’s health coverage
scheme are described in Box 12.
Box 12. Information provision to the public on the Universal Coverage Scheme, its
benefits and payment system in Thailand
In 2001, Thailand introduced the Universal Coverage Scheme (UCS), a health coverage
scheme for all ci zens outside formal sector employment funded from general
government revenues. The UCS introduced DRG payments for inpa ent care. The
Na onal Health Security Offi ce (NHSO), in charge of opera ng the UCS, recognized the
need for eff ec ve communica on with all stakeholders and used various channels to
support smooth implementa on from the beginning.
The launch of the UCS was publicized through the slogan “30 Baht treats all diseases”.
“30 Baht” (less than US$ 1) refers to the co-payment per visit or per admission at health
facili es, while “treats all diseases” describes the comprehensive benefi t package
covering all diseases, except for a few treatments (e.g. an retroviral treatment, renal
replacement therapy). The list of excluded health services was broadcast widely through
mul ple channels such as television, radio, newspapers and the Internet.
In parallel, providers at subdistrict level were required to register eligible members
through a door-to-door census. These measures made the UCS widely known to eligible
persons, health providers and the general public. Health facili es were further informed
about the new payment method through workshops, mee ngs and offi cial instruc ons
published in the Royal Thai Government Gaze e.
Another measure was to ins tu onalize mechanisms for which health-care providers’
and ci zens’ voices are heard and concerns are rec fi ed. For instance, the NHSO conducts
technical hearings with providers before an upgraded DRG version is implemented. DRG
manuals are published on the NHSO website and in paper. For providers, in addi on to
an annual mee ng with health-care providers on opera onal details of the UCS, the
NHSO also established a helpdesk reachable both by telephone and online to provide
clarifi ca ons related to capita on and DRG payments. Benefi ciaries can reach the NHSO
through a 24/7 call centre to obtain informa on on their benefi ts and rights under the
UCS and to receive assistance to resolve confl icts with providers. Moreover, the NHSO
collects feedback from providers and UCS members during annual public hearings, as
mandated by law. Finally, broadcas ng of rights and en tlements linked to the benefi t
package con nues through mul ple channels, including new pla orms such as YouTube,
Facebook, Instagram and on the NHSO’s own mobile applica on.
In sum, easily understandable messages, clear instruc ons, real- me support for both
benefi ciaries and providers, as well as the use of mul ple communica on channels,
were crucial for smooth implementa on of the UCS and for adop on and acceptance of
the new payment methods.
Na adhanai Rajatanavin, Hathairat Kosiyaporn, Nithiwat Saengruang, Walaiporn
Patcharanarumol and Viroj Tangcharoensathien.
FINAFINA
L D
FINA
T
AL D
RAF
TT
ce of
at Saengruang, Walaip
ol and Viroj Tangcha
46 HEALTH FINANCING GUIDANCE NO. 10
CONCLUDING REMARKS
Using case-based groups for informa on
and payment is one instrument for moving
towards strategic purchasing. However, the
eff ects of a CBG system strongly depend on
the system’s design and how well it aligns
with the overall health system.
The following points must be considered
throughout the processes of design, pilot,
implementa on and monitoring:
• Policy-makers and prac oners should
cri cally refl ect whether the problems
of health-system performance can be
addressed with the envisaged payment
reform – i.e. the introduc on and
opera on of a CBG system.
• When se ng the objec ves to be
achieved by opera ng a CBG system, one
must consider whether the proposed
design of the CBG is appropriate for
maximizing these objec ves or whether
alterna ve design features may be more
appropriate.
• The introduc on and opera on of a CBG
system should be part of a longer-term
strategy with clear objec ves supported
by an implementa on plan that includes
intermediate steps, responsible bodies
and melines.
• A CBG system should be updated
con nually to ensure that it fi ts the
respec ve health system context.
Administra ve measures – including
control – need to be in place to address
any expected undesirable eff ects of a
CBG system.
• There should be con nuing assessment
of whether the key stakeholders – i.e.
the Ministry of Health, the purchaser(s)
and the providers – have the required
capaci es and, if not, how these
capaci es can be generated.
• Policy-makers need to evaluate whether
the proposed design of the CBG system
is in line with the capacity of the health
informa on management system.
• Policy-makers and prac oners should
ask whether supporters and opponents
of the CBG have been iden fi ed and
whether strategies have been developed
to address any concerns.
• Integrated people-centred care and the
objec ves of universal health coverage
must remain at the heart of the CBG, and
any risk that the economic objec ves of
purchasers and providers take priority
over pa ents’ needs must be constantly
mi gated.
FINA
L DR
AFT
47
GLOSSARY
adjustment coeffi cient. A coeffi cient applied to the
base payment rate to adjust payment for the cost
of mee ng the health-service needs of diff erent
popula on groups or legi mate cost diff erences
related to specifi c provider characteris cs (e.g.
being located in a rural or remote area or serving as
a teaching facility).
average length of stay. The average number of bed-
days (inpa ent days) for each pa ent discharged
from the hospital. The average length of stay can be
calculated for an en re facility, a department, or a
diagnosis-related group classifi ca on.
balance billing. The prac ce of a health-care
provider charging a pa ent for the diff erence
between what the purchaser agrees to pay for the
service and what the provider chooses to charge.
base rate. The average payment rate paid by the
purchaser to the provider per unit of service, bundle
of services, or registered individual.
bed-day. A day during which a pa ent stays overnight
in a hospital. Also referred to as an inpa ent day or
pa ent day.
blended payment system. A payment system that
combines elements of mul ple payment methods
to maximize the benefi cial incen ves (and minimize
the perverse incen ves) of each. Also called a
mixed-model payment system.
bundled service payments. The alloca on of a
fi xed payment to a health-care provider to cover all
services, tests and procedures grouped into a higher
aggregated unit (e.g. a service package or hospital
discharge) rather than payment for each individual
service.
capita on payment. A payment method in which
all providers in the payment system are paid a
predetermined fi xed rate in advance to provide a
defi ned set of services to each individual enrolled
with the provider for a fi xed period. Also called per
capita provider payment.
case-based payment. A hospital payment method
that pays hospitals a fi xed amount per admission
or discharge, depending on the pa ent and clinical
characteris cs, which may include department of
admission, diagnosis and other factors. The payment
rate covers all tests, procedures and other services
provided during the hospital stay. If the cases are
grouped according to diagnosis, it is a case-based
diagnosis-related group (DRG) payment method.
case group. A group of hospital cases defi ned for a
case-based hospital payment system that includes
cases with similar clinical characteris cs and that
require similar resources to diagnose and treat
cases or complete a phase of case management.
case group weight. A measure that refl ects the
rela ve use of resources linked to a specifi c case
group in comparison to the overall average for all
case groups. It is calculated by dividing the average
cost per case in a given case group by the global
average cost per case. It refl ects the resource
intensity of diagnosing and trea ng cases in the
case group rela ve to the average (54).
case mix. The average rela ve complexity and
resource intensity of services required to diagnose
and treat pa ents in a hospital due to diagnosis,
disease severity and personal characteris cs such
as age.
GLOSSARY
This glossary is reproduced from: Cashin C, editor. Assessing health provider payment
systems: a practical guide for countries working toward universal health coverage.
Washington (DC): Joint Learning Network for Universal Health Coverage; 2015. There are
a couple of terms included from other sources. For those entries, the source is referenced
directly in the text.
FINA
L DR
AFT
48 HEALTH FINANCING GUIDANCE NO. 10
classifi ca on algorithm/grouper. An algorithm that
assigns hospital cases to groups with associated
rela ve case weights to calculate case mix or fi nal
payment rates for each case in a case-based hospital
payment system. The grouper is o en accompanied
by a computer program, grouper so ware, to run
the algorithm automa cally.
comorbidity. A condi on that is not related causally
to a pa ent’s principal disease process but increases
a pa ent’s total burden of illness (54). It is also
refl ected through secondary diagnoses.
cost. The value of resources (inputs), expressed in
monetary terms, used to produce a good or service,
carry out an ac vity, or achieve a goal.
cost accoun ng methods. Methods that use
accoun ng principles to classify and measure all
costs incurred in producing a good or service, carry
out an ac vity, or achieve a goal.
cost item. An input, or resource, used by providers to
deliver health services to which costs are a ached.
Cost items include both capital and recurrent items.
diagnosis-related group (DRG). A classifi ca on of
hospital case types into groups that are clinically
similar and are expected to have similar hospital
resource use. The groupings are based on diagnoses
and may also include procedures, age, sex and the
presence of complica ons or comorbidi es. DRGs
are an example of a system of case groups and
rela ve case weights. See also case-based payment.
fee-for-service provider payment. A payment
method that pays providers for each individual
service provided. Fees or tariff s are fi xed in advance
for each service or bundle of services.
fi xed-fee schedule. The list of fees or tariff s set in
advance in a fee-for-service payment system.
formula-based payment calcula on. Calcula on
of payments to providers based on a transparent
mathema cal formula with predefi ned parameters
rather than the use of fi xed tariff s or other non-
formula bases.
global budget provider payment. A payment method
that allocates a fi xed amount to a provider for a
specifi ed period to cover aggregate expenditures to
provide an agreed-upon set of services. The budget
can be used fl exibly and is not ed to specifi c line
items for input expenses (e.g. personnel, medicines,
u li es).
hard budget cap. Total amount of resources
allocated to the health sector, or a subsector such
as the hospital sector or all facili es paid using CBG,
which serves as a fi rm limit on expenditures in that
sector during the budget period, which means that
budget or volume overruns will not be compensated
(54).
health purchaser. An en ty that transfers pooled
health care resources to providers to pay for covered
health-care goods, services and interven ons.
Purchasers can include health ministries, social
insurance funds, private insurance funds and other
en es that manage health funds on behalf of the
popula on.
health purchasing. The alloca on of pooled
resources to health-care providers on behalf of the
covered popula on to pay for covered health-care
goods, services and interven ons.
incen ve. An economic signal that directs individuals
(e.g. health workers) or organiza ons (e.g. health
provider ins tu ons) toward self-interested
behaviour. The incen ves created by a provider
payment system will aff ect provider decisions about
the services they deliver, how they deliver them,
and the mix of inputs they use for delivery.
input. A resource (e.g. personnel me, supplies,
equipment) that is used to produce a good or
service, carry out an ac vity, or achieve a goal.
Interna onal Classifi ca on of Diseases (ICD). A tool
for recording, repor ng and grouping condi ons
and factors that infl uence health, medicaments,
infec ous agents, severity, substances, buildings,
devices and all aspects relevant to the descrip on
of injuries. The purpose of the ICD is to allow the
systema c recording, analysis, interpreta on
and comparison of mortality and morbidity data
collected in diff erent countries or areas and at
diff erent mes and across all sectors of the health
system, for epidemiology, case mix, pa ent safety,
primary care and much more. The classifi ca on
system is currently in its 11th edi on (ICD-11) and
is published by the World Health Organiza on (55).
major diagnos c category (MDC). A broad category
of diagnoses generally based on organ systems or
disease e ology that is generally associated with a
par cular medical specialty. MDCs cover the whole
range of the ICD (54, 56).
line-item budget provider payment. The alloca on
of a fi xed amount to a health-care provider for a
49GLOSSARY
specifi ed period to cover specifi c input costs (e.g.
personnel, medicines, u li es).
outlier case. Extreme case which falls outside
predefi ned limits for cost or length of stay. The
outlier case threshold is some mes called the trim
point (54, 56).
output. The result of a produc on process – a good
or service, a completed ac vity, or an achieved goal.
See also unit of payment.
payment cap. A limit on the total payments to a
provider or group of providers under a payment
system.
per diem provider payment. A payment method that
pays a fi xed amount per inpa ent day to hospitals
for each admi ed pa ent. The per diem rate may
vary by department, pa ent, clinical characteris cs,
or other factors.
prospec ve payment. Payment system in which
rates are set in advance and/or providers are paid
before services are delivered.
provider autonomy. Decision rights of a health-care
provider to make key management decisions such
as those related to staffi ng, salaries and bonuses,
use of other inputs, physical assets, organiza onal
structure, resource mobiliza on, output mix and
use of surplus revenue.
provider benchmarking. Comparing the
performance of health-care providers to average
performance or against high performers, using
specifi c indicators or measures.
provider payment. The alloca on of resources to a
health-care provider to deliver the covered package
of health-care goods, services and interven ons to
the covered popula on.
provider payment method. The way in which a
purchaser pays health-care providers to deliver
a service or set of services. A provider payment
method is defi ned primarily by the unit of payment.
See also unit of payment.
provider payment rate. The amount of money that
a purchaser pays to a provider to deliver a service or
set of services under the payment system.
provider payment system. One or more payment
methods and all suppor ng systems, such as
contrac ng and repor ng mechanisms, informa on
management systems and fi nancial management
systems.
public fi nancial management system. The rules
governing how public budgets are created, disbursed
and tracked.
rate-se ng. The process of determining provider
payment rates.
rela ve case weight. A coeffi cient applied to the
base rate in a case-based hospital payment system to
adjust the payment for a case upward or downward
to refl ect the cost of trea ng cases in a par cular
group rela ve to the average cost per case for all
cases. Used to calculate case mix. See also case mix.
rela ve cost. The cost of a good or service as it
compares with the cost of other goods and services,
expressed in terms of a ra o between two costs or
between one cost and a weighted average of all
other goods or services available.
so budget cap. The amount of fi nancial resources
allocated to the health sector, or subsector such as
hospitals or facili es paid using CBG, which serves
as a target, although providers are compensated
for overruns if expenditures or volume of services
provided exceed the predefi ned target of the budget
period (54).
unbundling of services. The prac ce of repor ng
ungrouped individual service components (e.g.
hospital discharge that is ungrouped into bed-days
and into all tests and procedures) instead of the
related aggregate case in order to charge or claim
reimbursement for each unit rather than the higher-
level “bundled” unit (54).
unit of payment. The unit of output for which a
health-care provider is paid under the payment
method – per service, per visit, per case, per bed-
day, or per person per year.
universal health coverage. Ensured access to
essen al health services for an en re popula on
without risk of fi nancial hardship or impoverishment.
upcoding. The o en fraudulent prac ce of coding
hospital cases so that they are assigned to a case
group that is reimbursed at a higher rate or has a
more advantageous rela ve case group weight than
the case group to which the case actually belongs
based on the observed clinical characteris cs of the
case (54).
FINA
L DR
AFT
FINA
L DR
AFT
REFERENCES
50 HEALTH FINANCING GUIDANCE NO. 10
1. Mathauer I, Dale E, Jowe M, Kutzin J. Purchasing health services for universal health
coverage. How to make it more strategic? Geneva: World Health Organiza on; 2019
(Health Financing Policy Brief, No. 6, accessed 29 September 2019).
2. World health report. Health systems fi nancing: the path to universal coverage. Geneva;
World Health Organiza on; 2010).
3. Mathauer I, Dkhimi F. Analy cal guide to assess a mixed provider payment system. Geneva:
World Health Organiza on; 2019 (h ps://apps.who.int/iris/handle/10665/311020,
accessed 13 September, 2019).
4. O’Dougherty S, Cashin C, Samyshkin E, Ibraimova A, Katsaga A, Kutanov E et al. Case-
based hospital payment systems. In: Langenbrunner JC, Cashin C, O’Dougherty S,
editors. Designing and implemen ng health care provider payment systems. How-to
manuals. Washington (DC): The World Bank; 2009 (h ps://openknowledge.worldbank.
org/bitstream/handle/10986/13806/48599 ?sequence=1, accessed 29 September
2019).
5. Kobel C, Thuilliez J, Bellanger M, Pfeiff er K-P. DRG systems and similar pa ent
classifi ca on systems in Europe. In: Busse R, Geissler A, Quen n W, Wiley M, editors.
Diagnosis-related groups in Europe: moving towards transparency, effi ciency and
quality in hospitals. Maidenhead and New York (NY): Open University Press – McGraw-
Hill Educa on; 2011 (h p://www.euro.who.int/__data/assets/pdf_fi le/0004/162265/
e96538 , accessed 29 September 2019).
6. Mathauer I, Wi enbecher F. Hospital payment systems based on diagnosis-related
groups: experiences in low- and middle-income countries. Bull World Health Organ.
2013;91:746–56A. doi: 10.2471/BLT.12.115931.
7. Mathauer I, Wi enbecher F. DRG-based payment systems in low-and middle-income
countries: Implementa on experiences and challenges. Geneva: World Health
Organiza on; 2012 (h p://www.who.int/iris/handle/10665/75150, accessed 29
September 2019.
8. Annear PL, Kwon S, Lorenzoni L, Ducke S, Hun ngton D, Langenbrunner JC et al.
Pathways to DRG-based hospital payment systems in Japan, Korea, and Thailand.
Health Policy. 2018;122:707–13. doi: 10.1016/j.healthpol.2018.04.013.
9. Paat-Ahi G, Aaviksoo A, Swiderek M. Cholecystectomy and diagnosis-related groups
(DRGs): pa ent classifi ca on and hospital reimbursement in 11 European countries. Int
J Health Policy Manag. 2014;3:383–91. doi: 10.15171/ijhpm.2014.121
FINA
L DR
AFT
51REFERENCES
10. Quen n W, Scheller-Kreinsen D, Geissler A, Busse R. Appendectomy and diagnosis-
related groups (DRGs): pa ent classifi ca on and hospital reimbursement in 11 European
countries. Langenbecks Arch Surg. 2012;397:317–26. doi: 10.1007/s00423-011-0877-5
11. Peltola M, Quen n W. Diagnosis-related groups for stroke in Europe: pa ent classifi ca on
and hospital reimbursement in 11 countries. Cerebrovasc Dis. 2013;35:113–23. doi:
10.1159/000346092
12. Cots F, Chiarello P, Salvador X, Castells X, Quen n W. DRG-based hospital payment:
intended and unintended consequences. In: Busse R, Geissler A, Quen n W, Wiley M,
editors. Diagnosis-related groups in Europe: moving towards transparency, effi ciency and
quality in hospitals. Maidenhead and New York (NY): Open University Press – McGraw-
Hill Educa on; 2011 (h p://www.euro.who.int/__data/assets/pdf_fi le/0004/162265/
e96538 , accessed 29 September 2019).
13. Geissler A, Scheller-Kreinsen D, Quen n W, Busse R. Germany: understanding DRGs. In:
Busse R, Geissler A, Quen n W, Wiley M, editors. Diagnosis-related groups in Europe:
moving towards transparency, effi ciency and quality in hospitals. Maidenhead and New
York (NY): Open University Press – McGraw-Hill Educa on; 2011 (h p://www.euro.
who.int/__data/assets/pdf_fi le/0004/162265/e96538 , accessed 29 September
2019).
14. Ducke S. Issues in the design and implementa on of case-based payment. In: Annear
PL, Hun ngton D, editors. Case-based payment systems for hospital funding in Asia: an
inves ga on of current status and future direc ons. Geneva: World Health Organiza on;
2015 (Asia Pacifi c Observatory on Health Systems and Policies, Compara ve Country
Studies 1:2, h p://apps.searo.who.int/PDS_DOCS/B5325 , accessed 29 September
2019).
15. Quen n W, Geissler A, Scheller-Kreinsen D, Busse R. Understanding DRGs and DRG-
based hospital payment in Europe. In: Busse R, Geissler A, Quen n W, Wiley M, editors.
Diagnosis-related groups in Europe: moving towards transparency, effi ciency and
quality in hospitals. Maidenhead and New York (NY): Open University Press – McGraw-
Hill Educa on; 2011 (h p://www.euro.who.int/__data/assets/pdf_fi le/0004/162265/
e96538 , accessed 29 September 2019).
16. Arrêté du ministre de la santé n° 239-07 du 16 safar 1428 (6 mars 2007) portant
approba on de l’avenant n° 1 à la conven on na onale conclue entre les organismes
ges onnaires de l’assurance maladie obligatoire et les médecins et les établissements
de soins du secteur privé. Rabat: Ministry of Health; 2007 (h ps://www.sante.gov.ma/
Reglementa on/Pages/ASSURANCE-MALADIE.aspx, accessed 27 August 2019).
17. New PhilHealth case rates for selected medical cases and surgical procedures and the no
balance billing policy. Manila: Philippines Health Insurance Corpora on; 2011 (h ps://
www.philhealth.gov.ph/circulars/2011/circ11_2011 , accessed 29 September
2019).
18. Kipp R, Baveja L. Developing na ons can gain from be er bundled payment informa on.
Milliman White Paper, 2013.
52 HEALTH FINANCING GUIDANCE NO. 10
19. Cashin C, Ankhbayar B, Phuong HT, Jamsran G, Nanzad O, Phuong NK et al. Assessing
health provider payment systems: a prac cal guide for countries working toward
universal health coverage. Washington (DC): Joint Learning Network for Universal
Health Coverage; 2015 (accessed 29 September 2019).
20. Langenbrunner JC, Cashin C, O’Dougherty S. What, how and who: an introduc on to
provider payment systems. In: Langenbrunner JC, Cashin C, O’Dougherty S, editors.
Designing and implemen ng health care provider payment systems. How-to manuals.
Washington (DC): The World Bank; 2009 (h ps://openknowledge.worldbank.org/
bitstream/handle/10986/13806/48599 ?sequence=1, accessed 29 September
2019).
21. Annear PL, Xu K. Contextual issues for DRG-based hospital payment. In: Annear PL,
Hun ngton D, editors. Case-based payment systems for hospital funding in Asia: an
inves ga on of current status and future direc ons. Geneva: World Health Organiza on;
2015 (Asia Pacifi c Observatory on Health Systems and Policies, Compara ve Country
Studies 1:2, h p://apps.searo.who.int/PDS_DOCS/B5325 , accessed 29 September
2019).
22. O’Reilly J, Busse R, Häkkinen U, Or Z, Street A, Wiley M. Paying for hospital care: the
experience with implemen ng ac vity-based funding in fi ve european countries.
Health economics, policy and law. 2012;7:73-101.
23. Or Z, Häkkinen U. DRGs and quality: for be er or worse? In: Busse R, Geissler A, Quen n
W, Wiley M, editors. Diagnosis-related groups in Europe: moving towards transparency,
effi ciency and quality in hospitals. Maidenhead and New York (NY): Open University
Press – McGraw-Hill Educa on; 2011 (h p://www.euro.who.int/__data/assets/pdf_
fi le/0004/162265/e96538 , accessed 29 September 2019).
24. Street A, O’Reilly J, Ward P, Mason A. DRG-based hospital payment and effi ciency:
theory, evidence, and challenges. In: Busse R, Geissler A, Quen n W, Wiley M, editors.
Diagnosis-related groups in Europe: moving towards transparency, effi ciency and
quality in hospitals. Maidenhead and New York (NY): Open University Press – McGraw-
Hill Educa on; 2011 (h p://www.euro.who.int/__data/assets/pdf_fi le/0004/162265/
e96538 , accessed 29 September 2019).
25. Geissler A, Quen n W, Scheller-Kreinsen D, Busse R. Introduc on to DRGs in Europe:
common objec ves across diff erent hospital systems. In: Busse R, Geissler A, Quen n
W, Wiley M, editors. Diagnosis-related groups in Europe: moving towards transparency,
effi ciency and quality in hospitals. Maidenhead and New York (NY): Open University
Press – McGraw-Hill Educa on; 2011 (h p://www.euro.who.int/__data/assets/pdf_
fi le/0004/162265/e96538 , accessed 29 September 2019).
26. Pongpirul K, Walker DG, Winch PJ, Robinson C. A qualita ve study of DRG coding
prac ce in hospitals under the Thai universal coverage scheme. BMC Health Serv Res.
2011;11:71. doi: 10.1186/1472-6963-11–71).
FINA
L DR
AFT
53REFERENCES
27. Streveler DJ, Sherlock SM. Health management informa on systems: linking purchasers
and providers. In: Langenbrunner JC, Cashin C, O’Dougherty S, editors. Designing and
implemen ng health care provider payment systems. How-to manuals. Washington
(DC): The World Bank; 2009 (h ps://openknowledge.worldbank.org/bitstream/
handle/10986/13806/48599 ?sequence=1, accessed 29 September 2019).
28. Bredenkamp C, Bales S, Kahur K (eds). Transi on to DRG Payments for Health: Lessons
from Case Studies. Interna onal Development in Focus. Washington (DC): The World
Bank; 2019.
29. Langenbrunner JC. Conclusions and lessons for the region. In: Annear PL, Hun ngton
D, editors. Case-based payment systems for hospital funding in Asia: an inves ga on
of current status and future direc ons. Geneva: World Health Organiza on; 2015 (Asia
Pacifi c Observatory on Health Systems and Policies, Compara ve Country Studies 1:2,
h p://apps.searo.who.int/PDS_DOCS/B5325 , accessed 29 September 2019).
30. Cashin C, Bloom D, Sparkes S, Barroy H, Kutzin J, O’Dougherty S. Aligning public
fi nancial management and health fi nancing: sustaining progress towards universal
health coverage. Geneva: World Health Organiza on; 2017 (h ps://apps.who.int/iris/
bitstream/handle/10665/254680/9789241512039-eng ?sequence=1, accessed 29
September 2019).
31. Cashin C, Ankhbayar B, Tsilaajav T, Nanzad O, Jamsran G, Somanathan A. Assessment of
systems for paying health care providers in Mongolia: implica ons for equity, effi cency
and universal health coverage. Washington (DC): The World Bank; 2016.
32. Quen n W, Scheller-Kreinsen., Busse R. Technological innova on in DRG-based hospital
payment systems across Europe. In: Busse R, Geissler A, Quen n W, Wiley M, editors.
Diagnosis-related groups in Europe: moving towards transparency, effi ciency and
quality in hospitals. Maidenhead and New York (NY): Open University Press – McGraw-
Hill Educa on; 2011 (h p://www.euro.who.int/__data/assets/pdf_fi le/0004/162265/
e96538 , accessed 29 September 2019).
33. Sparkes SP, Bump JB, Özçelik EA, Kutzin J, Reich MR. Poli cal economy analysis for health
fi nancing reform. Health Syst Reform. 2019:1–12. doi: 10.1080/23288604.2019.1633874.
34. Roberts M, Hsiao W, Berman P, Reich MR. Ge ng health reform right: a guide to
improving performance and equity. New York (NY): Oxford University Press; 2004.
35. Kwon S, Shon C. Implementa on experience with DRG-based payments. In: Annear PL,
Hun ngton D, editors. Case-based payment systems for hospital funding in Asia: an
inves ga on of current status and future direc ons. Geneva: World Health Organiza on;
2015 (Asia Pacifi c Observatory on Health Systems and Policies, Compara ve Country
Studies 1:2, h p://apps.searo.who.int/PDS_DOCS/B5325 , accessed 29 September
2019).
FINA
L DR
AFT
FINA
L DR
AFT
54 HEALTH FINANCING GUIDANCE NO. 10
36. Busse R, Quen n W. Moving towards transparency, effi ciency, and quality in hospitals:
conclusions and recommenda ons. In: Busse R, Geissler A, Quen n W, Wiley M, editors.
Diagnosis-related groups in Europe: moving towards transparency, effi ciency and
quality in hospitals. Maidenhead and New York (NY): Open University Press – McGraw-
Hill Educa on; 2011 (h p://www.euro.who.int/__data/assets/pdf_fi le/0004/162265/
e96538 , accessed 29 September 2019).
37. Tan SS, Serden L, Geissler A, Van Ineveld M, Redekap K, Heugren M et al. DRGs and cost
accoun ng: which is driving which? In: Busse R, Geissler A, Quen n W, Wiley M, editors.
Diagnosis-related groups in Europe: moving towards transparency, effi ciency and
quality in hospitals. Maidenhead and New York (NY): Open University Press – McGraw-
Hill Educa on; 2011 (h p://www.euro.who.int/__data/assets/pdf_fi le/0004/162265/
e96538 , accessed 29 September 2019).
38. Cashin C, Gubonova O, Kadyrova N, Khe N, Kutanov E, McEuen M et al. Primary
health care per capita payment systems. In: Designing and implemen ng health care
provider payment systems. How-to manuals. Washington (DC): The World Bank; 2009
(https://openknowledge.worldbank.org /bitstream/handle/10986/13806/48599.
pdf?sequence=1, accessed 29 September 2019).
39. Hopfe M, Stucki G, Marshall R, Twomey CD, Üstün TB, Prodinger B. Capturing pa ents’
needs in casemix: a systema c literature review on the value of adding func oning
informa on in reimbursement systems. BMC Health Serv Res. 2016;16:40. doi: 10.1186/
s12913-016-1277-x.
40. Interna onal classifi ca on of health interven ons. Geneva: World Health Organiza on;
2019 (h ps://mitel.dimi.uniud.it/ichi/docs/#introduc on, accessed 29 August 2019).
41. Sabi W. An evalua on of the DRG-based payment system of the Na onal Health
Insurance Authority of Ghana. 2013. (unpublished)
42. Lucyk K, Lu M, Sajobi T, Quan H. Disease groupings: what are they, how are they used,
and how do they compare interna onally? Perspect Health Inf Manag. 2016:1–12.
43. Design and development of the diagnosis-related group (DRG). Bal more: Centers for
Medicare & Medicaid Services; 2016 (h ps://www.cms.gov/ICD10Manual/version34-
fullcode-cms/fullcode_cms/Design_and_development_of_the_Diagnosis_Related_
Group_(DRGs)_PBL-038 , accessed 31 October 2019).
44. Strengthening health fi nancing systems in the Eastern Mediterranean Region towards
universal health coverage. Health fi nancing atlas 2018. Cairo: World Health Organiza on
Regional Offi ce for the Eastern Mediterranean; 2019 (h p://applica ons.emro.who.
int/docs/EMROPUB_2019_EN_22347 ?ua=1, accessed 27 August 2019).
45. Georgia: building an effi cient and transparent fi nancing system for UHC. Geneva:
Universal Health Coverage Partnership; 2019 (h ps://uhcpartnership.net/wp-
content/uploads/2019/04/JWT-Georgia-story-from-the-fi eld-web-fi nal-April-2019 ,
accessed 30 August 2019).
FINA
L DR
AFT
55REFERENCES
46. 20 years of compulsory health insurance and 25 years of the Na onal Health
Insurance Fund under the Ministry of Health – An overview. Vilnius: Na onal Health
Insurance Fund of Lithuania; 2017 (h p://www.vlk.lt/sites/en/news/Documents/VLK_
AP%c5%bdVALGA%202017 , accessed 31 October 2019).
47. Overview of Estonian experiences with DRG system. Tallinn: Estonian Health Insurance
Fund; 2009 (h ps://www.haigekassa.ee/uploads/userfi les/Implementa on_DRG_
EST_291209_cover%281%29 , accessed 9 September 2019).
48. Özal n A, Cashin C. Cos ng of health services for provider payment: a prac cal manual
based on country cos ng challenges, trade-off s, and solu ons. Washington (DC): Results
for Development Ins tute/Joint Learning Network for Universal Health Coverage; 2014.
(h p://www.jointlearningnetwork.org/uploads/fi les/resources/JLN_Cos ng_Toolkit_
Interac ve_FINAL , accessed 9 September 2019).
49. Na onal effi cient price determina on 2019–2020. Sydney: Independent Hospital
Pricing Authority; 2019 (h ps://www.ihpa.gov.au/sites/default/fi les/publica ons/
na onal_effi cient_price_determina on_2019-20 , accessed 9 September 2019).
50. de Lagasnerie G, Paris V, Mueller M, Kumar A. Tapering payments in hospitals.
Experiences in OECD countries. Paris: Organisa on for Economic Co-opera on and
Development; 2015 (OECD Health Working Papers, No. 78) doi: doi:h ps://doi.
org/10.1787/5js4h5rlkxxr-en.
51. OECD, WHO, Barber SL, Lorenzoni L, Ong P. Price se ng and price regula on in health
care. Paris: Organisa on for Economic Co-opera on and Development; 2019 (h ps://
www.oecd-ilibrary.org/content/publica on/ed3c16ff -en, accessed 29 September
2019).
52. Murakami Y, Lorenzoni L. Assessing the impact of case-based payment. In: Annear PL,
Hun ngton D, editors. Case-based payment systems for hospital funding in Asia: an
inves ga on of current status and future direc ons. Geneva: World Health Organiza on;
2015 (Asia Pacifi c Observatory on Health Systems and Policies, Compara ve Country
Studies 1:2, h p://apps.searo.who.int/PDS_DOCS/B5325 , accessed 29 September
2019).
53. NordDRG users’ manual, version 2012 NC PR1. Helsinki: NordCase Nordic Casemix
Centre; 2012 (h p://www.norddrg.net/norddrgmanual/NordDRG_2012_NC/index.
htm, accessed 10 September 2019).
54. Langenbrunner J, Cashin C, O’Dougherty S, editors. Designing and implemen ng health
care provider payment systems. How-to manuals. Washington (DC): The World Bank;
2009 (h ps://openknowledge.worldbank.org/bitstream/handle/10986/13806/48599.
pdf?sequence=1, accessed 29 September 2019).
55. Interna onal classifi ca on of diseases for mortality and morbidity sta s cs, eleventh
revision. Reference guide. Geneva: World Health Organiza on; 2019 (h ps://icd.who.
int/icd11refguide/en/index.html, accessed 29 September 2019).
56. Bardsley M, Coles J, Jenkins L. DRGs and health care. The management of case mix.
London: King’s Fund Publishing Offi ce; 1989.
ANNEX 1: EXAMPLES OF
CLASSIFICATION SYSTEMS
WITH DIFFERENT LEVELS OF
COMPLEXITY
56 HEALTH FINANCING GUIDANCE NO. 10
FINA
L
57ANNEX 1
FINA
L DR
AFT
58 HEALTH FINANCING GUIDANCE NO. 10
ANNEX 2: MAIN PROVIDER
PAYMENT METHODS AND THE
INCENTIVES THEY CREATE
Payment
method
Defi ni on Incen ves
Line-item
budget
Providers receive a fi xed amount to cover specifi c input expenses
(e.g. staff , drugs, …).
Under-provision
Per diem Hospitals are paid a fi xed amount per day that an admi ed
pa ent is treated in the hospital.
Extended length of stay,
reduced cost per case;
cream-skimming
Case-based
(“DRG”)
Hospitals are paid a fi xed amount per admission depending on
pa ent and clinical characteris cs.
Increase of volume,
reduc on of costs per
case, avoidance of severe
cases
Global
budget
Providers receive a fi xed amount of funds for a certain period to
cover aggregate expenditures. Budget is fl exible and not ed to
line items.
Under-provision, also in
terms of quality
Fee-for-
service
Providers are paid for each individual service provided. Fees are
fi xed in advance for each service or group of services.
Over-provision
Capita on Providers are paid a fi xed amount in advance to provide a defi ned
set of services for each individual enrolled for a fi xed period of
me.
Under-provision
Source: References (1, 19).
Reproduced from Mathauer I, Dale E, Jowe M, Kutzin J. Purchasing health services
for universal health coverage: How to make it more strategic? Geneva: World Health
Organiza on; 2019.
FINA
L DR
AFT
59ANNEX 3
ANNEX 3: LIST OF MAJOR
DIAGNOSTIC CATEGORIES (MDC)
OF TWO DRG VARIANTS
List of MDC of the Medicare DRGs 1985 (56)
MDC 1 – Diseases and disorders of the nervous system
MDC 2 – Diseases and disorders of the eye
MDC 3 – Diseases and disorders of the ear, nose and
throat
MDC 4 – Diseases and disorders of the respiratory system
MDC 5 – Diseases and disorders of the circulatory system
MDC 6 – Diseases and disorders of the diges ve system
MDC 7 – Diseases and disorders of the hepatobiliary system
and pancreas
MDC 8 – Diseases and disorders of the musculoskeletal
system and connec ve ssue
MDC 9 – Diseases and disorders of the skin, subcutaneous
ssue and breast
MDC 10 – Endocrine, nutri onal and metabolic diseases
and disorders
MDC 11 – Diseases and disorders of the kidney and urinary
tract
MDC 12 – Diseases and disorders of the male reproduc ve
system
MDC 13 – Diseases and disorders of the female
reproduc ve system
MDC 14 – Pregnancy, childbirth and the puerperium
MDC 15 – Newborns and other neonates with condi ons
origina ng in the principal period
MDC 16 – Diseases and disorders of the blood and blood-
forming organs and immunological disorders
MDC 17 – Myeloprolifera ve diseases and disorders and
poorly diff eren ated neoplasms
MDC 18 – Infec ous and parasi c diseases (systemic or
unspecifi ed sites)
MDC 19 – Mental diseases and disorders
MDC 20 – Substance use and substance induced organic
mental disorders
MDC 21 – Injury, poisoning and toxic eff ects of drugs
MDC 22 – Burns
MDC 23 – Factors infl uencing health status and other
contacts with health services
List of MDC of the NordDRG Users’ Manual Version 2012 (53)
MDC 1 – Diseases of the nervous system
MDC 2 – Diseases and disorders of the eye
MDC 3 – Diseases and disorders of ear, nose, mouth and
throat
MDC 4 – Diseases and disorders of the respiratory system
MDC 5 – Diseases and disorders of the circulatory system
MDC 6 – Diseases and disorders of the diges ve system
MDC 7 – Diseases and disorders of the hepatobiliary system
and pancreas
MDC 8 – Diseases and disorders of musculosceletal system
and connec ve ssue
MDC 9 – Diseases and disorders of skin and subcutaneous
ssue
MDC 30 – Diseases and disorders of breast
MDC 10 – Endocrine, nutri onal and metabolic diseases and
disorders
MDC 11 – Diseases and disorders of kidney and urinary tract
MDC 12 – Diseases and disorders of the male reproduc ve
system
MDC 13 – Diseases and disorders of the female reproduc ve
system
MDC 14 – Diseases and disorders of the pregnancy, childbirth
and puerperium
MDC 15 – Neonatal DRG
MDC 16 – Diseases and disorders of the blood and blood
forming organs and immunological disorders
MDC 17 – Myeloprolifera ve diseases and disorders, poorly
diff eren ated neoplasms
MDC 18 – Infec ous and parasi c diseases
MDC 19 – Mental diseases and disorders
MDC 20 – Alcohol and drug use and alcohol and drug induced
mental disorders
MDC 21 – Injuries, poisonings and toxic eff ects of drugs
MDC 22 – Burns
MDC 23 – Factors infl uencing health status and other
contacts with health services
MDC 24 – Mul ple signifi cant trauma
MDC 25 – HIV infec on
FINA
L DR
AFT
European Observatory on Health Systems and Policies Series
Diagnosis-Related Groups in Europe
Moving towards transparency, efficiency and quality in hospitals
Diagnosis-Related Group (DRG) systems were introduced in Europe to increase
the transparency of services provided by hospitals and to incentivize greater
efficiency in the use of resources invested in acute hospitals. In many countries,
these systems were also designed to contribute to improving – or at least
protecting – the quality of care. After more than a decade of experience with
using DRGs in Europe, this book considers whether the extensive use of DRGs
has contributed towards achieving these objectives.
Written by authors with extensive experience of these systems, this book is a
product of the EuroDRG project and constitutes an important resource for health
policy-makers and researchers from Europe and beyond. The book is intended
to contribute to the emergence of a ‘common language’ that will facilitate
communication between researchers and policy-makers interested in
improving the functioning and resourcing of the acute hospital sector. The book
includes:
• A clearly structured introduction to the main ‘building blocks’ of DRG
systems
• An overview of key issues related to DRGs including their impact on efficiency,
quality, unintended effects and technological innovation in health care
• 12 country chapters – Austria, England, Estonia, Finland, France, Germany,
Ireland, the Netherlands, Poland, Portugal, Spain and Sweden
• Clearly structured and detailed information about the most important DRG
system characteristics in each of these countries
• Useful insights for countries and regions in Europe and beyond interested
in introducing, extending and/or optimizing DRG systems within the
hospital sector
Reinhard Busse is Professor and Head of the Department of Healthcare
Management at Berlin University of Technology, Germany.
Alexander Geissler is a research fellow in the Department of Health Care
Management at Berlin University of Technology, Germany.
Wilm Quentin is a research fellow in the Department of Health Care
Management at Berlin University of Technology, Germany.
Miriam M. Wiley is Professor and Head of the Health Research and Information
Division at the Economic and Social Research Institute in Dublin, Ireland.
www.openup.co.uk
Diagnosis-Related
Groups in Europe
Moving towards transparency, efficiency
and quality in hospitals
Edited by
Reinhard Busse
Alexander Geissler
Wilm Quentin
Miriam Wiley
D
ia
gn
o
sis-R
ela
ted
G
ro
u
p
s in
Eu
ro
p
e
Busse, G
eissler, Q
uentin and W
iley
Diagnosis Related Groups…pb24.5_Diagnosis Related Groups…pb24.5 23/10/2011 11:32 Page 1
Diagnosis-Related
Groups in Europe
The European Observatory on Health Systems and Policies supports and
promotes evidence-based health policy-making through comprehensive and
rigorous analysis of health systems in Europe. It brings together a wide range of
policy-makers, academics and practitioners to analyse trends in health reform,
drawing on experience from across Europe to illuminate policy issues.
The European Observatory on Health Systems and Policies is a partnership
between the World Health Organization Regional Offi ce for Europe, the
Governments of Belgium, Finland, Ireland, the Netherlands, Norway, Slovenia,
Spain, Sweden and the Veneto Region of Italy, the European Commission, the
European Investment Bank, the World Bank, UNCAM (French National Union
of Health Insurance Funds), the London School of Economics and Political
Science, and the London School of Hygiene & Tropical Medicine.
Diagnosis-Related
Groups in Europe
Moving towards transparency,
effi ciency and quality in hospitals
Edited by
Reinhard Busse, Alexander Geissler,
Wilm Quentin, Miriam Wiley
Open University Press
Open University Press
McGraw-Hill Education
McGraw-Hill House
Shoppenhangers Road
Maidenhead
Berkshire
England
SL6 2QL
email: enquiries@openup.co.uk
world wide web: www.openup.co.uk
and Two Penn Plaza, New York, NY 10121-2289, USA
First published 2011
Copyright © World Health Organization 2011 on behalf of the European Observatory on
Health Systems and Policies.
The views expressed by authors or editors do not necessarily represent the decisions or the stated
policies of the European Observatory on Health Systems and Policies or any of its partners. The
designations employed and the presentation of the material in this publication do not imply
the expression of any opinion whatsoever on the part of the European Observatory on Health
Systems and Policies or any of its partners concerning the legal status of any country, territory,
city or area or of its authorities, or concerning the delimitations of its frontiers or boundaries.
Where the designation ‘country or area’ appears in the headings of tables, it covers countries,
territories, cities, or areas. Dotted lines on maps represent approximate border lines for which
there may not yet be full agreement.
The mention of specific companies or of certain manufacturers’ products does not imply
that they are endorsed or recommended by the European Observatory on Health Systems and
Policies in preference to others of a similar nature that are not mentioned. Errors and omissions
excepted, the names of proprietary products are distinguished by initial capital letters. The
European Observatory on Health Systems and Policies does not warrant that the information
contained in this publication is complete and correct and shall not be liable for any damages
incurred as a result of its use.
Rights to translate into German, Spanish, French and Russian should be sought from WHO at
WHO Regional Office for Europe, Scherfigsvej 8, DK-2100 Copenhagen, Denmark or by email
at pubrights@euro.who.int. Rights to translate into all other world languages should be sought
from Open University Press.
All rights reserved. Except for the quotation of short passages for the purpose of criticism
and review, no part of this publication may be reproduced, stored in a retrieval system, or
transmitted, in any form or by any means, electronic, mechanical, photocopying, recording
or otherwise, without the prior written permission of the publisher or a licence from the
Copyright Licensing Agency Limited. Details of such licences (for reprographic reproduction)
may be obtained from the Copyright Licensing Agency Ltd of Saffron House, 6–10 Kirby Street,
London, EC1N 8TS.
A catalogue record of this book is available from the British Library
ISBN-13: 978-0-33-524557-4
ISBN-10: 0-33-524557-9
eISBN: 978-0-33-524558-1
Library of Congress Cataloging-in-Publication Data
CIP data applied for
Typeset by Refi neCatch Limited, Bungay, Suffolk
Printed in Great Britain by Ball and Bain Ltd, Glasgow
Fictitious names of companies, products, people, characters and/or data that may be used herein
(in case studies or in examples) are not intended to represent any real individual, company,
product or event.
European Observatory on Health Systems and Policies Series
The European Observatory on Health Systems and Policies is a unique project that
builds on the commitment of all its partners to improving health systems:
• World Health Organization Regional Offi ce for Europe
• Government of Belgium
• Government of Finland
• Government of Ireland
• Government of the Netherlands
• Government of Norway
• Government of Slovenia
• Government of Spain
• Government of Sweden
• Government of Veneto Region of Italy
• European Commission
• European Investment Bank
• World Bank
• UNCAM (French National Union of Health Insurance Funds)
• London School of Economics and Political Science
• London School of Hygiene & Tropical Medicine
The series
The volumes in this series focus on key issues for health policy-making in Europe. Each
study explores the conceptual background, outcomes and lessons learned about the
development of more equitable, more effi cient and more effective health systems in
Europe. With this focus, the series seeks to contribute to the evolution of a more evidence-
based approach to policy formulation in the health sector.
These studies will be important to all those involved in formulating or evaluating national
health policies and, in particular, will be of use to health policy-makers and advisers,
who are under increasing pressure to rationalize the structure and funding of their health
system. Academics and students in the fi eld of health policy will also fi nd this series
valuable in seeking to understand better the complex choices that confront the health
systems of Europe.
The Observatory supports and promotes evidence-based health policy-making through
comprehensive and rigorous analysis of the dynamics of health care systems in Europe.
Series Editors
Josep Figueras is Director of the European Observatory on Health Systems and Policies,
and Head of the European Centre for Health Policy, World Health Organization Regional
Offi ce for Europe.
Martin McKee is Head of Research Policy and Head of the London Hub of the European
Observatory on Health Systems and Policies. He is Professor of European Public Health at
the London School of Hygiene & Tropical Medicine as well as a co-director of the School’s
European Centre on Health of Societies in Transition.
Elias Mossialos is the Co-director of the European Observatory on Health Systems and
Policies. He is Brian Abel-Smith Professor in Health Policy, Department of Social Policy,
London School of Economics and Political Science and Director of LSE Health.
Richard B. Saltman is Associate Head of Research Policy and Head of the Atlanta Hub of the
European Observatory on Health Systems and Policies. He is Professor of Health Policy and
Management at the Rollins School of Public Health, Emory University in Atlanta, Georgia.
Reinhard Busse is Associate Head of Research Policy and Head of the Berlin Hub of the
European Observatory on Health Systems and Policies. He is Professor of Health Care
Management at the Berlin University of Technology.
European Observatory on Health Systems and
Policies Series
Series Editors: Josep Figueras, Martin McKee, Elias Mossialos,
Richard B. Saltman and Reinhard Busse
Published titles
Regulating entrepreneurial behaviour in European health care systems
Richard B. Saltman, Reinhard Busse and Elias Mossialos (eds)
Hospitals in a changing Europe
Martin McKee and Judith Healy (eds)
Health care in central Asia
Martin McKee, Judith Healy and Jane Falkingham (eds)
Funding health care: options for Europe
Elias Mossialos, Anna Dixon, Josep Figueras and Joe Kutzin (eds)
Health policy and European Union enlargement
Martin McKee, Laura MacLehose and Ellen Nolte (eds)
Regulating pharmaceuticals in Europe: striving for effi ciency, equity and quality
Elias Mossialos, Monique Mrazek and Tom Walley (eds)
Social health insurance systems in western Europe
Richard B. Saltman, Reinhard Busse and Josep Figueras (eds)
Purchasing to improve health systems performance
Josep Figueras, Ray Robinson and Elke Jakubowski (eds)
Human resources for health in Europe
Carl-Ardy Dubois, Martin McKee and Ellen Nolte (eds)
Primary care in the driver’s seat
Richard B. Saltman, Ana Rico and Wienke Boerma (eds)
Mental health policy and practice across Europe: the future direction of mental
health care
Martin Knapp, David McDaid, Elias Mossialos and Graham Thornicroft
(eds)
Decentralization in health care
Richard B. Saltman, Vaida Bankauskaite and Karsten Vrangbæk (eds)
Health systems and the challenge of communicable diseases: experiences from
Europe and Latin America
Richard Coker, Rifat Atun and Martin McKee (eds)
Caring for people with chronic conditions: a health system perspective
Ellen Nolte and Martin McKee (eds)
Nordic health care systems: recent reforms and current policy challenges
Jon Magnussen, Karsten Vrangbæk and Richard B. Saltman (eds)
Forthcoming titles
Migration and health in the European Union
Bernd Rechel, Philipa Mladovsky, Walter Devillé, Barbara Rijks, Roumyana
Petrova-Benedict and Martin McKee (eds)
Contents
List of tables, fi gures and boxes xi
Abbreviations xvii
Foreword 1 xxi
Foreword 2 xxv
Acknowledgements xxvii
List of contributors xxix
Part One
one From the origins of DRGs to their
implementation in Europe 3
Miriam Wiley
two Introduction to DRGs in Europe: Common
objectives across different hospital systems 9
Alexander Geissler, Wilm Quentin,
David Scheller-Kreinsen and Reinhard Busse
three Understanding DRGs and DRG-based hospital
payment in Europe 23
Wilm Quentin, Alexander Geissler,
David Scheller-Kreinsen and Reinhard Busse
viii Contents
four DRG systems and similar patient
classifi cation systems in Europe 37
Conrad Kobel, Josselin Thuilliez,
Martine Bellanger and Karl-Peter Pfeiffer
fi ve DRGs and cost accounting: Which is
driving which? 59
Siok Swan Tan, Lisbeth Serdén, Alexander Geissler,
Martin van Ineveld, Ken Redekop, Mona Heurgren
and Leona Hakkaart-van Roijen
six DRG-based hospital payment: Intended and
unintended consequences 75
Francesc Cots, Pietro Chiarello, Xavier Salvador,
Xavier Castells and Wilm Quentin
seven DRG-based hospital payment and effi ciency:
Theory, evidence, and challenges 93
Andrew Street, Jacqueline O’Reilly,
Padraic Ward and Anne Mason
eight DRGs and quality: For better or worse? 115
Zeynep Or and Unto Häkkinen
nine Technological innovation in DRG-based
hospital payment systems across Europe 131
Wilm Quentin, David Scheller-Kreinsen and
Reinhard Busse
ten Moving towards transparency, effi ciency
and quality in hospitals: Conclusions and
recommendations 149
Reinhard Busse and Wilm Quentin
Part Two
eleven Austria: Inpatient care and the LKF
framework 175
Conrad Kobel and Karl-Peter Pfeiffer
twelve England: The Healthcare Resource
Group system 197
Anne Mason, Padraic Ward and Andrew Street
thirteen France: Implementing homogeneous patient
groups in a mixed market 221
Zeynep Or and Martine Bellanger
Contents ix
fourteen Germany: Understanding G-DRGs 243
Alexander Geissler, David Scheller-Kreinsen, Wilm Quentin
and Reinhard Busse
fi fteen Ireland: A review of casemix applications
within the acute public hospital system 273
Jacqueline O’Reilly, Brian McCarthy
and Miriam Wiley
sixteen NordDRG: The benefi ts of coordination 293
Miika Linna and Martti Virtanen
seventeen Estonia: Developing NordDRGs within
social health insurance 301
Kristiina Kahur, Tõnis Allik, Ain Aaviksoo,
Heli Laarmann and Gerli Paat
eighteen Finland: DRGs in a decentralized health
care system 321
Kirsi Kautiainen, Unto Häkkinen and Jorma Lauharanta
nineteen Sweden: The history, development and
current use of DRGs 339
Lisbeth Serdén and Mona Heurgren
twenty Poland: The Jednorodne Grupy
Pacjentów – Polish experiences with DRGs 359
Katarzyna Czach, Katarzyna Klonowska,
Maria Świderek and Katarzyna Wiktorzak
twenty one Portugal: Results of 25 years of
experience with DRGs 381
Céu Mateus
twenty two Spain: A case study on diversity of DRG use –
The Catalan experience 401
Francesc Cots, Xavier Salvador, Pietro Chiarello,
Montse Bustins and Xavier Castells
twenty three The Netherlands: The Diagnose Behandeling
Combinaties 425
Siok Swan Tan, Martin van Ineveld, Ken Redekop
and Leona Hakkaart-van Roijen
Index 447
List of tables, fi gures and boxes
Tables
2.1 Years of introduction and purposes of DRG systems over time 10
2.2 Hospital payment systems and their theoretical advantages and
disadvantages 13
2.3 Key fi gures of the European acute care landscape in 1995 and 2008 17
2.4 Share of ownership types across countries (% of acute care
beds), 2008 19
3.1 DRG weights and monetary conversion example 29
3.2 Key questions to be answered by this book 33
4.1 Basic characteristics of DRG-like PCSs in Europe (based on 2008) 41
4.2 Coding of diagnoses and procedures 43
4.3 Classifi cation variables and severity levels in European
DRG-like PCSs 50
4.4 Trends in coverage of services in DRG-like PCSs in Europe 53
5.1 Different approaches to collecting cost data in Europe 65
5.2 Different characteristics of the cost-accounting methods in eight
European countries 68
6.1 DRG weights, monetary conversion and their applicability in
12 European countries 78
6.2 DRG-based hospital payment for acute care hospitals 81
6.3 Incentives of DRG-based hospital payment systems and their
effects on quality and effi ciency 83
xii List of tables, fi gures and boxes
6.4 Defi nition of outliers and associated deductions/surcharges 87
7.1 Main differences across hospital payment systems 100
7.2 Incentives offered by three hospital payment models 101
7.3 Summary of recent studies examining the impact of
DRG-based hospital payment on hospital effi ciency 103
7.4 Summary of recent studies examining changes in indicators
of hospital effi ciency following the introduction of DRG-based
hospital payment 107
9.1 Possible effects of technological innovation on hospital costs 133
9.2 Incentives of DRG-based hospital payment systems and
effects related to technological innovation 135
9.3 Instruments to encourage the use of technological innovation
and types of associated payments 138
9.4 The application of short-term reimbursement instruments in 11
European countries (plus Catalonia) 140
9.5 Frequency of updates and time-lag to data used for updates
across 12 European countries 141
10.1 Options for integrating quality into DRG-based hospital payment
systems and examples from selected countries in Europe and
the United States 165
11.1 Hospital infrastructure in Austria 178
11.2 Payments to the SHFs in 2005 178
11.3 LKF-PCS versions 181
11.4 Content of the MBDS 183
12.1 Average daily number of available and occupied beds, 2008/2009 198
12.2 Evolution of the English DRG system 203
12.3 HRG prices, 2009/2010 212
12.4 Impacts of PbR: Overview of evidence 216
13.1 Distribution of acute care beds and activity between public
and private hospitals, 2007 223
13.2 Distribution of hospitals by size and ownership status 224
13.3 GHM versions used from 1986 to 2009 227
13.4 Number of hospitals and stays included in the National
Cost Study 228
13.5 Cost categories included in GHM prices for public and
private hospitals, 2010 235
14.1 Key hospital fi gures by size and ownership, 2007 245
14.2 Main facts relating to the G-DRG system 252
14.3 G-DRG modular costing approach 259
14.4 Strengths and weaknesses of the G-DRG system 267
15.1 Health care charges for public and private patients 275
15.2 Distribution of hospitals by number of beds, ownership
and type, 2008 276
15.3 Summary of hospital activity and beds, 2000 and 2008 276
15.4 Overview of inpatient and day-case classifi cation systems
used in Ireland 277
List of tables, fi gures and boxes xiii
15.5 Key facts on DRG systems used in Ireland since 1999 278
15.6 Inpatient and day-case blend rates, 2000–2010 casemix models 286
17.1 Hospital indicators, 1998–2008 (selected years) 305
17.2 Overview of discharges, outpatient contacts and insured
individuals used specialized medical care services 305
17.3 Summary description of the DRG system 307
17.4 Proportion of different payment mechanisms for inpatient
care in acute care hospitals, 2006–2009 315
18.1 Acute hospital key fi gures by type and ownership, 2009 323
18.2 Main facts relating to the different DRG versions 327
18.3 Distribution of cost by categories used in cost accounting in
the HUS, 2009 331
19.1 Hospital beds in acute care, psychiatric and long-term care,
and beds per 1000 people, 2009 340
19.2 Number of NordDRG codes in different settings in 2009 341
19.3 Various NordDRG versions in Sweden, 1995–2008 (selected years) 343
20.1 Number of hospitals, hospital beds and patients in Poland,
2000, 2005 and 2007 361
20.2 Main facts relating to the fi rst national DRG version in
Poland: the JGP system 364
20.3 Sections of the JGP system 365
20.4 Differences between British HRGs and Polish JGPs 371
20.5 Example of JGP scores and supplementary points for selected
JGP groups 374
21.1 Number of NHS (public) and private hospitals, 2003–2007 382
21.2 Versions of DRG groupers used in Portugal 385
21.3 Sources of data for the calculation of Portuguese prices by DRG 387
21.4 AP-DRG hierarchy in the pre-MDC process 388
21.5 Homogenous sections and cost centres 391
21.6 Percentage of funding based on DRGs 395
21.7 Cases treated, ALOS and CMI in the period 1997–2006 397
22.1 Acute hospitals in Catalonia 403
22.2 Main facts about the use of DRGs in Spain 406
22.3 Main facts about the use of DRGs in Catalonia 407
22.4 Main Spanish cost-accounting systems 412
22.5 Evaluation of unit cost using internal DRG weights 414
23.1 Key fi gures for university and general hospitals in 2009 427
23.2 Traditional DRGs versus DBCs 430
23.3 Hospital services resource-use categories 431
23.4 The main facts relating to the DBC versions, at its introduction
(2005) and the current version 432
23.5 Fictional average resource-use profi le for the DBC ‘surgery/
regular care/ arthrosis knee/ surgery with clinical episode’ 435
23.6 The patient classifi cation system logic: Surgery example 436
23.7 Fictional cost calculation of the hospital cost component 440
23.8 Negotiated tariffs in 2007 compared to those of 2004 443
xiv List of tables, fi gures and boxes
Figures
2.1 From DRG introduction to DRG-based budget allocation
and payment 11
3.1 Framework for navigating through the book 25
4.1 Historical development 39
4.2 Comparison of MDCs across eight DRG systems and similar PCSs 42
4.3 Classifi cation algorithm of different HCFA-DRG derived
PCSs, including system-specifi c modifi cations 45
4.4 Classifi cation algorithm in self-developed DRG-like PCSs
(HRG, JGP and LKF) 47
4.5 Trends in the number of groups in DRG-like PCSs in Europe 54
5.1 Cost accounting in the DRG era 60
5.2 Direct method for overhead allocation 62
5.3 Step-down method for overhead allocation 62
5.4 Reciprocal method for overhead allocation 63
5.5 Methodology matrix: Level of accuracy of the identifi cation and
valuation of hospital services 64
6.1 DRG-based hospital payment within the framework of this book 76
6.2 Selected incentives of DRG-based hospital payment for
a hypothetical standard patient 84
6.3 Trimming and reimbursement for outliers 88
7.1 Hospital revenues under global budgets 96
7.2 Hospital revenue under ‘pure’ DRG-based hospital payment 98
7.3 Hospital revenue under ‘mixed’ DRG-based hospital payment 99
7.4 Hospital revenue under ‘mixed’ DRG-based hospital payment
with marginal pricing 100
9.1 Effects of innovation on cost and quality 133
9.2 Short-term payment instruments and long-term
updating mechanisms 137
10.1 Extension of DRG systems from acute inpatient care to
other sectors 158
10.2 DRG-based hospital payment within the mix of total
hospital revenues 166
11.1 MBDS grouping process 184
11.2 Calculation of total LDF score per patient 189
12.1 Overview of previous and current DRG systems 202
12.2 HRG4 – Classifi cation fl ow chart for inpatients 206
12.3 Unbundled activity: Components for unbundled HRGs 207
12.4 English cost-accounting system: Initial stages 209
12.5 English cost-accounting system: Latter stages 210
13.1 Overview of hospital activity in France, 2006 223
13.2 Information used for classifying patients into the GHM 230
13.3 GHM classifi cation with level of severity 231
13.4 Range of services included in the GHM-based hospital
payment system 235
List of tables, fi gures and boxes xv
13.5 Breakdown of total hospital expenditure, 2008 236
14.1 Typical episode of care across sectoral borders 247
14.2 Phases involved in introducing DRGs in Germany 249
14.3 Phase of convergence 250
14.4 Types of data used for reimbursement and further development
of the G-DRG system 253
14.5 G-DRG grouping algorithm 254
14.6 Reimbursement components of inpatient care in Germany 260
14.7 Deductions and surcharges related to the length of stay 261
14.8 Components of G-DRG reimbursement and G-DRG
implementation 262
14.9 Prototypical regulatory pathways for introducing new
technologies into the regular system of G-DRG reimbursement 265
15.1 Assignment to DRGs using the AR-DRG classifi cation scheme 281
15.2 Distribution of CMUs for inpatient discharges 285
15.3 Calculation of the hospital inpatient casemix-adjusted budget 285
15.4 Casemix-adjusted inpatient average cost and blend rate for
Group I hospitals, 2001–2010 288
16.1 The development of the NordDRGs country versions 295
17.1 Overview of areas served by the Estonian hospital network 304
17.2 Main logic of the NordDRG system 309
17.3 Trimming method used in the calculation of the DRG price 315
17.4 Procedure for amending the benefi ts package 317
18.1 Funding of hospital care in Finland 324
18.2 Grouping process in the NordDRG system 329
18.3 Defi nition process of the average cost of a DRG group 333
19.1 Grouping algorithm in the Swedish version of NordDRG
system 346–347
19.2 The Swedish case costing model 349
19.3 An example of a patient in the Swedish National Case
Costing Database: femur fracture 350
20.1 Typical episode of care across sectoral borders 362
20.2 Timeline: introduction of the national JGP system in Poland
and prior sub-national uses of DRGs 366
20.3 JGP grouping algorithm 370
20.4 Calculation of hospital payment in the era of JGPs 375
21.1 AP-DRG (Version 21.0) classifi cation algorithm 389
21.2 Structure/rules for cost allocation 392
22.1 CMS-DRG version 24 grouping algorithm 410
22.2 The hospital payment system in Catalonia 416
22.3 Relative resource intensity values for Catalan hospitals, 2009 418
23.1 Role of hospitals and ZBCs in the delivery of hospital services 427
23.2 Fictional overview of types of DBCs 429
23.3 Data-collection process for the national database 434
23.4 The grouping algorithm for the new generation of DBCs 444
xvi List of tables, fi gures and boxes
Boxes
9.1 Separate payment to German hospitals under NUB regulations 138
9.2 Coverage with evidence development in the Netherlands 144
Abbreviations
A&E Accident & Emergency (department, services)
AC Autonomous Community
ACCC Australian Casemix Clinical Committee
ACHI Australian Classifi cation of Health Interventions
ACRA Advisory Committee on Resource Allocation
ACSS Central Administration of the Health System (Portugal)
ADRG Adjacent DRG
AFS Annual fi nancial statement(s)
AFSSAPS Agency for Safety of Medical Products (France)
(A)LOS (Average) Length(s) of stay
AMI Acute myocardial infarction
AN-DRG Australian National DRG
AOTM Health Technology Assessment Agency (Poland)
AP-DRG All-Patient DRG
APG Ambulatory Patient Group(s)
AR-DRG Australian Refi ned DRG
ARH Regional Hospital Agencies (France)
ARS Regional Health Agencies (France)
ATIH Technical Agency for Hospital Information (France)
ATU Temporary access for treatment (France)
AWBZ Exceptional Medical Expenses Act (Netherlands)
BQS Federal Offi ce for Quality Assurance (Germany)
CART Classifi cation and Regression Trees
CAT Computerized axial tomography
xviii Abbreviations
CC Complication and co-morbidity
CCAM French classifi cation of procedures (France)
CCSAE Clinical Costing Standards Association of England
CED Coverage with Evidence Development
CHS Catalan Health Service
CM Major category (France)
CMA Severe/acute/complicated episode(s) (France)
CMBD Joint standardized minimum basic dataset (Spain)
CMD Major diagnostic categories (France)
CMI Casemix index
CMS Centers for Medicare and Medicaid Services
CMU Casemix unit (Ireland)
COPD Chronic obstructive pulmonary disease
CQUIN Commissioning for Quality and Innovation (framework)
CT Computerized tomography
CVZ Healthcare Insurance Board (Netherlands)
DBC Diagnosis–treatment combinations (Netherlands)
DBC-DIS DBC information system (Netherlands)
DBC-O DBC onderhoud (Netherlands DBC administrative organisation)
DEA Data envelopment analysis
DG Day grouper
DIM Medical information units (France)
DIMDI German Institute of Medical Documentation and Information
DoHC Department of Health and Children (Ireland)
DPG Day-patient grouper
DRG Diagnosis-related group
ED Emergency department
EHESP Ecole des hautes études en santé publique (School of Public Health)
EHIF Estonian Health Insurance Fund
ENC National Cost Study (France)
ENCC Hospital cost database (France)
ESRI Economic and Social Research Institute (Ireland)
EU European Union
EU15 European Union Member States prior to May 2004
FPG Case Fees Act (Germany)
G-BA Federal Joint Committee (Germany)
GHCA General Health Care Act (Spain)
GHJ Homogeneous groups of days (French patient classifi cation tool)
GHM Homogeneous groups of patients (French patient classifi cation
system)
GP General practitioner
HAS High Health Authority (France)
HCAT HIPE Coding Audit Toolkit (Ireland)
HCFA Health Care Financing Administration (United States)
HDG Main diagnosis group (Austria)
HES Hospital Episode Statistics
HIPE Hospital In-Patient Enquiry (Ireland)
HNDP Hospital Network Development Plan
HPID Health Policy and Information Division (Ireland)
Abbreviations xix
HRG Healthcare Resource Group(s) (Ireland)
HRID Health Research and Information Division (Ireland)
HSE Health Service Executive
HTA Health Technology Assessment
ICD International Classifi cation of Diseases (WHO)
ICD-9-CM International Classifi cation of Diseases, Ninth Revision, Clinical
Modifi cation
ICD-10 International Classifi cation of Diseases, Tenth Revision
ICD-10-AM International Classifi cation of Diseases, Tenth Revision, Australia
Modifi cation
ICD-10-GM International Classifi cation of Diseases, Tenth Revision, German
Modifi cation
ICHI International Classifi cation of Health Interventions
ICPM International Classifi cation of Procedures in Medicine
ICS Catalan Health Care Institute (Spain)
ICU Intensive care unit
IGZ Healthcare Inspectorate (Netherlands)
InEK Institute for the Hospital Remuneration System (Germany)
IRDES Institute for Research and Information in Health Economics
(France)
IR-DRGs International Refi ned DRGs
IRF-PPS American Inpatient Rehabilitation Facility Prospective Payment
System
IT Information technology(ies)
JGP Homogeneous groups of patients (Polish patient classifi cation
system)
KAKuG Federal Hospitals Act (Austria)
KHEntgG Hospital Remuneration Act (Germany)
KMÅ Swedish national classifi cation system for non-surgical procedures
KRUS Agricultural Social Insurance Fund (Poland)
KVÅ Swedish national classifi cation system for surgery and non-
surgical procedures
LDF Procedure- and diagnosis-oriented case groups (Austria)
LKF Performance-oriented hospital fi nancing framework (Austria)
LTR Swiss performance-oriented payment system for rehabilitation
MAS Major ambulatory surgery
MBDS Minimum basic dataset
MDC Major diagnostic category
MEL Single medical procedure (Austria)
MERRI Missions d’enseignement, de recherche, de reference et d’innovation
(teaching, research, recourse and innovation, France)
MFF Market forces factor
MIGAC Missions of general interest and assistance with contracting,
including payments for education, research and public health
programmes
MRI Magnetic resonance imaging
MS-DRG Medicare Severity DRG
NCECI NOMESCO Classifi cation of External Causes of Injury
NCSP NOMESCO Classifi cation of Surgical Procedures
xx Abbreviations
NFZ National Health Fund (Poland)
NHS National Health Service
NOMESCO Nordic Medico-Statistical Committee
NordDRG Nordic patient classifi cation system
NPR National Patient Register
NUB New Diagnostic and Treatment Methods Regulation (Germany)
NZA Dutch Healthcare Authority
OECD Organisation for Economic Co-operation and Development
OPCS Offi ce of Population Censuses and Surveys (British
Classifi cation of Surgical Operations and Procedures)
OPS Procedure classifi cation codes (Germany)
OR Operating room
P4P Pay for performance
PbR Payment by Results
PCCL Patient Clinical Complexity Level
PCS Patient classifi cation system
PCSI Patient Classifi cation Systems International
PCT Primary Care Trust
PERFECT PERFormance, Effectiveness and Cost of Treatment episodes
PLICS Patient-Level Information and Costing Systems
PMSI Hospital activity database (France)
PRIKRAF Private Hospitals Financing Fund (Austria)
QMS Quality management system(s) (Germany)
RBG Rehabilitation Treatment Groups (Germany)
R&D Research and development
RRI Relative resource intensity
RSS Patient discharge summary (France)
RUM Departmental discharge summary (France)
RV Relative value(s) (Ireland)
SALAR Swedish Association of Local Authorities and Regions
SAM State Agency of Medicines (Estonia)
SD Skewed distribution
SGB V German Social Law, Fifth book
SHF State Health Fund (Austria)
SHI Statutory health insurance
STAKES National Research and Development Centre for Health and
Welfare
TB Tuberculosis
TUB Berlin University of Technology
UBA Basic Care Unit (Spain)
UMBDS Uniform minimum basic datasets (Portugal)
VAT Value-added tax
VHI Voluntary health insurance
VWS Ministry of Health, Welfare and Sport (Netherlands)
XHUP Public Hospital Network (Spain)
ZBC Independent Treatment Centre (Netherlands)
ZUS Social Insurance Institution (Poland)
ZVW Health Insurance Act (Netherlands)
Foreword 1
This book has global application; it will fi nd an eager audience among
policy leaders, technicians, hospitals, and physicians. In Organisation for
Economic Co-operation and Development (OECD) countries, in which DRGs
have been implemented since the 1980s, there is still little or no comparison
three decades later across countries, in terms of key building blocks, and
whether or how variations in design make a difference. How many categories
are enough, and when are there too many? When do DRGs begin to look
like a fee-for-service model? Which cost-accounting system works best?
Almost all DRG experts are conversant in one or two, or perhaps a small
number of systems, without any in-depth knowledge of the larger number of
countries in the European Union (EU) which have implemented some form of
DRGs.
As with other mechanisms and policies in the health sector, the book sheds
light and presents evidence on the importance of history and context. As far as
DRGs are concerned, the book suggests, there is no “one size fi ts all” situation.
At least not yet. While a uniform approach across Europe may emerge at some
point, it is clear that the experience of countries to date is defi ned by EU Member
States taking different approaches in terms of clinical categories, patient
classifi cation systems, costing and allocation, quality, and their readiness to
respond to the somewhat euphemistically termed “unintended consequences”
that seem to emerge in every implementation process. The diagnostic across
countries is both interesting and useful, and will be enlightening to students
in any country looking for ways to improve the casemix system, either under
design or already fully implemented. The variation in the short term takes the
xxii Foreword 1
form of an opportunity to provide a menu of options for solving technical
issues within each of the building blocks.
Still, it is remarkable that, from a broader vantage point, there is a path of
convergence in payment models for hospitals across Europe, with some mix of
DRGs and global budgets. Within the bigger picture, guidance is also given, not
only on what to do about individual building blocks, but also in terms of the
need to constantly “mind” or monitor and update the system in place. A former
United States Medicare administrator in the late 1980s, Dr. William Roper, once
argued that a system of DRGs would collapse under its own complicated and
technocratic weight, to be replaced by a simpler and more powerful capitation
model for the full benefi ts package. That day has not yet arrived, although the
book competently offers a glimpse of the future, which shows a system that
includes outpatient stays and the emergence of payment for entire episodes of
care, as is the case in the Netherlands.
Yet, this book will be appreciated beyond the EU and other OECD countries.
As a peripatetic World Bank health economist working in middle- and low-
income countries since the 1990s, upon arriving in a country and visiting the
leadership, a clear pattern of priorities emerges from the fi rst meeting with a
Minister of Health. The discussion typically starts with a series of questions
about how to mobilize more funds for services in the health sector. This is
often an ambiguous and meandering discussion, which highlights the need to
assess fi scal space, and raises some questions (from me back to the Minister)
regarding sectoral effi ciency and performance. Resource allocation, not new
money, quickly becomes front and centre. Like the Europeans and North
Americans of the 1970s and 1980s, the Minister agrees that the system needs to
restructure incentives to improve performance, while simultaneously facing a
landscape of changing demographics, changing disease profi les, and increasing
citizen dissatisfaction regarding responsiveness. Almost without exception,
the Minister then pronounces that the sector needs DRGs for hospitals, and
in quick succession wants to know in how many weeks might “we” (together)
implement the system in the country. Such a scenario has often played out
in the countries of the former USSR in the 1990s, in the Middle East in the
fi rst decade of this century, and in South and East Asia in the last few years.
My colleagues report that examples of this type of discussion are increasing in
number in Latin America and (most recently) Africa; for example, in Ghana,
Kenya, and South Africa.
Starting with the hospital sector in non-OECD countries makes sense. That
is where the money is. The share of expenditure for inpatient acute care is
typically more than 50 per cent of all spending. In China, it is 58 per cent, in
Brazil over 60 per cent and in the countries of the former USSR it was often above
70 per cent. Most countries face signifi cant challenges with both technical and
allocative effi ciency. Some effort to move from line-item budgets and/or fee-for-
service payment holds the promise of addressing multiple objectives related to
improved sectoral performance.
At the same time, the move to some form of DRGs is not risk free. Non-OECD
countries have often bought a software grouper from Australia, the United States
or the Nordic countries. More recently, they have begun to download from
the United States Medicare web site an open-source DRG grouper, with 350
Foreword 1 xxiii
categories and 3 levels of severity. “We can start right away” is often the remark
heard from the Minister’s staff. However, the European country experience is
that this model takes time to implement well – typically 5–10 years, and it took
even longer in the United States. This book is an insightful and helpful guide
on the multiplicity of paths that need to be followed – at times in parallel, at
times in concert – while at the same time providing options for fi nding the
fastest and most direct path to implementation.
A few years ago, the World Bank published a manual to help countries design,
build and implement new payment systems. The chapter on casemix was
certainly the centerpiece of the book. The book was written because countries
wanted to know not only “what” to do, but also “how” to do it. That book drew
on a very small number of countries in Central Asia, but most OECD and non-
OECD countries aspire to have a health sector similar to those found in Europe
today. Countries such as Germany, the United Kingdom, the Netherlands,
Denmark and Sweden have in place models that are often held up as examples,
if not actually emulated. Estonia has become the prime example of what can be
done well, from a rather dismal start point, and within a short space of time.
This list and mix of advanced and yet relatively similar European countries
identifi ed in the book become an optimal platform from which to really assess
the impact and potential of DRGs in terms of transparency, effi ciency, quality,
and so on. The book shows that, while there are predictable patterns of impact,
such as reduced length of stay, changes in numbers of beds, admissions and
occupancy, there are also signifi cant variations across the EU. And while most
of the world sees Europe as relatively homogeneous, the book also shows that
organization, fi nancing and delivery models continue to vary from country to
country. Finally, the country case studies are quite rich in detailing the political,
economic and technocratic approaches used in these individual countries, and
(again) provide a strong message to learn from others, but perhaps also to
develop unique and innovative solutions that refl ect history and the special
issues in any single country. The key message of good design is mingled
well with certain “preconditions” of success relating to political support, the
necessary legal framework, autonomy in the delivery system, good information
systems, and proactive quality assurance systems.
Enjoy, learn, compare, and be careful at the design and implementation
stages. The World Bank is a founding partner in the European Observatory on
Health Systems and Policies, and Bank experts will greatly appreciate this work.
With some Observatory books and publications, the experts contribute, but in
every case we also learn. We learn along with our many client countries, most
of which aspire to a system like those found in Europe today.
Jack Langenbrunner
The World Bank
Foreword 2
As a starting point, I think it is a good idea to use a private example to present a
book dealing with issues common to, and important for, all European countries.
Since the 1990s, Bulgaria has moved from the socialist model of centrally
planned health care to a single-payer health insurance system. The payment
of hospitals, formerly based on annual line-item budgets, gradually began
to be based on reported activities, known as “clinical pathways”. Each of
these pathways is defi ned for a set of similar diagnoses and has a fi xed price.
Prices were negotiated between the insurer and professional physicians’
organizations. This was just the opposite of what had taken place for 40
years – after centrally planned budgets, prices and wages, the country set out
optimistically, with the hope of a free market in hospital care! A few years
later, however, it became clear that the agreed prices of clinical pathways were
infl uenced by medical lobbyists and had no direct connection to the costs
actually incurred in hospitals – neither in respect of the ratios between the
different diagnoses and conditions, nor in terms of the varying degrees of
severity within a diagnosis. This was due to the fact that the clinical pathways
were based on the main diagnosis and procedure, but neglected the severity of
the patient’s condition and concomitant diseases. Thus, the more pathways a
hospital reported, the more money it received, and the milder the cases that
were treated, the more “cost-effective” (that is, profi table) the hospital was.
As a result, within ten years hospitalizations in the country increased by 68
per cent and the statistics reported a “growth” in diagnoses, mainly for the
well-paid clinical pathways. Part of this increase was also due to newly opened
private hospitals specializing precisely in these well-paid areas. Paradoxically (or
xxvi Foreword 2
actually, not surprisingly), despite the increase in fi nancial resources, citizens’
dissatisfaction with the health care system also increased.
Given the rapidly increasing hospitalizations and associated costs, global
budgets at hospital level were introduced, while the accounting continued to
be carried out through clinical pathways. Immediately, questions emerged: how
can we determine a fair global budget? How can we encourage those performing
well and limit those that are ineffi cient? How can we ensure transparency? How
can we ensure access and quality, without stimulating excessive consumption?
The system of clinical pathways was not able to provide adequate answers
to these questions, so Bulgaria began to look for alternatives, and intends to
introduce a DRG-based payment system, following the example of many other
countries in Europe. However, to reveal and compare the strengths, weaknesses,
opportunities and threats of European DRG-systems, as well as their design –
which is clearly different across countries due to their intended use – is a big
challenge for countries such as Bulgaria that want to introduce DRG-based
payments which are based on reliable data refl ecting patient needs and actual
costs, and which incentivize the provision of appropriate, high-quality and
effi cient care.
Therefore, this book – with Part One focusing on the main issues relating
to DRGs, as well as Part Two presenting structured DRG system comparisons
across twelve European countries – imparts extremely interesting information
for countries which are about to introduce DRGs to fi nance hospitals. It will
certainly be useful, not only for me, but for all others engaged with this issue.
It is essential reading for people who ask questions, share problems, offer
solutions and disseminate best practices.
Stefan Konstantinov
Minister of Health, Bulgaria
Acknowledgements
The editors would like to thank all EuroDRG project partners and related authors
who contributed their knowledge and broad experiences to this book; their
great patience with the editors ensured that countless reviews and revisions
did not threaten its realization. We would like to give special thanks to Dana
Forgione and Frank Hartmann, who reviewed some of the chapters in Part One,
encouraging the authors to improve these chapters. Furthermore, we would like
to thank Nicole Russell for the extensive editing work and Jonathan North for
coordinating the production process.
As the EuroDRG workshops were fundamental for collaboration among
project partners in order to develop this book, the editors would like to thank
Alexandra Starke, Carola Haring, Esther Martinez Amor and Reelika Ermel,
who organized the EuroDRG workshops in Potsdam (2009), Berlin (2009), Paris
(2010), Barcelona (2010), Tallinn (2011) and Berlin (again, in 2011). We would
also like to thank Pascal Garel, Bernhard Gibis, Luca Lorenzoni, Predrag Djukic
and Predrag Stojicic, who participated in EuroDRG workshops and provided
insight into related fi elds in different countries, which clearly contributed to
the development of the project as a whole.
Earlier versions of various chapters were presented and discussed at
numerous national and international conferences and workshops, such
as the EHMA conference ‘Hospital Financing: Diagnosis Related Groups –
Leading the Debate’ in Brussels (2009), the 4th Nordic Casemix Conference
in Helsinki (2010), the EHMA Annual Conference in Lahti (2010), the 8th
European Conference on Health Economics (ECHE) in Helsinki (2010), the
‘Polish DRG System Directions of Future Developments’ conference in Warsaw
xxviii Acknowledgements
(2010), the ‘Funding Models to Support Quality and Sustainability – A Pan-
Canadian Dialogue’ conference in Edmonton/Canada (2010), the Patient
Classifi cation Systems International (PCSI) Winter School in Dublin (2011), the
3rd German Association of Health Economics Conference in Bayreuth (2011),
the World Bank Workshop on ‘Health Financing Reforms to Improve Effi ciency
and Quality of Care’ in Moscow (2011), the 12th European Federation of
National Associations of Orthopaedics and Traumatology (EFORT) congress in
Copenhagen (2011), the European Observatory on Health Systems and Policies’
policy dialogue ‘Changing the Payment System for Hospital Care in Bulgaria
to Improve Equity and Effi ciency’ in Sofi a (2011) and the 8th World Congress
on Health Economics in Toronto (2011). The editors would like to thank the
auditorium of all these presentations for their inspiring thoughts, which have
infl uenced the fi nal versions presented in this volume.
European projects with multilateral relationships are subject to considerable
fi nancial administration efforts. The editors would therefore like to thank
Anette Schade, Silke Hönert and Katrin Ludwig from the European liaison offi ce
at the Berlin University of Technology for their thoughts to clarify budget-
related queries.
In addition, without funding from the European Commission (within the
Seventh Framework Programme, research area HEALTH-2007-3.2-8, reference:
223300) the Diagnosis-Related Groups in Europe: towards effi ciency and quality
research project in this highly relevant area would never have begun.
Finally, the authors would like to thank the numerous supporters of the
project, such as Philipp Seibert, Claudia Brendler, Claudia Reiche and Julia
Röttger.
Reinhard Busse
Alexander Geissler
Wilm Quentin
Miriam Wiley
List of contributors
Ain Aaviksoo, PRAXIS Center for Policy Studies, Estonia
Tõnis Allik, North Estonian Medical Centre, Estonia
Martine Bellanger, National School of Public Health, France
Reinhard Busse, Berlin University of Technology, Germany
Montse Bustins, Catalan Health Service, Spain
Xavier Castells, Hospital del Mar Research Institute, Spain
Pietro Chiarello, Hospital del Mar Research Institute, Spain
Francesc Cots, Hospital del Mar Research Institute, Spain
Katarzyna Czach, National Health Fund, Poland
Alexander Geissler, Berlin University of Technology, Germany
Leona Hakkaart-van Roijen, Erasmus University Rotterdam, the Netherlands
Unto Häkkinen, National Institute for Health and Welfare, Finland
Mona Heurgren, National Board of Health and Welfare, Sweden
Kristiina Kahur, Estonian Health Insurance Fund, Estonia
Kirsi Kautiainen, National Institute for Health and Welfare, Finland
Katarzyna Klonowska, National Health Fund, Poland
Conrad Kobel, Innsbruck Medical University, Austria
Heli Laarmann, PRAXIS Center for Policy Studies, Estonia
Jorma Lauharanta, Hospital District of Helsinki and Uusimaa, Finland
Miika Linna, National Institute for Health and Welfare, Finland
Anne Mason, University of York, United Kingdom
Céu Mateus, Universidade Nova de Lisboa, Portugal
Brian McCarthy, Economic and Social Research Institute Dublin, Ireland
Zeynep Or, Institute of Research and Information on Health Economics, France
xxx List of contributors
Jacqueline O’Reilly, Economic and Social Research Institute Dublin, Ireland
Gerli Paat, PRAXIS Center for Policy Studies, Estonia
Karl-Peter Pfeiffer, Innsbruck Medical University, Austria
Wilm Quentin, Berlin University of Technology, Germany
Ken Redekop, Erasmus University Rotterdam, the Netherlands
Xavier Salvador, Catalan Health Service, Spain
David Scheller-Kreinsen, Berlin University of Technology, Germany
Lisbeth Serdén, National Board of Health and Welfare, Sweden
Andrew Street, University of York, United Kingdom
Maria Świderek, National Health Fund, Poland
Siok Swan Tan, Erasmus University Rotterdam, the Netherlands
Josselin Thuilliez, National School of Public Health, France
Martin van Ineveld, Erasmus University Rotterdam, the Netherlands
Martti Virtanen, Nordic Casemix Centre, Finland
Padraic Ward, University of York, United Kingdom
Katarzyna Wiktorzak, National Health Fund, Poland
Miriam Wiley, Economic and Social Research Institute Dublin, Ireland
Part One
chapter o n e
From the origins of DRGs
to their implementation
in Europe
Miriam Wiley
1.1 The starting point
Really the whole hospital problem rests on one question: What happens to
the cases? [. . .] We must formulate some method of hospital report showing
as nearly as possible what are the results of the treatment obtained at
different institutions. This report must be made out and published by each
hospital in a uniform manner, so that comparison will be possible. With
such a report as a starting-point, those interested can begin to ask questions
as to management and effi ciency.
(Dr Eugene Codman, Address to the
Philadelphia County Medical Society, 1913)1
The ‘hospital problem’, as presented by Dr Codman – a surgeon at Massachusetts
General Hospital – at the beginning of the 20th century continues to present a
challenge today, almost 100 years later (Fetter, 1991). The work which was
initiated by Codman was revisited and further developed by Professor
Robert Fetter and his colleagues at Yale University in the late 1960s, when they
were invited to assist with the development of a programme of utilization
review and quality assurance for their local university hospital. The questions
posed of Fetter and his team relating to their work on this issue began what he
later described as a 20-year process of ‘measuring hospital production as a
means of evaluating what takes place in the hospital’ (Fetter, 1991, p. 4). It is
interesting that the original initiative was prompted by the requirements of
registration for the Medicare Program, which had been established in 1965, and
it was the Medicare Program in 1983 that fi rst implemented the diagnosis-
related group (DRG) system that emerged from this lengthy development
process.
4 Diagnosis-Related Groups in Europe
Because of the requirements to process very large sets of hospital data,
developments in information technology (IT) were critical to the work that
took place throughout the 1970s relating to ‘fi nding a way to measure and cost
the output of hospitals’ (Fetter, 1993). The fi rst version of what became the
DRG system was developed in 1973 and comprised 54 major diagnostic
categories (MDCs) and 333 fi nal groups. The second version was developed for
the Federal Social Security Administration and comprised 83 MDCs and 383
DRGs (Fetter et al., 1980), while the third version in 1978 was developed for the
State of New Jersey, which was proceeding with putting in place a DRG-based
hospital payment system. The fi nal (original) version of the DRG system was
developed by the Health Systems Management Group at Yale University within
the framework of a contract with the Health Care Financing Administration
(HCFA) for the purpose of developing ‘an inpatient classifi cation system that
differentiated the amount of hospital resources required to provide care and
was clinically coherent in the sense that the groups were expected to evoke a set
of clinical responses which resulted in a similar pattern of resources’ (Rodrigues,
1993). The so-called ‘prospective payment system’, which was introduced for
the Medicare Program in 1983 mandated that payments for hospital services
were determined on the basis of the fi rst version of the HCFA-DRG system,
which at that time comprised 470 groups across 23 MDCs.
The enactment of Medicare’s prospective payment system was considered to
be ‘the single most infl uential post-war innovation in medical fi nancing’ by
Mayes (2007, p. 21), who notes that ‘Medicare’s new prospective payment sys-
tem with DRGs triggered a shift in the balance of political and economic power
between the providers of medical care (hospitals and physicians) and those who
paid for it – power that providers had successfully accumulated for more than
half a century’ (ibid, p. 21). The view put forward by Mayes that this change
went virtually unnoticed by the general public is particularly interesting because
what this book attempts to track is how this innovation worked its way around
the world to the point where, almost 30 years later, the DRG system is the single
most important patient classifi cation system (PCS) in use internationally.
1.2 Crossing the Atlantic and the Pacifi c
Living, as we do, in an era of almost ‘instant’ communication, it would be easy
to underestimate the signifi cance of the international ripple-effect associated
with the adoption of the DRG system by the United States Government in
1983. While we are now accustomed to being immediately informed about
signifi cant world events or important developments in our areas of interest, in
the early 1980s we had to order journals by post, go to libraries to access
literature and communicate with our international colleagues by fax or ‘phone!
Despite such challenges, however, the international impact of the move to a
prospective payment system by the United States Medicare Program was rapid,
with developments in Europe and Australia proceeding quickly by the standards
of the era (and even by current standards). In Europe, a meeting hosted by the
Ministry of Health in France in 1984 included Professor Robert Fetter, the leader
of the team which developed the DRG system, and involved fi ve countries
From the origins of DRGs to their implementation in Europe 5
(Belgium, France, Ireland, the Netherlands and Portugal). A further inter-
national meeting was held just two years later in Dublin, already involv-
ing 11 European countries. When 15 countries participated in a meeting in
Lisbon in 1987, they agreed to set up a network for those interested in working
on issues related to the classifi cation of patients, and Patient Classifi cation Sys-
tems International (PCSI)’s Patient Classifi cation Systems Network continues to
function today.2
In parallel with the European developments, a National Seminar on DRGs
was held in Australia in 1984. Following this seminar, the funding of a number
of research projects sowed the seeds which quickly fl ourished into a substantial
research area, producing the evidence base on which subsequent developments
in DRG systems and their applications in Australia were founded.
The momentum in international developments regarding the portability and
suitability of DRGs for use in health systems outside of the United States was
given some additional support from international meetings organized by the
Yale development team in London in 1986, Washington in 1987, and Sydney
in 1988. In addition to profi ling the activities in an increasing number of
countries, these conferences enabled the researchers and policy-makers to make
personal contacts which facilitated more rapid exchange of information and
sharing of experiences than would otherwise have been possible (in the era
before the World Wide Web). These meetings, together with those organized by
PCSI, helped to foster a spirit of cooperation amongst those in a position at the
fore in this fi eld, such that each new entrant could quickly benefi t from those
who had gone before.
The momentum for international collaboration on developments and appli-
cations for DRG-type systems also benefi tted from initiatives supported by a
number of international organizations. In 1985 the Council of Europe sup-
ported a review of the research being undertaken in Europe at that time on
DRGs, while the Organisation for Economic Co-operation and Development
(OECD) began to publish international comparisons of average lengths of stay
by DRG (Rodrigues, 1989). The European Union (EU) programme of the late
1980s which was concerned with supporting medical and health research also
supported a number of projects relating to costing and using DRGs, and sup-
porting the development of PCSs appropriate for European hospitals (Casas &
Wiley, 1993; Leidl et al., 1990). Over the same period, WHO supported a number
of planning meetings regarding the use of DRGs for hospital budgeting and
performance measurement (Wiley, 1990).
While the lead-in to the application of DRGs within the United States
prospective payment system was not particularly lengthy when viewed in terms
of the pace at which translating research into policy applications usually takes
place, it is interesting to note that a much more truncated period predated the
fi rst national applications of DRG-based payment systems in Europe and
Australia. In Europe, Portugal was the fi rst country to begin operating a DRG-
based hospital payment system for payments from occupational health
insurance schemes in 1988 (see Chapter 21), which accounted for about 30 per
cent of hospital activity at the time. Norway followed, with the introduction of
a DRG-based payment system in selected hospitals in the period 1991–1993
(Magnussen & Solstad, 1994), and Ireland began the introduction of a
6 Diagnosis-Related Groups in Europe
DRG-based budget allocation system for a limited number of acute care hospitals
in 1993 (see Chapter 15). The fi rst initiative in Australia dates back to 1988,
when the then Australian Federal Health Department incorporated DRGs into
the 1988–1993 Medicare Agreements between the Commonwealth and eight
states and territories, and began funding the development of an Australian
version of DRGs (Australian National (AN-)DRGs), introduced in 1992. Victoria
was the fi rst state to use DRGs (in 1993) to set budgets for its public hospitals
(McNair & Duckett, 2002).
1.3 Where are we now? Aims of the book
It is evident that the development of the DRG casemix classifi cation system –
together with advancing a range of applications – could be described as an
international phenomenon (Kimberly et al., 2008). It is rare in the world of
health systems development to identify an initiative which has progressed so
rapidly from the research phase to implementation and international
dissemination. This book aims to bring readers up to date on developments in
this fi eld in European countries in more recent times. While it is clear that most
countries have introduced DRG systems and DRG-based hospital payment
systems with the aims of increasing transparency, improving effi ciency and
assuring quality in hospitals, it remains relatively unknown whether countries
are really moving towards achieving these goals. This book therefore summarizes
experiences and developments in European DRG systems.
The focus of the book on Europe relates to the fact that the EuroDRG project3
that gave rise to this initiative has been funded by the Seventh Framework Pro-
gramme (FP 7) of the EU. The 12 countries (Austria, England, Estonia, Finland,
France, Germany, Ireland, the Netherlands, Poland, Portugal, Spain, Sweden)
which take part in the EuroDRG project and which are included in this book
were selected based on their geographical region (for example, Portugal versus
Finland, and Poland versus France), health system typology (such as National
Health Service (NHS) versus Statutory Health Insurance (SHI)) and their duration
of affi liation to the EU (for example, Estonia versus the Netherlands), in order to
ensure a comparison of countries with truly different characteristics. However, it
is recognized that there would be scope for a companion volume tracking devel-
opments in Australia, Asia, Africa, and Central and South America.
The book is addressed to health policy-makers and researchers from Europe
and beyond and is intended to contribute to the emergence of a ‘common
language’ that will facilitate communication between those researchers and
policy-makers, from different countries. Both the overview of the key issues
(Part One) and the experience from the 12 countries analysed herein (Part Two)
should be particularly useful for countries and regions that want to introduce,
extend, or optimize their DRG systems. However, in the context of the increasing
importance of cross-border movement of patients and payments, this book also
aims to draw attention to the potential for coordinating and eventually
harmonizing DRG systems and DRG-based hospital payment in Europe. Clearly,
the book demonstrates that progress has been made since the work undertaken
by Codman a century ago, and that countries are continuously striving to
From the origins of DRGs to their implementation in Europe 7
optimize their DRG systems in order to better understand what Robert Fetter
termed ‘the rather strange cost behaviour of hospitals’ (Fetter, 1993, p. v).
1.4 Notes
1 See Codman, 1913–1917.
2 More information on the network can be found at the PCSI web site (www.pcs
international.org, accessed 26 July 2011).
3 More information is available at the EuroDRG project web site (www.eurodrg.eu,
accessed 26 July 2011).
1.5 References
Casas, M., Wiley, M., eds. (1993). Diagnosis-Related Groups in Europe, Uses and Perspectives.
Berlin: Springer-Verlag.
Codman, E.A. (1913–1917). The product of a hospital (Philadelphia address) (Box 4,
Folder 77), in E.A. Codman. Ernest Amory Codman Papers, 1849–1981: Finding Aid.
Boston, MA: Boston Medical Library and Francis A. Countway Library of Medicine (B
MS c60).
Fetter, R.B., ed. (1991). DRGs: Their Design and Development. Ann Arbor, MI: Health
Administration Press.
Fetter, R.B. (1993). Foreword, in M. Casas, M. Wiley, eds. Diagnosis-Related Groups in
Europe, Uses and Perspectives. Berlin: Springer-Verlag.
Fetter, R.B., Shin, Y., Freeman, J.L., Averill, R.F., Thompson, J.D. (1980). Casemix defi nition
by diagnosis-related groups. Medical Care, 18(2):1–53.
Kimberly, J.R., de Pouvourville, G., D’Aunno, T., eds. (2008). The Globalization of Managerial
Innovation in Health Care. Cambridge: Cambridge University Press.
Leidl, R., Potthoff, P., Schwefel, D., eds. (1990). European Approaches to Patient Classifi cation
Systems. Berlin: Springer-Verlag.
Magnussen, J., Solstad, K. (1994). Case-based hospital fi nancing: the case of Norway.
Health Policy, 28:23–36.
Mayes, R. (2007). The origins, development and passage of Medicare’s revolutionary
prospective payment system. Journal of the History of Medicine and Allied Sciences,
62(1):21–55.
McNair, P., Duckett, S. (2002). Funding Victoria’s public hospitals: the casemix policy of
2000–2001. Australian Health Review, 25(1):72–99.
Rodrigues, J.M. (1989). L’Europe des DRG. Sozial- und Präventivmedizin, 34:152–5.
Rodrigues, J.M. (1993). DRGs: origin and dissemination throughout Europe, in M. Casas,
M. Wiley, eds. Diagnosis-Related Groups in Europe, Uses and Perspectives. Berlin:
Springer-Verlag.
Wiley, M. (1990). Patient classifi cation systems: overview of experiments and applications
in Europe, in R. Leidl, P. Potthoff, D. Schwefel, eds. European Approaches to Patient
Classifi cation Systems. Berlin: Springer-Verlag.
chapter t w o
Introduction to DRGs in
Europe: Common objectives
across different hospital
systems
Alexander Geissler, Wilm Quentin, David
Scheller-Kreinsen and Reinhard Busse
2.1 Introduction
Since 1983, when Medicare adopted diagnosis-related groups (DRGs) as the
basis for paying hospitals in the United States, DRG-based hospital payment
systems have become the basis for paying hospitals and measuring their activity
in most high-income countries, albeit to different extents (Paris et al., 2010).
However, the term DRG is widely used with different meanings across and
within countries. Some countries use DRGs mostly as a measure for assessing
hospital casemix (for example, Sweden and Finland), whereas in other countries
DRGs are used as a synonym for payment rates (such as in France and Germany).
This is partly due to different DRG implementation processes that took place in
different decades, and partly due to the fact that DRG systems were adopted
and designed primarily based on the needs of the health system concerned
(Busse et al., 2006; Schreyögg et al., 2006).
The second section (2.2) of this chapter summarizes the purposes of the
introduction of DRGs in European countries and the expectations associated
with their implementation, as well as illustrating the complexity of this process
by highlighting the extended periods of time that sometimes evolved from the
initial application of DRGs in hospitals to their use for hospital payment. In
many countries, this process was highly controversial because of the potential
unintended consequences of DRG-based hospital payment systems (see Chapter
6), and it is diffi cult to understand the international success of these systems
without being aware of the alternatives. Therefore, section 2.3 presents the
basic incentives of fee-for-service systems and global budgets that were
10 Diagnosis-Related Groups in Europe
traditionally used in most countries. Section 2.4 then turns to the large structural
differences in the hospital sector – both those that existed between countries at
the time when DRGs were introduced for hospital payment and those that
continue to persist today, despite the fact that DRGs are used for hospital
payment in all countries. This serves to illustrate the original aim of this book;
namely, to identify similarities and differences in the use of DRG systems across
Europe. The chapter closes with a brief overview of the structure of this book.
2.2 Expectations and purposes of DRG introduction
Independent of the type of hospital system in place (see section 2.4), DRG
systems were internationally introduced for similar reasons, which can be
grouped into two broad categories: fi rst, they should increase the transparency
of services which are effectively provided in hospitals (that is, through patient
classifi cation, measuring hospital output, etc); and second, DRG-based payment
systems should give incentives for the effi cient use of resources within hospitals
by paying hospitals on the basis of the number and type of cases treated. In
addition, the combination of increased transparency and effi cient use of
resources was assumed to contribute to improving – or at least assuring –
the level of quality of care.
Table 2.1 shows how the purpose of DRG introduction varied according to
when the country in question introduced the DRG-based system. Interestingly,
Table 2.1 Years of introduction and purposes of DRG systems over time
Country Year of DRG Original purpose(s) Principal purpose(s) in 2010
introduction
Austria 1997 Budgetary allocation Budgetary allocation,
planning
England 1992 Patient classifi cation Payment
Estonia 2003 Payment Payment
Finland 1995 Description of hospital Planning and management,
activity, benchmarking benchmarking, hospital
billing
France 1991 Description of hospital Payment
activity
Germany 2003 Payment Payment
Ireland 1992 Budgetary allocation Budgetary allocation
Netherlands 2005 Payment Payment
Poland 2008 Payment Payment
Portugal 1984 Hospital output Budgetary allocation
measurement
Spain (Catalonia) 1996 Payment Payment, benchmarking
Sweden 1995 Payment Benchmarking,
performance measurement
Source: Authors’ own compilation based on information presented in the country-specifi c
chapters of Part Two of this volume.
Note: Even if the stated original purpose was to pay hospitals on the basis of DRGs, most
countries began this process only after a conversion period (see Figure 2.1).
Introduction to DRGs in Europe 11
countries that were early adopters of DRGs primarily did so with the aim of
increasing transparency (such as Portugal and France). Countries that introduced
DRGs later (such as the Netherlands and Poland) generally did so with the
intention of paying hospitals on the basis of DRGs.
Figure 2.1 illustrates the DRG introduction process in different countries
since the early 1980s. Every country took a different route at a different time to
introduce a DRG-based system, often initially for the purpose of patient classi-
fi cation, and later also for payment purposes. Some countries used DRGs over
an extended period of time exclusively for the purpose of patient classifi cation
and increasing transparency (for example, up to ten years in England), in order
to become acquainted with the DRG grouping logic before they started paying
hospitals on the basis of DRGs. Others introduced DRGs after a short period of
conversion (for example, in Ireland DRGs were introduced in 1992 and fi rst
used for budgetary allocation in 1993).
The reason why the introduction of DRG systems was thought to improve
transparency is that such systems condense the extremely large number of
patients that all appear to be unique into a limited number of groups that have
a set of certain characteristics in common (Fetter et al., 1980). By categorizing
patients with similar resource utilization and clinical characteristics into groups,
DRGs describe hospital activity in standardized units and enable analyses, which
Figure 2.1 From DRG introduction to DRG-based budget allocation and payment
Source: Authors’ own compilation based on information presented in the country-specifi c
chapters of Part Two of this volume.
Note: Ireland started with HCFA-DRGs in 1992 and switched in 2003 to AR-DRGs.
12 Diagnosis-Related Groups in Europe
otherwise would not be possible. For example, hospital managers and policy-
makers can compare length of stay, costs, and quality of patients within the
same DRG across different hospitals or across different hospital departments. In
addition, DRGs offer a framework for an accurate assessment of the costs of
treating a given patient, taking account of observable and measurable patient
and service characteristics such as diagnoses and performed procedures. Conse-
quently, DRGs facilitate performance comparisons and benchmarking, as well as
contributing to increased transparency in an area of policy-making that previ-
ously was characterized by extreme agency problems. Especially in countries
that traditionally used global budgets as their mode of hospital payment, the
hospital management had very little information on what types of services were
delivered and at what costs clinicians delivered these within their wards or
departments.
The main purpose behind the introduction of DRGs in the countries that
introduced them in the late 1990s and 2000s – namely, to use DRGs as a basis
for hospital payment – was extremely ambitious. This is because the aim is
not only to pay providers fairly, but also to discourage the provision of
unnecessary services and to encourage the effi cient delivery of appropriate
care. In the context of the increasing health care costs in many European
countries, DRG-based hospital payment systems fi tted well with the paradigm
of designing public policy according to general economic principles, in order
to exert fi nancial pressure and to incentivize effi cient resource use (see Chapter
7) by mimicking product markets that produce at marginal costs (Shleifer,
1985).
In Europe, Portugal was the front-runner in operating a DRG-based hospital
payment system for payments from occupational health insurance schemes in
the late 1980s. More recently, in many other European countries (such as
England, France and Germany) DRG-based hospital payment systems have
evolved to become the main hospital payment system, with the objectives
generally comprising, inter alia, increasing effi ciency, activity and transparency;
reducing waiting times and length of stay; supporting patient choice; enhancing
quality of care; and encouraging competition between hospitals. In Sweden and
Finland, however, DRGs are still primarily used to aid transparency in the
planning and management of hospital services.
As illustrated in the top row of Figure 2.1, most countries are using country-
specifi c DRG systems. Only Ireland, Portugal and Spain are operating DRG sys-
tems that were imported from Australia (Australian Refi ned (AR-)DRGs) or the
United States (All Patient (AP-)DRGs, Centers for Medicare and Medicaid Ser-
vices (CMS-)DRGs). However, many other countries also imported DRG systems
from abroad and used these as the starting point for developing their own sys-
tems (see Chapter 4). The Nordic countries (Finland, Sweden and Estonia) are
special, in that they decided to collaborate and share the development effort in
order to create a common NordDRG system that is further adjustable to country-
specifi c conditions (see Chapter 16), which may serve as an example for a pan-
European model of coordinating DRG models or even developing a uniform
system.
It is important to emphasize that countries introduced DRG-based payment
systems irrespective of (1) which kind of hospital payment system was in place
Introduction to DRGs in Europe 13
before (see section 2.3) and (2) their very different structural circumstances (see
section 2.4).
2.3 Hospital payment systems and incentives
The move in most countries towards DRG-based hospital payment systems was
driven by the objective of incentivizing hospitals to improve their perform-
ance (Langenbrunner & Wiley, 2002). Prior to the introduction of DRG-based
hospital payment systems, countries used two basic mechanisms to pay for
hospital care: fee-for-service payments and global budgets. These systems
provide a specifi c set of incentives, which are different from the incentives of
DRG-based systems. Therefore, in order to understand the international success
of DRG-based systems, it is necessary to be aware of the incentives of theses
alternative systems, and of the objectives that hospital payment systems are
supposed to achieve. Hospital payment systems should motivate providers to
treat patients in need of care and to deliver an adequate number of necessary
services (level of activity), while taking into account the appropriateness of the
services and patient outcomes (i.e. quality). Finally, a hospital payment system
should balance activity and expenditure control incentives, thus con-
tributing to increasing effi ciency, while minimizing administrative effort and
maximizing transparency. This demonstrates two things: (1) the design of
‘good’ payment systems needs to take into account various dimensions; namely,
those of patients and of providers, of the provided services, of payers, and
possibly of society at large; and (2) because of this complexity, it simply cannot
be expected that any payment system is ‘optimal’ in all respects. Rather, all
payment systems have their strengths and weaknesses in relation to the various
objectives. Table 2.2 summarizes the advantages and disadvantages of the
above-mentioned payment systems by evaluating their characteristics in
relation to the requirements of modern hospital payment systems.
Table 2.2 Hospital payment systems and their theoretical advantages and disadvantages
Activity
Number Number Expenditure Technical Quality Administrative Transparency
System of cases of services/ control effi ciency simplicity
case
Fee-for-service/
Cost
reimbursement + + – 0 0 – 0
DRG-based
payment + – 0 + 0 – +
Global budget – – + 0 0 + –
Sources: Authors’ own compilation, based on Barnum et al. (1995) and WHO (2000).
Notes: +/–: increase/decrease; 0: neutral or unclear; for a defi nition of technical effi ciency, see Chapter 7 of
this volume.
Dimension
14 Diagnosis-Related Groups in Europe
2.3.1 Fee-for-service payments
In the United States and some European countries (such as Estonia), the
(‘retrospectively’ determined) fee-for-service system was the principal means of
allocating resources to hospitals prior to the use of – in comparison to fee-for-
service payment – ‘prospectively’ determined DRG-based hospital payment
systems. The sum of the fees in fee-for-service payment systems should ideally
refl ect the actual individual patient costs. This approach was often considered
as fair or favourable by providers as long as fees covered at least their costs –
preferably costs plus profi t, of course. Fee-for-service payment provides strong
incentives to be productive and to offer a large of number services per patient
and therefore ensures that those hospitals treating more complex patients are
adequately reimbursed. However, fee-for-service payment may lead to the
provision of unnecessary services or may even encourage oversupply of inap-
propriate services, which negatively affects patient outcomes and the effi cient
delivery of services. In addition, providers under a pure fee-for-service regime
(that is, without budget limitations) are incentivized to neglect expenditure
considerations, which also contributes to an ineffi cient service delivery. Paying
hospitals according to a fee-for-service scheme is administratively complex, as
such systems require detailed and up-to-date price lists, as well as registration
and billing of all service items provided. Furthermore, the only instrument for
cost control is the specifi cation of the price list, which details the unit payment
for each item (Street et al., 2007).
2.3.2 Global budgets
In Europe global budgets were a common approach used for allocating fi nancial
resources to hospitals before the introduction of DRG-based hospital payment
systems. In the context of global budgets a fi xed payment for a certain activity
level (typically determined in terms of number of cases or number of bed days)
was negotiated and agreed between payers and hospitals, usually for the
approaching year; namely, really ‘prospectively’. In some countries global bud-
gets were defi ned at or adjusted for specialty. Global budgets are administra-
tively simple and can effectively contribute to cost-containment because of
their expenditure cap characteristic. However, they run the risk of hospitals
not producing suffi cient services to meet patient or population needs, hence
disregarding patient needs and therefore health outcomes. Some European
countries were using target budgets blended with per diem payments as billing
units (for example, Germany). Consequently, hospitals were provided with
clear incentives to increase bed occupancy by prolonging the length of stay.
Fee-for-service systems and global budgets provide confl icting incentives for
‘activity’ and ‘expenditure control’ (see Table 2.2). Both are problematic in
terms of ensuring high-quality care due to the inherent incentive to over-
provide (fee-for-service) or to under-provide (global budgets) hospital services.
Policy-makers (fi rst in the United States and later in Europe) were therefore
attracted by the idea of paying hospitals through DRGs, which to a certain
Introduction to DRGs in Europe 15
extent provide incentives somewhere in between a fee-for-service system and
global budgets.
2.3.3 DRG-based payments
The term DRGs is used here to highlight the theoretical incentives of DRG-
based payments, which do not necessarily correspond to the actual incentives
of the systems operated in the countries included in this book. Theoretically,
DRG-based payments provide strong incentives to increase the number of cases
treated and to reduce the number of services per case. In contrast to fee-for-
service systems, DRGs incentivize hospitals to limit their activity to necessary
services and – in contrast to global budgets – DRGs incentivize hospitals to treat
more patients. In terms of expenditure control, the effect of DRG-based pay-
ments thus depends on which effect prevails: increasing the number of cases or
reducing the number of services per case. In principle, this will also depend on
the previous system in place; that is, moving from fee-for-service payment to
DRGs can result in cost-containment, while moving from global budgets to
DRGs does not.
If DRGs do not suffi ciently control for differences between patient groups or
for differences in provided services (within DRGs), payments for highly com-
plex cases are too low, while payments for less-complex cases are too high.
Consequently, hospitals could try to avoid the risk of treating more complex
patients. Furthermore, DRG-based payment systems are administratively com-
plex as they require detailed and standardized coding of diagnoses and proce-
dures, as well as information on the average resource consumption (costs) per
DRG.
However, as already outlined, each of the presented payment systems has
certain advantages and disadvantages (see Table 2.2). Therefore, policy-makers
across Europe have combined features of the different systems: current DRG-
based hospital payment relies heavily on service characteristics to defi ne DRGs.
Consequently, hospitals are paid partly on the basis of the services that they
provide, which introduces aspects of fee-for-service payment into DRG-based
hospital payment. Furthermore, the systems are operated within global budgets
and provide additional payments for specifi ed services, high-cost drugs and
patients with exceptionally long lengths of stay (see Figure 10.2). Interestingly,
these payment reforms have been implemented in very diverse hospital
environments, which are described in the following section (2.4).
2.4 The hospital landscape
For a long time, a hospital was seen simply as ‘an institution which provides
beds, meals, and constant nursing care for its patients while they undergo
medical therapy at the hands of professional physicians. In carrying out these
services, the hospital is striving to restore its patients to health’ (Miller, 1997).
Clearly, this defi nition describes very broadly the activities of a hospital and
must therefore be constantly refi ned and extended by taking into account the
16 Diagnosis-Related Groups in Europe
key previous, ongoing and future changes in hospital care. Since the early
1980s, many European countries have shifted inpatient treatments towards
outpatient settings in order to reduce and improve effi ciency in the use of
hospital resources. This development has led to enormous structural challenges
for hospitals (McKee & Healy, 2002). Technological improvements and
redesigned care pathways made it possible to extend the number of day cases
and outpatient surgery cases treated outside the hospital or in specialized
departments within the hospital. However, countries vary in terms of their level
of integration between the ambulatory and inpatient sectors.
Table 2.3 provides an overview of these differences based on selected hospital-
related indicators for the 12 countries included in Part Two of this book – for
1995 (that is, before DRGs were introduced for payment purposes) and 2008.
The numbers and change rates (trends) indicate that different treatment
patterns and organizational differences existed before DRGs were in use across
Europe, and continue to exist.
All countries (except the United Kingdom) reduced to a different extent the
amount of acute care hospitals and beds between the mid-1990s and 2008.
However, the number of acute care hospitals and beds per capita differs by a
factor of between 5 and 3 across the 12 countries for the year 2008, only slightly
down from the sixfold and threefold differences seen in 1995. In terms of the
trend in acute care hospital admissions, the picture is less clear: France and the
United Kingdom show reduction rates between 1995 and 2008 of 18.1 per cent
and 42.5 per cent, respectively, while the Nordic countries (Estonia, Finland
and Sweden) and Ireland only slightly reduced the number of acute care
admissions (from 2.6 per cent in Estonia up to 7.2 per cent in Ireland). In
contrast, in Austria and the Netherlands the number of admissions to acute care
hospitals increased by 22 per cent and 15 per cent, respectively.
The average length of stay (ALOS) in acute care hospitals decreased more
(Estonia: 45 per cent) or less (France: 2 per cent) in each country except Sweden.
However, as in 1995, in 2008 the ALOS still differed by up to a factor of 2
between countries (for example, Germany versus Finland). Unlike Estonia,
Germany and the Netherlands, four countries (Austria, France, Spain and the
United Kingdom) were able to increase the bed occupancy rates during 1995
and 2008. Nevertheless, in 2008 the bed occupancy rates varied by a factor of
1.6 between Ireland (89 per cent) and the Netherlands (56 per cent), a larger
variation than in 1995.
In addition, comparing the inpatient expenditure as a share of the total
health expenditure (which decreased in each country) – as a proxy for the rela-
tive importance of the hospital sector – shows that countries rely on different
strategies to treat the same patients in different settings (namely, inpatient ver-
sus outpatient). This also becomes evident when comparing the number of
hospital-based physicians across countries in 2008. Compared to Finland, only
half as many physicians work in Dutch hospitals. Despite the fact that interna-
tional comparisons are always accompanied by inconsistencies in the defi ni-
tion of variables (for example, acute care hospital beds were defi ned slightly
differently across European countries), it becomes apparent that the range of
services delivered in acute care hospitals is somehow different from one coun-
try to another.
T
a
b
le
2
.3
K
ey
fi
g
u
re
s
o
f
th
e
E
u
ro
p
ea
n
a
cu
te
c
ar
e
la
n
d
sc
ap
e
in
1
9
9
5
a
n
d
2
0
0
8
C
o
u
n
tr
y
A
u
st
ri
a
E
st
o
n
ia
F
in
la
n
d
F
ra
n
c
e
G
e
rm
a
n
y
Ir
e
la
n
d
F
a
c
to
r
m
in
im
u
m
–
m
a
x
im
u
m
2
0
0
8
D
a
ta
y
e
a
rs
1
9
9
5
2
0
0
8
C
h
a
n
g
e
ra
te
1
9
9
5
2
0
0
8
C
h
a
n
g
e
ra
te
1
9
9
5
2
0
0
8
C
h
a
n
g
e
ra
te
1
9
9
5
2
0
0
8
C
h
a
n
g
e
ra
te
1
9
9
5
2
0
0
8
C
h
a
n
g
e
ra
te
1
9
9
5
2
0
0
8
C
h
a
n
g
e
ra
te
A
cu
te
c
ar
e
h
o
sp
it
al
s
p
er
1
0
0
0
0
0
1
.8
1
.6
–1
2
.8
5
.1
2
.8
–4
5
.7
n
.a
.
n
.a
.
n
.a
.
3
.8
d
3
.2
–1
6
.8
2
.6
2
.2
–1
4
.9
1
.7
1
.2
–2
9
.3
5
.0
A
cu
te
c
ar
e
h
o
sp
it
al
b
ed
s
p
er
1
0
0
0
6
.5
5
.6
–1
3
.2
6
.6
3
.9
–4
1
.9
3
.0
1
.9
–3
6
.2
4
.9
a
3
.6
–2
7
.1
6
.9
5
.7
–1
8
.2
3
.1
2
.7
j
–1
3
.1
3
.0
A
cu
te
c
ar
e
h
o
sp
it
al
ad
m
is
si
o
n
s
p
er
1
0
0
2
1
.9
2
6
.7
2
2
.0
1
7
.2
1
6
.8
–2
.6
1
9
.9
1
9
.0
–4
.7
2
0
.3
1
6
.7
–1
8
.1
1
9
.2
2
0
.7
8
.0
1
4
.6
1
3
.6
–7
.2
2
.4
A
L
O
S,
a
cu
te
c
ar
e
h
o
sp
it
al
s
o
n
ly
9
.7
6
.8
–2
9
.9
1
0
.5
5
.8
–4
5
.2
5
.4
3
.9
–2
7
.8
5
.9
5
.8
–2
.2
1
0
.8
7
.6
–2
9
.6
6
.6
6
.2
–5
.8
1
.9
B
ed
o
cc
u
p
an
cy
ra
te
i
n
%
,
ac
u
te
ca
re
h
o
sp
it
al
s
o
n
ly
7
7
.0
8
0
.4
4
.4
7
4
.9
6
9
.7
–7
.0
7
4
.0
n
.a
.
n
.a
.
7
5
.3
7
6
.4
1
.5
8
1
.8
7
6
.2
–6
.8
8
2
.5
8
8
.9
7
.8
1
.6
T
o
ta
l
in
p
at
ie
n
t
ex
p
en
d
it
u
re
a
s
%
o
f
to
ta
l
h
ea
lt
h
ex
p
en
d
it
u
re
4
0
.4
4
0
.0
–1
.0
3
5
.7
c
3
3
.0
–7
.6
4
2
.1
3
5
.0
–1
6
.9
4
0
.9
3
7
.3
–8
.8
3
6
.4
3
4
.1
–6
.3
n
.a
.
n
.a
.
n
.a
.
1
.9
P
h
y
si
ci
an
s
p
er
1
0
0
0
3
.5
4
.6
3
2
.4
3
.2
3
.3
4
.9
2
.2
2
.7
2
4
.1
3
.3
3
.4
4
.5
3
.1
3
.6
1
6
.1
2
.1
3
.1
4
8
.2
2
.1
%
o
f
p
h
y
si
ci
an
s
w
o
rk
in
g
i
n
h
o
sp
it
al
s
5
6
.2
5
5
.1
–1
.9
n
.a
.
6
7
.1
n
.a
.
8
5
.1
8
1
.2
j
–4
.6
4
7
.1
5
4
.1
1
4
.9
4
8
.2
5
1
.1
6
.0
5
3
.2
4
7
.1
–1
1
.4
2
.1
C
on
ti
n
u
ed
o
ve
rl
ea
f
C
o
u
n
tr
y
N
e
th
e
rl
a
n
d
s
P
o
la
n
d
P
o
rt
u
g
a
l
S
p
a
in
S
w
e
d
e
n
U
n
it
e
d
K
in
g
d
o
m
F
a
c
to
r
m
in
im
u
m
–
m
a
x
im
u
m
2
0
0
8
D
a
ta
y
e
a
rs
1
9
9
5
2
0
0
8
C
h
a
n
g
e
ra
te
1
9
9
5
2
0
0
8
C
h
a
n
g
e
ra
te
1
9
9
5
2
0
0
8
C
h
a
n
g
e
ra
te
1
9
9
5
2
0
0
8
C
h
a
n
g
e
ra
te
1
9
9
5
2
0
0
8
C
h
a
n
g
e
ra
te
1
9
9
5
2
0
0
8
C
h
a
n
g
e
ra
te
A
cu
te
c
ar
e
h
o
sp
it
al
s
p
er
1
0
0
0
0
0
0
.8
0
.6
–2
3
.2
2
.2
g
2
.0
–6
.9
1
.4
1
.3
–7
.1
1
.5
1
.2
–1
7
.6
1
.1
0
.9
f
–2
1
.6
n
.a
.
n
.a
.
n
.a
.
5
.0
A
cu
te
c
ar
e
h
o
sp
it
al
b
ed
s
p
er
1
0
0
0
3
.5
2
.9
–1
9
.2
5
.8
4
.4
–2
3
.4
3
.2
2
.8
–1
3
.6
3
.0
2
.5
–1
8
.0
3
.0
2
.2
h
–2
8
.2
2
.3
2
.7
1
9
.2
3
.0
A
cu
te
c
ar
e
h
o
sp
it
al
ad
m
is
si
o
n
s
p
er
1
0
0
9
.6
1
1
.1
1
5
.4
n
.a
.
n
.a
.
n
.a
.
1
1
.1
1
1
.2
j
0
.9
1
0
.7
1
1
.4
6
.7
1
6
.3
1
5
.2
j
–6
.6
2
1
.2
1
2
.2
–4
2
.5
2
.4
A
L
O
S,
a
cu
te
c
ar
e
h
o
sp
it
al
s
o
n
ly
9
.9
5
.9
–4
0
.4
n
.a
.
n
.a
.
n
.a
.
7
.9
6
.8
j
–1
3
.9
8
.2
6
.5
–2
0
.7
5
.2
6
.0
j
1
5
.4
8
.5
d
7
.4
–1
2
.8
1
.9
B
ed
o
cc
u
p
an
cy
ra
te
i
n
%
,
ac
u
te
ca
re
h
o
sp
it
al
s
o
n
ly
7
3
.3
5
5
.7
–2
4
.0
n
.a
.
n
.a
.
n
.a
.
7
2
.6
7
2
.6
j
0
.0
7
6
.4
7
7
.7
1
.7
7
5
.9
n
.a
.
n
.a
.
8
2
.3
d
8
4
.8
3
.1
1
.6
T
o
ta
l
in
p
at
ie
n
t
ex
p
en
d
it
u
re
a
s
%
o
f
to
ta
l
h
ea
lt
h
ex
p
en
d
it
u
re
4
9
.1
3
9
.6
e
–1
9
.3
n
.a
.
3
2
.0
n
.a
.
3
3
.9
2
0
.8
i
–3
8
.6
3
1
.0
2
8
.2
–9
.0
5
4
.7
2
9
.6
j
–4
5
.9
n
.a
.
n
.a
.
n
.a
.
1
.9
P
h
y
si
ci
an
s
p
er
1
0
0
0
2
.9
b
3
.7
j
2
6
.1
2
.3
2
.2
–6
.7
2
.9
3
.7
2
5
.2
2
.7
3
.5
3
0
.1
2
.9
3
.6
i
2
4
.0
1
.8
2
.6
4
6
.4
2
.1
%
o
f
p
h
y
si
ci
an
s
w
o
rk
in
g
i
n
h
o
sp
it
al
s
n
.a
.
3
8
.1
j
n
.a
.
n
.a
.
5
2
.1
n
.a
.
6
0
.2
5
4
.1
–1
0
.1
5
3
.1
6
1
.1
1
5
.1
n
.a
.
n
.a
.
n
.a
.
n
.a
.
n
.a
.
n
.a
.
2
.1
K
ey
:
a 1
9
9
3
;
b
1
9
9
7
;
c 1
9
9
9
;
d
2
0
0
0
;
e 2
0
0
1
;
f 2
0
0
3
;
g
2
0
0
4
;
h
2
0
0
5
;
i 2
0
0
6
;
j 2
0
0
7
d
u
e
to
l
ac
k
o
f
av
ai
la
b
le
d
at
a
fo
r
1
9
9
5
/2
0
0
8
.
So
u
rc
e:
W
H
O
R
eg
io
n
al
O
ffi
c
e
fo
r
E
u
ro
p
e,
2
0
1
1
.
N
ot
es
:
n
/a
n
o
t
av
ai
la
b
le
;
m
in
im
u
m
a
n
d
m
ax
im
u
m
v
al
u
es
a
re
p
re
p
ar
ed
i
n
i
ta
li
cs
.
T
a
b
le
2
.3
C
on
ti
n
u
ed
Introduction to DRGs in Europe 19
The ownership structure of hospitals also varies widely. In some countries
(such as France and Germany) private profi t-making hospitals play an important
role in the health system, but in many others, most hospital beds are operated
under public or private non-profi t-making ownership (Table 2.4).
In summary, DRGs were introduced in countries that were characterized by
large structural differences in their hospital sectors. Furthermore, despite the
introduction of DRG-based hospital payment systems, structural differences
persist. Apparently, DRG-based hospital payment systems can be fl exibly
implemented in different settings and do not prescribe any clear development
path. Therefore, the analysis of how these systems have been implemented in
different health care contexts – as well as their impact on the effi ciency and
quality of service delivery – is the main aim of this book.
2.5 The book and its structure
Comparative information from different European countries regarding the
specifi c characteristics of their DRG systems and how these characteristics
contribute to achieving the aims of transparency, effi ciency and quality in
hospitals is largely absent. This book aims to contribute towards fi lling this gap.
This book is structured in two parts: Part One (up to and including Chapter
10) deals with the essential building blocks of DRG-based hospital payment
systems and discusses the impact of these systems on quality and effi ciency of
service delivery, and on adoption and use of technological innovation. The aim
of this fi rst part is to highlight similarities and differences between different
countries’ systems, and to provide an overview of the key issues that need to be
considered when developing and optimizing DRG-based hospital payment
systems. At the same time, the discussion of these issues paves the way for
Table 2.4 Share of ownership types across countries (% of acute care beds), 2008
Country Publically owned Non-profi t- Profi t-making,
hospitals (%) making, privately privately owned
owned hospitals (%) hospitals (%)
Austria 73 19 9
Finland 89 0 11
France 66 9 25
Germany 49 36 15
Ireland 88 0 12
Netherlands 0 100 0
Poland 95 0 5
Portugal 86 7 8
Spain 74 17 9
Sweden 98 0 2
United Kingdom 96 4 0
Source: Paris et al., 2010.
Note: In Estonia all hospitals operate under private law and most of them are publically owned
(see Chapter 17 of this volume).
20 Diagnosis-Related Groups in Europe
considering the potential for harmonization of DRG systems and of DRG-based
hospital payment across Europe. The ‘building blocks’ of DRG systems and of
DRG-based hospital payment systems are described in detail in Chapter 3. The
differences in DRG systems and similar patient classifi cation systems (PCSs)
across Europe are analysed in Chapter 4. The main challenges and differences
in cost-accounting systems used in European countries are highlighted in
Chapter 5. The intended and unintended consequences of using DRGs for
hospital payment (rather than just patient classifi cation) are discussed in
Chapter 6. The available evidence of the impact of DRGs on the effi ciency and
quality of hospital care is discussed in chapters 7 and 8. Chapter 9 describes
how the 12 countries included in this book attempt to overcome the potential
problems relating to technological innovation that are associated with DRG-
based hospital payment systems. Finally, the fi rst part of the book closes in
Chapter 10 with a summary of the main fi ndings, and provides policy
recommendations for further DRG developments.
Part Two of the book – that is, chapters 11 to 23 – provides clearly structured
and detailed information about the most important DRG system characteristics
in each of the 12 countries that participated in the EuroDRG project (Austria,
England, Estonia, Finland, France, Germany, Ireland, Poland, Portugal, Spain,
Sweden and the Netherlands). Each country has an interesting story to tell,
which is contextualized within the prevailing health system. Part Two aims to
overcome some of the diffi culties that have existed for both researchers and
policy-makers aiming to compare European DRG systems: information used to
be mostly available only in national languages, and national descriptions of
DRG systems often used country-specifi c terminology that complicated the task
of making cross-border comparisons. Each country-specifi c chapter starts with
a background section that provides an overview to hospital services and to the
role of DRGs in the country (section 1). The developments and updates of the
DRG systems are outlined in the second section. The current DRG system(s),
which is (are) used to group patients into clinically meaningful and cost-
homogeneous groups, is (are) described in section 3 of each country-specifi c
chapter, while section 4 in each case deals with the countries’ cost-accounting
systems that are essential for determining DRG-based payments rates. The
DRG-based hospital payment system of the country in question is described in
section 5, and the countries’ methods to integrate new and innovative
treatments into their existing DRG-based payment systems are presented in
section 6. Finally, each chapter closes with an assessment of the country’s DRG
system (section 7) and a summary of the outlook, in terms of future developments
and reform (section 8).
2.6 References
Barnum, H., Kutzin, J., Saxenian, H. (1995). Incentives and provider payment methods.
International Journal of Health Planning and Management, 10(1):23–45.
Introduction to DRGs in Europe 21
Busse, R., Schreyögg, J., Smith, P.C. (2006). Hospital case payment systems in Europe.
Health Care Management Science, 9(3):211–13.
Fetter, J., Shin, Y., Freeman, J.L., Averill, R.F., Thompson, J.D. (1980). Casemix defi nition
by diagnosis-related groups. Medical Care, 18(Suppl. 2):1–53.
Langenbrunner, J.C., Wiley, M. (2002). Hospital payment mechanisms: theory and
practice in transition countries, in M. McKee, J.H. Healy, eds. Hospitals in a Changing
Europe. Buckingham: Open University Press.
McKee, M., Healy, J.H. (2002). Hospitals in a Changing Europe. Buckingham: Open
University Press.
Miller, T.S. (1997). The Birth of the Hospital in the Byzantine Empire. Baltimore, MD: Johns
Hopkins University Press.
Paris, V., Devaux, M., Wei, L. (2010). Health Systems Institutional Characteristics: A Survey of
29 OECD Countries. Paris: Organisation for Economic Co-operation and Development
(OECD Health Working Papers No. 50).
Schreyögg, J., Stargardt, T., Tiemann, O., Busse, R. (2006). Methods to determine
reimbursement rates for diagnosis related groups (DRG): a comparison of nine
European countries. Health Care Management Science, 9(3):215–23.
Shleifer, A. (1985). A theory of yard stick competition. RAND Journal of Economics,
16(3):319–27.
Street, A., Vitikainen, K., Bjorvatn, A., Hvenegaard, A. (2007). Introducing Activity-Based
Financing: A Review of Experience in Australia, Denmark, Norway and Sweden. York:
University of York Centre for Health Economics (CHE Research Paper 30).
WHO (2000). The World Health Report 2000 – Health Systems: Improving Performance.
Geneva: World Health Organization.
WHO Regional Offi ce for Europe (2011). European Health for All Database (HFA-DB) [online
database]. Copenhagen: WHO Regional Offi ce for Europe (http://data.euro.who.int/
hfadb/, accessed 20 July 2011) (July 2011 update).
chapter t h r e e
Understanding DRGs
and DRG-based hospital
payment in Europe
Wilm Quentin, Alexander Geissler, David
Scheller-Kreinsen and Reinhard Busse
3.1 Introduction
Despite the fact that diagnosis-related groups (DRGs) have been adopted in an
increasingly large number of countries around the world (Kimberly et al., 2008),
understanding of DRG systems and DRG-based hospital payment systems
remains surprisingly limited. On the one hand, there is no good overview of
alternative options for designing these systems because systematic comparisons
of the specifi c system characteristics in different countries are extremely rare
(France, 2003). Consequently, there is no agreed consensus on how best to
design DRG systems and DRG-based hospital payment systems, because the dif-
ferences between countries’ systems remain poorly understood. On the other
hand, despite the existence of numerous studies concerning the effects of DRG-
based hospital payment systems on hospital effi ciency, quality and technological
innovation, these effects remain relatively unclear (Brügger, 2010) – also because
the specifi c design features in different countries are rarely taken into account.
Nevertheless, a thorough understanding of international experiences with
DRG systems and DRG-based hospital payment systems could inform countries
when developing and optimizing their national systems. In addition, in a
context of growing patient mobility facilitated by the European Union (EU)
Directive on the Application of Patients’ Rights in Cross-Border Healthcare (European
Parliament and Council, 2011), an increasingly important issue relates to
whether there is scope for harmonization of DRG systems within Europe. This
is because if harmonization is not possible, it will remain diffi cult (or at least
not transparent) to pay hospitals in one EU Member State for care provided to
patients from another EU Member State. Furthermore, cross-border comparisons
of hospital prices and performance – which are increasingly being conducted in
24 Diagnosis-Related Groups in Europe
attempts to improve the understanding of differences in terms of effi ciency and
costs (see, for example, Chapko et al., 2009 or Busse et al., 2008) – will continue
to be complicated by the lack of a common basis for comparison.
The fi rst part of this book aims to contribute to a better understanding of
DRG systems and of how they are used for hospital payment in Europe by (1)
systematically comparing DRG systems and DRG-based hospital payment
systems across 12 European countries; and (2) by providing an overview of the
effects of these systems on hospital effi ciency, quality, and on the adoption and
use of technological innovation. This chapter develops a framework for com-
paring DRG systems and DRG-based hospital payment systems in Europe. It
presents the main building blocks of DRG-based hospital payment systems and
introduces some of the assumed effects of these systems on hospital effi ciency,
quality and on the adoption and use of technological innovation. It highlights
certain key concepts and raises a number of questions that should be considered
when developing or optimizing DRG systems and DRG-based hospital payment
systems. The six chapters that follow (chapters 4 to 9) develop these points in
more detail. The scope for harmonization of DRG systems – or at least for more
cooperation – in Europe is a cross-cutting issue that is discussed in several of the
chapters. Chapter 10 picks up the questions raised here and draws conclusions
from all the chapters in Part One.
3.2 How to understand DRG systems and DRG-based hospital
payment systems in Europe
Trying to understand DRG systems and how they are used for hospital payment
across countries requires a common framework. Without one, it is easy to get
lost in the specifi cities of each country’s system and confused by the diversity
of terms that are used for describing similar things in different countries. Figure
3.1 presents a framework that we developed to guide the reader through the
chapters of the fi rst part of this book. All countries included in this book have
a DRG system (Chapter 4), a system to collect cost information from hospitals
(Chapter 5), and they use DRGs for hospital payment (Chapter 6). These
building blocks are presented in more detail in subsection 3.2.1. Under-
standing each of these building blocks and how they interact is essential for
understanding the effects of DRG-based hospital payment systems on effi ciency
(Chapter 7), quality (Chapter 8) and innovation (Chapter 9), which are
introduced in subsection 3.2.2. The effect of DRG systems on transparency of
service provision is not discussed in a separate chapter. However, increased
transparency resulting from the use of DRGs is thought to contribute to both
improved effi ciency and quality of service provision.
3.2.1 Understanding the building blocks
DRG systems (Chapter 4)
A DRG system is a patient classifi cation system (PCS) that has four main char-
acteristics: (1) routinely collected patient discharge data (mostly concerning
Understanding DRGs and DRG-based hospital payment in Europe 25
patient, treatment and provider characteristics) are used to classify patients into
(2) a manageable number of groups (that is, DRGs), which are intended to be
(3) clinically meaningful and (4) economically homogeneous. DRGs summarize the
confusingly large number of different (individual) patients treated by hospitals
into a manageable number of clinically meaningful and economically hom-
ogeneous groups, thus providing a concise measure of hospital activity or, in
other words, they defi ne hospital products. Consequently, they facilitate com-
parisons of hospital costs, quality and effi ciency, and contribute to increased
transparency in hospitals.
When introducing DRG systems, two alternative options exist: DRG systems
can either be adopted from abroad or they can be developed from scratch.
Many countries originally adopted DRG systems from abroad and later used
these systems as the basis for further developing their own systems (Chapter 4).
Consequently, eight countries included in this book (Estonia, Finland, France,
Germany, Ireland, Portugal, Spain and Sweden) use DRG systems that are at
least remotely related to the Health Care Financing Administration (HCFA-)
DRG system originally introduced in the United States in the early 1980s; two
of them (Germany and Ireland) via the Australian Refi ned (AR-)DRG system.
Austria, England, and the Netherlands have developed their own systems from
scratch, while Poland used the English version to develop its own system.
Although these self-developed systems do not defi ne DRGs in the strictest sense
Figure 3.1 Framework for navigating through the book
26 Diagnosis-Related Groups in Europe
of the word (that is, groups are not necessarily diagnosis-related), this book uses
the term DRG system for all PCSs that share the above-mentioned four main
characteristics.
The actual classifi cation of patients into DRGs is almost always performed by
computerized grouping software. Since diagnoses and procedures are the most
important classifi cation variables, an essential requirement for the operation of
DRG systems is that diagnoses and procedures are coded in hospitals according
to standardized classifi cation systems, such as modifi cations of the International
Classifi cation of Diseases 10th revision (ICD-10) for diagnoses and country-
specifi c classifi cations of procedures, such as the English Offi ce of Population
Censuses and Surveys (OPCS) Classifi cation of Surgical Operations and Proce-
dures. However, how this information is used for defi ning DRGs depends on
the specifi c DRG grouping algorithm. While the general structure of many DRG
systems is relatively similar (see Chapter 4), the precise defi nition of specifi c
DRGs can be quite diverse (Quentin et al., in press).
Ideally, DRG systems should consider the most important determinants of
resource consumption as classifi cation variables; that is, they should defi ne
DRGs on the basis of those diagnoses, procedures or other classifi cation vari-
ables that make treating one patient with (or without) a specifi c procedure
more expensive than treating another patient with (or without) another pro-
cedure. Otherwise, if DRG systems fail to defi ne economically homogeneous
groups, performance comparisons on the basis of DRGs do not adequately con-
trol for differences of patients within DRGs. Further, DRG-based hospital pay-
ment may be inappropriate for a considerable number of patients – it can be
either too high or too low. However, because hospitals may try to manipulate
the classifi cation of patients into DRGs by changing their coding or practice
patterns (see Chapter 6), the selection of classifi cation variables also needs to
consider whether those variables are easy to manipulate or not.
In order to ensure that DRGs remain clinically meaningful and economically
homogeneous, even when technological innovation or other factors lead to
changes in practice patterns and costs, DRG systems need to be updated at
regular intervals (see Chapter 9). Most countries use some kind of hospital cost
information (see Chapter 5) to develop and update DRG systems, as illustrated
by an arrow in Figure 3.1.
Furthermore, different alternatives exist in terms of the unit for which
patients are classifi ed into DRGs. For example, some DRG systems classify
patients into one DRG per hospital admission; other systems classify patients
into DRGs for every stay in a hospital department (see the Finnish system,
described in Chapter 18); and still other systems classify patients into DRGs for
a specifi c treatment related to a specifi c diagnosis, which may include several
inpatient stays and outpatient visits (see Chapter 23 on the Netherlands).
Finally, it is important to consider which patient groups are to be included in
DRG systems. For some groups of patients, it may be more diffi cult to defi ne
clinically meaningful and economically homogeneous groups of patients. For
example, psychiatric patients were originally excluded from DRG systems in
most countries because it appeared to be more diffi cult to defi ne economically
homogeneous groups on the basis of diagnoses and procedures for this group of
patients (Lave, 2003; McCrone & Phelan, 1994). However, several countries are
Understanding DRGs and DRG-based hospital payment in Europe 27
now also in the process of developing or introducing DRGs for psychiatric
patients (see Chapters 4 and 10).
Hospital cost information (Chapter 5)
As illustrated in Figure 3.1, hospital cost information is used to (1) defi ne DRGs
and (2) determine (adjust) payment rates. The availability of high-quality
hospital cost information is essential for developing and updating DRG systems
and for ensuring fair DRG-based hospital payment systems. If hospital cost
information does not allow differences to be indentifi ed between costs of
individual patients, it is impossible to use a data-driven approach to develop
economically homogeneous DRGs. In addition, if hospital cost information is
imprecise, calculated weights for certain DRGs could be falsely estimated to be
higher or lower than they really are and, consequently, hospitals will be over- or
underpaid for specifi c DRGs. Therefore, the fairness of DRG-based hospital
payment systems and the ability of these systems to encourage effi ciency are to
a large extent determined by the quality of the hospital cost information used
to develop these systems and to calculate DRG weights.
Unfortunately, the availability of standardized and (therefore) comparable
high-quality cost information is limited in many European countries. This is
one of the reasons why some countries without high-quality patient-level cost
information have imported DRG systems including weights from abroad (for
example, Ireland, Portugal and Spain) and have only adjusted the imported
DRG weights to the local cost context by using more aggregated cost-accounting
fi gures, for example at the department level. Other countries have developed
their own DRG systems on the basis of length-of-stay data as a proxy for costs,
which makes it diffi cult to ensure that groups are economically homogeneous.
Furthermore, hospital charges from fee-for-service payment systems (that in
some countries existed prior to the introduction of DRGs), individual costing
studies or even expert opinions have occasionally been used as a proxy for costs
when calculating weights of DRGs (for example, in Estonia and Poland).
However, in many European countries, the introduction of DRGs has also
encouraged changes in hospitals’ cost-accounting systems. Following the intro-
duction of DRGs for classifi cation purposes – and even more so following the
introduction of DRG-based hospital payment systems – standardized (sometimes
mandatory) cost-accounting systems have been introduced in at least a sample
of hospitals in most countries. Countries have often introduced national cost-
accounting handbooks, which provide detailed rules and defi nitions concerning
the types of cost centres and cost categories to be used, and which specify the
allocation methods and allocation bases for distribution of costs to fi nal cost
centres and patients.
Yet, signifi cant differences exist between countries in terms of (1) the number
of hospitals that participate (voluntarily or not) in collecting standardized cost-
accounting information; and (2) the level of detail required according to the
national cost-accounting standards. In fact, there may be a trade-off between
collecting detailed patient-level data using a bottom-up micro-costing approach
(see Chapter 5) and the goal of having a large representative sample of hospitals
contributing to a national cost database (Schreyögg et al., 2006). This is because
28 Diagnosis-Related Groups in Europe
a more complex cost-accounting system increases the costliness of the data-
collection exercise, which may become prohibitively costly if data collection is
extended to a large number of hospitals. In addition, as the importance of
hospital cost information has increased, most countries have introduced regular
data checks, with the aim of assuring the validity of reported hospital cost
information.
While changes to cost-accounting systems resulting from regulations have
been important, the introduction of DRG-based hospital payment systems has
also increased the intrinsic motivation for hospital managers to introduce or
optimize existing cost-accounting systems (see Berki, 1985). Without high-
quality cost-accounting systems, hospital managers do not know whether
hospitals are able to ‘produce’ DRGs at costs that are below the payment rate.
Consequently, they do not know whether hospitals are making a profi t or are
incurring a loss by providing these DRGs. In addition, in order to be able to
identify the cost drivers of hospital products (that is, of DRGs) and to manage
the production of DRGs, hospital managers require detailed information about
the costs of different inputs in the production process – an element that can be
provided by high-quality cost-accounting systems. In summary, as illustrated
by the arrows in Figure 3.1, hospital cost information is a necessary input for
effective DRG-based hospital payment systems, and (ideally) improved hospital
cost information is also an outcome of the changed incentive structure following
the introduction of DRG-based hospital payment.
DRG-based hospital payment (Chapter 6)
While the use of DRGs for reporting purposes and for managing hospitals is
important, most countries included in this book use DRGs primarily as the basis
for hospital payment. In general, two main models of DRG-based hospital pay-
ment system can be distinguished. On the one hand, in DRG-based case pay-
ment systems, each discharged patient is grouped into the applicable DRG, and
hospitals receive a payment per case that is determined by the weight of that
DRG (after monetary conversion and relevant adjustments). On the other hand,
in DRG-based budget allocation systems, the available regional or national hos-
pital budget is distributed to individual hospitals on the basis of the number
and type of DRGs that these hospitals produced during one of the previous
years or that they are expected to produce in the next year. The casemix (that
is, the sum of the weights of all DRGs produced by a hospital) and the casemix
index (CMI) (that is, the casemix divided by the number of discharges) are usu-
ally the determining factors for distributing the budget. However, adjustments
for structural indicators and for certain high-cost cases are also considered, and
an implicit monetary conversion rate exists that can be used to estimate the
implicit revenue contribution to the hospital budget of one patient in a specifi c
DRG. In addition, some countries with DRG-based case payment systems, such
as Germany or Finland, use DRGs to negotiate global hospital budgets, which
limit (to a certain extent) the total amount of money that hospitals can earn
from DRG-based case payments.
There are three main incentives for hospitals resulting from DRG-based
hospital payment systems: (1) to reduce costs per treated patient, (2) to increase
Understanding DRGs and DRG-based hospital payment in Europe 29
revenues per patient, and (3) to increase the number of patients. These
incentives can have both intended and unintended consequences on effi ciency,
quality and technological innovation. However, the strength of these incentives
is determined by the type of DRG-based hospital payment systems (case-based
payment versus budget allocation), by the proportion of total hospital revenues
related to DRG-based hospital payment, and by the degree to which DRG
weights and monetary conversion rates are adjusted to refl ect hospital-specifi c
cost structures.
In all DRG-based hospital payment systems (except for that operating in the
Netherlands), the actual payment rate is not the same as the DRG weight. As
illustrated in Figure 3.1, three main approaches for expressing DRG weights
exist in the countries included in this book: (1) relative weights, (2) raw tariffs,
and (3) scores. Each of these approaches corresponds to a specifi c monetary
conversion method. In countries using a relative weight approach, the relative
weight provides a measure that relates the average costs of treating patients
within one DRG to the average costs of treating all patients included in the
DRG system in the country (see Chapter 6). Table 3.1 provides an example of
how DRG weights for a hypothetical DRG determine hospital payment within
the framework of the different approaches for expressing DRG weights. The
idea of all three approaches is the same: the DRG weight (almost always)
provides a measure of the average or expected costs of treating patients falling
into that DRG. The actual hospital payment rate is calculated by multiplying
the DRG weight with a country-specifi c monetary conversion/adjustment
rate, which often takes into account structural, regional or hospital-specifi c
differences in the costs of service provision.
Monetary conversion/adjustment rates may differ between types of hospitals,
for example, by degree of specialization or geographic location, according to
the country-specifi c choices for adjusting the DRG-based payment rate. Some-
times, monetary conversion rates are hospital specifi c, and are calculated in a
way that shelters hospitals from budget cuts, which means that the incentives
of DRG-based hospital payment are much reduced (see, for example, the Finn-
ish system described in Chapter 18). In addition, DRG weights are generally
adjusted in order to account for certain high-cost patients that stay in hospital
much longer than the average case, or that receive additional services, which
are not adequately reimbursed on the basis of the DRG-based payment system.
Furthermore, most countries operating DRG-based case payment systems
prevent an excessive increase in costs by applying macro-level price/volume
Table 3.1 DRG weights and monetary conversion example
Hypothetical example
DRG weight DRG weight (unit) Monetary conversion/ Hospital payment
approach adjustment (unit) rate (€)
Relative weight 1.95 � 2 000 € = 3 900 €
Raw tariff 3 000 € � 1.3 = 3 900 €
Score 130 points � 30 € = 3 900 €
30 Diagnosis-Related Groups in Europe
control measures (such as global hospital budgets, sectoral budgets, or price
reductions).
3.2.2 Understanding the effects of
DRG-based hospital payment
Effects on hospital effi ciency (Chapter 7)
In many countries, one of the most important purposes of introducing DRG-
based hospital payment systems was to increase effi ciency of hospital care.
Because DRG-based hospital payment provides incentives to increase activity
and to minimize costs, there is reason to believe that these systems contribute
to improved effi ciency. However, ‘effi ciency’ is a widely used term that can have
various meanings. Economists generally differentiate between technical
effi ciency – that is, maximizing outputs for a given level of inputs, or minimizing
inputs for a given level of outputs; allocative effi ciency – namely, ensuring the
appropriate mix of inputs and outputs to maximize utility; and cost-effi ciency –
that is, minimizing costs for a given level of output.
DRG-based hospital payment systems are often discussed as representing a
form of ‘yard stick competition’ (Shleifer, 1985). The idea of yard stick com-
petition is that prices for a given product (for example, a specifi c DRG) are set
at the level of average costs of other fi rms producing the same product (that is,
the same DRG). With DRG-based hospital payment, if hospitals produce DRGs
at costs that are below the average costs of other hospitals, they benefi t directly
by retaining the generated fi nancial surplus; if they underperform, they gener-
ate defi cits and, ultimately, risk bankruptcy. All hospitals, including the
most effi cient ones, are incentivized to continually reduce costs. In practice,
numerous options exist for hospitals to increase (technical and cost-) effi ci-
ency: care pathways can be optimized to reduce the length of stay; duplicate and
unnecessary tests can be avoided; and costly treatments can be replaced by
similarly effective but less costly alternatives.
However, unfortunately, if the incentives for cost reduction are too strong,
and if regulatory authorities do not have suffi cient capacity to monitor ade-
quately the quality of care, DRG-based hospital payment can lead to unintended
consequences (see Chapter 6). For example, hospitals could discharge patients
inappropriately early, and service intensity could be reduced to a level at which
necessary services are withheld from patients – thus leading to cost reductions
but not to improvements in effi ciency. Consequently, the effects of DRG-based
hospital payment systems on effi ciency have been highly controversial.
Although improving hospital effi ciency is generally a key motivation for
introducing DRG-based hospital payment, relatively few studies have explicitly
identifi ed and quantifi ed its impact using established methods, such as data
envelopment analyses (DEAs) or stochastic frontier analyses ( Jacobs et al.,
2006). Rather, most research has concentrated on indicators of effi ciency – such
as activity, length of stay and costs – which are more easily measured, but by
defi nition provide only a partial picture. Given the challenges inherent in
undertaking cross-country effi ciency comparisons, most available studies have
Understanding DRGs and DRG-based hospital payment in Europe 31
adopted a longitudinal perspective, comparing hospital effi ciency before and
after the introduction of DRG-based hospital payment. Chapter 7 reviews both
types of studies, effi ciency analyses and studies of indicators of effi ciency.
When interpreting the results of these (longitudinal) studies, it is important
to consider the diffi culties that often arise in clearly separating the effect of the
introduction of DRG-based hospital payment from other concurrent infl uences,
such as changes in medical technology or new legislation. Furthermore, in
longitudinal studies, the measured effect of introducing DRG-based hospital
payment depends on the hospital payment system that existed prior to the
introduction of the system. In the United States, where DRG-based hospital
payment replaced a fee-for-service system, the DRG-based hospital payment
system provided strong incentives to reduce costs (Berki, 1985). In contrast, in
Europe, where DRG-based hospital payment systems often replaced global
hospital budgets, the incentives of DRG-based hospital payment would be
expected to lead to an increase in hospital activity, which could also result in
increased costs.
Effects on hospital quality (Chapter 8)
The effect of DRGs on hospital quality is not straightforward (Davis & Rhodes,
1988; Farrar et al., 2009). On the one hand, because DRGs provide a concise and
meaningful measure of hospital activity and thus facilitate monitoring and
comparisons of hospital quality, they could contribute to better quality of care.
In addition, cost-reduction incentives of DRG-based hospital payment systems
could lead to increased efforts to improve quality, if quality contributes to
reduced costs. For example, improved coordination between hospitals, out-
patient providers and long-term care facilities would reduce costs but could
also contribute to better quality of care. However, on the other hand, and this
has been a reason for continuous concern (Rogers et al., 1990), hospitals may
be tempted to reduce costs by reducing quality, if DRG-based payments do
not depend on quality. For example, because DRGs do not specify which ser-
vices must be provided when treating a specifi c patient, hospitals can ‘skimp’
on quality by avoiding certain diagnostic tests, disregarding hygiene standards,
or by lowering staffi ng ratios per bed.
Assessments of the effect of DRGs on hospital quality are often complicated
by the fact that the notion of quality is rather diffuse (Legido-Quigley et al.,
2008). This book defi nes quality as any aspect of hospital services that benefi ts
patients during the process of treatment or improves health outcome after
treatment (Chalkley & Malcomson, 1998). To measure ‘quality’, a common
framework developed by Donabedian (1966) differentiates between structural,
process and outcome indicators of quality. Structural indicators, such as
qualifi cation(s) of medical staff or available equipment are easy to measure and
are relevant to quality if they represent conditions for the delivery of a given
quality of health care. Process indicators can also be measured relatively easily,
but should be based on the available evidence of what constitutes ‘good’ quality
of care in the treatment of a specifi c patient and in a specifi c situation (Smith
et al., 2010). Thus, they usually provide clear pathways for action. Outcome
32 Diagnosis-Related Groups in Europe
indicators assess what is most meaningful for policy-makers and patients (for
example, mortality), but it is not always possible to determine the contribution
of health care to health outcomes because outcomes are also infl uenced by
(unobserved) patient-level factors. Therefore, careful risk adjustment is neces-
sary if outcome indicators are to be used.
The effect of DRGs and of DRG-based hospital payment systems on the
quality of hospital care has been assessed in numerous studies from the United
States and several studies from Europe, using a range of indicators. Again, when
interpreting the results, which are presented in Chapter 8, it is important to
consider that the effects of the introduction of DRGs and of DRG-based hospital
payment on quality may be different depending on the hospital payment
system previously in existence.
In theory, DRG-based hospital payment systems could be modifi ed to
explicitly consider quality of care. However, basic information on the quality of
services provided is still lacking in most countries in which DRGs are used for
hospital payment. Yet, the availability of information regarding the quality of
services (in terms of structure, process or outcomes) is a prerequisite for any
attempts to explicitly integrate fi nancial incentives for quality into DRG-based
hospital payment systems (see Chapter 8).
Effects on technological innovation (Chapter 9)
Since the introduction of DRG-based hospital payment systems, there have
been concerns that these systems may not provide suffi cient incentives to
encourage the desired adoption and use of technological innovations in health
care (OTA, 1983; MedPAC, 2003; Shih & Berliner, 2008). However, the effect of
DRG-based hospital payment systems on any specifi c technological innovation
depends on how the technological innovation infl uences total hospital costs
(both capital and operating costs) per admission.
Technological innovations may increase or decrease capital costs, operat-
ing costs or both. DRG-based hospital payment systems encourage hospitals
to invest in technological innovations that reduce total costs per patient
and discourage hospitals from introducing technological innovations that
lead to higher costs per patient. Yet, whether this effect of DRG-based hospital
payment on technological innovation is socially desirable or not depends
on whether the innovations in question really improve the quality of care.
In cases in which technological innovation is more costly but does not improve
quality of care, the effect of DRG-based hospital payment (namely, prevent-
ing hospitals from adopting these innovations) is in line with societal objec-
tives. However, when technological innovations increase quality of care and
are associated with higher costs, DRG-based hospital payment becomes
problematic.
The problem is that hospitals are paid on the basis of cost information that
was collected in hospitals in the past. Consequently, when technological
innovations fi rst enter the market, the higher costs of those innovations are not
yet accounted for in current DRG weights. Only once hospitals have started
using these technological innovations, and when data relating to the costs of
using these innovations in routine practice have been collected, can DRG
Understanding DRGs and DRG-based hospital payment in Europe 33
systems and DRG weights be updated to account for the change in practice
patterns and costs. Therefore, the ability of DRG-based hospital payment
systems to respond to technological innovation is determined by (1) the
frequency of updates of DRG systems and of DRG weights, and (2) the time-lag
to data used for these updates (see Chapter 9).
Furthermore, most countries included in this book have developed additional
payment incentives to encourage the use of quality-increasing technological
innovations that also increase costs, within the time period during which the
DRG-based hospital payment system does not yet account for the innovation.
Because the available evidence relating to the effects of DRGs on technological
innovation is virtually non-existent, Chapter 9 is less focused on reviewing the
limited available literature than on providing an overview regarding how
European countries deal with technological innovation.
3.3 In summary: What do we want to understand?
This chapter provides a framework for understanding and comparing DRG
systems and DRG-based hospital payment systems in Europe. It introduces the
building blocks of DRG-based hospital payment systems and highlights their
likely effects on effi ciency, quality and technological innovation. The chapter
also outlines some alternative options that exist when designing DRG-based
hospital payment systems, indicating that the specifi c design features will
infl uence the effects of those systems.
Table 3.2 summarizes key questions that are raised in this chapter and that
are addressed within the chapters that follow (Part One of this book). The
concluding chapter of Part One (Chapter 10) draws on the fi ndings of chapters
4–9: (1) in order to make specifi c recommendations for policy-makers regarding
how best to design DRG-based hospital payment systems given country-specifi c
aims and objectives; and (2) to explore the potential for harmonization of DRG
systems and DRG-based hospital payment systems across Europe.
Table 3.2 Key questions to be answered by this book
Chapter Key questions
Chapter 4: DRG
systems and similar
patient classifi cation
systems in Europe
1. What are the advantages and disadvantages of importing
DRG systems?
2. How are diagnoses and procedures coded?
3. Which classifi cation variables can be used?
4. What should be the scope of included services?
5. How many groups are justifi ed?
Chapter 5: DRGs
and cost accounting:
Which is driving
which?
1. Why is cost accounting important?
2. How many hospitals should be included in the data sample?
3. What incentives exist for hospitals to calculate their costs?
4. What cost-accounting methods should be used?
5. Which cost categories should be included?
Continued overleaf
34 Diagnosis-Related Groups in Europe
3.4 References
Berki, S.E. (1985). DRGs, incentives, hospitals, and physicians. Health Affairs (Millwood),
4(4):70–6.
Brügger, U. (2010). Impact of DRGs: Introducing a DRG Reimbursement System. A Literature
Review. Zurich: SGGP (Schriftenreihe der SGGP, Vol. 98).
Busse, R., Schreyögg, J., Smith, P.C. (2008). Variability in healthcare treatment costs
amongst nine EU countries – results from the HealthBASKET project. Health Economics,
17(Suppl. 1):1–8.
Chalkley, M., Malcomson, J.M. (1998). Contracting for health services when patient
demand does not refl ect quality. Journal of Health Economics, 17(1):1–19.
Chapko, M.K., Liu, C., Perkins, M. et al. (2009). Equivalence of two healthcare costing
methods: bottom-up and top-down. Health Economics, 18(10):1188–201.
Davis, C., Rhodes, D.J. (1988). The impact of DRGs on the cost and quality of health care
in the United States. Health Policy, 9(2):117–31.
Donabedian, A. (1966). Evaluating the quality of medical care. The Milbank Memorial Fund
Quarterly, 44(3)2:166–203.
Chapter 6: DRG-
based hospital
payment: Intended
and unintended
consequences
1. How can hospitals be paid using DRGs?
2. What are the incentives of DRG-based hospital payment?
3. What determines the strength of these incentives?
4. How can unintended consequences be avoided?
5. How can DRG-based hospital payment be adjusted?
Chapter 7: DRG-
based hospital
payment and
effi ciency: Theory,
evidence and
challenges
1. Why should DRG-based hospital payment improve effi ciency
of hospitals?
2. How has the effect of DRG-based hospital payment on
effi ciency been measured?
3. Does DRG-based hospital payment improve effi ciency?
4. What challenges need to be overcome?
Chapter 8: DRGs and
quality: For better or
worse?
1. Why should DRGs and DRG-based hospital payment
infl uence the quality of hospital care?
2. How has the effect of DRG-based hospital payment on
quality been measured?
3. Does DRG-based hospital payment lead to better or worse
quality of care?
4. How can DRG-based hospital payment be modifi ed to
improve quality of care?
Chapter 9:
Technological
innovation in
DRG-based hospital
payment systems
across Europe
1. Why should DRG-based hospital payment infl uence the
adoption of technological innovation?
2. How do countries in Europe encourage technological
innovation?
3. How does technological innovation become incorporated
into DRG-based hospital payment?
4. How could innovation management be improved in DRG-
based hospital payment systems?
Table 3.2 Continued
Chapter Key questions
Understanding DRGs and DRG-based hospital payment in Europe 35
European Parliament and Council (2011). Directive 2011/24/EU on the Application of
Patients’ Rights in Cross-Border Healthcare. Brussels: Offi cial Journal of the European
Union (L88/45–L88/65).
Farrar, S., Yi, D., Sutton, M. et al. (2009). Has payment by results affected the way that
English hospitals provide care? Difference-in-differences analysis. British Medical
Journal, 339:b3047.
France, F.H.R. (2003). Casemix use in 25 countries: a migration success but international
comparisons failure. International Journal of Medical Informatics, 70(2–3):215–19.
Jacobs, R., Smith, P., Street, A. (2006). Measuring Effi ciency in Health Care. Cambridge:
Cambridge University Press.
Kimberly, J.R., de Pouvourville, G., D’Aunno, T., eds (2008). The Globalization of Managerial
Innovation in Health Care. Cambridge: Cambridge University Press.
Lave, J.R. (2003). Developing a Medicare prospective payment system for inpatient
psychiatric care. Health Affairs (Millwood), 22(5):97–109.
Legido-Quigley, H., McKee, M., Nolte, E., Glinos, I.A. (2008). Assuring the Quality of Health
Care in the European Union: A Case for Action. Copenhagen: WHO Regional Offi ce for
Europe on behalf of the European Observatory on Health Systems and Policies.
McCrone, P., Phelan, M. (1994). Diagnosis and length of psychiatric inpatient stay.
Psychological Medicine, 24(4):1025–30.
MedPAC (2003). Payment for new technologies in Medicare’s prospective payment
system, in MedPAC. Report to the Congress: Medicare Payment Policy. Washington, DC:
Medicare Payment Advisory Commission.
OTA (1983). Diagnosis-Related Groups (DRGs) and the Medicare Program: Implications for
Medical Technology – A Technical Memorandum. Washington, DC: Offi ce of Technology
Assessment.
Quentin, W., Scheller-Kreinsen, D., Geissler, A., Busse, R. (in press). Appendectomy and
diagnosis-related groups (DRGs): patient classifi cation and hospital reimbursement
in 11 European countries. Langenbeck’s Archives of Surgery (in press).
Rogers, W.H., Draper, D., Kahn, K.L. et al. (1990). Quality of care before and after
implementation of the DRG-based prospective payment system. A summary of
effects. Journal of the American Medical Association, 264(15):1989–94.
Schreyögg, J., Stargardt, T., Tiemann, O., Busse, R. (2006). Methods to determine
reimbursement rates for diagnosis related groups (DRG): a comparison of nine
European countries. Health Care Management Science, 9(3):215–23.
Shih, C., Berliner, E. (2008). Diffusion of new technology and payment policies: coronary
stents. Health Affairs (Millwood), 27(6):1566–76.
Shleifer, A (1985). A theory of yard stick competition. The RAND Journal of Economics,
16(3):319–27.
Smith, P.C., Mossialos, E., Papanicolas, I., Leatherman, S. (2010). Conclusions, in P.C.
Smith, ed. Performance Measurement for Health System Improvement. Cambridge:
Cambridge University Press.
chapter f o u r
DRG systems and similar
patient classifi cation
systems in Europe
Conrad Kobel, Josselin Thuilliez,
Martine Bellanger and Karl-Peter Pfeiffer
4.1 Introduction
The idea of any patient classifi cation system(s) (PCSs) is to combine the confus-
ingly large number of different patients, all appearing to be unique, into a lim-
ited number of groups with roughly similar features. Diagnosis-related group
(DRG) systems are PCSs that have four main characteristics: (1) routinely col-
lected data on patient discharge are used to classify patients into (2) a manage-
able number of groups that are (3) clinically meaningful and (4) economically
homogeneous. In addition, all DRG systems are at least remotely related to the
original DRG system that was developed by a group of researchers including
Robert Fetter at Yale University during the 1970s (Fetter et al., 1980; Fetter,
1999).
Today, DRG systems are the most widely employed PCS in Europe. They are
used in eight countries (Estonia, Finland, France, Germany, Ireland, Portugal,
Spain and Sweden) out of the 12 countries covered in this book. Only Austria,
England, the Netherlands and Poland have introduced PCSs that do not
originate from the original United States Health Care Financing Administra-
tion (HCFA-)DRG system (Fischer, 2008). However, most of the self-developed
systems are similar to DRG systems in that they share the basic characteristics.
Only the Dutch PCS differs to a great extent from the DRG approach (see
Chapter 23 of this volume). All PCSs of countries included in this book are
referred to as ‘DRG-like patient classifi cation systems’.
Yet, in spite of many similarities in the basic characteristics of different DRG-
like PCSs, each country’s system is unique, and thus defi nes patient groups or
hospital products in a different way. On the one hand, it is very likely that this
38 Diagnosis-Related Groups in Europe
ability to adapt DRG systems to country-specifi c needs was one of the reasons
for their success and their widespread application in European countries. On
the other hand, in a context of increasing patient mobility and growing interest
in cross-border comparisons of hospital performance, the lack of a common
defi nition of hospital products is starting to become problematic (European
Parliament and Council, 2011). Therefore, this chapter intends to provide a
systematic overview of the similarities and differences between DRG-like PCSs
in Europe.
The chapter is organized as follows: the next section (4.2) fi rst describes the
historical origins of DRG-like PCSs in the countries included in this book.
Section 4.3 provides an overview of some of the main characteristics of these
systems and compares major diagnostic categories (MDCs) or similar categories
that play an important role in most systems across the countries concerned.
Section 4.4 presents the coding systems for diagnoses and procedures that form
the basis of all PCSs. Subsequently, section 4.5 describes the classifi cation
algorithms of the systems, before section 4.6 looks in more detail at the specifi c
classifi cation variables used. Section 4.7 describes current trends in European
DRG-like PCSs and last, but not least, the fi nal section (4.8) concludes the
chapter with a discussion of the opportunities and requirements for the
harmonization of DRG-like PCSs in Europe.
4.2 Historical origins of DRG-like PCSs in Europe
Figure 4.1 illustrates the historical origins of DRG-like PCSs used in the European
countries included in this book. It shows that all currently existing DRG systems
are at least remotely related to the original HCFA-DRGs, while this is not true
for the other ‘DRG-like’ PCSs (shown at the far right of Figure 4.1) (Fischer,
2008). The fi rst DRG system, Yale DRG, developed at Yale University and
introduced in the late 1970s was initially intended as a tool to measure hospital
resource utilization. However, recognizing the potential of a system that enabled
assessment of hospital production, the United States’ HCFA adapted the system
for the purpose of monitoring and reimbursing hospital care delivered to elderly
patients insured under Medicare (the federal tax-funded old-age insurance in
the United States) (Fischer, 1997; Chilingerian, 2008).
In 1986, France modifi ed the HCFA-DRG system and developed its own
national DRG system called groupes homogènes des malades (GHMs) (ATIH, 2010),
translated as ‘homogeneous groups of patients’. Later, in 1988, 3M™ Health
Information Systems adapted and extended HCFA-DRGs in order to better
refl ect the pathologies of non-elderly populations (3M, 2005). The resulting All
Patients (AP-)DRG system was widely applied in the United States and, subse-
quently, updated versions of AP-DRGs were adopted in various European coun-
tries, such as Spain and Portugal, as well as infl uencing the development of
national DRG systems, such as those of France and Australia. AP-DRGs were
later refi ned by changing the determination of severity levels in order to
respond to demands for more accurate assessment of case severity and differ-
ences in resource intensity, thus leading to the All Patient Refi ned (APR-)
DRGs (3M, 2003). Together, AP-DRGs and APR-DRGs formed the basis for the
F
ig
u
re
4
.1
H
is
to
ri
ca
l
d
ev
el
o
p
m
en
t
So
u
rc
es
:
A
u
th
o
rs
’
o
w
n
c
o
m
p
il
at
io
n
,
b
as
ed
o
n
F
is
ch
er
,
2
0
0
8
;
Sc
h
re
y
ö
g
g
e
t
al
.,
2
0
0
6
;
an
d
i
n
fo
rm
at
io
n
p
ro
v
id
ed
i
n
A
T
IH
,
2
0
1
0
;
3
M
,
2
0
0
5
;
A
u
st
ra
li
an
G
o
v
er
n
m
en
t,
2
0
0
4
;
In
E
K
,
2
0
0
9
;
N
H
S
In
fo
rm
at
io
n
C
en
tr
e
fo
r
H
ea
lt
h
a
n
d
S
o
ci
al
C
ar
e,
2
0
1
0
;
an
d
t
h
e
re
le
v
an
t
co
u
n
tr
y
-s
p
ec
ifi
c
c
h
ap
te
rs
i
n
P
ar
t
T
w
o
o
f
th
is
v
o
lu
m
e.
*
D
is
co
n
ti
n
u
ed
d
ev
el
o
p
m
en
ts
a
re
n
o
t
sh
o
w
n
,
su
ch
a
s
th
e
u
se
o
f
H
C
FA
-D
R
G
s
in
I
re
la
n
d
a
n
d
P
o
rt
u
g
al
o
r
th
e
fl
at
-r
at
e/
ad
d
it
io
n
al
p
ay
m
en
t
(F
a
ll
pa
u
sc
h
a
le
/
So
n
d
er
en
tg
el
t)
a
p
p
ro
ac
h
i
n
G
er
m
an
y.
40 Diagnosis-Related Groups in Europe
Australian National (AN-)DRG system, which was renamed to Australian
Refi ned (AR-)DRGs after further modifi cations had been introduced into the
system (Australian Government, 2004). In 2003, Ireland adopted AR-DRGs (see
Chapter 15 of this volume), while Germany used AR-DRGs as the basis for
developing its own German (G-)DRG system (see Chapter 14).
The Nordic countries are special in that they started to collaborate in 1996 in
order to develop a common Nordic DRG system, called NordDRG – a PCS based
on HCFA-DRGs. NordDRGs are jointly updated and then imported by each
country before country-specifi c modifi cations are added to each new version of
NordDRGs (see Chapter 16 of this volume). Of the countries covered in this
book, Sweden and Finland are using NordDRGs. In addition, Estonia adopted
NordDRGs in 2003 and has continued to use the same version of the system
until the fi rst update in 2010 (see Chapter 17). Unless otherwise explained, the
term ‘NordDRG’ refers to the common Full NordDRG system that is jointly
developed among the Nordic countries.
England, Austria and the Netherlands decided to develop their own PCSs. In
1992, the English Healthcare Resource Group (HRG) system was developed, and
was later adopted by Poland, with a number of modifi cations. This led to the
emergence of the Jednorodne Grupy Pacjentów (JGP), which can be translated
(like the French GHMs) as ‘homogeneous groups of patients’. In Austria a
national self-developed PCS, described as a performance-oriented hospital
fi nancing system (Leistungsorientierte Krankenanstaltenfi nanzierung; LKF) has
been used since 1997 (see Chapter 11 of this volume). The Netherlands devel-
oped its own – very special – system of diagnosis–treatment combinations
(Diagnose Behandeling Combinaties; DBCs), which has been in use since 2005
(see Chapter 23).
4.3 DRG-like PCSs in Europe: Overview
As illustrated by the historical origins of DRG-like PCSs in Europe, current PCSs
are either self-developed or have their (remote) origins in various successors of
the original Yale DRG system. Table 4.1 describes some basic characteristics of
nine DRG-like PCSs. First, the systems differ in the number of groups they
defi ne: most systems contain between 650 and 2300 groups. The Polish JGP
system defi nes fewer groups than all other systems (only 518), while the Dutch
DBC system is an extreme outlier, comprising about 30 000 DBCs in the 2010
version.
In all HCFA-derived DRG systems, DRGs are organized within MDCs. Even
the DRG-like PCSs – HRGs and JGPs – categorize their groups into ‘chapters’;
only in LKF and DBC is this technique of subdivision not used. The chapters/
MDCs cover certain parts of the body or certain disease entities and are similar
across all systems. While the total number of DRGs differs greatly across PCSs,
the number of chapters/MDCs is around 25 for all systems, except the JGP
system, which eliminated a number of chapters when adopting the English
HRGs. Since in most systems, each MDC/chapter represents one organ
system, the MDC/chapter structure of PCSs parallels the structure of medical
specialties.
DRG systems and similar patient classifi cation systems in Europe 41
Furthermore, all DRG-like PCSs except the DBC system defi ne ‘partitions’ to
further divide cases into more homogeneous groups. These partitions are
defi ned by the kind of treatment, namely ‘surgical’ (or ‘operating room’ (OR))
versus ‘medical’ treatment. In addition, in some systems, partitions distinguish
between OR procedures and non-OR procedures. Only the French GHM
contains a fourth partition in certain MDCs, whereby the classifi cation process
does not check for the type of procedure (ATIH, 2010).
Figure 4.2 presents a graphical illustration of the distribution of DRG-like
groups into MDCs (or chapters). On the left-hand side of the fi gure is a list of
the MDCs as currently used in Medicare Severity (MS-)DRGs (the successor to
HCFA-DRGs), which served as the reference for this comparison. Since MDCs
are not used in the LKF system, LKF groups were mapped to MS-DRGs on the
basis of the LKF group names. The Dutch DBC system was excluded from this
comparison, since no mapping seemed feasible. Each cell represents one MDC
in a PCS. The letters within the cells are the codes that are used in the different
PCSs as names for each category. The ordering of the codes demonstrates that
in all countries almost exactly the same categories are used to form MDCs, and
that they follow in almost exactly the same order. Even the self-developed HRG
system uses similar categories in a similar order. However, some MDCs are only
used by a specifi c PCS. This is the case for ‘Vascular disease’ (JGPs), ‘Breast
problems’ (NordDRGs) and ‘HIV infection’ (AP-DRGs, G-DRGs, GHMs). These
are highlighted in Figure 4.2.
Figure 4.2 can be interpreted thus: the wider a column is, the higher the total
number of groups of this DRG-like PCS in comparison to the others. The higher
a cell is, the higher the share of groups in this system’s MDC. For example, the
column representing the GHM system is more than four times wider than the
column representing the JGP system. Comparing the height of the cells shows
that the distribution of DRGs into MDCs/chapters is similar across all DRG-like
PCSs. This illustrates that all systems need similar shares of their total groups to
describe cases within a specifi c category of diseases. However, some minor
differences exist: for example, the MDC ‘Circulatory system’ represents around
10 per cent of the total number of groups in most PCSs, but only 4.5 per cent of
all groups in the HRG system. Furthermore, the category ‘Pre-MDC’ is defi ned
either explicitly or only implicitly (for example, as ‘Organ transplants’ in the
GHM system). However, as this analysis does not assess the specifi c groups
Table 4.1 Basic characteristics of DRG-like PCSs in Europe (based on 2008)
AP- AR- G- GHM Nord- HRG JGP LKF DBC
DRG DRG DRG DRG
Groups 679 665 1 200 2 297 794 1 389 518 979 ≈30 000
MDCs/Chapters 25 24 26 28 28 23 16 – –
Partitions 2 3 3 4 2 2* 2* 2* –
Source: Authors’ own compilation based on data provided by the Nordic Casemix Centre
(2011), as well as information contained in the relevant chapters of Part Two of this volume.
* HRG, JGP, and LKF do not defi ne partitions per se, but distinguish between treatment- and
diagnosis-driven episodes.
F
ig
u
re
4
.2
C
o
m
p
ar
is
o
n
o
f
M
D
C
s
ac
ro
ss
e
ig
h
t
D
R
G
s
y
st
em
s
an
d
s
im
il
ar
P
C
Ss
So
u
rc
es
:
A
u
th
o
rs
’
o
w
n
c
o
m
p
il
at
io
n
b
as
ed
o
n
i
n
fo
rm
at
io
n
a
v
ai
la
b
le
i
n
B
M
G
,
2
0
0
9
;
N
H
S
In
fo
rm
at
io
n
C
en
tr
e
fo
r
H
ea
lt
h
a
n
d
S
o
ci
al
C
ar
e,
2
0
1
0
;
A
T
IH
,
2
0
1
0
;
In
E
K
,
2
0
0
9
;
A
u
st
ra
li
an
G
o
v
er
n
m
en
t,
2
0
0
4
;
3
M
,
2
0
0
5
;
d
at
a
p
ro
v
id
ed
b
y
t
h
e
N
o
rd
ic
C
as
em
ix
C
en
tr
e
(2
0
1
1
);
a
n
d
i
n
fo
rm
at
io
n
co
n
ta
in
ed
i
n
C
h
ap
te
r
2
0
(
P
o
la
n
d
)
o
f
th
is
v
o
lu
m
e.
DRG systems and similar patient classifi cation systems in Europe 43
included within MDCs/chapters in different PCSs, it cannot be ignored that
differences in the distribution of groups might be either greater or smaller than
they appear.
4.4 Data requirements: Coding of diagnoses and procedures
In all DRG-like PCSs, the coding of diagnoses and procedures is important, since
this information forms the basis for the defi nition of patient groups. For coding
of diagnoses, an international standard exists: most countries use the 10th
revision of the WHO’s International Classifi cation of Diseases (ICD-10). Only in
Spain and Portugal is the previous version of the ICD system (ICD-9) still in use
because the AP-DRG system requires ICD-9 codes. However, even within the
group of countries using the ICD-10 version, signifi cant differences exist, since
almost all countries are using ICD-10 codes with country-specifi c modifi cations
(see Table 4.2). Country-specifi c modifi cations usually add a fi fth digit to the
general structure of ICD-10 codes, which allows for more detailed specifi cation
of certain conditions. However, sometimes country-specifi c modifi cations even
deviate from the ICD-10 logic for specifi c conditions. For example, the German
Modifi cation ICD-10-GM does not contain the O84 code for multiple deliveries.
Instead, Z37 codes are used, which specify the outcome of delivery (for example
single birth, multiple births). Furthermore, each country has its own coding
standards and guidelines.
Table 4.2 Coding of diagnoses and procedures
Country Diagnoses coding Procedure coding
Austria ICD-10-BMSG-2001 Leistungskatalog
England ICD-10 OPCS
Estonia ICD-10 NCSP
Finland ICD-10-FI NCSP-FI
France CIM-10 CCAM
Classifi cation Commune des Actes Médicaux
Germany ICD-10-GM OPS
Operationen- und Prozedurenschlüssel
Ireland ICD-10-AM ACHI
Australian Classifi cation of Health Interventions
The Netherlands ICD-10 Elektronische DBC Typeringslijst
Poland ICD-10 ICD-9-CM
Portugal ICD-9-CM ICD-9-CM
Spain ICD-9-CM ICD-9-CM
Sweden ICD-10-SE KVÅ
Klassifi kation av vårdåtgärder (Swedish
adaption of NCSP)
NordDRG ICD-10 NCSP
Nomesco Classifi cation of Surgical Procedures
Sources: Authors’ own compilation based on data provided by the Nordic Casemix Centre
(2011), as well as information contained in the relevant country-specifi c chapters of Part Two
of this volume.
44 Diagnosis-Related Groups in Europe
For procedure coding, the differences between countries are even greater,
since no similar international standard exists. Almost every country has
developed its own procedure coding system tailored to its needs. Consequently,
these systems are very heterogeneous. They range from sequential numbered
lists, such as the Australian Classifi cation of Health Interventions (ACHI) to
multi-axial procedure classifi cations, such as the French classifi cation of pro-
cedures (classifi cation commune des actes médicaux, CCAM), or the Austrian
Leistungskatalog. In addition, granularity differs to a great extent. The LKF sys-
tem includes only selected procedures and therefore contains only 1500 items.
At the other end of the scale, the German procedure classifi cation codes
(Operationen- und Prozedurenschlüssel, OPS) – designed to include all procedures
– contain more than 30 000 items; 20 times more than the Austrian system.
4.5 The classifi cation algorithm in European DRG-like PCSs
DRG-like PCSs group patients into a manageable number of groups. In order to
do so, they follow a certain classifi cation algorithm. This is similar across all the
DRG systems that are based on different modifi cations of the original HCFA-
DRGs. In particular, diagnoses are the predominating classifi cation criteria. The
classifi cation algorithm in other DRG-like PCSs (for example in England, Poland
and Austria) differs in that procedures become more important at an earlier
stage and diagnoses only play a subordinate role (NHS Information Centre for
Health and Social Care, 2010; BMG, 2009). In the Netherlands, the medical
specialty department forms the fi rst step in the grouping process (see Chapter
23 of this volume).
The following subsections contain descriptions of classifi cation algorithms in
PCSs derived from the HCFA-DRG system and other DRG-like PCSs, and they
describe both similarities and differences within and between these groups of
classifi cation systems.
4.5.1 PCSs derived from HCFA-DRGs
Figure 4.3 shows the general grouping algorithm of PCSs derived from HCFA-
DRGs and DRG system-specifi c modifi cations of the basic algorithm. The Nord-
DRG system is not mentioned explicitly in the diagram because its developments
do not change the general grouping algorithm.
There are six major steps common to all systems. Before the actual classifi ca-
tion starts, the data are (1) checked to exclude cases with incorrect or missing
information. Then, (2) very high-cost cases (for example, cases with transplan-
tations) are isolated from all other cases into a special category of groups called
‘Pre-MDCs’. Subsequently, (3) cases are allocated to mutually exclusive MDCs
based on the principal diagnosis (although some systems sporadically use other
variables, such as age, to assign cases to a neonatal MDC).
In the next step, (4) the grouping algorithm checks whether or not an OR
procedure was performed and separates patients into a ‘surgical’ or into a ‘medi-
cal’ partition. In addition, the AR-DRG, the (derived) G-DRG, and the GHM
F
ig
u
re
4
.3
C
la
ss
ifi
c
at
io
n
a
lg
o
ri
th
m
o
f
d
if
fe
re
n
t
H
C
FA
-D
R
G
d
er
iv
ed
P
C
Ss
,
in
cl
u
d
in
g
s
y
st
em
-s
p
ec
ifi
c
m
o
d
ifi
c
at
io
n
s
So
u
rc
es
:
A
u
th
o
rs
’
o
w
n
c
o
m
p
il
at
io
n
b
as
ed
o
n
i
n
fo
rm
at
io
n
p
ro
v
id
ed
a
v
ai
la
b
le
i
n
A
T
IH
,
2
0
1
0
;
In
E
K
,
2
0
0
9
;
A
u
st
ra
li
an
G
o
v
er
n
m
en
t,
2
0
0
4
;
3
M
,
2
0
0
5
;
an
d
in
fo
rm
at
io
n
c
o
n
ta
in
ed
i
n
c
h
ap
te
rs
1
3
(
Fr
an
ce
),
1
4
(
G
er
m
an
y
)
an
d
1
5
(
Ir
el
an
d
)
o
f
th
is
v
o
lu
m
e.
46 Diagnosis-Related Groups in Europe
systems differentiate between cases with relevant non-OR procedures (that is,
relevant within a specifi c MDC), which are then assigned to the ‘other’/‘non-
OR’ partition. Consequently, the medical partition in NordDRG countries may
contain cases which are found in other countries’ systems within the ‘other’/
’non-OR’ partition; the actual name varies according to the system. A particu-
larity of the GHM system is that an undifferentiated partition exists within
certain MDCs (see Chapter 13 of this volume).
After assignment of the partition, (5) all DRG systems check for further charac-
teristics of the case (complexity of the principal and sometimes secondary
diagnoses, type of procedures, combinations of both, and sometimes age,
length of stay or treatment setting) in order to assign it to a class (in the AP-DRG
system) or to a ‘base-DRG’ (in other systems). The algorithm usually checks fi rst
for more complicated procedures or conditions in order to make sure that
patients are classifi ed into the base-DRG/class that best refl ects resource
consumption of the case (illustrated by the arrow between base-DRGs/classes in
Figure 4.3).
A particularity of the AP-DRG system is that a list of secondary diagnoses is
checked in order to identify cases with major complications and co-morbidities
(major CCs), which are then collected in a specifi c major-CC class (3M, 2005).
This is different from other DRG systems, where CCs are usually only considered
in the last step of the grouping algorithm (although exceptions to this rule
exist, for example in the G-DRG system). Furthermore, the AP-DRG system has
explicit classes for symptoms and ‘other’ conditions that do not exist in other
DRG systems. Yet, the AP-DRG is similar to the AR-DRG and G-DRG systems, in
terms of their approach to identifying cases with surgery unrelated to the MDC.
For example, cases with hip surgery within the nervous system MDC are
classifi ed into the unrelated surgery class/base-DRG, which will determine an
Error DRG in the fi nal AR-DRG and G-DRG assignment process.
In the last step of the classifi cation algorithm, (6) each case is grouped into
the fi nal DRG. Often, the class/base-DRG is split into several DRGs (the arrows
between the DRGs in Figure 4.3 indicate that there may be more than two) in
order to refl ect different levels of resource consumption. Other classes/base-
DRGs are not split if the group of patients within the base-DRG is relatively
homogeneous. In these cases, the fi nal DRG is identical to the base-DRG/class.
The assignment to the fi nal DRG is based on classifi cation variables, which
differ across systems. Most systems consider secondary diagnoses, procedures,
age, and type of discharge (including, for example, death) in order to assign the
fi nal DRG. The section that follows (4.5.2) explores these variables in more
detail.
4.5.2 Self-developed DRG-like PCSs in
England, Poland and Austria
Figure 4.4 illustrates the basic structure of the classifi cation algorithm for the
self-developed DRG-like HRG, JGP and LKF systems. Since JGPs were derived
from an earlier HRG version, it is not surprising that a number of similarities
exist between these two systems (see Chapter 20 of this volume). The grouping
DRG systems and similar patient classifi cation systems in Europe 47
algorithm of all three systems consists of between fi ve and seven consecutive
steps, similar to those shown in Figure 4.4. However, the steps do not necessarily
coincide, and the most important difference in comparison to the PCSs derived
from HCFA-DRGs is that procedures play the dominant role in the grouping
algorithm, while diagnoses are less important.
In the fi rst step, all three systems identify whether the patients in each case
received certain well-defi ned specialized services, for example computerized
tomography (CT) scans, intensive care unit (ICU) treatment or chemotherapy.
If patients received specifi c procedures (in the HRG and JGP systems), or if they
Figure 4.4 Classifi cation algorithm in self-developed DRG-like PCSs (HRG, JGP and
LKF)
Sources: Authors’ own compilation based on information available in BMG, 2009; NHS
Information Centre for Health and Social Care, 2010; and information contained in chapters
12 (England) and 20 (Poland) of this volume.
48 Diagnosis-Related Groups in Europe
were treated in specialist departments (for example, geriatrics in the LKF system)
the PCSs classify patients into certain add-on groups that are assigned in
addition to the fi nal groups. The idea is to separate services that are provided to
heterogeneous groups of patients (but not necessarily to all patients within
these groups) from all other services, in order to increase the ability to defi ne
homogeneous groups of patients. In PCSs derived from HCFA-DRGs, similar
mechanisms exist to identify certain well-defi ned specialized services and to
reimburse them separately (see section 4.8 and Chapter 6), but these are not
always directly integrated within the grouping algorithm.
The second step, which is similar to Pre-MDC assignment, exists only in the
HRG system: cases with procedures that indicate trauma of more than two sites
of the body are separated as multiple trauma cases into a type of Pre-MDC cat-
egory and are assigned to HRGs. In the next step, all systems separate cases with
signifi cant procedures into a procedure-driven partition, while cases with no
signifi cant procedures are assigned to a diagnosis-driven partition. Subse-
quently, the HRG and JGP systems determine the most important (dominant)
procedure, either using a rank list of procedures (in the HRG system) or accord-
ing to the decision of the provider, who can manually select the dominant
procedure (in the JGP system). In both systems, this is followed by the assign-
ment of cases to chapters and sub-chapters, which represent medical specialties
similar to those of MDCs in systems derived from HCFA-DRGs.
In the penultimate step of the grouping algorithm, the LKF system differs
again from the HRG and JGP systems. Within the procedure-driven partition in
the HRG and JGP systems, the highest ranked procedure determines the ‘base-
group’/root to which each case is assigned. For major procedures, which are
identifi ed through a procedure rank above a certain threshold, base-groups/
roots are determined directly. In contrast, for cases with procedures of a rank
below the threshold, the principal diagnosis is also checked. In the Austrian
LKF system, no explicit ranking of procedures takes place. Instead, for all proce-
dures, the score of the corresponding group is calculated. The one with the
highest score is then selected. In the diagnosis-driven partition, the base-group
is always determined by the principal diagnosis.
In the fi nal step of the grouping algorithm, which is similar to that of PCSs
derived from the HCFA-DRG system, patients are classifi ed into the fi nal group.
Base-groups are either split into several fi nal groups, in order to differentiate
between different levels of resource consumption, or they remain unsplit. In
the HRG and JGP systems, it depends on the chapter as to whether specifi c CCs
are considered to be relevant in the grouping algorithm or not. In the LKF
system, age is used most often to separate groups.
4.5.3 The Dutch DBC classifi cation
The DBC classifi cation system is very different from all the other systems. In
most cases it consists of four dimensions: (1) the fi rst dimension specifi es one
of 27 medical specialties, under which the patient was treated. Then (2) one of
fi ve types of care is determined (for example, regular inpatient care or ICU treat-
ment). Subsequently (3) the diagnosis of the patient is considered, before
DRG systems and similar patient classifi cation systems in Europe 49
fi nally (4) the treatment dimension differentiates between the treatment set-
ting (inpatient versus outpatient) and the type of care (conservative treatment,
type of surgery). For certain medical specialties, a fi fth dimension exists, which
identifi es whether certain conditions existed that are expected to result in
higher-than-average resource consumption (such as age < 11 years or require-
ment for a second surgeon). Any specifi cation can be chosen for each dimen-
sion, resulting in a very high number of groups (Warners, 2008; see also Chapter
23 of this volume).
4.6 Classifi cation variables and severity levels
in European DRG-like PCSs
This section provides more details regarding the variables considered in the
classifi cation process, with an emphasis on the fi nal split into DRGs or DRG-like
groups. In addition, differences in the number of severity/complexity levels per
base-group are explored and the approaches to using CCs are explained.
4.6.1 Classifi cation variables
All DRG-like PCSs in Europe use routinely collected patient discharge data in
order to classify patients. Table 4.3 provides an overview of clinical, demo-
graphic/administrative and resource-consumption variables used in European
DRG-like PCSs. Clinical information (relating to diagnoses and procedures) is
used as classifi cation variables in all systems. In addition, all PCSs except the
Netherlands’ DBC system use the concept of one principal diagnosis as the
highest ranked diagnosis for hospital discharge. However, the defi nition of
what constitutes the principal diagnosis differs. In some countries the principal
diagnosis is defi ned as the ‘main reason’ for a hospital stay (in, for example, the
AR-DRG, G-DRG and LKF systems). In other countries, where the hospital dis-
charge is aggregated based on several departmental discharges (the GHM or
HRG systems, for example), a diagnosis hierarchy is used to determine the most
important diagnosis. Procedures are also used, in all systems, but their impor-
tance in the classifi cation algorithm varies – even between similar systems. For
example, procedures play a more prominent role in the classifi cation algorithm
in the G-DRG system than in the AR-DRG system, on which the German sys-
tem was originally based (InEK, 2009). In the self-developed HRG, JGP and LKF
systems, information about procedures actually dominates information about
diagnoses (see section 4.5.2).
Demographic and administrative variables, especially age and discharge type
(for example, death or transfer) are frequently used variables in all systems, except
the DBC system. Gender is a relevant classifi cation variable only in the NordDRG
system, although many systems use it to verify consistency of data (for example,
where obstetric diagnosis codes are accepted only for female patients).
Similarly, resource-consumption variables are used in many DRG-like PCSs.
Length of stay is the most frequently used explicit resource-consumption vari-
able. However, even if systems do not explicitly include resource-consumption
T
a
b
le
4
.3
C
la
ss
ifi
c
at
io
n
v
ar
ia
b
le
s
an
d
s
ev
er
it
y
l
ev
el
s
in
E
u
ro
p
ea
n
D
R
G
-l
ik
e
P
C
Ss
Sy
st
em
s
A
P
-D
R
G
A
R
-D
R
G
G
-D
R
G
G
H
M
N
or
d
D
R
G
H
R
G
JG
P
L
K
F
D
B
C
C
la
ss
ifi
c
at
io
n
v
ar
ia
b
le
s
C
li
n
ic
a
l
va
ri
a
b
le
s
D
ia
g
n
o
se
s
×
×
×
×
×
×
×
×
×
P
ro
ce
d
u
re
s
×
×
×
×
×
×
×
×
×
N
eo
p
la
sm
s/
M
al
ig
n
an
cy
×
×
×
–
–
–
–
–
–
T
y
p
e
o
f
ca
re
–
–
–
–
–
–
–
–
×
A
d
m
in
is
tr
a
ti
ve
/d
em
og
ra
ph
ic
v
a
ri
a
b
le
s
A
d
m
is
si
o
n
t
y
p
e
–
–
–
–
×
×
–
–
A
g
e
×
×
×
×
×
×
×
×
–
B
ir
th
w
ei
g
h
t
(n
ew
b
o
rn
)
×
×
×
×
–
–
–
–
D
is
ch
ar
g
e
ty
p
e
×
×
×
×
×
×
×
–
–
G
en
d
er
–
–
–
–
×
–
–
–
–
M
en
ta
l
h
ea
lt
h
l
eg
al
s
ta
tu
s
–
×
×
–
–
–
–
–
–
R
es
ou
rc
e
co
n
su
m
pt
io
n
v
a
ri
a
b
le
s
L
O
S/
Sa
m
e-
d
ay
s
ta
tu
s
–
×
×
×
×
×
×
–
–
M
ec
h
an
ic
al
v
en
ti
la
ti
o
n
–
–
×
–
–
–
–
–
–
Se
tt
in
g
–
–
–
×
–
–
–
–
×
St
ay
a
t
sp
ec
ia
li
st
d
ep
ar
tm
en
ts
–
–
–
–
–
–
–
×
–
M
ed
ic
al
s
p
ec
ia
lt
y
–
–
–
–
–
–
–
–
×
D
em
an
d
s
fo
r
ca
re
–
–
–
–
–
–
–
–
×
Se
v
er
it
y
/c
o
m
p
le
x
it
y
l
ev
el
s
3
*
4
n
o
t
li
m
it
ed
5
**
2
3
3
n
o
t
li
m
it
ed
–
A
g
g
re
g
at
e
ca
se
c
o
m
p
le
x
it
y
m
ea
su
re
–
P
C
C
P
C
C
×
–
–
–
–
–
So
u
rc
es
:
A
u
th
o
rs
’
o
w
n
c
o
m
p
il
at
io
n
b
as
ed
o
n
A
T
IH
,
2
0
1
0
;
In
E
K
,
2
0
0
9
;
A
u
st
ra
li
an
G
o
v
er
n
m
en
t,
2
0
0
4
;
3
M
,
2
0
0
5
;
B
M
G
,
2
0
0
9
;
N
H
S
In
fo
rm
at
io
n
C
en
tr
e
fo
r
H
ea
lt
h
a
n
d
S
o
ci
al
C
ar
e,
2
0
1
0
;
W
ar
n
er
s,
2
0
0
8
;
d
at
a
p
ro
v
id
ed
b
y
t
h
e
N
o
rd
ic
C
as
em
ix
C
en
tr
e
(2
0
1
1
),
a
s
w
el
l
as
i
n
fo
rm
at
io
n
c
o
n
ta
in
ed
i
n
t
h
e
re
le
v
an
t
co
u
n
tr
y
-
sp
ec
ifi
c
c
h
ap
te
rs
o
f
P
ar
t
T
w
o
o
f
th
is
v
o
lu
m
e.
*N
o
t
ex
p
li
ci
tl
y
m
en
ti
o
n
ed
(
m
aj
o
r
C
C
s
at
M
D
C
l
ev
el
p
lu
s
2
l
ev
el
s
o
f
se
v
er
it
y
a
t
D
R
G
l
ev
el
)
**
4
l
ev
el
s
o
f
se
v
er
it
y
p
lu
s
o
n
e
G
H
M
f
o
r
sh
o
rt
s
ta
y
s
o
r
o
u
tp
at
ie
n
t
ca
re
DRG systems and similar patient classifi cation systems in Europe 51
variables, such as mechanical ventilation, these variables are regularly consid-
ered in the classifi cation algorithms by other means. For example, while the
G-DRG system explicitly considers duration of mechanical ventilation, other
systems use procedure codes for tracheostomy in order to identify cases with
mechanical ventilation.
4.6.2 Severity levels
Table 4.3 also shows the number of severity levels in different DRG-like PCSs.
Most countries limit the number of possible severity levels. For example, the
number of severity levels is restricted to only two in NordDRG systems and to
three in the HRG system. The same logic of splitting base-groups only when
necessary is also used in other systems (AR-DRG and HRG systems). However,
in GHM, if a base-group is split, it is almost always split into four levels, plus
one additional group for short stays or day cases. At the other end of the scale,
the G-DRG and LKF systems do not limit the number of severity levels. They
subdivide base-groups into as many fi nal groups as necessary in order to achieve
relative homogeneity of resource consumption within each group. The DBC
system is the only system that does not split base-groups during the fi nal step
of the grouping algorithm.
4.6.3 Dealing with CCs
In all systems, except for the DBC and LKF PCSs, secondary diagnoses determine
to a large degree the classifi cation of cases into the appropriate level of severity
or complexity. In most DRG-like PCSs, lists of secondary diagnoses are defi ned
that represent CCs. The same CC list usually applies to all cases, except in the
HRG system, which has one specifi c CC list for each chapter. However, even
systems with global CC lists always defi ne certain exclusion criteria – mostly
usually principal diagnosis, for which specifi c secondary diagnoses are not
considered a CC. Depending on the number of severity/complexity levels of the
PCS, CC lists specify different levels of severity for each CC.
Furthermore, a number of approaches to dealing with multiple secondary
diagnoses exist. While in the AR-DRG and G-DRG systems a cumulative measure
(called Patient Clinical Complexity Level (PCCL)) of all secondary diagnoses is
applied, in most other DRG systems it is the highest ranked secondary diagnosis
that defi nes the severity. In the GHM system, another cumulative approach is
used: the highest ranked secondary diagnosis together with age, length of stay
and death during admission defi ne the severity for a number of DRGs. In the
Netherlands’ DBC system, secondary diagnoses are not taken into account.
Instead, a new DBC is allocated if patients are treated for additional diagnoses.
4.7 Trends
When analysing the developments of DRG-like PCSs over time, three main
developments come to light: (1) DRG-like PCSs are progressively being applied
52 Diagnosis-Related Groups in Europe
to settings that are beyond the acute care hospital inpatient sector for which
they were originally developed; (2) the number of groups has continued to
increase in all systems; and (3) systems increasingly develop measures to ensure
that specifi c complicated, high-cost services are adequately refl ected.
4.7.1 Coverage of services
Since the early 1990s, researchers have tried to expand the concept of
DRGs into settings other than inpatient acute hospital care (Goldfi eld, 2010.)
Table 4.4 shows that the majority of countries are also using DRG-like PCSs
for day care – or they are planning to do so. In order to use DRG-like PCSs
for day care, countries have either extended their PCS (for example Finland,
France and Sweden) or assigned different weights for DRGs in different
settings.
Countries using the same PCS for inpatients and day cases should introduce
additional algorithms into their classifi cation systems in order to identify day
cases. For example, the French GHM system splits base-DRGs according to the
length of stay (LOS) in order to identify day cases as cases with a LOS = 0 (ATIH,
2010). In the Swedish and Finnish versions of the NordDRG system, a split is
used in the grouping algorithm in order to separate day cases from inpatients
according to the treatment setting (see Chapter 19 of this volume). In Austria,
England and Germany, day cases are not identifi ed explicitly as part of the
grouping process. For reimbursement purposes, LKF groups, HRGs, and G-DRGs
are adjusted for cases with a LOS = 0. In addition, the English HRG system
identifi es certain procedures as being only applicable to day cases (NHS Infor-
mation Centre for Health and Social Care, 2010).
Furthermore, many countries are planning to develop DRG-like PCSs for psy-
chiatric and rehabilitation care (see Table 4.4). For rehabilitation care, several
PCSs have been proposed but heterogeneous duration and resource consump-
tion – as well as the absence of dominant procedures – make it diffi cult to defi ne
homogeneous groups of patients. However, in contrast to acute hospital care,
grouping can be used to classify cases or days (or weeks). The German Rehabilita-
tion Treatment Groups (RBG) system (Neubauer & Pfi ster, 2008) or the American
Inpatient Rehabilitation Facility Prospective Payment System (IRF-PPS) (Med-
PAC, 2009) classify cases. These systems take into account scores relating to
impairment, possible co-morbidities and age. The French Groupes homogène de
journées (Homogeneous groups of days, GHJ) (Metral et al., 2008) and the Swiss
Leistungsorientiertes Tarifmodell Rehabilitation (Performance-oriented payment
system for rehabilitation, LTR) (Fischer et al., 2010) classify days or weeks.
4.7.2 Number of groups
Figure 4.5 illustrates changes in the number of groups in different DRG-like
PCSs in Europe over time. It shows that the number of groups has continued to
increase in all systems. In most cases, there are only minor changes from year
to year. However, in France (GHMs) and England (HRGs), major revisions of the
DRG systems and similar patient classifi cation systems in Europe 53
grouping algorithm have taken place in recent years, and consequently the
number of groups has more than doubled in both countries. The G-DRG system
is the only PCS with large increases in the number of groups every year before
2010, when this trend was stopped.
There are several reasons for which the number of groups in DRG-like PCSs is
increasing: fi rst, most systems have tried to improve their ability to refl ect dif-
ferences in the complexity of treating different patients. In the G-DRG system,
the number of fi nal DRGs per base-DRG (refl ecting case complexity) has con-
tinuously increased over time. In France, the recent revision of the coding algo-
rithm introduced four severity levels for most base-DRGs; and in England, the
increase in the number of groups can be mostly attributed to the introduction
of more severity levels. Second, countries are increasingly moving to incorpo-
rating day care into their DRG-like PCSs. If day care is included within the same
classifi cation system, this may necessitate the creation of new groups to specifi -
cally refl ect resource consumption of day cases. Third, new medical devices,
drugs and medical knowledge become available, infl uence treatment patterns,
and may necessitate separating certain cases of one group of patients into a new
group, in order to assure medical and economic homogeneity of groups (see
Chapter 9 of this volume). In addition, the underlying coding systems (for both
diagnoses and procedures) are regularly updated in most countries. If the accu-
racy (granularity) of the coding systems is improved, this enables the creation
of patient groups that better refl ect specifi c characteristics of procedures or
patients, and are thus more homogeneous. Finally, improved cost accounting
Table 4.4 Trends in coverage of services in DRG-like PCSs in Europe
Country Inpatient Day cases Psychiatry Rehabilitation
Austria X Xa – –
England X X in the process of extension
Estonia X Xe – –
Finland X X Xb Xb
France X X in the process of extension
Germany X Xa planned for 2013 –
The Netherlands X X X X
Ireland X X – –
Poland X Xa in the process of extension
Portugal X Xa –c –c
Spain X –d – –
Sweden X X X X
Source: Authors’ own compilation based on information contained in the relevant country-
specifi c chapters of Part Two of this volume.
Notes: a Not explicitly part of the grouping algorithm but day-case status is explicitly considered
for payment purposes; b The DRG system is designed to cover such cases, but ‘in all hospitals,
psychiatric patients and patients requiring long-term intensive treatment (such as patients
suffering from respiratory arrest) are excluded’ from DRG billing (see Chapter 18 of this volume,
subsection 18.5.1); c Studies have been undertaken regarding the possibility of including
psychiatry and rehabilitation, but nothing concrete has come of this research; d Surgical day
cases are grouped and fi nanced using AP-DRGs in the same way as for inpatient care; Ongoing
research is taking place regarding International Refi ned (IR-)DRGs; e Only surgical day cases are
grouped and fi nanced using DRGs.
54 Diagnosis-Related Groups in Europe
in hospitals increases the ability of regulators to identify determinants of the
costs of treating patients and to adapt the PCS accordingly (see Chapter 5).
The Dutch DBC system is not included in Figure 4.5, since the number of
DBCs differs greatly from the number of groups in all other systems. However,
it is interesting to note that the DBC system is reducing the number of groups
with each revision of the system. Having started with about 100 000 DBCs in
2005, the number of groups was reduced to about 30 000 by 2010, and the
intention is to defi ne about 3000 DBCs, including severity levels similar to
DRGs (Warners, 2008).
4.7.3 Specifi c high-cost services: Unbundling, séances,
and supplementary payments
All DRG-like PCSs are faced with the problem of how to ensure that certain
specifi c high-cost services required by heterogeneous patients belonging to dif-
ferent DRG-like groups are adequately refl ected in the grouping process. In
order to do so, the English HRG system has developed the concept of ‘unbun-
dling’. This separates a set of certain services, such as chemotherapy, radio-
therapy, diagnostic imaging, renal dialysis, and high-cost drugs, from the core
HRGs (NHS Information Centre for Health and Social Care, 2009). By separat-
ing these services, the economic homogeneity of core HRGs is improved and, at
the same time, adequate reimbursement through supplementary payments can
Figure 4.5 Trends in the number of groups in DRG-like PCSs in Europe
Sources: Authors’ own compilation based on data provided by the Nordic Casemix Centre (2011),
as well as information contained in the relevant country-specifi c chapters of Part Two of this
volume and complemented by personal communications with the authors of those chapters.
DRG systems and similar patient classifi cation systems in Europe 55
be guaranteed (see Chapter 6 of this volume). In the French GHM system, there
is a category called ‘sessions’ (séances), which fulfi ls a similar purpose, also sepa-
rating renal dialysis, chemotherapy and radiotherapy from other services. In
Germany, an increasingly large number of supplementary payments exist (see
Chapter 14), which are not directly part of the grouping process but still fulfi l
the same purpose as unbundling or séances.
Another trend in DRG-like PCSs internationally is that attempts are being
made to differentiate better between co-morbidities on the one hand, and
complications attributable to poor-quality care, on the other. However, until
now only the United States MS-DRG system differentiates in this way, by
requiring providers to assign codes revealing whether diagnoses were present
on admission. If certain diagnoses that should not occur during hospitalization
were not identifi ed (coded) on admission, they are considered to refl ect poor
quality of care. How this information is used to adjust payment rates is discussed
by Or and Häkkinen in Chapter 8 of this volume.
4.8 Conclusions: Likelihood of a common Euro-DRG system?
In the context of an emerging European hospital market, a common defi nition
of hospital products through a common DRG-like PCS could be a major catalyst
to facilitate cross-border movements of patients and payments. Therefore,
establishing the likelihood of harmonization of DRG-like PCSs or, alternatively,
the development of a common European DRG-like system is of high relevance
for politicians and patients. In the introduction to this chapter, DRG-like PCSs
were defi ned as systems that have four main characteristics. (1) routinely collected
data on patient discharge are used to classify patients into (2) a manageable
number of groups that are (3) clinically meaningful and (4) economically homo-
geneous. These points can also be used to guide discussion about the possibility
of a common ‘Euro-DRG’ PCS.
Regarding the availability of routine data, section 4.6 discussed the fact that
similar information is used to classify patients in all systems, and is readily avail-
able from hospital discharge summaries, while section 4.4 demonstrated that
information is often coded in different ways. Therefore, an initial requirement
for a common European DRG-like system would be to harmonize coding of
diagnoses and procedures or to develop a mapping system that would allow the
translation of codes from different coding systems into a common European
coding system. The Hospital Data Project as part of the European Union (EU)’s
Health Monitoring Programme has suggested a common format for hospital
activity data, to improve comparability (Magee, 2003). For the coding of diag-
noses, an agreement on a coding system should be relatively unproblematic,
since the WHO ICD-10 system is already used for cause-of-death statistics in all
countries and the next revision, ICD-11 is currently being developed.
A question that is changing over time relates to what is regarded as a
manageable number of groups. Current developments of European DRG-like
PCSs seem to indicate that a number of between 1000 and 1500 groups is
necessary to describe the activity of hospitals. Since all countries use software
tools to classify patients into groups, the manageability of a system depends
56 Diagnosis-Related Groups in Europe
mostly on the ability to reliably calculate average costs of patients within each
group. In a European DRG-based system, the population basis for calculating
average costs of patients within each group would be much larger. Therefore, it
would be possible to develop a more detailed DRG system than currently exists
in each individual Member State.
In order to defi ne economically homogeneous groups of patients, the group-
ing algorithm of the DRG-like PCS needs to refl ect the most important deter-
minants of costs. If the determinants of costs are the same across European
countries, it should be possible to classify patients using the same DRG-like
PCS. Current research projects – such as the EU-funded EuroDRG project – aim
to identify the most important determinants of costs in 11 European countries.
The results of this project should be able to inform decisions about the feasibility
of developing a common European DRG-based system. However, if such a
system is to be developed, detailed cost-accounting information from a suf-
fi ciently large and representative sample of hospitals is essential (see Chapter 5
of this volume). In addition, mechanisms to ensure that the system is regularly
updated must be developed (see Chapter 9).
As shown in section 4.2, all currently existing DRG systems originate from
the original HCFA-DRG system, and even the self-developed DRG-like PCSs
share many elements of these systems. The most likely scenario for developing
a Euro-DRG system according to European needs seems to be that the existing
systems will form the basis for this work. In order to ensure that these
modifi cations do not change the principal of clinically meaningful groups, a
process would need to be set up to incorporate consultation with medical
professionals in developing and refi ning the DRG system.
In conclusion, while a European DRG system is unlikely to emerge within a
medium- to short-term time frame, the development of such a system does not
appear to be impossible. On the one hand, a number of requirements would
need to be fulfi lled, such as the development of common coding systems, cost-
accounting systems, and consultation mechanisms. On the other hand, over a
decade of experience using DRG-like PCSs in most countries has resulted in
several highly refi ned DRG-like PCSs that could serve as the starting point for
developing a new Euro-DRG system. Empirical analyses will be needed to
identify the system that best refl ects resource-consumption patterns in European
hospitals. However, similar to the historical emergence of DRG-like PCSs as a
result of political decisions, a common European PCS is only likely to emerge if
there is suffi ciently strong political will to support the emergence of a common
European hospital market.
4.9 Note
1 The authors thank Wilm Quentin for his efforts in revising this chapter and Caroline
Linhart for her work on the graphical representations.
DRG systems and similar patient classifi cation systems in Europe 57
4.10 References
3M (2005). All Patient DRG Defi nitions Manual (Version 23.0). St. Paul, MN: 3M Health
Information Systems.
3M (2003). All Patient Refi ned DRGs (Version 20.0) Methodology Overview. St. Paul, MN: 3M
Health Information Systems.
ATIH (2010). Manuel des GHM, Version 11b. Lyon: Agence Technique de l’Information sur
l’Hospitalisation (http://www.atih.sante.fr/index.php?id=000250002DFF, accessed
10 August 2010).
Australian Government (2004). Australian Refi ned Diagnosis Related Groups Version 5.1.
Defi nitions Manual Volume One (DRGs A01Z-I78B). Canberra: Commonwealth of
Australia Department of Health and Ageing.
BMG (2009). Leistungsorientierte Krankenanstaltenfi nanzierung – LKF – Modell 2010. Vienna:
Bundesministerium für Gesundheit.
Chilingerian, J. (2008). Origins of DRGs in the United States: A technical, political and
cultural story, in J. Kimberly, G. de Pouvourville, T. D’Aunno, eds. The Globalization
of Managerial Innovation in Health Care. Cambridge: Cambridge Universtiy Press.
European Parliament and Council (2011). Directive 2011/24/EU of the European Parliament
and of the Council of 9 March 2011 on the Application of Patients’ Rights in Cross-Border
Health Care. Brussels: Offi cial Journal of the European Union.
Fetter, R.B. (1999). Casemix classifi cation systems. Australian Health Review, 22(2):
16–34.
Fetter, R.B., Shin, Y., Freeman, J., Averill, R.F., Thompson, J.D. (1980). Casemix defi nition
by diagnosis-related groups. Medical Care, 18(2):i–53.
Fischer, W. (1997). Patientenklassifi kationssysteme zur Bildung von Behandlungsfallgruppen
im stationären Bereich – Prinzipien und Beispiele. Wolfertswil: Zentrum für Informatik
und wirtschaftliche Medizin (ZIM).
Fischer, W. (2008). Die DRG-Familie. Wolfertswil: Zentrum für Informatik und wirtschaftliche
Medizin (ZIM) (http://www.fi scher-zim.ch/textk-pcs/index.htm, accessed 10 July 2011).
Fischer, W., Blanco, J., Butt, M., Hund, M., Boldt, C. (2010). Leistungsorientiertes
Tarifmodell Rehabilitation (LTR). Neurologie & Rehabilitation, (16)3:1–18.
Goldfi eld, N. (2010). The evolution of diagnosis-related groups (DRGs): from its
beginnings in casemix and resource-use theory, to its implementation for payment
and now for its current utilization for quality within and outside the hospital. Quality
Management in Healthcare, 19(1):1–16.
InEK (2009). German Diagnosis-Related Groups Version 2010. Defi nitionshandbuch Kompakt-
version Band 1 (DRGs A01A-I98Z). Siegburg: Institut für das Entgeltsystem im
Krankenhaus gGmbH.
Magee, H.F. (2003). The hospital data project: comparing hospital activity within Europe.
European Journal of Public Health, 13(Suppl. 1):73–9.
MedPAC (2009). Rehabilitation Facilities (Inpatient) Payment System: Payment Basics.
Washington, DC: Medicare Payment Advisory Commission (http://www.medpac.
gov/documents/MedPAC_Payment_Basics_09_IRF , accessed 20 July 2010).
Metral, P., Ducret, N., Patris, A., Steunou, P. (2008). Improving casemix for description
and funding in rehabilitation in France: additive model is better than tree-
classifi cation. BMC Health Services Research, 8(Suppl. 1):A2.
Neubauer, G., Pfi ster, F. (2008). Entwicklung einer leistungsorientierten, fallgruppenspezifi schen
Vergütung in der Rehabilitation: Abschlussbericht. Munich: Institut für Gesundheit-
sökonomik (http://www.bdpk.de/media/fi le/358.Abschlussbericht_Prof._Neubauer_
Februar%202008 , accessed 20 July 2010).
NHS Information Centre for Health and Social Care (2009). The Casemix Service HRG4,
Guide to Unbundling. Leeds: NHS Information Centre for Health and Social Care.
58 Diagnosis-Related Groups in Europe
NHS Information Centre for Health and Social Care (2010). The Casemix Service HRG4
Design Concepts. Leeds: NHS Information Centre for Health and Social Care.
Nordic Casemix Centre (2011). Nordic Common Version (based on NCSP+) (NordDRG
manuals). Helsinki: Nordic Casemix Centre (http://www.nordcase.org/eng/norddrg_
manuals/versions/common/, accessed 26 July 2011).
Schreyögg, J., Stargardt, T., Tiemann, O., Busse, R (2006). Methods to determine reimburse-
ment rates for diagnosis related groups (DRG): a comparison of nine European
countries. Health Care Management Science, 9(3):215–23.
Warners, J. (2008). Redesigning the Dutch DBC-system; towards increased transparency.
From ideas and directions for improvement to implementation. Patient Classifi cation
Systems International: 2008 Case Mix Conference, Lisbon, 8–11 October.
chapter f i v e
DRGs and cost accounting:
Which is driving which?
Siok Swan Tan, Lisbeth Serdén,
Alexander Geissler, Martin van Ineveld,
Ken Redekop, Mona Heurgren and
Leona Hakkaart-van Roijen
5.1 Introduction
Cost-accounting systems could enable hospital managers to collect, summarize,
analyse and control the most relevant information regarding the allocation of
resources and reimbursement of hospital services (Finkler et al., 2007; Horngren
et al., 2006). Comprehensive cost-accounting systems are able to identify the
costs which are generated by some unit of analysis (such as by a diagnosis-
related group (DRG)) and could support the development of DRG-based
payment rate-setting mechanisms based on standardized cost data (Nathanson,
1984).
In the past, cost accounting has not been of high priority to hospitals in
conventional payment systems, such as fee-for-service reimbursement and
global budgets. With respect to fee-for-service reimbursement, prices charged
for typical conditions linked with standard services (that is, charges/bills
invoiced to payers) did not necessarily represent a good estimate of the cost of
individual services (Cohen et al., 1993; Ott, 1996). However, costs were, if at all,
likely to be registered in decentralized and mutually incompatible information
systems (Feyrer et al., 2005). Global budgets used to be the common funding
model in most European health care systems. One of the key advantages of
global budget arrangements was cost control; a fi xed payment was agreed in
advance for a target level of activity and hospitals’ level of reimbursement was
not directly related to the costs per patient (see Chapter 6 of this volume). The
inability of prospective budgeting to provide insight into hospital activity
restricted the planning – and possibly also the control – of the ever-growing
hospital costs.
60 Diagnosis-Related Groups in Europe
With the aim of improving the effi ciency of hospital care, DRG-based hospital
payment systems have been introduced in many European countries since 1983
(see Chapter 2). This development fundamentally changed hospital services
from being sources of incremental revenue (revenue centres) to being sources of
incremental costs (cost centres) (Berki, 1985).
Figure 5.1 provides an overview of the relevance of cost-accounting systems
in the DRG era. Regulatory authorities throughout Europe came to realize that
DRGs could not serve as payment rate-setting mechanisms without a function-
ing cost-accounting system (Feyrer et al., 2005); that is, effective and fair DRG-
based hospital payment systems to a large extent depend on high-quality and
accurate cost-accounting systems within hospitals (see Chapter 6 of this vol-
ume). Therefore, many countries started to routinely collect cost-accounting
data from a representative sample of hospitals in order to calculate and con-
tinuously update national DRG weights (for example, England, France and
Germany). Other countries have imported relative weights from abroad. In any
case, the use of DRGs as a payment mechanism increased the awareness of the
importance of accurate cost accounting in all hospitals, including those which
did not collect data for calculating national DRG prices, since erroneous cost
information would lead to inadequate relative weights and, ultimately, unin-
tended incentives for the delivery of services. However, the collection of hospi-
tal cost information has led to greater transparency. Moreover, hospital
managers recognized that cost accounting could support other purposes than
simply payment rate-setting, such as systematic benchmarking and managed
competition (Schuster et al., 2006; van de Ven & Schut, 2009). Precise cost
information enabled hospital managers to detect sources of resource consump-
tion in order to redesign treatment processes more effi ciently.
The aims of this chapter are to give a short introduction to cost accounting in
health care (section 5.2), to provide an overview of the different cost-accounting
methods across Europe (section 5.3) and to examine the interaction between
DRGs and cost accounting (section 5.4). The fi nal section (5.5) contains some
conclusions regarding the interdependency of DRGs and cost accounting, and
the prospects for harmonizing cost-accounting systems across Europe. The 12
Figure 5.1 Cost accounting in the DRG era
DRGs and cost accounting: Which is driving which? 61
countries considered were Austria, England, Estonia, Finland, France, Germany,
Ireland, the Netherlands, Poland, Portugal, Spain and Sweden.
5.2 Cost accounting in health care
In theory, there are three subsequent steps involved in allocating hospital costs
either to individual patients or groups of patient cases that are both medically
coherent and cost-homogeneous (St-Hilaire & Crepeau, 2000; Tan et al.,
2009c):
1. overhead cost allocation: allocation of hospital overhead costs to medical
departments (subsection 5.2.1);
2. indirect cost allocation: allocation of department overhead costs to patients
(subsection 5.2.2)
3. direct cost allocation: allocation of department direct costs to patients
(subsection 5.2.3).
5.2.1 Overhead allocation
The available literature describes different frameworks for allocating hospital
overhead costs to medical departments (Drummond et al., 2005; Williams et al.,
1982). The most commonly used framework is cost-centre allocation (Finkler et
al., 2007; St-Hilaire & Crepeau, 2000). In cost-centre allocation, a distinction is
made between medical departments and overhead departments. Medical depart-
ments provide patient care and may involve in- and outpatient clinics, labora-
tories, operating rooms (ORs) and radiology departments. Overhead departments
do not provide patient care and may include departments for administration,
facility management, logistics and security. Overhead costs from such depart-
ments may be assigned to medical departments by means of various allocation
bases, such as the number of inpatient days or the amount of direct costs
(Finkler et al., 2007; Horngren et al., 2006).
An alternative, very similar framework is ‘activity-based costing’. Activity-
based costing does not refer to a separate allocation methodology, but instead
emphasizes the importance of identifying the most accurate allocation base;
an allocation base should most closely refl ect a cause-and-effect relation-
ship between the overhead costs and the medical department (Cooper &
Kaplan, 1988). Hospital overhead costs are allocated to medical departments
based on the activities which drive them (for example, the area (m2) to allocate
costs of accommodation, and the number of full-time equivalents to allocate
administration costs), instead of using a more generic allocation base for all
overhead departments, such as inpatient days or direct costs (Drummond et al.,
2005).
Within either cost-centre allocation or activity-based costing, the available
literature describes three methods for allocating hospital overhead costs to
medical departments. The simplest method is ‘direct allocation’, in which
overhead costs are allocated to medical departments without interaction
between overhead departments (Figure 5.2).
62 Diagnosis-Related Groups in Europe
A second method, ‘step-down allocation’, partially adjusts for interaction
between overhead departments. The method appoints overhead costs to both
the medical departments and the remaining overhead departments in a stepwise
fashion. The step-down method accounts for unilateral deliveries between
overhead departments. This means that the sequence in which overhead
departments allocate their costs is important (Figure 5.3); that is, costs of the
second overhead department cannot be allocated to the fi rst one.
A fi nal method for allocating costs to medical departments is the ‘reciprocal’
method in which overhead costs are appointed to both the medical departments
and to all other overhead departments. The reciprocal method takes into
account bilateral deliveries between overhead departments. This means that
the procedure should be repeated a number of times to eliminate residual
unallocated amounts (Figure 5.4).
5.2.2 Indirect cost allocation
Department overhead costs (indirect costs) are those costs incurred by medical
departments that are not directly related to patients, such as the personnel costs
of non-medical staff and inventory. Cost-centre allocation and activity-based
costing are not applicable to the allocation of costs to patients, because they
assume a cause-and-effect relationship with the medical department, rather
Figure 5.2 Direct method for overhead allocation
Notes: OD: overhead department; MD: medical department.
Figure 5.3 Step-down method for overhead allocation
Notes: OD: overhead department; MD: medical department.
DRGs and cost accounting: Which is driving which? 63
than with patients. Instead, department overhead costs may be assigned to
patients using the following methods (Finkler et al., 2007; Tan et al., 2009c):
• marginal mark-up percentages: indirect costs distributed to direct costs by
raising the direct costs with a mark-up percentage;
• weighting statistics: service time, for example, used as a proxy for resource
consumption, yielding a cost per treatment minute or inpatient day;
• relative value units: establish the relative cost of each patient by assigning a
base value to the base-line resource use of the hospital service and adding
relative values to this base value when the patient uses additional resources.
5.2.3 Direct cost allocation
Direct costs refer to the costs incurred by medical departments which are
directly linked to patients, such as the personnel costs of medical staff (especially
if clearly involved with a particular procedure), medications and materials.
Direct costs per patient are calculated by subsequently identifying the quantities
of hospital services a patient consumed and valuing these hospital services with
their unit costs (Drummond et al., 2005; Jackson, 2000). Overall, there are four
methodologies to calculate the direct costs per patient (Figure 5.5). These
methodologies differ in terms of the level of accuracy with which they identify
hospital services (‘gross-costing’ versus ‘micro-costing’) and value hospital services
(the ‘top-down’ versus ‘bottom-up’ approaches) (Tan et al., 2009b; Tan, 2009).
Gross-costing identifi es hospital services at a highly aggregated level; often
inpatient days are defi ned as the only hospital service ( Jackson, 2000; Tan et al.,
2009b). Top-down gross-costing values inpatient days per average patient,
whereas bottom-up gross-costing values inpatient days per individual patient.
Top-down micro-costing identifi es all relevant hospital services at the most
detailed level, but values each hospital service per average patient (Tan et al.,
2009b; Wordsworth et al., 2005). Hospital services may comprise staff time,
laboratory services, medical imaging services, medications, medical materials
and (surgical) procedures. As the methodology does not require patient-level
data, statistical analyses of costs cannot be carried out, and differences between
patients cannot be detected (Clement Nee Shrive et al., 2009).
Figure 5.4 Reciprocal method for overhead allocation
Notes: OD: overhead department; MD: medical department.
64 Diagnosis-Related Groups in Europe
Bottom-up micro-costing identifi es and values all relevant hospital services at
the most detailed level. Because the methodology values hospital services per
individual patient, bottom-up micro-costing enables statistical analyses to
determine whether there are cost differences between patients for each single
hospital service and combinations of hospital services (Tan et al., 2009b;
Wordsworth et al., 2005).
5.3 Costing approaches across Europe
5.3.1 Ascertaining cost data across Europe
Nearly every European country has a unique approach to collecting cost data in
order to further develop their DRG-based system(s).
Mandatory cost-accounting systems
Most countries allow their hospitals to use a cost-accounting system which best
fulfi ls their own needs, but some countries require their hospitals to have
mandatory cost-accounting systems (such as England and Portugal). However,
despite the presence of mandatory cost-accounting systems, some variations
between systems may still exist within countries.
Presence of national costing guidelines
In addition, the absence of mandatory cost-accounting systems does not pre-
vent some countries from encouraging systematic cost accounting by means of
national costing guidelines. For example, hospitals in France are recommended
to apply the hospital cost-accounting model ‘analytical accounting’. In Ireland,
regulation relating to the collection of cost data is enforced centrally, using a
national costing manual and auditing. In Sweden, national guidelines have
been developed for cost-per-case calculations.
Figure 5.5 Methodology matrix: Level of accuracy of the
identifi cation and valuation of hospital services
Sources: Tan, 2009; Tan et al., 2009b.
DRGs and cost accounting: Which is driving which? 65
Cost-accounting data used for calculating DRG prices
The majority of countries use nationally collected cost-accounting data to cal-
culate DRG prices (Austria, England, Estonia, Finland, France, Germany, the
Netherlands and Sweden). Other countries have imported DRG weights from
abroad (Ireland, Poland, Portugal and Spain), but each of these countries uses at
least some cost-accounting data to adjust imported DRG weights to their local
situation. Ireland initially adopted a slightly modifi ed version of the Victorian
Cost Weights methodology for casemix modelling. These relative cost weights
have been refreshed, adjusted to the local context and updated for the Irish
health care system in subsequent years. Poland only calculates DRG weights for
specifi c procedures; costs for the remaining procedures are determined relative
to the costs of the United Kingdom Healthcare Resource Groups (HRG) system
(Version 3.5). Portugal allocates inpatient costs to DRGs based on the cost
weights of Maryland in the United States and on the lengths of stay in
Portuguese hospitals. Spain also uses relative weights from the United States,
adjusting them to the Spanish context. Table 5.1 presents an overview of the
different approaches to collecting cost data in 12 European countries.
5.3.2 Cost-accounting methods across Europe
Number (share) of cost-collecting hospitals
In most of the countries that use national cost-accounting data for calculating
DRG weights, the calculation is based on a selected number of hospitals from
which reliable cost-accounting data are collected and pooled. Selected hospitals
typically use comparable cost-accounting systems meeting predefi ned quality
standards (for example, Finland, Germany, the Netherlands and Sweden)
(Schreyögg et al., 2006). DRG weights may also be based on cost-accounting
data from a sample of hospitals which have contracts with the country’s
national health insurance fund (such as the EHIF in Estonia), or which partici-
pate in ongoing projects (for example, in France). Other countries require all
Table 5.1 Different approaches to collecting cost data in Europe
Mandatory cost- National Cost-accounting data used
accounting system costing guidelines for developing DRG prices
Austria – – ×
England × × ×
Estonia – – ×
Finland – – ×
France – × ×
Germany – × ×
Ireland – × –
Poland – – –
Portugal × × –
Netherlands × × ×
Spain – – –
Sweden – × ×
66 Diagnosis-Related Groups in Europe
hospitals to report their activity and unit costs annually to their regulatory
authority (as is the case in England).
For example, the subset of 15–25 cost-collecting hospitals in the Netherlands
were required to implement cost-accounting systems that were able to capture
patient-level data for the allocation of costs to the individual patient and to
support the maintenance, registration and validation of the 30 000 diagnosis–
treatment combinations (Diagnose Behandeling Combinaties, DBCs).
The regulatory authorities in some countries started to provide special monetary
incentives to hospitals which complied with predefi ned standards for cost
accounting. In France, the regional health authorities award the yearly salary for
a fi nancial controller, by means of the ‘payment for general interest missions’
(MiGAC)1 to each of the hospitals providing cost-accounting data. After having
calculated relative weights without using monetary incentives for participating
hospitals in the years 2003 and 2004, Germany introduced such payments in
2005. Currently, the Institute for the Hospital Remuneration System (InEK)
reimburses hospitals with an additional fee for voluntarily collecting patient-level
cost-accounting data. This consists of a lump sum and a variable amount related
to the number of delivered cases and their data quality. In 2008, the InEK spent
€9 million to compensate hospitals for their additional efforts (InEK, 2009).
There is a trade-off between ensuring high-quality data standards and
obtaining a representative number of cost-collecting hospitals. A large number
of hospitals may provide a clearer picture of differences in the severity of cases,
or in the structure of hospitals in a particular country, insofar as these factors
have already not been adjusted for separately. However, a small number of cost-
collecting hospitals – with comparable, high-quality cost-accounting systems –
may allow the data quality obtained to be higher, but with the disadvantage
that data on rare treatments (for multiple trauma patients, for example) might
not be available (Schreyögg et al., 2006).
Overhead allocation
For the allocation of overhead costs to the medical departments, European coun-
tries either use the direct method (England, Estonia, Finland, the Netherlands
and Sweden) or the step-down method (France). Germany intends to use the
step-down method in the hospitals from which cost-accounting data are col-
lected. If this is not feasible, however, a combination of the step-down cost-
accounting method and other methods (such as the direct method) can be used.
In the Netherlands, hospitals are free to choose the method to be used for the
allocation of hospital costs. As the allocation method was found to have only a
minor impact on individual patients’ costs, hospitals commonly use the simple
method of direct allocation.
Indirect and direct cost allocation
Most countries require their cost-collecting hospitals to report minimum data-
sets containing patient and/or hospital characteristics, some clinical parameters
(such as diagnoses, status at discharge) as well as cost-accounting data. Mini-
mum datasets containing cost-accounting data are fairly similar across European
countries. For example, Finland collects resource-use and unit cost data relating
DRGs and cost accounting: Which is driving which? 67
to inpatient days, outpatient visits, laboratory services, medical imaging ser-
vices, medications, blood products, surgical procedures and pathological ser-
vices for each treated patient.
In Austria, hospitals can implement cost-accounting systems to suit their
needs. However, hospitals fi nanced by State Health Funds report highly aggre-
gated and standardized data (113 out of 264 hospitals). In contrast, most coun-
tries apply various weighting statistics in combination with the micro-costing
methodology to allocate costs from the medical department to patients. Coun-
tries recording data on itemized resource consumption apply the bottom-up
approach to allocate hospital costs to individual patients (or hospital services)
(Finland, Germany, the Netherlands and Sweden). Countries in which patient-
level data are not available apply the top-down approach to allocate hospital
costs to average patients (inpatient admissions) (England, Estonia and France).
In England, a working group of costing experts has been established to support
the implementation of Patient-Level Information and Costing Systems (PLICS)
within the National Health Service (NHS). As of yet, the implementation of
PLICS is not mandatory and the number of hospitals that have introduced
patient-level costing is not known. In Estonia and France, the calculation of
DRG costs is a combination of the top-down accounting model with a (small)
proportion of costs being identifi ed at the patient level.
Data checks on reported cost data
In most countries, data checks on reported cost-accounting data initially take
place internally at the hospitals. In addition, data checks are commonly carried
out annually either by the national authority (in England, Estonia, Germany
and the Netherlands) or by the regional authority (in Austria and France). In
Finland, ensuring data quality is the sole responsibility of the hospitals, as no
offi cial data quality and plausibility checks are undertaken at the national or
regional levels. In Sweden, the National Board of Health and Welfare publishes
reports on coding activity and quality based on information from the National
Patient Register (NPR) but it is the county councils’ responsibility to check the
quality of data through case record audits.
In most countries, national/regional data checks on reported cost-accounting
data primarily focus on resource-use information, in terms of technical and
clinical validity; that is, coded hospital services are held against certain patient
and/or hospital characteristics. For example, a check is performed to establish
whether a procedure is allowed/plausible for a specifi c hospital or patient. In
some countries, data checks are additionally performed on unit cost information
(Germany, the Netherlands and Sweden). In Germany, unit costs are compared
to minimum and maximum values, to unit cost ratios between hospital services,
and to corresponding resource-use information; for example, costs for a hip
replacement must refl ect the material cost of implants. Cost-accounting data
are either checked for all hospitals (in Germany, the Netherlands and Sweden)
or for random samples (in Austria, England, Estonia and France).
Table 5.2 presents some characteristics of the cost-accounting methods in
eight European countries, using their own cost-accounting data for calculating
DRG prices.
T
a
b
le
5
.2
D
if
fe
re
n
t
ch
ar
ac
te
ri
st
ic
s
o
f
th
e
co
st
-a
cc
o
u
n
ti
n
g
m
et
h
o
d
s
in
e
ig
h
t
E
u
ro
p
ea
n
c
o
u
n
tr
ie
s
O
ve
rh
ea
d
c
os
t
a
ll
oc
a
ti
on
to
m
ed
ic
a
l
d
ep
a
rt
m
en
ts
In
d
ir
ec
t
co
st
a
ll
oc
a
ti
on
t
o
pa
ti
en
ts
D
ir
ec
t
co
st
a
ll
oc
a
ti
on
to
p
a
ti
en
ts
N
u
m
b
er
(
sh
a
re
)
of
c
os
t-
co
ll
ec
ti
n
g
h
os
pi
ta
ls
D
a
ta
c
h
ec
ks
(
re
gu
la
ri
ty
)
A
u
st
ri
a
V
ar
y
in
g
b
y
h
o
sp
it
al
V
ar
y
in
g
b
y
h
o
sp
it
al
G
ro
ss
-c
o
st
in
g
2
0
r
ef
er
en
ce
h
o
sp
it
al
s
(~
8
%
o
f
al
l
h
o
sp
it
al
s)
R
eg
io
n
al
a
u
th
o
ri
ty
(i
rr
eg
u
la
rl
y
)
E
n
g
la
n
d
D
ir
ec
t
W
ei
g
h
ti
n
g
s
ta
ti
st
ic
s
T
o
p
-d
o
w
n
m
ic
ro
-c
o
st
in
g
A
ll
h
o
sp
it
al
s
N
at
io
n
al
a
u
th
o
ri
ty
(a
n
n
u
al
ly
)
E
st
o
n
ia
D
ir
ec
t
M
ar
k
-u
p
p
er
ce
n
ta
g
e
T
o
p
-d
o
w
n
m
ic
ro
-c
o
st
in
g
H
o
sp
it
al
s
co
n
tr
ac
te
d
w
it
h
t
h
e
E
H
IF
N
at
io
n
al
a
u
th
o
ri
ty
(a
n
n
u
al
ly
)
Fi
n
la
n
d
D
ir
ec
t
W
ei
g
h
ti
n
g
s
ta
ti
st
ic
s
B
o
tt
o
m
-u
p
m
ic
ro
-c
o
st
in
g
5
r
ef
er
en
ce
h
o
sp
it
al
s
m
ee
ti
n
g
p
ar
ti
cu
la
r
co
st
-a
cc
o
u
n
ti
n
g
st
an
d
ar
d
s
(~
3
0
%
o
f
sp
ec
ia
li
ze
d
ca
re
)
N
o
,
re
sp
o
n
si
b
il
it
y
o
f
h
o
sp
it
al
s
Fr
an
ce
St
ep
d
o
w
n
W
ei
g
h
ti
n
g
s
ta
ti
st
ic
s
T
o
p
-d
o
w
n
m
ic
ro
-c
o
st
in
g
9
9
v
o
lu
n
te
er
in
g
h
o
sp
it
al
s
p
ar
ti
ci
p
at
in
g
i
n
t
h
e
E
N
C
C
(~
1
3
%
o
f
in
p
at
ie
n
t
ad
m
is
si
o
n
s)
R
eg
io
n
al
a
u
th
o
ri
ty
(a
n
n
u
al
ly
)
G
er
m
an
y
St
ep
d
o
w
n
(
p
re
fe
ra
b
ly
)
W
ei
g
h
ti
n
g
s
ta
ti
st
ic
s
B
o
tt
o
m
-u
p
m
ic
ro
-c
o
st
in
g
1
2
5
v
o
lu
n
te
er
in
g
h
o
sp
it
al
s
m
ee
ti
n
g
I
n
E
K
c
o
st
-a
cc
o
u
n
ti
n
g
st
an
d
ar
d
s
(~
6
%
o
f
al
l
h
o
sp
it
al
s)
N
at
io
n
al
a
u
th
o
ri
ty
(a
n
n
u
al
ly
)
N
et
h
er
la
n
d
s
D
ir
ec
t
W
ei
g
h
ti
n
g
s
ta
ti
st
ic
s
B
o
tt
o
m
-u
p
m
ic
ro
-c
o
st
in
g
U
n
it
c
o
st
s:
1
5
–2
5
v
o
lu
n
te
er
in
g
g
en
er
al
h
o
sp
it
al
s
(~
2
4
%
o
f
al
l
h
o
sp
it
al
s)
N
at
io
n
al
a
u
th
o
ri
ty
(a
n
n
u
al
ly
)
Sw
ed
en
D
ir
ec
t
W
ei
g
h
ti
n
g
s
ta
ti
st
ic
s
B
o
tt
o
m
-u
p
m
ic
ro
-c
o
st
in
g
H
o
sp
it
al
s
w
it
h
c
as
e
co
st
in
g
sy
st
em
s
(~
6
2
%
o
f
in
p
at
ie
n
ts
ad
m
is
si
o
n
s)
N
at
io
n
al
a
n
d
re
g
io
n
al
a
u
th
o
ri
ti
es
(a
n
n
u
al
ly
)
DRGs and cost accounting: Which is driving which? 69
5.3.3 Breadth of costs covered by European
DRG-based payments
The relative importance of any DRG-based hospital payment system is deter-
mined by the share of hospital costs that are covered by DRG-based payments
(see Chapter 6, section 6.3, Table 6.2). DRG payment is the principal means of
reimbursing hospitals in the majority of European countries. For example, in-
patient care funding through DRGs represents 75–85 per cent of hospital costs
in Germany and Portugal. However, most countries exclude some (medical)
specialties and/or hospital services due to (Schreyögg et al., 2006):
• the usual incentive set by the DRG system to shorten the patient’s length of
stay, which is considered harmful in these specialties (for example, intensive
care);
• coding problems in hospital services for which DRG prices cannot be reliably
calculated because they are rarely provided (for example, for multiple trauma
care);
• circumstances involving specialties in which a diagnosis seems to be a bad
predictor for costs (for example, psychiatric care).
Some countries therefore exclude the costs for rehabilitation (France, Germany,
Ireland and Sweden), psychiatric services (England, Finland, France, Germany
and Ireland), and intensive and emergency care (Finland, France and Poland).
Other costs excluded from the system may involve primary care services,
community services and ambulance services (England); neonatology, dialysis
and radiotherapy performed during hospitalization (France); geriatric services
(Ireland); and burn treatment (Sweden). Costs for excluded hospital services are
mostly reimbursed via supplementary fees, fee-for-service reimbursement and/
or surcharges (see Chapter 6 of this volume).
In addition, the costs for expensive drugs (in France, Germany, the Netherlands,
Poland and Sweden) and/or expensive materials (in France, Spain and Sweden)
are not commonly reimbursed using DRGs.
With respect to specialties, the costs for education and research are not
commonly funded through DRGs. Some countries also exclude capital costs
and interest (for example Austria, Finland, Germany and Ireland) and allowance
for debts (for example Germany and Ireland). Other disregarded costs may
relate to taxes, charges and insurance (in Germany), pensions (in Ireland) and
accreditation (in Sweden).
5.4 Developing DRG systems with cost data
5.4.1 Relevance of cost-accounting systems
Cost-accounting data play an important role in calculating DRG weights
(Nathanson, 1984). If the data given by cost-accounting systems are imprecise,
hospitals are likely to be over- or underpaid for specifi c DRGs. In practice,
profi table DRGs may compensate for less-profi table DRGs (cross-subsidizing).
However, if cost-accounting data lead to an overestimated payment for a specifi c
70 Diagnosis-Related Groups in Europe
DRG, hospitals are disincentivized to reorganize treatment processes in order to
improve effi ciency for certain groups of patients. On the other hand, if cost-
accounting data lead to an underestimated payment for a specifi c DRG, hospitals
are disincentivized to provide high-quality care as this may lead to costs above
the payment level. These hospitals may start to compromise quality in order
to reduce their costs (or losses). Consequently, the appropriate level of hos-
pital payment to a large extent determines the effectiveness and fairness of
DRG-based hospital payment systems. Hospital managers, as well as regulating
authorities, should consider whether the benefi ts of more reliable cost data
justify the additional costs and complexity incurred in improving the cost-
accounting systems to obtain accurate and detailed information. The choice
that they make between costing methods should refl ect the importance of
accurate cost estimates, feasibility and the costs associated with introducing
the system (Clement Nee Shrive et al., 2009).
Several previous studies have demonstrated that DRG-based hospital pay-
ments do not always adequately refl ect costs (Busse et al., 2008; Heerey et al.,
2002; Skeie et al., 2002; Tan et al., 2009a). This may be explained by inaccura-
cies in the patient classifi cation and cost-accounting systems.
Inaccuracies in the patient classifi cation
Although health care providers have long contended that every patient is
unique, the reality of DRG-based hospital payment systems is that patients are
grouped together, and that some groups represent a mixture of diagnoses to a
greater extent than others. Countries with itemized resource use per patient
commonly use cost accounting to support adequate resource allocation, to
assess the homogeneity of resource consumption within each DRG, to calculate
separate DRG payments for patients requiring more complex resource use (for
example, due to complications and co-morbidities (CCs)) and/or to test the
effect of changes in the PCS (for example France, Germany, the Netherlands
and Sweden). For instance, medical DRGs were found to be less homogeneous
than surgical DRGs in France, but the creation of new DRGs was restricted by
the small number of cases which would be affected in different medical stays.
Cost accounting has shown that the dispersion around the mean costs varies
greatly between DRGs, with highly variable DRGs most likely to comprise a
wide variety of different diagnoses and treatments ( Jackson, 2000).
Inaccuracies in cost-accounting systems
The extent to which cost-accounting systems could support the effi cient and
fair use of DRG systems as a reimbursement tool relies on:
• the number and composition (sample characteristics) of hospitals from which
cost-accounting data are collected and the quality of data delivered by these
hospitals;
• the accuracy of the cost-accounting method (see subsection 5.4.2);
• the ability to maintain/update the cost-accounting data in a timely manner.
DRGs and cost accounting: Which is driving which? 71
5.4.2 The accuracy of the cost-accounting method
The extent to which cost-accounting systems could support the effective and
fair development of DRG-based hospital payment systems as a reimbursement
tool is determined by the accuracy of the cost-accounting method used in the
respective country or region. However, the nature of costs is such that the more
refi ned the analysis, the more costly it generally is. The reciprocal method
theoretically allocates hospital costs to medical departments most precisely. At
the same time, it is more time-consuming than the methods used across Europe
(‘step-down’ and ‘direct’ methods). Earlier studies have revealed no statistically
signifi cant relationship between alternative cost-accounting methods and the
unit costs produced (St-Hilaire & Crepeau, 2000; Zuurbier & Krabbe-Alkemade,
2007). Likewise, relative value units are believed to most closely refl ect actual
resource consumption for the allocation of hospital costs to patients, but their
calculation requires more detailed data than operational methods require
(weighting statistics and marginal mark-up percentages). Weighting statistics
have been shown to provide reasonably similar cost estimates, while marginal
mark-up percentages result in substantially different cost estimates compared to
those based on relative value units (Tan et al., 2009c).
Bottom-up micro-costing may be the preferred methodology for calculating
DRG weights because it helps hospital managers to understand whether cost
differences between and within DRGs arise from variation in resource-use
intensity or from variations in the costs of hospital services; it can also help to
understand the distributional form of the cost-accounting data on which DRG
payments are based (Jackson, 2000; Tan et al., 2009b). Unlike the alternative
methodologies, bottom-up micro-costing allows for insight into the costs
directly employed for individual patients, cost homogeneities and high-cost
outliers. However, countries need to rely on top-down micro-costing (or gross-
costing) if their hospitals’ cost-accounting systems do not collect itemized
resource-use data for each individual patient (as is the case in England, for
example). Top-down micro-costing has proven to be a strong alternative to
bottom-up micro-costing in terms of accuracy, and the approach is fairly
feasible with respect to data availability, costs and complexity (Tan et al.,
2009b). In contrast, both economic theory and empirical studies support the
notion that gross-costing results in rather inaccurate cost estimates. For exam-
ple, the patient’s diagnosis has an important effect on the use of resources, and
this is something which is not generally refl ected by gross-costing methods
(Jackson, 2000; Swindle et al., 1999).
5.5 Impact of cost accounting on hospitals
Currently, cost-accounting systems certainly represent an improvement over
the information that was formerly available in many institutions. Cost-
accounting systems offer an effi cient and clinically sound approach for describ-
ing and managing hospital activity, in order to offer greater transparency in
the fi nancing of health care. Hospitals across Europe recognized that cost-
accounting data are fundamental for systematic benchmarking and for
72 Diagnosis-Related Groups in Europe
managed competition approaches that can improve the effi ciency of hospital
service delivery (Schuster et al., 2006; van de Ven & Schut, 2009). First, cost-
accounting systems facilitate the comparison of performance indicators, along
with productivity and effi ciency parameters. Benchmarking has also helped
hospitals to manage and control operating processes and thus improves
their performance; for example, it encouraged the use of DRGs in assessing the
budgetary impact of anticipated changes in the volume and casemix of patients
and in monitoring actual expenditure versus expected levels (for example in
England, France and Estonia). Second, managed competition has allowed
authorities in many European countries to provide powerful incentives to other
actors in the system, such as health insurers/sickness funds (Busse et al., 2006).
It led to the use of DRGs to negotiate on service quality and access, as well as
on detailed cost- and volume-based fi nancial components (for example in
England, Estonia, the Netherlands and Spain). Finally, cost-accounting data
enable regulatory authorities to monitor unintended incentives that are sup-
posed to accompany DRG-based payments, such as treatment of patients whose
expected costs are lower than the associated reimbursement, up-coding of
expensive DRGs to increase revenue, cost minimization or shifting of treatment
costs onto other parties, and compromising quality of care (see Chapter 6 of
this volume).
5.6 Conclusions: Which is driving which?
One may argue that cost accounting is driving the further development of
DRGs. The introduction of DRG-based hospital payment systems in Europe
partly originated from the absence, or inadequacy, of information relating to
cost data (Feyrer et al., 2005). Cost-accounting data made it possible to validate
cost homogeneities and to detect cost-outliers in the patient population. This
led to revisions and refi nements of the existing DRG systems. However, one
could also argue that DRGs are driving cost accounting. The introduction of
comprehensive and standardized cost-accounting systems was encouraged by
the need to collect data for calculating DRG weights as well as supporting hos-
pital management and auditing. Revisions of existing cost-accounting systems
are undertaken to improve the effectiveness and fairness of DRG-based hospital
payment systems. Regardless of which argument one chooses, the following
observations cannot be disputed: (1) DRG systems cannot function well with-
out accurate cost accounting; and (2) the necessity of cost-accounting systems
to use costs based on a unit of analysis is met by DRGs. However, it is crucial to
note that DRG and cost-accounting systems should be developed independ-
ently of each other; otherwise it will be impossible to validate the systems’
performance individually.
The way in which cost-accounting data are collected for developing DRG-
based hospital payment systems and the way in which DRG weights are
calculated differ substantially among the European countries concerned. Two
observations are important in this regard. First, the characteristics of the DRG
and cost-accounting systems refl ect the current situation but are, in effect,
subject to (rapid) change based on the dynamics of the systems they represent.
DRGs and cost accounting: Which is driving which? 73
Second, there is no ‘best’ cost-accounting system in general, because the choice
of the system must be made based on the characteristics of the cost-collecting
hospitals and the national health systems concerned, as well as on the objective
that health policy-makers intend to fulfi l by using DRG systems. However, the
only way to truly evaluate DRGs in terms of medical coherence and cost
homogeneity is to defi ne costs at the individual patient level. Cost-accounting
systems may not be suffi ciently meaningful to measure, compare and improve
effi ciency of hospital care if DRG costs are not defi ned according to bottom-up
micro-costing.
Each DRG-based hospital payment system has similar aims (for example, to
increase transparency, to ensure adequate hospital reimbursement) but reaching
these is to be achieved within different nation-specifi c health system contexts
(see Chapter 2). Therefore, it is unlikely that cost-accounting systems across
Europe will be harmonized in the near future. However, European countries are
likely to deal with many of the same issues concerning the ongoing process of
developing and updating DRGs and cost-accounting systems in the years to
come. An overall similarity in terms of the problems they encounter may in
time lead to greater interest in fi nding common solutions that are adjustable for
each country.
5.7 Note
1 Missions d’intérêt général et de l’aide à la contractualisation: Missions of general interest
and assistance with contracting, including payments for education, research and
public health programmes.
5.8 References
Berki, S.E. (1985). DRGs, incentives, hospitals, and physicians. Health Affairs (Millwood),
4:70–6.
Busse, R., Schreyögg, J., Smith, P.C. (2006). Hospital case payment systems in Europe.
Health Care Management Science, 9:211–13.
Busse, R., Schreyögg, J., Smith, P.C. (2008). Variability in healthcare treatment costs
amongst nine EU countries – Results from the HealthBASKET project. Health
Economics, 17:1–8.
Clement Nee Shrive, F.M., Ghali, W.A., Donaldson, C., Manns, B.J. (2009). The impact of
using different costing methods on the results of an economic evaluation of cardiac
care: micro-costing vs gross-costing approaches. Health Economics, 18:377–88.
Cohen, D.J., Breall, J.A., Ho, K.K. et al. (1993). Economics of elective coronary
revascularization. Comparison of costs and charges for conventional angioplasty,
directional atherectomy, stenting and bypass surgery. Journal of the American College
of Cardiology, 22:1052–9.
Cooper, R., Kaplan, R.S. (1988). Measure costs right. Make the right decisions. Harvard
Business Review, Sep/Oct:96–103.
Drummond, M.F., Sculpher, M.J., Torrance, G.W., O’Brien, B.J., Stoddart, G.L. (2005).
Methods for the Economic Evaluation of Health Care Programmes. Oxford: Oxford
University Press.
74 Diagnosis-Related Groups in Europe
Feyrer, R., Rosch, J., Weyand, M., Kunzmann, U. (2005). Cost unit accounting based on a
clinical pathway: a practical tool for DRG implementation. The Thoracic and
cardiovascular surgeon, 53:261–6.
Finkler, S.A., Ward, D.M., Baker, J.J. (2007). Essentials of Cost Accounting for Health Care
Organizations. New York, NY: Aspen Publishers.
Heerey, A., McGowan, B., Ryan, M., Barry, M. (2002). Micro-costing versus DRGs in the
provision of cost estimates for use in pharmacoeconomic evaluation. Expert Review of
Pharmacoeconomics and Outcomes Research, 2:29–33.
Horngren, C.T., Datar, S.M., Foster, G. (2006). Cost Accounting: A Managerial Emphasis.
Upper Saddle River, NJ: Pearson Prentice Hall.
InEK (2009). Abschlussbericht zur Weiterentwicklung des G-DRG-Systems für das Jahr 2010.
Siegburg: Institut für das Entgeltsystem im Krankenhaus gGmbH.
Jackson, T. (2000). Cost estimates for hospital inpatient care in Australia: evaluation of
alternative sources. Australian and New Zealand Journal of Public Health, 24:234–41.
Nathanson, M. (1984). DRG cost-per-case management. Comprehensive cost accounting
systems give chains an edge. Modern Healthcare, 14(3):122, 124, 128.
Ott, K. (1996). A comparison of craniotomy and Gamma Knife charges in a community-based
Gamma Knife Center. Stereotactic and Functional Neurosurgery, 66(Suppl 1.):357–64.
Schreyögg, J., Stargardt, T., Tiemann, O., Busse, R. (2006). Methods to determine
reimbursement rates for diagnosis related groups (DRG): a comparison of nine
European countries. Health Care Management Science, 9:215–23.
Schuster, M., Kuntz, L., Hermening, D. et al. (2006). The use of diagnosis-related-groups
data for external benchmarking of anesthesia and intensive care services. Anaesthesist,
55(1):26–32.
Skeie, B., Mishra, V., Vaaler, S., Amlie, E. (2002). A comparison of actual cost, DRG-based
cost, and hospital reimbursement for liver transplant patients. Transplant International,
15:439–45.
St-Hilaire, C., Crepeau, P.K. (2000). Hospital and unit cost-allocation methods. Healthcare
Management Forum, 13:12–32.
Swindle, R., Lukas, C.V., Meyer, D.A., Barnett, P.G., Hendricks, A.M. (1999). Cost analysis
in the department of veterans affairs: consensus and future directions. Medical Care,
37:AS3–8.
Tan, S.S. (2009). Micro-costing in Economic Evaluations: Issues of Accuracy, Feasibility,
Consistency and Generalizability. Rotterdam: Erasmus Universiteit Rotterdam.
Tan, S.S., Oppe, M., Zoet-Nugteren, S.K. et al. (2009a). A micro-costing study of diagnostic
tests for the detection of coronary artery disease in the Netherlands. European Journal
of Radiology, 72:98–103.
Tan, S.S., Rutten, F.F., van Ineveld, B.M., Redekop, W.K., Hakkaart-van Roijen, L. (2009b).
Comparing methodologies for the cost estimation of hospital services. European
Journal of Health Economics, 10:39–45.
Tan, S.S., van Ineveld, B.M., Redekop, W.K., Hakkaart-van Roijen, L. (2009c). Comparing
methodologies for the allocation of overhead and capital costs to hospital services.
Value in Health, 12:530–5.
Van de Ven, W.P., Schut, F.T. (2009). Managed competition in the Netherlands: still work-
in-progress. Health Economics, 18:253–5.
Williams, S.V., Finkler, S.A., Murphy, C.M., Eisenberg, J.M. (1982). Improved cost
allocation in casemix accounting. Medical Care, 20:450–9.
Wordsworth, S., Ludbrook, A., Caskey, F., Macleod, A. (2005). Collecting unit cost data in
multicentre studies. Creating comparable methods. European Journal of Health
Economics, 6:38–44.
Zuurbier, J., Krabbe-Alkemade, Y. (2007). Onderhandelen Over DBC’s [Negotiating on DBCs].
Maarssen: Elsevier Gezondheidszorg.
chapter s i x
DRG-based hospital
payment: Intended and
unintended consequences
Francesc Cots, Pietro Chiarello,
Xavier Salvador, Xavier Castells
and Wilm Quentin1
6.1 Introduction
Almost 30 years after the fi rst introduction of a diagnosis-related group (DRG)-
based hospital payment system in the United States in 1983 (Fetter, 1991),
DRG systems have become the basis for hospital payment in most European
countries, and in many other countries around the world (Kimberly et al., 2008).
In fact, as illustrated in Chapter 2, one of the main purposes of the use of
DRG systems in all countries discussed in this volume is to enable DRG-based
hospital payment. Figure 6.1 illustrates the basic set-up of DRG-based hospital
payment systems: (1) a patient classifi cation system (PCS) is used to group
patients with similar clinical characteristics and relatively homogeneous
resource consumption into DRGs (see Chapter 4); (2) some kind of hospital cost
information is used to determine DRG weight levels, usually at (about) the
average treatment costs of patients falling within a specifi c DRG (see Chapter 5);
(3) DRG weights are converted into monetary values and the payment rate may
be adjusted for structural (teaching status, region) and further resource-
consumption variables (length of stay, utilization of high-cost drugs or services);
before (4) hospitals are paid on the basis of the number and type of DRGs that
they produce.
DRG-based hospital payment systems provide a specifi c set of incentives that
is different from other hospital payment systems, and the popularity of DRG-
based systems is related to the fact that they are thought to have a number of
(predominantly desirable) effects on quality and effi ciency, which are discussed
76 Diagnosis-Related Groups in Europe
in Chapters 7 and 8 of this volume. Yet, since the introduction of the fi rst DRG-
based hospital payment system, discussions regarding negative or ‘unintended’
consequences of these systems have persisted in the United States (Lave, 1989;
Ellis, 1998) as well as in Europe, following the adoption of DRGs for hospital
payment (Böcking et al., 2005; Steinbusch et al., 2007; Farrar et al., 2009).
This chapter focuses on DRG-based hospital payment systems in the 12 coun-
tries included in this book, and on the main incentives generally attributed to
these systems, which may generate both intended and unintended conse-
quences. The chapter starts with an overview of how the payment rate is deter-
mined in DRG-based hospital payment systems (section 6.2). This is followed in
section 6.3 by an introduction to the two main models of DRG-based hospital
payment; namely, DRG-based case payment and DRG-based budget allocation.
Section 6.4 provides a theoretical discussion of the main incentives attributed
to the most basic model of DRG-based hospital payment and the related poten-
tial intended and unintended consequences. In section 6.5 we discuss how the
countries analysed attempt to overcome some of the unintended consequences
by modifying the basic model of DRG-based payment. Section 6.6 summarizes
the fi ndings and concludes that the incentives related to DRG-based payment
systems have the potential to contribute to achieving the intended conse-
quences, as long as they succeed at avoiding the unintended ones by imple-
menting adequate control measures.
Figure 6.1 DRG-based hospital payment within the framework of this book
DRG-based hospital payment: Intended and unintended consequences 77
6.2 Paying hospitals on the basis of DRGs:
Determining the payment rate
As illustrated in Figure 6.1, the hospital payment rate in DRG-based hospital
payment systems is determined on the basis of the DRG weights and their
conversion into monetary values. In general, the aim is that DRG weights are
set at about the average costs of treating patients within a DRG. However,
countries differ in terms of how they express DRG weights, usually calculated
on the basis of information about average costs (see Chapter 5), or alternatively,
a ‘best practice’ approach (see Chapter 7). Table 6.1 shows the distribution of
the three main approaches that prevail in the 12 countries included in this
book: (1) relative weights, (2) raw tariffs, and (3) scores. Each of these approaches
corresponds to a specifi c monetary conversion method. The applicability of
DRG weights and monetary conversion factors differs between countries and
allows for adjusting hospital payments according to structural factors and/or
national or regional priorities.
6.2.1 DRG weights and their applicability in
12 European countries
Most countries calculate DRG relative weights or use imported and adapted
relative weights from other countries. The idea behind using DRG relative
weights is that the weight of a DRG is expressed in relation to the average
treatment costs of all cases in a country. In greatly simplifi ed terms, DRG relative
weights are computed by dividing the average costs of cases falling within a
DRG through the average treatment costs of all cases in a country. Consequently,
a DRG with a relative weight of one implies that average treatment costs of
patients falling into that DRG are equal to the average treatment costs of all
cases within the country. If relative weights are imported from abroad, DRG
relative weights do not necessarily refl ect national practice patterns. However,
by adapting relative weights to national cost data (as in Ireland and Portugal,
for example), countries aim to ensure that the adapted relative weights are
linked to national resource-consumption patterns and that DRG weights can be
interpreted in terms of above/below-average treatment costs.
England, France, the Netherlands and Spain calculate raw tariffs. These differ
from relative weights in that they are expressed in monetary terms. In general,
raw tariffs are calculated directly from the average treatment costs of patients
within a DRG, even though they may be adjusted for infl ation (as in England)
or for national global budget control purposes (as in France). In Spain, raw
tariffs are calculated on the basis of imported internal relative weights and
national cost-accounting data. Interestingly, although the raw tariff is already
expressed in monetary terms, it is not necessarily equal to the actual hospital
payment, which is determined after further adjustments (discussed in more
detail in the sections that follow).
Austria and Poland are the only two countries that express DRG weights as
scores. The difference between scores and raw tariffs is that the score is not
expressed in monetary terms but as a number of points. At the same time, and
T
a
b
le
6
.1
D
R
G
w
ei
g
h
ts
,
m
o
n
et
ar
y
c
o
n
v
er
si
o
n
a
n
d
t
h
ei
r
ap
p
li
ca
b
il
it
y
i
n
1
2
E
u
ro
p
ea
n
c
o
u
n
tr
ie
s
C
ou
n
tr
y
D
R
G
w
ei
gh
t
(u
n
it
)
A
pp
li
ca
b
il
it
y
of
D
R
G
w
ei
gh
t
M
on
et
a
ry
c
on
ve
rs
io
n
A
pp
li
ca
b
il
it
y
of
c
on
ve
rs
io
n
r
a
te
A
u
st
ri
a
Sc
o
re
N
at
io
n
w
id
e
(I
m
p
li
ci
t)
P
o
in
t
v
al
u
e
D
ep
en
d
in
g
o
n
s
ta
te
E
n
g
la
n
d
R
aw
t
ar
if
f
N
at
io
n
w
id
e
(s
ep
ar
at
e
ta
ri
ff
s
fo
r
em
er
g
en
ci
es
,
el
ec
ti
v
e
ca
se
s,
d
ay
c
as
es
,
ch
il
d
re
n
,
o
rt
h
o
p
ae
d
ic
a
ct
iv
it
y
)
M
ar
k
et
f
o
rc
es
f
ac
to
r
H
o
sp
it
al
-s
p
ec
ifi
c
E
st
o
n
ia
R
el
at
iv
e
w
ei
g
h
t
N
at
io
n
w
id
e
B
as
e
ra
te
N
at
io
n
w
id
e
Fi
n
la
n
d
R
el
at
iv
e
w
ei
g
h
t
N
at
io
n
al
(
8
d
is
tr
ic
ts
),
D
is
tr
ic
t-
sp
ec
ifi
c
(
5
d
is
tr
ic
ts
)
B
as
e
ra
te
H
o
sp
it
al
-s
p
ec
ifi
c
Fr
an
ce
R
aw
t
ar
if
f
N
at
io
n
w
id
e
(s
ep
ar
at
e
ta
ri
ff
s
fo
r
p
u
b
li
c
an
d
p
ri
v
at
e
h
o
sp
it
al
s)
(1
)
R
eg
io
n
al
a
d
ju
st
m
en
t
(2
)
T
ra
n
si
ti
o
n
c
o
ef
fi
ci
en
t
(u
n
ti
l
2
0
1
2
)
(1
)
R
eg
io
n
-s
p
ec
ifi
c
(2
)
H
o
sp
it
al
-s
p
ec
ifi
c
G
er
m
an
y
R
el
at
iv
e
w
ei
g
h
t
N
at
io
n
w
id
e
B
as
e
ra
te
St
at
e-
w
id
e
Ir
el
an
d
(A
d
ap
te
d
)
R
el
at
iv
e
w
ei
g
h
t
(l
o
ca
ll
y
r
ef
er
re
d
to
a
s
re
la
ti
v
e
v
al
u
e)
N
at
io
n
w
id
e
(s
ep
ar
at
e
w
ei
g
h
ts
f
o
r
p
ae
d
ia
tr
ic
h
o
sp
it
al
s)
B
as
e
ra
te
s
(1
)
Sp
ec
ifi
c
t
o
o
n
e
o
f
fo
u
r
h
o
sp
it
al
p
ee
r
g
ro
u
p
s
(2
)
H
o
sp
it
al
-s
p
ec
ifi
c
N
et
h
er
la
n
d
s
R
aw
t
ar
if
f
N
at
io
n
w
id
e
(6
7
%
o
f
D
R
G
s)
,
h
o
sp
it
al
-
sp
ec
ifi
c
(
3
3
%
o
f
D
R
G
s)
D
ir
ec
t
(n
o
c
o
n
v
er
si
o
n
)
N
o
t
ap
p
li
ca
b
le
P
o
la
n
d
Sc
o
re
N
at
io
n
w
id
e
(s
ep
ar
at
e
ta
ri
ff
s
fo
r
em
er
g
en
ci
es
,
el
ec
ti
v
e
ca
se
s,
d
ay
c
as
es
)
P
o
in
t
v
al
u
e
N
at
io
n
w
id
e
P
o
rt
u
g
al
(A
d
ap
te
d
)
R
el
at
iv
e
w
ei
g
h
t
N
at
io
n
w
id
e
B
as
e
ra
te
H
o
sp
it
al
p
ee
r
g
ro
u
p
Sp
ai
n
(
C
at
al
o
n
ia
)
(1
)
(A
d
ap
te
d
)
R
aw
t
ar
if
f
(A
P
-D
R
G
s)
;
(2
)
(I
m
p
o
rt
ed
)
R
el
at
iv
e
w
ei
g
h
t
(C
M
S-
D
R
G
s)
(1
)
N
at
io
n
w
id
e
(A
P
-D
R
G
s)
(2
)
R
eg
io
n
-w
id
e
(C
M
S-
D
R
G
s)
(1
)
D
ir
ec
t
(n
o
c
o
n
v
er
si
o
n
)
(2
)
B
as
e
ra
te
(1
)
N
o
t
ap
p
li
ca
b
le
(2
)
R
eg
io
n
-w
id
e
(C
M
S-
D
R
G
s)
Sw
ed
en
R
el
at
iv
e
w
ei
g
h
t
N
at
io
n
w
id
e,
c
o
u
n
ty
-s
p
ec
ifi
c
(
so
m
e
co
u
n
ti
es
)
B
as
e
ra
te
C
o
u
n
ty
-s
p
ec
ifi
c
So
u
rc
e:
B
as
ed
o
n
t
h
e
co
u
n
tr
y
-s
p
ec
ifi
c
c
h
ap
te
rs
i
n
P
ar
t
T
w
o
o
f
th
is
v
o
lu
m
e
(m
o
st
r
ec
en
t
in
fo
rm
at
io
n
a
v
ai
la
b
le
).
DRG-based hospital payment: Intended and unintended consequences 79
in contrast to relative weights, the score does not relate the DRG weight of one
DRG to the average treatment costs of all cases within a country. Therefore,
unlike relative weights, scores and raw tariffs do not facilitate the calculation of
the casemix indices (CMIs) of hospitals, which are frequently used in countries
using a relative/weight approach to comparing differences in patient populations
across hospitals.
In the majority of countries, the same DRG weights apply to all hospitals and
regions, nationwide. However, in Ireland, a separate set of relative weights is
calculated for paediatric hospitals, and in England, different tariffs exist for
(amongst others) day cases, emergencies, elective cases, children, and ortho-
paedic activity. In Finland and Sweden, national relative weights exist as well as
district- or county-specifi c relative weights, and districts or counties are free to
choose the set of relative weights that best suits their needs. In France, separate
tariffs are calculated for public and private hospitals, as physicians’ salaries are
included only in DRG-based payments for public hospitals. In the Netherlands,
national tariffs apply only for 67 per cent of DRGs, while 33 per cent of tariffs
are the result of negotiations between hospitals and insurers. In Spain, raw
tariffs for All Patient (AP-)DRGs apply only for patients treated in regions in
which they are not ordinarily resident, while relative weights of Centers for
Medicare and Medicaid Services (CMS-)DRGs apply only to the autonomous
community of Catalonia. Furthermore, even if the DRG weight is the same for
all hospitals and regions within the country, monetary conversion and ad-
justments may result in different payments for hospitals in different regions (see
subsection 6.2.2).
6.2.2 Monetary conversion and structural adjustments
Table 6.1 also shows that all countries except for the Netherlands and Spain (AP-
DRGs) multiply DRG weights with some sort of monetary conversion factor in
order to determine actual DRG-based hospital payments. All countries that use
a ‘relative weight’ approach multiply the relative weight with a so-called ‘base
rate’. However, signifi cant differences exist between countries in terms of the
applicability of the base rate to different hospitals. Estonia is the only country
in which the same base rate is applied to all hospitals, nationwide. In Finland
and Sweden, base rates are calculated specifi cally for every hospital according to
predetermined global budgets and the expected hospital activity. In Germany,
different base rates are negotiated between the self-governmental bodies (most
importantly the social health insurance associations and the hospitals) for every
Land (federal state). In Ireland, an increasing share of the DRG-based budgets is
determined on the basis of a hospital peer-group (for example, major teaching
hospitals, other hospitals, paediatric hospitals) base rate, which ensures that
similar hospitals are grouped together for payment purposes; the hospital-
specifi c base rate, which sheltered hospitals from excessive budget cuts during
the introduction period, currently determines only 20 per cent of hospitals’
budgets. In Portugal, a similar approach is taken, as hospitals are also paid on
the basis of a base rate that is specifi c to the hospitals’ peer group. In Catalonia,
the base rate is the same for the entire region.
80 Diagnosis-Related Groups in Europe
In England and France, raw tariffs are multiplied by adjustment factors, which
ensure that certain structural characteristics (such as higher salary levels), are
taken into account in DRG-based payments. In addition, DRG weights in France
are still adjusted for a transition coeffi cient, refl ecting historical cost patterns of
hospitals in order to shelter them from excessive budget cuts, which is similar to
the approach used in Ireland and other countries during the introduction period
of DRG-based payments. In the Netherlands and Spain (for nation-wide use of
AP-DRGs) raw tariffs are not adjusted but are directly used for payment.
In Poland, DRG scores are multiplied by a point value that is the same for the
entire country. In Austria, the point value is implicit and state specifi c, as point
values are not published and hospital budget allocations are determined on the
basis of state-specifi c rules. For example, some states infl ate the scores of teach-
ing hospitals or hospitals located in areas with higher salary levels. The implicit
point value is then determined by dividing the entire inpatient budget available
for DRG-based payment within a state through the total (adjusted) scores pro-
duced by all hospitals. In Poland, the point value depends on the available
national hospital budget and is determined through negotiations between the
National Health Fund (NFZ), the Ministry of Health and representatives of asso-
ciations of medical professionals.
6.3 DRG-based hospital payment
After monetary conversion and structural adjustments, hospitals in all 12
countries are paid – to at least some extent – on the basis of DRGs. Table 6.2
shows the distribution of DRG-based case payment and DRG-based budget
allocation systems across countries. It also indicates the percentage of hospital
revenues related to DRGs in acute care hospitals, and specifi es further payment
components of the hospital payment system in each country.
Most countries included in this book use a variant of DRG-based case payment
systems, whereby each discharged patient is grouped into an applicable DRG,
and hospitals receive a payment per case that is determined by the weight of
that DRG (after monetary conversion and relevant adjustments). Several other
countries use some kind of a DRG-based budget allocation system, whereby the
available regional or national hospital budget is distributed to individual
hospitals on the basis of the DRGs that they produced during one of the
previous years, or that they are expected to produce in the next year. In these
cases, the casemix (that is, the sum of the weights of all DRGs produced by a
hospital) and the CMI (the casemix divided by the number of discharges) are
usually the determining factors. In addition, some countries with DRG-based
case payment systems – such as Germany or Finland – use DRGs to negotiate
global hospital budgets, which limit (to a certain degree) the total amount of
money that hospitals can earn from DRG-based case payments.
Table 6.2 shows that DRG-based hospital payment accounts for the majority
of hospital revenues in all countries except for Spain (that is, Catalonia) and
Estonia. Consequently, the incentives related to these payment systems are par-
ticularly important. However, it should be borne in mind that in most coun-
tries, psychiatric, rehabilitation and long-term care hospitals are not fi nanced
on the basis of DRGs, although several countries plan to extend their DRG
DRG-based hospital payment: Intended and unintended consequences 81
systems beyond the acute care hospital sector (see Chapter 4). In addition, the
hospital payment system in almost all countries includes other payment com-
ponents aside from DRG-based hospital payment, such as global budgets and
additional payments for certain activities or cost categories. For example, in
some countries, DRG-based payments do not include capital costs (see Chapter
5), and almost all countries have additional payments for certain innovative
and high-cost services (see Chapters 4 and Chapter 9 of this volume), as well as
additional budgets for teaching and research or availability of emergency care.
To appreciate fully the discussion that follows regarding the incentives
associated with DRG-based payment systems, it is important to be aware of the
differences between and within the two basic models of DRG-based hospital
payment. First, in theory, DRG-based case payment systems could provide
stronger incentives to hospitals than DRG-based budget allocation systems
Ta ble 6.2 DRG-based hospital payment for acute care hospitals
Country DRG-based hospital payment
model
% of hospital
revenues related
to DRGs
Other payment components
Austria DRG-based budget allocation ≈ 96 Per diems
England DRG-based case payments ≈ 60 GB, additional payments
Estonia DRG-based case payments ≈ 39 FFS (33%), per diem (28%)
Finland In 13 out of 21 districts:
DRG-based case payments
(within GB)
Varies Varies
France DRG-based case payments,
MLPC
≈ 80 GB, additional payments
Germany DRG-based case payments
(within GB)
≈ 80 GB, additional payments
Ireland DRG-based budget allocation ≈ 80 GB, additional payments
Netherlands DRG-based case payments
(within GB for 67% of
DRGs)
≈ 84 GB, additional payments
Poland DRG-based case payments,
MLPC
≥ 60 GB, additional payments
Portugal (1) DRG-based budget
allocation (NHS)
(2) DRG-based case payments
(health insurance)
≈ 80 Additional payments
Spain
(Catalonia)
DRG-based budget allocation
(Catalonia)
≈ 20 GB (based on structural
index), FFS, additional
payments
Sweden DRG-based case payments
with volume ceilings
or GBs (region-specifi c
allocation methods)
Varies Varies
Source: Based on the country-specifi c chapters in Part Two of this volume (most recent
information available).
Notes: FFS: fee-for-service (payment); GB: global budget; MLPC: macro-level price control.
82 Diagnosis-Related Groups in Europe
because the link between hospital service provision and payment is more direct
and transparent: hospitals know how much money they can expect if providing
a specifi c set of services to a specifi c patient. In contrast, in DRG-based budget
allocation systems, hospitals only know that the provision of a specifi c set of
services to a specifi c patient in one year will increase the DRG-based budget for
one of the following years, but the exact size of the payment remains unknown:
it depends on the number of DRGs produced by other hospitals and on the
available budget in the following year. Consequently, it is more diffi cult for
hospitals to predict whether the provision of certain DRGs is profi table or not.
Second, incentives relating to DRG-based case payment systems can be more
or less intensive, depending on country-specifi c modifi cations. For example, in
Germany or the Netherlands, where DRG-based case payment systems are
operated within global budgets, the incentives to increase hospital activity are
less strong than in England, where hospital activity is not (yet) limited by global
budgets or volume thresholds. Furthermore, the situation in countries with
DRG-based case payment systems that operate within global budgets differs
depending on whether or not hospitals are allowed to exceed the budgets. For
example, in Germany, hospitals are allowed to exceed the budget but are paid
at a reduced rate for those cases that are treated in addition to the negotiated
budget. In the Netherlands, however, hospitals must pay back at the end of the
year all revenue from DRG-based case payments that they received in excess of
the global budget.
Third, independent of the model of DRG-based hospital payment, the strength
of the theoretical incentives (see section 6.4) depends on how the monetary
conversion rate is determined. For example, on the one hand, in Poland, where
a nationally uniform monetary conversion rate is used, hospitals face strong
incentives to reduce their treatment costs to below the DRG payment rate. On
the other hand, in Finland, hospitals are paid according to a hospital-specifi c
payment rate, as the base rate is determined by dividing the negotiated hospital
budget by the predicted activity. Consequently, there are no incentives for
Finnish hospitals to lower their costs to the level of treatment costs in other
hospitals – and even less so since any potential defi cits accruing to hospitals are
compensated by the municipalities, which are both the purchasers and providers
of hospital care.
6.4 DRG-based hospital payment in theory:
Incentives and their consequences
As already mentioned, the principal reason for the popularity of DRG-based
hospital payment systems is that they are thought to have predominantly desir-
able effects on hospital effi ciency and quality. In general, there are three main
incentives attributed to DRG-based hospital payment systems (Lave, 1989).
Hospitals are incentivized (1) to reduce costs per treated patient, (2) to increase
revenues per patient, and (3) to increase the number of patients. Table 6.3 sum-
marizes these basic incentives of DRG-based hospital payment systems, pres-
ents the most important response strategies of hospitals, and indicates whether
these imply positive or negative effects on effi ciency and quality.
DRG-based hospital payment: Intended and unintended consequences 83
Tab le 6.3 Incentives of DRG-based hospital payment systems and their effects on
quality and effi ciency
Incentives of
DRG-based
hospital
payment
Strategies of hospitals Effects
1. Reduce
costs per
patient
a) Reduce length of stay
• optimize internal care pathways
• transfer to other providers
– improve coordination/integration
with other providers
– transfer/avoidance of unprofi table
cases (‘dumping’ or ‘cost-shifting’)
• inappropriate early discharge
(‘bloody discharge’)
• quality ↑, effi ciency ↑
• quality ↑, effi ciency ↑
• quality ↓
• quality ↓
b) Reduce intensity of provided
services
• avoid delivering unnecessary
services
• substitute high-cost services with
low-cost alternatives (labour/capital)
• withhold necessary services
(‘skimping/undertreatment’)
• effi ciency ↑
• effi ciency ↑
• quality ↓
c) Select patients
• specialize in treating patients
for which the hospital has a
competitive advantage
• select low-cost patients within
DRGs (‘cream-skimming’)
• effi ciency ↑, quality ↑
• effi ciency ↓
2. Increase
revenue
per
patient
a) Change coding practice
• improve coding of diagnoses and
procedures
• fraudulent reclassifi cation of
patients, e.g. by adding inexistent
secondary diagnoses (‘up-coding’)
• quality↑
• effi ciency ↓
b) Change practice patterns
• provide services that lead to
reclassifi cation of patients into
higher paying DRGs (‘gaming/
overtreatment’)
• effi ciency ↓, quality ↓
3. Increase
number of
patients
a) Change admission rules
• reduce waiting list
• split care episodes into multiple
admissions
• admit patients for unnecessary
services (‘supplier-induced demand’)
• effi ciency ↑
• effi ciency ↓↑, quality ↓↑
• effi ciency ↓
b) Improve reputation of hospital
• improve quality of services
• focus efforts exclusively on
measurable areas
quality ↑
quality ↓↑
84 Diagnosis-Related Groups in Europe
Taking a step back: how are these incentives generated? The fi rst
main incentive (that is, to reduce costs per case) is generated because, in
mathematical terms, the most basic DRG-based hospital (case) payment system
is one in which hospital revenue for treating a patient falling into a specifi c
DRG (R1) is determined by the fi xed payment rate per DRG
1
( p̂
1
) (Ellis &
McGuire, 1986):
R1 = p̂
1
(1)
Figure 6.2 provides a simplifi ed graphical illustration of the relationship
between costs, length of stay, revenue of hospitals, and the incentives related to
this basic DRG-based hospital payment system for a hypothetical standard
patient in DRG
1
. As hospital revenue (R1) per patient in DRG
1
does not depend
on the costs of service provision, hospitals are strongly incentivized to reduce
their costs below the payment rate ( p̂
1
).
The three most important response strategies for hospitals trying to reduce
costs per case are (Berki, 1985; Miraldo et al., 2006): (1a) to reduce the length of
stay, (1b) to reduce the intensity of the services provided, and (1c) to select
patients for whom hospitals can provide care at costs that are below the DRG
payment rate (not shown in Figure 6.2). On the one hand, reducing the length
of stay and the intensity of services are intended effects of DRG-based hospital
payment systems because both can contribute to increased effi ciency of hospital
care. For example, length of stay can be reduced by optimizing internal care
pathways (Kahn et al., 1990); and intensity of services may be reduced by not
providing unnecessary services. However, on the other hand, reducing length
of stay could result in inappropriately early (‘bloody’) discharges and service
intensity could be reduced to a level at which necessary services begin to be
withheld from patients (‘skimping’; Ellis, 1998), both leading to reductions in
quality (see Table 6.3). Similarly, the selection of patients can have intended
and unintended consequences. On the one hand, hospitals could specialize in
treating those patients for whom they have a competitive advantage (for
example, better qualifi ed personnel or better care pathways), which could lead
Figure 6.2 Selected incentives of DRG-based hospital payment for a hypothetical
standard patient
DRG-based hospital payment: Intended and unintended consequences 85
to greater effi ciency and higher quality. However, on the other hand, there is
also the danger that hospitals engage in ‘cream-skimming’ (Levaggi &
Montefi ori, 2003; Martinussen & Hagen, 2009); that is, they attempt to admit
only those patients within each DRG that can be expected to have costs below
the payment rate (for example, by selecting patients without co-morbidities, if
these are not adequately accounted for in the DRG system) or that they ‘dump’
unprofi table patients by transferring them to other providers or avoiding them
altogether (Ellis, 1998; Newhouse & Byrne, 1988).
The second main incentive of basic DRG-based hospital payment systems –
that is, to increase revenues per case (see Table 6.3) – can be achieved by
hospitals through one of two strategies: (2a) changing coding practices, or (2b)
changing practice patterns. As illustrated in Figure 6.2, the aim of both strategies
is to reclassify patients into a different DRG (DRG
2
) with an associated higher
payment rate ( p̂
2
). While more thorough coding of secondary diagnoses and
procedures is an intended effect of the introduction of DRG-based hospital
payment systems, the attempt by hospitals to increase revenues through
fraudulent coding practices – such as adding inexistent secondary diagnoses or
inverting primary and secondary diagnoses (known as ‘up-coding’ or ‘DRG
creep’; Simborg, 1981; Steinbusch et al., 2007; Silverman & Skinner, 2004) – is
not intended because it leads to unjustifi ed payments to hospitals. Furthermore,
changed practice patterns would be an unintended consequence if hospitals
provide additional (unnecessary) procedures that lead to the reclassifi cation
of patients into higher paying DRGs (‘gaming/overtreatment’). However, this
should be relevant only if these procedures can be performed at marginal costs
that are below the level of the additional obtainable revenue as a result of the
reclassifi cation.
Finally, because hospital revenue in basic DRG-based hospital payment sys-
tems is determined simply by multiplying activity in each DRG by the fi xed
payment per DRG, the third main incentive for hospitals is to increase the
number of admitted patients. Again, an increase in activity can be both an
intended and an unintended consequence of the introduction of this type of
hospital payment. On the one hand, if waiting lists existed under the old hos-
pital payment system, an increase in hospital activity is an intended conse-
quence that can contribute to increasing effi ciency of hospitals. On the other
hand, if activity is increased by admitting patients for services that could
be provided in outpatient settings, effi ciency is reduced. Furthermore, in com-
petitive environments, hospitals’ efforts to attract more patients may result in
strategies to improve the reputation of hospitals by providing higher quality
services, but could also lead to strategies that focus all efforts on improving
only those services that are visible to patients or measurable by quality assur-
ance programmes.
In summary, the intended and unintended consequences of DRG-based
hospital payment systems are deeply intertwined. Most importantly, they are
related to the fact that payment in these systems is independent of the costs of
care provided to a specifi c patient. This becomes particularly problematic in the
context of health care markets, in which information asymmetries are highly
prevalent and make it diffi cult for payers to monitor and control providers’
activity or behaviour (Lave, 1989). Furthermore, unintended consequences are
86 Diagnosis-Related Groups in Europe
related to the fact that DRG-based hospital payment systems can be interpreted
as providing highly powerful incentives (Frant, 1996) because hospital payment
depends directly on provider behaviour.
6.5 DRG-based hospital payment in practice: Modifi cations
and instruments to avoid unintended consequences
In practice, DRG-based hospital payment systems in the 12 countries included
in this book are far more complicated than the basic model of DRG-based
hospital payment presented in the previous section. As already mentioned,
different models of DRG-based hospital payment systems, selective applicability
of DRG weights and monetary conversion and adjustment factors, and structural
payment adjustments modify the basic incentives of DRG-based payments.
This section focuses more closely on the explicit attempts of the 12 countries to
avoid and control unintended consequences.
6.5.1 Fairness of payment: assuring adequate payment
for outliers and high-cost services
While DRG-based hospital payment systems can be considered to provide ade-
quate reimbursement for the average patient within each DRG, they overpay
hospitals for patients with below-average resource consumption and underpay
for patients with above-average costs. In general, most of these differences are
compensated automatically, as relatively more expensive cases within a DRG are
compensated by cheaper cases within the same DRG, and even unprofi table
DRGs may be compensated by highly profi table DRGs within the same hospital.
However, ensuring that DRGs comprise cases with relatively homogeneous costs
has been a major concern in all countries, as evidenced by the increasingly large
number of DRGs in all systems (see Chapter 2). On the one hand, if DRG sys-
tems adequately account for differences between patients (by considering all
relevant secondary diagnoses) and necessary treatments (by considering all rel-
evant procedures), the incentives for certain unintended consequences, such
as cream-skimming and skimping/undertreatment, could be greatly reduced.
On the other hand, refi ned DRG systems with more narrowly defi ned DRGs
also increase the scope for other unintended consequences, such as up-coding
(if DRGs are defi ned on the basis of classifi cation criteria that are easy to
manipulate) and gaming/overtreatment (if procedural classifi cation criteria
introduce strong incentives to deliver certain services) (Hafsteinsdottir &
Siciliani, 2010).
Yet, in spite of the continuous refi nement of DRG systems, DRGs in all sys-
tems incorporate patients that require much more resources than most patients
belonging to the same DRG. These high-cost ‘outlier’ cases often account for a
sizeable share of total hospital costs and consequently have a strong infl uence
on the average costs of cases within a DRG (Cots et al., 2003). If DRG weights
were calculated based on the average costs of patients within a DRG, including
the outlier cases, this would lead to hospitals being overpaid for the majority of
DRG-based hospital payment: Intended and unintended consequences 87
patients. Furthermore, if outlier cases were not paid for separately, hospitals
would experience particularly strong incentives to avoid these high-cost cases
(‘dumping’), or to discharge them inappropriately early (‘bloody’ discharge).
Consequently, most of the countries analysed in this book have developed
mechanisms to identify outlier cases and to pay hospitals separately for the
extra costs of treating such patients. Table 6.4 shows that most countries defi ne
outlier cases on the basis of a length-of-stay threshold (a certain number of days
beyond which cases are considered outliers), as cost data are usually available
only for a sample of patients across the country (see Chapter 5). The Nordic
countries (Estonia, Finland, and Sweden) are an exception as they defi ne outlier
cases on the basis of costs. However, while all countries (except for the
Netherlands) defi ne outliers, the trimming methods determining the outlier
threshold differ. They are either based on a variant of the interquartile method
or the parametric method (Schreyögg et al., 2006), leading to varying percentages
of all cases being considered outliers.
Figure 6.3 illustrates how hospital payment systems in most countries ensure
adequate payment for outlier patients. Most often, the DRG-based payment
rate is increased for long-stay outlier cases by a surcharge that depends on the
number of days that patients were in hospital beyond the specifi ed threshold.
In Estonia and Finland, where outliers are defi ned on the basis of costs, the
extra costs of outlier patients are reimbursed directly through a fee-for-service
system. In Catalonia and the Netherlands, hospitals do not receive surcharges
Tab le 6.4 Defi nition of outliers and associated deductions/surcharges
Defi nition of outliers
(trimming method)
Outliers as
% of total
cases
Outlier payment
Deductions/
payments
Surcharges
Austria LOS (interquartile) ~ 12–15 Per day Per day
England LOS (interquartile) 7 No (but short-
stay tariff)
Per day
Estonia Cost (parametric) 9 ? FFS
Finland Cost (parametric) 5 No FFS
France LOS (interquartile) 0.4 (public
hospitals)
Per day Per day
Germany LOS (parametric) 22 Per day Per day
Ireland LOS (parametric) 6 Per day Per day
Netherlands – – Not applicable Not
applicable
Poland LOS (interquartile) ~ 2 No (but short-
stay tariff)
Per day
Portugal LOS (interquartile) – Per day Per day
Spain (Catalonia) LOS (interquartile) 5 No No
Sweden Cost/LOS (parametric) 5 Varies varies
Source: Based on the country-specifi c chapters in Part Two of this volume (most recent
information available).
88 Diagnosis-Related Groups in Europe
for outlier cases. In Catalonia, the extra costs of outlier cases are supposed to be
taken into account in the structural adjustments that determine the majority of
hospital budgets.
In the Netherlands, the problem of outlier cases is dealt with very differently:
if a patient has more than one diagnosis requiring treatment, this additional
‘diagnosis–treatment combination’ (the Dutch equivalent of a DRG, known as
a DBC) triggers an additional DRG-based payment to the hospital. Interestingly,
in Finland (and, prior to the recent update, also in England), a similar approach
exists, whereby hospitals can assign more than one DRG for patients that were
treated in several departments during one hospital stay.
Furthermore, several countries also determine lower length-of-stay outlier
thresholds, which are sometimes an explicit attempt at avoiding inappropriately
early (‘bloody’) discharges. These countries calculate a reduced payment rate for
patients that are discharged prior to the lower length-of-stay threshold, either by
deducting a certain amount from the standard DRG rate for each day that
patients are discharged before the lower length-of-stay threshold, or by calculat-
ing the sum of a minimum payment plus a certain daily rate. England and
Poland do not determine lower length-of-stay thresholds but have specifi c short-
stay (one-day) weights ( p̂
1–short
) for certain DRGs. In the Finnish and Swedish
versions of the NordDRG system, specifi c DRGs exist for day-care patients.
Aside from the issue of outlier cases, all DRG systems are confronted with the
problem that certain high-cost services are provided to a heterogeneous group
of patients that fall into different DRGs. As discussed in Chapter 3, most DRG
systems have instituted additional payment mechanisms for certain services
that cannot be assigned to a specifi c DRG. Consequently, these services are
exempt from the incentives that tend to apply to DRG-based hospital payment,
and this can therefore be interpreted as an attempt to avoid skimping/under-
treatment relating to these services. In addition, most countries have developed
similar payment mechanisms for certain innovative drugs and treatments that
are not adequately accounted for by their DRG systems (see Chapter 9).
Figure 6.3 Trimming and reimbursement for outliers
DRG-based hospital payment: Intended and unintended consequences 89
6.5.2 Asymmetry of information: Monitoring and controlling
unintended consequences
The asymmetry of information between providers and payers gives rise to
several unintended consequences of DRG-based hospital payment systems: for
example, payers do not necessarily know whether a specifi c patient was in need
of a specifi c procedure; whether the patient really needed to be admitted as an
inpatient; whether non-existing secondary diagnoses were coded; and whether
certain secondary diagnoses resulted from medical errors. Consequently, pro-
viders can perform procedures that lead to the reclassifi cation of patients into
higher paying DRGs (gaming/overtreatment); they can increase the volume of
admitted patients; they can up-code their patients; and they may receive higher
payments for providing services of poorer quality (see Chapter 8).
In order to control some of these unintended consequences, several countries
have implemented auditing systems that aim to reduce the asymmetry of infor-
mation. For example, in Germany, the regional medical review boards of the
sickness funds send teams to randomly selected hospitals to evaluate the coding
and treatment of patients by auditing patients’ medical records (MDS, 2011). In
2009, 12 per cent of all hospital cases were audited by the sickness funds, result-
ing in average costs of around €800 per audited case being recovered. In France,
1 per cent of hospital discharges were audited by the Regional Hospitalization
Agencies (ARH) in 2006, which found that 60 per cent of evaluated records had
some kind of coding error.
Other control mechanisms aim to limit the ability of hospitals to exploit the
asymmetry of information by determining global budgets or volume thresholds
(see Table 6.2), which ensure that hospitals do not increase their activity beyond
predetermined limits. Furthermore, in order to control frequent readmissions,
Germany and England fi nancially penalize hospitals if patients are readmitted
for the same problem within 30 days after initial discharge: for these patients,
hospitals do not receive a second DRG-based payment. In addition, if countries
in Europe were to follow the example of the United States in obligating hospitals
to specify whether secondary diagnoses were present on admission (see Chapter
8), they would be able to differentiate between hospital-acquired (potentially
avoidable) conditions, and those that were beyond the control of hospitals.
Finally, the regular recalculation of DRG weights and monetary conversion
factors (see Chapter 9) reduces the ability of hospitals to benefi t from up-
coding: if all hospitals engage in up-coding, the recalculation of DRG weights
and monetary conversion factors will lead to reduced payment rates for previ-
ously higher paying DRGs. However, if some hospitals engage in up-coding and
others do not, the honest hospitals are likely to be penalized by reduced pay-
ment rates. Therefore, the readjustment of payment rates is useful as an effec-
tive mechanism for cost control, but it does not replace the need for thorough
auditing of hospital coding activities.
90 Diagnosis-Related Groups in Europe
6.5.3 The power of incentives: Reducing the share of
DRG-based payment in total hospital revenues
As illustrated in Table 6.2, DRG-based hospital payment systems never deter-
mine the entirety of hospital revenues.2 These other sources of revenue contrib-
ute to reducing the power of the incentives related to DRG-based hospital
payment, as hospitals can focus their efforts on maximizing revenues through
other strategies.
For example, in Spain (Catalonia), where DRG-based hospital payment
accounts for only 20 per cent of hospital revenues, the power of the incentives
related to DRG-based hospital payment are relatively weak. As pointed out by
Cots and colleagues (see Chapter 22, subsection 22.7.2, p. 420), ‘since hospital
revenues are mostly determined by their SRI [structural relative index], hospitals
are more likely to focus on introducing new and advanced technologies in
order to increase their SRI, rather than focusing on improving performance as
measured by DRGs’.
Furthermore, most countries that have introduced DRG-based hospital pay-
ment systems have phased in the systems over several years, with the size of total
hospital revenues related to DRG-based payment slowly increasing over time.
Consequently, the incentives of such hospital payment were minimal at fi rst, giv-
ing hospitals time to slowly adjust to the changing fi nancial environment.
6.6 Conclusions: Maximizing the intended and
avoiding unintended consequences
This chapter illustrates that DRG-based hospital payment systems in the 12
countries analysed for this volume do not conform to the basic model presented
in section 6.4. All countries’ hospital payment systems include other payment
components: the cumulative effect of structural adjustments of weights or mon-
etary conversion factors, of outlier payments, and additional payments can be
assumed to moderate the incentives associated with the basic model of DRG-
based hospital payment. The resulting intricately blended hospital payment
systems are more likely to contribute to achieving the societal objectives of
securing high-quality hospital care at affordable costs than any other hospital
payment mechanism alone (Ellis & Mcguire, 1986).
One advantage of determining hospital payment on the basis of DRGs is that
hospitals will be incentivized to increase their efforts in terms of coding of
diagnoses and procedures, which will contribute to generating better hospital
activity data. Yet, it is important to be aware that DRG-based hospital payment
systems should always be accompanied by thorough monitoring systems that
enable payers to reduce the information asymmetries, which would give rise to
unintended consequences of the incentives that are inherent to DRG-based
hospital payment systems. Furthermore, continuous refi nement of DRG systems
(see Chapter 4), and high-quality cost-accounting data (see Chapter 5) are
essential for optimizing DRG-based hospital payment systems, and for assuring
that payment rates are suffi ciently related to the costs of care.
DRG-based hospital payment: Intended and unintended consequences 91
However, the country experiences presented in this book also suggest that
governments do not need to be afraid of introducing DRG-based payment
systems – as long as they do so carefully and over extended time periods, slowly
increasing the share of DRG-based payments within the overall hospital
payment system. The large number of alternative models – ranging from DRG-
based case payment systems (operating within DRG-based negotiated budgets
or not) to DRG-based budget allocation systems – illustrate that countries can
tailor DRG-based hospital payment systems to the specifi c structure of their
existing hospital payment system. If the effects of DRG-based hospital payment
systems are carefully re-evaluated at regular intervals, ideally in close
collaboration with all actors concerned, the incentives of DRG-based payment
systems have the potential to contribute to achieving the intended consequences,
as long as the unintended ones can be adequately controlled through the
mechanisms described.
6.7 Notes
1 Even though some DRG-like PCSs do not defi ne DRGs in the strict sense of the word
(that is, groups are not diagnosis-related), this chapter uses the term DRGs to
summarize all groups of patients defi ned by DRG systems or DRG-like PCSs (for further
details see Chapter 4 or the relevant country-specifi c case study chapters in Part
Two).
2 In fact, the extraordinarily high share of hospital revenues appearing to be determined
by the DRG-based payment system in Austria is somewhat misleading, for example, as
the state-specifi c monetary conversion factors in several of the federal states adjust –
to a signifi cant extent – for the structural characteristics of hospitals.
6.8 References
Berki, S.E. (1985). DRGs, incentives, hospitals, and physicians. Health Affairs (Millwood),
4:70–6.
Böcking, W., Ahrens, U., Kirch, W., Milakovic, M. (2005). First results of the introduction
of DRGs in Germany and overview of experience from other DRG countries. Journal
of Public Health, 13:128–37.
Cots, F., Elvira, D., Castells, X., Sáez, M. (2003). Relevance of outlier cases in casemix
systems and evaluation of trimming methods. Health Care Management Science,
6:27–35.
Ellis, R.P. (1998). Creaming, skimping and dumping: provider competition on the
intensive and extensive margins. Journal of Health Economics, 17:537–55.
Ellis, R.P., McGuire, T.G. (1986). Provider behavior under prospective reimbursement –
cost-sharing and supply. Journal of Health Economics, 5:129–51.
Farrar, S., Yi, D., Sutton, M. et al. (2009). Has payment by results affected the way that
English hospitals provide care? Difference-in-differences analysis. British Medical
Journal, 339:1–8.
Fetter, R.B. (1991). Diagnosis-related groups – understanding hospital performance.
Interfaces, 21:6–26.
Frant, H. (1996). High-powered and low-powered incentives in the public sector. Journal
of Public Administration Research and Theory, 6(3):365–81.
92 Diagnosis-Related Groups in Europe
Hafsteinsdottir, E.J.G., Siciliani, L. (2009). DRG prospective payment systems: refi ne or
not refi ne? Health Economics, 19(10):1226–39.
Kahn, K.L., Rogers, W.H., Rubenstein, L.V. et al. (1990). Measuring quality of care with
explicit process criteria before and after implementation of the DRG-based prospective
payment system. Journal of the American Medical Association, 264:1969–73.
Kimberly, J.R., de Pouvourville, G., D’Aunno, T., eds (2008). The Globalization of Managerial
Innovation in Health Care. Cambridge: Cambridge University Press.
Lave, J.R. (1989). The effect of the Medicare prospective payment system. Annual Review
of Public Health, 10:141–61.
Lave, J.R. (2003). Developing a Medicare prospective payment system for inpatient
psychiatric care. Health Affairs (Millwood), 22:97–109.
Levaggi, R., Montefi ori, M. (2003). Horizontal and Vertical Cream-Skimming in the Health
Care Market. DISEFIN Working Paper, 11/2003. Rochester, NY: Social Science Research
Network (http://ssrn.com/abstract=545583, accessed 10 July 2011).
Martinussen, P.E., Hagen, T.P. (2009). Reimbursement systems, organizational forms and
patient selection: evidence from day surgery in Norway. Health Economics, Policy and
Law, 4:139–58.
MDS (2011). Abrechnungsprüfungen der MDK in Krankenhäusern sind angemessen, wirtschaftlich
und zielführend. Zahlen und Fakten der MDK-Gemeinschaft. Essen: Medizinischer Dienst
des Spitzenverbandes Bund der Krankenkassen e.V.
Miraldo, M., Goddard, M., Smith, P. (2006). The Incentive Effects of Payment by Results.
York: University of York Centre for Health Economics (CHE Research Paper 19).
Newhouse, J.P., Byrne, D.J. (1988). Did Medicare’s prospective payment system cause
length of stay to fall? Journal of Health Economics, 7:413–16.
Schreyögg, J., Stargardt, T., Tiemann, O., Busse, R. (2006). Methods to determine reimburse-
ment rates for diagnosis-related groups (DRG): a comparison of nine European
countries. Health Care Management Science, 9:215–23.
Silverman, E., Skinner, J. (2004). Medicare up-coding and hospital ownership. Journal of
Health Economics, 23:369–89.
Simborg, D.W. (1981). DRG creep: a new hospital-acquired disease. New England Journal of
Medicine, 304:1602–4.
Steinbusch, P.J., Oostenbrink, J.B., Zuurbier, J.J., Schaepkens, F.J. (2007). The risk of up-
coding in casemix systems: a comparative study. Health Policy, 81:289–99.
chapter s e v e n
DRG-based hospital payment
and effi ciency: Theory,
evidence, and challenges
Andrew Street, Jacqueline O’Reilly,
Padraic Ward and Anne Mason
7.1 Introduction
Diagnosis-related groups (DRGs) were fi rst used to pay hospitals in 1983 under
the Medicare Program in the United States. This development was born out of
a need to move away from an approach to hospital fi nancing based on fee-for-
service payments, which was seen as inherently ineffi cient and increasingly
expensive. Since then, DRG-based hospital payment has been widely adopted
internationally with the explicit objective of improving effi ciency, principally
because of its three overarching strengths, summarized here (see Chapter 2 for
further details).
1. By relating provider revenue directly to their workload, DRG-based hospital
payment offers greater transparency in the fi nancing of health care.
2. Payments are based on patient characteristics (predominantly demographic
and clinical). Fundamental to effective DRG-based hospital payment is an
accurate description of the type of patients treated (casemix).
3. DRG-based hospital payment is a form of ‘yard stick competition’, designed
to encourage greater effi ciency in the absence of market competition.
Concentrating on the third strength, this chapter considers the relationship
between DRG-based hospital payment and effi ciency from theoretical and
empirical perspectives. It thus fi rst discusses the concepts ‘effi ciency’ and ‘yard
stick competition’. Different hospital payment models are then compared in
section 7.2, with the intention of indicating in each case the incentives for
hospitals to pursue effi cient behaviour, particularly in terms of maximizing
94 Diagnosis-Related Groups in Europe
output and minimizing cost. The empirical evidence regarding the impact of
DRG-based hospital payment on effi ciency is reviewed in section 7.3 by look-
ing at studies that consider effi ciency as defi ned by economists and those that
focus on indicators of effi cient practice. Finally, the chapter outlines in section
7.4 some key challenges associated with the use of DRG-based hospital pay-
ment. While economic theory suggests that this hospital payment system
may provide incentives to encourage effi ciency, there could be barriers (such
as the system’s particular design and operation) to realizing these incentives
in practice.
‘Effi ciency’ is a widely used term that can have various meanings. Economists
make distinctions between technical, cost- and allocative effi ciency. Technical
effi ciency is defi ned as maximizing output for given input levels or, in this
context, treating as many patients as possible given the resources available.
Hospitals are cost-effi cient when they minimize costs for any given output level
(closely related to, but distinct from, technical effi ciency). Allocative effi ciency
can be defi ned for both outputs and inputs. The optimal output mix depends
on the value of each output, which requires judgements to be made on the
relative values of an appendectomy operation, a heart bypass and all other
health care interventions. The optimal mix of inputs depends on the relative
price of each input type, such as the salaries of doctors and nurses. Alongside
these economic terms, reference is often made to things thought to be indicative
of effi cient behaviour, which – in the hospital sector – might include the
number and type of patients treated, unit costs and length of stay, for example.
The extent to which DRGs contribute to achieving these forms of effi ciency
depends on how they are used for payment purposes, which helps to determine
the incentives hospitals face to pursue effi cient behaviour.
Yard stick competition is designed to encourage providers to reduce their
costs in contexts in which they face limited competitive pressure (Shleifer,
1985). If providers outperform others they benefi t directly by retaining the
generated fi nancial surplus; if they underperform they generate defi cits and,
ultimately, risk bankruptcy. All providers, including the most effi cient, are
incentivized to continually reduce costs. Yard stick competition is effective
when regulated prices are virtually independent of an individual provider’s
costs. Ideally, prices should refl ect the supply costs of effi cient providers,
determined across all providers within the same industry.
However, it is not straightforward to identify effi cient providers, especially if
the regulator is poorly informed about the provider’s costs, the exogenous
infl uences on these costs and the level of effort expended by the provider
(that is, their effi ciency). This asymmetry of information is particularly
problematic in the health care sector. In practice, price is often determined on
the basis of the average cost of all or a sample of providers (see Chapter 5),
although it may remain preferable to base it on ‘best practice’, set at the level of
effi cient high-quality providers that deliver care at costs below the average costs
in other hospitals. In England, such ‘best practice tariffs’ have recently been
introduced for certain high-volume areas (such as cholecystectomy, hip
fractures, cataracts, and stroke), with signifi cant unexplained variation in
quality of clinical practice and clear evidence of what constitutes best practice
(see Chapter 12).
DRG-based hospital payment and effi ciency 95
7.2 Hospital payment models
To understand the role of DRG-based hospital payment in enhancing effi ciency,
we compare (simplifi ed versions of) the three main forms of provider payment
models used in hospital fi nancing: cost-based reimbursement (also known as
fee-for-service payment), the global budget model, and DRG-based payment.
7.2.1 Cost-based or fee-for-service reimbursement
With cost-based reimbursement, payments to hospitals are based on the cost
incurred by each individual patient (plus potentially a profi t margin). The main
method of cost control is to specify a price list that details the unit payment for
each ‘item of service’ (for example, medication, X-ray, procedure). Hospitals
must therefore provide itemized bills for every patient treated, but there is no
incentive to limit what treatments they provide per insured patient – the more
diagnostic tests they perform, the more they get paid.
Stated formally, with cost-based reimbursement, hospital revenue (R
C
)
amounts to the number of patients treated (Q
i
) multiplied by the unit cost of
treatment (c
i
), where i indicates a particular patient:
∑
I
[Q
i × ci] + Z
C
R
C =
i = 1
(1)
Z
C
captures all other forms of revenue that hospitals receive, such as funds for
teaching and research. In the hospital sector, cost-based reimbursement was
primarily used in the United States during the 1960s and 1970s. This fuelled
escalation in health care costs as hospitals engaged in a ‘medical arms race’,
spending ever more on technologies and facilities to attract patients. Hospitals
knew that they could reclaim the costs from health insurance companies as
well as Medicare and Medicaid, the public insurance programmes for older
people and those with low incomes.
7.2.2 Global budgets
Cost control is one of the key advantages of global budget arrangements, which
have been used in many European health care systems, at least if the budget
constraint is credible and binding, and a separation exists between a payer (also
known as ‘purchaser’) and hospitals as providers of care. This division has
traditionally been present in social health insurance systems and, since the 1990s,
increasingly also in tax-funded systems (Robinson et al., 2005). A fi xed payment
is agreed in advance for a target level of activity – often specifi ed at specialty level.
Figure 7.1 illustrates the case in which a hospital receives a fi xed payment (R
–
) for
carrying out a pre-specifi ed volume of health service activity (Q).
Diffi culties arise if there are deviations from the pre-specifi ed volume. Some
form of penalty must be imposed if the volume is not achieved. If the pre-
specifi ed volume – usually defi ned as the number of hospital cases – is exceeded
96 Diagnosis-Related Groups in Europe
(‘overperformance’), the funder must either provide extra money or the hospital
will refuse to do extra work, thereby creating waiting lists. ‘Cost and volume’
contracts were developed to deal with these problems, and we return to a DRG-
based hospital payment form of these in the following subsection (7.2.3).
In more advanced global budget systems, activity is specifi ed by specialty.
Negotiations between the payer (whether this is a sickness fund or a health
authority) and the hospital revolve around the monetary value of each specialty-
level contract (B
s
) and how much activity (Q
s
) – usually defi ned as cases per
specialty – will be provided under this contract. The local specialty-level price
( p
s
) is the by-product of negotiations relating to total contract value and the
volume of activity. In formal terms, with the approach to fi nancing that uses
global budgets, hospital revenue comprises the sum of its contracts across
specialties (Bs):
∑ ∑RG = Bs + ZG = [Qs × ps] + ZG
S
s = 1 s = 1
S
(2)
where ZG captures all other forms of revenue that hospitals receive within the
framework of these payment arrangements.
7.2.3 DRG-based hospital payment
There are two key features of DRG-based hospital payment. (1) Activity is de-
scribed using DRGs rather than by specialty. For instance, payment is made for
a patient receiving a hip replacement rather than a patient treated in trauma
and orthopaedics. (2) The reimbursement per DRG is to a large extent fi xed in
Figure 7.1 Hospital revenues under global budgets
Source: Street et al., 2007.
DRG-based hospital payment and effi ciency 97
advance, as patient characteristics (especially the main diagnosis) determine
the DRG category with its fi xed ‘price’. As this constituted a major shift from
the ‘retrospective’ system of cost-based reimbursement, payment by DRGs was
thus termed ‘prospective’ in the United States – a term which was inappropriate
for systems with a global budget approach to fi nancing (where instead ‘activity-
based’ was used to describe the new payment system). As shown in Chapter 4,
the ‘prospective’ nature of DRGs is also weaker if to a large extent procedures
determine the DRG classifi cation. However, whether driven by diagnosis or pro-
cedure, the ‘price’ of a DRG is wholly or at least partially independent of an
individual provider’s costs (see Chapter 5). In many jurisdictions, this fi xed
price is set nationally rather than locally (see Chapter 6).
The relationship between the unit price and amount of activity can take a
number of forms. The main ones discussed here are:
1. linear payments, whereby the total payment equals price multiplied by
quantity;
2. mixed payments, whereby hospitals receive additional payments (often in
the form of lump sums) that are unrelated to activity levels;
3. marginal payments, whereby different prices are payable for the same type
of activity, depending on the quantity provided;
4. mixed and marginal payments, which are a combination of (2) and (3).
To understand the differences between these payment arrangements, we con-
sider how the total revenue received by a particular hospital is calculated.
Linear payments
With the most straightforward DRG-based hospital payment system, using
linear payments, hospital revenue is determined simply by multiplying activity
in each DRG (Q
j
) by the fi xed price per DRG ( p̂
j
), where j indicates a DRG:
∑
j = 1
]ˆ[Q
j
×RA = pj
J
(3)
Using this formulation, hospital revenue increases linearly with activity, as
illustrated in Figure 7.2. If the hospital treats Q
0
patients it receives revenue
amounting to only R
0
; if Q
1
patients are treated, revenue increases to R
1
. Clearly,
then, the revenue consequences of changes in activity are much more trans-
parent than within a system based on global budget arrangements.
Mixed payments
In almost all countries that have introduced DRG-based hospital payment, hos-
pital revenue is not determined solely by the number of patients treated. Hos-
pitals also receive revenue in other forms – for instance, to fund teaching and
research, to compensate for different geographical costs, or to cover some ele-
ment of the fi xed costs of providing services. It has been formally demonstrated
that such a ‘mixed’ hospital payment system creates better incentives than
‘pure’ systems (Ellis & McGuire, 1986; Barnum et al., 1995). The composition
98 Diagnosis-Related Groups in Europe
of these other revenue forms is a matter of negotiation between the payer (or
‘purchaser’) and the hospital sector, and may vary between hospitals, between
countries and over time. We defi ne ZA as capturing all these sources of revenue
not related to health care activity within the category ‘DRG-based hospital pay-
ment’. Then the revenue function becomes:
[Q
j
× p
j
] + ZA= ∑RA ˆ
j = 1
J
(4)
Figure 7.3 shows how this arrangement changes the relationship between
revenue and activity. Hospitals receive a fi xed amount ZA irrespective of the
number of patients treated. On top of this, hospitals receive revenue in line
with activity – but the unit price (p̂
j
) will be lower within the framework of this
‘mixed’ arrangement than within a ‘pure’ DRG-based system.
Marginal payments
DRG-based hospital payment can be modifi ed to allow incentives to vary
with supply. Quite often, DRG-based hospital payment is introduced to
stimulate activity beyond existing levels. But unconstrained growth in activity
may be undesirable. First, it undermines control over global expenditure –
under the simple formulation (see equation (3)), expenditure may simply
keep rising in line with activity. Second, hospitals may be able to expand
activity at low marginal cost – perhaps because they have underutilized
resources available – and, thus, this differential pricing may be used to exploit
economies of scale. If so, there is an argument for reducing the unit price for
additional activity.
Figure 7.2 Hospital revenue under ‘pure’ DRG-based hospital payment
Source: Street et al., 2007.
DRG-based hospital payment and effi ciency 99
The resulting arrangements are akin to ‘cost and volume’ contracts. Two
policy decisions are required.
1. A ‘target’ level of activity (Q
j
) should be defi ned for each hospital. In some
countries, this is based on historical activity. Agreeing a target is more
diffi cult where there is decentralized purchasing, such as in England, because
the target has to be agreed between each purchaser and provider.
2. The price that should be paid for activity above the target level must be
agreed – this is usually defi ned as some proportion (�) of the price up to the
target level. Formally the revenue function can be expressed as:
∑ ∑ +×+×= ])[(][ α ZAˆjp p̂jQj Qj − QjRA
j = 1 j = 1
J J
(5)
where (Qj – Qj) is non-negative and represents activity above the target and
α p̂
j
is the price paid per unit of additional activity. If α = 0.5, the price for
additional activity is 50 per cent of that paid for activity up to the target;
if α = 1, the same price is paid (in which case equations (4) and (5) are
equivalent); if α = 0, the marginal price is zero, so there is no incentive for
hospitals to undertake more activity; and if α > 1, additional payments are
higher than the base price, which creates very strong incentives to undertake
additional work. This may be justifi ed if marginal costs are high, as expansions
in activity require additional investment.
Figure 7.4 shows how revenue changes under this arrangement, when the
marginal price for additional activity is below the price for activity up to the
target; that is, 0 < α < 1. This results in a ‘kinked’ revenue function.
Figure 7.3 Hospital revenue under ‘mixed’ DRG-based hospital payment
Source: Street et al., 2007.
100 Diagnosis-Related Groups in Europe
7.2.4 Summary
Table 7.1 summarizes the main differences between the three hospital payment
systems. Of course, it is important to be cognisant that the hospital payment
systems implemented in practice are usually more complicated variants of the
simplifi ed models in the previous subsections.
The three models offer different incentives for achieving objectives relating
to activity levels, expenditure control, quality of care and the three types of
effi ciency (Table 7.2). Incentives to increase activity exist in both cost-based and
DRG-based hospital payment systems, with the relative strength of the incentives
depending on how closely the link between reimbursement and activity levels is
Figure 7.4 Hospital revenue under ‘mixed’ DRG-based hospital payment
with marginal pricing
Source: Street et al., 2007.
Table 7.1 Main differences across hospital payment systems
System Description of
patients
Amount of
activity
Price per unit
of activity
Basic formulation of
revenue function
Cost-based/fee-
for-service
Individual Unrestricted Item of
service ∑ [Qi × ci]RC =
i = 1
I
Global budget Per hospital/
specialty
Target/
historical
Locally
agreed
1 1
S S
s s= =
∑ ∑ [Qs × ps]RG = Bs =
‘Pure’ DRG-based
hospital payment
DRG Unrestricted Fixed
prospectively [Qj × pj]∑
=j 1
ˆRA =
J
Source: Adapted from Street et al., 2007.
DRG-based hospital payment and effi ciency 101
specifi ed (WHO, 2000; Langenbrunner et al., 2005; Moreno-Serra & Wagstaff,
2010). DRG-based hospital payment performs better than cost-based reimburse-
ment with regard to expenditure control, but not as well as global budgets
(assuming that budgets are enforced). The potential for quality improvement
under a DRG-based hospital payment system may be dependent on whether
payments are adjusted for quality of care (see Chapter 8).
Where DRG-based hospital payment provides a fi xed price per unit of activity,
hospitals are incentivized to increase activity and minimize cost and, therefore,
to improve technical effi ciency. While cost-based reimbursement also
encourages increased activity, there is no motivation to minimize inputs/costs
(unless there is a fi xed fee schedule). DRG-based hospital payment may offer
incentives to improve allocative and cost-effi ciency by encouraging providers
to consider the prices and amount of inputs they use. It may also promote an
effi cient allocation of outputs if prices refl ect their relative value but, in practice,
most jurisdictions still base prices on costs. Nevertheless, overall, DRG-based
hospital payment is likely to provide stronger incentives for effi ciency compared
to either of the alternatives.
7.3 Review of empirical evidence on DRG-based hospital
payment and effi ciency
The preceding discussion suggests that DRG-based hospital payment may en-
hance hospital effi ciency, either by changing the focus of cost-based reimburse-
ment from retrospective to prospective (as was the case in the United States), or
by explicitly linking payment to activity in systems with global budgets (as in
most European countries).
This section reviews recent empirical evidence from developed countries.
Although improving hospital effi ciency is generally a key motivation for intro-
ducing DRG-based hospital payment, relatively few studies have explicitly
Table 7.2 Incentives offered by three hospital payment models
Objective
Increase
activity
Expenditure
control
Improve
quality
Enhance effi ciency
Technical Cost Allocative
Cost-based/
fee-for-
service
Strong Weak Strong* Weak Weak Weak
Global
budget
Weak Strong Moderate Weak Moderate Moderate
‘Pure’
DRG-based
hospital
payment
Moderate Moderate Moderate Strong Strong Moderate
*However, quality of care could be adversely affected, as the incentive to increase activity may
lead to the provision of inappropriate and potentially harmful services (see Chapter 8 of this
volume).
102 Diagnosis-Related Groups in Europe
identifi ed and quantifi ed its impact. Rather, most research has concentrated on
indicators of effi ciency – such as activity and costs – which are more easily mea-
sured, but by defi nition provide only a partial picture. It is important to note,
moreover, that the different starting points in the United States and Europe also
imply different hypotheses about the impact of DRG-based payment; that is,
moving from cost-based reimbursement to DRGs weakens the activity incentive
and strengthens the expenditure control incentive – while the opposite is the
case when moving from global budgets to DRG-based payment.
7.3.1 Impact on effi ciency
Studies of the impact of DRG-based payment on hospital-level effi ciency
typically focus on technical effi ciency and/or the broader concept of productiv-
ity (which incorporates scale, as well as technical, effi ciency; see Coelli et al.,
2005; Street & Häkkinen, 2010). Data envelopment analysis (DEA) – a well-
established non-parametric method – is the most commonly applied approach,
although some studies use regression-based (parametric) stochastic frontier
analysis. Both methods have advantages and disadvantages (inter alia, Jacobs
et al., 2006; Street & Häkkinen, 2010; Street et al., 2010) yet, reassuringly, stud-
ies that applied both techniques produce broadly consistent results (Gerdtham
et al., 1999a, b; Dismuke & Sena, 1999).
Given the challenges inherent in undertaking cross-country effi ciency com-
parisons, all but two of the studies summarized in Table 7.3 adopted a longitudi-
nal perspective, comparing hospital effi ciency before and after the introduction
of a DRG system. However, the length of follow-up periods varies, complicating
interpretation: where the time horizon is short, changes may not be sustained;
conversely, a longer time frame may fail to establish a causal relationship, par-
ticularly if other reforms are implemented in the interim. Several studies expli-
citly highlight the diffi culty in attributing changes in effi ciency, or any of its
indicators, to the introduction of DRG-based payment (Farrar et al., 2007; Audit
Commission, 2008). Moreover, few studies assess the quality of care, despite
the potential trade-off between quality and effi ciency (see Chapter 8 of this
volume).
Methodological caveats aside, fi ndings relating to the impact of DRG-based
payment on hospital effi ciency are mixed. The reformed hospital payment
system was associated with improved technical effi ciency in Portugal (albeit
narrowly assessed; Dismuke & Sena, 1999, 2001), Sweden (Gerdtham et al.,
1999a, b) and Norway (Biørn et al., 2003; Hagen et al., 2006). By contrast, no
positive impact was observed in the United States (Borden, 1988; Chern & Wan,
2000) and there were technological improvements but no technical effi ciency
gains in Austria (Sommersguter-Reichmann, 2000). The limited evidence on
time-series changes to cost-effi ciency – confi ned to Norwegian data – is also
mixed (Biørn et al., 2003; Hagen et al., 2006). These divergent results may be
explained by the country-specifi c starting points and contexts in which the
hospital payment reforms were implemented, including different incumbent
reimbursement mechanisms, the specifi cation of DRG-based payment, and/or
the simultaneous introduction of other health care reforms.
T
a
b
le
7
.3
Su
m
m
ar
y
o
f
re
ce
n
t
st
u
d
ie
s
ex
am
in
in
g
t
h
e
im
p
ac
t
o
f
D
R
G
-b
as
ed
h
o
sp
it
al
p
ay
m
en
t
o
n
h
o
sp
it
al
e
ffi
c
ie
n
cy
C
ou
n
tr
y,
Y
ea
r
of
ch
a
n
ge
t
o
D
R
G
-
b
a
se
d
h
os
pi
ta
l
pa
ym
en
t
St
u
d
y
M
et
h
od
ol
og
y
V
a
ri
a
b
le
s
R
es
u
lt
s/
C
on
cl
u
si
on
s
U
n
it
ed
S
ta
te
s,
1
9
8
3
B
o
rd
en
,
1
9
8
8
a
M
et
h
od
:
D
E
A
,
ra
ti
o
a
n
d
re
g
re
ss
io
n
a
n
al
y
si
s
Sa
m
pl
e:
5
2
h
o
sp
it
al
s
in
N
ew
J
er
se
y
St
u
d
y
pe
ri
od
:
1
9
7
9
–1
9
8
4
O
u
tp
u
ts
:
(1
)
C
as
es
t
re
at
ed
i
n
e
ac
h
o
f
th
e
ei
g
h
t
D
R
G
ca
te
g
o
ri
es
w
it
h
t
h
e
h
ig
h
es
t
v
o
lu
m
es
;
(2
)
C
as
es
tr
ea
te
d
i
n
t
h
e
re
m
ai
n
in
g
D
R
G
c
at
eg
o
ri
es
In
pu
ts
:
(1
)
T
o
ta
l
FT
E
s;
(
2
)
N
u
rs
in
g
F
T
E
s;
(
3
)
O
th
er
n
o
n
-
p
ay
ro
ll
e
x
p
en
se
s;
(
4
)
B
ed
s
Q
u
a
li
ty
:
N
o
t
in
cl
u
d
ed
T
h
e
h
o
sp
it
al
p
ay
m
en
t
re
fo
rm
d
id
n
o
t
h
av
e
a
p
o
si
ti
v
e
ef
fe
ct
o
n
te
ch
n
ic
al
e
ffi
c
ie
n
cy
C
h
er
n
&
W
an
,
2
0
0
0
M
et
h
od
:
D
E
A
Sa
m
pl
e:
8
0
h
o
sp
it
al
s
in
V
ir
g
in
ia
St
u
d
y
pe
ri
od
:
1
9
8
4
&
1
9
9
3
O
u
tp
u
ts
:
(1
)
C
as
em
ix
-a
d
ju
st
ed
i
n
p
at
ie
n
t
d
is
ch
ar
g
es
;
(2
)
V
is
it
s
to
t
h
e
E
R
a
n
d
o
u
tp
at
ie
n
t
fa
ci
li
ti
es
In
pu
ts
:
(1
)
B
ed
s
an
d
s
er
v
ic
e
co
m
p
le
x
it
y
;
(2
)
N
o
n
-
p
h
y
si
ci
an
F
T
E
s
an
d
w
ei
g
h
te
d
n
u
m
b
er
o
f
p
ar
t-
ti
m
e
p
er
so
n
n
el
;
(3
)
O
p
er
at
in
g
e
x
p
en
se
s
ex
cl
u
d
in
g
p
ay
ro
ll
,
ca
p
it
al
a
n
d
d
ep
re
ci
at
io
n
Q
u
a
li
ty
:
N
o
t
in
cl
u
d
ed
T
h
er
e
w
as
n
o
s
ta
ti
st
ic
al
ly
si
g
n
ifi
c
an
t
d
if
fe
re
n
ce
i
n
te
ch
n
ic
al
e
ffi
c
ie
n
cy
b
et
w
ee
n
1
9
8
4
a
n
d
1
9
9
3
,
b
u
t
th
e
p
er
ce
n
ta
g
e
o
f
ef
fi
ci
en
t
h
o
sp
it
al
s
w
as
h
ig
h
er
i
n
1
9
9
3
P
o
rt
u
g
al
,
1
9
9
0
D
is
m
u
k
e
&
Se
n
a,
1
9
9
9
b
M
et
h
od
:
T
w
o
s
ta
g
es
:
(1
)
D
E
A
a
n
d
m
ax
im
u
m
li
k
el
ih
o
o
d
e
st
im
at
io
n
o
f
st
o
ch
as
ti
c
in
p
u
t
re
q
u
ir
em
en
t
fr
o
n
ti
er
;
(2
)
R
eg
re
ss
io
n
Sa
m
pl
e:
2
D
R
G
s:
(
1
)
H
ea
rt
Fa
il
u
re
a
n
d
S
h
o
ck
;
(2
)
Sp
ec
ifi
c
C
er
eb
ro
v
as
cu
la
r
D
is
o
rd
er
s
ex
ce
p
t
T
ra
n
si
en
t
Is
ch
ae
m
ic
A
tt
ac
k
St
u
d
y
pe
ri
od
:
1
9
9
2
–1
9
9
4
O
u
tp
u
ts
:
(1
)
N
u
m
b
er
o
f
li
v
e
d
is
ch
ar
g
es
w
it
h
in
e
ac
h
D
R
G
;
(2
)
N
u
m
b
er
o
f
d
ea
d
d
is
ch
ar
g
es
w
it
h
in
e
ac
h
D
R
G
In
pu
ts
:
U
ti
li
za
ti
o
n
o
f:
(
1
)
C
A
T
s
ca
n
n
er
;
(2
)
E
le
ct
ro
ca
rd
io
g
ra
m
;
(3
)
E
ch
o
ca
rd
io
g
ra
m
Q
u
a
li
ty
:
D
is
ti
n
g
u
is
h
es
b
et
w
ee
n
d
es
ir
ab
le
o
u
tp
u
ts
(l
iv
e
d
is
ch
ar
g
es
)
an
d
u
n
d
es
ir
ab
le
o
u
tp
u
ts
(
d
ea
d
d
is
ch
ar
g
es
)
P
er
ce
n
ta
g
e
p
ai
d
t
h
ro
u
g
h
D
R
G
s
h
ad
a
p
o
si
ti
v
e
im
p
ac
t
o
n
p
ro
d
u
ct
iv
it
y
C
on
ti
n
u
ed
o
ve
rl
ea
f
C
ou
n
tr
y,
Y
ea
r
of
ch
a
n
ge
t
o
D
R
G
-
b
a
se
d
h
os
pi
ta
l
pa
ym
en
t
St
u
d
y
M
et
h
od
ol
og
y
V
a
ri
a
b
le
s
R
es
u
lt
s/
C
on
cl
u
si
on
s
P
o
rt
u
g
al
,
1
9
9
0
D
is
m
u
k
e
&
Se
n
a,
2
0
0
1
b
M
et
h
od
:
M
al
m
q
u
is
t-
L
u
en
b
er
g
er
i
n
d
ex
Sa
m
pl
e:
2
D
R
G
s
St
u
d
y
pe
ri
od
:
1
9
9
2
–1
9
9
4
O
u
tp
u
ts
:
(1
)
N
u
m
b
er
o
f
li
v
e
d
is
ch
ar
g
es
w
it
h
in
e
ac
h
D
R
G
;
(2
)
N
u
m
b
er
o
f
d
ea
d
d
is
ch
ar
g
es
w
it
h
in
e
ac
h
D
R
G
In
pu
ts
:
U
ti
li
za
ti
o
n
o
f:
(
1
)
C
A
T
s
ca
n
n
er
;
(2
)
E
le
ct
ro
ca
rd
io
g
ra
m
;
(3
)
E
ch
o
ca
rd
io
g
ra
m
Q
u
a
li
ty
:
A
s
p
er
D
is
m
u
k
e
&
S
en
a,
1
9
9
9
D
R
G
-b
as
ed
p
ay
m
en
t
ap
p
ea
rs
to
h
av
e
im
p
ro
v
ed
t
h
e
p
ro
d
u
ct
iv
it
y
o
f
th
e
d
ia
g
n
o
st
ic
te
ch
n
o
lo
g
ie
s
co
n
si
d
er
ed
Sw
ed
en
,
E
ar
ly
1
9
9
0
s
G
er
d
th
am
et
a
l.
,
1
9
9
9
b
M
et
h
od
:
T
w
o
s
ta
g
es
:
(1
)
M
o
d
ifi
e
d
D
E
A
;
(2
)
R
eg
re
ss
io
n
Sa
m
pl
e:
2
6
c
o
u
n
ty
co
u
n
ci
ls
St
u
d
y
pe
ri
od
:
1
9
9
3
&
1
9
9
4
O
u
tp
u
ts
:
(1
)
Su
rg
ic
al
d
is
ch
ar
g
es
;
(2
)
Sh
o
rt
-t
er
m
in
te
rn
al
m
ed
ic
in
e
d
is
ch
ar
g
es
;
(3
)
Su
rg
ic
al
o
p
er
at
io
n
s
in
s
h
o
rt
-t
er
m
c
ar
e;
(
4
)
P
h
y
si
ci
an
v
is
it
s
in
s
h
o
rt
-t
er
m
s
u
rg
ic
al
c
ar
e;
(
5
)
P
h
y
si
ci
an
v
is
it
s
in
in
te
rn
al
m
ed
ic
in
e
In
pu
ts
:
(1
)
T
o
ta
l
co
st
f
o
r
sh
o
rt
-t
er
m
c
ar
e;
(
2
)
B
ed
s
Q
u
a
li
ty
:
N
o
t
in
cl
u
d
ed
H
o
sp
it
al
s
er
v
ic
es
w
er
e
m
o
re
ef
fi
ci
en
t
in
c
o
u
n
ty
c
o
u
n
ci
ls
w
it
h
i
n
te
rn
al
m
ar
k
et
s
an
d
o
u
tp
u
t-
b
as
ed
r
ei
m
b
u
rs
em
en
t,
co
m
p
ar
ed
t
o
t
h
o
se
w
it
h
a
b
u
d
g
et
-b
as
ed
a
p
p
ro
ac
h
.
P
o
te
n
ti
al
c
o
st
-s
av
in
g
s
o
f
ap
p
ro
x
im
at
el
y
1
3
%
b
y
sw
it
ch
in
g
f
ro
m
b
u
d
g
et
-
to
o
u
tp
u
t-
b
as
ed
r
ei
m
b
u
rs
em
en
t
G
er
d
th
am
et
a
l.
,
1
9
9
9
a
M
et
h
od
:
M
u
lt
ip
le
-o
u
tp
u
t
st
o
ch
as
ti
c
ra
y
f
ro
n
ti
er
m
o
d
el
Sa
m
pl
e:
2
6
c
o
u
n
ty
co
u
n
ci
ls
St
u
d
y
pe
ri
od
:
1
9
8
9
–1
9
9
5
D
ep
en
d
en
t
va
ri
a
b
le
s:
(
1
)
O
p
er
at
io
n
s;
(
2
)
D
is
ch
ar
g
es
;
(3
)
P
h
y
si
ci
an
v
is
it
s
In
d
ep
en
d
en
t
va
ri
a
b
le
s:
(
1
)
C
o
st
;
(2
)
A
v
ai
la
b
le
b
ed
s;
(3
)
Y
ea
r;
(
4
)
V
ar
ia
b
le
s
to
c
ap
tu
re
t
h
e
le
ad
e
ff
ec
ts
o
f
re
fo
rm
;
(5
)
V
ar
ia
b
le
s
fo
r
th
e
n
ew
r
ei
m
b
u
rs
em
en
t
sy
st
em
;
(6
)
P
o
li
ti
ca
l
m
aj
o
ri
ty
;
(7
)
P
ro
p
o
rt
io
n
o
f
p
o
p
u
la
ti
o
n
a
g
ed
o
v
er
7
0
y
ea
rs
;
(8
)
P
ro
p
o
rt
io
n
o
f
p
ri
v
at
e
v
is
it
s;
(
9
)
U
n
iv
er
si
ty
h
o
sp
it
al
Q
u
a
li
ty
:
N
o
t
in
cl
u
d
ed
M
o
v
e
to
o
u
tp
u
t-
b
as
ed
h
o
sp
it
al
p
ay
m
en
t
in
cr
ea
se
d
t
ec
h
n
ic
al
ef
fi
ci
en
cy
b
y
9
.7
%
o
n
a
v
er
ag
e
T
a
b
le
7
.3
C
on
ti
n
u
ed
C
ou
n
tr
y,
Y
ea
r
of
ch
a
n
ge
t
o
D
R
G
-
b
a
se
d
h
os
pi
ta
l
pa
ym
en
t
St
u
d
y
M
et
h
od
ol
og
y
V
a
ri
a
b
le
s
R
es
u
lt
s/
C
on
cl
u
si
on
s
A
u
st
ri
a,
1
9
9
7
So
m
m
er
sg
u
te
r-
R
ei
ch
m
an
n
,
2
0
0
0
M
et
h
od
:
D
E
A
/
M
al
m
q
u
is
t
Sa
m
pl
e:
2
2
h
o
sp
it
al
s
St
u
d
y
pe
ri
od
:
1
9
9
4
–1
9
9
8
O
u
tp
u
ts
:
(1
)
P
at
ie
n
ts
t
re
at
ed
i
n
t
h
e
o
u
tp
at
ie
n
t
ca
re
u
n
it
;
(2
)
C
re
d
it
p
o
in
ts
r
ep
o
rt
ed
b
y
e
ac
h
h
o
sp
it
al
,
m
u
lt
ip
li
ed
b
y
a
s
te
er
in
g
f
ac
to
r
(t
o
d
if
fe
re
n
ti
at
e
b
et
w
ee
n
h
o
sp
it
al
t
y
p
es
)
In
pu
ts
:
(1
)
L
ab
o
u
r
FT
E
s;
(
2
)
H
o
sp
it
al
b
ed
s;
(
3
)
E
x
p
en
se
s
fo
r
ex
te
rn
al
m
ed
ic
al
s
er
v
ic
es
Q
u
a
li
ty
:
N
o
t
in
cl
u
d
ed
T
h
er
e
w
as
a
n
i
m
p
ro
v
em
en
t
in
te
ch
n
o
lo
g
y
b
et
w
ee
n
1
9
9
6
an
d
1
9
9
8
,
b
u
t
th
er
e
w
as
n
o
im
p
ro
v
em
en
t
in
t
ec
h
n
ic
al
ef
fi
ci
en
cy
N
o
rw
ay
,
1
9
9
7
B
iø
rn
e
t
al
.,
2
0
0
3
M
et
h
od
:
T
w
o
s
ta
g
es
:
(1
)
D
E
A
;
(2
)
R
eg
re
ss
io
n
Sa
m
pl
e:
4
8
h
o
sp
it
al
s
St
u
d
y
pe
ri
od
:
1
9
9
2
–2
0
0
0
O
u
tp
u
ts
:
(1
)
C
as
em
ix
-a
d
ju
st
ed
d
is
ch
ar
g
es
(
in
cl
u
d
in
g
d
ay
c
ar
e)
;
(2
)
O
u
tp
at
ie
n
t
v
is
it
s
w
ei
g
h
te
d
b
y
t
h
e
fe
e
p
ai
d
b
y
t
h
e
st
at
e
fo
r
ea
ch
v
is
it
In
pu
ts
:
(1
)
P
h
y
si
ci
an
F
T
E
s;
(
2
)
O
th
er
l
ab
o
u
r
FT
E
s;
(3
)
M
ed
ic
al
e
x
p
en
se
s;
(
4
)
T
o
ta
l
ru
n
n
in
g
e
x
p
en
se
s
(f
o
r
an
al
y
si
s
o
f
co
st
-e
ffi
c
ie
n
cy
)
Q
u
a
li
ty
:
N
o
t
in
cl
u
d
ed
T
h
e
in
tr
o
d
u
ct
io
n
o
f
D
R
G
-b
as
ed
h
o
sp
it
al
p
ay
m
en
t
im
p
ro
v
ed
t
ec
h
n
ic
al
e
ffi
c
ie
n
cy
,
b
u
t
re
su
lt
s
re
la
ti
n
g
t
o
t
h
e
im
p
ac
t
o
n
c
o
st
-e
ffi
c
ie
n
cy
w
er
e
v
ar
ie
d
H
ag
en
e
t
al
.,
2
0
0
6
M
et
h
od
:
T
w
o
s
ta
g
es
:
(1
)
D
E
A
;
(2
)
R
eg
re
ss
io
n
Sa
m
pl
e:
4
8
h
o
sp
it
al
s
St
u
d
y
pe
ri
od
:
1
9
9
2
–2
0
0
3
O
u
tp
u
ts
:
(1
)
C
as
em
ix
-a
d
ju
st
ed
d
is
ch
ar
g
es
;
(2
)
O
u
tp
at
ie
n
t
v
is
it
s
w
ei
g
h
te
d
b
y
g
o
v
er
n
m
en
t
re
im
b
u
rs
em
en
t
p
er
v
is
it
In
pu
ts
:
(1
)
P
h
y
si
ci
an
F
T
E
s;
(
2
)
O
th
er
l
ab
o
u
r
FT
E
s;
(3
)
M
ed
ic
al
e
x
p
en
se
s;
(
4
)
T
o
ta
l
o
p
er
at
in
g
c
o
st
s
(f
o
r
an
al
y
si
s
o
f
co
st
-e
ffi
c
ie
n
cy
)
Q
u
a
li
ty
:
N
o
t
in
cl
u
d
ed
T
ec
h
n
ic
al
e
ffi
c
ie
n
cy
i
n
cr
ea
se
d
af
te
r
th
e
re
im
b
u
rs
em
en
t
re
fo
rm
,
b
u
t
th
e
ef
fe
ct
o
n
c
o
st
-
ef
fi
ci
en
cy
w
as
i
n
si
g
n
ifi
c
an
t
N
o
rw
ay
,
1
9
9
7
an
d
F
in
la
n
d
L
in
n
a
et
a
l.
,
2
0
0
6
M
et
h
od
:
D
E
A
Sa
m
pl
e:
F
in
la
n
d
–
4
7
h
o
sp
it
al
s
an
d
N
o
rw
ay
–
5
1
h
o
sp
it
al
s
St
u
d
y
pe
ri
od
:
1
9
9
9
O
u
tp
u
ts
:
(1
)
D
R
G
-w
ei
g
h
te
d
a
d
m
is
si
o
n
s;
(
2
)
W
ei
g
h
te
d
o
u
tp
at
ie
n
t
v
is
it
s;
(
3
)
W
ei
g
h
te
d
d
ay
c
ar
e;
(
4
)
In
p
at
ie
n
t
d
ay
s
In
pu
ts
:
(1
)
N
et
o
p
er
at
in
g
c
o
st
s
Q
u
a
li
ty
:
N
o
t
in
cl
u
d
ed
T
h
e
av
er
ag
e
le
v
el
o
f
co
st
-e
ffi
c
ie
n
cy
w
as
l
o
w
er
i
n
N
o
rw
eg
ia
n
h
o
sp
it
al
s
So
u
rc
e:
C
o
m
p
il
ed
b
y
t
h
e
au
th
o
rs
b
as
ed
o
n
t
h
e
w
o
rk
s
li
st
ed
i
n
t
h
e
‘S
tu
d
y
’
co
lu
m
n
.
N
ot
es
:
E
R
:
em
er
g
en
cy
r
o
o
m
;
FT
E
:
fu
ll
-t
im
e
eq
u
iv
al
en
t;
a
St
u
d
ie
s
h
o
sp
it
al
s
in
N
ew
J
er
se
y,
i
n
w
h
ic
h
D
R
G
-b
as
ed
p
ay
m
en
ts
w
er
e
in
tr
o
d
u
ce
d
f
o
r
al
l
p
ay
er
s
in
1
9
8
0
;
b
A
ss
es
se
s
th
e
p
ro
d
u
ct
iv
it
y
a
n
d
t
ec
h
n
ic
al
e
ffi
c
ie
n
cy
o
f
d
ia
g
n
o
st
ic
t
ec
h
n
o
lo
g
ie
s
o
n
ly
.
106 Diagnosis-Related Groups in Europe
Taking the fi rst of these, the potential for effi ciency gains may depend on the
pre-existing hospital payment system. Thus, where global budgets preceded
DRG-based payment (as in Sweden, Portugal and Norway, detailed in Table 7.3),
hospitals’ technical effi ciency apparently improved (although Linna and
colleagues (2006) found lower cost-effi ciency in Norwegian hospitals com-
pared to their Finnish counterparts, despite the latter being understood to
operate within a global budget framework). Conversely, DRG-based payment
did not improve technical effi ciency when it replaced retrospective, cost-based
reimbursement (as in the United States) or per diem payments (as in Austria).
This apparent greater potential for effi ciency gains when moving from global
budgets cannot be regarded as defi nitive, because the operation of the national
DRG-based payment system may itself act as a constraint. Hence, initial effi -
ciency improvements in Sweden were subsequently negated when ceilings were
imposed on hospital-activity levels (Gerdtham et al., 1999a, b; Anell, 2005;
Kastberg & Siverbo, 2007), and analogous restrictions may also help to explain
the lack of improvements in the United States and Austria (US Congress Offi ce
of Technology Assessment, 1985; Sommersguter-Reichmann, 2000; Böcking et
al., 2005). Finally, it is diffi cult to isolate the impact of DRG-based payment
when it is introduced as part of a wider health care reform programme, as was
the case in Sweden when an internal market was also established (Gerdtham
et al., 1999a, b).
7.3.2 Impact on indicators of effi ciency: Activity, length
of stay and costs
Table 7.4 summarizes studies that examined country-specifi c changes in indica-
tors of effi ciency. Following the introduction of DRG-based payment, hospital
admissions increased in Australia (Ettelt et al., 2006; Street et al., 2007), Denmark
(Street et al., 2007), England (Farrar et al., 2007; Audit Commission, 2008; Farrar
et al., 2009), France (Or, 2009), Germany (Böcking et al., 2005; Hensen et al.,
2008), Norway (Biørn et al., 2003; Kjerstad, 2003; Hagen et al., 2006; Magnussen
et al., 2007), Spain (Cots, 2004, cited in Ellis & Vidal-Fernández, 2007) and, at
least initially, in Sweden (Anell, 2005; Kastberg & Siverbo, 2007). However, in line
with the hypotheses derived from the incentives indicated in Table 7.2, activity
did not increase in the United States (US Congress Offi ce of Technology Assess-
ment, 1985; Davis & Rhodes, 1988; Guterman et al., 1988; Manton et al., 1993;
Muller, 1993; Rosenberg & Browne, 2001). Results for Italy are mixed (Louis et al.,
1999; Ettelt et al., 2006), while Moreoa-Serra & Wagstaff (2010) found no effect
on activity countries in central and eastern Europe and central Asia that had in-
troduced DRGs or other activity-based reimbursement systems. Of course, the
aforementioned points regarding country-specifi c contexts and the diffi culties in
assigning causality also apply here.
The fi na ncial incentive to minimize costs under DRG-based hospital payment
has often contributed to a shift from inpatient to day-case and/or outpatient
settings (for example, in the United States and England, see Rosenberg &
Browne, 2001 and Farrar et al., 2009, respectively) – this may also improve the
quality of care, as well as effi ciency, ceteris paribus. Indeed, DRG-based tariffs
T
a
b
le
7
.4
Su
m
m
ar
y
o
f
re
ce
n
t
st
u
d
ie
s
ex
am
in
in
g
ch
an
ge
s
in
i
n
d
ic
at
o
rs
o
f
h
o
sp
it
al
e
ffi
c
ie
n
cy
f
o
ll
o
w
in
g
th
e
in
tr
o
d
u
ct
io
n
o
f
D
R
G
-b
as
ed
h
o
sp
it
al
p
ay
m
en
t
C
ou
n
tr
y,
Y
ea
r
of
c
h
a
n
ge
to
D
R
G
-b
a
se
d
h
os
pi
ta
l
pa
ym
en
t
St
u
d
y
St
u
d
y
pe
ri
od
M
et
h
od
ol
og
y
H
os
pi
ta
l
a
ct
iv
it
y
A
L
O
S
C
os
ts
U
n
it
/a
ve
ra
ge
T
ot
a
l
U
n
it
ed
S
ta
te
s,
1
9
8
3
U
S
C
o
n
gr
es
s
O
ffi
c
e
o
f
Te
ch
n
o
lo
gy
A
ss
es
sm
en
t,
1
9
8
5
1
9
8
3
–1
9
8
4
a
n
d
R
ev
ie
w
D
es
cr
ip
ti
v
e
–
–
G
u
te
rm
an
e
t
al
.,
1
9
8
8
1
9
8
3
–1
9
8
6
D
es
cr
ip
ti
v
e
(w
it
h
s
o
m
e
su
b
-g
ro
u
p
a
n
al
y
si
s)
–
–
+
+ (b
u
t
at
a
sl
o
w
er
r
at
e)
D
av
is
&
R
h
o
d
es
,
1
9
8
8
1
9
8
4
–1
9
8
5
D
es
cr
ip
ti
v
e
–
–
K
ah
n
e
t
al
.,
1
9
9
0
1
9
8
1
/1
9
8
2
&
1
9
8
5
/1
9
8
6
R
et
ro
sp
ec
ti
v
e
o
b
se
rv
at
io
n
al
st
u
d
y
(
fo
cu
si
n
g
o
n
fi
v
e
d
is
ea
se
s)
–
M
an
to
n
e
t
al
.,
1
9
9
3
1
9
8
2
/1
9
8
3
&
1
9
8
4
/1
9
8
5
D
es
cr
ip
ti
v
e
(u
si
n
g
l
if
e
ta
b
le
m
o
d
el
s)
–
–
M
u
ll
er
,
1
9
9
3
1
9
7
0
–1
9
9
2
A
u
to
re
g
re
ss
iv
e-
in
te
g
ra
te
d
m
o
v
in
g
a
v
er
ag
e
m
o
d
el
s
–
–
R
o
se
n
b
er
g
&
B
ro
w
n
e,
2
0
0
1
R
ev
ie
w
R
ev
ie
w
–
–
A
u
st
ra
li
a,
1
9
9
3
a
E
tt
el
t
et
a
l.
,
2
0
0
6
R
ev
ie
w
R
ev
ie
w
+
–
–
St
re
et
e
t
al
.,
2
0
0
7
R
ev
ie
w
R
ev
ie
w
+
Sw
ed
en
,
ea
rl
y
1
9
9
0
s
A
n
el
l,
2
0
0
5
R
ev
ie
w
R
ev
ie
w
+
–
+
K
as
tb
er
g
&
S
iv
er
b
o
,
2
0
0
7
R
ev
ie
w
R
ev
ie
w
+
–
+
It
al
y,
1
9
9
5
L
o
u
is
e
t
al
.,
1
9
9
9
1
9
9
3
–1
9
9
6
D
es
cr
ip
ti
v
e
(w
it
h
s
o
m
e
su
b
-g
ro
u
p
a
n
al
y
si
s)
?
–
E
tt
el
t
et
a
l.
,
2
0
0
6
R
ev
ie
w
R
ev
ie
w
+
–
C
at
al
o
n
ia
,
Sp
ai
n
,
1
9
9
7
C
o
ts
,
2
0
0
4
b
1
9
9
3
–2
0
0
0
D
es
cr
ip
ti
v
e
+
–
+ (b
u
t
at
a
sl
o
w
er
r
at
e)
C
on
ti
n
u
ed
o
ve
rl
ea
f
T
a
b
le
7
.4
C
on
ti
n
u
ed
C
ou
n
tr
y,
Y
ea
r
of
c
h
a
n
ge
to
D
R
G
-b
a
se
d
h
os
pi
ta
l
pa
ym
en
t
St
u
d
y
St
u
d
y
pe
ri
od
M
et
h
od
ol
og
y
H
os
pi
ta
l
a
ct
iv
it
y
A
L
O
S
C
os
ts
U
n
it
/a
ve
ra
ge
T
ot
a
l
N
o
rw
ay
,
1
9
9
7
B
iø
rn
e
t
al
.,
2
0
0
3
1
9
9
2
–2
0
0
0
D
es
cr
ip
ti
v
e
+
K
je
rs
ta
d
,
2
0
0
3
1
9
9
5
–1
9
9
8
D
if
fe
re
n
ce
-i
n
-d
if
fe
re
n
ce
m
o
d
el
+
H
ag
en
e
t
al
.,
2
0
0
6
1
9
9
2
–2
0
0
0
D
es
cr
ip
ti
v
e
+
M
ag
n
u
ss
en
e
t
al
.,
2
0
0
7
R
ev
ie
w
R
ev
ie
w
+
A
u
st
ri
a,
1
9
9
7
T
h
eu
rl
&
W
in
n
er
,
2
0
0
7
1
9
8
9
–2
0
0
3
E
co
n
o
m
et
ri
c
m
o
d
el
w
it
h
fi
x
ed
e
ff
ec
ts
–
D
en
m
ar
k
,
2
0
0
2
St
re
et
e
t
al
.,
2
0
0
7
R
ev
ie
w
R
ev
ie
w
+
G
er
m
an
y,
2
0
0
3
B
ö
ck
in
g
e
t
al
.,
2
0
0
5
R
ev
ie
w
R
ev
ie
w
+
–
Sc
h
re
y
ö
g
g
e
t
al
.,
2
0
0
5
2
0
0
3
–2
0
0
4
a
n
d
R
ev
ie
w
D
es
cr
ip
ti
v
e
–
H
en
se
n
e
t
al
.,
2
0
0
8
c
2
0
0
3
–2
0
0
6
D
es
cr
ip
ti
v
e
+
–
E
n
g
la
n
d
,
2
0
0
3
/2
0
0
4
Fa
rr
ar
e
t
al
.,
2
0
0
7
2
0
0
2
/2
0
0
3
&
2
0
0
5
/2
0
0
6
D
if
fe
re
n
ce
-i
n
-d
if
fe
re
n
ce
fr
am
ew
o
rk
+
–
–
A
u
d
it
C
o
m
m
is
si
o
n
,
2
0
0
8
2
0
0
3
/2
0
0
4
&
2
0
0
6
/2
0
0
7
D
es
cr
ip
ti
v
e
+
–
?
Fa
rr
ar
e
t
al
.,
2
0
0
9
2
0
0
3
/2
0
0
4
&
2
0
0
5
/2
0
0
6
D
if
fe
re
n
ce
-i
n
-d
if
fe
re
n
ce
an
al
y
si
s
+
–
–
Fr
an
ce
,
2
0
0
4
/2
0
0
5
O
r,
2
0
0
9
2
0
0
5
a
n
d
R
ev
ie
w
D
es
cr
ip
ti
v
e
+
O
E
C
D
c
o
u
n
tr
ie
s
Fo
rg
io
n
e
&
D
’A
n
n
u
n
zi
o
,
1
9
9
9
1
9
8
4
–1
9
8
6
&
1
9
9
4
–1
9
9
6
D
es
cr
ip
ti
v
e
–
+
V
ar
io
u
s
co
u
n
tr
ie
s
in
ce
n
tr
al
a
n
d
e
as
te
rn
E
u
ro
p
e
as
w
el
l
as
ce
n
tr
al
A
si
a
M
o
re
n
o
-S
er
ra
&
W
ag
st
af
f,
2
0
1
0
1
9
9
0
–2
0
0
4
D
if
fe
re
n
ce
-i
n
-d
if
fe
re
n
ce
m
o
d
el
=
–
+
So
u
rc
e:
C
o
m
p
il
ed
b
y
t
h
e
au
th
o
rs
b
as
ed
o
n
t
h
e
w
o
rk
s
li
st
ed
i
n
t
h
e
‘S
tu
d
y
’
co
lu
m
n
.
N
ot
es
: T
h
e
ch
an
ge
s
fo
ll
o
w
in
g
th
e
in
tr
o
d
u
ct
io
n
o
f
D
R
G
-b
as
ed
h
o
sp
it
al
p
ay
m
en
t
ar
e
d
en
o
te
d
b
y
+
f
o
r
an
i
n
cr
ea
se
, –
f
o
r
a
d
ec
re
as
e,
=
f
o
r
n
o
c
h
an
ge
, a
n
d
?
f
o
r
m
ix
ed
e
ff
ec
ts
; a
D
R
G
-b
as
ed
h
o
sp
it
al
p
ay
m
en
t
w
as
i
n
tr
o
d
u
ce
d
i
n
V
ic
to
ri
a
in
1
9
9
3
; b
N
o
t
in
te
n
d
ed
a
s
a
st
u
d
y
o
f
th
e
ef
fe
ct
o
f
D
R
G
-b
as
ed
h
o
sp
it
al
p
ay
m
en
t;
c
R
el
at
es
t
o
d
er
m
at
o
lo
gy
.
DRG-based hospital payment and effi ciency 109
can be used to explicitly incentivize hospitals to increase day-case activity,
as for example in England, where a common national tariff has been applied
to most elective activity across inpatient and day-case settings (Epstein &
Mason, 2006; Street et al., 2007). In the United States, the shift towards
outpatient care may also be explained by the operation (until 2000) of a parallel
retrospective cost-based reimbursement system for such treatment (Rosenberg
& Browne, 2001).
Average length of stay generally declined following the move to DRG-based
payment (for example, Kahn et al., 1990; Böcking et al., 2005; Moreno-Serra &
Wagstaff, 2010), although some argue that this was merely consistent with a
general trend (Rosenberg & Browne, 2001; Schreyögg et al., 2005). Discharge
rates to post-acute institutions (typically less costly than acute facilities) usually
increased. On average, the recorded severity of patients remaining in acute
settings increased (Böcking et al., 2005), and assuming this was not simply
changed coding practice, suggests limited potential for further reductions in
length of stay ceteris paribus (Guterman et al., 1988; Rosenberg & Browne, 2001).
Finally, in the majority of cases, the introduction of DRG-based hospital
payment was associated with higher total costs, partly due to higher activity
levels (Forgione & D’Annunzio, 1999; Anell, 2005; Kastberg & Siverbo, 2007;
Moreno-Serra & Wagstaff, 2010), whereas unit costs appear to have declined
(Böcking et al., 2005; Farrar et al., 2009). In the United States the overall impact
was reduced infl ation in aggregate costs (Guterman et al., 1988). The initial ex-
perience with DRG-based payment in the Netherlands has been a lower rate of
increase where prices are negotiated rather than set centrally and there is increased
competition among hospitals and health insurers (see Chapter 23 of this volume).
In short, in some cases, hospital-level effi ciency has improved following the
introduction of DRG-based hospital payment, but establishing causation is
diffi cult, due to confounding factors. Elsewhere its theoretically benefi cial
effects may have been somewhat offset by other features of the national health
care system – such as limitations on activity and/or expenditure, or the pre-
existing reimbursement system – leading to mixed results.
7.4 What are the key challenges?
7.4.1 Categorization problems may lead to unfair
reimbursement or patient selection
Like any categorization system, DRGs cannot group patients perfectly on the
basis of their expected resource requirements. Much hea lth care is highly
individualized, so defi ning a ‘standardized package of care’ is not straightfor-
ward. This would not create hospital payment problems if differences across
providers were random, but if the differences across providers are systematic,
then the reimbursement system becomes potentially unfair and may encourage
hospitals to engage in up-coding or to ‘dump’ (that is, avoid) high-cost patients.
These adverse consequences could be avoided, however, if the fi nancial risks of
such cases were shared between payer and hospital (see Part Two of this volume,
along with chapters 5 and 6).
110 Diagnosis-Related Groups in Europe
7.4.2 Independence in price-setting
In some countries, the number of hospitals may be insuffi cient to ensure that
prices are independent of each hospital’s costs. This has two implications. First,
the regulator may be unable to determine whether costs are contaminated by
ineffi cient behaviour, especially if provision is concentrated in only one or two
hospitals. DRG-based hospital payment is then in danger of reducing to cost-
based reimbursement – which embodies little incentive to improve effi ciency.
Second, this form of reimbursement may encourage collusion between providers
in their reporting behaviour or in their efforts to reduce their costs. The likelihood
of such behaviour increases if there are few providers that are well informed
about each other’s behaviour. Collusion will limit the scope for DRG-based
hospital payment to deliver effi ciency improvements. Where data are collected
on a sampling basis, as in Germany, the sample must be representative of all
hospitals; otherwise, unfair reimbursement may result (see chapters 5 and 14).
7.4.3 Control of expenditure
DRG-based hospital payment that adopts a simple price-per-unit-of-activity
approach offers direct incentives to suppliers to increase activity levels. If
marginal cost is lower than marginal revenue, the more providers ‘do’, the
larger their fi nancial surplus/profi t. Increases in activity levels may therefore
place severe pressure on funders’ budgets. Consequently, a number of countries
attempt to contain expenditure by a system of operating DRG-based hospital
payment within a global budget framework (for example, Catalonia (Spain) and
Sweden – see chapters 22 and 19, respectively). In France, local-level contracts
were found to be more effective at controlling spending than macro-level
mechanisms (see Chapter 13).
7.5 Conclusions
DRG-based hospital payment systems have the potential to enhance effi ciency
in the delivery of hospital services, more so than other hospital payment
models. This is because there are clear incentives for hospitals to work harder,
because they are paid according to the number of patients they treat, as well as
to control their costs, because the prices they face are set independently of their
own costs. These payment characteristics encourage providers to improve their
technical and cost-effi ciency and to seek allocative effi ciency in their choice of
input mix. In theory, DRG-based hospital payment can be used to support
allocative effi ciency in the overall mix of outputs produced by the hospital
sector as a whole. This requires the price attached to each DRG to refl ect its
societal value. In practice, though, DRG prices are based on costs in almost all
countries, so the pursuit of allocative effi ciency in this sense has not been a
feature of DRG-based hospital payment policy.
Empirical evidence is mixed in terms of the extent to which DRG-based
hospital payment has improved effi ciency. This is partly because of cross-
DRG-based hospital payment and effi ciency 111
country heterogeneity in how DRG-based hospital payment systems are
operated (detailed in Part Two of this volume) and because attribution is
complicated by the existence of confounding factors (such as changes being
part of a wider reform package, or the country-specifi c design and operation of
the reimbursement regime). It is generally agreed that DRG-based hospital
payment affects indicators of effi ciency, such as activity and length of stay,
although the same caveats apply. Unintended consequences may include
skimping (on quality), cost-shifting, patient selection or up-coding to higher
priced DRGs (see Chapter 6).
While we have outlined simplifi ed forms of DRG-based hospital payment,
in practice the payment arrangements implemented in each country can be
quite sophisticated (see the country-specifi c chapters in Part Two of this book).
More complex formulations may refl ect concerns over the ability of DRG
classifi cations to describe casemix accurately, if the need to moderate incen-
tives to undertake more activity in the pursuit of quality, or other regulatory
objectives, such as an equitable geographical distribution of hospital pro-
vision. Such sophistication is not surprising: the provision of hospital care is a
complex process, often requiring packages of care tailored to the individual
patient and delivered under conditions of crisis and uncertainty, requiring
co-ordination of health professionals both within and beyond the hospital.
In the face of such complexity, the method by which payments are made
must be sophisticated enough to provide clear incentives for what is desirable
and to avoid creating perverse responses. Compared to cost-based reimburse-
ment and global budgets, DRG-based hospital payment is able to embody such
sophistication and, thereby, to provide clearer incentives for hospitals to
improve their effi ciency.
7.7 References
Anell, A. (2005). Swedish health care under pressure. Health Economics, 14:S237–54.
Audit Commission (2008). The Right Result? Payment by Results 2003–2007. London: Audit
Commission (http://www.audit-commission.gov.uk/SiteCollectionDocuments/Audit
CommissionReports/NationalStudies/The_right_result_PbR_2008 , accessed 29
June 2011).
Barnum, H., Kutzin, J., Saxenian, H. (1995). Incentives and provider payment methods.
International Journal of Health Planning & Management, 10:23–45.
Biørn, E., Hagen, T.P., Iversen, T., Magnussen, J. (2003). The effect of activity-based
fi nancing on hospital effi ciency: a panel data analysis of DEA effi ciency scores
1992–2000. Health Care Management Science, 6:271–83.
Böcking, W., Ahrens, U., Kirch, W., Milakovic, M. (2005). First results of the introduction
of DRGs in Germany and overview of experience from other DRG countries. Journal
of Public Health, 13:128–37.
Borden, J.P. (1988). An assessment of the impact of diagnosis-related group (DRG)-based
reimbursement on the technical effi ciency of New Jersey hospitals using data
envelopment analysis. Journal of Accounting and Public Policy, 7:77–96.
Chern, J.Y., Wan, T.T. (2000). The impact of the prospective payment system on the
technical effi ciency of hospitals. Journal of Medical Systems, 24:159–72.
Coelli, T.J., Rao, D.S.P., O’Donnell, C.J. (2005). An Introduction to Effi ciency and Productivity
Analysis. New York, NY: Springer Science+Business Media.
112 Diagnosis-Related Groups in Europe
Cots, F. (2004). La sostenibilidad del sistema hospitalario en Cataluña. El balance de una
década [Viability of the hospital system in Catalonia. Balance after a decade]. Gaceta
Sanitaria, 18:64–7.
Davis, C., Rhodes, D.J. (1988). The impact of DRGs on the cost and quality of health care
in the United States. Health Policy, 9:117–31.
Dismuke, C., Sena, V. (1999). Has DRG payment infl uenced the technical effi ciency and
productivity of diagnostic technologies in Portuguese public hospitals? An empirical
analysis using parametric and non-parametric methods. Health Care Management
Science, 2:107–16.
Dismuke, C., Sena, V. (2001). Is there a trade-off between quality and productivity? The case
of diagnostic technologies in Portugal. Annals of Operations Research, 107:101–16.
Ellis, R.P., McGuire, T.G. (1986). Provider behaviour under prospective reimbursement:
cost sharing and supply. Journal of Health Economics, 5:129–51.
Ellis, R.P., Vidal-Fernández, M. (2007). Activity-based payments and reforms of the English
hospital payment system. Health Economics, Policy and Law, 2:435–44.
Epstein, D., Mason, A. (2006). Costs and prices for inpatient care in England: mirror twins
or distant cousins? Health Care Management Science, 9:233–42.
Ettelt, S., Thomson, S., Nolte, E., Mays, N. (2006). Reimbursing Highly Specialised Hospital
Services: The Experience of Activity-Based Funding in Eight Countries. London: London
School of Hygiene & Tropical Medicine.
Farrar, S., Sussex, J., Yi, D. et al. (2007). National Evaluation of Payment by Results. Aberdeen:
University of Aberdeen Health Economics Research Unit.
Farrar, S., Yi, D., Sutton, M. et al. (2009). Has payment by results affected the way that
English hospitals provide care? Difference-in-differences analysis. British Medical
Journal, 339 b3047doi:10.1136/bmj.b3047.
Forgione, D.A., D’Annunzio, C.M. (1999). The use of DRGs in health care payment
systems around the world. Journal of Health Care Finance, 26:66–78.
Gerdtham, U., Löthgren, M., Tambour, M., Rehnberg, C. (1999a). Internal markets and
health care effi ciency: a multiple-output stochastic frontier analysis. Health Economics,
8:151–64.
Gerdtham, U., Rehnberg, C., Tambour, M. (1999b). The impact of internal markets on
health care effi ciency: evidence from health care reforms in Sweden. Applied Economics,
31:935–45.
Guterman, S., Eggers, P.W., Riley, G., Greene, T.F., Terrell, S.A. (1988). The fi rst 3 years of
Medicare prospective payment: an overview. Health Care Financing Review, 9:67–77.
Hagen, T.P., Veenstra, M., Stavem, K. (2006). Effi ciency and Patient Satisfaction in Norwegian
Hospitals. Oslo: Health Organization Research Norway (HORN Working Paper
2006:1).
Hensen, P., Beissert, S., Bruckner-Tuderman, L. et al. (2008). Introduction of
diagnosis-related groups in Germany: evaluation of impact on inpatient care in a
dermatological setting. European Journal of Public Health, 18:85–91.
Jacobs, R., Smith, P.C., Street, A. (2006). Measuring Effi ciency in Health Care: Analytic
Techniques and Health Policy. Cambridge: Cambridge University Press.
Kahn, K.L., Keeler, E.B., Sherwood, M.J. et al. (1990). Comparing outcomes of care before
and after implementation of the DRG-based prospective payment system. Journal of
the American Medical Association, 264:1984–8.
Kastberg, G., Siverbo, S. (2007). Activity-based fi nancing of health care – experiences from
Sweden. International Journal of Health Planning and Management, 22:25–44.
Kjerstad, E. (2003). Prospective funding of general hospitals in Norway: incentives for
higher production? International Journal of Health Care Finance and Economics,
3:231–51.
DRG-based hospital payment and effi ciency 113
Langenbrunner, J.C., Orosz, E., Kutzin, J., Wiley, M.M. (2005). Purchasing and paying
providers, in J. Figueras, R. Robinson, E. Jakubowski, eds. Purchasing to Improve Health
Systems Performance. Maidenhead: Open University Press.
Linna, M., Häkkinen, U., Magnussen, J. (2006). Comparing hospital cost effi ciency
between Norway and Finland. Health Policy, 77:268–78.
Louis, D.Z., Yuen, E.J., Braga, M. et al. (1999). Impact of a DRG-based hospital fi nanc-
ing system on quality and outcomes of care in Italy. Health Services Research, 34:
405–15.
Magnussen, J., Hagen, T.P., Kaarboe, O.M. (2007). Centralized or decentralized? A case
study of Norwegian hospital reform. Social Science & Medicine, 64:2129–37.
Manton, K.G., Woodbury, M.A., Vertrees, J.C., Stallard, E. (1993). Use of Medicare services
before and after introduction of the prospective payment system. Health Services
Research, 28:269–92.
Moreno-Serra, R., Wagstaff, A. (2010). System-wide impacts of hospital payment reforms:
evidence from central and eastern Europe and central Asia. Journal of Health Economics,
29:585–602.
Muller, A. (1993). Medicare prospective payment reforms and hospital utilization:
temporary or lasting effects? Medical Care, 31:296–308.
Or, Z. (2009). Activity-based payment in France. Euro Observer, 11:5–6.
Robinson, R., Jakubowski, E., Figueras, J. (2005). Organization of purchasing in Europe, in
J. Figueras, R. Robinson, E. Jakubowski, eds. Purchasing to Improve Health Systems
Performance. Maidenhead: Open University Press.
Rosenberg. M.A., Browne, M.J. (2001). The impact of the inpatient prospective payment
system and diagnosis-related groups: a survey of the literature. North American
Actuarial Journal, 5:84–94.
Schreyögg, J., Tiemann, O., Busse, R. (2005). The DRG reimbursement system in Germany.
Euro Observer, 7:4–6.
Shleifer, A. (1985). A theory of yard stick competition. Rand Journal of Economics, 16:
319–27.
Sommersguter-Reichmann, M. (2000). The impact of the Austrian hospital fi nancing
reform on hospital productivity: empirical evidence on effi ciency and technology
changes using a non-parametric input-based Malmquist approach. Health Care
Management Science, 3:309–21.
Street, A., Häkkinen, U. (2010). Health system productivity and effi ciency, in P.C. Smith,
E. Mossialos, I. Papanicolas, S. Leatherman, eds. Performance Measurement for Health
System Improvement: Experiences, Challenges and Prospects. Cambridge: Cambridge
University Press.
Street, A., Scheller-Kreinsen, D., Geissler, A., Busse, R. (2010). Determinants of Hospital
Costs and Performance Variation: Methods, Models and Variables for the EuroDRG Project.
Berlin: Universitätsverlag der Technischen Universität Berlin (working Papers in
Health Policy and Management Volume 3).
Street, A., Vitikainen, K., Bjorvatn, A., Hvenegaard, A. (2007). Introducing Activity-Based
Financing: A Review of Experience in Australia, Denmark, Norway and Sweden. York:
University of York Centre for Health Economics (CHE Research Paper 30).
Theurl, E., Winner, H. (2007). The impact of hospital fi nancing on the length of stay:
evidence from Austria. Health Policy, 82:375–89.
US Congress Offi ce of Technology Assessment (1985). Medicare’s Prospective Payment
System: Strategies for Evaluating Cost, Quality, and Medical Technology. Washington, DC:
United States Government Printing Offi ce.
WHO (2000). The World Health Report 2000 – Health Systems: Improving Performance.
Geneva: World Health Organization.
114 Diagnosis-Related Groups in Europe
7.8 Summary of terms used in the equations
Symbol Description
RC, RG, RA Hospital revenue under, respectively, cost-based reimbursement
(C), global budgets (G) and DRG-based funding (A)
ZC, ZG, ZA All sources of revenue not related to health to care activity under
cost-based reimbursement, global budgets and DRG-based funding
Q Activity
Q Target activity
I Individual patient
S Specialty
J DRG
C Unit cost
B
s
Specialty contract value
p
s
Locally agreed specialty-level price
p̂
j
Prospectively fi xed DRG price
α Proportion of fi xed DRG price paid for additional activity
chapter e i g h t
DRGs and quality:
For better or worse?
Zeynep Or and Unto Häkkinen
8.1 Introduction
Initially, in most European countries, diagnosis-related groups (DRGs) were in-
troduced to better describe hospital services and to improve the measurement
and management of hospital production (services). Increasing the transparency
of care procedures and hence facilitating comparisons of hospitals’ activity was
seen as a way of improving quality of care in hospitals. Over time, DRGs have
increasingly become the basis for hospital payment. However, the impact of
DRG-based hospital payment systems on quality of care is not straightforward.
These systems may present an inherent risk to quality of care because they
directly incentivize hospitals to reduce the cost per stay, irrespective of out-
comes. Hospitals are expected to reduce costs by cutting down unnecessary
services and by improving effi ciency through organizational changes. On the
one hand, these changes may improve quality, if they improve clinical process
and care management. On the other hand, providers may also ‘skimp’ on qual-
ity as a way of cost-saving, potentially placing the patient’s health at risk.
There are many different ways through which DRG-based hospital payment
systems may create perverse incentives (Ellis & McGuire, 1996; Miraldo et al.,
2006), which could negatively affect care quality. In particular, hospitals may
discharge patients earlier than clinically appropriate, omit medically indicated
tests and therapies, or over-provide certain services, pushing the patient into a
higher paying DRG in order to optimize the payments they receive. Despite
greater awareness of the need for better monitoring of care quality and patient
outcomes, basic information relating to the quality of services provided is
lacking in most countries in which DRGs are used for hospital payment.
This chapter explores the possible impact of DRG-based hospital payment on
quality of care. We fi rst provide a theoretical discussion of how the quality of
care might be affected in DRG-based payment systems. Quality is defi ned as any
116 Diagnosis-Related Groups in Europe
aspect of the service that benefi ts patients during the process of treatment, or
improves health outcome after treatment (Chalkley & Malcomson, 1998). We
do not focus here on issues related to patient selection and overspecialization
(discussed further in Chapter 6 of this volume), but instead on care quality
following hospital admission. We then review the available evidence concerning
the impact of DRG-based payment on quality of care, including the experience
of countries participating in the EuroDRG project. Finally, based on a review of
the available literature on contracting and the results of a few experimental
payment designs that explicitly take into account quality of care, we discuss
how DRG-based payment systems can be adjusted for quality. We conclude
with some recommendations for ensuring a DRG design which will not lead to
deterioration in the quality of care.
8.2 What the theory suggests
In the health sector, the notion of quality is rather diffuse, since it is diffi cult to
observe and quantify the quality of care provided. Quality of care is a multi-
dimensional concept, covering effectiveness (appropriateness), safety, accessi-
bility and responsiveness of care (Kelly & Hurst, 2006) but there is no agreement
on how these should be measured. A useful and widely used approach is the
one conceptualized by Donabedian (2003) that describes quality measures as
being either structure-, process- or outcome-oriented in nature. Structural
measures – such as qualifi cation of medical staff or equipment levels – may
represent conditions for the delivery of a given quality of health care, but they
are not suffi cient to ensure an appropriate care process. Process measures should
be based on clinical evidence of the effectiveness of the process concerned and
consistent with current professional knowledge (IOM, 2001). However, there is
not always agreement on what is ‘appropriate’ in health care. Thus, process
indicators may be more vulnerable to ‘gaming’ than outcome or structure
measures. While outcome indicators are attractive, it is not always possible to
assess the contribution of care to health outcomes which are infl uenced by
other patient-level factors.
Information about quality – whether in terms of the care structure, process or
medical outcomes – is particularly diffi cult to obtain. Moreover, in the hospital
sector there are several sources of information asymmetries. Patients and
purchasers may not be able to distinguish whether a bad medical outcome is
attributable to the underlying disease or poor quality of care. Individual patients
would have little experience with their specifi c problem to be able to compare
different providers or care procedures. Finally, in some health systems, patients
may not have a choice regarding which hospital to attend. The existence of
information asymmetry implies that payers and patients will have to rely on
the decisions made by the providers. In the ‘agency theory’ framework, the
providers are ‘experts’ who act on behalf of their patients, but patients, pro-
viders and purchasers may have confl icting interests (Forgione et al., 2005).
Providers are interested in recovering their costs, or maximizing profi ts, while
achieving an acceptable level of quality in the market place. Public purchasers
are interested in meeting the health care needs of the population, while
DRGs and quality: For better or worse? 117
controlling costs. They become economically concerned only when the margi-
nal cost of lower quality exceeds the marginal benefi ts from cost-saving policies.
Therefore, the type and quality of treatment provided (clinical discretion) is
a choice variable of the provider and is determined by multiple incentives.
While some incentives are non-fi nancial and can be induced by organizational
culture, leadership, information systems, quality regulations, and so on, the
economic incentives provided by the payment policy would also infl uence how
providers behave in different situations.
As discussed in Chapter 7, under the most basic DRG-based hospital pay-
ment system, hospital revenue (RA) increases linearly with the quantity of
patients treated,1 as follows:
∑
j = 1
]ˆ[Q
j
×RA = pj
J
(1)
where j refers to each DRG category, ( p̂
j
) refers to the fi xed payment for each
patient treated in each DRG and (Q
j
) to the volume of patients. Thus, under this
formulation, hospitals will seek to increase the volume of their activity and
are not incentivized fi nancially to improve the quality of care provided (Street
et al., 2007). In systems where there is an ‘excess demand’ (or undersupply),
stimulating higher production by itself may help to improve quality by reducing
long waiting times (accessibility).
However, in conditions in which it is possible to manipulate treatment
thresholds or in which clinical discretion is high, quality of care may be at risk.
For example, it is diffi cult to ascertain the right amount of diagnostic tests to be
carried out, or in which circumstances a surgical procedure (such as a caesarean
section) is justifi ed.
In the literature on contracting, it is widely recognized that when some
dimensions of the product/service are not visible (not specifi ed in the contract)
providers will be incentivized to withhold or ‘economize’ on the dimensions
that are not verifi able (Chalkley & Malcomson, 1998; Levaggi, 2005). Given
that the treatments provided in a DRG (content) are not always known (badly
defi ned) the providers could decrease resources devoted to the services covered
by the fi xed (DRG) payment and seek to transfer the cost related to other aspects
of care to other providers (cost-shifting). Moreover, Siciliani (2006) shows that
when the information on average severity of the patient is known only by the
provider, they (the hospital) are incentivized to over-provide high-intensity
(surgical) treatment to low-severity patients.
In several countries in which DRG-based hospital payment has been intro-
duced, there has been a signifi cant reduction in the average length of stay
(ALOS) (see Chapter 7). As reducing the length of stay in hospitals has been a
policy objective in many countries (with or without DRG-based hospital
payment), this could be seen as desirable. Shorter hospital stays reduce the risk
of morbidity and may be preferred by patients. However, providers can also
discharge patients prematurely, in an unstable condition. Unfortunately, it is
diffi cult to assess to what extent reductions in length of stay are ‘legitimate’ and
to what extent they are the result of premature discharges.
The way prices are set will have a signifi cant impact on the cost-effi ciency
effort of providers (see Chapter 6) and, consequently, on quality. For example,
118 Diagnosis-Related Groups in Europe
moving from local prices to a national tariff would increase incentives to
control costs. This may reduce incentives for improving quality if quality
implies extra costs. Of course, quality would not be a concern in situations in
which better quality induces costs savings. Moreover, if providers can increase
their profi ts by treating more patients, they have an incentive to attract more
patients – if multiple providers exist – by increasing quality (Farrar et al., 2007).
8.3 Evidence from the literature
The earliest and most comprehensive evidence on the impact of DRG-based
hospital payment on quality comes from the Unites States, where a DRG-based
hospital payment system known as the ‘prospective payment system’ was
implemented in 1983, replacing a cost-based (or fee-for-service) reimbursement
model. The following subsections fi rst review evidence from the United States,
before turning to experiences from Europe.
8.3.1 Evidence from United States studies
In one of the earliest and most signifi cant studies, using a nationally repre-
sentative sample of 14 012 patients hospitalized between 1981/1982 and
1985/1986, the RAND Corporation showed that a prospective payment system
led to a 20 per cent rise in the likelihood that a patient was discharged from
hospital in an unstable condition. However, mortality at 30 and 180 days follow-
ing hospitalization was unaffected (Rogers et al., 1990). The study also looked
at changes in a large number of variables defi ning the process of care, including
cognitive skills of physicians and nurses, as well as technical diagnostic and
therapeutic scales, and it suggested that while the process of care improved after
the introduction of a prospective payment system (better nursing care, better
physician cognitive performance), these improvements in hospital process
began prior to the introduction of the prospective payment system and have
continued after its implementation. Moreover, after the implementation of
the prospective payment system, the ALOS decreased considerably, with no
signifi cant impact on readmission rates, and patients were diagnosed as having
been more ill at the time of admission (Keeler et al., 1990: Kahn et al., 1991).
Other studies also suggested that since the introduction of the prospective
payment system, hospitals have been treating a more severely ill inpatient
population, since less severely ill patients were shifted to outpatient settings
(Newhouse & Byrne, 1988), but it is not clear to what extent this refl ects an
improvement in care organization (better management of cases), and to what
extent it is a selection effect or shift in coding practices. Some of the increase in
severity of illness refl ects hospitals’ efforts to input more co-morbidity codes,
leading to better fi nancial rewards (Feinglass & Holloway, 1991).
In general, the introduction of DRG-based payment has signifi cantly
decreased both the ALOS and the rate of hospital admissions in the United
States (Feinglass & Holloway, 1991). Despite the evidence of some adverse
effects, some of the decline in the number of admissions and the ALOS appears
DRGs and quality: For better or worse? 119
to be related to improvements in organizational effi ciency and quality (utiliza-
tion of new technologies/procedures, development of home or ambulatory
care, and so on).
For example, Schwartz & Tartter (1998) compared the experiences of patients
who underwent colorectal cancer surgery before and after the implementation of
DRG-based hospital payment, in order to identify changes in health care delivery.
Studying a sample of 446 patients treated in a New York hospital they showed
that the mean length of stay was 2.6 days shorter after the introduction of the
DRG system, with a 1.1-day decrease in preoperative and 1.5-day decrease in
postoperative length of stay. DRG patients had signifi cantly less operative blood
loss, fewer transfusions, shorter duration of surgery, and fewer post-operative
complications than the patients treated before the DRG system was implemented.
Measures of disease severity (admission hematocrit, tumour differentiation, and
tumour size) and patient mix (age and gender) did not change. Schwartz and
Tartter (1998) suggested that there have been improvements in operative tech-
niques, but the surgeons may have modifi ed certain aspects of treatment in order
to reduce length of stay without adversely affecting the quality. The signifi cant
decrease in preoperative length of stay may be due to organizational changes,
shifting preoperative assessment to out-patient settings.
Nevertheless, the prospective payment systems may have had contradictory
effects for different patient groups, depending on the price incentives provided
by the different DRGs. Gilman (2000) examined the effect of DRG refi ne-
ment for HIV infection in 1994 in the United States, where the prices of non-
procedural DRGs were generally lowered and those of procedural DRGs were
raised. He demonstrated that in the New York State hospital length of stay for
lower priced non-procedural DRGs declined by 3.3 days from 1992 to 1995,
while length of stay for better paid procedural DRGs increased by 1.1 days on
average over the same period.
However, the pressure for cost-containment created by the DRG-based pay-
ment system can also adversely affect care quality. Cutler (1995) demonstrated
that the impact of prospective payment systems may depend on the hospitals’
economic situation (effi ciency) before the prospective payment system was
implemented. Using a longitudinal dataset of about 40 000 hospital admissions
(from 1981 to 1988) in New England, he showed that hospitals experiencing
average price declines (historical costs higher than DRG prices) had a ‘compres-
sion’ of mortality rates, with more deaths occurring in hospital or within two
months after discharge, while overall one-year death rates remained the same.
Reductions in average prices (revenues) may force hospitals to cut back on treat-
ment intensity and/or other inputs. Cutler also found that there was an increase
in readmission rates caused by the introduction of the prospective payment
system, without any apparent change in sickness levels.
In a similar (more recent) study, Shen (2003) showed that fi nancial pressure
from the prospective payment system adversely affected short-term health
outcomes after treatment for acute myocardial infarction (AMI), but did not
affect patient survival beyond one year after admission.
Some evidence from the United States suggests that the introduction of DRG-
based payment in rehabilitation/nursing facilities had a similar impact on the
quality of rehabilitative and post-acute care. After the implementation of the
120 Diagnosis-Related Groups in Europe
new payment approach, patients appeared to have shorter lengths of stay, with
lower functional levels at discharge and higher institutional discharge rates
(Gillen et al., 2007; Buntin et al., 2009). Moreover, both emergency readmissions
and deaths within 60 days of discharge increased signifi cantly for patients with
chronic obstructive pulmonary disease (COPD), although some other outcomes
of post-acute care were not affected (McCall et al., 2003).
8.3.2 Evidence from Europe
The evidence from Europe is scarce and less clear cut. In Sweden and Finland,
where the incentives of DRG-based hospital payment are moderated by locally
adjusted monetary conversion rates and additional payment components
(see Chapter 6), it is believed that DRGs have helped with homogenizing care
procedures and have improved inpatient care organization. However, in both
countries there are no direct indicators of care quality, treatment and access
associated with the DRG system. In Sweden, most hospitals contribute to
quality registers, but quality monitoring appears to be independent of the DRG-
based payment system. A longitudinal study of patient-reported quality of care
in two Swedish hospitals suggested that the quality of care as perceived by
patients – especially with respect to treatment by staff – decreased after the
introduction of DRG-based payment (Ljunggren & Sjödén, 2001) but had no
effect on quality of life after surgery (Ljunggren & Sjödén, 2003). The evidence
from Sweden also confi rmed that the introduction of DRG-based payments
contributed to an increase in re-coding diagnoses and increased the number of
secondary diagnoses recorded per case (Serdén et al., 2003). In Finland, com-
parison of outcomes across hospitals is based on specifi c diseases or procedures,
and this information is used only for benchmarking.
An early study of four Norwegian hospitals suggests that the DRG-based pay-
ment system did not have any impact on hospital-acquired infections (Pettersen,
1995), although there was some evidence of cream-skimming in the immediate
period after DRG-based hospital payment was introduced in Norway in 1997
(Martinussen & Hagen, 2009).
Using data from one region (Friuli) and 32 hospitals over the period 1993–
1996, Louis and colleagues (1999) found for Italy – where a DRG-based pay-
ment system was introduced at national level in 1995 – that the total number
of hospital admissions decreased by 17 per cent, while day-case hospital use
increased sevenfold. They also found that the mean length of stay decreased
(resulting in a 21 per cent decrease in hospital bed days) for most conditions,
while severity of illness increased without any signifi cant change in mortality
or readmission rates.
A formal evaluation of DRG-based hospital payment in England, locally
referred to as Payment by Results (PbR), also showed that while the ALOS has
decreased signifi cantly in settings in which PbR was implemented, little measur-
able change has occurred in the quality of care in terms of inpatient (in-hospital)
mortality, 30-day post-surgical mortality and emergency readmissions after treat-
ment for hip fracture (Farrar et al., 2009). The Audit Commission (2008) con-
cluded that PbR has not had a measurable impact on quality of care in England.
DRGs and quality: For better or worse? 121
No other formal evaluation of the impact of DRG-based payment on quality
of care is available from other European countries. It appears that in most
countries in which a DRG-based hospital payment system is introduced, the
monitoring and reporting of care quality remains inadequate. For example,
both in Germany and France, there is still no systematic information system to
monitor readmission rates, postoperative mortality and complication rates.
In Germany, a survey of 30 hospitals in Lower Saxony suggested that the
introduction of DRG-based payment did not create cream-skimming or early
discharge problems in these hospitals (Sens et al., 2009). Based on interviews
with hospital managers, health professionals and patients, the study suggested
that service quality appeared to be steady over the period 2007–2008, and may
even have improved due to better care organization, especially in large hospitals.
Nevertheless, this study did not analyse any concrete measures of patient
outcomes or care quality.
In France, there is evidence that up-coding might be a concern. External
control efforts by the health insurance fund(s) revealed quickly that a signifi cant
proportion of the increase in day cases was due to incorrect coding of outpatient
consultations (CNAM, 2006). While this problem has been partly resolved with
stricter coding rules for day cases, introduced in 2007, further attention was
required to address the pertinence of some day-case procedures, which have
been increasingly signifi cantly (see Chapter 13 of this volume).
8.4 Integrating quality into payment
Unintended adverse effects of DRG-based hospital payment systems on care
quality could potentially be avoided by modifying the incentives of the
payment system. If the payer/purchaser wants to improve quality of care,
payments need to be adjusted in a way that rewards hospitals for the additional
costs/effort involved in raising quality. Chalkley & Malcomson (1998) suggest,
furthermore, that the form of the payment contract should take into account
the type of provider (public, profi t-making, non-profi t-making) and should be
adjusted carefully by the purchasers, depending on the objectives pursued
(maintaining a certain level of quality while reducing costs, improving quality,
and so on).
Different options exist for adjusting DRG-based hospital payment systems on
the basis of quality of care. Simplifi ed, there are three options: (1) the hospital
level, (2) the level of a DRG-or all DRGs for one condition, and (3) the individual
patient level.
Under the fi rst option, total hospital income could be adjusted on the basis
of hospital-level quality indicators:
∑
j = 1
ˆ[Q
j
×RA = pj] + p
hqh
J
(2)
where qh is an index of quality measured at hospital level and ph is payment
(price) per unit change on this quality scale. Given the diffi culties and cost
of measuring treatment-specifi c outcomes at patient level, hospitals can be
rewarded for quality improvements or progress in the care process, given a
122 Diagnosis-Related Groups in Europe
national framework. This is appropriate if quality is independent of the volume
of activity. Otherwise, contracts using a price which varies by volume of patients
treated could be more effi cient (Chalkley & Malcomson, 1998).
One example of hospital-level quality adjustment is the approach adopted in
England according to the Commissioning for Quality and Innovation (CQUIN)
framework, which came into effect in April 2009. Within this framework, all
acute trust hospitals collect patient-related outcome measures and report on
quality in order to publish ‘quality accounts’ alongside their fi nancial accounts.
Subsequently, Primary Care Trusts (PCTs) can link a specifi c modest proportion
of providers’ income (agreed nationally) to the achievement of realistic locally
agreed goals. In 2009/2010 the CQUIN payment framework covered 0.5 per
cent of a provider’s annual contract income (Department of Health, 2008),
and this proportion increased to 1.5 per cent in 2010/2011 (Department of
Health, 2010). Along a similar line, the Centers for Medicare and Medicaid
Services (CMS) in the United States will lower DRG payments for all patients in
hospitals – initially by up to 1 per cent – with above-average readmission rates
for congestive heart failure, pneumonia and AMI from October 2012. Two years
later, COPD, coronary artery bypass graft, percutaneous coronary intervention
and other vascular procedures will be included in the calculation – and penalties
will increase to 2 per cent in 2014 and 3 per cent in 2015.
Under the second option, when patient-level data are available on outcomes
and/or treatment process(es), payments can be adjusted for certain DRGs based
on the quality of all patients treated within that DRG. The aim is to encourage
medical practice that is considered to be ‘good quality’ by moving away from
pricing simply based on average observed costs per episode. However, this
requires reliable indicators of patient-level data and agreement on what
constitutes ‘good quality’. In this case, both quality measurement and payments
are DRG specifi c, as follows:
∑ ∑
= 1
ˆ[Q
j
× p
j
] + [(qi
j
Q
j
) × p�
j
]RA =
j = 1j
J J
(3)
where p�
j
corresponds to the price for the ‘good quality’ care practice for patients
of a given DRG ( j), and quality is measured at individual DRG level. The price
paid for good quality ( p�
j
) could be higher or lower than the average cost of an
episode, depending on what is considered ‘good’ or ‘best’ compared to average/
common practice. In England, ‘best practice tariffs’ have recently been
introduced for four areas (cholecystectomy, hip fractures, cataracts and stroke),
whereby signifi cant unexplained variation in quality of clinical practice is
observed and clear evidence of what constitutes best practice is available
(Department of Health, 2011). Best practice tariffs are set to incentivize day-case
activity for cholecystectomy, while for cataract treatment the price covers the
entire care pathway, so that commissioners only pay for events in the best
practice (streamlined elective cataract) pathway, in which patients are treated
in a ‘joined-up’ and effi cient manner. For hip fracture and stroke, prices are
adjusted upwards if key clinical characteristics of best practice care are met
(with corresponding lower payment for non-compliance).
In practice, outcome-based adjustment can also be carried out for specifi c
diseases, such as AMI, stroke (Ash et al., 2003; Iezzoni, 2003) or for procedures
DRGs and quality: For better or worse? 123
deemed effective (Nashef et al., 1999) that are not related to specifi c DRGs. In
Germany, one example of such a quality adjustment is the ‘integrated care’
contract between a large German sickness fund (Techniker Krankenkasse) and
the Karlsruhe heart surgery hospital, which has been in place since 2005. Under
the terms of the contract, the hospital receives higher payments for coronary
bypass surgery patients if it scores above the national average on a set of heart
surgery quality indicators, which are collected as part of the German external
quality assurance system (see Busse et al., 2009). Similarly, in the Netherlands,
the original purpose of introducing DBCs was to allow insurers to negotiate
with hospitals regarding price, volume and quality of care (which purchasers
are currently allowed to do for about 30 per cent of DBCs). However, it would
appear that insurers and hospitals negotiate predominantly on price and
volume, whereas quality plays only a minor role in the negotiation process.
However, it is challenging to integrate in the payment system an implicit set
of clinical guidelines defi ning how to treat a homogeneous group of patients,
approximating a contract that specifi es what is ‘good quality’ for specifi c DRGs
(Newhouse, 2003). Clearly, the condition for such contracts is a consensus on
what constitutes ‘good-quality’ care in different clinical contingencies. The lack
of clinical consensus on the guidelines to be used – even in cases of common
problems, such as heart attacks – is well documented (Baker et al., 2008; Phelps,
2000) and remains a major obstacle to quality-based contracting. Whether or
not best practice tariffs can contribute to improving quality remains to be seen.
The third option is to adjust payments for individual patients based on the
quality of their treatment, independent of the DRG to which they are allocated.
Hospital contracts could be simply modifi ed to take into account the quality of
care provided, as follows:
∑ ∑
= 1
ˆ[Q
j
× p
j
] + [(qi
j
Q
j
) × pi]RA =
j = 1j
J J
(4)
where qi is the patient-level quality index (which could be simply 0, 1) and pi is
the price for individual-level quality (or non-quality). The revenue (R) of
providers depends on the number of patients treated Q
j
as well the quality of
treatment and its price, irrespective of the DRG in which patients are placed.
This requires reliable indicators of patient outcomes.
Developing such indicators is not always straightforward, as attributing a
certain patient outcome to provider behaviour (rather than to patient health
status) can be controversial. Indicators for bad (or good) quality, on which such
penalties (or rewards) are based, will thus need to be very robust and subject to
as little controversy as possible.
Patient-level quality adjustment policies so far have focused on disentangling
complications (caused by the hospital) from co-morbidities (which the patient
already has upon admission), as well as on readmissions.
The best-known example of this is the United States Medicare policy, whereby
the CMS require hospitals to use ‘present-on-admission’ codes for both primary
and secondary diagnoses when submitting claims for discharges. Since October
2008, diagnosis codes for ten selected conditions – such as pressure ulcers;
‘dislocation of patella open’ due to a fall; catheter-associated urinary tract
infection – are excluded from consideration during the grouping process if they
124 Diagnosis-Related Groups in Europe
were not coded as being present on admission (that is, they were contracted
during the hospital stay) (Department of Health and Human Services, 2008).
Consequently, these codes cannot lead to the classifi cation of patients into
higher-paying DRGs, and Medicare no longer has to pay for the extra costs of
these avoidable hospital-acquired conditions. It is estimated that about 15 per
cent of the claims had a ‘non-present on admission’ diagnosis (Zhan et al.,
2007). While this approach to reducing adverse events is considered attractive
by some (McNair et al., 2009), others highlight the diffi culty of determining
what are avoidable adverse events (Provonost et al., 2008). Furthermore, ensur-
ing accurate and thorough coding of hospital diagnoses is challenging. Penaliz-
ing or rewarding hospitals based on their diagnosis coding could heighten the
risks of ‘gaming’ or coding manipulation (Iezzoni, 2009).
Another patient-based alternative for integrating quality into DRG-based
hospital payment systems is to extend the treatment episode for which a DRG-
based payment is granted; that is, by including outpatient visits, readmissions,
and so on. In England and Germany, hospitals do not receive a second DRG
payment if a patient is readmitted for the same condition within 30 days after
discharge. Ideally, it is desirable to extend the payment for an integrated set of
treatments, including outpatient visits, rehabilitation, and so on, but this is
challenging and requires a sophisticated integrated information system. In the
Netherlands, the DBC-based DRG system covers the whole spectrum of inpatient
and outpatient care provided at hospitals, relating to a specifi c diagnosis from
the fi rst specialist visit to the end of the care process (treatment completed) and
including inpatient days, outpatient visits, laboratory services, medical imaging
services, medications, medical materials, (surgical) procedures, and so on.
Consequently, as long as a patient is treated for the same condition, the hospital
does not receive an extra payment. However, the Dutch system does not provide
incentives to reduce postoperative infections or readmission rates, since these
are coded as new DBCs.
Of course, it is also possible to have a system which combines different
approaches, for example: quality adjustments at the patient level with a global
payment/adjustment for quality at the hospital level. However, and essential
prerequisite for any quality-based payment adjustments to the hospital pay-
ment system is the availability of information on quality of care. Therefore,
several countries have increased their efforts to collect quality information (for
example, BQS/AQUA2 in Germany (Busse et al., 2009), COMPAQH in France)
but routinely available information on patient outcomes is still scarce. The
importance of having better information regarding the quality of care is evi-
denced by the existence of specifi c fi nancial incentives to hospitals for report-
ing quality information. For example, Medicare in the United States encourages
hospitals to participate in public reporting of quality information. Those hospi-
tals that do not report on 10 measures of quality (defi ned by the Hospital Qual-
ity Alliance) receive a 0.4 per cent reduction in their DRG prices. In Germany,
hospitals are fi nancially penalized if they report quality information for less
than 80 per cent of treated cases (Busse et al., 2009). The pertinence of using the
act of reporting quality data as a proxy for quality of care delivery is question-
able, but – when data are available – hospitals can also be offered positive incen-
tives for their effort or extra payments can be made for stimulating innovative
DRGs and quality: For better or worse? 125
approaches to improving quality and patient safety. However, caution is called
for before implementing any such schemes, as providers could be destabilized
if their revenues fl uctuate signifi cantly from one year to another.
8.5 Conclusions
The effects of DRG-based hospital payment systems on patient outcomes and
quality of care have long been debated. In many countries, health profes-
sionals have expressed concern that these systems may lead to a focus on cost-
containment efforts at the expense of quality of care. Based on theoretical
considerations and a review of the available literature, this chapter suggests
that DRG-based payment systems may represent risks for quality of care, but
may also provide opportunities for quality improvements. The introduction of
DRGs has increased transparency and has facilitated comparison and
standardization of care. The pressure for effi ciency introduced by DRG-based
payment systems might help to improve organization of care, accelerate the
adoption of technology, and hence improve quality. Nevertheless, hospitals
can also skimp on quality as a way of saving costs by manipulating the services/
care provided to patients. Technology adoption rates may decelerate if new
technologies do not induce cost-savings (see Chapter 9). At the same time,
these potential adverse effects are not inevitable consequences of DRG-based
hospital payment and can be addressed by carefully designing the payment
scheme.
The evidence from the United States suggests that, on the one hand, the
introduction of DRG-based payment has improved organizational effi ciency
and quality of care in some areas, in particular by stimulating better options for
ambulatory and home care. On the other hand, there is evidence that the cost-
containment pressure created by the introduction of DRG-based payment can
have an adverse impact on patient outcomes in terms of readmission and mor-
tality rates. Different patient groups can also experience various impacts,
depending on the price incentives provided by different DRGs. Particular atten-
tion appears to be necessary to ensure that high-severity groups are adequately
accounted for in the DRG system, in order to avoid quality of care being
adversely affected for these patients.
In Europe, despite the widespread introduction of DRG-based hospital pay-
ment systems since the early 2000s, the available research evaluating the sys-
tems’ impact on care quality and patient outcomes is too limited to draw any
fi rm conclusions. The limited evidence so far does not suggest that the intro-
duction of DRG-based hospital payment had a signifi cant impact on patient
outcomes (as measured by readmission and mortality rates). Thus, some of the
adverse effects observed in the United States are not confi rmed by evidence
from Europe. Clearly, the impact of DRGs on quality would depend on the
model adopted and the regulatory and health care context of each country.
Because DRG-based hospital payment systems in Europe generally speaking did
not replace fee-for-service systems, but rather replaced per diem-based pay-
ments or global budgets (see Chapter 2), the effect of DRG-based hospital pay-
ments on quality of care might also be different in Europe from that experienced
126 Diagnosis-Related Groups in Europe
in the United States. In addition, the pressure to contain costs is possibly weaker
in many European countries than in the United States, because of the stronger
presence of both public providers and public regulator in the hospital sector.
If this is true, any adverse impact on quality would also be weaker.
In basic DRG-based hospital payment systems, health care providers are not
explicitly rewarded for improving quality. Therefore, these schemes need to
be refi ned in order to integrate direct incentives for improving quality. This
chapter provides some examples of how this could be carried out. Nevertheless,
caution is called for when implementing any such schemes. A balance needs
to be struck between the positive motivational effects and the potentially
destabilizing effect of penalties for providers (Maynard & Bloor, 2010). Also,
providers may focus too much on those areas in which payments are linked
to measured quality improvements, to the detriment of some other (non-
measured) aspect(s) of care. Therefore, careful piloting and evaluation of such
schemes is essential.
DRG-based hospital payment provides an opportunity to better measure
quality of care in hospitals. Thus, it becomes possible to improve quality by
providing explicit incentives for higher quality procedures/treatments,
penalizing ‘poor-quality care’ or granting funds for improving patient outcomes.
This requires continuous refi nement of data and indicators for monitoring
quality of care. In many countries, information on patient outcomes and
process quality is not routinely collected. However, if fi nancing arrangements
become more sophisticated, the demand for and supply of information
regarding quality of health care will surely increase.
8.6 Notes
1 In practice, in all countries, hospitals receive some fi xed payments independent of
their activity to cover the fi xed costs of providing certain services, such as education
and research. For the sake of simplicity, these are not discussed here.
2 Federal Offi ce for Quality Assurance/AQUA-Institute for Applied Quality Improvement
and Research in Health Care.
8.7 References
Ash, A.S., Posner, M.A., Speckman, J. et al. (2003). Using claim data to examine mortality
trends following hospitalization for heart attack in Medicare. Health Services Research,
38:1253–62.
Audit Commission (2008). The Right Result? Payment by Results 2003–2007. London: Audit
Commission (http://www.audit-commission.gov.uk/SiteCollectionDocuments/Audit
CommissionReports/NationalStudies/The_right_result_PbR_2008 , accessed 29
June 2011).
Baker, L., Fisher, E.S., Wennberg, J. (2008). Variations in hospital resource use for Medicare
and privately insured populations in California. Health Affairs, 27(2):123–34.
Buntin, M., Colla, C., Escarce, J. (2009). Effects of payment changes on trends in post-
acute care. Health Services Research, 44(4):1188–210.
Busse, R., Nimptsch, U., Mansky, T. (2009). Measuring, monitoring, and managing quality
in Germany’s hospitals. Health Affairs (Millwood), 28:294–304.
DRGs and quality: For better or worse? 127
Chalkley, M., Malcomson, J.M. (1998). Contracting for health services when patient
demand does not refl ect quality. Journal of Health Economics, 17:1–19.
CNAM (2006). Contrôles et lutte contre les abus et les fraudes. Paris: Caisse National
d’Assurance Maladie (http://www.securite-sociale.fr/institutions/fraudes/fraude.htm,
accessed 4 July 2011).
Cutler, D.M. (1995). The incidence of adverse medical outcomes under prospective
payment. Econometrica, 63(1):29–50.
Department of Health (2008). Using the Commissioning for Quality and Innovation (CQUIN)
Payment Framework. London: Department of Health.
Department of Health (2010). Using the Commissioning for Quality and Innovation (CQUIN)
Payment Framework – A Summary Guide. London: Department of Health.
Department of Health (2011). Best Practice Tariffs. London: Department of Health (http://
www.dh.gov.uk/en/Managingyourorganisation/NHSFinancialReforms/DH_105080,
accessed 27 July 2011).
Department of Health and Human Services (2008). Changes to the Hospital Inpatient
Prospective Payment Systems and Fiscal Year 2009 Rates. Washington, DC: Federal
Register (Vol. 73, No. 84).
Donabedian, A. (2003). An Introduction to Quality Assurance in Health Care. Oxford: Oxford
University Press.
Ellis, R.P., McGuire, T.G. (1996). Hospital response to prospective payment: Moral hazard,
selection and practice style effects. Journal of Health Economics, 15:257–77.
Farrar, S., Sussex, J., Yi, D. et al. (2007). National Evaluation of Payment by Results. Aberdeen:
University of Aberdeen Health Economics Research Unit.
Farrar, S., Yi, D., Sutton, M. et al. (2009). Has payment by results affected the way that
English hospitals provide care? Difference-in-differences analysis. British Medical
Journal, 339:1–8.
Feinglass, J., Holloway, J.J. (1991). The initial impact of the Medicare prospective payment
system on United States health care: a review of the literature. Medical Care Review,
48(1):91–115.
Forgione, D.A., Vermeer, T.E., Surysekar, K., Wrieden, J.A., Plante, C.A. (2005). DRGs, costs
and quality of care: an agency theory perspective. Financial Accountability & Manage-
ment, 21(3):291–307.
Gillen, R., Tennen, H., McKee, T. (2007). The impact of the inpatient rehabilitation facility
prospective payment system on stroke program outcomes. American Journal of Physical
Medicine & Rehabilitation, 86(5):356–63.
Gilman, B.H. (1999). Measuring hospital cost-sharing incentives under refi ned prospective
payment. Journal of Economics and Management Strategy, 8(3):433–52.
Gilman B.H. (2000). Hospital response to DRG refi nements: the impact of multiple reim-
bursement incentives on inpatient length of stay. Health Economics, 9(4):277–94.
Iezzoni, L.I. (2003). Risk Adjustment for Measuring Health Care Outcomes. Chicago: Health
Administration Press.
Iezzoni, L. (2009). Reinvigorating the quality improvement incentives of hospital
prospective payment. Medical Care, 47(3):269–71.
IOM (2001). Crossing the Quality Chasm: A New Health System for the 21st Century.
Washington DC: National Academy Press.
Kahn, K., Draper, D., Keeler, E. et al. (1991). The Effects of the DRG-Based Prospective Pay-
ment System on Quality of Care for Hospitalized Medicare Patients. Pittsburgh, PA: RAND
Corporation.
Keeler, E.B. (1990). What proportion of hospital cost differences is justifi able? Journal of
Health Economics, 9:359–65.
Kelly, E., Hurst, J. (2006). Health Care Quality Indicators Project: Conceptual Framework Paper.
Paris, Organisation for Economic Co-operation and Development (OECD Health
Working Papers, No. 23).
128 Diagnosis-Related Groups in Europe
Levaggi, R. (2005). Hospital health care: pricing and quality control in a spatial model
with asymmetry of information. International Journal of Health Care Finance and
Economics, 5(4):327–49.
Ljunggren, B., Sjödén, P.O. (2001). Patient-reported quality of care before vs. after the
implementation of a diagnosis-related groups (DRG) classifi cation and payment system
in one Swedish county. Scandinavian Journal of Caring Sciences, 15(4):283–94.
Ljunggren, B., Sjödén, P.O. (2003). Patient-reported quality of life before, compared with
after a DRG intervention. International Journal for Quality in Health Care, 15(5):
433–40.
Louis, D., Yuen, E.J., Braga, M. et al. (1999). Impact of a DRG-based hospital fi nancing
system on quality and outcomes of care in Italy. Health Services Research, 34(1 Pt 2):
405–15.
Martinussen, P., Hagen, T. (2009). Reimbursement systems, organisational forms and
patient selection: evidence from day surgery in Norway. Health Economics, Policy and
Law, 4:139–58.
Maynard, A., Bloor, K. (2010). Will fi nancial incentives and penalties improve hospital
care? British Medical Journal, 340:c88.
McCall, N., Korb, J., Petersons, A., Moore, S. (2003). Reforming Medicare payment: early
effects of the 1997 Balanced Budget Act on post-acute care. The Milbank Quarterly,
81(2):277–303.
McNair, P., Borovnicar, D., Jackson, T., Gillett, S. (2009). Prospective payment to encourage
system-wide quality improvement. Medical Care, 47(3):272–8.
Miraldo, M., Goddard, M., Smith, P. (2006). The Incentive Effects of Payment by Results.
York: University of York Centre for Health Economics (CHE Research Paper 19).
Nashef, S.A.M., Roques, F., Michel, E. et al. (1999). European system for cardiac operative
risk evaluation (EuroSCORE). European Journal of Cardio-thoracic Surgery, 16:9–16.
Newhouse, J.P. (2003). Reimbursing for health care services. Economie Publique, 13:3–31.
Newhouse, J.P., Byrne, D.J. (1988). Did Medicare’s prospective payment system cause
length of stay to fall? Journal of Health Economics, 7:413–16.
Pettersen, K.I. (1995). Hospital infections as quality indicators. DRG-based fi nancing, did
it change therapeutic quality? [in Norwegian] Journal of the Norwegian Medical
Association, 115(23):2923–7.
Phelps, C.E. (2000). Information diffusion and best practice adoption, in J.P. Newhouse,
A. Culyer, eds. Handbook of Health Economics. Amsterdam: North Holland Press.
Pronovost, P., Goeschel, C., Wachter, R. (2008). The wisdom and justice of not paying for
‘preventable complications’. Journal of the American Medical Association, 299(18):
2197–9.
Rogers, W.H., Draper, D., Kahn, K.L. et al. (1990). Quality of care before and after
implementation of the DRG-based prospective payment system. A summary of
effects. Journal of the American Medical Association, 264:1989–94.
Schwartz, M.H., Tartter, P.I. (1998). Decreased length of stay for patients with colorectal
cancer: implications of DRG use. Journal for Healthcare Quality, 20(4):22–5.
Sens, B., Wenzlaff, P., Pommer, G., von der Hardt, H. (2009). DRG-induzierte Veränderungen
und ihre Auswirkungen auf die Organisationen, Professionals, Patienten und Qualität.
Hanover: Zentrum für Qualität und Management im Gesundheitswesen, Einrichtung
der Ärztekammer Niedersachsen.
Serdén, L., Lindqvist, R., Rosen, M. (2003). Have DRG-based prospective payment systems
infl uenced the number of secondary diagnoses in health care administrative data?
Health Policy, 65(2):101–7.
Shen, Y.C. (2003). The effect of fi nancial pressure on the quality of care in hospitals.
Journal of Health Economics, 22(2):243–69.
Siciliani, L. (2006). Selection of treatment under prospective payment systems in the
hospital sector. Journal of Health Economics, 25(3):479–99.
DRGs and quality: For better or worse? 129
Street, A., Vitikainen, K., Bjorvatn, A., Hvenegaard, A. (2007). Introducing Activity-Based
Financing: A Review of Experience in Australia, Denmark, Norway and Sweden. York:
University of York Centre for Health Economics (CHE Research Paper 30).
Zhan, C., Elixhauser, A., Friedman, B., Houchens, R., Chiang, Y. (2007). Modifying DRG-
PPS to include only diagnosis present at admission. Medical Care, 45(4):288–91.
chapter n i n e
Technological innovation in
DRG-based hospital payment
systems across Europe
Wilm Quentin, David Scheller-Kreinsen
and Reinhard Busse
9.1 Introduction
Technological innovation in health care is highly valued by patients, clinicians
and politicians (Rettig, 1994), as advances in medical technology have greatly
improved the ability to prevent, diagnose and treat a large number of diseases
and conditions, leading to reduced mortality and better quality of life in many
countries (Atella & The TECH Investigators, 2003; Cutler & McClellan, 2001;
Cutler, 2007; Tunstall-Pedoe et al. 2000). At the same time, technological
innovation is a major driver of increasing health care costs (Weisbrod, 1991;
Cutler et al., 1998a; Congressional Budget Offi ce 2008), and policies have been
devised with the aim of balancing technological innovation and affordability
(Schreyögg et al., 2009).
The hospital payment system is one important factor infl uencing the
implementation of technological innovation in health care (Greenhalgh et al.,
2004; Banta, 1983; Torbica & Cappellaro, 2010; Atella & the Tech Investigators,
2003; Cappellaro et al., 2011), especially as many new technologies are fi rst
used in the inpatient sector. Ever since the introduction of diagnosis-related
group (DRG)-based hospital payment systems, there have been concerns that
these systems may not provide the right set of incentives to encourage the
desired adoption and use of technological innovations in health care (OTA,
1983; Garrison & Wilensky, 1986; MedPAC, 2001; Shih & Berliner, 2008). Con-
sequently, mechanisms have been developed by most countries using DRG-
based hospital payment systems to account for technological innovation in
health care (MedPAC, 2003; Clyde et al., 2008; Schreyögg et al., 2009; Henschke
et al., 2010).
132 Diagnosis-Related Groups in Europe
This chapter aims to (1) clarify the relationship between DRG-based hospital
payment systems and technological innovation; and (2) to describe how the 12
countries included in this book attempt to overcome the potential problems for
technological innovation associated with DRG-based hospital payment systems.
The following section (9.2) provides a theoretical overview of the relationship
between technological innovation, hospital costs and quality, in order to ex-
plain how DRG-based hospital payment systems can potentially affect the
adoption and diffusion of technological innovations. Subsequently, section 9.3
presents a comparative analysis of the analysed countries’ policy responses to
the problems of encouraging technological innovations and incorporating
them formally into DRG-based hospital payment systems. Section 9.4 sum-
marizes the fi ndings and draws some conclusions for European countries
regarding how best to deal with technological innovations in the context of
DRG-based hospital payment systems.
9.2 Technological innovation and DRG-based hospital
payment in theory: Costs, quality and the
adequacy of payment
9.2.1 Technological innovation: Costs and quality
Technological innovation is often defi ned as the practical application and
diffusion of ideas or knowledge (Goodman, 2004). In health care, innovations
can potentially refer to all categories of medical technology, such as drugs,
devices, equipment and supplies, medical and surgical procedures, support
systems and organizational and managerial systems (Banta et al., 1978). Techno-
logical innovation may be incremental, consisting of small improvements of
existing services; or it may comprise radical changes, such as replacing surgi-
cal therapy with new medical therapy. Finally, technological innovation may
occur as a transfer or adaptation of existing technology from one setting to
another, for example the shift of certain procedures from inpatient settings to
day care.
When analysing the implications of DRG-based hospital payment systems
for the adoption and diffusion of technological innovation in health care, it is
essential to consider the effects of technological innovations on hospital costs
and quality. Table 9.1 illustrates possible effects of technological innovations
on hospital costs: such innovations may increase or decrease capital costs, oper-
ating costs or both (OTA, 1983). Yet, the overall effect on hospital costs depends
on the interplay of various factors. For example, in large or highly specialized
hospitals, an increase in capital costs might be compensated by reductions in
operating costs, if capital costs can be distributed among a suffi ciently large
number of patients. In small or less-specialized hospitals, the effect of the same
technological innovation on costs may be different. Furthermore, it is impor-
tant to bear in mind that technological innovations may be related to costs of
all hospital services (for example, the introduction of electronic medical
records) or may only affect the costs of treating a small and very specifi c group
of patients (for example, the introduction of drug-eluting stents).
Technological innovation in DRG-based hospital payment systems 133
Quality in health care can be defi ned as any aspect of health services that ben-
efi ts patients during the process of treatment or improves health outcome after
treatment (see Chapter 8). In theory, the effect of technological innovation on
‘quality’ can be positive, neutral or negative. Figure 9.1 illustrates different theo-
retical combinations of costs and quality that can result from the introduction of
technological innovations, using a graphical illustration similar to that of the
cost–effectiveness plane (Black, 1990). Technological innovations can increase
both costs and quality (A), increase quality while decreasing costs (B), decrease
both costs and quality (C), or increase costs while decreasing quality (D).
Whether or not the incentives of DRG-based hospital payment systems pro-
duce socially desirable effects depends on the specifi c combination of costs and
quality; that is, it depends on the quadrant (A to D) into which the new tech-
nology would be classifi ed. Technologies falling into quadrant B would be
always desirable, whereas technologies falling into quadrant D should never be
Table 9.1 Possible effects of technological innovation on hospital costs
Technological innovation Effect on costs
capital operating total
Cost-increasing technology + + +
Cost-decreasing technology – – –
Capital cost-increasing technology + – +/–
Operating cost(s)-increasing technology – + +/–
Source: OTA, 1983, with modifi cations.
Figure 9.1 Effects of innovation on cost and quality
Source: Adapted from Black, 1990.
134 Diagnosis-Related Groups in Europe
adopted. For technologies falling into quadrants A and C, things are more com-
plicated (Cutler & McClellan, 2001). For quadrant A, the country-specifi c will-
ingness to pay for a given increase in quality determines whether these
technological innovations should be used in hospitals. Conversely, for tech-
nologies falling into quadrant C, it depends on the extent to which countries
are willing to forego quality in exchange for a decrease in costs.
However, the effects (especially long-term effects) of technological innovation
on quality (and on costs) are often diffi cult to identify at the time, when
innovations are fi rst introduced into hospital practice (Mowatt et al., 1997). In
fact, a variety of technologies have been found to be ineffective or even harmful
after having been widely adopted and used (Goodman, 2004). Consequently,
policy-makers face considerable uncertainty when making decisions about
technological innovations.
9.2.2 DRG-based hospital payment systems: Incentives
against technological innovation?
Under DRG-based hospital payment systems, clinicians are free to decide on the
exact set of technologies that they want to employ when treating a given
patient. However, as outlined in Chapter 6, DRG-based hospital payment
systems provide a specifi c set of incentives to hospitals that are likely to have an
effect on shaping the clinicians’ decisions. This subsection investigates how
these incentives may infl uence the use of technological innovations in hospitals.
Under the most basic DRG-based hospital payment system, introduced in
Chapter 6, hospitals are paid a predetermined fi xed payment rate per case.
Consequently, hospitals are encouraged to keep their average costs below the
payment rate in order to avoid making a loss. Thus, the two dominant incentives
of a basic DRG-based hospital payment system encourage hospitals to (1) reduce
costs per admission, and (2) increase the number of admissions (OTA, 1983).
The effects of these incentives on the hospitals’ willingness to adopt and to
use technological innovations are summarized in Table 9.2. Hospitals are likely
to invest in technological innovations that reduce total costs per admission.
They may purchase new diagnostic equipment or electronic drug interaction
monitoring systems if these can be shown to reduce costs per stay – for example,
by reducing length of stay. In cases in which technological innovations are cost
neutral, or in which increases in one area can be compensated by decreasing
costs in another area, DRG-based hospital payment systems should have no
effect on the introduction of technological innovations. Furthermore, as tech-
nological innovation often increases capital costs, DRG-based hospital payment
systems might encourage the specialization of hospitals (if separate funding for
capital costs is unavailable), concentrating the adoption of technological inno-
vations in centres with suffi ciently large numbers of patients. In addition, as
hospitals bear the fi nancial risk of average costs rising above the payment rate,
hospitals are likely to make use of economic evaluations before introducing
certain technological innovations.
As far as many cost-decreasing, cost-neutral or cost-increasing but quality-
decreasing technological innovations are concerned (quadrants B to D in
Technological innovation in DRG-based hospital payment systems 135
Figure 9.1), DRG-based hospital payment provides incentives that are likely to be
in line with societal objectives: they encourage adoption of technological inno-
vations in quadrant B and C, and inhibit technological innovation in quadrant
D. However, economic evaluations and country-specifi c value-judgements are
required in cases in which cost-decreasing technological innovations are accom-
panied by decreases in quality (quadrant C), as it should be determined whether
the decrease in quality outweighs the reduction in costs (Drummond et al., 2005).
Problems with DRG-based hospital payment occur when technological
innovations improve quality but are associated with increased costs per
admission (quadrant A in Figure 9.1). In most countries, DRG-based payment
rates are at least remotely related to the average costs of treating cases in other
hospitals in the past (see Chapter 5). When technological innovations are
introduced, hospitals are paid according to historical cost patterns that do not
refl ect the (potentially) higher costs of using technological innovations.
Consequently, disincentives exist for hospitals to adopt and use cost-increasing
technological innovations until the payment system is updated to account for
their extra costs. Patient access to quality-increasing technological innovations
that also increase costs could be delayed because, in general, it takes some time
for enough information regarding the costs of using a technological innovation
in routine practice to be generated.
In some cases, the disincentive for using technological innovations under
DRG-based hospital payment systems might be counterbalanced by the second
kind of incentive (Table 9.2), which is to increase the number of admissions. In
competitive environments, and if certain technological innovations are thought
to improve hospital reputation or to stimulate admissions by physicians,
hospitals are likely to react by offering these services (OTA, 1983). Of course,
Table 9.2 Incentives of DRG-based hospital payment systems and effects related to
technological innovation
Main incentives Effects related to technological innovation
1. Reduce costs per admission • Promoting the use of cost-decreasing
technological innovations
• Encouraging the concentration of capital cost-
increasing innovations in fewer institutions,
leading to specialization of hospitals for certain
technologies
• No effect on technological innovations that are
cost neutral
• Discouraging the introduction of cost-increasing
technologies
• Encouraging HTAs before introduction of new
technologies
2. Increase number of admissions • Encouraging the use of technologies promoting
hospital reputation
• Promoting the use of technological innovations
valued by patients/admitting physicians
Source: Based on OTA, 1983.
136 Diagnosis-Related Groups in Europe
particular design features of each country’s DRG system and its DRG-based
hospital payment system (see Chapter 6 of this volume) are likely to modify the
strength of the basic incentives of these systems. For example, several countries
(such as France, Germany, Ireland, Poland, and Spain (Catalonia)) provide
additional funding for capital costs, thus exempting a signifi cant proportion of
hospital costs (particularly relevant in the context of innovations that increase
capital costs) from the incentives of DRG-based hospital payment. Similarly,
the availability of funding from sources other than the DRG-based hospital
payment system may modify the incentives of DRG-based hospital payment
systems. For example, hospitals receiving extra funding for teaching or research
are more likely to be in a better position to adopt technological innovations.
Yet, as evidenced by the existence of specifi c payment instruments for
technological innovations in most countries across Europe (see section 9.3),
DRG-based hospital payment systems alone seem to be perceived as providing
insuffi cient incentives for the desired introduction of technological innovations
that increase quality but also increase cost.
9.3 Technological innovation and DRG-based hospital
payment in practice: 12 European countries in comparison
As illustrated in the country-specifi c studies in the Part Two of this volume,
and as shown by Scheller-Kreinsen et al. (2011) DRG-based hospital payment
systems in most countries are updated at regular intervals. These long-term
mechanisms ensure that technological innovations are eventually formally
incorporated into the DRG-based hospital payment system, either through
updates of the DRG system (see Chapter 4), or through updates of the payment
rate (see Chapter 6). In addition, almost all countries have developed certain
short-term payment instruments that encourage the use of quality-increasing
technological innovations that also increase costs, within the time period
during which the DRG-based hospital payment system does not yet account for
the technological innovation.
Figure 9.2 illustrates the short-term payment instruments and long-term
updating mechanisms used to encourage and incorporate technological inno-
vation in the DRG-based hospital payment system. On the left, the fi gure shows
the short-term payment instruments used to encourage the use of quality-
increasing technological innovations that also increase costs. These can be
completely outside the system (extreme left) or can be associated to the DRG-
based hospital payment system (in the middle). On the right, the fi gure presents
mechanisms to incorporate technological change into the systems, either by
updating the DRG system – that is, the patient classifi cation system (PCS) – or
by adjusting the payment rate. When updating the PCS, several options exist:
(1) cases can be reassigned to different DRGs, (2) existing DRGs can be split, and
(3) new DRGs can be created when necessary.
A common challenge for policy-makers when devising payment policies is to
fi nd the right balance between two confl icting goals (Schreyögg et al., 2009).
On the one hand, they need to provide suffi cient incentives for hospitals to
make use of quality-increasing technological innovations that also increase
Technological innovation in DRG-based hospital payment systems 137
costs, in order to assure patient access. On the other hand, they need to keep
expenditure for technological innovations under control. If short-term instru-
ments provide additional payments for selected technological innovations,
these payments introduce incentives that may distort clinical decision-making
and can lead to ineffi ciencies related to over-provision of these services and
escalating health care costs (MedPAC, 2001). These confl icting incentives are
illustrated by the bars at the bottom of Figure 9.2. Specifi c incentives to use
technological innovations should decrease as technologies become more
formally incorporated into the system (although exceptions are conceivable, as
discussed later). Conversely, incentives for providers to make effi cient use of
resources increase once the use of technological innovations is no longer
encouraged through specifi c payment incentives.
9.3.1 Short-term instruments across Europe to encourage the
use of technological innovation
In the countries analysed in this book, three different short-term instruments
are employed to encourage the use of quality-increasing technological
innovations that also increase costs: (1) separate payments, (2) supplementary
payments and (3) special funding for cost-outliers. Table 9.3 shows that these
three types of short-term instruments can be represented in different forms.
Separate payments can take two forms: (1) fee-for-service payment, negotiated
nationally or locally, as is used in Germany (see Box 9.1); or (2) retrospective
reimbursement of hospital-reported total costs, as is used by some county
Figure 9.2 Short-term payment instruments and long-term updating mechanisms
138 Diagnosis-Related Groups in Europe
councils in Sweden. Both payment instruments are designed to encourage the
use of technological innovations when information regarding associated costs
and effects is still relatively scarce. Separate payments do not necessarily require
procedure codes to have been assigned to procedures, or drugs to be entered in
specifi c lists. The associated fl exibility allows some countries, such as France, to
make decisions at the level of the individual patient; for example, whether or
not to pay for experimental cancer drugs (see Chapter 13). Consequently, access
to new therapies in France (particularly in terms of cancer treatment) remains
one of the most generous in Europe (de Pouvourville, 2009). However, the
disadvantages of extensively using a system of separate payments have also
been experienced in France, where expenditure on new drugs has rocketed
(Cour de Comptes, 2009).
Box 9.1 Separate payment to German hospitals under NUB regulations
In Germany, the New Diagnostic and Treatment Methods Regulation (NUB)
was introduced as part of the 2005 Hospital Remuneration Act (KHEntgG).
It provides extrabudgetary funding in the form of negotiated fee-for-
service payments to selected hospitals using technological innovations.
Hospitals wishing to be reimbursed via NUBs for their use of technological
innovations must take several steps before being reimbursed. First, hospitals
must apply – with a description of the new technology and of associated
costs – to the Institute for the Hospital Remuneration System (InEK),
which is responsible for managing the German diagnosis-related groups
(G-DRG) system. If the application is accepted, individual providers must
successfully negotiate with the sickness funds concerning the size of the
payments to be made. Finally, each hospital must conclude a contractual
Table 9.3 Instruments to encourage the use of technological innovation and types of
associated payments
Instrument Type of payment
Outside DRG system
Separate payments • fee-for-service (based on weighted costs or
negotiated payment)
• retrospective reimbursement of reported costs per
case
Inside DRG system
Supplementary payments • fee-for-service
• retrospective reimbursement of costs above
standard rate
• payment of weighted costs
Special funding for cost-outliers • retrospective reimbursement of costs above a
statistically determined threshold
• fi xed payments (based on weighted costs or
negotiated payment)
• payment of weighted costs
Technological innovation in DRG-based hospital payment systems 139
agreement with the sickness funds to receive NUB reimbursement for its
use of the technology. Negotiated NUB payments are valid for only one year
and hospitals need to reapply to the InEK if they want to continue to use
a technology in subsequent years. Since the introduction of this approach
in 2005, it has been shown that acceptance for NUB reimbursement often
represents the fi rst step in the process of incorporating new technologies
into the DRG system (Henschke et al., 2010).
Source: Henschke et al., 2010.
In contrast to separate payments, supplementary payments and cost-outlier
funding are relevant for technological innovations, but are also used to improve
the general coherence of the DRG system by excluding certain high-cost tech-
nologies or high-cost patients and reimbursing them separately (see Chapter 4
and Chapter 6). Both instruments take a specifi c DRG payment rate as a starting
point and justify additional payments in terms of substantial differences
between incurred costs and standard payment rates.
Supplementary payments are made on top of the ‘standard’ DRG payment
rate if specifi c technologies (including new and innovative ones) are applied.
The amount to be paid on top of the standard rate can be negotiated or can take
the form of retrospective reimbursement of reported costs (per case) above the
standard rate of individual providers. In some countries costs are weighted
across providers before being paid; that is, average costs per patient category are
calculated and reimbursed (‘payment of weighted costs’). The necessary admini-
strative processes for establishing a relationship between a procedure (a tech-
nological innovation) and a DRG require some time, which may contribute to
slowing down the adoption of technological innovations by hospitals. In some
cases, a procedure code needs to be assigned to a technological innovation
before supplementary payments can be made, thus prolonging the process of
providing reimbursement for technological innovations.
In countries in which special funding for cost-outliers is available, the way
technologies (including new and innovative ones) infl uence homogeneity of
resource use of patients within DRGs determines whether special funding is
made available on top of standard payment rates. Cost-outlier funding builds
on detailed retrospective statistical analysis of cost data. Different variants of
this instrument exist (see Table 9.3). In addition, many countries provide extra
payments for length-of-stay outliers (see Chapter 6) but these instruments are
not particularly relevant to technological innovation, as technological innova-
tion may contribute to a reduction in the length of stay, for example, when
new, minimally invasive surgical procedures lead to faster patient recovery and
discharge (Simpson et al., 2005).
Table 9.4 presents the distribution of the outlined short-term payment
instruments across the 12 countries. Separate payments are the most frequently
used payment instrument. Surprisingly, cost-outlier funding for cost-increasing
technological innovation is used only in Estonia, Finland and some Swedish
county councils. Some countries with DRG-based budget-allocation systems
(such as Austria and Portugal) do not make use of any short-term payment
instruments.
140 Diagnosis-Related Groups in Europe
Table 9.4 The application of short-term reimbursement instruments in 11 European
countries (plus Catalonia)
Instruments used to provide extra payments for technological innovations
Separate payments Supplementary payments Cost-outlier funding
Austria No No No
Catalonia Yes (for certain No No
(Spain)* high-cost procedures)
England/ Yes (for up to Yes (for certain No
United Kingdom three years) high-cost services)
Estonia Yes (for certain No Yes
high-cost services)
Finland Depending on hospital district, both No
instruments are used
France Yes Yes No
Germany Yes Yes (for certain No
high-cost services)
Ireland Yes No No
Netherlands Yes (for certain Yes (envisaged to No
high-cost drugs) start in 2011)
Poland No Yes (for certain No
high-cost services)
Portugal No No No
Sweden Depending on the county council, all instruments are used
Source: Compiled by the authors on the basis of information presented in the country-specifi c
chapters in Part Two of this volume.
*In Spain hospital fi nancing is decentralized. The information presented here refers to
Catalonia, where a DRG system is used that determines 35 per cent of hospital reimbursement.
All extra payments provide strong incentives to hospitals to apply techno-
logical innovations, as they exempt the selected technologies from the incen-
tives of DRG-based hospital payment. However, as already mentioned, they
may favour the use of certain procedures, drugs or technological equipment
over existing technologies included within the DRG-based system, which may
reduce effi ciency of hospital care (MedPAC, 2001). Furthermore, extra funding
may produce ‘winners’ and ‘losers’ in the hospital market, as it is likely to lead
to higher payments for hospitals that play a strong role in technology dissemi-
nation (for example, at university hospitals, at the expense of other hospitals)
(MedPac, 2001).
9.3.2 Long-term updating mechanisms in European
DRG-based hospital payment systems: Incorporating
technological innovations
In terms of incorporating technological innovations into DRG-based hospital
payment systems, the processes of updating the PCS and the payment rate are
Technological innovation in DRG-based hospital payment systems 141
essential. Table 9.5 presents the frequency of updates and the time-lag to data
used for updates in 12 countries across Europe, as these two factors determine
how fast a DRG-based hospital payment system is able to respond to techno-
logical innovations. Neither of the updating mechanisms are specifi cally
targeted at incorporating technological innovations, but they are intended to
Table 9.5 Frequency of updates and time-lag to data used for updates across
12 European countries
DRG-based hospital payment system
PCS Payment rate
Frequency of
updates
Time-lag to
data
Frequency of
updates
Time-lag to data
Austria Annual 2–4 years 4–5 years
(updated
when
necessary)
2–4 years
England/
United
Kingdom
Annual Minor revisions
annually;
irregular
overhauls
about every
5–6 years
Annual 3 years (but
adjusted for
infl ation)
Estonia Irregular (fi rst
update after
7 years)
1–2 years Annual or
following
update of fee-
for-service fees
1–2 years
Finland Annual 1 year Annual 0–1 year
France Annual 1 year Annual 2 years
Germany Annual 2 years Annual 2 years
Ireland Every 4 years,
linked to
Australian
updates of
AR-DRGs
Not applicable
(imported
AR-DRGs)
Annual (linked
to Australian
relative-weight
updates)
1–2 years
Netherlands Irregular Not
standardized
Annual or when
considered
necessary
2 years, or
based on
negotiations
Poland Irregular (planned
twice per year)
1 year Annual update
only of base
rate
1 year
Portugal Irregular Not applicable
(imported
AP-DRGs)
Irregular 2–3 years
Spain
(Catalonia)
Biennial Not applicable
(imported
3-year-old
CMS-DRGs)
Annual 2–3 years
Sweden Annual 1–2 years Annual 2 years
Source: Compiled by the authors on the basis of information presented in the country-specifi c
chapters in Part Two of this volume.
142 Diagnosis-Related Groups in Europe
ensure that the DRG-based payment systems are always adapted to current
practice patterns and treatment costs.
Both the PCS and the payment rate are updated annually in the majority of
countries, but there are remarkable exceptions. In 2010, Estonia updated its
DRG system for the fi rst time since the introduction of the Nordic PCS
(NordDRGs) to the country in 2003. Ireland currently uses Australian Refi ned
(AR-)DRGs, which are updated every four years (see Chapter 15 of this volume).
Austria is an interesting outlier with regard to the adjustment of payment rates,
as DRG weights are not updated regularly, but are adjusted only for specifi c
DRGs when deemed necessary by policy-makers. The data used for updates vary
considerably between countries. In Finland, data are used from the current year
to update the DRG system for the next year, and DRG weights are recalculated
as soon as data become available (during the same year). In most countries,
however, data both for updating the PCS and for adjusting DRG weights or
prices are at least two years old.
In addition, the mechanisms to introduce new codes for new procedures,
drugs and medical devices affect the way in which DRG systems can incorporate
technological innovations. Frequent updates of codes facilitate more rapid
adoption and incorporation of technological innovations into DRG systems.
Rare updates increase the length of time before technological innovations can
be systematically incorporated.
As already mentioned, technological innovations can alter treatment costs
in different ways. Countries collecting detailed bottom-up hospital cost-
accounting information (see Chapter 5) are clearly in a better position to pre-
cisely identify the effect of technological innovations on hospital costs using
routinely available information. When technological innovations increase (or
decrease) costs for a well-defi ned subset of patients, adjusting the PCS is the best
method of incorporating technological innovations into the DRG-based hospi-
tal payment system. However, the incentives to modify the PCS should be
closely monitored: if a new DRG is introduced – for example, for using a specifi c
innovative medical device in a broadly defi ned group of patients – providers
could be incentivized to over-provide the technological innovation to patients
that would not benefi t from the innovative technology.
When technological innovations increase the costs of all services bundled in
one DRG or the costs of all hospital services, updates to the payment rate are
the best approach to incorporating them into the DRG-based hospital payment
system. In order to increase payment for a specifi c DRG, DRG weights can be
recalculated. In order to increase funding for all hospital services, countries not
operating a relative-weight approach can infl ate raw tariffs by the appropriate
amount. Countries using a relative-weight approach have different options.
They can either adjust the base rate to account for proportionate increases in
costs (for example, a 5 per cent increase of all hospital costs), or they can adjust
the base rate and recalculate relative weights if technological innovations
increase costs for all cases by a fi xed amount.
Technological innovation in DRG-based hospital payment systems 143
9.4 Conclusions: Encouraging and incorporating
technological innovations in European DRG-based
hospital payment systems: Scope for improvement
In many European countries, there are concerns that DRG-based hospital
payment systems do not provide the right set of incentives to ensure that
patients have timely access to technological innovations. Our discussion of the
theoretical incentives of DRG-based hospital payment systems to adopt and use
technological innovations in hospitals has revealed that these concerns should
be important only for the specifi c case of those technological innovations that
increase quality and are accompanied by a signifi cant increase in total costs per
case.
The second part of the chapter illustrates that most (but not all) countries
analysed in this book have complemented their DRG-based payment systems
with specifi c short-term payment instruments targeted at encouraging the
adoption and use of technological innovations. However, additional payments
for technological innovations exempt these technologies from the inherent
effi ciency incentives of DRG-based hospital payment systems. In fact, generous
separate payment methods (such as fee-for-service payments) may lead to a
distortion of clinical decision-making and a signifi cant increase in spending on
those technological innovations for which separate payments are available (as
evidenced in France). Furthermore, as short-term payment incentives are often
introduced for technological innovations at a time when rigorous analyses of
their (long-term) effects are not yet available, there is a risk that the additional
payments inadvertently incentivize the use of cost-increasing technological
innovations that are quality neutral or even result in a decrease in the quality
of health care.
Therefore, short-term payment instruments should be employed very
carefully, and incorporated only after careful assessments have been made
concerning the likely effects of the concerned technology on quality of care. In
the United States, short-term payment instruments are intended to be limited
to technological innovations offering either considerable quality improvements
over existing technologies, or offering options for diagnosis or treatment of
previously untreatable conditions (Clyde et al., 2008). Unfortunately, in several
European countries (such as Germany and France), the introduction of short-
term payment instruments for technological innovations seems to be more
directly linked to the criteria of higher costs than to the criteria of demonstrating
considerable quality improvements.
If countries should want to provide short-term payment incentives for
technological innovations with expected signifi cant quality improvements but
for which the evidence remains uncertain, one possible approach is the so-called
Coverage with Evidence Development (CED) (Hutton et al., 2007). Under CED
approaches, payments for technological innovations are provided only for a
limited period of time and on the condition that continuing evaluation is
carried out (see Box 9.2 for an example from the Netherlands).
144 Diagnosis-Related Groups in Europe
Box 9.2 Coverage with evidence development in the Netherlands
In 2006, new regulations were introduced in the Netherlands regarding
expensive (and orphan) inpatient drugs. The regulations specify that
an innovative drug can be provisionally included on the expensive (or
orphan) drug list(s) for up to four years, which allows hospitals to receive
separate payments for these drugs even before their cost–effectiveness has
been formally established. The conditions for a drug to be included on
a list are that (1) added therapeutic value is demonstrated; (2) a plan for
the assessment of cost–effectiveness in daily clinical practice is approved
by the pharmaceutical advisory committee; and (3) the drug expenses
account for over 0.5 per cent (for the expensive drugs category) or 5 per
cent (for orphan drugs) of the annual hospital drug budget. If all three
conditions are met, hospitals can receive separate payments amounting
to 80 per cent (for expensive drugs) (and 100 per cent for orphan drugs)
of the purchase price of drugs placed on the expensive (and orphan)
drug list(s). After three years, the data generated in the context of the
assessment plan are used to inform decisions about providing further
funding for the innovative or (orphan) drug(s).
Source: Delwel, 2008.
Given that most DRG-based hospital payment systems are updated at regular
intervals, the change of treatment patterns and costs resulting from the
introduction of technological innovations should ultimately be refl ected by
the DRG-based hospital payment system. Countries with frequent updates of
their DRG system and of the payment rate – and with a short time-lag between
data collection and using the information collected for DRG-based hospital
payment – are clearly in a better position to incorporate technological
innovations into their systems. However, if updates of the system lead to the
introduction of specifi c new DRGs for technological innovations (such as for
drug-eluting stents), the effect may be similar to that of introducing separate
payments for technological innovations; namely, introducing strong incentives
to make use of the specifi c technology. More generally, therefore, the issue of
incorporating technological innovations into DRG systems highlights the
trade-off that exists between providing adequate funding for specifi c procedures
and the intention to promote effi ciency by leaving to clinicians the decisions
regarding which procedures to use.
Furthermore, in the context of an emerging common European hospital
market, there is scope for increasing cooperation across countries in terms of
technological innovations. Cooperation appears to be particularly benefi cial in
the fi eld of assessing the effect of technological innovations on quality. As
envisaged by article 15 of the recently adopted European Union (EU) Directive on
the Application of Patients’ Rights in Cross-Border Healthcare (European Parliament
and Council, 2011), a European network of health technology assessment (HTA)
agencies could assess technological innovations using a common set of criteria
Technological innovation in DRG-based hospital payment systems 145
in order to avoid duplication of work and individual analyses in each Member
State (Kristensen, 2008). If suffi cient evidence is available to demonstrate con-
siderable improvements in quality, decentralized decisions regarding whether or
not to introduce short-term payment instruments for these technologies could
then be made by governments, self-governing bodies or local payers within
Member States, in a manner similar to the decentralized approaches used in
Finland, Germany (see Box 9.1) or Sweden. The advantage would be that the
available evidence could be assessed more effi ciently, while payment decisions
would still be made according to national or local value-judgements, which is
necessary because differences are likely to exist in the willingness to pay for a
given increase in quality.
Empirical research on the effects of DRG-based hospital payment systems in
terms of the adoption and diffusion of technological innovations is diffi cult to
design and is relatively scarce. Research relating to the effects of DRG-based
payment systems on the adoption, implementation and use of technological
innovations has rarely taken into account the different approaches to encourag-
ing and incorporating technological innovations within the family of DRG-
based payment systems (Torbica & Cappellaro, 2010; Packer et al., 2006; Bech et
al., 2009). Short-term payment instruments and long-term updating mechanisms
differ greatly across countries. Future empirical cross-country investigations – for
example of the determinants of the implementation and use of technological
innovation – should take these differences into account and test empirically
whether and how the different identifi ed approaches affect the implementation
and use of technological innovation.
9.5 References
Atella, V., the TECH Investigators (2003). The relationship between health policies, medi-
cal technology trends, and outcomes: a perspective from the TECH global research
network, in OECD. A Disease-Based Comparison of Health Systems. What is Best and at
What Cost? Paris: Organisation for Economic Co-operation and Development.
Banta, H.D. (1983). Social science research on medical technology: utility and limitations.
Social Science & Medicine, 17(18):1363–1369.
Banta, H.D., Behney, C.J., Andrulid, D.P. (1978). Assessing the Effi cacy and Safety of Medical
Technologies. Washington, DC: Offi ce of Technology Assessment.
Bech, M., Christiansen, T., Dunham, K., et al. (2009). The infl uence of economic incen-
tives and regulatory factors on the adoptions of treatment technologies: a case
study of technologies used to treat heart attacks. Health Economics, 18(10):1114–
1132.
Black, W.C. (1990). The CE plane: a graphic representation of cost–effectiveness. Medical
decision-making: an international journal of the Society for Medical Decision-Making,
10(3):212–214.
Cappellaro, G., Ghislandi, S., Anessi-Pessina, E. (2011). Diffusion of medical technology:
the role of fi nancing. Health Policy, 100(1):51–59.
Clyde, A.T., Bockstedt, L., Farkas, J.A., Jackson, C. (2008). Experience with Medicare’s new
technology add-on payment program. Health Affairs, 27(6):1632–1641.
Congressional Budget Offi ce (2008). Technological Change and the Growth of Health Care
Spending. Washington, DC : Congressional Budget Offi ce.
146 Diagnosis-Related Groups in Europe
Cour de Comptes (2009). La Securité Sociale, chapitre 7, La mise en place de la T2A: Bilan à
mi-parcours. Paris: Cour de Comptes.
Cutler, D.M. (2007). The lifetime costs and benefi ts of medical technology. Journal of
health Economics, 26(6):1081–1100.
Cutler, D.M., McClellan, M. (2001). Is technological change in medicine worth it? Health
Affairs, 20(5):11–29.
Cutler, D.M., McClellan, M., Newhouse, J.P. (1998a). What has increased medical care
spending bought? The American Economic Review, 88(2):132–136.
Cutler, D.M., McClellan, M., Newhouse, J.P., Remler, D. (1998b). Are medical prices
declining? Evidence for heart attack treatment. The Quarterly Journal of Economics;
113(4):991–1024.
Delwel, G.O. (2008). Leidraad voor Uitkomstenonderzoek ten behoeve van de beoordeling
doelmatigheid intramurale geneesmiddelen [Guideline for Outcomes Research in Support of
Assessing Effi ciency of Intramural Drugs]. Diemen: College voor Zorgverzekeringen.
Drummond, M.F., Sculpher, M.J., Torrance, G.W., O’Brien, B.J., Stoddart, G.L. (2005).
Methods for the Economic Evaluation of Health Care Programmes. Oxford: Oxford Univer-
sity Press.
European Parliament and Council (2011). Directive 2011/24/EU of 9 March 2011 on the
Application of Patients’ Rights in Cross-Border Healthcare. Brussels: Offi cial Journal of the
European Union (L88/45-L88/65).
Garrison, L.P., Wilensky, G.R. (1986). Cost-containment and incentives for technology.
Health Affairs, 5(2):46–58.
Goodman, C.S. (2004). HTA 101. Introduction to Health Technology Assessment. Falls
Church, VA: The Lewin Group.
Greenhalgh, T., Robert, G., MacFarlane, F., Bate, P., Kyriakidou, O. (2004). Diffusion of
innovations in service organizations: systematic review and recommendations.
Milbank Quarterly, 82(4):581-629.
Henschke, C., Bäumler, M., Weid, S., Gaskins, M., Busse, R. (2010). Extrabudgetary (‘NUB’)
payments – a gateway for introducing new medical devices into the German inpatient
reimbursement system? Journal of Management and Marketing in Healthcare, 3(2):
119–33.
Hutton, J., Trueman, P., Henshall, C. (2007). Coverage with evidence development: an
examination of conceptual and policy issues. International Journal of Technology
Assessment in Health Care, 23:425–32.
Kristensen, F.B. (2008). Transnational collaboration on health technology assessment – a
political priority in Europe, in M.V. Garrido, F.B. Kristensen, C.P. Nielsen, R. Busse, eds.
Health Technology Assessment and Health Policy-Making in Europe: Current Status,
Challenges and Potential. Copenhagen: WHO Regional Offi ce for Europe on behalf of
the European Observatory on Health Systems and Policies.
MedPAC (2001). Accounting for new technology in hospital prospective payment
systems, in MedPAC Report to the Congress. Medicare Payment Policy. Washington, DC:
Medicare Payment Advisory Commission.
MedPAC (2003). Payment for new technologies in Medicare’s prospective payment
system, in MedPAC Report to the Congress: Medicare Payment Policy. Washington, DC:
Medicare Payment Advisory Commission.
Mowatt, G., Bower, D.J., Brebner, J.A. et al. (1997). When and how to assess fast-changing
technologies: a comparative study of medical applications of four generic tech-
nologies. Health Technology Assessment, 1:1–149.
OTA (1983). Diagnosis-Related Groups (DRGs) and the Medicare Program: Implications for
Medical Technology – A Technical Memorandum. Washington, DC: Offi ce of Technology
Assessment.
Technological innovation in DRG-based hospital payment systems 147
Packer, C., Simpson, S., Stevens, A. (2006). International diffusion of new health tech-
nologies: a ten-country analysis of six health technologies. International Journal of
Technology Assessment in Health Care, 22(4):419–28.
de Pouvourville, G. (2009). Les hôpitaux français face au paiement prospectif au cas: la
mise en ouvre de la tarifi cation à l’activité. Revue Economique, 60(2):457–70.
Rettig, R.A. (1994). Medical innovation duels cost-containment. Health Affairs, 13(3):
7–27.
Scheller-Kreinsen, D., Quentin, W., Busse, R. (2011). DRG-based hospital payment systems
and technological innovation in 12 European countries. Value in Health, 14(8).
Schreyögg, J., Bäumler, M., Busse, R. (2009). Balancing adoption and affordability of
medical devices in Europe. Health Policy, 92(2–3):218–24.
Shih, C., Berliner, E. (2008). Diffusion of new technology and payment policies: coronary
stents. Health Affairs, 27(6):1566–76.
Simpson, S., Packer, C., Stevens, A., Raftery, J. (2005). Predicting the impact of new health
technologies on average length of stay: development of a prediction framework.
International Journal of Technology Assessment in Health Care, 21(4):487–91.
Torbica, A., Cappellaro, G. (2010). Uptake and diffusion of medical technology innovation
in Europe: what role for funding and procurement policies? Journal of Medical
Marketing, 10(1):61–9.
Tunstall-Pedoe, H., Vanuzzo, D., Hobbs, M. et al. (2000). Estimation of contribution of
changes in coronary care to improving survival, event rates, and coronary heart disease
mortality across the WHO MONICA Project populations. Lancet, 355(9205):688–700.
Weisbrod, B.A. (1991). The health care quadrilemma: an essay on technological change,
insurance, quality of care, and cost-containment. Journal of Economic Literature,
29(2):523–52.
chapter t e n
Moving towards
transparency, effi ciency
and quality in hospitals:
Conclusions and
recommendations
Reinhard Busse and Wilm Quentin
10.1 Introduction
Part One of this book has provided comparative information from 12 European
countries about the specifi c characteristics of their diagnosis-related group (DRG)
systems, about how these systems are used for hospital payment and about the
progress that has been made in moving towards transparency, effi ciency and
quality in hospitals. Part Two provides more detailed information from the 12
European countries and facilitates insights into the strengths and problems of
DRG systems and DRG-based hospital payment systems in each of these
countries. Together, the two parts of the book demonstrate a great degree of
diversity in the specifi c design features of DRG systems and DRG-based hospital
payment systems across countries, but at the same time they reveal that most
countries are struggling with similar issues in their pursuit of common goals.
This chapter draws together the fi ndings from Part One and Part Two in order
to address the question raised in the title of this book; namely, whether we
are moving towards transparency, effi ciency and quality in European hospitals.
In addition, the chapter makes specifi c recommendations for policy-makers
regarding how best to design DRG-based hospital payment systems given
country-specifi c aims and objectives, and it explores the potential for coopera-
tion across European countries in designing and developing DRG systems and
DRG-based hospital payment systems – a process, which could ultimately lead
to the emergence of European DRGs as the answer to common problems in this
fi eld in European countries.
150 Diagnosis-Related Groups in Europe
The next section (10.2) summarizes the country experiences and draws on
fi ndings of the available literature presented, especially in Chapters 7 and 8, in
order to provide an overview of the status quo. That is: where are we now, in
terms of transparency, effi ciency, and quality in hospitals? Subsequently, sec-
tion 10.3 makes recommendations for both types of countries – those preparing
the introduction of DRG systems, and those optimizing existing systems. This
section is structured according to the three building blocks introduced in Chap-
ter 3; namely, the DRG system itself, hospital cost information, and actual
DRG-based hospital payment. Finally, section 10.4 draws conclusions from the
vast experience summarized in this book and aims to look into the future of
DRGs in Europe, including the potential for coordinating, and eventually har-
monizing DRG systems and DRG-based hospital payment in Europe.
10.2 Where are we now?
In almost all European countries in which DRGs have been introduced since
the mid-1980s, the most important aims related to their introduction included
increasing transparency, improving effi ciency and assuring quality of hospital
care (see Chapter 2). Today, after more than a decade of experience with using
DRGs in most European countries, it is time to consider whether the extensive
use of DRGs in the 12 countries included in this book has contributed towards
achieving these aims. The country chapters in Part Two and the extensive
literature searches carried out for chapters 7 and 8 on the effects of DRG-based
hospital payment systems on effi ciency and quality of care provide a solid
foundation for approaching this question. The following subsections discuss
how far European countries have moved towards achieving each of these aims.
10.2.1 Moving towards transparency?
Following the introduction of DRGs and DRG-based hospital payment systems,
transparency of hospital services and costs has substantially improved in all
countries, essentially for four interrelated reasons: (1) DRGs provide a concise
measure for reporting hospital activity; (2) DRGs facilitate performance com-
parisons of costs, effi ciency and quality; (3) hospitals are incentivized to increase
their efforts in coding diagnoses and procedures; and (4) hospitals are en-
couraged to improve their cost-accounting systems (see Chapter 5).
First, because DRGs aggregate the confusingly large number of patients
treated by hospitals into a small number of groups of patients with similar clini-
cal characteristics and similar resource-consumption patterns, they provide
a concise and meaningful measure of hospital outputs (Fetter et al., 1976;
Goldfi eld, 2010). Prior to the introduction of DRGs, hospital activity was
reported either on the basis of highly aggregated measures, such as the number
of provided bed days and the number of discharged patients, or on the basis of
very detailed measures, such as the main diagnoses or procedures of all patients.
However, because none of these measures summarized patients with similar
clinical characteristics and similar resource-need patterns, they could not mean-
ingfully refl ect hospital activity. Today, the vast majority of hospitals in all
Moving towards transparency, effi ciency and quality in hospitals 151
countries – often including hospitals with different ownership (profi t-making
versus non-profi t-making) and different levels of specialization (for example,
teaching hospitals versus general hospitals) – are required to prepare detailed
activity reports that specify the number and type of DRGs provided. These are
usually made available to the public and help to overcome agency problems
that existed prior to the introduction of DRGs, because purchasers did not have
a meaningful measure for hospital activity.
Second, regulators, payers and hospital managers in most countries (for
example, Finland, France, Ireland and Spain) are starting to use DRGs for hos-
pital performance comparisons. They compare resource use of hospitals by
assessing whether patients in one DRG are staying signifi cantly longer in one
hospital than in another, for example, or whether one hospital is signifi cantly
more costly than another when treating patients within the same DRG. Simi-
larly, quality is compared by determining whether patients assigned to a par-
ticular DRG have a higher rate of complications in one hospital than in another,
and effi ciency is assessed by using DRGs as a measure of hospital output.
Third, because hospitals receive higher DRG-based payments if they ‘code’
(input) all relevant diagnoses and procedures, they have strong incentives
to improve their coding practices. In many countries, clinicians, nurses or
documentation assistants are specifi cally trained in order to improve their
coding skills. Consequently, almost all countries fi nd that information about
diagnoses and procedures in hospitals has improved considerably since the
introduction of DRGs. In addition, payers have introduced auditing systems
to assure that the provided information is correct, which further increases
the reliability of the available information. However, at the same time, the
coding-related administrative workload has been of concern for clinicians in
many countries.
Fourth, as discussed in Chapter 5, the introduction of DRG-based hospital
payment has infl uenced cost-accounting practices in hospitals. On the one
hand, regulators have mandated improved and standardized cost-accounting
systems in hospitals, while on the other hand, hospitals have been incentivized
to improve their cost-accounting systems for management purposes. Conse-
quently, the quality of cost information has improved in most countries.
However, if patients within a DRG do not adequately account for differences
between patients – that is, if DRGs are not suffi ciently homogeneous – they are
an inadequate measure of hospital activity and, consequently, hospital perfor-
mance comparisons on the basis of DRGs will be unfair. Therefore, the methods
used for ensuring that DRG systems are an adequate measure of hospital activ-
ity are highly important, particularly because innovations are continuously
changing the way hospital services are provided. In addition, performance
comparisons on the basis of DRGs need to take into account that certain factors
may be beyond the control of hospitals (for example, treating a larger share of
socially disadvantaged patients or having higher labour costs), which are not
accounted for in the DRG system. Furthermore, while DRGs have contributed
to increased transparency of hospital services within countries, transparency of
hospital services across countries remains limited because different DRG sys-
tems are used in different countries, thus preventing – or at least severely com-
plicating – comparisons of hospital activity and performance across borders
except where the same systems are in use.
152 Diagnosis-Related Groups in Europe
10.2.2 Moving towards effi ciency?
As discussed in Chapter 7, although improving hospital effi ciency is generally a
key motivation for introducing DRG-based hospital payment systems, there are
relatively few studies that have explicitly identifi ed and quantifi ed the impact
of these systems on effi ciency using established data-driven methods such as
data envelopment analysis (DEA) or stochastic frontier analyses. Rather, most
research has concentrated on indicators of effi ciency – such as activity and
length of stay – which are more easily measured, but by defi nition provide only
a partial picture of effi ciency.
Existing studies using DEA or stochastic frontier analyses – both with their
own particular limitations (Street et al., 2010) – have produced mixed evidence
on the extent to which DRG-based hospital payment has contributed to higher
effi ciency levels in hospitals. The studies reviewed in Chapter 7 reported that
the introduction of DRG-based hospital payment was associated with improved
technical effi ciency in Portugal, Sweden and Norway, but that no positive
impact was observed in the United States and in an Austrian study. On the one
hand, this mixed evidence could be related to the considerable differences in
the design and operation of DRG-based hospital payment systems in different
countries and to heterogeneity in the hospital payment systems that existed
prior to the introduction of DRG-based hospital payment. On the other hand,
studies may have underestimated (or overestimated) the effect of DRG-based
hospital payment on effi ciency because attribution of effi ciency changes to a
hospital payment reform in longitudinal studies is complicated by the existence
of confounding factors, such as changes being part of wider reform packages,
and because detected changes could merely represent changes in documentation
practice. In addition, because of an almost complete absence of data, it remains
unknown whether the effect of the potential unintended consequences of
DRG-based hospital payment systems (such as overtreatment of admitted
patients (‘gaming’) or increased admissions of patients for unnecessary services
(see Chapter 6)) could have led to reduced allocative (output) effi ciency.
There is generally agreement in the literature that the introduction of DRG-
based hospital payment systems has led to increased activity and reduced
length of stay, and it is consequently often assumed that hospital effi ciency has
improved. For example, studies have found that hospital admissions increased
following the introduction of DRG-based hospital payment in Australia,
Denmark, England, France, Germany, Norway and, at least initially, in Sweden,
while results for Italy are mixed (see Table 7.4 in Chapter 7). Hospital activity
did not increase in the United States, but this is in line with the expected effects
of DRG-based hospital payment when replacing a fee-for-service system. Yet, of
course, the aforementioned points regarding country-specifi c contexts and the
diffi culties in attributing causality also apply here.
Mostly based on the evidence of these studies, the authors of the country-
specifi c chapters in Part Two come to similar conclusions. In Austria and
England, for example, DRG-based hospital payment is thought to have contrib-
uted to increased effi ciency. In the chapters on Estonia, Germany, Ireland,
Poland, Portugal, Sweden and the Netherlands, the authors do not directly
comment on the effect of DRG-based hospital payment on effi ciency but
Moving towards transparency, effi ciency and quality in hospitals 153
they highlight rather positive trends in costs, length of stay or productivity.
For Finland and Spain (see chapters 18 and 22), DRG-based hospital payment
is thought to have had only minimal effects on effi ciency because country-
specifi c design features imply that hospitals are not exposed to strong incen-
tives for effi ciency improvement (see Chapter 6). By contrast, in Chapter 13,
Or and Bellanger come to a rather negative conclusion about the effect of the
French GHM system on effi ciency, which seems to be strongly infl uenced by
the results of an evaluation by the Auditor’s Offi ce (Cour des Comptes, 2009).
In summary, while the evidence remains limited because of the above-
mentioned diffi culties in measuring and detecting effi ciency changes (and in
attributing them to the introduction of a specifi c payment system), the bulk
of the literature and most of the authors in this book assume that DRG-
based hospital payment systems have had a somewhat positive effect on
effi ciency. However, it is also clear that DRG-based hospital payment systems
can have unintended consequences, such as ‘cream-skimming’, ‘up-coding’,
overtreatment/‘gaming’, supplier-induced demand, and so on (see Chapter 6).
If these unintended consequences are not accounted for by the specifi c
design features of the payment system or by the regulatory and institutional
context, they might threaten to outweigh any effi ciency improvements that
could be expected as a result of the introduction of DRG-based hospital pay-
ment systems.
10.2.3 Moving towards quality?
The effect of DRGs on quality of care has always been highly controversial:
there have been major concerns on the part of health professionals in many
countries that DRG-based hospital payment systems might compromise quality
of care because hospitals are incentivized to reduce costs. However, at the same
time, proponents of the use of DRGs have argued that quality of care could
in fact be improved, because DRGs contribute to increased transparency in
the quality of care and because hospitals are incentivized to invest in quality
improvements that lead to reduced costs (for example, infection control
measures or improved surgical techniques).
As discussed in Chapter 8, the effect of DRGs on quality of care has been
assessed in numerous studies from the United States and – more recently – also
in studies from Europe. The reviewed evidence from the United States has
produced a multifaceted picture: some studies found that processes of care (for
example, as measured by physician and nurse cognitive performance) improved
following the introduction of DRG-based hospital payment (Kahn et al., 1990b),
even though these changes could not be directly attributed to the hospital
payment reform (Rogers et al., 1990). At the same time, a larger proportion of
patients were found to have been discharged in unstable conditions after the
implementation of DRG-based payment (Kosecoff et al., 1990), but mortality at
30 and 180 days following hospitalization was unaffected (Kahn et al., 1990a).
It appeared that quality of care improved in certain hospitals and certain areas
of care, such as colorectal cancer surgery (Schwartz & Tartter, 1998), but was
worse in other areas of care (Gilman, 2000), in particular in hospitals for which
154 Diagnosis-Related Groups in Europe
the introduction of DRG-based payment implied high levels of fi nancial
pressure (Cutler, 1995). In summary, studies from the United States suggest
that quality of care was, in general, not signifi cantly affected by the introduc-
tion of DRG-based hospital payment, as it did not compromise the long-term
trend towards improved quality of care in hospitals (Rogers et al., 1990). How-
ever, the effect on quality needs to be closely monitored because there could
be adverse effects for certain patient groups in certain hospitals and because
a trend towards more unstable discharges emerged after the implementation
of DRGs.
In Europe, the available research evaluating the impact on care quality and
patient outcomes is too limited to draw any fi rm conclusions, in particular
because evidence is available only from a limited number of countries. In
England, little measurable change was found in the quality of care following
the introduction of DRG-based payment, in terms of in-hospital mortality,
30-day post-surgical mortality, and emergency readmissions after treatment for
hip fracture (Farrar et al., 2009). In Germany, 30-day post-discharge mortality
signifi cantly decreased during the introduction period of DRG-based hospital
payment, and a large number of quality indicators were found to have improved
over the same period of time (Fürstenberg et al., 2011). In Norway and Italy,
studies did not fi nd that quality decreased following the introduction of DRG-
based payment (see Chapter 8), while one study from Sweden showed that
patient-perceived quality of care decreased after the introduction of DRG-based
hospital payment (Ljunggren & Sjödén, 2003).
In general, it seems that quality was not adversely affected by the introduc-
tion of DRG-based hospital payment in most European countries. However, of
course, the impact of DRG-based hospital payment on quality of care always
depends on the country-specifi c design features of the systems and the regula-
tory and health care context(s) in question. The effect of DRG-based hospital
payments on quality of care might be different in Europe from that in the
United States because (1) DRG-based hospital payment systems in most coun-
tries did not replace fee-for-service systems (as was the case in the United States)
but rather global budgets, which were already partly adjusted for activity mea-
sured in cases or bed days (see Chapters 2 and 7); and because (2) there is a
much stronger public sector presence in the provision of health care in Europe
than in the United States.
Surprisingly, only very few countries explicitly adjust DRG-based hospital
payments on the basis of information regarding quality in hospitals. One
notable exception is England, where the Commissioning for Quality and
Innovation (CQUIN) framework allows purchasers to link a moderate proportion
of hospitals’ income (that is, 1.5 per cent in 2010/2011) to the achievement of
locally negotiated quality goals. In the Netherlands, insurers can negotiate with
hospitals regarding price, volume and quality of care for about 30 per cent of
Dutch DRGs (Diagnose Behandeling Combinaties, DBCs – see Chapter 23).
However, apparently insurers and hospitals negotiate predominantly on price
and volume, while quality plays only a minor role in the negotiation process.
Instead of adjusting DRG-based hospital payment for quality, most countries
reward quality improvements through specifi c budgets that are independent
from DRG-based hospital payment (see Chapter 8).
Moving towards transparency, effi ciency and quality in hospitals 155
One problem relating to quality adjustments of DRG-based hospital payments
is that in many European countries, information on quality in hospitals is still
insuffi cient. However, data quality (at least in terms of diagnoses and procedures)
has been found to have improved considerably following the introduction of
DRGs in many countries. In addition, the authors of the country-specifi c
chapters in Part Two of this book (see, for example, Chapter 13 on France or
Chapter 14 on Germany) highlight that national quality measurement
programmes have been introduced in recent years. If these data are found to
provide valid and reliable indicators for the quality of care, it is likely that there
will be increased efforts to use such data also for payment purposes, called pay-
for-performance (P4P).
10.3 Improving transparency, effi ciency and quality
in hospitals: Recommendations for DRG systems and
DRG-based hospital payment systems
As highlighted in the previous section (10.2), the specifi c design features of
DRG systems and of DRG-based hospital payment systems are of utmost
importance because they determine whether countries will be able to reap the
potential benefi ts of these systems in terms of transparency, effi ciency and
quality in hospitals. This section takes up again the three main building blocks
of DRG-based hospital payment systems introduced in Chapter 3; namely, the
DRG system for patient classifi cation purposes, hospital cost information, and
the actual DRG-based hospital payment (see section 3.2 and Figure 3.1 in
Chapter 3), and makes recommendations concerning the most important issues
that need to be considered when introducing, revising, extending or harmo-
nizing DRG systems and DRG-based hospital payment systems. The section
does not provide detailed instructions in the sense of a ‘how to’ manual, as
readers interested in this kind of information can fi nd it in existing publications
(see Langenbrunner et al., 2009; Cashin et al., 2005).
However, before turning to the building blocks of DRG-based hospital
payment systems, three questions should be explored, which must represent
the starting point for introducing DRGs.
First, is the political situation favourable to the introduction of a DRG system or of a
DRG-based hospital payment system?
While this may seem to be an obvious point, the politics of health policy-
making are too often overlooked (Eggleston et al., 2008). The introduction of
DRG systems has been infl uenced by political agendas, along with the structure
of political and health care systems, by the presence or absence of supporters
and by the general economic and political context (D’Aunno et al., 2008). If
these factors are not conducive to the introduction of a DRG system, the
adoption of DRGs could be delayed or the application of DRGs could be limited
to only certain regions or to a subset of hospitals. Furthermore, as noted in the
chapter on Poland (Chapter 20), in a generally positive economic environment,
the availability of additional fi nancial resources may be able to assure support
from various actors that would otherwise be opposed to the reform.
156 Diagnosis-Related Groups in Europe
Second, is the institutional and legal context adequate for the introduction of DRGs
and DRG-based hospital payment?
One prerequisite for DRG-based hospital payment to work is that purchasers
and providers are separate entities. Public hospitals need to have a certain
degree of autonomy for managing health care resources, for example, as
autonomized organizations with decision rights regarding how to manage
hospital resources (Busse et al., 2002; Langenbrunner et al., 2009). Purchasers
need to have the capacity for managing the DRG system, for monitoring
potential unintended consequences, and for negotiating contracts with private
(profi t-making or non-profi t-making) hospitals. Furthermore, the legal and
institutional context should not prevent the (intended) reorganization of care;
for example, moving the provision of certain services from acute inpatient
hospital care to outpatient care or long-term care settings.
Third, what is the intended purpose of using DRGs?
As illustrated in the country-specifi c chapters in Part Two and as summarized in
Chapter 2, the purpose of using DRGs can change over time. Often countries
begin using DRGs with the aim of improving transparency of hospital activity.
While this can already be ambitious – in terms of DRG system development/
adjustment, management capacities and hospital data requirements – the
(intended and unintended) effects of using DRGs merely as a measure of hospital
activity are likely to be rather limited. Once countries have gathered experience
with a DRG system and have gained confi dence in the ability of the system to
refl ect adequately hospital activity, countries have always started moving
towards using DRGs for determining a progressively increasing proportion of
hospital revenues. Other countries have introduced DRGs directly with the
purpose of using them for hospital payment. The purpose – namely, hospital
activity measurement or hospital payment (in DRG-based case payment or DRG-
based budget allocation systems) – implies different requirements for the
capacity of purchasers and providers, and for the building blocks of the systems.
10.3.1 DRG systems
Countries planning to introduce DRG systems have two options: (1) they can
develop a new DRG system from scratch (as described by Cashin and colleagues
(2005)), or (2) they can import one of the already-existing DRG systems from
abroad. Chapter 4 shows that most countries included in this book have
adopted DRG systems that were originally developed abroad. Those fi rst experi-
menting with DRGs in England, Portugal, France and Ireland used different
versions of DRG systems developed in the United States as the starting point.
Subsequently, several countries adopted DRG systems from the United States,
either Health Care Financing Administration (HCFA-)DRGs or All Patient (AP-)
DRGs (as in Ireland, Spain and Portugal). More recently, Australian Refi ned
(AR-)DRGs have been adopted by a large number of countries in Europe, also
going beyond those included in this book (for example, Ireland as well as
Slovenia (Don, 2003), Croatia (Voncina et al., 2007) and Romania (Radu et al.,
2010)). AR-DRGs served as the origin for developing the German DRG (G-DRG)
Moving towards transparency, effi ciency and quality in hospitals 157
system, which have in turn become the starting point for the development of
DRGs in Switzerland. Finally, Poland has developed its own DRG system on the
basis of the English system. Given that developing a new DRG system is a
highly complex process, requiring several years of work (and which will not
necessarily lead to a superior system compared to the existing ones), adopting
a DRG system from abroad – at least as a starting point for country-specifi c
modifi cations – appears to be the preferable solution.
When deciding which DRG system to adopt, countries need to consider a
wide range of issues, such as the adequacy of the system for the national hospi-
tal context (in terms of clinical acceptability, cost homogeneity, and exist-
ing coding systems for diagnoses and procedures), the availability of training
material and technical support systems (for example, software applications),
and the costs related to obtaining copyright for using the system, in particular
if the DRG system is produced by private enterprises (Don, 2003). Ideally, alter-
native DRG systems are evaluated using available data from hospital discharge
summaries in order to reveal differences in the adequacy of alternative systems
for the country-specifi c context (Aisbett et al., 2007). The additional adminis-
trative costs of coding diagnoses and procedures, installing necessary informa-
tion technology (IT) systems, and enabling data transfer between providers and
purchasers should also be considered when introducing DRGs. In particular,
start-up costs may be higher if a DRG system is chosen that is based on coding
systems for diagnoses and procedures that are not yet used in the country – but
this does not need to be prohibitive, as shown by the case of Ireland, which
adopted the Australian coding system when changing from HCFA-DRGs to
AR-DRGs in 2003 (see Chapter 15).
Historically, most countries that introduced DRG systems initially did so for
the classifi cation of acute hospital inpatients. The reason for excluding out-
patients, day cases, rehabilitation and psychiatric care from DRGs was that
diagnoses were found to be a bad predictor of resource consumption and that
dominant procedures were absent in psychiatric and rehabilitation facilities
(Lave, 2003; Cotterill & Thomas, 2004). However, in recent years many countries
have extended their DRG systems to account for day cases and sometimes have
even included outpatient activity (see Figure 10.1 and Chapter 4). Furthermore,
similar to the situation in the United States, where DRG-like systems were
introduced for rehabilitation facilities in 2002 and for psychiatric facilities in
2005 (MedPAC, 2008, 2010), a number of European countries (such as England,
France and Germany) are extending the concept of DRGs to other types of
hospital care (namely, rehabilitation or psychiatric facilities) or have plans to
do so in the near future (see Chapter 4 and the relevant country-specifi c chapters
in Part Two).
Because hospital activity in most European countries is progressively
expanding into day-case and/or outpatient settings, it is important for countries
to explicitly consider these areas of care when designing or updating their DRG
systems. Some DRG systems – such as NordDRGs, AR-DRGs, and the French
system of patient classifi cation (GHMs) – have been explicitly designed to take
into account day-case and/or outpatient activity, which is important because
otherwise an increasingly important share of hospital activity would be left out
of the systems.
158 Diagnosis-Related Groups in Europe
The purposes of using DRG systems – that is, contributing to transparency in
the hospital sector and paying hospitals fairly for provided services – can only
be achieved if the defi ned groups of patients are suffi ciently homogeneous in
terms of treatment costs. Otherwise, performance comparisons on the basis
of DRGs do not adequately control for differences in patients within the
same groups; and hospital payment for a large number of patients is not
appropriate – it can be either too high or too low. In order to ensure homogeneous
groups of patients, DRG systems need to consider the most important
determinants of resource consumption as classifi cation variables. This can be
achieved only if detailed information relating to treatment costs in hospitals
(see subsection 10.3.2 and Chapter 5) is available for designing and updating
the system. In addition, consultation mechanisms must be established, which
can ensure that input from medical professionals is considered by the responsible
DRG institutions during the process of updating and designing the system. This
is also important because the selection of classifi cation variables must carefully
consider the incentives of using certain variables (such as specifi c procedures)
for defi ning DRGs. If the DRG system is used for hospital payment, that system
should ideally produce neutral incentives for alternative treatment options, in
order to ensure that patients are treated according to their medical needs – and
not according to profi t considerations. Under such circumstances, decisions
regarding which treatment options to choose can be left to clinicians.
As part of the attempt to increase resource homogeneity of DRGs, almost all
systems have seen an expansion in the number of groups over the past few
years (see Chapter 4). Today, the German G-DRG system defi nes 1200 DRGs,
the English HRG system consists of about 1400 HRGs, and the French GHM
system comprises almost 2300 groups. However, an increasingly large number
of groups also brings about problems. First, with an increasingly large number
of groups, it becomes more diffi cult to reliably calculate relevant and signifi cant
differences in the average treatment costs of patients within different DRGs.
Therefore, it does not seem to be a coincidence that larger European countries
Figure 10.1 Extension of DRG systems from acute inpatient care to other sectors
Moving towards transparency, effi ciency and quality in hospitals 159
are operating systems with a larger number of groups because larger countries
should be better able to reliably calculate average costs of patients within
relatively poorly populated DRGs.1 Second, a more complex system is likely to
defi ne groups which are less clearly distinguishable from each other. The
problem is not so much that hospitals would have diffi culties grouping patients
into the appropriate DRGs, because all countries use software tools for the
classifi cation of patients, but rather that, if the criteria used for grouping of
patients into different DRGs are less distinguishable, it becomes increasingly
diffi cult for purchasers or regulators to audit hospital activity and to detect
whether hospitals are engaging in up-coding or gaming (see Chapter 6).
In addition, regular updates of DRG systems are important in order to account
for changes in medical practice and hospital resource consumption, as well as
to incorporate technological innovation. Chapter 9 has shown that most
countries regularly update their DRG systems, albeit at different frequencies
and with a different time-lag between data collection and using those data for
updating the DRG system. Obviously, countries with frequent updates of their
DRG system and with a short time-lag between data collection and use of the
information for DRG-based hospital payment are in a better position to (1)
correct the DRG system if unintended consequences of using a particular
classifi cation variable are detected (for example, unexplained increases in
certain procedures); and (2) incorporate technological innovations into their
systems. In fact, while it is important to have a good DRG system, it is at least
as important to have a well-designed system for monitoring the effects of DRGs
and to update and optimize the system over time.
Finally, DRG systems can be designed to facilitate attempts to incorporate
quality into DRG-based hospital payment systems (for details see subsection
10.3.3). For example, Medicare in the Unites States demands that hospitals code
into the system whether primary and secondary diagnoses were present on
admission (Department of Health and Human Services, 2008). If certain diag-
noses were not present on admission, they are excluded from consideration
during the grouping process. Additionally, for certain high-volume DRGs in
disease areas in which clear consensus exists regarding what constitutes best
practice (for example, cholecystectomy, hip fracture, and stroke), it would be
worth expanding on the concept introduced in the United Kingdom (Depart-
ment of Health, 2011), to explicitly use such best practice care processes instead
of the average across all hospitals. This would ideally lead to more clearly speci-
fi ed groups of patients with more homogeneous care processes, aligned with
best practice guidelines. Obviously, if used for payment, such a process needs to
be accompanied by appropriate measures to ensure that hospitals do not cut
costs by under-providing services.
10.3.2 Hospital cost information
Chapters 2 and 5 have highlighted the importance of accurate cost-accounting
information for the development of DRG systems and for the calculation of
DRG payment rates. However, the availability of high-quality cost-accounting
information is not a prerequisite for the introduction of DRG systems. Many
160 Diagnosis-Related Groups in Europe
countries originally introduced DRG systems and cost weights from abroad also
because they did not have the necessary information for developing their own
systems (as was the case in Ireland, Poland, Portugal and Spain). These countries
adjusted imported DRG weights to the local cost context, using highly
aggregated cost-accounting data and a set of internal DRG cost weights (see for
example Chapter 22) or used data from a previously existing fee-for-service
system for the calculation of weights (see for example Chapter 20). Nevertheless,
even though it is possible to start using DRGs without having high-quality cost-
accounting information, countries usually realize with the passing of time that
better data are required in order to verify that the system and payment rates are
adequate for the local cost context.
Therefore, standardized (sometimes mandatory) cost-accounting systems
have been introduced in at least a sample of hospitals in most of the countries
included in this book. Most frequently, data for the refi nement of DRG systems
and for the calculation of DRG weights are collected from a selected number of
hospitals that use comparable cost-accounting systems meeting predefi ned
quality standards (for example, in Finland, France, Germany and Sweden). How-
ever, the size of the hospital sample varies considerably, between 6 per cent in
Germany and 62 per cent in Sweden. Other countries require all hospitals to
report their activity and unit costs annually to their regulatory authority, but
have fewer demands in terms of the quality and level of detail of this information
(for example, England). In addition, the time-lag varies between data collection
and the use of these data to readjust the DRG system and the DRG payment
rates (see Chapter 9).
In countries in which cost-accounting data are collected from hospitals, this
information is generally used to set DRG weights (the basis of DRG payment
rates) at the average costs of cases within a DRG. However, average costs are
usually calculated only after having excluded outliers through trimming
(Schreyögg et al., 2006). This is because a relatively small number of high-cost
outliers usually accounts for a relatively large proportion of total costs of all
cases within a DRG. Consequently, calculating DRG weights on the basis of
average costs of all cases (including outliers) would lead to an overvaluation of
DRG weights for most cases. Recently, England has moved away from the
concept of using average costs for determining DRG weights for a small number
of high-volume DRGs (for example, hip fracture, cholecystectomy, stroke). For
these DRGs, weights are set to refl ect costs of effi cient high-quality providers
instead of average costs. However, this does not mean that cost-accounting data
become less important. Quite the contrary; very reliable and comparable cost-
accounting data are needed to be able to identify effi cient providers, and to be
sure that lower costs in certain hospitals are not the result of inaccuracies in the
cost-accounting methodology.
When cost-accounting information is used to determine DRG weights, it is
important that only those cost categories are included in the calculation of
average costs that are paid for through the DRG-based hospital payment system.
This is important because many countries use specifi c budgets or other payment
systems for certain cost categories or certain activities (see subsection 10.3.3).
For example, capital costs are not included in DRG weights in some countries
(such as Germany, Ireland and Spain), whereas other countries include capital
Moving towards transparency, effi ciency and quality in hospitals 161
costs in the calculation. Whether to include capital costs when setting DRG
weights depends on the objectives that countries want to achieve. Including
capital costs in DRG weights will imply stronger incentives of the DRG-based
hospital payment system for the reorganization of care, possibly leading to the
concentration of large-scale equipment or certain specialties in fewer hospitals.
While the reorganization of care can be an intended objective, it must be borne
in mind that this could also compromise accessibility of services in poorly
populated rural areas.
There has been some debate about which cost-accounting methodology is
preferable (Tan, 2009). At a theoretical level, there is consensus that bottom-up
micro-costing generates the highest quality of data for developing DRG systems
and for calculating DRG weights (but also for hospital managers in terms of
planning and controlling) because it allows differences in resource consumption
and costs for individual patients to be identifi ed. However, bottom-up micro-
costing is also very demanding in terms of its impact on hospital information
systems, data requirements and analytical complexity. Top-down micro-costing
is more feasible because consumed resources are not valued for individual
patients but for the average patient (see Chapter 5). In addition, top-down
micro-costing has been found to be a fairly accurate alternative to bottom-up
micro-costing, and it is possible to combine both methods and to restrict
bottom-up micro-costing only to the most important cost components (Tan et
al., 2009). By contrast, gross-costing produces relatively inaccurate estimates
because it is unable to trace consumed resources to individual patients.
When deciding on the size of the data sample of cost-collecting hospitals,
there seems to be a trade-off between collecting high-quality cost-accounting
information and the goal of ensuring that a large and representative sample of
hospitals contributes to a national cost database. More complex cost-accounting
systems – collecting more detailed patient-level information using a bottom-up
micro-costing approach – are also more costly to operate, which may make the
data-collection exercise prohibitively costly if it is extended to a large number
of hospitals. Concerning this trade-off, the Netherlands seem to have struck an
interesting balance between representativeness and data quality, by collecting
resource-use data from all hospitals (assuring representativeness of the data)
and unit costs using bottom-up micro-costing from a small sample of hospitals.
Because collecting detailed cost-accounting information requires additional
work from hospitals, regulatory authorities in some countries have started to
provide monetary incentives to hospitals if they comply with predefi ned cost-
accounting standards. For example, in France, the Regional Health Agencies
(ARSs) pay the equivalent of the yearly salary for a fi nancial controller for
hospitals contributing to the national cost database (ENCC). In Germany, the
national DRG institute (the InEK) pays hospitals a lump sum for participating
in the data-collection exercise, and a variable amount of money related to the
number of delivered cases and their data quality. In addition, because cost-
accounting information is of such high importance, almost all countries that
collect cost-accounting data have also implemented monitoring systems to
verify the accuracy of the delivered data. However, if better cost-accounting
information is collected in hospitals, this does not only contribute to more
accurate data for regulators; in addition, hospital managers fi nd this information
162 Diagnosis-Related Groups in Europe
useful because it enables the identifi cation of the most important cost com-
ponents and facilitates comparisons of resource consumption for similar
patients across different hospitals.
10.3.3 DRG-based hospital payment
The countries included in this book have, in general, implemented one of two
main models of DRG-based hospital payment systems: (1) DRG-based case
payment systems (in Estonia, England, Finland, France, Germany, Poland, the
Netherlands and Sweden) and (2) DRG-based budget allocation systems (in
Austria, Ireland, Portugal and Spain; see Chapter 6). In DRG-based case payment
systems, each discharged patient is grouped into the applicable DRG, and
hospitals receive a payment per case that is determined by the weight of that
DRG (after monetary conversion and relevant adjustments). In DRG-based
budget allocation systems, the available regional or national hospital budget is
distributed to individual hospitals on the basis of the number and type of DRGs
that those hospitals produced (namely, the casemix of the hospitals) during one
of the previous years, or that they are expected to produce in the current year.
The existence of these alternative models facilitates the adjustment of the DRG-
based hospital payment system to the country-specifi c context and to the pre-
existing hospital payment system.
Adjusting DRG-based hospital payment to the country-specifi c context and
to take account of the pre-existing payment system is important because new
hospital payment systems should be introduced carefully over extended periods
of time, in order to allow purchasers to monitor the potential unintended
consequences and to give hospitals the necessary time to adjust to the changing
context. Almost all countries included in this book have introduced DRG-based
hospital payment systems over many years, usually operating the new DRG-
based hospital payment system simultaneously with the pre-existing system
and slowly increasing the share of total hospital revenues related to DRGs. For
example, in Ireland, the share of hospital budgets that is determined on the
basis of DRG-based budget allocation has increased progressively from 15 per
cent in 2001 to 80 per cent in 2010. In Estonia, DRG-based case payment
initially accounted for only 10 per cent of hospital payment in 2004, with
the rest being determined on the basis of fee-for-service charges. Later, the
proportion of DRG-based case payments as a percentage of total hospital
payments per discharge was progressively increased to 70 per cent in 2007.
As explained in Chapter 6, there are three main incentives for hospitals
resulting from DRG-based hospital payment systems:
(1) to reduce costs per treated patient,
(2) to increase revenues per patient, and
(3) to increase the number of patients.
These incentives can have both intended and unintended consequences.
Therefore, it is important for countries to take into account the unintended
consequences when designing their DRG-based hospital payment systems as
part of the overall hospital payment system.
Moving towards transparency, effi ciency and quality in hospitals 163
Concerning the fi rst incentive, it is important that hospitals are adequately
paid for the costs of provided services because, otherwise, they may reduce costs
beyond acceptable levels and, in particular, may try to avoid high-cost patients.
Therefore, a whole set of different mechanisms is used by European countries in
order to avoid these unintended consequences: fi rst, in order to adequately
account for high-cost cases, all countries except for the Netherlands and Spain
provide per diem-based additional payments to hospitals for outlier cases that
stay in hospitals for longer than a specifi ed length-of-stay threshold (for
example, Austria, Germany, France, England, Ireland and Portugal) or additional
fee-for-service payments for cases that exceed a specifi ed cost threshold (for
example, Estonia, Finland and Sweden). Second, additional payments are
provided for certain high-cost services that are not adequately fi nanced through
the normal DRG-based payment system (for example, for certain high-cost
drugs or devices), and some countries have defi ned specifi c DRGs for intensive
care treatment according to the length of stay in these departments (such as in
Germany), or fi nance treatment in intensive care units (ICUs) on the basis of
per diem-based surcharges (such as Austria). Third, procedures have come to
play a much more important role in most European DRG systems (see Chapter
4) compared to the DRG systems originally developed in the United States.
Fourth, several countries have introduced adjustment factors to take into
account structural differences between hospitals and to provide adequate
payments to different kinds of hospitals (see Chapter 6).
In order to avoid hospitals being able to increase revenues per treated patient
through up-coding or gaming, several countries have installed systems for
regular auditing. For example, in Germany, the regional medical review boards
of the sickness funds send teams to randomly selected hospitals to audit
patients’ medical records in order to evaluate whether they are correctly coding
and treating patients (MDS, 2011). In 2009, 12 per cent of all hospital cases
were audited by the sickness funds, resulting in average claw-back sums of
around €800 per audited case. In France, a total of 1 per cent of hospital
discharges were audited by the Regional Hospitalization Agencies (ARHs) in
2006, which found that 60 per cent of evaluated records had some kind of
coding error. It is important for regulators to monitor both the adequacy of
hospital treatment and whether it was really necessary for patients to be treated
as inpatients.
Countering the third incentive of DRG-based hospital payment systems –
that is, to increase the number of patients – several countries have introduced
global expenditure control measures. For example, some countries are operating
their DRG-based case payment systems within predefi ned volume limits. In
Germany, DRGs are used to negotiate ‘revenue budgets’, which limit (to a
certain degree) the total amount of money that hospitals can earn from
DRG-based case payments. If hospitals provide more DRGs than agreed, they
have to pay back at the beginning of the next year a certain percentage of the
DRG-based case payments that they earned in excess of the negotiated revenue
budgets (and they are rewarded with increased payments per case if they
remained below the budget). By contrast, in the Netherlands, hospitals do not
receive any payments for those cases treated in excess of the budget set
prospectively which cannot be negotiated between insurers and hospitals.
164 Diagnosis-Related Groups in Europe
Similarly, aiming to achieve expenditure control, France and Poland adjust
national DRG-based case-payment rates in order to stay within global expen-
diture targets. However, in France, this approach is criticized for not being
transparent enough, because payment rates are progressively set independently
of average costs. Instead, Or and Bellanger (see Chapter 13 of this volume) argue
in favour of clear volume targets for hospitals.
As chapters 6 and 8 have shown, all three DRG-inherent incentives may both
improve or compromise quality of care. Although quality has been of continuous
concern for policy-makers across Europe, it is still relatively rarely explicitly
taken into account in existing DRG-based hospital payment systems. However,
as evidenced by the examples presented in Chapter 8, it is possible to refi ne
these systems to integrate direct incentives for improving quality. For example,
DRG-based payments can be adjusted at the hospital level by increasing
payments for all patients treated by one hospital, if one hospital provides
above-average quality as measured through hospital-level quality indicators.
Similarly, it is possible to increase payments to a hospital for all patients falling
into one DRG if the hospital scores above average on DRG-specifi c quality
indicators, or to adjust payments for individual patients if quality can be more
robust monitored at the individual patient level (see Table 10.1). Yet, an essential
prerequisite is that reliable quality indicators are developed and that more
robust data about quality of care are collected in hospitals. Consequently, most
countries have started collecting more detailed information regarding quality
in hospitals in order to ensure that care quality is not compromised by the cost-
reduction incentives of DRG-based hospital payment systems.
In addition, as discussed in Chapter 9, it is important that countries take into
account the effect that DRG-based hospital payment may have on the adoption
and use of technological innovations. Chapter 9 showed that most (but not all)
countries included in this book have complemented their DRG-based payment
systems with specifi c short-term payment instruments targeted at encouraging
the adoption and use of technological innovations. However, short-term
payment instruments should be employed very carefully, and granted only
after careful assessment of the likely effects of the technology in question on
costs, as well as quality of care. They should be limited to technological inno-
vations that offer either considerable quality improvements over existing
technologies, or options for diagnosis and treatment of previously untreatable
conditions. Otherwise, if countries should want to provide short-term payment
incentives for technological innovations with expected signifi cant quality
improvements but for which the evidence remains uncertain, one possible
approach is that of so-called Coverage with Evidence Development (CED) (see
Chapter 9, and Hutton and colleagues (2007)).
The payment of hospitals in all countries therefore consists of a highly
sophisticated mix of different payment mechanisms that aim to modify the
type and strength of the incentives of DRG-based hospital payment. The
resulting intricately blended payment systems – incorporating elements of fee-
for-service payment, per diem payment and global budgets – are more likely to
contribute to achieving the societal objectives of securing high-quality hospital
care at affordable costs than any other hospital payment mechanism alone
(Ellis & McGuire, 1986).
Moving towards transparency, effi ciency and quality in hospitals 165
Table 10.1 Options for integrating quality into DRG-based hospital payment systems
and examples from selected countries in Europe and the United States
Type of payment
adjustment/
calculation
Mechanism Examples
Hospital based • Payment for entire
hospital activity is
adjusted upwards or
downwards by a certain
percentage
• Hospital receives specifi c
budgetary allocation
unrelated to activity
• Predefi ned quality results are met/not
met (for example, in England)
• Overall readmission rate is below/above
average or below/above agreed target
(for example, in the United States)
• Hospitals install new quality
improvement measures (for example,
in France)
DRG/disease
based
• Payment for all patients
with a certain DRG (or a
disease entity) is adjusted
upwards or downwards
by a certain percentage
• DRG payment is not
based on average costs
but only on costs of
those hospitals delivering
‘good quality’
• Insurers negotiate with hospitals
that DRG payment is higher/lower if
certain quality standards are met/not
met (for example, in Germany and the
Netherlands)
• DRG payment for all hospitals is
based on ‘best practice’; that is, costs
incurred by effi cient, high-quality
hospitals (for example, in England)
Patient based • No payment is made for
a case
• Payment for an indivi-
dual patient is adjusted
upwards or downwards
by a certain amount
• Readmissions within 30 days are not
paid separately but as part of the
original admission (for example, in
England and Germany)
• Complications (that is, certain
conditions that were not present upon
admission) cannot be used to classify
patients into DRGs that are weighted
more heavily (for example, in the
United States)
Figure 10.2 illustrates that DRG-based hospital payments generally account for
only part of total hospital revenues. In the fi gure, the DRG-based payments –
which are often operated partially and/or during the implementation phase
simultaneously with a pre-existing hospital payment system (such as global
budgets or fee-for-service payments) – constitute the basis of hospital revenues.
These DRG-based hospital payments are often already adjusted for high-cost
cases through outlier payments, as well as for quality and/or for structural
differences between hospitals through structural adjustment factors.
On top of this, hospitals may receive additional payments for specifi c
activities for DRG-classifi ed patients, for example for certain expensive drugs,
for certain services that are not adequately accounted for in the DRG system,
and for certain cost-increasing technological innovations. Such payments may
be integrated to different degrees into the DRG-based hospital payment systems,
for example in the form of ‘unbundled HRGs’ in England or supplementary
payments in Germany.
166 Diagnosis-Related Groups in Europe
A further element of hospital revenue originates from payments for patients
not classifi ed into the DRG system. The extent of these payments depends on
both the types of activities hospitals are undertaking (for example, whether
outpatients constitute a large part of their activity) and whether these have
been incorporated into the DRG system (see Figure 10.1).
Furthermore, certain hospitals usually receive additional payments or budgets
for non-patient care activities, such as teaching and research (although some
countries may account for the extra costs of teaching and research within their
DRG-based budget allocation model by operating separate systems for teaching
hospitals and non-teaching hospitals – see, for example, Chapter 15) or emer-
gency availability. Finally, several countries pay separately for capital invest-
ments (buildings and expensive equipment) or for certain structural quality
measures, such as infection control programmes.
10.4 Conclusions: Future of DRG systems in Europe
Based on the experiences of the 12 countries included in Part Two of this book,
the previous section has made recommendations regarding how best to design
DRG systems, how to improve hospital cost information and how to maximize
the intended consequences of DRG-based hospital payment systems, while
Figure 10.2 DRG-based hospital payment within the mix of total hospital revenues
Moving towards transparency, effi ciency and quality in hospitals 167
avoiding the unintended ones. These recommendations may contribute towards
improved national DRG systems and better DRG-based hospital payment
systems in different countries. However, because the goals of European countries
and the problems they face are highly similar, it is at least worth considering the
benefi ts of increased cooperation, coordination and harmonization of DRG sys-
tems in Europe.
Currently, six of the twelve countries included in this book develop, update
and operate their own national DRG systems. The other six countries use either
imported DRG systems from abroad (for example, from Australia and the United
States) or a national version of the common Nordic system of patient classi-
fi cation (NordDRGs). Each country with a national DRG system analyses its
own national database to improve resource homogeneity of DRGs; develops
and updates its own cost-accounting guidelines; has developed its own national
consultation mechanisms with medical professionals; develops national
software applications; evaluates technological innovations; updates national
procedure coding systems, and so on. This raises two important questions: (1)
Do all countries have the fi nances and skills to do this? And (2) is it worth it?
In regard to the fi rst question, the answer – at least for smaller countries – is
a clear ‘no’. For practical reasons, without pan-European cooperation, these
countries will always need to import certain important elements of their DRG
systems. Further, if they have to do so anyway, it is not evident why imported
DRG systems from outside Europe – which are used in several European
countries – should be better able to defi ne homogeneous groups of patients in
these countries than a common European DRG system. In regard to the second
question, one might argue that these efforts were worthwhile if the resulting
national DRG systems were really tailor-made to achieve national objectives
and better adjusted to the country-specifi c context than a multi-country
solution. Before the EuroDRG project (which inspired this book), we did not
know whether this was the case, because the ability of different DRG systems to
defi ne homogeneous groups of patients (in terms of clinical meaningfulness
and costs) had not been assessed across European countries.
If the factors to explain cost differences (in terms of the patient characteristics
and diagnoses as well as procedures performed and services provided) were
suffi ciently similar across European hospitals (and the parallel work of the
EuroDRG project – to be published in 2012 – shows that this is the case), there
would be a case for cooperation in terms of the development of DRG systems in
Europe. The benefi ts would include: (1) avoiding duplication of work, (2)
improving knowledge exchange in the refi nement of DRG systems, (3) increas-
ing transparency of hospital services across countries, and (4) facilitating cross-
border movements of patients and payments. However, similar to the historical
emergence of DRG systems as a result of political decisions, a coordination of
European DRG systems – and, ultimately, possibly a harmonized DRG system
– is likely to emerge only if there is suffi ciently strong political will to support
the emergence of a common European hospital market, as well as an increasing
level of mobility of European patients. While this may be an unrealistic scenario
in the short term, the recent Directive on the Application of Patients’ Rights in
Cross-Border Healthcare (European Parliament and Council, 2011) demonstrates
that now is the time to start such a discussion.
168 Diagnosis-Related Groups in Europe
The NordDRG system (see Chapter 16) provides an example of the feasibility
of developing a common DRG system for a group of countries. NordDRGs
emerged from existing cooperation between Nordic countries in the develop-
ment of a common procedure classifi cation system, and the presence of a
common problem across Nordic countries during the mid-1990s; namely, how
to convert the national or imported DRG systems from using the International
Classifi cation of Diseases (ICD) 9th revision for the coding of diagnoses to the
ICD-10 codes. Consequently, countries amalgamated their efforts to develop a
common DRG system that would replace existing national systems and
imported DRG systems from abroad. The example of NordDRGs shows that a
common DRG system does not prevent the adaptation of the common system
to meet country-specifi c needs. Since the very beginning of NordDRGs, several
countries have developed national versions of the system, and DRG weights are
always calculated separately for each country. In addition, country-specifi c
modifi cations of the underlying classifi cations of diagnoses and procedures
exist, adding further detail where necessary but conforming to the general logic
of the systems. Every year, NordDRGs are jointly updated by the Nordic Casemix
Centre and country-specifi c modifi cations are then added to the updated
version of the common NordDRG system.
The example of NordDRGs suggests that a fi rst requirement for a common
European DRG system (which could be called the ‘EuroDRG’ system) would be
to harmonize the coding of diagnoses and procedures, or – as a second-best
option – to develop a mapping system that would allow translation of codes
from different coding systems into a common European coding system. The
Hospital Data Project as part of the European Union (EU)’s Health Monitoring
Programme has suggested a common – albeit for patient classifi cation purposes,
too rudimentary – format for hospital activity data, to improve comparability
(Kiwa Prismant, 2008). For the coding of diagnoses, an agreement on a coding
system should be relatively unproblematic, since the ICD-10 is already used for
cause-of-death statistics in all countries. For procedures, an agreement could be
more diffi cult to reach. This is testifi ed by four decades of work, but the as yet
unfi nished attempt to develop such an international classifi cation system,
initially termed the International Classifi cation of Procedures in Medicine
(ICPM), and later the International Classifi cation of Health Interventions
(ICHI). European countries may consider not waiting for this development to
be fi nished but to coordinate their efforts based on their own coding and
patient classifi cation systems.
As a starting point, the EuroDRG project has not only compared the DRG
systems and their effects on transparency, effi ciency and quality (in this
volume), but has also compared in depth the classifi cation of patients into
DRGs across the DRG systems for 10 episodes of care. A common EuroDRG
system could draw on the best features of national DRG systems, such as the
most relevant classifi cation variables, concepts for the defi nition of severity
groups (for example, the patient clinical complexity levels (PCCLs), as used in
AR-DRGs and G-DRGs; see Chapter 4) or the defi nition of short-stay groups, as
in NordDRGs. However, detailed cost information collected on the basis of a
standardized cost-accounting system from a suffi ciently large and representative
sample of hospitals from all participating countries would be necessary in order
Moving towards transparency, effi ciency and quality in hospitals 169
to test the ability of such a EuroDRG system to defi ne homogeneous groups of
patients across different countries.
A common EuroDRG system would not need to be employed in all countries
and all hospitals from the beginning. It could initially be used only for the
purpose of increasing transparency, possibly even coexisting simultaneously
with national DRG systems, which could continue to be used for payment
purposes for a limited time period – similar to the current situation in Spain (see
Chapter 22). Furthermore, country examples included in this book show that it
is possible to use DRG systems only for a subset of voluntarily participating
hospitals (for example, as is the case in Ireland; see Chapter 15) or for certain
regions (as in Spain; see Chapter 22). Similarly, EuroDRGs could initially be
used only in certain countries, in certain hospitals interested in international
performance comparisons, or for those patients treated in countries in which
they are not permanent residents.
The starting point of this book (see Chapter 1) was the problem formulated
by Dr. Eugene Codman in 1913: ‘Really the whole hospital problem rests on
one question: What happens to the cases? [. . .] We must formulate some method
of hospital report showing as nearly as possible what are the results of the
treatment obtained at different institutions.’ While this book has demonstrated
that DRGs have contributed to improved transparency within hospitals, the
concept of DRGs has one important drawback: they are almost always restricted
to one hospital stay, and providers are not encouraged to take into account the
long-term effects of their treatment(s) in terms of continuity of care, patient
outcomes and, ultimately, population health. While certain modifi cations of
DRG-based hospital payment systems – such as not-paying for readmissions in
England and Germany – aim to overcome (parts of) these problems, measuring
the wider performance of hospitals in terms of the named outcomes remains a
major obstacle. In this respect, only the Dutch DBCs have the advantage of
defi ning groups on the basis of the treatment that is necessary for a specifi c
condition, independent of the number of outpatient visits, diagnostic tests and
inpatient admissions. Therefore, future developments of DRGs should be linked
to efforts that aim to measure and ultimately increase the performance of health
systems as a whole.
10.5 Note
1 The Netherlands may be considered an exception to this rule but they are currently
in the process of reducing the complexity of their system; see Chapter 23.
10.6 References
Aisbett, C., Wiley, M.M., McCarthy, B., Mulligan, A. (2007). Measuring Hospital Casemix:
Evaluation of Alternative Approaches for the Irish Hospital System. Dublin: Economic and
Social Research Institute (ESRI Working Paper No. 192).
Busse, R., van der Grinten, T., Svensson, P. (2002). Regulating entrepreneurial behaviour
in hospitals: theory and practice, in R.B. Saltman, R. Busse, E. Mossialos, eds.
Regulating Entrepreneurial Behaviour in European Health Care Systems. Buckingham:
Open University Press.
170 Diagnosis-Related Groups in Europe
Cashin, C., O’Dougherty, S., Yevgeniy, S. et al. (2005). Case-Based Hospital Payment Systems:
A Step-By-Step Guide for Design and Implementation in Low- and Middle-Income Countries.
Bethesda, MD: United States Agency for International Development (USAID)
ZdravPlus Project.
Cotterill, P.G., Thomas, F.G. (2004). Prospective payment for Medicare inpatient psychia-
tric care: assessing the alternatives. Health Care Financing Review, 26(1):85–101.
Cour des comptes (2009). La Securité Sociale, chapitre 7, La mise en place de la T2A: Bilan à
mi-parcours. Paris, Cour des comptes (February).
Cutler, D. (1995). The incidence of adverse medical outcomes under prospective payment.
Econometrica, 63(1):29–50.
D’Aunno, T., Kimberly, J.R., de Pouvourville, G. (2008). Conclusions: the global diffusion
of casemix, in J.R. Kimberly, G. de Pouvourville, T. D’Aunno, eds. The Globalization of
Managerial Innovation in Health Care. Cambridge: Cambridge University Press.
Department of Health (2011). Best Practice Tariffs. London: Department of Health (http://
www.dh.gov.uk/en/Managingyourorganisation/NHSFinancialReforms/DH_105080,
accessed 27 July 2011).
Department of Health and Human Services (2008). Changes to the Hospital Inpatient
Prospective Payment Systems and Fiscal Year 2009 Rates. Washington, DC: United States
Department of Health and Human Services.
Don, H. (2003). Implementing DRGs in Slovenia: why the Australian variant was selected.
Australian Health Review, 26:50–60.
Eggleston, K., Shen, Y.C., Lau, J., Schmid, C.H., Chan, J. (2008). Hospital ownership and
quality of care: what explains the different results in the literature? Health Economics,
17(12):1345–62.
Ellis, R.P., McGuire, T.G. (1986). Provider behavior under prospective reimbursement –
cost-sharing and supply. Journal of Health Economics, 5(2):129–51.
European Parliament and Council (2011). Directive 2011/24/EU of 9 March 2011 on the
Application of Patients’ Rights in Cross-Border Healthcare. Brussels: Offi cial Journal of the
European Union (L88/45-L88/65).
Farrar, S., Yi, D., Sutton, M. et al. (2009). Has payment by results affected the way that
English hospitals provide care? Difference-in-differences analysis. British Medical
Journal, 339:b3047.
Fetter, R.B., Thompson, J.D., Mills, R.E. (1976). A system for cost and reimbursement
control in hospitals. Yale Journal of Biology and Medicine, 49:123–36.
Fürstenberg, T., Laschat, M., Zich, K. et al. (2011). G-DRG Begleitforschung gemäß §17b Abs.
8 KHG: Endbericht des zweiten Forschungszyklus. Siegburg: Institut für das Entgeldsystem
in Krankenhaus (InEK).
Gilman, B.H. (2000). Hospital response to DRG refi nements: the impact of multiple reim-
bursement incentives on inpatient length of stay. Health Economics, 9(4):277–94.
Goldfi eld, N. (2010). The evolution of diagnosis-related groups (DRGs): from its
beginnings in casemix and resource use theory, to its implementation for payment
and now for its current utilization for quality within and outside the hospital. Quality
Management in Health Care, 19(1):3–16.
Hutton, J., Trueman, P., Henshall, C. (2007). Coverage with evidence development: an
examination of conceptual and policy issues. International Journal of Technology
Assessment in Health Care, 23:425–32.
Kahn, K.L., Keeler, E.B., Sherwood, M.J. (1990a). Comparing outcomes of care before and
after implementation of the DRG-based prospective payment system. Journal of the
American Medical Association, 264(15):1984–8.
Kahn, K.L., Rogers, W.H., Rubenstein, L.V. et al. (1990b). Measuring quality of care with
explicit process criteria before and after implementation of the DRG-based prospective
payment system. Journal of the American Medical Association, 264(15):1969–73.
Moving towards transparency, effi ciency and quality in hospitals 171
Kiwa Prismant (2008). Hospital Data Project Phase 2: Final Report. The Need for Metadata and
Data. Utrecht: Kiwa Prismant.
Kiwa Prismant (2010). Kengetallen Nederlandse Ziekenhuizen 2009 [Key Numbers Dutch
Hospitals 2009]. Utrecht: Dutch Hospital Data.
Kosecoff, J., Kahn, K.L., Rogers, W.H. et al. (1990). Prospective payment system and im-
pairment at discharge. The ‘quicker-and-sicker’ story revisited. Journal of the American
Medical Association, 264(15):1980–3.
Langenbrunner, J., Cashin, C., O’Dougherty, S. (2009). Designing and Implementing Health
Care Provider Payment Systems. How-To Manuals. Washington, DC: World Bank (http://
siteresources.worldbank.org/HEALTHNUTRITIONANDPOPULATION/Resources/Peer-
Reviewed-Publications/ProviderPaymentHowTo , accessed 5 September 2011).
Lave, J.R. (2003). Developing a Medicare prospective payment system for inpatient
psychiatric care. Health Affairs, 22(5):97–109.
Ljunggren, B., Sjödén, P. (2003). Patient-reported quality of life before, compared with
after a DRG intervention. International Journal for Quality in Health Care, 15(5):
433–40.
MDS (2011). Abrechnungsprüfungen der MDK in Krankenhäusern sind angemessen,
wirtschaftlich und zielführend. Zahlen und Fakten der MDK-Gemeinschaft. Essen:
Medizinischer Dienst des Spitzenverbandes Bund der Krankenkassen e.V.
MedPAC (2008). Rehabilitation Facilities (Inpatient) Payment System. Washington, DC:
Medicare Payment Advisory Commission.
MedPAC (2010). Psychiatric Hospital Services Payment System. Washington, DC: Medicare
Payment Advisory Commission.
Radu, C., Chiriac, D.N., Vladescu, C. (2010). Changing patient classifi cation system for
hospital reimbursement in Romania. Croatian Medical Journal, 51(3):250–8.
Rogers, W.H., Draper, D., Kahn, K.L. et al. (1990). Quality of care before and after
implementation of the DRG-based prospective payment system. A summary of
effects. Journal of the American Medical Association, 264(15):1989–94.
Schreyögg, J., Stargardt, T., Tiemann, O., Busse, R. (2006). Methods to determine reim-
bursement rates for diagnosis related groups (DRG): a comparison of nine European
countries. Health Care Management Science, 9(3):215–23.
Schwartz, M.H., Tartter, P.I. (1998). Decreased length of stay for patients with colorectal
cancer: implications of DRG use. Journal for Healthcare Quality, 20(4):22–5.
Street, A., Scheller-Kreinsen, D., Geissler, A., Busse, R. (2010). Determinants of hospital
costs and performance variation: methods, models and variables for the EuroDRG
project. Working Papers in Health Policy and Management, 3:1–44.
Tan, S.S. (2009). Micro-costing in Economic Evaluations: Issues of Accuracy, Feasibility,
Consistency and Generalizability. Rotterdam: Erasmus Universiteit Rotterdam.
Tan, S.S., Rutten, F.F., van Ineveld, B.M., Redekop, W.K., Hakkaart-van Roijen, R.L. (2009).
Comparing methodologies for the cost estimation of hospital services. European
Journal of Health Economics, 10:39–45.
Voncina, L., Strizrep, T., Dzakula, A. (2007). The introduction of DRGs in Croatia.
Eurohealth, 13(1):4–5.
Part Two
chapter e l e v e n
Austria: Inpatient care
and the LKF framework
Conrad Kobel and Karl-Peter Pfeiffer
11.1 Hospital services and the role of DRGs in Austria
11.1.1 The Austrian health system
In Austria responsibility for coordination and planning of health care provision
and fi nancing has been traditionally shared between the federal Government,
the nine states, called Länder in German, and their municipalities. In addition,
several self-governing bodies representing physicians and pharmacists play an
important role. Although the Ministry of Health delegates many tasks to the
states or to the self-governing bodies, it remains by far the most infl uential
actor in health policy-making and nationwide planning. It is responsible for
supervision of the health insurance funds, enforcement of bills, and regulation
of the training of health care professionals (Hofmarcher & Rack, 2006).
In general the Ministry of Health is responsible for enacting legislation, while
the implementation of health care-related legislation is the responsibility of the
states. However, states can also pass legislation concerning the inpatient sector.
In practice, the Federal Health Commission and the Health Platforms at state
level – composed of the state governments, the social health insurance institu-
tions and the federal Government – are the most infl uential actors in terms of
the actual implementation. In addition, the physicians’ chambers, municipali-
ties, patients and hospital owners are represented in the Health Platforms
(Hofmarcher & Rack, 2006).
An agreement according to the Federal Constitution (B-VG) Article 15a
between the states and the federal Government is signed and adjusted every
four years to establish and coordinate the constitutional duties. As part of this
agreement the Austrian Structural Plan for Health is elaborated (ÖBIG, 2008),
which guarantees nationwide standards in health care. Regarding inpatient
care, the plan specifi es standards for bed capacities and the availability of
176 Diagnosis-Related Groups in Europe
infrastructure (such as major technological equipment) in hospitals. The states
are obliged to ensure that these standards are met, which is achieved through
regional health plans that specify regional standards for health care providers
and their structural characteristics.
Since the early 1990s, expenditures on health have increased rapidly, both in
absolute terms and as a share of gross domestic product (GDP). Total health
expenditures (not including long-term care) grew from about €10 billion (7.4
per cent of GDP) in 1990 to almost €26 billion (9.2 per cent of GDP) in 2008
(Statistics Austria, 2010). In the same period, the share of public expenditure for
health slightly increased from 73.4 per cent in 1990 to 76.9 per cent in 2008
(Statistics Austria, 2010).
In the statutory health insurance (SHI) system, membership is mandatory for
almost 99 per cent of the population and depends either on profession or region
of residence. SHI funds therefore do not compete. Income-related contributions
to SHI are shared equally between employers and employees. Together this
amounts to about 46 per cent of health care expenditure. Another 30 per cent
of expenditure is paid from taxes. A total of 24 per cent is paid as co-payments
either by additional private health insurance or out-of-pocket payments by
patients (Thomson et al., 2009).
All SHI funds are organized in the Federation of Austrian Social Insurance
Institutions, which administers delegated tasks from the federal Government
and the states, such as contracting with the Austrian Physicians’ Chamber or
the Austrian Pharmacists’ Association on prices for outpatient services and
reimbursement of medications. Outpatient care is mainly provided by three
types of providers: individual self-employed physicians, outpatient clinics and
hospital outpatient departments.
Inpatient care is to a great extent provided by public or non-profi t-making
hospitals (see subsection 11.1.2) and is fi nanced through State Health Funds
(Landesgesundheitsfonds in German; SHF), which are supervised and managed
by the respective State Health Platforms (BMGFJ, 2008e). Since 1997, resources
from the SHFs have been allocated to hospitals on the basis of the Austrian
performance-oriented hospital fi nancing framework (Leistungsorientierte Kranke-
nanstaltenfi nanzierung in German; LKF). The introduction of the LKF framework
shifted hospital budget allocation for most inpatient services in most hospitals
away from a per diem-based system to one that allocates a signifi cant propor-
tion of hospital budgets on the basis of a diagnosis-related group (DRG)-like
patient classifi cation system.
To distinguish between the general hospital fi nancing framework, that is, the
LKF framework and the patient classifi cation system (PCS), we will refer to the
latter as the LKF-PCS, although the term is not used in Austria. The LKF frame-
work applies to all inpatient services covered by SHI, including rehabilitation
and psychiatric care. The LKF-PCS applies only to (most) areas of acute in-
patient hospital care. All public and non-profi t-making hospitals are primarily
fi nanced through the LKF framework.
Private profi t-making hospitals, on the other hand, obtain their funds
primarily through out-of-pocket payments by patients and from private health
insurance companies. In addition, the nationwide Private Hospitals Financing
Fund (PRIKRAF), which is exclusively funded by SHIs, pays private profi t-
Austria: Inpatient care and the LKF framework 177
making hospitals for health services covered by private health insurance.
PRIKRAF allocates fi nancial resources to hospitals exclusively on the basis of the
same system of DRG-type hospital budget allocation that is used in public and
private non-profi t-making hospitals (Hofmarcher & Rack, 2006).
A more detailed description of the Austrian health care system can be found
in Hofmarcher and Rack (2006).
11.1.2 Hospital services in Austria
Austrian hospitals provide a broad range of services. Amongst others, the
Federal Hospitals Act (KAKuG) defi nes as hospitals (BMGFJ, 2008b) the following
types of institutions:
• general hospitals for all patients without distinction according to gender, age
or type of medical care provided;
• special hospitals for the examination and treatment of patients with par-
ticular illnesses or patients in particular age groups, or for other special
purposes;
• convalescent homes for patients in need of medical care and special nursing
care while they convalesce;
• homes for chronically ill patients in need of medical care and special nursing
care;
• maternity clinics and maternity homes;
• sanatoria – hospitals specially equipped to provide higher standards of board
and accommodation;
• independent outpatient health care centres (X-ray clinics, dental care centres
and similar facilities) – organizationally independent facilities for the
examination or treatment of patients who do not require inpatient care.
Table 11.1 provides an overview of the hospital infrastructure in Austria.
In 2006 there were 264 registered hospitals equipped with 63 354 beds, which
corresponds to 7.66 beds per 1000 residents and a utilization level of 32.5
admissions per 100 residents. All Austrian hospitals together provided more
than 18 million bed days (BMGFJ, 2008b) in 2006.
There were 183 acute care hospitals in 2006, providing 52 894 beds. Of those,
133 were public or private non-profi t-making hospitals funded by the SHFs.
They provided 92 per cent of beds in acute care hospitals. A total of 43 hospitals
were private profi t-making hospitals. The remaining 7 were prison or military
hospitals. Hospitals funded by the SHF had an average length of stay (ALOS) of
5.71 days (without day cases and long-term care) (BMGFJ, 2008b). For the
remainder of this chapter we focus on acute care hospitals.
The sources of fi nancing for inpatient care in Austria are regulated nationwide
by the Article 15a B-VG treaties. The federal Government, states, local authorities
and health insurance funds contribute to a global national budget, which is
then allocated to the SHFs based on state quotas. While health insurance funds
pay a fl at fee, the federal Government, the states and local authorities pay based
on fi xed percentages of value-added taxes. In addition, in most states, payments
to balance structural defi cits by local authorities and the state are also included
178 Diagnosis-Related Groups in Europe
Table 11.1 Hospital infrastructure in Austria
Hospitals Beds Hospitals Beds
Absolute no. % of total
Non-acute care hospitals 81 10 460 30.7 16.5
Acute care hospitals 183 52 894 69.3 83.5
hospitals funded by SHF 133 48 870 50.4 77.1
hospitals funded by PRIKRAFa 43 4031 16.3 6.4
Total 264 63 354 100.0 100.0
Source: BMGFJ, 2008b.
a Applies to certain cases/services.
Table 11.2 Payments to the SHFs in 2005
%
Federal States Local authorities Health insurance
Governmenta institutions
Burgenland 11.3 45.3 3.6 39.8
Carinthia 11.7 33.2 8.0 47.2
Lower Austria 10.7 31.0 20.9 37.5
Salzburg 12.4 32.1 10.7 44.8
Styria 18.7 34.6 1.3 45.4
Tyrol 22.8 17.9 15.5 43.8
Upper Austria 13.0 22.5 16.8 47.7
Vienna 15.6 39.1 1.7 43.5
Vorarlberg 12.5 25.6 14.3 47.6
Austria 14.7 31.6 9.9 43.8
Source: Based on Grossmann & Hauth, 2007.
a Including payments for university teaching hospitals.
in the budgets (Hofmarcher & Rack, 2006). The sum of all fi nancial contributions
to SHFs determines the global budget of resources available for the fi nancing of
hospital services provided in that state for a given year (Table 11.2).
Financing of acute care hospitals is regulated by the national LKF framework,
which provides the basis for resource allocation from SHFs to hospitals. The LKF
framework consists of two areas: (1) the core area (Kernbereich in German) that is
made up of the nationwide DRG-like patient classifi cation system (the LKF-PCS);
and (2) a state-specifi c steering area (Steuerungsbereich in German). When the LKF
framework was introduced, the idea of the steering area was to enable quality-
related payments and to allow for a transitional period before hospital resource
allocation would be exclusively based on the LKF-PCS. Yet, states continue to use
the steering area for different purposes, such as assuring higher payments to
university hospitals, providing additional resources to hospitals in alpine areas or
compensating for defi cits of hospitals (see subsection 11.5.3).
Austria: Inpatient care and the LKF framework 179
11.1.3 Purpose of the Austrian LKF-PCS
During the 1980s and 1990s, cost infl ation in the Austrian health care system
and in particular in the inpatient sector exceeded annual growth of GDP. Austria
was spending more on inpatient care than the average of the EU15 countries
(belonging to the European Union before May 2004) and had one of the longest
ALOS (OECD, 2010). The old per diem-based hospital fi nancing system did not
provide incentives for cost–effectiveness or for a reduction in lengths of stay. In
addition, transparency in terms of hospital activity was poor, as no detailed
structured data relating to diagnoses or procedures were available.
Inspired by DRG-based hospital payment systems in other countries, it was
decided to introduce a new hospital payment system to overcome these
problems. Using a DRG-like patient classifi cation system was expected to increase
cost–effectiveness, limit cost infl ation and contribute to a reduction of the
ALOS – all while guaranteeing high-quality health care. In particular, using
DRGs for hospital payment was intended to reduce multiple diagnostic proce-
dures and to promote a shift from inpatient care to ambulatory care, thus con-
tributing to a reduction in hospital beds (BMGFJ, 2008e). Furthermore, it was
hoped that the introduction of a DRG-like system would increase transparency
and improve documentation quality (BMGFJ, 2008f).
Today, the primary purpose of the LKF-PCS as part of the LKF framework is to
enable activity-based budget allocations to Austrian acute care hospitals. Beyond
that, the LKF framework provides a catalogue of diagnoses, an Austrian modifi -
cation of the World Health Organization’s International Classifi cation of Dis-
eases (ICD-10-WHO), and a catalogue of selected procedures; ‘selected’ to the
effect that only expensive and highly frequent procedures are listed. Together
with the Austrian Structural Plan for Health, the LKF framework is also used as a
planning and steering instrument by stipulating minimum department sizes,
staffi ng standards or volume thresholds as prerequisites for the fi nancing of cer-
tain services.
11.2 Development of and updates relating to the LKF-PCS
11.2.1 The LKF-PCS at a glance
The Ministry of Health is the owner of the LKF-PCS and as such responsible for
the content, structure, development and maintenance of the system. As already
mentioned, the Austrian LKF-PCS is the backbone of the core area (Kernbereich)
of the LKF framework. The LKF-PCS defi nes procedure- and diagnosis-oriented
case groups (Leistungsorientierte Diagnosefallgruppen in German; LDF).
The LKF-PCS classifi es each hospital case into exactly one of 979 LDFs. There
are two main steps (for details see section 11.3). First, in the event that a
patient has undergone at least one grouping relevant procedure listed in the
Austrian catalogue of procedures, (s)he is grouped into one of the 209 single
medical procedure-based groups (medizinische Einzelleistungen in German;
MEL). Otherwise, the patient is grouped into one of 219 main diagnosis
groups (Hauptdiagnose-Gruppen in German; HDG). These groups are similar to
180 Diagnosis-Related Groups in Europe
base-DRGs, adjacent DRGs, or classes used in other DRG-like patient classifi ca-
tion systems (see Chapter 4 of this volume). Second, depending on the patient’s
characteristics (that is, age, diagnoses or treatments), MEL groups may be split
into one of 427 procedure-oriented MEL-LDFs; and HDGs may be split into one
of 552 diagnosis-oriented HDG-LDFs.1
Each LDF has a specifi c score that is determined based on information about
average costs of treating patients within that LDF (see section 11.4). These LDF
scores, together with add-on scores for additional expensive procedures or for
stays in specialist departments serve as the basis for hospital budget allocation
from SHFs and the PRIKRAF (see subsection 11.5.3).
The steering area (Steuerungsbereich) of the LKF framework allows each state to
determine hospital budget allocations from its SHF according to state-specifi c
priorities. As each state uses different criteria to adjust hospital budgets, there
are 10 different ways of paying hospitals based on the LKF-PCS of the core area:
one for each of the nine SHFs, plus one in the nationwide PRIKRAF (Hofmarcher
& Riedel, 2001).
11.2.2 Development of the LKF-PCS
In the 1980s, different performance-oriented hospital payment systems were
tested with the conclusion that none of them fi tted exactly the special needs of
the Austrian health care system and that an appropriate legislative framework
for documentation standards was still lacking (BMGFJ, 2008f). It was therefore
decided to introduce documentation standards for hospitals and to develop an
Austrian patient classifi cation system from scratch.
Working in close collaboration with medical experts from various fi elds, an
interdisciplinary team of economists and statisticians developed a system that
would be tailored to the specifi c needs of the Austrian inpatient sector. The
result was a precursor version of today’s LKF-PCS. Between 1988 and 1990 the
system was tested by a sample of 20 hospitals across the country. The outcome
was that further development was needed.
Meanwhile, an obligation to document diagnoses was introduced in all
Austrian hospitals. Between 1991 and 1996 an Austrian catalogue of procedures
was developed by the Ministry of Health, which was the prerequisite for defi n-
ing procedure-oriented LDFs. Around that time, the potential effects of using
the LKF-PCS for hospital fi nancing were calculated for testing purposes. A pilot
project in Vorarlberg in 1995 and in Lower Austria in 1996 tested the potential
of using the LKF-PCS for hospital budget allocation. The pilot project showed
good results as the LKF-PCS seemed to fulfi l the aims connected with its intro-
duction (see section 11.3). After fi nal evaluation and further adjustments, the
LKF-PCS was introduced within the LKF framework for all acute care hospitals
in 1997.
Table 11.3 summarizes some main features of the Austrian LKF-PCS. Since
its introduction, the LKF-PCS has undergone annual revisions. However, the
Ministry of Health distinguishes years during which revisions are related to
‘maintenance’ (meaning that only absolutely essential corrections are imple-
mented) from those during which ‘amendments’ are implemented that imply
Austria: Inpatient care and the LKF framework 181
far-reaching further developments of the system (BMG, 2010). Therefore, the
table shows only the three LDF versions resulting from major amendments.
One of the main aspects of annual maintenance work is the introduction of
new procedures to keep the system up to date (see section 11.6). In addition, in
2001 the minimum basic data set (MBDS) was extended and the coding of
diagnoses was changed to ICD-10-BMSG-2001 (see Table 11.4 in subsection
11.3.1), an Austrian modifi cation of the ICD 10th revision (ICD-10) that was
developed together with the German Institute of Medical Documentation and
Information (Deutsches Institut für medizinische Dokumentation und Information
in German; DIMDI).
In 2002, the fi rst major amendment took place. Based on data from 15
reference hospitals, LDF scores, ALOS and trim-points were recalculated for
every LDF. LDFs were rearranged or new ones created and new procedures were
introduced. Since then, the LDF scores for each LDF consist of a day component
and a performance component. The performance component includes costs
directly connected to procedures (for example, personnel costs during surgery
and medical products) calculated by the reference hospitals’ data. The day
component includes costs that accrue during the whole hospital stay, such as
nursing and hotel costs.
Again, years of minor maintenance revisions followed and in 2006/2007 the
LDF scores for day patients were recalculated to discourage unnecessary longer
hospital stays. Extra scores for necessary stays in specialist departments and
intensive care unit (ICU) were introduced on a per diem basis.
The next major amendment was performed in 2009. As in 2002, all LDF
scores, the ALOS and trim-points were recalculated for each LDF, and LDFs were
rearranged if necessary. For the calculation of LDF scores, cost data from 20
reference hospitals from 2005 were used. In addition, new procedures were
added to the catalogue of procedures, and its structure was revised.
A detailed description of the development has been given by the BMGFJ
(2008f).
Table 11.3 LKF-PCS versions
Year 1997 2002 2009
Purpose DRG-based budget allocation, planning, performance measurement
DRG system LKF-PCS Version 1997 LKF-PCS Version 2002 LKF-PCS Version 2009
Data used for
development
Cost data from 20
reference hospitals,
activity data from all
hospitals
Cost data from 15
reference hospitals,
activity data from all
hospitals
Cost data from 20
reference hospitals,
activity data from all
hospitals
Number of LDFs 916 842 979
Applied to Public and non-
profi t-making acute
care hospitals
All acute care hospitals (including those paid
through PRIKRAF)
Included
services
All acute inpatient care (including day cases), excluding psychiatric,
rehabilitation and long-term care
Source: Authors’ own compilation based on BMGFJ 2008f.
182 Diagnosis-Related Groups in Europe
11.2.3 Data used for development and updates of the LKF-PCS
Two main databases are used for updates of the LKF-PCS. First, a hospital activity
database is maintained at the Ministry of Health, containing the aggregated
information of all MBDS from all hospitals (see Table 11.4). For each admission,
hospitals complete the MBDS, which is then sent to the SHFs or the PRIKRAF.
There, the data are integrated into a state discharge database, plausibility checks
are performed and the data are searched for errors. If necessary, hospitals are
asked to correct the data. After approval by the SHF, the data are forwarded to
the Ministry of Health.
Second, for the two revisions in 2002 and 2009, detailed resource-consumption
data were provided by 20 reference hospitals. The data from these hospitals
were merged into a resource-consumption database maintained at the Ministry
of Health that contains detailed information for the procedures listed in the
catalogue of procedures, including personnel hours spent (by type of personnel
and type of treatment), costs for medical consumables and investment costs for
large-scale medical equipment. In addition, department-level cost information
is available for the calculation of the day component of LDF scores. (For further
details, see section 11.4 and subsection 11.5.2.)
Updates of the system rely on information relating to changes in the Austrian
catalogue of procedures, which is updated annually. As part of the yearly revi-
sions of the LKF-PCS, the Austrian catalogue of procedures is reviewed for pro-
cedures that are used only rarely or no longer fulfi l the criteria of the Ministry
of Health. After consultation with the SHFs, those procedures are deleted
(BMGFJ, 2008c).
11.2.4 Regularity and method of system updates
Revisions of the LKF-PCS are carried out by the LKF team within the Ministry of
Health and are divided into two major areas: (1) recalculation of scores for each
LDF; and (2) revisions of the LKF-PCS.
LDF scores have been recalculated only twice since the introduction of the
LKF-PCS. Recalculations in 2002 relied on data from the year 1999. Recalculations
in 2009 were based on data from 2005. As already described, the scores for each
LDF comprise two components, which are calculated separately: a performance
component related to the direct resource consumption of procedures, and a day
component related to hotel costs of keeping a patient in hospital.
Revisions of the LKF-PCS rely on all the types of information described.
On a yearly basis, updates (which are almost always related to the inclusion
of new procedures) concern only a small number of LDFs, for which the ALOS,
thresholds and (if necessary) LDF scores are recalculated (see section 11.6).
This is carried out for all LDFs in years of systematic revision. ALOS and
thresholds are calculated on the basis of the hospital activity database (BMGFJ,
2008a).
Before a new version of the system comes into effect, simulation calculations
are performed to estimate the fi nancial impact for hospitals. No major system
amendments are planned before 2013.
Austria: Inpatient care and the LKF framework 183
11.3 The current patient classifi cation system
11.3.1 Information used to classify patients
The LKF-PCS classifi es patients on the basis of information provided in the
MBDS that hospitals are required to prepare for every admission. Table 11.4
shows the kind of data that are included in the MBDS.
For the grouping process, only the following data are used as classifi cation
variables: (1) procedures, (2) main diagnosis, (3) age classes, (4) secondary diag-
noses, and (5) treatment at specialist departments (that is, acute geriatric care,
remobilization, palliative departments or neuropsychiatric departments for
children and youths).
The procedure catalogue is particularly important for the LKF-PCS and was
specifi cally developed for the purpose of supporting the system. As already men-
tioned, procedures in the catalogue are called single medical procedures (MELs).
Consequently, MEL groups in the LKF-PCS simply summarize the procedure-
based groups, similar to base-DRGs in an operation-room partition of a DRG
system (see Chapter 4). In contrast to the extensive procedure catalogues used in
other countries, the Austrian catalogue of procedures contains only 1500 selected
procedures that range from surgical, through cancer treatment, to diagnostic
procedures using large-scale equipment. In addition, possible plausibility infor-
mation is provided for each procedure, such as age, gender, and day-case fl ag.
The close relationship between the procedure catalogue and the LKF-PCS is
also illustrated by the fact that the catalogue indicates a group of expensive and
highly frequent procedures that qualify individual cases to be grouped into
procedure-oriented LDFs. By contrast, secondary diagnoses and treatment at
specialist departments are very rarely used for the classifi cation of patients.
Table 11.4 Content of the MBDS
Administrative data Admission data
• hospital code
• admission code and date
• type of admission
• departments and transfers
• discharge date and type
Patient data
• date of birth
• gender
• citizenship
• principal residence
• insurance or funding body
Medical data • main diagnosis (ICD-10 BMSG 2001)
• any additional diagnoses (ICD-10 BMSG 2001)
• any medical services from the catalogue of procedures
LKF dataa • LDF
• LDF score
• score for outliers
• extra scores for ICU stays
• extra scores for multiple treatments
• scores for specialist departments
• total scores
Source: BMGFJ, 2008f.
a Only if required by the respective SHF.
184 Diagnosis-Related Groups in Europe
11.3.2 Grouping algorithm
On the basis of information contained in the MBDS (see subsection 11.3.1),
every hospital admission is grouped into exactly one LDF. The classifi cation
process follows a series of iterative steps that are illustrated in Figure 11.1.
Coding and grouping is carried out by medical doctors within hospitals using
special software provided by the Ministry of Health.
The objective of the LKF-PCS (as with any other DRG-like patient classifi cation
system) is to assign cases into medically meaningful and economically homo-
geneous groups. In order to do so, the grouping algorithm checks fi rst whether
patients were treated in specialist departments (for example, acute geriatric care
Figure 11.1 LKF grouping process
Source: Based on BMGFJ, 2008e; Grubinger et al., 2010.
Austria: Inpatient care and the LKF framework 185
or remobilization). If this was the case, the system assigns extra scores based on
the number of days spent in these departments. As a second step, the grouping
process also checks whether patients have received signifi cant procedures.
These are procedures that signifi cantly infl uence the total costs of the hospital
stay. If no such procedures have been carried out, the case is assigned into one
of the 219 HDGs on the basis of the main diagnosis. If patients received any
signifi cant procedures during the hospital stay, the algorithm checks whether
more than one such procedures was performed. If this is the case the procedure
and LDF that returns the highest LDF score is selected and add-on scores for all
other signifi cant procedures are added, based on the performance component
of their LDF. If only one procedure was performed, the case is directly assigned
into one of the 204 MEL groups.
Each of the 219 HDG groups and of the 204 MEL groups is characterized by a
four-digit number code, which follows a prefi x (either ‘HDG’ or ‘MEL’). As a
result of the grouping process, HDG groups comprise cases with similar
diagnoses, and MEL groups pool procedures that are medically similar and have
similar resource-consumption levels. MEL and HDG groups are similar to base-
DRGs or classes in other DRG systems. In order to increase economical homo-
geneity of the fi nal LDFs, MEL and HDG groups are either split into several
severity levels according to age classes, principal diagnoses, procedures and
secondary diagnoses, or they remain as one group (unsplit). If groups remain
unsplit, the letter A is added to the four-digit MEL or HDG code and defi nes the
fi nal LDF. If MEL or HDG groups are split, additional letter codes are assigned
(B, C, D). However, the LDF codes are not ordered by resource consumption.
The system does not limit the number of splits per MEL or HDG group but
subdivides them into as many LDFs as necessary in order to achieve relative
homogeneity of resource consumption within each group. In total, there are
979 LDFs in the 2009 LKF-PCS: 427 MEL-LDFs and 552 HDG-LDFs.
After assignment of the fi nal LDF, the system checks whether patients were
treated in an ICU and assigns additional per diem points per day of ICU treat-
ment. Four types of ICU exist, with differing per diem scores: intensive monitor-
ing units, and three stages of ICU. Hospitals need approval by the SHF or the
PRIKRAF in order to be able to provide ICU treatment. If a patient stays in an
ICU, the patient’s status has to be documented every 24 hours according to
standardized reporting schemes, for example TISS-28, SAPS and TRISS (BMGFJ,
2008a; BMGFJ, 2008d). A minimum score is required to justify ICU stays.
Although the assignment of per diem-based add-on scores is not – strictly
speaking – the result of the grouping process, these scores are determined
during the grouping process and are an integral part of the Austrian LKF-PCS
(BMGFJ, 2008e). Since the LKF-PCS does not use the concept of major diagnostic
categories (MDCs) used in other DRG systems, the main diagnosis of procedure-
oriented groups is often not checked during the grouping process.
11.3.3 Data quality and plausibility checks
The grouping software, which is available free of charge from the Ministry of
Health includes a set of data quality and plausibility checks. The most important
plausibility and data quality rules are:
186 Diagnosis-Related Groups in Europe
• plausibility of diagnoses and age and gender
• plausibility of procedure and age and gender
• for each procedure, there must be at least one diagnosis
• whether a certain procedure is allowed/possible in a certain hospital.
In each SHF and in the PRIKRAF a data quality group is responsible for
carrying out data quality controls at hospitals. Usually, random samples from
the hospital activity database are taken and the structured documentation is
compared with the patient history documented at the hospital. Questionnaires
are used to identify the main problems, such as incorrect main diagnosis, wrong
fourth digit of the ICD-10 codes assigned, missing or too many procedures or
secondary diagnoses, and so on. Some states apply special algorithms to identify
suspicious datasets (Pfeiffer, 2002a). This also allows a data quality profi ling of
hospitals.
11.3.4 Incentives for up- or wrong-coding
Up-coding is not an issue in Austria, probably because severity levels are rarely
based on secondary diagnoses. They are most frequently defi ned on the basis of
relatively objective criteria, such as age, or specifi c procedures. Since the intro-
duction of the LKF framework, some companies have offered so-called ‘optim-
ization’ software. However, software has also been developed by the Ministry of
Health that allows a certain level of control, if systematic optimization has been
used.
In principle, sanctions are possible, if up-coding is found to be an issue.
However, thus far hospitals have never been sanctioned, even where misuse has
been detected.
11.4 Cost accounting within hospitals
In general, cost accounting within hospitals is not part of the LKF framework
and it is not even regulated. Hospitals can therefore implement cost-accounting
systems suited to their own needs. However, hospitals fi nanced by SHFs report
highly aggregated and standardized data to the SHF. These data include, for
example, total costs for consumables (medical and non-medical), energy, fees
and administration; the number of full-time equivalents by type of personnel
and the respective total costs (reported at department level).
Hospitals that act as reference hospitals for the recalculation of the LDF scores
provide data on average resource consumption for procedures. For each
procedure, average working time by type of personnel is reported. A distinction
is drawn between times for preparation, anaesthesia and actual treatment or
surgery. Average costs are reported for ‘expensive’ medical consumables, such as
blood products, implants, prostheses and operation linen. For non-surgical
procedures, usage times for large-scale equipment and its costs are reported.
These costs include acquisition, depreciation, interest and maintenance
(BMGFJ, 2008a). In addition, total costs at the departmental level are reported
Austria: Inpatient care and the LKF framework 187
(including the allocated share of overheads), which are used for calculation of
the day component.
11.5 Hospital fi nancing on the basis of the LKF-PCS
11.5.1 Range of services and costs paid through
the LKF framework
The LKF framework serves as the reimbursement framework for all hospital
stays covered by the SHI in all acute care hospitals (public, private non-profi t-
making and private profi t-making hospitals), including day care and stays in
specialist departments, such as acute geriatric care, rehabilitation, palliative
departments or neuropsychiatric departments for children and youths. How-
ever, payment based on the LKF-PCS does not apply to stays in specialist
departments which are fi nanced on a per diem basis. In addition, the SHFs can
fi nance investments as part of their steering activity from within their own
budgets. Private profi t-making hospitals are fi nanced partly by the LKF-PCS.
The PRIKRAF pays directly for those treatments that are listed in the catalogue
of procedures covered by the SHI.
Besides fi nancing by the SHFs or PRIKRAF, hospitals also receive payments
from various other sources. Treatment of private patients is paid for by private
health insurance or out-of-pocket payments by the patients themselves. The
federal Government pays a lump sum to university hospitals to cover additional
expenses for teaching and research.
11.5.2 Calculation of the LDF scores
The LDF scores are calculated based on the hospital activity data from all
hospitals and the resource-consumption data for procedures from the reference
hospitals (see subsections 11.2.3 and 11.2.4). As already described, the score of
each LDF contains two components: (1) a performance component that
includes all resource consumption directly connected to procedures; and (2) a
day component which comprises the sum of all remaining costs accruing
during the hospital stay.
Performance component
Based on the data for procedures provided by the reference hospitals, the
average procedure-related costs for each LDF are determined.
For each procedure the reference hospitals provide the average cost, which is
the sum of four categories: (1) personnel costs, which are calculated as working
time multiplied by average salary according to type of personnel; (2) costs of
expensive consumables, established by multiplying the quantity by the price;
(3) large-scale equipment costs; and (4) procedure-related overheads.
188 Diagnosis-Related Groups in Europe
Day component
At department level all costs that cannot be allocated to one of the procedures are
divided by the total number of bed days, which provides the adjusted costs of a
bed day. The day component of an LDF score is then calculated as the average
sum of the adjusted costs for each bed day, assuming the ALOS of this LDF.
The ALOS in MEL-LDFs is calculated as a 10 per cent trimmed mean, and as a
20 per cent trimmed mean in HDG-LDFs. In MEL-LDFs, ‘outliers’ are defi ned as
patients staying longer than the minimum of 1.5*ALOS and the 90th percentile
of the length of stay or staying less time than the maximum of 0.3*ALOS and
the 10th percentile of the length of stay. In HDG-LDFs, outliers are defi ned by
a slightly different method. Long stays are those longer than the minimum of
1.5*ALOS and the 80th percentile, while short stays are those shorter than the
maximum of 0.5*ALOS and the 20th percentile (BMGFJ, 2008e).
Adjustments for outliers
The additional daily score for long-stay outliers is reduced for each following
outlier day, but remains stable at the minimum of half of the daily day
component. The calculation is carried out as follows (BMGFJ, 2008e):
DC � ,
t
x
DC⎧
⎨
⎩
⎫
⎬
⎭2
Score (x) = max
x = number of hospital days (and has to be above the trim-point)
Score(x) = extra points for day x
DC = day component per day
t = trim-point = bound for long-stay outliers
The LDF score of short-term outliers contains the full performance component,
whereas the day component is reduced (BMGFJ, 2008e).
Score = PC +
(LDF score – PC)* (x + 1)
t + 1
x = number of hospital days (and has to be below the trim-point)
Score = reduced LDF score
LDF score = score of the LDF-group
PC = performance component
t = trim-point = bound for short-stay outliers
For 0-day stays, approved treatments in day hospitals are reimbursed in the
same way as 1-day stays. 0-day stays with non-approved treatments receive the
full performance component but only 10 per cent of the reduced day component
calculated for short-stay outliers.
11.5.3 LDFs in actual hospital payment
SHFs allocate the majority of hospital budgets on the basis of LDF scores. As the
general rules that apply for determining the LDF score are the same nationwide,
these are part of the LKF core area. However, how SHFs make use of this
information in order to determine hospital budgets depends on state-specifi c
priorities and is defi ned in the LKF steering area, which is specifi c to each state.
Austria: Inpatient care and the LKF framework 189
LKF core area (Kernbereich)
The LKF core area defi nes how the LDF score per patient is determined on
the basis of the LKF-PCS. Figure 11.2 shows that the total score consists of two
main components: (1) the LDF-specifi c part; and (2) add-on scores for certain
services.
Every hospital discharge is assigned to an LDF on the basis of its diagnoses,
procedures, and so on (see section 11.3). Each LDF has a specifi c score which is
calculated based on average costs of treatment of patients within that LDF (see
subsection 11.5.2). In addition, the LDF score is adjusted for outlier patients,
that is, patients with exceptionally long or short lengths of stay. For long-stay
outliers, the day component is increased for every day that patients stayed
beyond the upper length-of-stay threshold. For short-stay outliers, the day com-
ponent of the LDF score is reduced for every day that patients were discharged
below the lower length-of-stay threshold. For day cases (for which length of
stay is 0), the reduction depends on whether day-care treatment is explicitly
allowed or not.
Add-on scores are assigned in two ways. Per diem-based add-on scores
are given for every day a patient spent in specialist departments (such as acute
geriatric care, remobilization, palliative departments or neuropsychiatric
departments for children and youths) or in an ICU. Additional procedure scores
are added if patients received more than one signifi cant procedure during
a hospital stay, as the performance component of an LDF for patients with
multiple procedures refl ects only the resource consumption of the most com-
plex procedure. Possible reductions apply for multiple treatments on the same
day.
As the basis for hospital payment, the sum of all LDF scores of each hospital
in a given state is calculated. In addition, the total sum of all LDF scores in a
state is calculated and serves as a reference point from which to determine the
share of total score provided by a specifi c hospital. A comparable procedure is
used to determine hospital budget allocations to private profi t-making hospitals
from the PRIKRAF. The main difference is that only one nationwide global
budget exists (fi nanced by SHIs) that is distributed to all private profi t-making
hospitals.
Figure 11.2 Calculation of total LDF score per patient
Source: Based on BMGFJ, 2008e.
190 Diagnosis-Related Groups in Europe
LKF steering area (Steuerungsbereich)
In the steering area of the LKF framework, states have the possibility to
determine hospital budgets for public and private non-profi t-making hospitals
funded by the SHF according to provincial criteria. Four main approaches can
be identifi ed.2
1. Upper Austria allocates the entire hospital budget on the basis of LDF scores;
that is, the SHF budget is distributed to the hospitals according to their share
of LDF scores. This is also the case for 98 per cent of the budget in Lower
Austria. Only 2 per cent of the SHF budget is allocated according to the hos-
pital type – that is, block grants are made to hospitals depending on struc-
tural characteristics of those hospitals, for example according to teaching
status or size.
2. In three states, fi xed rates are used between the core area and the steering
area. Tyrol and Burgenland allocate 70 per cent according to LDF scores. The
remaining 30 per cent of the budget is allocated after weighting the LDF
scores, depending on the hospital type (for example, LDF scores produced by
university hospitals are infl ated) or depending on the specifi c hospital. In
Vorarlberg, 15 per cent of the budget is allocated by adjusting LDF scores for
higher personnel costs in certain regional areas.
3. Carinthia, Styria and Vienna do not use fi xed rates between the core area and
the steering area, but weight LDF scores by certain criteria. While Vienna
combines personnel costs and additional costs as a factor, Carinthia and
Styria weight hospital types by different factors. This means that the LDF
scores of each hospital are weighted by a factor and the budget is allocated
based on the weighted shares.
4. In Salzburg, fi nancing is divided into two tiers. In the fi rst, 75 per cent of the
budget is allocated without weighting and 25 per cent is weighted by a
hospital-related factor. A total of 40 per cent of the second tier is based on
past defi cits and 60 per cent on weighted LDF scores.
11.5.4 Quality-related adjustments
Hospital budget allocations under the LKF framework are not adjusted for
quality. However, the Austrian Structural Plan for Health (ÖBIG, 2008) requires
minimum standards for certain treatments fi nanced through the LKF framework.
For example, there are specifi cations relating to hospital size, availability of
infrastructure and personnel, and minimum volume thresholds.
11.5.5 Main incentives for hospitals
The budgets of the SHFs are distributed mainly according to the amount of LDF
scores that hospitals produce in a given year. Therefore, hospitals may try to
increase their share of the budget by producing more LDF scores, for example
by treating more patients, especially day cases. However, as the LKF hospital
budget allocation system operates within a fi xed global budget, increased
Austria: Inpatient care and the LKF framework 191
production of LDF scores by one hospital reduces the value of LDF scores and
thus the budget available for other hospitals.
Several states were forced to take up counter-measures in order to avoid an
uncontrolled growth of hospital activity by defi ning score budgets for each
hospital. If a hospital produces more than the permitted amount of LDF scores,
the value of these LDF scores is reduced.
11.6 New/innovative technologies
The inclusion of new technologies into the Austrian catalogue of procedures
and the LKF framework follows several predefi ned steps.
Hospital departments that want to use a specifi c technology must prepare a
request containing a detailed description of the technology and a calculation of
costs. Subsequently, the request is submitted by the hospitals and their owners
to the respective SHF using a standardized form, which is now based on an
Internet platform maintained by the Ministry of Health.3
SHFs evaluate the requests and forward them to the LKF team at the Ministry
of Health, where medical and economic aspects are assessed. In the event of a
positive evaluation, new technologies are preliminarily included in the cata-
logue of procedures, nationwide, for two years. However, utilization is often
restricted to a limited number of hospitals based on certain structural criteria,
such as number of beds or hospital types.
In order to assure adequate reimbursement of hospitals, the LKF-PCS is
modifi ed by the LKF team. In most cases, new procedures can be assigned to
existing MEL groups comprising similar procedures, and the cost of a new
procedure is assumed to be similar to the existing ones. If no similar procedures
exist, a new LDF can be created and the performance component of the LDF
score is estimated based on the cost calculation included in the request for
inclusion in the catalogue. Consequently, new technologies are fi nanced
through the LKF framework in the same way as any existing procedure, and
there are no separate or supplementary payments for new technologies.
After the fi rst and the second years the new technology is evaluated based on
medical evidence. For the ultimate inclusion in the catalogue, the respective
LDF scores are calculated based on the collected data. If necessary, a new LDF is
created (BMGFJ, 2008c).
11.7 Evaluation of the LKF framework in Austria
Ten years after the introduction of the LKF-PCS as the basis for the hospital
budget allocation system, an evaluation process was initiated by the Ministry of
Health. As part of this evaluation, a group of international experts was
contracted to assess the status quo and to propose future developments. Their
fi ndings are programmed to have a great impact on the next Article 15a treaties
between the states and the federal Government. However, unfortunately, the
results have not yet been made public.
192 Diagnosis-Related Groups in Europe
11.7.1 Offi cial evaluations
Some research has assessed whether the switch from per diem-based fi nancing
to hospital budget allocation on the basis of the LKF-PCS has had an effect on
ALOS and hospitalization rates or not (Frick et al., 2001). Other research has
focused on the impact that the introduction of the LKF framework has had on
health care as a whole (Theurl & Winner, 2007). All authors found that the
introduction of the LKF framework had a decreasing effect on the overall ALOS,
on top of the long-term trend of declining lengths of stay (Frick et al., 2001,
Theurl & Winner, 2007). However, when looking at separate medical disciplines,
Frick and colleagues (Frick et al., 2001) showed that only 3 out of 21 disciplines
had displayed signifi cant reductions in lengths of stay. Furthermore, they found
that in 8 disciplines, hospitalization rates had increased signifi cantly.
Theurl & Winner (2007) came to similar conclusions and showed that 8 out
of 20 diagnostic groups (according to ICD-10) had signifi cant declines in
ALOS. The authors concluded, ‘Our evidence suggests that the Austrian hospital
sector has gained a substantial increase in effi ciency through the reform of
the fi nancing system. This conclusion is also confi rmed by the fact that the
annual increase of hospital costs declined after the implementation of the
LKF1997’ (Theurl & Winner, 2007).
In addition, the authors highlight that the introduction of the LKF-PCS could
only have a limited infl uence on shifting inpatient care to ambulatory care,
because organizational structures and fi nancing of different health care sectors
are highly segmented in Austria. SHI funds do not pay the full costs of inpatient
care, contributing only fl at fees per member to the SHFs. Consequently, they
are not particularly interested in shifts towards outpatient care for which they
would be required to cover a higher percentage of the total costs.
Pfeiffer (2002b) has observed that the amount of day-care treatment in
hospitals is increasing every year. However, the majority of these cases do not
replace inpatient admissions. Instead, they represent a shift from the outpatient
to the inpatient sector (Pfeiffer, 2002b) – the opposite of what was intended
when the LKF framework was introduced (see subsection 11.1.3). Pfeiffer
(2002b) explains this trend with reference to the strict separation of health care
in Austria. Beyond that, he fi nds that there is a missing link in health care
provision, highlighted by the example of clinics located between inpatient care
and care provided by private physicians.
11.7.2 Authors’ assessment
The introduction of the LKF framework was an important improvement in hos-
pital fi nancing in Austria, as hospital budget allocations under the LKF frame-
work are more closely related to hospital activity than under the old per diem
system. In addition, hospitals now report detailed activity data, contributing to
increased transparency in the hospital sector and improving documentation
quality.
Now, 13 years after the introduction of the LKF framework, it still serves its
purpose(s). However, reforms are needed that require a consensus by all relevant
Austria: Inpatient care and the LKF framework 193
stakeholders, including the federal Government and all nine states, which is
almost impossible to reach. Financial issues are particularly diffi cult to solve.
For example, it remains unclear how SHFs are to be compensated for fi nancing
the treatment of patients from other states, or how ‘fairness’ of budget alloca-
tions from the federal Government to the SHFs can be improved.
Other countries, such as France or England have recently updated their
patient classifi cation systems, especially in terms of their systematic assessment
of secondary diagnoses for the defi nition of severity levels. In Austria, secondary
diagnoses play a very minor role in defi ning severity levels. Aside from changes
in the catalogue of procedures, the classifi cation process as such has not
changed.
Although transparency and activity documentation have been improved,
research into the hospital sector is rarely carried out. Unfortunately, detailed
information – such as hospital activity data or resource-consumption data – is
only available to a limited number of people at the federal Government, the
states and a group of contracted experts at the Ministry of Health (the LKF team).
For researchers outside this circle of people it is very diffi cult, if not impossible,
to obtain hospital activity data. Yet, such research could help to improve health
care in Austria.
11.8 Outlook: Future developments and reform
In Austria there is a clear separation between inpatient and outpatient care. In
order to improve the continuity of care and to avoid unnecessary hospital
admissions, it is necessary to build interfaces. However, there are many obstacles
to this, such as the different fi nancing systems for inpatient and outpatient care
and varying interests on the part of stakeholders.
Concerning technical issues of the LKF framework and the LKF-PCS grouping
algorithm, further extensions and specifi cations of the catalogue of procedures
are necessary, along with updates to diagnosis coding. However, after the major
maintenance of the LKF framework in 2009, the system should remain rela-
tively unchanged for the next few years. Future development will include the
implementation of plausibility checks in the provided grouping software and
specifi c access criteria for ICUs. Furthermore, how to improve the severity clas-
sifi cation will be discussed, along with how to defi ne LDFs that extend beyond
individual hospital admissions and include transfers or readmissions of patients.
Currently, the development of a procedure classifi cation system for the out-
patient sector is coming to an end, and pilot tests have been initiated (BMG,
2009). This is important, as an outpatient procedure classifi cation is a prerequi-
site for any attempts to extend the LKF framework to the outpatient sector.
11.9 Notes
1 Figures correct for the 2009 LKF-PCS.
2 Further details can be found in Hofmarcher and Riedel (2001) or in Hofmarcher and
Rack (2006).
194 Diagnosis-Related Groups in Europe
3 More information can be obtained at the Ministry of Health web site (http://mel.lkf.
bmgf.gv.at, accessed 26 June 2011).
11.10 References
BMG (2009). Dokumentation im ambulanten Bereich – Bericht zur Entwicklung des
Leistungskatalogs 2010. Vienna: Bundesministerium fü r Gesundheit.
BMG (2010). The Austrian DRG system. Vienna: Bundesministerium für Gesundheit.
BMGFJ (2008a). Handbuch zur Dokumentation in landesgesundheitsfonds- fi nanzierten
Krankenanstalten (Anhang 1) 2004+. Vienna: Bundesministerium fü r Gesundheit,
Familie und Jugend.
BMGFJ (2008b). Krankenanstalten in Ö sterreich. Vienna: Bundesministerium fü r
Gesundheit, Familie und Jugend.
BMGFJ (2008c). Leistungsorientierte Krankenanstaltenfi nanzierung – LKF – Änderungen und
Neuerungen im Modell 2009. Vienna: Bundesministerium fü r Gesundheit, Familie und
Jugend.
BMGFJ (2008d). Leistungsorientierte Krankenanstaltenfi nanzierung – LKF – Medizinische
Dokumentation inklusive 17. Vienna: Rundschreiben. Bundesministerium fü r
Gesundheit, Familie und Jugend.
BMGFJ (2008e). Leistungsorientierte Krankenanstaltenfi nanzierung – LKF – Modell 2009.
Vienna: Bundesministerium fü r Gesundheit, Familie und Jugend.
BMGFJ (2008f). Leistungsorientierte Krankenanstaltenfi nanzierung – LKF – Systembeschreibung
2009. Vienna: Bundesministerium fü r Gesundheit, Familie und Jugend.
Frick, U., Barta, W., Zwisler, R., Filipp, G. (2001). Auswirkungen der leistungsorientierten
Krankenhausfi nanzierung (LKF) auf die Verweildauern und Hospitalisierungen im
Land Salzburg seit 1997. Gesundh ökon Qual manag, 6(04):95–104.
Grossmann, B., Hauth, E. (2007). Verwaltungs- und Pensionsreformen im öffentlichen Dienst
sowie Finanzierung des Krankenanstaltenwesens. Vienna: MANZ’sche Verlags- und
Universitätsbuchhandlung.
Grubinger, T., Kobel, C., Pfeiffer, K.P. (2010). Regression tree construction by bootstrap:
model search for DRG systems applied to Austrian health data. BMC Medical
Informatics and Decision-Making, 10:9.
Hofmarcher, M.M., Rack, H-M. (2006). Austria: Health system review. Health Systems in
Transition, 8(3):1–247.
Hofmarcher, M.M., Riedel, M. (2001). Gesundheitsausgaben in der EU: Ohne Privat kein
Staat, Schwerpunktthema. Das ö sterreichische Krankenanstaltenwesen – eines oder
neun Systeme? Health System Watch, 1/Frühjahr:1–24.
Ö BIG (2008). Ö sterreichischer Strukturplan Gesundheit 2008. Vienna: Bundesministerium
fü r Gesundheit.
OECD (2010). OECD Health Data 2010: Statistics and Indicators. Paris: Organisation for
Economic Co-operation and Development (http://www.oecd.org/document/30/
0,3343,en_2649_34631_12968734_1_1_1_1,00.html, accessed 20 August 2010).
Pfeiffer, K.P. (2002a). Documentation, data quality and continuous observation of the hospital
sector. 18th PCS/E Conference. Innsbruck, Austria. 2–5 October 2002:398–406.
Pfeiffer, K.P. (2002b). Fünf Jahre Erfahrung mit der Leistungsorientierten
Krankenanstaltenfi nanzierung (LKF) in Österreich, in M. Arnold, J. Klauber, H.
Schellschmidt. Krankenhaus-Report 2001. Stuttgart: Schattauer Verlag.
Statistics Austria (2010). Health expenditure in Austria. Vienna: Statistik Austria (http://
www.statistik.gv.at/web_en/static/health_expenditure_in_austria_according_to_the_
system_of_health_accounts_1_027971.xlsx, accessed 20 December 2010).
Austria: Inpatient care and the LKF framework 195
Theurl, E., Winner, H. (2007). The impact of hospital fi nancing on the length of stay:
evidence from Austria. Health Policy, 82(3):375–89.
Thomson, S., Foubister, T., Mossialos, E. (2009). Financing health care in the European Union.
Challenges and policy responses. Copenhagen: WHO Regional Offi ce for Europe on
behalf of the European Observatory on Health Systems and Policies.
chapter t w e l v e
England: The Healthcare
Resource Group system
Anne Mason, Padraic Ward and
Andrew Street1
12.1 Hospital services and the role of DRGs
12.1.1 The English health care system
The United Kingdom spends about 8 per cent of its gross domestic product
(GDP) on health, and 87 per cent of expenditure comes from the public sector
(Hawe, 2009). The National Health Service (NHS) is funded by general taxation
(80.3 per cent), national insurance contributions (18.4 per cent) and patients’
out-of-pocket payments for prescriptions, dental and optometry services (1.3
per cent). Private expenditure constitutes about 13 per cent of total health care
expenditure, which is lower than the average (23 per cent) for the countries
belonging to the European Union prior to May 2004 (EU15) (Hawe, 2009).
In England, the Department of Health has overall responsibility for the NHS
and is under the direction of a politician – the Secretary of State for Health. In
2010, 10 Strategic Health Authorities managed the local NHS, overseeing
provision, capacity and quality on behalf of the Secretary of State. Special
health authorities, such as the National Blood Authority and the National
Institute for Health and Clinical Excellence (NICE), provided national services
for the English NHS.
Most NHS services are delivered by public providers. In the primary care
sector, general practitioners (GPs) typically work in group practices. Although
their income originates from public funds, GPs are effectively self-employed
and their practices employ nurses, health visitors and administrative staff. The
GP contract was revised in 2004 and pays GPs for the provision of basic services,
as well as rewarding GP practices for the achievement of specifi c ‘quality’
targets. Some practices also act as purchasers (‘Practice-based Commissioners’).
198 Diagnosis-Related Groups in Europe
In secondary care, hospitals are grouped into legal bodies known as NHS
Trusts. These are mostly acute trusts (168) but there are also 73 mental health
trusts. These trusts cover around 1600 hospitals and specialist centres.2 There
are also 10 ‘care trusts’ that provide health and social care, and 12 ambulance
trusts, which provide emergency and non-emergency patient transport.
12.1.2 Hospital services in England
The number of NHS hospital beds by provider type available in England in
2008/2009 is summarized in Table 12.1. Statistics on the number of beds
available in the private sector are not available at a national level.
Acute trusts provide elective and non-elective care, surgical and diagnostic
procedures, Accident & Emergency (A&E) services, and some maternity services.
Inpatient and outpatient psychiatric services are mostly provided by mental
health trusts. Some NHS patients are treated in Independent Sector Treatment
Centres, although this amounts to less than 1.5 per cent of elective care patients
(Mason et al., 2009).
In 2010, most NHS hospital care was purchased by 152 Primary Care Trusts
(PCTs), each of which covered populations of between 300 000 and 350 000
individuals. PCTs can purchase elective services from any hospital or treatment
centre in England, including private providers. For services covered by the pros-
pective payment system, known as ‘Payment by Results’ (PbR), PCTs pay hospi-
tals a fi xed price (tariff) for each patient treated. For services not covered by PbR,
such as mental health care and high-cost pharmaceuticals, volume-based con-
tracts are agreed between PCTs and hospitals and prices are negotiated locally.
Information on English NHS hospital services originates from two main
databases; in both, data relate to a fi nancial year which ends on the last day of
the month of March. First, the Hospital Episode Statistics (HES) database
comprises activity data, including individual patient records for all inpatient
Table 12.1 Average daily number of available and occupied beds, 2008/2009
Sector Available beds
(share (%))
Occupied beds
(share (%))
Occupancy (%)
All ward types 160 254 136 860 85.4
General and acute (acute + geriatric) 122 538 106 142 86.6
– Acute 101 520 86 779 85.6
– Geriatric 21 018 19 363 92.1
Mental illness 26 448 22 793 86.2
Learning disabilities 2 882 2 393 83.0
Maternity 8 386 5 532 66.0
NHS organizations
Acute trusts 121 448 (76) 103 407 (76) 85
Mental health trusts 29 512 (18) 25 465 (19) 86
PCTs 8 737 (5.0) 7 492 (5.0) 86
Care trusts 116 (0.1) 92 (0.1) 80
Source: Department of Health, 2009a.
England: The Healthcare Resource Group system 199
admissions, outpatient appointments and A&E attendances. England is unusual
in that the recording ‘unit’ in the HES refers to the time spent under the care of
each consultant during the hospital stay. However, these records can be linked
to construct a ‘provider spell’, which corresponds to the usual measure of an
inpatient stay, defi ned as the period between admission and discharge. Second,
the NHS Reference Cost database contains provider costs for inpatient spells,
outpatient and A&E attendances, psychiatric care, and critical care amongst
other specialist services. Reference costs are used as the basis for the PbR tariff
(Street & Maynard, 2007b).
Healthcare Resource Groups (HRGs), an English version of diagnosis-related
groups (DRGs) are the unit of analysis both for activity and cost. The main focus
of this chapter is on HRGs, the classifi cation system used to describe patients
admitted to hospital. We outline their development and construction; the uses
to which they have been put; how costs are calculated for each HRG; and the use
of HRGs for reimbursement. Recent years have also seen the development of
classifi cation systems for patients treated in other hospital settings – notably
outpatient and A&E departments – and these are addressed briefl y.
12.1.3 Purpose of the DRG system
In most countries, the purpose of the DRG system has evolved from benchmark-
ing to reimbursement. The evolution has been similar in England and by the
mid-1990s HRGs were being used for three main purposes (Sanderson, 1995).
First, HRGs were used for benchmarking, providing the basis for comparative
performance assessment. The (then) National Casemix Offi ce constructed an
interactive national database that hospitals could use to assess the average
length of stay for their patients compared to the national average or compared
to a selective set of hospitals for each HRG. The database could also be used to
identify patients with excessive lengths of stay (so-called ‘outliers’) and to
produce specialty-level and hospital-level comparisons.
Second, hospitals were encouraged to use HRGs to assist with internal
resource management. HRGs were used to assess the budgetary impact of
anticipated changes in the volume and casemix of patients within specialties or
clinical directorates, as well as to monitor actual versus expected expenditure.
Third, HRGs were used to inform the contracting process. In the 1990s,
hospitals received their income via three main types of contractual arrangement.
Block contracts specifi ed payment for a fi xed volume of activity; cost-and-volume
contracts allowed for payments to be withheld (or made) if volumes fell below
(or surpassed) expectations; and cost-per-case contracts involved patient-specifi c
payments. Originally, contracts distinguished patients according to the specialty
in which they were treated but, from 1994 onward, increasingly more contracts
were specifi ed using HRGs. This required hospitals to undertake HRG-level
costing, applying a standardized method of cost allocation (see section 12.4).
In 1997, the incoming Labour Government announced that they would be
developing a national schedule of ‘reference costs’ itemizing the cost of HRGs
across the NHS (NHS Executive, 1997). It was intended that, by benchmarking
costs in a standardized manner, purchasers would be able to identify and
address ineffi ciency. However, the provision of benchmarking information
200 Diagnosis-Related Groups in Europe
alone probably did not provide suffi cient incentive for hospitals to address cost
differentials (Dawson & Street, 2000). In 2002, therefore, the Government
published proposals to introduce a prospective payment system, with hospitals
receiving a fi xed national payment per patient according to the HRG to which
they are allocated (Department of Health, 2002). PbR – as these reimbursement
arrangements have been called – was introduced for a small number of HRGs in
2003, and coverage has gradually expanded to other HRGs.
12.2 Development and updates of the DRG system
12.2.1 The current DRG system at a glance
All patients admitted to hospital are classifi ed by HRGs, which are clinically
similar and resource homogeneous (Anthony, 1993). Allocation is carried out
according to which (if any) procedures are received, primary diagnosis, age and
level of complications. The current system, known as HRG4 contains about
1400 groups (in 22 ‘chapters’).
Psychiatric inpatient care is not currently covered by HRG4, but the inten-
tion is that it will be incorporated in future (Mason & Goddard, 2009). There
are also plans to extend the tariff system to a number of other types of care
(see subsection 12.8.2). Figure 12.1 provides an overview of HRG4 and
Table 12.4 details how HRG4 is used to reimburse inpatient and day-case
activity.
Patient-level outpatient activity data have been collected within the HES
since 2003/2004 (NHS Information Centre for Health and Social Care, 2009).
Attendances are classed by specialty, subdivided by fi rst or subsequent appoint-
ment, primary diagnosis, main procedures and interventions, as well as by hos-
pital provider. However, providers are not obliged to submit outpatient data or
to code procedures and diagnoses, and data quality is consequently poor (NHS
Information Centre for Health and Social Care, 2009). HRGs are used only for
outpatient procedures, with other classifi cations based on specialty. Reimburse-
ment of outpatient activity is not based on HRGs, but differentiated according
to whether attendance is a fi rst or follow-up appointment, the number of clini-
cians seen and by specialty.
Patient-level data on A&E activity were fi rst collected in 2007/2008 and cover
attendances at major A&E departments, single-specialty departments, minor
injury units and walk-in centres. However, as the submission of records is not
mandatory for all providers, the data quality is sometimes poor, as already
mentioned in the context of outpatient activity data. In 2009/10 there were
12 tariffs for A&E reimbursement that varied by investigation or procedure cost.
12.2.2 Development of the DRG system
Development of an English version of DRGs fi rst commenced in 1981 when
the Department of Health funded a research project to assess the ability of the
contemporary version of DRGs in the United States to explain variation in the
length of stay of English patients (Coles, 1993). Further research eventually led
England: The Healthcare Resource Group system 201
to the development of the United Kingdom’s own categorization system of
HRGs, launched in 1991. While DRGs were based on major diagnostic categories
(MDCs) that correspond to a single organ system, HRGs were (and remain)
more directly related to specialties. They also differ from (historical) DRGs in
using local procedure codes, developed by the Offi ce of Population Censuses
and Surveys (OPCS), in addition to the International Classifi cation of Diseases
(ICD) codes for diagnoses.
As shown in Table 12.2, the fi rst version comprised 534 categories (including
12 undefi ned categories) but did not cover all acute activity, lacking groups for
psychiatry, radiotherapy and oncology (Anthony, 1993). HRG version 2 was
released in 1994, with a reduction in the number of categories to 533, including
6 undefi ned (‘U’) groups, but also including psychiatric HRGs. Further refi ne-
ments led to the release of HRG3.1 in 1997, comprising 572 groups and
including chemotherapy (Benton et al., 1998). This version remained in use for
a number of years, becoming the basis for the reporting and benchmarking of
hospital ‘reference’ cost data (Street & Dawson, 2002). A less-dramatic revision
appeared with the release of HRG3.5 in 2003, expanding the number of groups
to 610. It was this version that was in place when the Government started to
use HRGs explicitly for reimbursement purposes, with the phased introduction
of PbR (Department of Health, 2002), which commenced in 2003/2004.
The HRG4 design represents a major development from HRG3.5, and uses
ICD-10 (10th revision) diagnoses and OPCS-4.5 procedure codes. It was fi rst used
in the 2006/2007 reference cost collection exercise and replaced HRG3.5 as the
basis for reimbursement in 2009/2010 (Information Standards Board for Health
and Social Care, 2009). HRG4 is designed to evolve year on year, so the number
of categories is not constant, containing approximately 1400 groups, only one
of which is an undefi ned category.
HRG4 differs from the previous version in various respects (NHS Information
Centre for Health and Social Care, 2008a), as detailed here.
1. HRG3.5 covered only inpatient and day-case activity, but HRG4 covers non-
admitted (outpatient) care, emergency medicine and some specialty areas
not covered by HRG3.5, such as critical care (NHS Health and Social Care,
2008a).
2. Under HRG3.5, each episode of care generated a single HRG and all elements
of treatment were subsumed under this base- (core) HRG. Under HRG4,
some (high-cost) elements of treatment are separated from the base-HRG,
generating unbundled HRGs that can be reimbursed as additions to base-
HRGs. Therefore, one patient can have several HRGs. To qualify for unbun-
dling, there must be at least 600 cases expected annually, or the total annual
cost must be at least £1.5 million.
3. HRG4 refi nes the classifi cation of complications and co-morbidities (CCs) to
better refl ect variations in severity.
4. This latest version also provides spell-based HRGs that cover the whole stay
from admission to discharge (including one or more episodes) to make
reimbursement ‘fairer’.
Table 12.2 and Figure 12.1 provide an overview of the evolution of the English
DRG system.
202 Diagnosis-Related Groups in Europe
Figure 1 2.1 Overview of previous and current DRG systems
Sources: Anthony, 2993; Sanderson et al., 1995; NHS Information Centre for Health and Social
Care, 2006a, b, 2008b.
Notes:
HRG4 CHAPTERS DEFINITIONS
A Nervous system N Obstetrics
B Eyes and periorbita P Diseases of childhood and neonates
C Mouth, head, neck and ears Q Vascular system
D Respiratory system R Diagnostic imaging and interventional
radiology
E Cardiac surgery and primary cardiac
condition
S Haematology, chemotherapy,
radiotherapy and specialist palliative
care
F Digestive system T Reserved for mental health currencies
G Hepatobiliary and pancreatic system U Undefi ned groups
H Musculoskeletal system V Multiple trauma, emergency and urgent
care, and rehabilitation
J Skin, breast and burns W Immunology, infectious diseases and
other contacts with health services
K Endocrine and metabolic system X Critical care, high-cost drugs and devices
L Urinary tract and male reproductive
system
Y Empty
M Female reproductive system and
assisted reproduction
Z Unbundled
T
a
b
le
1
2
.2
E
v
o
lu
ti
o
n
o
f
th
e
E
n
g
li
sh
D
R
G
s
y
st
em
1
st
D
R
G
v
er
si
on
2
n
d
D
R
G
v
er
si
on
3
rd
D
R
G
v
er
si
on
4
th
D
R
G
v
er
si
on
5
th
D
R
G
v
er
si
on
D
a
te
o
f
in
tr
o
d
u
c
ti
o
n
M
ay
1
9
9
2
A
u
g
u
st
1
9
9
4
Ju
n
e
1
9
9
7
O
ct
o
b
er
2
0
0
3
In
t
ra
n
si
ti
o
n
a
(M
a
in
)
P
u
rp
o
se
P
at
ie
n
t
cl
as
si
fi
ca
ti
o
n
P
at
ie
n
t
cl
as
si
fi
ca
ti
o
n
P
at
ie
n
t
cl
as
si
fi
ca
ti
o
n
P
at
ie
n
t
cl
as
si
fi
ca
ti
o
n
P
at
ie
n
t
cl
as
si
fi
ca
ti
o
n
D
R
G
s
y
st
e
m
H
R
G
1
H
R
G
2
H
R
G
3
.1
H
R
G
3
.5
H
R
G
4
D
a
ta
u
se
d
f
o
r
d
e
v
e
lo
p
m
e
n
t
A
d
ap
ta
ti
o
n
o
f
U
n
it
ed
S
ta
te
s
D
R
G
s
D
at
a
an
al
y
si
s
o
f
g
ro
u
p
in
g
s
C
li
n
ic
al
r
ev
ie
w
t
o
re
fi
n
e
fo
r
IC
D
-1
0
.
St
at
is
ti
ca
l
an
al
y
si
s
C
li
n
ic
al
W
o
rk
in
g
G
ro
u
p
s
re
fi
n
ed
c
at
eg
o
ri
es
.
St
at
is
ti
ca
l
an
al
y
si
s
E
x
p
er
t
w
o
rk
in
g
g
ro
u
p
s’
m
ic
ro
-c
o
st
in
g
d
at
a
N
u
m
b
e
r
o
f
D
R
G
s
5
3
4
5
3
3
5
7
2
6
1
0
1
3
8
9
A
p
p
li
e
d
t
o
P
u
b
li
c
h
o
sp
it
al
s
P
u
b
li
c
h
o
sp
it
al
s
P
u
b
li
c
h
o
sp
it
al
s
P
u
b
li
c
h
o
sp
it
al
s
/
p
ri
v
at
e
h
o
sp
it
al
s
o
r
tr
ea
tm
en
t
ce
n
tr
es
tr
ea
ti
n
g
N
H
S
p
at
ie
n
ts
P
u
b
li
c
h
o
sp
it
al
s
/
p
ri
v
at
e
h
o
sp
it
al
s
o
r
tr
ea
tm
en
t
ce
n
tr
es
tr
ea
ti
n
g
N
H
S
p
at
ie
n
ts
A
cu
te
a
d
m
is
si
o
n
s
A
cu
te
ad
m
is
si
o
n
s
A
cu
te
a
d
m
is
si
o
n
s
A
cu
te
a
d
m
is
si
o
n
s
A
cu
te
a
d
m
is
si
o
n
s
O
u
tp
at
ie
n
ts
a
T
h
e
v
er
si
o
n
s
h
o
w
n
i
s
fo
r
2
0
0
7
/2
0
0
8
.
204 Diagnosis-Related Groups in Europe
In evaluating alternative arrangements for the classifi cation architecture,
performance has always been judged by considering reductions in variance in
length of stay, this being the primary defi nition of ‘resource’ for grouping
purposes. This refl ects the fact that patient-level cost data are not available in
England (see section 12.4).
12.3 The current patient classifi cation system
12.3.1 Information used to classify patients
HRGs are standard groupings of clinically similar treatments that use comparable
levels of health care resources (NHS Information Centre for Health and Social
Care, 2007a). Developed under the auspices of 33 Clinical Working Groups,
HRG4 was devised by clinicians, fi nance specialists, statisticians, health econo-
mists, users, as well as the PbR (reimbursement) team and casemix experts.
HRG4 uses a fi ve-character code structure (AANNA). The fi rst two characters
represent the chapter/sub-chapter (for example, BZ = Eyes and Periorbita
Procedures and Disorders). The next two numeric characters represent the HRG
number within the chapter (for example, BZ0 6A = Oculoplastics category 2: 19
years and over). The fi nal character (BZ06A) signifi es the ‘split’ level applicable
to the episode (for example, an age split or a severity split). In general, ‘A’ codes
signify greater resource use than ‘B’ codes, which in turn signify greater resource
use than ‘C’ codes. An HRG ending with Z indicates that no splits are applied to
that HRG. Episodes that cannot be grouped because of data insuffi ciency or
data validation issues are allocated to an ‘uncoded’ HRG (for example, UZ01Z).
Of the 1390 HRGs within HRG4 (2007/2008 version), 511 are not adjusted by
age, gender or any other modifi er. Patients are classifi ed within HRGs based on
clinical data (diagnoses (ICD-10), procedures (OPCS) and severity (presence and
level of CCs)); demographic data (age, gender); and resource use (length of
stay). HRGs are not defi ned by patient weight or by disease stage.
HRG4 uses the latest procedure codes (currently OPCS-4.5) to ensure better
specifi city of grouping than OPCS-4.2/HRG3.5. The OPCS-4 classifi cation, which
was fully implemented across the NHS in 1990, is based on a statistical classifi -
cation of surgical operations fi rst introduced in the United Kingdom in 1944.
The OPCS system is updated annually to refl ect modern clinical practice (NHS
Information Centre for Health and Social Care, 2007a). HRG4 uses an improved
mapping of CCs that modifi es HRG assignment to refl ect the additional cost of
more complex cases. For many HRGs there are three splits to refl ect the scale of
complexity: ‘Without CC’, ‘Intermediate CC’ and ‘Major CC’. Where no relevant
secondary diagnoses are recorded, the activity will group to the ‘Without CC’
variant of the relevant HRG, designed to be the lowest resource use category.
12.3.2 Classifi cation algorithm
Clinical Working Groups make judgements on resource homogeneity within
HRGs, which are tested on patient-level data. The principal data source is the
HES database. The HES comprises individual patient records and all NHS trusts
England: The Healthcare Resource Group system 205
in England routinely provide HES data for every inpatient and day-case patient
they treat, so the full dataset comprises about 15 million records each year. Each
patient record includes a number of variables containing demographic data
(such as age, gender); clinical information (such as diagnosis, procedures per-
formed); type of admission (such as elective, non-elective, day case); and length
of stay, which is used as a primary measure of resource use.
A variety of statistical techniques are employed to assist in the optimal design
of groupings and to measure statistical coherence. The main analytical approach
to the design of HRGs employs Classifi cation and Regression Trees (CARTs)
(NHS Information Centre for Health and Social Care, 2007a). This is a non-
parametric analysis technique that makes no assumptions about the underlying
distribution of values of the predictor variables (length of stay). Thus, CARTs
can handle numerical data that are highly skewed. CARTs will use variables
contained in the HES, such as procedures, diagnoses, age and so on, to identify
HRGs that best differentiate between cases with long or short lengths of stay.
These CART analyses are undertaken to support the Clinical Working Groups
whenever there is a major review of the HRG system.
Patients are grouped to a single HRG on the basis of several data elements. If
these data are missing or invalid, the patient is allocated to an ‘uncoded’ HRG
(UZ01Z). The stages of the algorithm are shown in Figure 12.2 and described in
more detail in the remainder of the chapter.
Unbundling is the fi rst step in the grouping process (NHS Information Centre
for Health and Social Care, 2007b). Unbundled activity is identifi ed and removed
as separate ‘unbundled’ HRGs (see Figure 12.3). The grouper then ignores these
unbundled components when deriving the core HRG for an episode or spell.
Second, a new mechanism has been defi ned to identify high-resource,
complex treatments associated with multiple trauma sites. This dominates all
other procedure hierarchies and so follows the unbundling step shown at the
top of Figure 12.2.
The third step in the grouping process concerns procedures. In HRG4,
procedures are ranked using a hierarchy based on cost data and clinical
knowledge. When several procedures are recorded, a procedure hierarchy list is
used to decide which procedure is dominant and should be used to assign the
HRG. The procedure hierarchy used for HRG version 3.5 has been extensively
updated for HRG4 and now contains 11 bands (from 2 (lowest resource use) to
12 (highest resource use)). In addition, Band 0 identifi es procedures that are not
valid for HRG assignment (for example, site-of-operation codes) and Band 1
identifi es minimal resource-use codes for non-operative procedures (such as
injections). If procedures are planned but not carried out, patients are allocated
to a bespoke HRG (WA14).
If no procedure is recorded, HRG is assigned by the primary diagnosis. This
includes respite or convalescent stays, and mental health diagnoses treated by
providers not strictly in the mental health care sector. In the 2009 version of
HRG4, there are just three mental health HRGs that are differentiated only by
age group.
CC splits are a way of incorporating variations in severity and complexity
within HRGs. Lists of CC splits are specifi c to each HRG chapter and are
particularly important for the medical HRGs (as these are driven by primary
206 Diagnosis-Related Groups in Europe
diagnosis). However, secondary diagnoses can be considered as CCs for both
surgical and medical episodes. Depending on their explanatory power in terms
of explaining cost variation, some HRGs are also split by secondary procedures,
age, gender, length of stay, anatomical region (for digestive system diagnoses or
procedures) or approach type (for example, laparoscopic surgery).
12.3.3 Data quality and plausibility checks
HRG4 is updated annually to meet clinical and costing requirements.
Continuing formal engagement with clinicians is ensured through the Expert
Working Groups and the Clinical Advisory Panel.
Since 2006, all acute trusts in England have received an external clinical
coding audit by the Audit Commission, an independent public body responsible
for ensuring value for money in the public sector (Audit Commission, 2006).
The audit process involves comparing a random sample of patients’ case notes
with the trust’s actual coding (Audit Commission, 2010). The Audit Commission
assesses coding accuracy and adherence to national standards for coding and
data defi nitions. An online national benchmarking tool is available to PCTs and
trusts so that organizations can compare their performance and identify areas
for further investigation.
Figure 12.2 HRG4 – Classifi cation fl ow chart for inpatients
Sources: Dawson & Street, 1998; NHS Information Centre for Health and Social Care, 2007a, b,
2008a.
England: The Healthcare Resource Group system 207
12.4 Cost accounting within hospitals
12.4.1 Regulation
The NHS Costing Manual sets out the mandatory practice of costing to be
applied in NHS hospitals (Department of Health, 2009b). Introduced in 1999,
Figure 12.3 Unbundled activity: Components for unbundled HRGs
Source: Information Centre for Health and Social Care, 2007b.
Notes: ECMO: extracorporeal membrane oxygenation; ECLS: extracorporeal life support.
208 Diagnosis-Related Groups in Europe
it brings a degree of consistency to the production and collection of cost
information.
The Clinical Costing Standards Association of England (CCSAE), a working
group of costing experts was established to develop clinical costing standards
for the acute care sector, while also supporting the implementation of Patient-
Level Information and Costing Systems (PLICS) within the NHS. The implemen-
tation of PLICS is currently not mandatory.
12.4.2 Main characteristics of the cost-accounting system(s)
All NHS hospitals are required to report their activity and unit costs annually to
the Department of Health (Department of Health, 2009b). Unit costs refl ect the
full cost of provision and include all operating expenses, staff costs and capital
costs (both interest and principal), but exclude the costs of teaching and
research. Total costs are reconciled to the fi nancial costs of the provider for the
previous fi nancial year.
As data on itemized resource use by individual patients are not collected in
England, costs are estimated using a top-down approach. Figure 12.4 illustrates
the initial and Figure 12.5 the latter stages of this costing exercise (Department
of Health, 2009b). The starting point for the costing process is the general
ledger. Here, total costs or ‘high-level control totals’ are established. Costs are
calculated on a full absorption basis; that is, all costs are allocated to the services
delivered. These costs are allocated and apportioned by maximizing direct
charging and, where this is not possible, using standard methods of apportion-
ment matched to the services that generate them.
Aggregate costing fi gures are then divided into one of three cost categories –
direct (D), indirect (I) and overhead (O) costs. Direct costs are those which can
be directly attributed to the service(s) that generated them. For instance, the
type and amount of nursing staff working in a particular specialty can be
estimated with reasonable precision. Costs that cannot be attributed directly
must be apportioned by other means. Indirect and overhead costs are pooled in
order to do this. These ‘cost pools’ bring together costs into identifi able groups
(for example, wards, pharmacies, theatres) and allow them to be apportioned to
the relevant services. Each type of cost pool can be identifi ed as being fi xed,
semi-fi xed or variable. The pooling of costs allows for the calculation of units of
activity (for fi xed and semi-fi xed pools) and time (for variable pools). Within
each costing pool, key cost drivers are established. These may include length of
stay for time-based ward costs, or event-based costs, such as the number of
prostheses used.
For all services not directly attributed to patients, the high-level control totals
are analysed by setting, indicating whether the patient was treated as a day
case or as an inpatient (elective or non-elective), whether (s)he underwent an
outpatient procedure, or was treated in ‘other’ settings (Figure 12.5).3 For
inpatient and day-case activity, as well as outpatient procedures, costs are
further disaggregated into HRGs. To do this, the main HRGs used by the pro-
vider are identifi ed within each specialty. These key HRGs should cover at
least 80 per cent of cost and activity within each setting. The main conditions
England: The Healthcare Resource Group system 209
or procedures of the provider are then identifi ed within each HRG. A weighted
average cost of each HRG is then calculated by:
• multiplying each diagnosis/procedure in a given HRG by the total number of
patients for that diagnosis/procedure;
• adding up all the costs of the diagnosis/procedure;
• dividing this total cost by the total number of patients in the HRG.
For each HRG there will be a small number of cases which have an abnormally
long length of stay. An upper trim-point is calculated for each HRG: the upper
quartile of the length of stay distribution for that HRG plus 1.5 times the
interquartile range (Schreyögg et al., 2006). Instead of excluding outlier cases,
only excess bed days beyond the upper trim-point are excluded, and a cost per
excess bed day is calculated. For clarity, the process of allocating HRGs to
elective inpatient activity is illustrated.
The outcome of this cost-allocation process is a cost per HRG (i=1. . . I)
according to the treatment setting (j=1. . .5) and type of admission. The formula
for the cost per HRG in each setting (c
ij
) is
c
ij
= D
ij
+ γ
ij
I + ϕ
ij
O, i = 1. . .I, j = 1. . .5 (1)
where D
ij
indicates the direct costs attributable to the HRG, and γ
ij
and ϕ
ij
represent, respectively, the shares of indirect and overhead costs attributed to
the HRG.
Figure 12.4 English cost-accounting system: Initial stages
Source: Based on Department of Health, 2009b.
210 Diagnosis-Related Groups in Europe
12.5 DR Gs for hospital payment
12.5.1 Range of services and costs included in
DRG-based hospital payments
Phased in since 2003, almost all hospital care in England is reimbursed under
the PbR system. PbR tariffs are based on average hospital costs, and include
labour, equipment and capital costs. In 2009/2010, HRG4 replaced HRG3.5 as
the system underpinning the PbR tariff (with the exception of payment for A&E
services). Clinical activity can now be coded more specifi cally and the increased
number of HRGs means that providers can be more fairly reimbursed for the
activities they carry out. HRG4 also allows for ‘unbundling’. This means that
some services now can be priced separately (Figure 12.3).
In 2009/2010, the national PbR tariff was payable for inpatient care (involving
admission to hospital), outpatient care and A&E services, and covers almost all
hospital activity. Services not covered by PbR included primary care services,
community services, mental health services and ambulance services. A full list
of exclusions from the tariff is available from the Department of Health web site
(Department of Health, 2009c). Although there are no published tariffs for
services that are not covered under PbR, prices have been disclosed to support
and guide local negotiations (Department of Health, 2009c).
12.5.2 Calculation of DRG prices
The HRG price (tariff) is determined for the year ahead by the Department of
Health according to a standard methodology (Department of Health, 2009c).
Figure 12.5 English cost-accounting system: Latter stages
Source: Based on Department of Health, 2009b.
England: The Healthcare Resource Group system 211
Details of the tariffs for admitted patients, outpatients and A&E attendances are
summarized in Table 12.3. Prices are set based on the average of the costs calcu-
lated by all hospitals for each of their HRGs, as detailed earlier in equation 1. The
tariff for each HRG and admission type for a given year t, p
ijt
, is calculated as:
p
ijt
= π
i
c̄
ijt–3
(2)
where c̄
ij
is the average cost for each HRG by admission type across all hospitals.
There is a three-year4 delay between hospitals submitting cost data and these
data being converted into prices, hence the t – 3 subscript attached to these
average costs. To take account of this delay, an infl ationary adjustment π
i
is
made to each HRG. This adjustment is HRG-specifi c, allowing for infl ationary
impacts such as clinical guidance and technology appraisals (issued by NICE)
that may have occurred in the intervening period.
12.5.3 DRGs in actual hospital payment
Originally, a single tariff was applied to elective patients treated on a day-case
and inpatient basis, to encourage providers to move patients to cheaper day-
case settings (Street & Maynard, 2007b). From 2009/2010 the same tariff no
longer applies to inpatient and day-case care.
A single PbR tariff applies to all providers regardless of geographical location.
However, it is argued there are some costs outside the control of hospitals that
mean they face higher-than-average overall costs, irrespective of how effi cient
they are. Thus, to refl ect unavoidable cost variations in factor prices, the
Department of Health (DH) makes a payment directly to providers based on a
single index known as the Market Forces Factor (MFF). This single MFF index is
based on three sub-indices – labour, land and buildings. Labour costs for each
hospital are based on local variation in wages in the private sector for analogous
service sector jobs. The land index is calculated for each hospital in the NHS
using data from the Valuation Offi ce on the NHS estate in 2004, and the
building index is based on a rolling average of tender prices for all public and
private contracts (Miraldo et al., 2008).
The MFF is adjusted periodically by the Department of Health in order to
ensure it relates to current, unavoidable cost variations. In 2009, following a
review of the staff component of the MFF by the Advisory Committee on
Resource Allocation (ACRA), there were changes to how the MFF index was
calculated and how it was paid (Department of Health, 2009c). PCTs now
pay the MFF payment to providers at the same time as activity payments,
whereas the MFF was previously paid directly by the Department of Health. To
smooth the impact of this change on provider income, the new index was
capped at plus or minus 2 per cent (Department of Health, 2009c).
The MFF can be represented as a hospital-specifi c adjustment to the tariff, so
that, in effect, the price paid per HRG is unique to each hospital k, with:
p
ijkt
= δ
kt
p
ijt (3)
where δ
kt
is the MFF adjustment applying to hospital k at time t.
T
a
b
le
1
2
.3
H
R
G
p
ri
ce
s,
2
0
0
9
/2
0
1
0
A
d
m
it
te
d
p
a
ti
en
ts
O
u
tp
a
ti
en
ts
A
&
E
C
u
rr
e
n
c
y
H
R
G
s
p
el
l
A
tt
en
d
an
ce
b
y
s
p
ec
ia
lt
y
(
fo
r
p
ro
ce
d
u
re
s:
H
R
G
s)
A
tt
en
d
an
ce
S
tr
u
c
tu
re
T
ar
if
fs
f
o
r:
•
el
ec
ti
v
es
•
n
o
n
-e
le
ct
iv
es
•
p
la
n
n
ed
s
am
e-
d
ay
a
ct
iv
it
y
(
d
ay
c
as
es
o
n
ly
i
n
2
0
0
9
/2
0
1
0
)
•
sh
o
rt
-s
ta
y
e
le
ct
iv
e
•
sh
o
rt
-s
ta
y
e
m
er
g
en
ci
es
T
ar
if
fs
f
o
r:
•
fi
rs
t
at
te
n
d
an
ce
•
fo
ll
o
w
-u
p
a
tt
en
d
an
ce
•
m
u
lt
i-
p
ro
fe
ss
io
n
al
a
s
w
el
l
as
s
in
g
le
p
ro
fe
ss
io
n
al
a
p
p
o
in
tm
en
ts
,
fo
r
tr
ea
tm
en
t
fu
n
ct
io
n
c
o
d
es
w
h
er
e
d
at
a
ar
e
av
ai
la
b
le
•
P
ro
ce
d
u
re
s
ca
rr
ie
d
o
u
t
in
o
u
tp
at
ie
n
t
se
tt
in
g
s
u
b
je
ct
t
o
n
o
n
-m
an
d
at
o
ry
t
ar
if
f
b
as
ed
o
n
H
R
G
s,
w
it
h
t
h
e
in
te
n
ti
o
n
t
h
at
th
is
a
ct
iv
it
y
i
s
co
v
er
ed
b
y
t
h
e
m
an
d
at
o
ry
‘P
la
n
n
ed
S
am
e
D
ay
’
ta
ri
ff
i
n
f
u
tu
re
y
ea
rs
•
N
o
n
-m
an
d
at
o
ry
t
ar
if
f
fo
r
o
u
tp
at
ie
n
t
ap
p
o
in
tm
en
ts
n
o
t
ca
rr
ie
d
o
u
t
fa
ce
t
o
f
ac
e
T
ar
if
fs
f
o
r:
•
H
ig
h
-c
o
st
a
tt
en
d
an
ce
•
St
an
d
ar
d
a
tt
en
d
an
ce
•
C
o
m
b
in
ed
M
in
o
r
A
&
E
/M
in
o
r
In
ju
ri
es
U
n
it
a
tt
en
d
an
ce
S
p
e
c
ia
li
z
e
d
se
rv
ic
e
a
d
ju
st
m
e
n
ts
•
T
o
p
-u
p
p
ay
m
en
t
fo
r
sp
ec
ia
li
ze
d
s
er
v
ic
es
f
o
r
ch
il
d
re
n
a
n
d
o
rt
h
o
p
ae
d
ic
a
ct
iv
it
y
•
E
x
cl
u
si
o
n
s
•
E
x
cl
u
si
o
n
s
•
N
o
t
ap
p
li
ca
b
le
O
u
tl
ie
rs
•
L
o
n
g
-s
ta
y
o
u
tl
ie
r
p
ay
m
en
t
tr
ig
g
er
ed
a
t
p
re
d
et
er
m
in
ed
l
en
g
th
o
f
st
ay
(
d
ep
en
d
en
t
o
n
H
R
G
).
D
ai
ly
r
at
e
sp
ec
ifi
c
t
o
H
R
G
•
N
o
o
u
tl
ie
r
p
o
li
cy
•
N
o
o
u
tl
ie
r
p
o
li
cy
F
le
x
ib
il
it
ie
s
•
U
n
b
u
n
d
li
n
g
o
f
ca
re
p
at
h
w
ay
s
u
b
je
ct
t
o
l
o
ca
l
ag
re
em
en
t
•
L
o
ca
l
‘p
as
s
th
ro
u
g
h
’
p
ay
m
en
ts
f
o
r
n
ew
te
ch
n
o
lo
g
y
•
E
m
er
g
en
cy
r
ea
d
m
is
si
o
n
s:
l
o
ca
l
ar
ra
n
g
em
en
ts
fo
r
d
et
er
m
in
in
g
a
p
p
ro
p
ri
at
e
re
im
b
u
rs
em
en
t
an
d
cr
it
er
ia
•
U
n
b
u
n
d
li
n
g
o
f
ca
re
p
at
h
w
ay
s
u
b
je
ct
t
o
lo
ca
l
ag
re
em
en
t
•
L
o
ca
l
‘p
as
s
th
ro
u
g
h
’
p
ay
m
en
ts
f
o
r
n
ew
te
ch
n
o
lo
g
y
•
L
o
ca
l
fl
ex
ib
il
it
ie
s
co
u
ld
b
e
ap
p
li
ed
t
o
s
u
p
p
o
rt
s
er
v
ic
e
re
d
es
ig
n
So
u
rc
e:
D
ep
ar
tm
en
t
o
f
H
ea
lt
h
,
2
0
0
9
c.
N
ot
e:
T
ea
ch
in
g
a
n
d
r
es
ea
rc
h
a
re
f
u
n
d
ed
e
n
ti
re
ly
s
ep
ar
at
el
y,
a
n
d
t
h
ei
r
co
st
s
ar
e
n
o
t
in
cl
u
d
ed
i
n
P
b
R
.
England: The Healthcare Resource Group system 213
12.5.4 Quality-related adjustments
Following recommendations in High-quality care for all (Darzi & Department of
Health, 2008), from 2009/2010 all acute trusts publish ‘quality accounts’
alongside their fi nancial accounts. The Commissioning for Quality and
Innovation (CQUIN) payment framework came into effect in April 2009. It
allows PCTs to link a specifi c, modest proportion of providers’ income (agreed
nationally) to the achievement of realistic locally agreed goals. In 2009/2010,
the CQUIN payment framework covered 0.5 per cent of a provider’s annual
contract income (Department of Health, 2008), increasing to 1.5 per cent in
2010/2011. CQUIN payments are made at monthly intervals, alongside
payment of regular income, and adjusted to refl ect achievement against
contractual goals. The CQUIN framework applies to all patient-related activity,
including activity reimbursed as part of the PbR system.
12.5.5 Main incentives for hospitals
The principal aims of PbR are to increase ‘throughput’ (activity), reduce wait-
ing times, support patient choice and improve effi ciency, as well as increas-
ing patient satisfaction while at the same time keeping costs under control
(Miraldo et al., 2006). Because hospitals are given a fi xed tariff per HRG for the
work they carry out, PbR encourages them to cut costs and reduce lengths
of stay, thus freeing up capacity to treat more patients. Increasing activity
means that patients are treated more quickly, improving access to health care
for patients on waiting lists (Mannion et al., 2008). PbR also facilitates choice
by encouraging new providers into the market, increasing competition in the
fi eld, and improving the mix of care provided by hospitals (Miraldo et al.,
2006).
International empirical evidence suggests that the introduction of a prospective
payment system can offer providers perverse incentives to improve their fi nancial
position. For example, hospitals can engage in ‘DRG drift’, up-coding patients to
more expensive DRGs and resulting in over-reimbursement.
PbR may encourage providers to ‘cream-skim’ (select less complicated cases)
in order to reduce costs. However, the Audit Commission – which now regularly
monitors and audits coding performance in English hospitals – found little
evidence of systemic gaming or deliberate up-coding: observed coding errors
were associated with both positive and negative fi nancial consequences (Audit
Commission, 2008).
12.6 New/innovative technologies
12.6.1 Steps required prior to introduction in hospitals
While adoption decisions for most new technologies are made by individual
hospitals, NICE also provides guidance to the NHS on health care technologies,
public health and clinical practice. The overall aim is to promote good health
and prevent and treat ill health (NICE, 2007).
214 Diagnosis-Related Groups in Europe
NICE assesses selected new and existing medicines. Almost all newly licensed
cancer drugs are appraised by NICE before they are made available routinely
to patients by the NHS. The appraisal process can introduce signifi cant delays
in patient access, partly due to the assessment process, but also because of
stakeholder appeals against NICE’s preliminary decisions.
Since January 2002, the NHS has been legally obliged to provide funding
and resources in England and Wales for medicines and treatments recom-
mended by NICE’s technology appraisal process. This means that when NICE
recommends a medicine, the NHS must ensure it is publicly available within
three months of the guidance being issued. If a new medicine is not recom-
mended by NICE (perhaps because other options are more cost-effective), it
should not be provided routinely by the NHS. However, the Richards report
recommended that patients be allowed to pay privately for medicines not
funded by the NHS, without losing their entitlement to NHS care that they
would otherwise receive (Richards, 2008).
12.6.2 Payment mechanisms
‘Pass through’ payments are used to fund new and innovative technologies.
These apply to new devices, drugs, treatments and technologies or a new
application of an existing technology. They give the purchaser the fl exibility
to make additional payments for higher quality care than the standard care
covered by the national tariff. Any such arrangement between a provider and
purchaser should be fi xed for a maximum of three years, and the price should be
agreed in advance and be directly related to the additional cost of the new
technology.
Some activity, including some high-cost drugs, devices and procedures are
excluded from the PbR tariff, such as magnetic resonance imaging (MRI) scans,
cochlear implants, orthopaedic prostheses and chemotherapy. Instead, pur-
chasers and providers agree local prices and local arrangements for monitoring
activity.
12.6.3 (Dis-)incentives for hospitals to use new technologies
The fi nancial incentive to the provider to innovate depends on the impact on
provider costs, and on whether the innovation improves patient outcomes but
increases provider costs (Boyle et al., 2007).
If the innovation is cost-saving in nature, there is a clear incentive to adopt
it. If the innovation is more expensive and more effective, the provider may be
reluctant to adopt the new technology. To address this disincentive, the price
under prospective funding must be adjusted to compensate providers for the
additional cost. Technology lags also result in higher costs for providers. When
new technologies enter the market, the HRG system will not refl ect the costs of
adopting them, which is why pass-through payments are used. However,
creating economic incentives for health care providers to adopt innovative
technologies can lead to a sharp rise in expenditure, diverting resources from
England: The Healthcare Resource Group system 215
other parts of the health system where they might be used to deliver greater
health benefi ts (Schreyögg et al., 2009).
12.7 Evaluation of the DRG system in England
12.7.1 Offi cial evaluation
Since its inception in 2004, various aspects of PbR have been studied, such as
the benefi ts and costs of the policy (Marini & Street, 2006; Miraldo et al., 2006)
and its incentives and disincentives (Mannion et al., 2008). The evidence
suggests that PbR has generally had a positive impact on hospital activity and
effi ciency, with no deterioration in the quality of care provided (Table 12.4).
A national evaluation of PbR used mixed methods to assess the effects on
hospital behaviour (Farrar et al., 2007). This included the exploration of an
appropriate theoretical framework, a series of semi-structured interviews, and
an econometric analysis of routine data. The theoretical framework hypothesized
that a fi xed national tariff would lower unit costs. The interviews revealed
positive attitudes toward PbR, despite some scepticism as to whether it would
achieve its objectives. The econometric analysis found that unit costs fell with
the introduction of PbR, with no adverse effect on the quality of care, suggesting
that lower unit costs were the result of effi ciency savings. Meanwhile, volume
of activity increased, while there was little evidence of a change in coding
patterns.
The Audit Commission’s evaluation of PbR (Audit Commission, 2008) found
broadly similar fi ndings to those of the national evaluation. PbR was associated
with increased activity and effi ciency in elective care, although PbR itself was
not considered to be the principal driver behind these changes. Findings on
quality of care were similar to those reported in the national study. The Audit
Commission made a number of recommendations on the future development
of PbR, including (1) strengthening of the information infrastructure so that
providers are more accurately reimbursed for activity carried out; (2) greater
fl exibility in the national tariff, with a greater scope for unbundling tariff prices
into separate components; (3) greater consideration given to the possibility of
separate funding streams for capital and quality; and (4) the introduction of
some normative tariffs for selected HRGs, whereby the tariff would be based not
on average prices but on the costs of higher performing providers (a view also
shared by others) (Street & Maynard, 2007a).
12.7.2 Authors’ assessment
Despite implementation diffi culties (Department of Health & Lawlor, 2006),
PbR has been rolled out as the funding mechanism that covers almost all NHS
inpatient care in England. Concerns regarding unintended consequences and
up-coding have proved largely unfounded and large increases in tariff-funded
hospital activity have not materialized. This implies that PbR has not led to
widespread fi nancial instability among purchasers (PCTs). This may be because
216 Diagnosis-Related Groups in Europe
Table 12.4 Impacts of PbR: Overview of evidence
Impact of PbR Study
Activity Proportion of elective care provided as day cases has
increased
2007a
‘PbR has had a positive effect on activity in elective
care. Day cases have increased and the LOS for elective
inpatients has fallen.’
‘Other policies have also encouraged increases in activity.
We consider that PbR has at most contributed to these
positive trends rather than driven them.’
2008b
Unit costs Unit costs have fallen more quickly where PbR was
implemented
2007a
Volume of spells Both Foundation Trusts and non-Foundation Trusts have
increased volumes, although these may be linked to other
initiatives, such as waiting-time targets, which have also
affected the volume of care provided
2007a
Effi ciency Improved effi ciency through reduction in unit costs with
no reduction in quality
2007a
Quality Little change in quality 2007a
The negative impact on quality which some feared has not
materialized
2008b
DRG coding Very limited evidence of a change in the pattern of coding 2007a
Financial
management
and information
systems
Have encouraged commissioners and providers to
strengthen their systems as well as their overall planning
2008b
Administrative
costs
Estimated to have increased in both hospital trusts and
PCTs
2006c
Sources: aFarrar et al., 2007; bAudit Commission, 2008; cMarini & Street, 2006.
PCTs have improved their monitoring of provider activity and performance;
some have tried to manage demand by investing in initiatives to reduce
avoidable hospital admissions (Audit Commission, 2008). However, they have
been less successful at restraining the strong incentives that motivate hospitals
to undertake more elective activity (Mannion et al., 2008).
HRG4 – the new classifi cation system underpinning the 2009/2010 tariff – has
potential for improving the fairness of the payment system. The role of
‘unbundling’ has been enhanced, and the increased number of categories and
greater separation of patients by different complexity levels should, in principle,
help to ensure that payments better refl ect casemix differences. However, there
is a risk that unbundling could lead to increased pressure on budgets, as activities
that were previously paid for by a single tariff are now funded separately.
England: The Healthcare Resource Group system 217
12.8 Outlook
12.8.1 Trends in hospital service or general care delivery
Historically, there has been a trend toward reduced use of inpatient care and
toward treating more patients on a day-case basis or in outpatient departments.
The development of HRG4 allows greater scope for ‘unbundling’ elements of
care from the base-HRG so that services can be provided in non-inpatient
settings where appropriate.
1 2.8.2 Trends in DRG application/coverage
PbR drives the refi nement of HRGs and the development of classifi cation
systems in non-hospital settings. The Department of Health has progressively
been extending the scope of PbR to cover adult mental health, long-term
conditions, preventative services, sexual health, community services, ambulance
services and out-of-hours primary care (Department of Health, 2007). In some
of these areas, pilot work is under way locally to determine the appropriate
units of activity (‘categories’). The NHS reference costs already collect cost data
for most of these areas, but local costing exercises are also being carried out to
test whether the use of tariffs for these ‘de novo’ categories is feasible, particularly
for mental health services (Mason & Goddard, 2009).
Perhaps the most important initiative is the development of ‘best practice’
tariffs for high-volume areas, with signifi cant unexplained variation in quality
of clinical practice and clear evidence of what constitutes best practice
(Department of Health, 2009d). In 2010, prices for cholecystectomy, fragility
hip fracture, cataracts and stroke were based on the most effi cient cost rather
than average cost. From 2011/2012, best practice tariffs are to be extended to
adult renal dialysis, interventional radiology, transient ischaemic attack, and
paediatric diabetes (Department of Health, 2011). This means that DRGs have
progressed gradually from a means of classifying activity, then to paying for
activity, and now to incentivizing quality and better outcomes for patients.
This welcome direction of travel represents the next challenge for policy
development and evaluation over the coming decade.
12.9 Notes
1 The authors thank Martine Bellanger (EHESP) and Alexander Geissler (TUB) for helpful
comments on an earlier draft. We are responsible for all remaining errors and omissions.
2 More information is available at the NHS Choices web site (http://www.nhs.uk/
NHSEngland/thenhs/about/Pages/nhsstructure.aspx, accessed 29 June 2011).
3 ‘Other’ here refers to all other hospital costs that are not part of day-case, inpatient or
outpatient activity. It includes community services, critical care services, A&E
medicine, radiotherapy and chemotherapy, renal dialysis, and kidney and bone
marrow transplantation, for example.
218 Diagnosis-Related Groups in Europe
12.10 References
Anthony, P. (1993). Healthcare resource groups in the NHS: a measure of success. Public
Finance and Accountancy, 23:8–10.
Audit Commission (2006). Payment by Results Assurance Framework: Pilot Results and
Recommendations. London: Audit Commission.
Audit Commission (2008). The Right Result? Payment by Results 2003–2007. London: Audit
Commission (http://www.audit-commission.gov.uk/SiteCollectionDocuments/Audit
CommissionReports/NationalStudies/The_right_result_PbR_2008 , accessed 29
June 2011).
Audit Commission (2010). Improving Data Quality in the NHS: Annual Report on the PbR
Assurance Programme 2010. London: Audit Commission.
Benton, P.L., et al. (1998). The development of healthcare resource groups. Version 3.
Journal of Public Health Medicine, 20:351–8.
Boyle, S., Hutton, J., Street, A., Sussex, J. (2007). Introducing Activity-Based Funding to
Financial Flows between Providers and Commissioners in Northern Ireland. Draft Report for
McClure Watters. York: York Health Economics Consortium.
Coles, J.M. (1993). England: ten years of diffusion and development, in J.R. Kimberley, G.
de Pourourville, eds. The Migration of Managerial Innovation. San Francisco: Jossey-Bass
Publishers.
Darzi, A., Department of Health (2008). High-Quality Care for All: NHS Next Stage Review
Final Report. London: Department of Health.
Dawson, D., Street, A. (1998). Reference Costs and the Pursuit of Effi ciency in the ‘New’ NHS.
York: University of York Centre for Health Economics.
Dawson, D., Street, A. (2000). Reference costs and the pursuit of effi ciency in the ‘new’
NHS, in P.C. Smith, ed. Reforming Markets in Health Care: An Economic Perspective.
Buckingham: Open University Press.
Department of Health (2002). Reforming NHS Financial Flows: Introducing Payment by
Results. London: Department of Health.
Department of Health (2007). Options for the Future of Payment by Results: 2008/2009 to
2010/2011. Leeds: Department of Health.
Department of Health (2008). Using the Commissioning for Quality and Innovation (CQUIN)
Payment Framework. London: Department of Health.
Department of Health (2009a). Department of Health Activity Statistics [web site]. London:
Department of Health (http://www.dh.gov.uk/en/Publicationsandstatistics/Statistics/
Performancedataandstatistics/Beds/DH_083781, accessed 29 June 2011).
Department of Health (2009b). NHS Costing Manual 2008/2009. London: Department of
Health (http://www.dh.gov.uk/prod_consum_dh/groups/dh_digitalassets/documents/
digitalasset/dh_095858 , accessed 12 July 2011).
Department of Health (2009c). Payment by Results Guidance for 2009–2010. Leeds:
Department of Health.
Department of Health (2009d). Payment by Results in 2010–2011: Letter from David Flory,
Director General, NHS Finance, Performance and Operations. London: Department of
Health.
Department of Health (2011). Payment by Results Guidance for 2011–2012. Leeds:
Department of Health.
Department of Health, Lawlor, J. (2006). Report on the Tariff-Setting Process for 2006/2007.
London: Department of Health.
Farrar, S., Sussex, J., Yi, D. et al. (2007) National Evaluation of Payment by Results. Aberdeen:
University of Aberdeen Health Economics Research Unit.
Hawe, E. (2009). OHE Compendium of Health Statistics: 2009. Abingdon: Radcliffe
Publishing.
England: The Healthcare Resource Group system 219
Information Standards Board for Health and Social Care (2009). Healthcare Resource Groups
4 (HRG4): Dataset Change Notice 17/2008. Leeds: Information Standards Board for
Health and Social Care.
Mannion, R., Marini, G., Street, A. (2008). Implementing payment by results in the
English NHS: changing incentives and the role of information. Journal of Health
Organization & Management, 22:79–88.
Marini, G., Street, A. (2006). The Administrative Costs of Payment by Results. York: University
of York Centre for Health Economics.
Mason, A., Goddard, M. (2009). Payment by Results in Mental Health: A Review of the
International Literature and an Economic Assessment of the Approach in the English NHS.
York: University of York Centre for Health Economics.
Mason, A., Street, A., Miraldo, M., Siciliani, L. (2009). Should prospective payments be
differentiated for public and private health care providers? Health Economics, Policy
and Law, 4:383–403.
Miraldo, M., Goddard, M., Smith, P.C. (2006). The Incentive Effects of Payment by Results.
London: Dr Foster Intelligence.
Miraldo, M., Siciliani, L., Street, A. (2008). Price Adjustment in the Hospital Sector. York:
University of York:
NHS Executive (1997). The New NHS: Modern, Dependable. Leeds: NHS Executive.
NHS Information Centre for Health and Social Care (2006a). Hospital Episode Statistics
1998/9. Ungrossed Data. Leeds: NHS Information Centre for Health and Social Care.
NHS Information Centre for Health and Social Care (2006b). Hospital Episode Statistics
2003/4. Ungrossed Data. Leeds: NHS Information Centre for Health and Social Care.
NHS Information Centre for Health and Social Care (2007a). The Casemix Service. HRG4
Design Concepts. Leeds: NHS Information Centre for Health and Social Care.
NHS Information Centre for Health and Social Care (2007b). The Casemix Service. HRG4
Guide to Unbundling. Leeds: NHS Information Centre for Health and Social Care
(http://www.ic.nhs.uk/webfi les/Services/casemix/Prep%20HRG4/Guide%20to%20
Unbundling_090213v4.0 , accessed 12 July 2011).
NHS Information Centre for Health and Social Care (2008a). HRG Version 3.5 & HRG4
Comparative Chapter Analysis: Version No 1.0. Leeds: NHS Information Centre for
Health and Social Care (http://www.ic.nhs.uk/webfi les/Services/casemix/Prep%20
HRG4/HRG%20v3.5%20and%20HRG4%20Comparative%20Chapter%20Analysis
%202008 , accessed 11 July 2011).
NHS Information Centre for Health and Social Care (2008b). The Casemix Service. HRG4
Reference Cost Grouper – Guide to File Preparation. Leeds: NHS Information Centre for
Health and Social Care.
NHS Information Centre for Health and Social Care (2009). Explanatory Notes: Outpatient
Data 2008–2009. Leeds: NHS Information Centre for Health and Social Care.
NICE (2007). About NICE Guidance: What Does It Mean For Me? Information for Patients,
Carers and the Public – An Interim Guide. London: National Institute for Health and
Clinical Excellence.
Richards, M. (2008). Improving Access to Medicines for NHS Patients. A Report for the Secretary
of State for Health by Professor Mike Richards CBE. London: Department of Health.
Sanderson, H.F. (1995). The use of healthcare resource groups in managing clinical
resources. British Journal of Hospital Medicine, 54:531–4.
Sanderson, H.F., Anthony, P., Mountney, L.M. (1995). Healthcare resource groups –
version 2. Journal of Public Health Medicine, 17:349–54.
Schreyögg, J., Baumler, M., Busse, R. (2009). Balancing adoption and affordability of
medical devices in Europe. Health Policy, 92:218–24.
Schreyögg, J., Stargardt, T., Tiemann, O., Busse, R. (2006). Methods to determine
reimbursement rates for diagnosis related groups (DRG): a comparison of nine
European countries. Health Care Management Science, 9:215–23.
220 Diagnosis-Related Groups in Europe
Street, A., Dawson, D. (2002). Costing hospital activity: the experience with healthcare
resource groups in England. European Journal of Health Economics, 3:3–9.
Street, A., Maynard, A. (2007a). Activity-based fi nancing in England: the need for
continual refi nement of payment by results. Health Economics, Policy and Law,
2:419–27.
Street, A., Maynard, A. (2007b). Payment by results: qualifi ed ambition? Health Economics,
Policy and Law, 2:445–8.
chapter t h i r t e e n
France: Implementing
homogeneous patient groups
in a mixed market
Zeynep Or and Martine Bellanger
13.1 Hospital services and the role of DRGs
13.1.1 The French health care system
The French health care system is based on social insurance, with universal
coverage. Health care provision relies heavily on private providers. Ambulatory
care is mainly provided on a private, and usually solo practice basis. Inpatient
care is delivered both by public hospitals and profi t-making and non-profi t-
making private hospitals. Patients can choose freely between public and private
providers without necessarily needing a referral.
Compared with most other European countries, the French system is charac-
terized by high levels of spending. France devotes about 11 per cent of its gross
domestic product (GDP) to the health sector, which contrasts with an average
of 9 per cent in Organisation for Economic Co-operation and Develop-
ment (OECD) countries (OECD, 2010). In terms of hospital fi nancing, about 91
per cent of total expenditure is fi nanced by the public health insurance funds,
while another 5 per cent is paid for by private complementary insurance. Direct
contributions from the state amounted to only about 1.3 per cent in 2008
(Fenina et al., 2008).
At the macro level, fi nancial stewardship of the health system is shared
between the Government and the health insurance funds. The Government
sets annual fi nancial targets to limit the expenditure of the health insurance
funds. There are separate targets for the hospital sector, the ambulatory sector
and social/long-term care. The hospital sector budget is further divided into
two components: one for acute care, which is fi nanced through diagnosis-
related group (DRG)-based hospital payment (including hospital care at home),
222 Diagnosis-Related Groups in Europe
and one for other hospital services (mainly psychiatric and rehabilitative care),
with separate objectives for the public and private sectors. The public health
insurance funds defi ne the baskets of benefi ts, regulate the prices of procedures,
drugs and devices, and defi ne the levels of co-payment (Mousques & Polton,
2005). The health insurance funds are also in charge of setting tariffs for health
professionals in private practice. Doctors working in private hospitals contract
with health insurance funds and they are paid according to a negotiated fee-for-
service schedule,1 while those in public hospitals are salaried. The salaries and
working conditions of the hospital staff – as well as the prices set for DRGs – are
regulated by the Government.
Budget targets for fi nancing the hospital sector are defi ned by the state for
each region. At the regional level, Regional Hospital Agencies (Agences Régional
d’Hospitalisation, ARH) are responsible for organizing and assuring the quality
of hospital care. In 2010 these agencies were replaced by the newly created
Regional Health Agencies (Agences Régionales de Santé, ARS), which will be
responsible not only for acute care but also for prevention, rehabilitation, long-
term and social care. Currently, the Government is reforming the governance
structure of the health care system in France, shifting the power to the ARS,
which will control the resources and defi ne the strategy for hospitals within a
given region (Or, 2008). Each hospital (including private ones) will have to sign
a contract to defi ne its activity and fi nancing needs. Despite this trend in the
shift of power towards local and regional authorities, regions have little
responsibility for hospital funding.
13.1.2 Hospital services in France
The hospital sector plays an important role in health care provision in France.
One person in six is hospitalized each year, either as an inpatient or on a day-
case basis. Hospitals are also signifi cant providers of outpatient care: they
account for about 33 million specialist consultations and an estimated 15.5
million emergency visits per year. Figure 13.1 compares the volumes of different
types of services provided by hospitals, including psychiatry and rehabilitation
care.
With about 4 beds per 1000 inhabitants, hospital bed capacity in France is at
a level comparable to the OECD average. Acute care (including day cases and
home hospitalizations) accounts for about 16 million cases and is administered
by a mixture of public and private facilities (Table 13.1).
Public hospitals represent 60 per cent of all hospitals and 65 per cent of all
acute inpatient beds (about 221 000 beds in 2007). These hospitals are obligated
by law to ensure continuity of care, which means providing 24-hour emergency
care, accepting any patient who seeks treatment and participating in activities
corresponding to national/regional public health priorities.
The private profi t-making sector represents 25 per cent of all inpatient beds in
France, including 46 per cent of surgical beds and over 70 per cent of ambulatory
beds (patient places). The market share of private hospitals depends heavily on
the type of hospital activity. About 56 per cent of all surgery and a quarter of
obstetric care services are provided by private profi t-making hospitals. Their
France: Implementing homogeneous patient groups in a mixed market 223
Figure 13.1 Overview of hospital activity in France, 2006
Source: HCAAM, 2009.
Table 13.1 Distribution of acute care beds and activity between public and private
hospitals, 2007
Total Public Private non-
profi t-making
Private profi t-
making
Number of beds 221 990 146 461 19 251 56 278
% 100.0 66.0 8.7 25.4
Surgical beds 88 280 41 307 8 151 38 822
% 100.0 46.6 7.5 45.9
Total hospital stays
(episodes, millions) 15.9 8.9 1.3 5.7
% 100.0 56.1 7.8 36.0
of which:
Surgery 5.6 2.0 0.45 3.1
% 100.0 36.4 7.4 56.2
Medicine 9.1 6.1 0.7 2.3
% 100.0 67.0 7.8 25.2
Obstetrics 1.3 0.9 0.086 0.33
% 100.0 67.6 6.6 25.8
ALOS (days) 4.0 4.9 4.2 2.5
Source: Authors’ own compilation based on 2007 data from the French hospital activity
database (PMSI).
market share goes up to more than 70 per cent in some areas of elective surgery,
such as eye surgery (cataract in particular), ear surgery and for endoscopies.
However, certain complex procedures are carried out almost exclusively by
public hospitals, for example in the case of burn treatments (92 per cent) or
treatment of patients needing surgery for serious multiple trauma (97 per cent).
Finally, private non-profi t-making hospitals specialize more in medium- to
long-term care; they represent about 8 per cent of acute care activity. Three
quarters of these hospitals have a special agreement with the state and they
have the same engagement terms as public hospitals for providing ‘public
224 Diagnosis-Related Groups in Europe
services’, such as (24-hour) continuous care, for example. In return they can
receive public subsidies. With the last reform, private profi t-making hospitals
will also have the opportunity to sign the same type of contract with the ARS.
Hospital profi les in terms of size (number of beds) vary largely by ownership
status (Table 13.2). Close to 60 per cent of private profi t-making hospitals have
fewer than 100 beds. This fi gure increases to 90 per cent for those specialized in
surgery. By contrast, the public sector is characterized by a diversity of profi les,
with about 30 per cent of general public hospitals having over 300 beds, while
20 per cent have fewer than 100 beds.
Until 2004/2005, two different funding arrangements were used to fi nance
public and private hospitals. Public and most private non-profi t-making hospi-
tals operated according to global budgets, mainly based on historical costs, while
private profi t-making hospitals were fi nanced through a mixture of per diem
and fee-for-service payments. Since 2004, DRG-based hospital payment has been
gradually introduced into French hospitals. In public hospitals, the share of all
acute care activities fi nanced by the system has progressively increased: from 10
per cent in 2004 to 25 per cent in 2005, reaching 100 per cent in 2008. Private
profi t-making hospitals have been fi nanced entirely by DRG-based hospital pay-
ment since February 2005. However, during a transition period that extends
Table 13.2 Distribution of hospitals by size and ownership status
No. of
hospitals
< 30
beds
(%)
30–99
beds
(%)
100–
199
beds (%)
200–
349 beds
(%)
350 +
beds
(%)
Private
profi t-
making
Establishments for
surgical care 171 18 70 12 0 0
Establishments for
medical care 35 14 60 26 0 0
Establishments for
multidisciplinary care 374 3 41 43 13 0
Total private, profi t-
making 580 8 51 32 8 0
Private
non-
profi t-
making
Establishments for
surgical care 12 17 83 0 0 0
Establishments for
medical care 45 24 53 18 2 2
Establishments for
multidisciplinary care 119 6 35 39 16 4
Total private, non-
profi t-making 176 11 43 31 11 3
Public
University hospitals 170 0 5 10 18 66
Local hospitals 355 25 58 15 2 0
General hospitals 643 1 19 26 23 31
Total public 1168 8 29 21 16 27
TOTAL 1924 8 37 25 13 17
Source: Or et al., 2009.
France: Implementing homogeneous patient groups in a mixed market 225
until 2012, national DRG prices are still adjusted to refl ect hospitals’ historical
cost patterns, in order to shelter them from excessive budget cuts.
13.1.3 Purpose of the DRG system
Initially introduced for reporting on hospital activity in France, the DRG
classifi cation system has since been used to adjust budget allocations and is
now used for hospital payment. The fi rst French patient classifi cation system,
Groupes Homogènes des Malades (GHM, translated into English as ‘homogeneous
groups of patients’) was introduced in 1986 for a sample of voluntary public
hospitals in order to better describe hospital activity. Following the hospital
reform measures passed in 1991, collecting/reporting data on hospital activity
using GHMs became mandatory for all public hospitals. Increasingly, these data
were used to compare hospital productivity and to make adjustments to global
budgets. However, during the 1990s, several Ministers of Health still declared
that DRG data will never be used for hospital payment.
Providing DRG data only became compulsory for private profi t-making
hospitals in 1998. It took another six years to use these data as a basis for pay-
ing hospitals, which is currently the main purpose of the DRG system. DRG-
based hospital payment was introduced in 2004/2005 for acute care services
(including home hospitalization), with the following objectives: to improve
effi ciency; to create a ‘level playing fi eld’ for payments to public and private
hospitals; to improve the transparency of hospital activity and management;
and to improve quality of care.
13.2 Developing and updating the DRG system
13.2.1 The current DRG system at a glance
There is one national DRG system in France – GHMs – used as the basis of
hospital payment in France since 2004/2005. The system applies to all hospitals
(public and private) and all patients (inpatients and day cases), except those
treated under psychiatry, rehabilitation and long-term care. Payments received
through this system account for 56 per cent of all hospital expenditures (ATIH,
2009).
The current GHM system (version 11) was introduced in January 2009. It
defi nes 2297 GHMs within 26 Major Diagnostic Categories (catégories majeures
de diagnostic, CMD (MDC in English)), one Pre-MDC group (catégorie majeure 27)
for organ transplantations and one undifferentiated group for ‘sessions’ (séance),
mainly for chemotherapy, radiotherapy or dialysis (CMD 28). Furthermore, it
differentiates between ‘surgical’, ‘other procedure’, ‘medical’, and ‘undifferen-
tiated’ categories. There are 606 base-GHMs, most of which are split into four
severity levels.
The institution responsible for developing the GHM patient classifi ca-
tion system and calculating prices is the Technical Agency for Hospital
Information (ATIH). The ATIH was created in 2002 and is an independent
226 Diagnosis-Related Groups in Europe
public administrative institution, co-funded by the Government and the natio-
nal health insurance funds. It includes an advisory committee, involving repre-
sentatives of public and private health care facilities, which make suggestions
based on their experiences of or within the system.
13.2.2 Development of the French DRG system
The initial idea of a French patient classifi cation system dates back to the early
1980s, when the Government decided to introduce global budgets at the hospi-
tal level to replace the previously existing poorly regulated per diem system. It
was planned to adjust the budgets allocated to hospitals by measuring their
clinical activity through the GHMs (Michelot & Rodrigues, 2008).
The initial French GHM classifi cation (tested between 1986 and 1990) was
inspired directly from the third DRG version of the United States Health Care
Financing Administration (HCFA-DRG) but the GHM system was later modifi ed
to include parts of the All-Patient DRG system. The most important modifi cation
was the introduction of a specifi c major category (CM 24) for day cases. In 1996
a National Cost Study (ENC) was set up with data from about 35 voluntary
public hospitals in order to calculate French GHM cost weights.
The fi rst GHM version was introduced in public hospitals between 1990 and
1993. Eleven versions have been implemented since then (Table 13.3). In earlier
versions of the GHM system, a closed list of secondary diagnoses (inspired from
the original Yale list) was used to identify ‘signifi cant complications’ (CMAs),
independent of the principal diagnosis of the patient. However, later versions
of the GHM used several lists of ‘exceptions’ in order to deal with specifi c cases.
Version 9 (2004–2005) introduced a separate list of diagnoses for episodes
which are acutely severe/complicated (the aforementioned CMAs).
Version 10 (2006–2008) aimed to improve the classifi cation system, taking
into account problems encountered in fi nancing hospitals. In response to
requests from the hospital federations and from the Ministry of Health, a
number of extra (mostly ambulatory) surgical groups and specifi c DRGs for
non-surgical ambulatory procedures were created.
The current version (11) has seen a major change: the number of GHMs
increased almost threefold through the introduction of four levels of case
severity applied to most base-GHMs (see Table 13.3). Information on length of
stay, secondary diagnoses and old age is now used in a more systematic way in
order to improve cost homogeneity of GHMs, especially of medical GHMs.
Moreover, day cases can now be identifi ed as a separate group for relevant
GHMs and, consequently, the old French specialty ‘CM 24’ (which was a
mixture of day cases and very short stays) was abandoned.
13.2.3 Data used to develop the DRG system
Two different databases have been used to develop the current DRG system.
The patient classifi cation system is based on the French hospital activity data-
base (PMSI), which contains information about patient characteristics, primary
T
a
b
le
1
3
.3
G
H
M
v
er
si
o
n
s
u
se
d
f
ro
m
1
9
8
6
t
o
2
0
0
9
D
R
G
s
ys
te
m
H
C
FA
-D
R
G
G
H
M
V
.1
G
H
M
V
.2
G
H
M
V
.3
G
H
M
V
.4
G
H
M
V
.5
G
H
M
V
.6
G
H
M
V
.7
G
H
M
V
.9
G
H
M
V
.
1
0
G
H
M
V
.1
1
D
at
e
1
9
8
6
–1
9
9
0
1
9
9
1
–
1
9
9
3
1
9
9
4
1
9
9
5
1
9
9
6
–
1
9
9
7
1
9
9
8
–
1
9
9
9
2
0
0
0
–
2
0
0
1
2
0
0
2
–
2
0
0
3
2
0
0
4
–
2
0
0
5
2
0
0
6
–2
0
0
8
2
0
0
9
t
o
p
re
se
n
t
P
u
rp
o
se
E
x
p
er
im
en
ta
ti
o
n
D
es
cr
ip
ti
o
n
o
f
h
o
sp
it
al
ac
ti
v
it
y
H
o
sp
it
al
b
u
d
g
et
al
lo
ca
ti
o
n
H
o
sp
it
al
p
ay
m
en
t
D
at
a
u
se
d
f
o
r
d
ev
el
o
p
m
en
t
D
at
a
fr
o
m
v
o
lu
n
ta
ry
p
u
b
li
c
h
o
sp
it
al
s
U
ti
li
za
ti
o
n
d
at
a
fr
o
m
s
o
m
e
p
u
b
li
c
h
o
sp
it
al
s
N
at
io
n
al
u
ti
li
za
ti
o
n
d
at
a
N
at
io
n
al
u
ti
li
za
ti
o
n
d
at
a
N
at
io
n
al
u
ti
li
za
ti
o
n
d
at
a
N
u
m
b
er
o
f
D
R
G
s
4
5
0
4
8
0
4
8
0
5
7
2
5
8
2
5
8
2
5
9
8
5
9
8
7
7
3
(5
7
3
*)
7
8
4
(5
7
5
*)
2
2
9
7
(6
0
6
*)
A
p
p
li
ed
t
o
V
o
lu
n
ta
ry
sa
m
p
le
o
f
h
o
sp
it
al
s,
o
n
ly
ac
u
te
c
ar
e
se
rv
ic
es
P
u
b
li
c
h
o
sp
it
al
s:
i
n
p
at
ie
n
ts
a
n
d
d
ay
c
as
es
,
ex
cl
u
d
in
g
p
sy
ch
ia
tr
y,
r
eh
ab
il
it
at
io
n
a
n
d
l
o
n
g
-t
er
m
c
ar
e
P
u
b
li
c
an
d
p
ri
v
at
e
h
o
sp
it
al
s:
i
n
p
at
ie
n
ts
a
n
d
d
ay
c
as
es
,
ex
cl
u
d
in
g
p
sy
ch
ia
tr
y,
r
eh
ab
il
it
at
io
n
a
n
d
l
o
n
g
-t
er
m
c
ar
e
*B
as
e
G
H
M
i
n
p
ar
en
th
es
es
.
Fo
r
in
st
an
ce
,
7
7
3
(
5
7
3
)
m
ea
n
s
5
7
3
c
at
eg
o
ri
es
f
o
r
d
es
cr
ib
in
g
b
as
e
G
H
M
(
co
m
b
in
at
io
n
s
o
f
d
ia
g
n
o
se
s/
p
ro
ce
d
u
re
s)
a
n
d
2
0
0
t
o
d
is
ti
n
g
u
is
h
c
as
e
se
v
er
it
y.
228 Diagnosis-Related Groups in Europe
and secondary diagnoses, procedures, and length of stay of treated patients, as
well as the GHM to which each patient is assigned. This is a national database
covering all public (since 1996) and private (since 1998) hospitals.
The information for calculating DRG cost weights comes from the French
hospital cost database (ENCC), which provides detailed cost information for
each hospital stay from 70–100 voluntary hospitals. Until 2006 the ENCC
covered only public and private non-profi t-making hospitals (about 40 in total)
representing about 3 per cent of these hospitals. Since 2006, cost information is
collected from a set of private profi t-making hospitals in order to calculate costs
in a comparable way across all hospitals for the ENC. The number of participating
hospitals increased slightly between 2006 and 2007 (Table 13.4). At present, the
ENCC covers 99 hospitals, representing 13 per cent of total stays.
13.2.4 Regularity and method of system updates
The GHM classifi cation algorithm has been revised continuously since its
introduction. Since 2005, the ATIH has introduced a process of regular revisions
of the patient classifi cation system in order to take account of changes in
medical practice and technology and to adjust for changes in the WHO
International Classifi cation of Diseases, 10th revision (ICD-10). Alterations to
the system are made on the basis of suggestions from an expert group set up by
the ATIH and composed mainly of physicians and statisticians (Patris et al.,
2001).
Table 13.4 Number of hospitals and stays included in the National Cost Studya
Hospital type 2006 2007
Number of
hospitals
in data
sampleb
Number of
episodes
includedc
Surveyed
episodes
as percent
of all
stays (%)
Number of
hospitals
in data
sampleb
Number of
episodes
includedc
Surveyed
episodes as
percent of
all stays
(%)
University hospital 10 512 707 11 13 823 440 17
General hospitals 16 508 520 7 22 718 893 10
Cancer centres 5 268 358 25 7 387 184 36
Private non-profi t-
making hospitals
11 168 616 15 13 224 590 20
Total public
hospitals
42 1 458 201 10 55 2 154 107 15
Private profi t-
making
hospitals
32 628 894 7 44 781 769 9
Total 74 2 087 095 9 99 2 935 876 13
Source: ATIH, 2007b.
Notes: a Data samples from 2006 and 2007, which are included in the ENCC 2008 and ENCC
2009. b Hospitals for which the data provided fi tted the quality standard to calculate costs;
c Number of episodes contributing to reference cost scale (after trimming procedure).
France: Implementing homogeneous patient groups in a mixed market 229
Information from the PMSI about length of stay, as well as information about
costs of treating patients within each GHM from the ENCC are used to assess
cost homogeneity of the diagnostic groups and the classifi cation system as a
whole. The impact of proposed changes to the classifi cation algorithm is tested
using the same data.
GHM cost weights are updated annually by the ATIH on the basis of infor-
mation from the ENCC. However, there is always a time-lag of two years between
the year of the data and the year of the application of prices in hospitals. For
example, data relating to hospital costs from the year 2008 were analysed during
the year 2009 in order to defi ne the GHM prices to be used for hospital payment
in 2010.
13.3 The current patient classifi cation system
13.3.1 Information used to classify patients
Classifi cation of patients into GHMs is based on administrative and clinical
information, both of which are available from the standard patient discharge
summary (RSS) (see Figure 13.2). Clinical data are reported by physicians and
are transmitted to the medical information units (DIMs) of hospitals, where
data are processed and checked before a specialized software programme uses
the information to select the appropriate GHM.
If a patient was transferred between medical wards during the hospital stay,
several departmental discharge summaries (RUMs) are combined into one RSS.
Until 2009, the main diagnosis was coded at admission (main diagnosis of fi rst
RUM) and any additional diagnoses were coded as secondary diagnoses. In the
current GHM version (11), the main diagnosis is assigned by the discharging
department (last RUM) and should represent the ‘cause’ of hospitalization.
Clinical information considered in the classifi cation process includes the
main diagnosis and secondary diagnoses coded using the ICD-10 and the
procedures coded according to the French classifi cation of procedures (CCAM).
In addition, birth weight and age (in days) of neonates are considered. Adminis-
trative data that are used to defi ne the severity level of patients include age,
length of stay and mode of discharge (death, transfer).
13.3.2 Classifi cation algorithm
Every discharged hospital patient is grouped into exactly one GHM on the basis
of information contained in the standard RSS. Figure 13.3 illustrates the grouping
algorithm. The fi rst test carried out is to see if the patient’s hospital stay
corresponded to a ‘session’ (séance) for chemotherapy, radiotherapy or dialysis.
If this is the case the patient is classifi ed into a separate CMD (CMD 28), which
is divided into 15 GHMs without any severity levels. The next step of the
grouping process identifi es a type of Pre-MDC group for organ transplantations
(catégorie majeure 27). Furthermore, ‘transversal’ cases with multiple trauma or
with a diagnosis of AIDS are assigned to specifi c CMDs (26 and 25).
F
ig
u
re
1
3
.2
In
fo
rm
at
io
n
u
se
d
f
o
r
cl
as
si
fy
in
g
p
at
ie
n
ts
i
n
to
t
h
e
G
H
M
So
u
rc
e:
P
o
w
er
P
o
in
t
p
re
se
n
ta
ti
o
n
p
re
p
ar
ed
i
n
2
0
0
9
b
y
R
.
C
as
h
f
o
r
M
is
si
o
n
T
2
A
.
France: Implementing homogeneous patient groups in a mixed market 231
All other patients are classifi ed into one of 23 mutually exclusive CMDs on
the basis of the main diagnosis. Afterwards, the grouping algorithm examines
the procedures that were carried out during the hospital stay. Cases with
operating room (OR) procedures are classifi ed into a ‘surgical’ partition. Cases
with relevant non-OR procedures are assigned to an ‘other procedure’ partition.
Cases without relevant procedures fall into the ‘medical; partition. In certain
CMDs, an ‘undifferentiated’ partition exists, which contains cases that were
assigned without testing to establish the type of procedures carried out.
Figure 13.3 GHM classifi cation with level of severity
Source: Adapted from Bellanger & Tardif, 2006.
232 Diagnosis-Related Groups in Europe
Within partitions, base-GHMs are selected for a specifi c combination of main
diagnosis and procedures, and often also considering age, complications and
length of stay. If several procedures were performed during the hospital stay,
the most complicated procedure (in terms of complexity and resource use)
determines the classifi cation of patients into base-GHMs. Error-GHMs can be
assigned at several stages of the grouping process if inconsistencies exist, for
example between diagnosis and patient gender or weight and patient age.
A new feature of the current GHM version (11) is that base-GHMs are
systematically split into four levels of severity. Severity levels are defi ned on the
basis of length of stay, age, and secondary diagnoses that represent complications
or co-morbidities (CCs). Lists of secondary diagnoses exist that defi ne their level
of complexity (levels 2 to 4) and specify excluding conditions (that is, a secon-
dary diagnosis is not considered to be a CC for certain main diagnoses).
Severity level 1 corresponds to cases without any CCs or with a length of stay
of less than 3 days. Severity level 2 requires a minimum length of stay of 3 days
and level-2 CCs. Severity level 3 requires a minimum length of stay of 4 days and
level-3 CCs. Severity level 4 requires a minimum length of stay of 5 days and
level-4 CCs. Under certain conditions, patients can be classifi ed into a higher
severity level if their age is either below 2 years or above 69 (or even 79) years. In
addition, death is also used within the system as a marker of case severity. If the
length of stay is more than 3 days, and the patient died during hospitalization,
a case without CCs can be reclassifi ed from level 1 into level 2. The idea is to give
hospitals suffi cient resources to cover the extra costs of dealing with death, but
it is not clear what the implications are for the quality of care.
In addition, for some base-GHMs (for example, cataract surgery, for which
day surgery is a recognized practice), an additional group is created to classify
cases involving ambulatory surgery, previously coded as CM 24.
13.3.3 Data quality and plausibility checks
DIMs within hospitals carry out internal controls to analyse the plausibility of
data. To this end, the ATIH provides them with a specifi c program (DATIM) that
checks consistency between length of stay, type of admission, CCs and severity
levels. In addition, the ATIH provides to each hospital reference means and
standard deviations (from a comparable group of hospitals), as well as an index
of outlier cases. The physicians within the DIM can use this information to
check and correct the data before validating the database.
External data quality and plausibility checks are performed at the regional level
by the ARH and the health insurance funds. The ATIH provides information to
support external controls for hospitals with too many ‘outlier cases’. The principal
objective of external controls is to identify ‘unjustifi ed’ billing of services and
up- or wrong-coding. In 2006, more than 150 000 hospital stays in about 530
hospitals (one third of all hospitals concerned) were inspected: over 60 per cent
of inpatient stays (and more than 80 per cent for ambulatory episodes) had some
kind of coding error or inconsistency in the procedures billed (CNAM, 2006). The
controls also revealed that use of innovative medications (fi nanced separately, on
top of the DRG price) was not justifi ed in about 30 per cent of cases.
France: Implementing homogeneous patient groups in a mixed market 233
If up-coding or incorrect coding is detected, hospitals must reimburse
payments received. In addition, hospitals may have to pay high fi nancial
penalties of up to 5 per cent of their annual budgets. The revenue recovered
from these controls amounted to €24 million in 2006. The number of controls
doubled for the year 2007, but results are not yet available.
13.3.4 Incentives for up- or wrong-coding
Since classifi cation of patients into GHMs determines hospital revenues, strong
incentives exist for hospitals to ‘optimize’ their coding practices. In 2006, a year
after the introduction of DRG-based payment, external controls from health
insurance funds demonstrated that a large number of hospitals either
intentionally up-coded patients or inadvertently classifi ed them into incorrect
GHMs. The up-coding of ambulatory consultations as day cases appeared to be
a real problem (CNAM, 2006). Therefore, the Ministry of Health issued a decree
in 2007 describing those procedures that should not be coded as day cases.
Between 2005 and 2008, the share of inpatient stays without any CCs decreased
signifi cantly in all hospitals, which could indicate DRG creep (see Chapter 6).
13.4 Cost accounting within hospitals
13.4.1 Regulation
The recommended hospital cost-accounting model is called ‘analytical account-
ing’, which is essentially a top-down accounting model distributing current
consumption of resources into various cost groups (Ministry of Health, 2007).
Since 1992, all hospitals participating in the ENC must provide data according to
this model. In 2007, in order to harmonize cost-accounting methods for private
hospitals joining the database, common accounting rules were defi ned by a
decree (Circulaire DHOS 2007/06/27). The rest of the public and private hospitals
use a far less detailed accounting system than the analytical one.
13.4.2 Main characteristics of the cost-accounting system
Hospitals participating in the joint ENC use a combination of top-down and
bottom-up cost accounting, with elements of both gross-costing and micro-
costing (see Chapter 5) (Bellanger & Tardif, 2006). Participating hospitals must
be able to provide patient-level information regarding all procedures performed
and relating to direct charges for certain specifi c drugs and medical devices,
blood, external laboratory tests and fees for private physicians.
Preparing the hospitals’ cost accounts for the analysis requires excluding all
expenditure related to activities that are not reimbursed through the GHMs (for
example teaching, research, psychiatry, rehabilitation, intensive care, neona-
tology, physicians’ fees in private hospitals), and excluding the costs of high-
cost drugs and medical consumables that can be directly attributed to patients.
All remaining costs are distributed into a number of cost centres.
234 Diagnosis-Related Groups in Europe
In order to calculate costs per hospital stay, unit costs of cost centres are
determined and allocated to patients on the basis of easily identifi able allocation
criteria. Total costs of each hospital stay are broken down into three main
components: medical costs, overheads and capital costs. Medical costs include: (1)
direct charges, which can be directly attributed to a patient, such as specifi c drugs
and medical devices, blood, outpatient tests and fees for private physicians; (2)
costs at direct cost centres – that is, clinical costs at the ward level (for example
medical and non-medical staff, drugs, materials and running costs of hospital
wards, equipment and maintenance), which are allocated to patients on the basis
of length of stay in the hospital ward; and (3) medico-technical costs (such as
anaesthesia, surgery, laboratory, radiology, pharmacy, including the running
costs of these departments). Since patient-level consumption of these services
(relating, for example, to the number of imaging tests or surgical procedures) is
recorded by hospitals, it is possible to allocate costs to patients on the basis of
services consumed and imputed costs per service at medico-technical cost centres.
Overhead costs include general administration, as well as and management
and support services such as laundry, catering, sterilization, pharmacy and
hospital hygiene. Capital costs include rental of buildings, interests, depreciation
of buildings, and taxes. Overheads and capital costs are allocated to patients on
the basis of calculated per diem costs. Despite this common methodology, the
cost components may not always cover exactly the same cost items in public
and private hospitals.
13.5 DRGs for reimbursement
13.5.1 Range of services and costs included in GHM-based
hospital payment
Since 2008, all acute care activity in public and private hospitals is fi nanced on
the basis of GHMs (see Figure 13.4). Pilot tests to include psychiatric care and
rehabilitative care services into GHM-based hospital payment are planned to
start in 2011.
Currently, GHM prices differ for public and private hospitals, since they
include different cost categories and are based on historical costs in each sector
(Table 13.5). The tariffs for public hospitals cover all costs linked to a stay
(including medical personnel, tests and procedures), while those for private
hospitals do not cover medical fees of doctors (paid for by fee-for-service
payments) and the cost of some technical equipment, paid for by a specifi c
allocation to concerned hospitals (forfeit haute téchnicité). Until 2010, certain
medical devices were billed separately by private hospitals, while they were
included in the DRG pricing in public hospitals. The objective is to harmonize
cost- and tariff-calculation methods between the two sectors by 2012.
Since 2008, capital costs (equipment, fi nancial and building costs) are
included in GHM prices. Hence, hospitals are expected to fund capital invest-
ments from these revenues. However, some (unmeasured) part of capital costs
is fi nanced through specifi c funding streams to help public hospitals to fi nance
weighty investment plans imposed by recent hospital reforms. This means that
France: Implementing homogeneous patient groups in a mixed market 235
Figure 13.4 Range of services included in the GHM-based hospital payment system
Source: HCAAM, 2009.
Table 13.5 Cost categories included in GHM prices for public and private hospitals,
2010
Hospitals Public Private, profi t-making
Cost categories Included in DRG price
Payment for physicians including social charges Yes No
Payment for other medical staff Yes Yes
Investment in technical equipment Yes 25%
Expensive drugs and devices from a closed list No No
All medical material, devices, drugs Yes Yes
Infrastructure/Overheads Partly Yes
the part of the capital costs covered by GHM prices is not completely transpar-
ent (Cour des comptes, 2009).
In 2008, payments made through GHM-based hospital payment repre-
sented about 56 per cent of hospital expenditure budgets (which amount to
€67 billion). The overall payments made for ‘missions of general interest’
(MIGAC)2 represented about 10 per cent of the public hospital budget, but there
are large variations between hospitals according to their size, ownership status,
and so on. Additional payments for expensive drugs and medical devices represent
on average about 6 per cent of hospital expenditure, while annual remuneration
for providing specifi c services such as intensive care, emergency care, and organ
transplants corresponds to 1.5 per cent of total hospital expenditure (see Figure
13.5). Global budgets are used for the fi nancing of rehabilitative, psychiatric and
long-term care and account for about 27 per cent of all hospital expenditure.
13.5.2 Calculation of reference costs and prices
Average costs per GHM (reference costs) are calculated from the ENC separately
for public and private hospitals (ATIH, 2007a).
236 Diagnosis-Related Groups in Europe
Using as a basis information relating to costs of individual patients, outlier
cases are detected for each GHM through two different ‘trimming’ procedures:
the fi rst on length of stay and the second on costs. Trimming by length of stay
is applied only to those GHMs for which the severity level is 1. This involves
excluding all GHMs for which length of stay is longer than: [average length of
stay (ALOS) � 2.5].
On average, 0.7 per cent of all public hospital stays and 0.4 per cent of all
private hospital stays are trimmed from the ENCC on this basis (ATIH, 2009). In
rare cases, this is followed by a second stage of trimming based on cost data.
However, according to the ATIH, only 92 stays were discarded in 2007 during
the cost data part of the trimming process.
Given that the ENCC does not cover all hospitals, but just a small group, costs
per GHM are weighted by the type of hospital. For the public sector, fi ve types of
hospitals are defi ned: general hospitals producing fewer than 16 000 episodes per
year; those producing more than 16 000 episodes; teaching hospitals; cancer
centres; and private non-profi t-making hospitals. The ALOS, the ALOS in wards
that provide services related to reanimation, and the average number of pro-
cedures performed by type of hospital are used to weight average costs per GHM
obtained from the ENCC. For private profi t-making hospitals, the ALOS for the
sector – as well as the ALOS in reanimation/intensive care (when relevant),
together with the average number of procedures – are used as weighting variables.
The reference costs are used to compute ‘raw’ tariffs per GHM given the total
budget for GHM-based payments (per sector). The actual prices per GHM are
determined by the Ministry of Health, taking into account the budget envelope
(expenditure target) for the acute care sector and other political priorities. The
result is a macro-level price/volume control mechanism: if the growth in total
volume of activity exceeds the target for the inpatient sector, GHM prices are
reduced. In 2009 the ATIH noted that GHM prices were modifi ed to adjust for
Figure 13.5 Breakdown of total hospital expenditure, 2008
Source: Adapted from ATIH, 2009
* Revenues for rehabilitative, psychiatric, and long-term care.
France: Implementing homogeneous patient groups in a mixed market 237
the increase in MIGAC budgets, the growth of expenditures for additional
payments on expensive drugs and the evolution of activity volumes and national
priorities (for cancer treatment and palliative care). However, it is not clear how
these different elements changed the prices of different GHMs. Consequently, it
is not possible to predict the evolution of GHM prices from one year to another.
13.5.3 DRGs in actual hospital payment
National GHM prices are set annually. They differ between public and private
hospitals since they do not cover the same cost items (see subsection 13.5.1) but
they are not affected by hospital size or teaching status.
Hospital payment is adjusted for extreme cases. An upper and a lower
threshold are calculated for each DRG in order to identify cases with extremely
long or extremely short lengths of stay. The GHM tariff applies to episodes with
a length of stay between these limits (inliers). For long-stay outlier cases,
hospitals receive GHM-specifi c surcharges (Tariff EXH) for every day that the
patient stayed above the upper length-of-stay threshold. Similarly, if patients
are discharged earlier than the lower length-of-stay threshold, the DRG payment
is reduced by per diem-based deductions (Tariff EXB). The lower threshold is
used to discourage providers from discharging patients earlier than clinically
appropriate. These low/high length-of-stay limits are not always the same for
public and private hospitals.
Currently, the national DRG prices are weighted with a hospital-specifi c
‘transition coeffi cient’ calculated for each hospital from its own historical costs/
prices. The transition coeffi cients aim to avoid large changes in hospital budgets
from one year to another. The objective is for the coeffi cients within public and
private sectors to converge to ‘1’ by 2012. A regional index is also applied to
hospitals in the Parisian area and those in overseas French territories, where
labour costs are higher.
The initial proposition to introduce one DRG price for public and private
hospitals in 2012 has been delayed to 2016 because of the strong reactions from
public hospital federations. However, experimentation with selected DRGs is
expected over the period 2011–2012.
As already mentioned (see subsection 13.5.1), hospitals receive additional
payments for certain services, drugs and medical devices, and if applicable for
teaching and research. Budget envelopes for public missions (MIGAC) are
distributed by the ARH according to nationally defi ned rules. The growing size
of the MIGAC budgets is currently an issue of concern, as the decision regarding
the amount of these budgets seems to be political rather than evidence based.
13.5.4 Quality-related adjustments
There is no specifi c adjustment for quality of care. GHM payments do not vary
according to differences in outcomes. The only GHM-related measure against
inappropriate early discharge (as a dimension of quality) is the use of per
diem-based deductions below the defi ned lower length-of-stay threshold(s).
238 Diagnosis-Related Groups in Europe
Otherwise, quality-related programmes, such as developing infection control
programmes, are negotiated and fi nanced through specifi c allocations from the
ARH as part of the MIGAC budget envelope.
At the same time, with the introduction of GHM-based payment, there has
been quite substantial work – led by the Ministry of Health and the High Health
Authority (HAS) – towards developing indicators to better monitor care quality
in hospitals. A battery of indicators measuring care process and structure/
organization quality – which has been tested and validated in a small number
of voluntary hospitals – will be generalized over the period 2011–2012.
Surprisingly, however, outcome indicators such as standardized mortality rates,
readmission and/or complication rates are not part of that battery of indicators
and they are currently not monitored routinely.
13.5.5 Main incentives for hospitals
The principal incentives provided by GHM-based hospital payment are to
increase activity and to improve effi ciency. Because hospitals are paid a fi xed
tariff per GHM, they are incentivized to reduce length of stay and to treat more
patients. However, since GHM prices are reduced if activity exceeds the target
for the inpatient sector, hospitals do not know whether increasing activity in a
given year will always lead to an increased income in the next year. Since it is
impossible to predict the evolution of GHM prices from one year to another, it
is not clear how much incentive there is for hospitals to increase productivity.
The most obvious perverse incentive for hospitals is for up-coding or wrong-
coding (see subsection 13.3.4). Other possible perverse incentives – such as
engaging in patient selection and cream-skimming – are seen to be less of an
issue for public hospitals since, by law, they cannot select their patients and
have to provide a comprehensive package of care.
13.6 New/innovative technologies
The effect of DRG-based hospital payment on development and introduction of
cost-increasing innovative technologies in hospitals (see Chapter 9) has been a
major preoccupation in France, where access to new therapies (particularly in
cancer treatment) remains one of the most generous in Europe (De Pouvourville,
2009).
Ultimately, the patient classifi cation system and/or GHM prices are updated
in order to refl ect the higher costs for innovative drugs and technologies.
However, two fi nancing mechanisms exist to encourage the development and
utilization of cost-increasing innovative drugs and technologies during the
early stages of introduction to hospitals, as detailed here.
1. Additional payments are made for a certain number of expensive innovative
drugs and medical devices, for which a list is defi ned at the national level.
These are funded on the basis of a maximum standard price. Total expenditure
on these drugs and devices increased by 37 per cent between 2005 and 2007,
reaching €2.4 billion in 2008.
France: Implementing homogeneous patient groups in a mixed market 239
2. The development of innovative technologies is funded by a specifi c budgetary
allocation within the global budget envelope of MERRI (Missions d’enseigne-
ment, de recherché, de reference et d’innovation – teaching, research, recourse
and innovation). These payments are to cover the general cost of innovation-
related activities, as well as specifi c innovative technologies on an experi-
mental basis (such as artifi cial hearts, new-generation ear implants, and so
on). Within this budget, there are specifi c separate payments to ensure quick
access to innovative drugs which have not yet been authorized to be mar-
keted called ‘temporary access for treatment’ (ATU). ATUs can be requested
for one patient or a group of patients. The Agency for Safety of Medical
Products (AFSSAPS) examines the request(s) and decides thereon after
consultation with medical experts. The authorization and funding for ATUs
is for one year, but can be renewed. The duration of the individual ATU
corresponds to the duration of the treatment.
13.7 Evaluation of the GHM system in France
13.7.1 Offi cial evaluations
Several public bodies have recently evaluated specifi c aspects of GHM-based
hospital payment in France. The Evaluation Committee set up by the Ministry
of Health published a report about the fi nancial effects of the hospital payment
reform in September 2009 (DREES, 2009). According to the report, the fi nancial
situation of private hospitals has improved since the introduction of GHM-
based hospital payment, while that of public hospitals has deteriorated. In
2007, one in three public hospitals was in defi cit, with a total budget defi cit of
about €500 million. The report points out that it has been diffi cult for the
public hospitals to reduce their costs, despite a slight increase in their activity.
The report also examined the organizational changes in hospitals through a
survey of 800 hospitals and found that efforts have been concentrated on
modifying the structure of hospital activity (through transfers, hospital mergers,
and so on) rather than on trying to improve effi ciency. There has been little
change in medical and human resource management. Finally, the report points
out the incoherence between the incentives provided by GHM-based hospital
payment and regional health plans aimed at ensuring a needs-based distribution
of hospital resources. Currently, the development of the regional health plans
is disconnected from fi nancial planning and often ignores the fi nancial
constraints faced by hospitals.
In 2009, the Auditor’s Offi ce (Cour des comptes), within the framework of its
annual evaluation of public accounts, presented an evaluation of GHM-based
hospital payment. The major conclusion of the report was that it had not
improved effi ciency in the hospital sector. The report suggests that (1) GHM-
based hospital payment has become a very opaque mechanism of cost control for
managers and local regulators; and (2) the measurement and follow-up of hospi-
tal resources (revenues) is insuffi cient. For example, it is not possible to establish
how hospital revenues (from public health insurance, patients and private com-
plementary insurance) have evolved with respect to their production/activity.
240 Diagnosis-Related Groups in Europe
The report also questions the incomprehensible nature of the price/volume
control mechanism, which makes it very hard for hospitals to predict their
income. Furthermore, it severely criticizes the ambiguous process for fi xing
prices, given that it is not always clear what is included in the price and what is
not.
Furthermore, the Auditor’s Offi ce report estimated that within the hospital
inpatient budgets, the categories which are not included in DRG prices escalated
between 2005 and 2007: the expenditure for expensive drugs and medical
devices increased by 37 per cent and other daily supplementary payments by 21
per cent, against an average of a 4 per cent increase in DRG prices.
13.7.2 Authors’ assessment
To date, GHM-based hospital payment in France appears to fall short of achiev-
ing its stated objectives in terms of improving effi ciency, transparency, fairness
of funding, and quality.
Cost data are not available to identify effi cient providers, to facilitate an
understanding of the differences in medical practices and to monitor changes
in behaviour of various actors. In terms of productivity improvement, it is not
clear to what extent the rise in ambulatory activity represents an increase in
effi ciency, and to what extent this is due to up- or wrong-coding or to oversupply
of services. Quality indicators – such as readmission and avoidable mortality
rates – are not available either.
In addition, the macro-level volume/price control mechanism appears to be
counterproductive. It creates an extremely opaque environment for hospitals,
whereby they cannot predict their income based on their activity. Prices are set
(progressively) independently of costs, which encourages health care facilities
(especially private ones) to opt for less expensive care/therapies.
In order to achieve expected benefi ts in terms of effi ciency and quality, it
is important to improve the monitoring and transparency of the GHM system
(methods used for cost/price calculations, data on individual providers, and
so on), as well as expenses alongside the GHM payments, which are still allo-
cated through an opaque mechanism. Furthermore, a contractual approach –
giving individual providers clear volume and quality signals – could improve
effi ciency.
13.8 Outlook: Future developments and reform
It is intended to introduce GHM-based hospital payment for other hospital
services which are currently fi nanced through global budgets: namely rehabili-
tative and psychiatric care. The construction of a DRG scale for psychiatric
care has proved to be diffi cult. The Ministry of Health (along with the ATIH)
has been developing a DRG system for rehabilitative care, using more or less
the same logic as that applied in inpatient care. The Ministry aims to test
this classifi cation system in a number of hospitals on a voluntary basis in
2011/2012.
France: Implementing homogeneous patient groups in a mixed market 241
13.9 Notes
1 These costs are not accounted for in the hospital sector budget, but are included in the
ambulatory sector.
2 Missions d’intérêt général et de l’aide à la contractualisation: Missions of general
interest and assistance with contracting, including payments for education, research
and public health programmes.
13.10 References
ATIH (2007a). Modalités de calcul du référentiel national de coûts 2007. Données ENCC 2007.
Lyon: Agence Technique de l’Information sur l’Hospitalisation (http://www.atih.
sante.fr/openfi le.php?id=2585, accessed 4 July 2011).
ATIH (2007b). Principaux résultats issus des données de coûts ENCC 2007. Lyon: Agence
Technique de l’Information sur l’Hospitalisation (http://www.atih.sante.fr/openfi le.
php?id=2586, accessed November 2010).
ATIH (2009). Manuel des GHM, version 11. Lyon: Agence Technique de l’Information sur
l’Hospitalisation (http://www.atih.sante.fr/index.php?id=000250002DFF, accessed 4
July).
Bellanger, M., Tardif, L. (2006). Accounting and reimbursement schemes for inpatient
care in France. Health Care Management Sciences, 9:295–305.
CNAM (2006). Contrôles et lutte contre les abus et les fraudes. Paris: Caisse National
d’Assurance Maladie (http://www.securite-sociale.fr/institutions/fraudes/fraude.htm,
accessed 4 July 2011).
Cour des comptes (2009). La Securité Sociale, chapitre 7, La mise en place de la T2A : Bilan à
mi-parcours. Paris: Cour des comptes (February).
DREES (2009). Rapport d’activité du Comité d’évaluation de la T2A, septembre 2009. Paris:
Direction de la recherche, des études, de l’évaluation et des statistiques, Ministère du
Travail, de l’Emploi et de la Santé.
Fenina, A., Geffroy, Y., Duée, M. (2008). Les comptes nationaux de la santé en 2007.
Etudes et Résultats, 655:1–8.
HCAAM (2009). Note sur la situation des établissements de santé, Avril 2009. Paris: Haut
Conseil de l’Avenir de l’Assurance Maladie.
Michelot, X., Rodrigues, J.M. (2008). DRGs in France, in J.R. Kimberly, G. de Pouvourville,
T. D’Aunno, eds. Globalization of Managerial Innovation in Health Care. Cambridge:
Cambridge University Press.
Ministry of Health (2007). Guide méthodologique de la comptabilité analytique hospitalière.
Paris: Ministère de la santé (éditions mise à jour 1997, 2004, 2007).
Mousques, J., Polton, D. (2004). Sickness funds reform: a new form of governance. Health
Policy Monitor, October (http://hpm.org/survey/fr/a4/3, accessed 4 July 2011).
OECD (2010). OECD Health Data 2010: Statistics and Indicators. Paris: Organisation for
Economic Co-operation and Development.
Or, Z. (2008). Changing regional health governance in France. Eurohealth, 14(4):7–8.
Or, Z., Renaud, T., Com-Ruelle, L. (2009). Les écarts des coûts hospitaliers sont-ils justifi ables?
Réfl exions sur une convergence tarifaire entre les secteurs public et privé en France. Paris:
Institut de Recherche et Documentation en Economie de la Santé (IRDES Working
Paper 25).
Patris, A., Blum, D., Girardier, M. (2001), A change in the French patient classifi cation
system. CASEMIX Quarterly, 34:128–38.
de Pouvourville, G. (2009). Les hôpitaux français face au paiement prospectif au cas: la
mise en ouvre de la tarifi cation à l’activité. Revue Economique, 60(2):457–70.
chapter f o u r t e e n
Germany: Understanding
G-DRGs
Alexander Geissler,
David Scheller-Kreinsen,
Wilm Quentin and Reinhard Busse
14.1 Hospital services and the role of DRGs in Germany
14.1.1 The German health system
A key characteristic of the German health care system is the sharing of decision-
making powers between the 16 Länder (states), the federal Government and
statutory civil society organizations. Moreover, Bismarckian principles dominate
statutory health insurance (SHI), that is, important competences are legally dele-
gated to membership-based, self-regulated organizations of payers and providers.
In the most important pillar of the German health care system, the SHI,
sickness funds, their associations and associations of SHI-affi liated physicians
have assumed the status of quasi-public corporations. These self-regulated
corporate structures operate the fi nancing and delivery of benefi ts covered by
SHI within a general legal framework. They are based on mandatory membership
and internal democratic legitimization. They have the power and a duty to
defi ne benefi ts, prices and standards (at federal level) and to negotiate horizontal
contracts to manage and sanction their members’ behaviour (at regional level).
The vertical implementation of decisions made at superior levels is combined
with strong horizontal decision-making and contracting among the legitimate
stakeholders involved in the various sectors of health care.
The corner-stone of health service provision in Germany is the fi fth book of
the German Social Law (SGB V). The SGB V separates the provision of out-
patient and inpatient services. Planning, resource allocation and fi nancing are
undertaken completely separately in each sector. Beyond the established
decision-making organizations, other organizations have been given formal
rights to contribute to decision-making bodies by consultation (for example,
244 Diagnosis-Related Groups in Europe
nurses and allied health professions), participation and proposals (for example,
patient organizations) or by becoming a decision-making and fi nancing partner
in the process (for example, private health insurance for case-based payments
in hospitals).
Financing
Germany spends about 10.4 per cent of gross domestic product (GDP) on health
care, with the three main sources being statutory health insurance (57.5 per
cent of total expenditure on health), private health insurance (9.3 per cent) and
out-of-pocket spending (13.5 per cent) (DESTATIS 2009; data for 2007).
Since 2009, health insurance has been mandatory in Germany, while
previously it was only mandatory for around 75 per cent of the population
(while de facto over 99.5 per cent were covered). About 86 per cent of the
German population are covered by SHI and 10 per cent are privately insured
(with the remainder falling under special provisions). Premiums in private
health insurance are risk related. One can opt for insurance under this type of
health insurance if the earned income passes a certain threshold (€49 950 per
year or €4162.50 per month in 2010) for three consecutive years. The SHI
system is based on wage-related contributions (since 1 July 2009: 14.9 per cent
on gross income up to a threshold of €3750 per month).
14.1.2 Hospital services in Germany
In Germany one can distinguish between three different types of hospital
ownership. Almost half of all beds are found in public hospitals. In terms of the
remaining capacity, ~35 per cent is provided by non-profi t-making hospitals
and ~16 per cent by private profi t-making hospitals, which have increased their
share since the beginning of the 1990s. Table 14.1 summarizes the key statistics
for the German hospital sector.
Planning and ensuring hospital capacities
In the inpatient sector, the reimbursement of hospitals follows the principal
of ‘duality’ introduced with the Hospital Financing Act (KHG) in 1972. This
means that hospitals are fi nanced from two different sources: investments in
infrastructure are covered directly by state budgets, while operating costs are
reimbursed by sickness funds and private health insurance.
Each of the 16 state governments is responsible for maintaining hospital
infrastructure. The main instruments used to do so are the so-called ‘hospital
requirement plans’, which are set by the state governments after input by the
respective hospital federation and the sickness funds. They specify hospital
capacity and the range of services to be delivered across all hospitals within a
state, as well as within individual hospitals.
The self-governing bodies – namely, provider associations and sickness
funds – are responsible both for providing substantive detail to the provisions
of the laws defi ning the framework of hospital fi nancing, and for the continual
T
a
b
le
1
4
.1
K
ey
h
o
sp
it
al
fi
g
u
re
s
b
y
s
iz
e
an
d
o
w
n
er
sh
ip
,
2
0
0
7
Si
ze
a
n
d
t
yp
e
H
os
pi
ta
ls
B
ed
s
B
ed
s
pe
r
O
cc
u
pa
n
cy
C
a
se
s
C
a
se
s
pe
r
A
L
O
S*
of
o
w
n
er
sh
ip
(o
ve
ra
ll
)
1
0
0
0
0
0
1
0
0
0
0
0
in
h
a
b
it
a
n
ts
in
h
a
b
it
a
n
ts
N
u
m
b
er
N
u
m
b
er
N
u
m
b
er
%
N
u
m
b
er
N
u
m
b
er
D
a
ys
(s
h
a
re
i
n
%
)
(s
h
a
re
i
n
%
)
H
o
sp
it
a
l
si
z
e
i
n
b
e
d
s
2
0
8
7
5
0
6
9
5
4
6
1
6
7
7
.2
1
7
1
7
8
5
7
3
2
0
8
8
3
8
.3
(
1
0
0
)
(1
0
0
)
<
4
9
4
0
7
7
5
7
2
9
6
4
.9
2
1
0
0
2
8
2
5
5
8
.5
5
0
–9
9
2
6
4
1
9
3
5
4
2
4
7
3
.3
5
2
9
5
7
9
6
4
4
9
.8
1
0
0
–1
4
9
3
0
2
3
6
9
9
5
4
5
7
4
.2
1
1
0
8
2
8
5
1
3
4
7
9
.0
1
5
0
–1
9
9
2
0
8
3
5
9
0
3
4
4
7
4
.8
1
1
7
9
1
3
7
1
4
3
3
8
.3
2
0
0
–2
9
9
3
2
6
7
9
5
7
8
9
7
7
6
.1
2
6
1
2
2
8
8
3
1
7
6
8
.5
3
0
0
–3
9
9
2
0
3
6
9
6
1
3
8
5
7
7
.4
2
3
6
1
3
5
2
2
8
7
1
8
.3
4
0
0
–4
9
9
1
3
1
5
8
2
5
8
7
1
7
7
.6
1
9
5
3
5
9
8
2
3
7
5
8
.4
5
0
0
–5
9
9
9
6
5
2
5
4
5
6
4
7
7
.1
1
8
7
0
3
2
5
2
2
7
4
7
.9
6
0
0
–7
9
9
6
4
4
3
6
5
4
5
3
7
8
.8
1
5
6
4
8
0
0
1
9
0
2
8
.0
>
8
0
0
8
6
1
0
3
4
8
2
1
2
6
8
0
.7
3
7
8
9
1
8
4
4
6
0
6
8
.0
P
u
b
li
c
h
o
sp
it
a
ls
6
7
7
2
5
0
3
4
5
3
0
4
7
9
8
6
9
7
7
5
5
1
0
5
7
3
8
(
3
2
.4
)
(4
9
.4
)
u
n
d
er
p
ri
v
at
e
la
w
3
8
0
1
3
3
9
5
7
1
6
3
7
7
.5
4
8
0
4
9
1
4
5
8
4
1
7
.9
u
n
d
er
p
u
b
li
c
la
w
2
9
7
1
1
6
3
8
8
1
4
1
8
0
.5
3
8
9
2
8
4
1
4
7
3
2
8
.8
–
le
g
al
ly
d
ep
en
d
en
t
1
6
1
5
4
3
1
9
6
6
7
9
.5
1
7
5
5
5
7
6
2
1
3
4
9
.0
–
le
g
al
ly
i
n
d
ep
en
d
en
t
1
3
6
6
2
0
6
9
7
5
8
1
.4
2
1
3
7
2
6
6
2
5
9
8
8
.6
N
o
n
-p
ro
fi
t-
m
a
k
in
g
7
9
0
1
7
7
6
3
2
2
1
6
7
5
5
9
7
0
3
2
4
7
2
5
8
8
h
o
sp
it
a
ls
(
3
7
.9
)
(3
5
.0
)
P
ri
v
a
te
h
o
sp
it
a
ls
6
2
0
7
8
9
7
7
9
6
7
6
2
5
1
0
4
9
4
3
0
5
2
9
(
2
9
.7
)
(1
5
.6
)
So
u
rc
e:
B
ö
lt
(
2
0
1
0
),
w
it
h
m
o
d
ifi
c
at
io
n
s.
246 Diagnosis-Related Groups in Europe
development of the German diagnosis-related group (G-DRG) system. The
G-DRG system applies to all hospitals, irrespective of ownership status, and all
patients (except rehabilitation and psychiatric, psychosomatic or psychothera-
peutic patients), regardless of whether or not they are members of the SHI
system, have private health insurance, or are self-funding patients (Tuschen &
Trefz, 2004). DRGs cover all clinical departments with the exception of institu-
tions or facilities providing psychiatric care, psychosomatic medicine, or psy-
chotherapy services. For these services the 2009 Hospital Financing Reform Act
(KHRG) mandated the German self-governing bodies to develop and intro-
duce a prospective payment system by the year 2013, which is to be based on
per diem payments adjusted for patient characteristics and procedures.
Range of activities and services in the hospital sector
German hospitals concentrate on inpatient care because sectoral borders are still
strict compared with the practice in other countries. Legally, hospitals still mainly
provide inpatient services. Ambulatory care, including emergency care, is pro-
vided by the regional physicians’ associations and their offi ce-based physicians.
Only university hospitals have formal outpatient facilities, offi cially for research
and teaching purposes, while in most other hospitals, head physicians need to be
authorized by the physicians’ association if they (as individuals – and not the
hospital as an institution) want to provide ambulatory services.
Activity levels for day surgery and ambulatory pre- and post-hospital care
have increased. Since 2004, hospitals have been granted additional competences
to provide services to outpatients that require highly specialized care on a
regular basis. Also, participation in integrated care models (which require a
contract between a sickness fund and providers from different sectors) offers
new opportunities to become active in ambulatory care if their partners on the
providers’ side also include ambulatory care providers.
Nevertheless, hospital care remains clearly separate from outpatient care deliv-
ered by general practitioners (GPs) or specialists (Figure 14.1). A typical episode of
care starts with a referral including patient’s case history and preliminary diagno-
sis from a GP (or an offi ce-based specialist) to a hospital and ends with a discharge
or a transfer back to the GP (or specialist). Diagnostics (such as tests for cancer)
are carried out in outpatient as well as inpatient settings.
Relationship with third party payers
As outlined above, the principle of ‘dual fi nancing’ means that hospitals receive
funds for infrastructure from the state governments, while operating costs are
covered via DRGs by the sickness funds. Reimbursement for such costs is, to a
certain extent, limited by volumes which are negotiated between every hospital
accredited in the hospital plan and the sickness funds. If a hospital treats more
cases than negotiated, the DRG reimbursement rate is reduced by a certain
percentage (and vice versa – it is increased if the number of treated cases is lower).
Long-term infrastructural assets require a case-by-case grant application by
each individual hospital. State governments distinguish between grants for
Germany: Understanding G-DRGs 247
construction of hospitals and initial procurement or replacement of other
assets. According to the KHG, a hospital acquires a legal claim to subsidy only
as long as it is included in the ‘hospital plan’ of the respective state. Inclusion
in the hospital plan also means that fl at-rate grants for short-term assets (3–15
years economic life) can be granted. In practice, infrastructural hospital invest-
ments are mainly determined by the budgetary situation of the states and by
political considerations. If a hospital is not included in a ‘hospital plan’ it
cannot make a claim for state investment fi nancing. The share of public
investment in hospitals has decreased continuously since the early 1990s.
14.1.3 Purpose of the DRG system
The introduction of the G-DRG system sought to achieve several objectives.
First, the primary motive for fundamentally reforming the old reimbursement
system based on budgets with per diem charges as the unit for reimbursement
was to achieve a more appropriate and fair allocation of resources by utilizing
DRGs. Related goals were to facilitate a precise and transparent measurement of
the casemix and the levels of services delivered by hospitals. Moreover, it was
assumed that effi ciency and quality of service delivery in the hospital sector
would increase due to the improved documentation of internal processes and
increased managerial capacity. As a consequence, a moderate contribution to
cost-containment based on a reduction of length of stay and bed capacity was
presumed (Braun et al., 2007).
14.2 Development and updates of the DRG system
14.2.1 The current DRG system at a glance
The national G-DRG system was introduced in 2003, based on the Australian
Refi ned Diagnosis-Related Groups (AR-DRG, version 4.1). Outpatient services
Figure 14.1 Typical episode of care across sectoral borders
248 Diagnosis-Related Groups in Europe
are not covered by the G-DRG system. The system has evolved so that the
number of groups increased from 664 in 2003 up to 1200 in 2010. The procedure
to assign treatment cases to a DRG is based on a grouping algorithm using the
inpatient hospital discharge dataset, containing: major diagnosis and other
diagnoses, medical procedures, patient characteristics (age, gender and weight
of newborns), length of stay, duration of ventilation, reason for hospital dis-
charge and type of admission (for example, emergency, referral from GP or
transfer from other hospital). Specialized ‘grouper’ software assigns these data
to a particular DRG (see section 14.3). Each DRG is assigned to one of 25 major
diagnostic categories (MDCs) and has a fi xed cost weight which is calculated by
the Institute for the Hospital Remuneration System (InEK) based on average
costs as documented by a sample of hospitals.
14.2.2 Development of the DRG system
In 2000, the Statutory Health Insurance Reform Act paved the way for the
G-DRG system. It represented the most signifi cant reform of the German
hospital sector since the system of ‘dual fi nancing’ was introduced in 1972 by
the KHG. The reform defi ned the fundamental features of the G-DRG system
for case-based reimbursement of inpatient services. However, under this
provision, the self-governing bodies at the federal level (that is, the Federal
Association of Sickness Funds, the Association of Private Health Insurance, and
the German Hospital Federation) were mandated to select (by June 2000) and
then to introduce a DRG-based reimbursement system themselves. As a guiding
principle they were required to ensure that the system would be guided by
universal and uniform application, performance orientation and case payments,
taking account of disease severity and case complexity. In June 2000 the German
self-governing bodies decided to use the AR-DRG system as the foundation for
the G-DRG system.
Four phases can be distinguished in the G-DRG introduction process (Figure
14.2): fi rst, the preparation phase, from 2000 until 2002, in which the selected
AR-DRG system was adapted to the German hospital environment in two major
steps, as detailed here.
1. The Australian procedure codes based on the WHO’s International
Classifi cation of Diseases ICD-9-CM (clinical modifi cation) were transformed
to the German procedure classifi cation codes (OPS) and the ICD-10-WHO
diagnosis codes were modifi ed to the ICD-10-GM (German modifi cation) by
the German Institute for Medical Documentation and Information (DIMDI).
2. A cost-accounting system for calculating Germany-specifi c relative cost
weights was developed by the InEK. The institute was founded for this
purpose by the self-governing bodies. In 2001 a small set of hospitals tested
the Australian grouper. The results were discussed in 2002 and requirements
for a German system were derived. By the end of 2002 the fi rst version of the
G-DRG system had been prepared. For this early version, approximately 100
hospitals (of ~1800 acute hospitals falling under the DRG system) voluntarily
shared their cost data with the InEK to calculate cost weights. Version 1 of
the G-DRG system included 664 DRGs in the Case Fee Catalogue.
Germany: Understanding G-DRGs 249
The second phase from 2003 until 2004 was the introduction of DRGs. This
phase was called the budget-neutral phase, as hospitals were receiving the bud-
gets as negotiated previously. The only difference was that the reimbursement
units were no longer per diem charges, but were the DRGs instead. In 2003,
hospitals could voluntarily group their patients using G-DRGs (incentivized
by the option to be able to negotiate higher budgets), then in 2004 they were
mandated to do so. In order to change from a budget based on per diem pay-
ment to one based on DRGs, it was necessary to transform the historically
developed budgets into ‘DRG budgets’ (‘revenue budgets’). This involved defi n-
ing cost categories within ‘DRG budgets’ as additional activities by hospitals
which continued to be reimbursed differently (for example, psychiatric ser-
vices, teaching of nursing students).
Whereas until 2002 the budget was based on the agreed number of patient
days to calculate the per diem charge, the budget in 2003/2004 was based on its
casemix (that is, the number of relative weights for all patients) to give the
hospital-specifi c base rate. For the fi rst time in the German hospital sector,
hospital effi ciency became visible as it became apparent which hospitals with a
high base rate (due to budgets set comparatively high for the patient casemix)
produced the same services comparatively less effi ciently than those with low
base rates. ‘Casemix’ and the ‘casemix index’ (CMI) have become common
terms in comparing hospitals. The casemix is equal to the sum of the cost
weights of all DRGs for a specifi ed time period. The average case weight or so
called CMI is calculated by dividing the casemix by the total number of cases.
The CMI is thus equal to the average DRG cost weight for a particular hospital
and is an important indicator of the costliness of cases treated by a particular
hospital. Small rural hospitals typically have CMIs of well below the average
of 1, while university hospitals may have CMIs above 1.5.
During the third phase of convergence from 2005–2010, hospitals’ individual
base rates converged to state-wide base rates (one for each of the 16 Länder). As
a starting point, state-wide base rates were negotiated for the fi rst time in 2005.
Figure 14.2 Phases involved in introducing DRGs in Germany
Source: Neubauer & Pfi ster, 2008, with modifi cations.
250 Diagnosis-Related Groups in Europe
These were used as a yard stick for the base rates of all hospitals in that state.
While hospital budgets (or rather revenue budgets) were still negotiated and
used to calculate hospital-specifi c base rates, the actual base rate used for each
hospital diverged year by year from the (calculated) hospital-specifi c base rate
to approach the state-wide base rate. In 2005, the individual base rate was
determined by 15 per cent of the difference to the state-wide base rate, in 2006
by 35 per cent (15 per cent plus 20 per cent), and so on, until in 2009 it was
meant to reach the state-wide base rate (Figure 14.3).
Initially, hospital-specifi c base rates varied considerably from ~€2200 (mostly
minor hospitals in rural areas) up to ~€3200 (for major hospitals in urban
areas), which to some extent refl ected historical differences in their reimburse-
ment negotiations (Friedrich et al., 2008). As the G-DRG system does not
account for organizational characteristics – such as size, differences in input
prices or the teaching status of a hospital – the convergence of the base rate put
high-cost hospitals under signifi cant pressure to lower costs.
To make the reform politically more acceptable, resulting losses of the
negotiated budget were limited, initially to 1 per cent in 2005 (compared to
2004), but then increasing up to 3 per cent in 2009 (compared to 2008). As a
result, not all hospitals with initially high hospital-specifi c base rates had
reached the state-wide levels by 2009. In 2010, however, there was no safety net
for losses so that the state-wide base rates were applied to all hospitals (and
hospital-specifi c base rates consequently ceased to exist) (Figure 14.3).
With the fourth phase from 2010/11 onwards, further modifi cations of the
G-DRG system are planned. Among them are:
• From 2010 onewards, a nationwide base rate will be calculated by the InEK.
Until 2014 state-wide base rates should converge towards a target corridor of
2.5% above and 1.25% below this rate.
Figure 14.3 Phase of convergence
Source: Neubauer & Pfi ster, 2008, with modifi cations.
Germany: Understanding G-DRGs 251
• The 2009 KHRG gave the state governments the opportunity to include the
investment costs in the cost calculation of the DRGs. This would result in
some states having a single payer approach to hospital reimbursement.
Currently, however, it is not clear how the money paid by the states for
hospital investment will be channelled into the system.
• Psychiatric services will also be reimbursed by a DRG-like system. This will
probably differ from the rest of the system by being a combination of length
of stay and resource intensity; that is, the case weights will be calculated on a
per diem basis.
Table 14.2 summarizes the main characteristics of the G-DRG system and
changes over time. Two developments stand out: (1) the sample for calculating
cost weights was substantially increased. Since 2004, an increasing number of
major and university hospitals with severe and rare cases have participated; (2)
the number of DRGs and supplementary fees (mostly used for the reimbursement
of high-cost drugs) increased dramatically as new DRGs were added and existing
ones were split.
14.2.3 Data used for the development and updates
of the DRG system
Three types of information are important for the development of the G-DRG
system: (1) adequate coding of clinical data, both to further develop the
grouping system and to facilitate precise reimbursement that takes account of
individual patient characteristics (reimbursement of individual hospitals); (2)
cost data to calculate cost weights; and (3) information on medical innovations
that allows regular updates of fee catalogues.
To calculate cost weights, the InEK relies on retrospective cost and performance
data collected in German hospitals (Table 14.2 and Figure 14.4). All German
hospitals are obliged to provide hospital-related structural data (relating to type
of hospital, ownership, number of beds, number of trainees, labour and total
costs) and case-related performance data (regarding diagnoses, procedures,
reason for admission, date of discharge) on an annual basis (§21 Hospital
Remuneration Act (KHEntG)) to the Data Centre.
Additionally, hospitals can participate voluntarily in the sample used to calcu-
late cost weights (section 14.4). In order to do so, they must provide patient-
level cost data, submitted to the InEK. To achieve uniform and comparable cost
data, the InEK has developed a standardized cost-accounting system based on a
‘Calculation Handbook’ (InEK, 2007). Each year up to the end of March the
hospitals must deliver all datasets of the previous year to the Data Centre
(operated from 3M Medica). After data checks (see subsection 14.3.3), the InEK
receives the data before 1 July in order to develop the Case Fee Catalogue for the
following year. For example, the G-DRG system for 2010 is based on retrospective
cost and structural data from the 2008 calendar year, while 2009 was used to
check the data on plausibility and recalculate the cost weights.
The third type of information is needed for the introduction of new diagnostic
and treatment options within the OPS, maintained and developed by the
DIMDI (subsection 14.2.2). The DIMDI has developed a process by which
institutions such as the InEK, the Federal Offi ce for Quality Assurance (BQS)
T
a
b
le
1
4
.2
M
ai
n
f
ac
ts
r
el
at
in
g
t
o
t
h
e
G
-D
R
G
s
y
st
em
Y
ea
r
2
0
0
3
2
0
0
4
2
0
0
5
2
0
0
6
2
0
0
7
2
0
0
8
2
0
0
9
2
0
1
0
D
R
G
s
to
ta
l
6
6
4
8
2
4
8
7
8
9
5
4
1
0
8
2
1
1
3
7
1
1
9
2
1
2
0
0
In
p
a
ti
e
n
t
D
R
G
s
to
ta
l
6
6
4
8
2
4
8
7
8
9
5
2
1
0
7
7
1
1
3
2
1
1
8
7
1
1
9
5
–
v
al
u
at
ed
(
B
2
*)
6
4
2
8
0
6
8
4
5
9
1
2
1
0
3
5
1
0
8
9
1
1
4
6
1
1
5
4
–
u
n
v
al
u
at
ed
(
D
1
*)
2
2
1
8
3
3
4
0
4
2
4
3
4
1
4
1
R
an
g
e
o
f
co
st
w
ei
g
h
ts
:
m
in
.-
m
ax
.(
ro
u
n
d
ed
)
0
.1
2
–2
9
.7
1
0
.1
1
–4
8
.2
7
0
.1
2
–5
7
.6
3
0
.1
2
–6
5
.7
0
0
.1
1
–6
4
.9
0
0
.1
1
–6
8
.9
7
0
.1
2
–7
8
.4
7
0
.1
3
–7
3
.7
6
D
a
y
c
a
re
D
R
G
s
to
ta
l
0
0
0
2
5
5
5
5
–
v
al
u
at
ed
(
B
2
*)
0
0
0
1
1
1
1
1
–
u
n
v
al
u
at
ed
(
D
3
*)
0
0
0
1
4
4
4
4
S
u
p
p
le
m
e
n
ta
ry
f
e
e
s
0
2
6
7
1
8
3
1
0
5
1
1
5
1
2
7
1
4
3
–
v
al
u
at
ed
(
C
1
*)
0
1
3
5
4
1
5
9
6
4
7
4
8
1
–
u
n
v
al
u
at
ed
(
D
2
*)
0
2
5
3
6
4
2
4
6
5
1
5
3
6
2
H
o
sp
it
a
ls
p
a
rt
ic
ip
a
ti
n
g
i
n
c
o
st
d
a
ta
c
o
ll
e
c
ti
o
n
1
2
5
1
4
4
1
4
8
2
1
4
2
6
3
2
4
9
2
5
1
2
5
3
–
ex
cl
u
d
ed
f
o
r
d
at
a
q
u
al
it
y
9
0
0
0
3
8
2
8
3
3
2
8
–
ac
tu
al
1
1
6
1
4
4
1
4
8
2
1
4
2
2
5
2
2
1
2
1
8
2
2
5
–
in
cl
u
d
ed
u
n
iv
er
si
ty
h
o
sp
it
al
s
0
1
2
1
0
9
1
0
8
1
0
1
0
–
n
u
m
b
er
o
f
ca
se
s
av
ai
la
b
le
f
o
r
ca
lc
u
la
ti
o
n
6
3
3
5
7
7
2
8
2
5
6
5
0
2
9
0
9
7
8
4
3
5
3
1
7
6
0
4
2
3
9
3
6
5
3
9
0
0
0
9
8
4
3
7
7
0
2
1
4
5
3
9
7
6
3
–
n
u
m
b
er
o
f
ca
se
s
u
se
d
f
o
r
ca
lc
u
la
ti
o
n
a
ft
er
d
at
a
ch
ec
k
s
4
9
4
3
2
5
2
3
9
5
4
1
0
2
2
8
3
8
7
4
2
8
5
1
8
1
9
2
8
6
3
1
1
5
2
8
1
1
6
6
9
3
0
7
5
3
7
8
3
2
5
7
4
9
7
R
2
a
ll
c
as
es
0
.4
5
5
6
0
.5
5
7
7
0
.6
3
8
8
0
.6
8
0
5
0
.7
0
7
2
0
.7
2
0
9
0
.7
4
4
0
.7
4
4
3
R
2
in
li
er
0
.6
2
1
1
0
.7
0
2
2
0
.7
7
9
6
0
.7
8
8
4
0
.8
0
4
9
0
.8
1
6
6
0
.8
3
4
5
0
.8
4
3
So
u
rc
e:
B
as
ed
o
n
d
at
a
fr
o
m
I
n
E
K
a
n
n
u
al
r
ep
o
rt
s.
*
H
o
sp
it
al
r
ei
m
b
u
rs
em
en
t
ac
co
rd
in
g
t
o
F
ig
u
re
1
4
.6
Germany: Understanding G-DRGs 253
(see subsection 14.5.4) and other professional (medical) associations can submit
suggestions to be considered for classifi cation (both within the OPS as well as
within the ICD-10-GM). All proposals are discussed and evaluated and further
refi ned in different working groups. Successful proposals result in a new or
modifi ed code. Both the OPS and the ICD are updated annually. New
technologies are incorporated sequentially and appended to existing medical
coding catalogues (see section 14.6).
The InEK is obliged to take the latest medical knowledge into account when
developing the DRG catalogue. Therefore, the InEK developed a proposal
process (structured dialogue) whereby medical experts are asked to contribute
their knowledge from clinical practice in order to refi ne certain DRGs. After
collecting the suggestions from clinicians, the InEK carries out statistical
analysis to prove the proposals empirically. About 37 per cent of the proposals
that were able to be tested empirically (410 out of 700) were implemented to
the G-DRG 2010 version (InEK, 2009a).
14.3 The current patient classifi cation system
14.3.1 Information used to classify patients
Diagnoses and medical procedures are the most important information used to
assign patients to a certain G-DRG. The ICD-10-GM is used to code diagnoses.
To code procedures, the OPS is used to assign a specifi c code to most procedures.
Figure 14.4 Types of data used for reimbursement and further development of the
G-DRG system
254 Diagnosis-Related Groups in Europe
Although the OPS originally contained procedure codes only for inpatient
surgical interventions, it has been used to code both these and general inpatient
medical procedures since 2004 and thus plays a key role in the implementation
of DRGs. Since 2005, ambulatory surgical procedures have also been included
in the OPS; it is thus also used in the ambulatory care sector, in which many
such surgical procedures are carried out. In addition to its role in the G-DRG
system, the OPS is designed to facilitate quality assurance (see subsection 14.5.4)
and the uptake of new technologies (see section 14.6).
14.3.2 Classifi cation algorithm
A simplifi ed version of the grouping process is presented in Figure 14.5. In cases
with extremely high resource consumption, certain procedure codes (for
example, transplantation) determine the DRG directly. The DRGs in this category
are referred to as ‘Pre-MDC’ DRGs. For all others, the major diagnosis determines
the classifi cation into one of 25 MDCs, numbered 1 to 23 (with 18 and 21 each
split into A and B). Essentially, an MDC corresponds to diseases of the body
system comparable to the classifi cation in ICD. While all DRGs relating to the
‘Pre-MDC’ start with an A, the 25 MDCs use a starting letter between B and Z, for
Figure 14.5 G-DRG grouping algorithm
Source: Updated and modifi ed from Schreyögg et al., 2006.
Germany: Understanding G-DRGs 255
example, MDC 1 (Nervous system) = B or MDC 14 (pregnancy, childbirth and
puerperium) = P.
After this step, data on the type of procedure are used to assign a case to a
‘base-DRG’, which is a group of closely related diagnoses and procedures that
have not been subdivided according to criteria such as co-morbidities or patient
age. Within each MDC, base-DRGs have a two-digit number, which also shows
the ‘partition’ of the DRG, with 01 to 39 for surgical DRGs (for example, B01–
B39 for diseases of the nervous system with surgery), 40 to 59 for DRGs with
other important procedures which are essential for the DRG, and 60 to 99 for
other DRGs. Since the 2005 system, the strict partitioning has been relaxed in
MDC 5 (Circulatory System) and MDC 8 (Musculoskeletal System and
Connective Tissue) so that DRGs above 39 can also contain surgical procedures.
Base-DRGs may be split into separate DRGs based on additional criteria, thus
refl ecting different degrees of resource consumption. A case is subsequently
assigned to its fi nal DRG (which is either a base-DRG that has not been split, or
one of at least two – but usually more – as a result of splitting) using information
such as co-morbidities, procedures and patient characteristics on the one hand,
and cost data on the other. If a base-DRG is not split, the fourth digit (again a
letter) is a Z, for example, B01Z, while split DRGs use A, B, C and so on in
descending order of resource intensity, such as B02A > B02B > B02C.
14.3.3 Data quality and plausibility checks
Cost data
Initially, the Data Centre (see Figure 14.4) checks the cost datasets for formal
and technical errors. As part of this process the fi le compatibility and data
encryption, as well as the existence of service and cost data in every dataset are
validated. Cases without DRG relevance (such as psychiatry) are excluded.
Next, the InEK conducts three further steps consisting of economic and medical
plausibility checks. First, minimum and maximum costs per module (such as
costs of the clinical staff per day, total cost of the hospital) and the ratios
between modules (such as costs of the cost centre ‘anaesthesia’ < costs of the
cost centre ‘operating room’) are given an economic check. Second, adherence
to the German DRG classifi cation codes (ICD-10-GM and OPS) is given a
medical check, and third, coherence between economic and medical infor-
mation is checked (for example, the costs per case of a hip replacement must
refl ect the material cost of implants; if radiology procedures are reported, the
costs must be part of cost centre 9 ‘radiology’, see Table 14.3). In 2009, after
these data plausibility checks, 3 257 497 out of 4 539 763 records were available
(~72 per cent) for the calculation (InEK, 2009a). The datasets that remain serve
as the basis for determining the cost weights and trim-points.
Clinical data
For reimbursement purposes, every hospital must deliver case data (§301 SGB
V) to the sickness funds, mainly comprising clinical data (diagnoses, procedures),
demographic data (age, gender) and administrative data (dates of admission,
256 Diagnosis-Related Groups in Europe
surgery and discharge). The coding quality of these data is regularly checked by
the regional medical review boards of the sickness funds. They evaluate the
assignment of cases to DRGs and their respective service utilization (§275 SGB
V; §17 KHG). In order to do so, they send teams to randomly selected hospitals
which have to disclose their medical and coding practices. In instances where
unintended up-coding is revealed, the hospitals must reimburse the sickness
funds for the respective revenues that they gained through up-coding. If it is
demonstrated that hospitals intentionally used up-coding as a means to increase
profi ts, then in addition to their reimbursement fee they are required to make a
penalty payment equal to the sum of their reimbursement fee. In 2009, 12
percent of all hospital cases (~1.7 million cases) were audited by the sickness
funds, resulting in average claw-back amounts of about €850 per audited case
(MDS, 2011).
14.3.4 Incentives for up- or wrong-coding
Up-coding, wrong-coding
The revenues of a German hospital depend on the number and value of the
services delivered. This may incentivize hospitals to encode more or higher
reimbursed services than actually delivered. The medical review board of the
sickness funds tries to detect this up- or wrong-coding by reviewing individual
cases which are randomly selected, as already described.
Cream-skimming or cherry-picking
Adverse selection is contrary to the function and maintenance mission of
hospitals, especially in rural areas. As the Case Fee Catalogue is updated annually
to refl ect current costs for inpatient treatments, it represents a systemic
(inherent) method to prevent cherry-picking as cost weights differ from one
year to the next. This approach makes it impossible to predict DRG contribu-
tion margins for certain treatments in the long run and reduces incentives to
adjust capacities accordingly, especially as the delivery of specifi c hospital
services often depends on special infrastructure and may require organizational
change.
Inappropriate early discharge
The risks of early discharge in order to cut costs have been well documented
ever since DRG systems were fi rst introduced. The G-DRG system tries to avoid
early discharge by the application of two major instruments. First, the annual
update of the Case Fee Catalogue and the recalculation of cost weights and
trim-points for the reimbursement of outliers (section 14.5) are designed to
reduce incentives for early discharge by reimbursing adequately for expensive
services, as well as deducting payments for short-stay outliers. Second, read-
missions for the same cause within 30 days after discharge are reimbursed by
the original DRG (§2 Case Fee Agreement (FPV) 2010) and receive no additional
Germany: Understanding G-DRGs 257
funds. This approach fi nancially penalizes inappropriate early discharge (at
least if it leads to readmission).
14.4 Cost accounting within hospitals
14.4.1 Regulation
Cost accounting within hospitals is neither obligatory nor directly regulated in
Germany. However, the introduction of the G-DRG system required medical
and cost-controlling systems to be implemented in order to control for their
resource consumption and the level of services delivered. Medical accounting is
a separate administrative unit in nearly every hospital in Germany. Medical
controllers (mostly physicians with further education in coding) examine
hospital cases in terms of correct coding to avoid a review by the sickness funds
and to maximize revenue. In addition, patient-level cost accounting is in-
creasingly applied to monitor cost structures and sources of resource waste. In
order to calculate cost weights, the InEK established a sample of hospitals that
voluntarily collect patient-level cost data (InEK, 2009a). Only hospitals that can
deliver cost data to a standard defi ned by the InEK (in the Calculation Handbook)
are eligible to participate. The extra effort is reimbursed via an additional fee,
which consists of a lump sum and a variable amount related to the number of
delivered cases and their data quality. In 2008 the InEK spent €9 million to
compensate hospitals for their additional efforts.
14.4.2 Main characteristics of the cost-accounting system
In this section we focus on hospitals that follow the cost-accounting standards
specifi ed by the InEK, as the cost-accounting characteristics of other hospitals
do not affect DRG calculation and differ widely. The participating hospitals
must meet certain cost-accounting standards. They must calculate costs per
case according to the full cost method, using actual costs. This means that all
DRG-related costs must be taken into account when calculating the costs of
DRG treatment cases. The actual costs are derived from the hospitals’ audited
annual accounts. Accordingly, the reference period for calculating costs per
case is an entire calendar year. The intention is that participating hospitals use
step-down cost accounting. However, if this is not feasible they are also allowed
to use a mixed calculation (using step-down cost accounting, with gross- (or
top-down) costing as a second option), or even make use of a kind of gross-
costing when necessary. When calculating costs per case, the only costs to be
taken into consideration are those that arise due to the performance of the
DRG-related services. The following cost elements are excluded:
• extraordinary expenses and expenses relating to other periods;
• investment costs;
• core business expenses, insofar as these are not related to general inpatient
services (for example, costs of scientifi c research/teaching and costs of psy-
chiatric and outpatient services are excluded);
258 Diagnosis-Related Groups in Europe
• taxes, charges, insurance for operational sections of the hospital that do not
provide general inpatient services, as well as tax on profi ts;
• specifi c and long-term allowance for bad debts;
• interest payable, insofar as this is not related to capital loans;
• imputed costs (for example, hospital building).
The process of calculating costs per case is based on a modular approach,
which is detailed in Table 14.3 (InEK, 2007). It entails arranging each set of case-
related data in the calculation according to cost-element groups and cost-centre
groups. Aggregating costs across cost-element groups and cost-centre groups
makes it possible to identify the costs per patient or per patient group (DRGs).
14.5 DRGs for reimbursement
14.5.1 Range of services and costs included in DRG-based
hospital payments
Figure 14.6 outlines the inpatient reimbursement components used in Germany.
In the Case Fee Catalogue for 2010, there are 1155 DRGs with national uniform
cost weights (B2), 45 DRGs without national cost weights (D1 & D3), and 143
sup plementary fees (C1 & D2) (see Table 14.2). The DRGs without national cost
weights (D1 & D3) are individually negotiated with each hospital as they were
excluded from the DRG national cost weights because their sample size was
insuffi cient for calculation, or their cost variance was too large. G-DRGs are
intended to cover medical treatment, nursing care, the provision of pharmaceu-
ticals and therapeutic appliances, as well as board and accommodation.
Supplementary fees cover certain complex or cost-intensive services, and/or
very expensive drugs. The supplementary fees are used due to a lack of suffi cient
data for calculating costs for certain DRGs, and the limited appropriateness (in
terms of refl ecting actual costs incurred) of the current cost weights (InEK,
2009a). These supplementary fees are generally calculated in a uniform manner
across Germany. Since the introduction of supplementary fees in 2004, their
number has increased from 26 to a total of 143 individual fees in 2010. These
include 81 supplementary fees, whereby the amounts were fi xed at the national
level in the 2010 DRG Case Fee Catalogue (C1). The other 62 treatment services
were included in a sub-list of supplementary fees in the Case Fee Catalogue that
are to be negotiated on a hospital-by-hospital basis (D2).
In addition, the contracting parties are authorized to negotiate additional
reimbursement by means of case-based or per diem remuneration for highly
specialized services if it can be proved that the service in question cannot yet be
appropriately reimbursed through DRGs or supplementary fees. There are also a
number of surcharges which are negotiated between the contracting parties and
are especially relevant for hospitals that are using new and innovative treatment
options. For instance, it is possible to negotiate surcharges for innovative
diagnostic and treatment procedures (E1; see section 14.6) and even to exclude
certain special facilities and hospital departments completely from the G-DRG
system, fi nancing them instead through individually negotiated fees (for further
T
a
b
le
1
4
.3
G
-D
R
G
m
o
d
u
la
r
co
st
in
g
a
p
p
ro
ac
h
C
os
t-
el
em
en
t
gr
ou
ps
L
a
b
ou
r*
M
a
te
ri
a
l†
In
fr
a
st
ru
ct
u
re
‡
C
o
st
-
C
e
n
tr
e
G
ro
u
p
s
1
2
3
4
a
4
b
5
6
a
6
b
7
8
H
o
sp
it
a
l
u
n
it
s
w
it
h
b
e
d
s
1
:
N
o
rm
al
w
ar
d
2
:
In
te
n
si
v
e
ca
re
u
n
it
3
:
D
ia
ly
si
s
u
n
it
1
.1
2
.1
3
.1
1
.2
2
.2
2
.3
1
.3
2
.3
3
.3
1
.4
a
2
.4
a
3
.4
a
1
.4
b
2
.4
b
3
.4
b
-
2
.5 -
1
.6
a
2
.6
a
3
.6
a
1
.6
b
2
.6
b
3
.6
b
1
.7
2
.7
3
.7
1
.8
2
.8
3
.8
D
ia
g
n
o
st
ic
a
n
d
t
re
a
tm
e
n
t
a
re
a
s
4
:
O
p
er
at
in
g
r
o
o
m
5
:
A
n
ae
st
h
es
ia
6
:
M
at
er
n
it
y
r
o
o
m
7
:
C
ar
d
ia
c
d
ia
g
n
o
st
ic
s/
th
er
ap
y
8
:
E
n
d
o
sc
o
p
ic
d
ia
g
n
o
st
ic
s/
th
er
ap
y
9
:
R
ad
io
lo
g
y
1
0
:
L
ab
o
ra
to
ri
es
1
1
:
O
th
er
d
ia
g
n
o
st
ic
a
n
d
t
h
er
ap
eu
ti
c
ar
ea
s
4
.1
5
.1
6
.1
7
.1
8
.1
9
.1
1
0
.1
1
1
.1
- - - - - - -
1
1
.2
4
.3
5
.3
6
.3
7
.3
8
.3
9
.3
1
0
.3
1
1
.3
4
.4
a
5
.4
a
6
.4
a
7
.4
a
8
.4
a
9
.4
a
1
0
.4
a
1
1
.4
a
4
.4
b
5
.4
b
6
.4
b
7
.4
b
8
.4
b
9
.4
b
1
0
.4
b
1
1
.4
b
4
.5 - -
7
.5
8
.5
9
.5
1
0
.5
1
1
.5
4
.6
a
5
.6
a
6
.6
a
7
.6
a
8
.6
a
9
.6
a
1
0
.6
a
1
1
.6
a
4
.6
b
5
.6
b
6
.6
b
7
.6
b
8
.6
b
9
.6
b
1
0
.6
b
1
1
.6
b
4
.7
5
.7
6
.7
7
.7
8
.7
9
.7
1
0
.7
1
1
.7
4
.8
5
.8
6
.8
7
.8
8
.8
9
.8
1
0
.8
1
1
.8
K
ey
:
*1
=
L
ab
o
u
r
co
st
s
o
f
th
e
o
th
er
m
ed
ic
al
s
ta
ff
;
2
=
L
ab
o
u
r
co
st
s
o
f
th
e
n
u
rs
in
g
s
ta
ff
;
3
=
L
ab
o
u
r
co
st
s
o
f
th
e
ad
m
in
is
tr
at
iv
e
an
d
t
ec
h
n
ic
al
s
ta
ff
† 4
a
=
D
ru
g
c
o
st
s;
4
b
=
D
ru
g
c
o
st
s
(i
n
d
iv
id
u
al
c
o
st
s/
ac
tu
al
c
o
n
su
m
ti
o
n
);
5
=
C
o
st
s
o
f
im
p
la
n
ts
a
n
d
g
ra
ft
s;
6
a
=
M
at
er
ia
l
co
st
s
(w
it
h
o
u
t
d
ru
g
s,
i
m
p
la
n
ts
a
n
d
g
ra
ft
s)
;
6
b
=
M
at
er
ia
l
co
st
s
(i
n
d
iv
id
u
al
c
o
st
s/
a
ct
u
al
c
o
n
su
m
p
ti
o
n
,
w
it
h
o
u
t
d
ru
g
s,
i
m
p
la
n
ts
a
n
d
g
ra
ft
s
‡
7
=
M
ed
ic
al
i
n
fr
as
tr
u
ct
u
re
c
o
st
s;
8
:
N
o
n
-m
ed
ic
al
i
n
fr
as
tr
u
ct
u
re
c
o
st
s.
So
u
rc
e:
I
n
E
K
,
2
0
0
7
,
w
it
h
m
o
d
ifi
c
at
io
n
s.
F
ig
u
re
1
4
.6
R
ei
m
b
u
rs
em
en
t
co
m
p
o
n
en
ts
o
f
in
p
at
ie
n
t
ca
re
i
n
G
er
m
an
y
So
u
rc
e:
U
p
d
at
ed
a
n
d
m
o
d
ifi
e
d
f
ro
m
S
ch
re
y
ö
g
g
e
t
al
.,
2
0
0
6
.
N
ot
es
:
a
E
x
ce
p
ti
o
n
:
cl
as
si
fi
ca
ti
o
n
a
s
a
sp
ec
ia
l
fa
ci
li
ty
(
FP
V
B
E
2
0
0
9
)†
;
b
O
n
ly
r
ei
m
b
u
rs
em
en
t
o
f
th
e
ad
d
it
io
n
al
i
n
te
g
ra
te
d
c
ar
e
se
rv
ic
e
w
h
ic
h
i
s
n
o
t
co
v
er
ed
b
y
th
e
h
o
sp
it
al
b
u
d
g
et
(
§
1
4
0
d
A
b
s.
4
S
G
B
V
)
†
C
as
e
Fe
e
A
g
re
em
en
t
fo
r
sp
ec
ia
l
fa
ci
li
ti
es
(
FP
V
B
E
),
u
p
d
at
ed
a
n
n
u
al
ly
.
Germany: Understanding G-DRGs 261
details, including the function of the revenue budget, see Busse and Riesberg,
2004)). Including other reimbursement components, for example for individuals
accompanying patients (A2) or quality assurance (A3), all reimbursement
components besides the uniformly weighted DRGs (B1–B3) currently account for
approximately 20 per cent of the total reimbursement for non-psychiatric
inpatient care. This remains so even though the political aim is to reimburse
hospitals solely through uniformly weighted DRGs.
14.5.2 Calculation of DRG prices/cost weights
In the G-DRG system cost weights are calculated, which defi ne a relationship
between the different DRG groups according to resource intensity. Using this
framework, the price for the reference treatment group with cost weight 1.0 is
equal to the base rate (average costs) and the prices for all other DRGs are
calculated by multiplying the DRG cost weight attached to each DRG with the
price set for the reference DRG cost weight of 1.0. The cost weight of each DRG
group refl ects the resource consumption relative to the reference DRG, which
adjusts prices for resources.
Trimming methods
The InEK applies a mathematical trimming method to account for extreme
cases (InEK, 2004). Because DRG systems attempt to translate inpatient cases
into medically coherent and cost-homogeneous groups, outliers are excluded
for the calculation of cost weights. The term ‘inlier’ denotes cases that are
treated within a length-of-stay interval. This is demarcated by a low trim-point
and a high trim-point, between which the average treatment cases are located
(Figure 14.7). Therefore, after data have been refi ned with plausibility checks,
the average costs of inlier cases are determined for each DRG. To determine the
Figure 14.7 Deductions and surcharges related to the length of stay
262 Diagnosis-Related Groups in Europe
cost weight for each DRG, the average costs of inlier cases for the DRG in
question are divided by the reference value for the respective year. The reference
value, defi ned as the arithmetic mean costs of all inlier cases, is calculated as the
sum of DRG-relevant costs (section 14.4.2) divided by the sum of the effective
casemix across Germany. The reference value used to develop the Case Fee
Catalogue 2010 was €2619.10 (InEK, 2009a).
14.5.3 DRGs in actual hospital payment
The conversion from cost weights into actual reimbursement rates is given by
multiplying the applicable base rate by the DRG specifi c cost weight (Figure
14.8). The calculation of cost weights is described in subsection 14.5.2.
14.5.4 Quality-related adjustments
The current G-DRG-system does not adjust reimbursement for quality. As
reimbursement is based on average treatment costs, hospitals with a higher-
than-average cost level are incentivized to cut expenditure. This can adversely
affect quality as hospitals may reduce quality without incurring reimbursement
penalties. To address incentives to increase profi ts without consideration of
quality implications, the legislator introduced regulatory measures, such as
mandatory quality reports, external quality assurance, quality management
system(s) (QMS) and minimum volume thresholds (§137 SBG V).
Quality reports
In 2002, the Case Fees Act (FPG) introduced hospital quality reports to simplify
comparisons between hospitals and to support physicians and sickness funds in
advising patients regarding elective hospital treatments. Since 2005, hospitals
have been obliged to submit quality reports every second year following a
structure mandated by a directive of the Federal Joint Committee (G-BA). The
reports are available publicly, online.
External quality assurance
Since the SHI Reform Act of 2000, hospitals have been obliged to participate in
an external and comparative quality assurance programme developed by the
Figure 14.8 Components of G-DRG reimbursement and G-DRG implementation
Germany: Understanding G-DRGs 263
BQS. This programme surveys treatment-related quality indicators and compares
them nationally. From 2001 to 2009, the BQS has published an annual quality
report detailing the results of the hospitals, which are not named. The BQS
methodology has been criticized because of the extra effort involved for
hospitals to obtain data which are not part of routine datasets. From 2010
onwards, the AQUA-Institute for Applied Quality Improvement and Research
in Health Care is charged with further developing and implementing the
external quality assurance programme.
Quality management systems
In 1999 the legislator introduced §135a of the SGB V, obliging hospitals to
launch and further develop a QMS. Hospitals have a free choice of which kind
of QMS they set up. Therefore, a wide range of different QMS from simple
(Cooperation for Transparency and Quality in Health Care) to more sophisticated
(Joint Commission) systems were introduced across Germany. However, most
patients are not able to distinguish between different quality certifi cates, which
led to confusion instead of clarifi cation on the part of patients.
Minimum volume thresholds
In addition to the quality reports, the FPG enacted an ordinance for defi ning
minimum volumes as thresholds to deliver certain (particularly elective) services
whereby the outcome is related to the volume of services delivered. In order to
determine these services, the G-BA is charged with developing a catalogue that
defi nes the minimum number of delivered services per physician or hospital
(Velasco-Garrido & Busse, 2004). Hospitals which do not reach the required
volume of services may not deliver the service. Since 2004, the catalogue has
contained six elective services (with the annual minimum number per hospital
shown in parentheses): liver transplantation (20), kidney transplantation (25),
complex procedures on the oesophagus (10), complex procedures on the
pancreas (10), stem cell transplantation (25) and knee replacement (50).
14.5.5 Main incentives for hospitals
Under the G-DRG system, hospitals are generally not incentivized to improve
their medical outcomes (see subsection 14.3.4). However, within the G-DRG
framework, hospitals are incentivized to create and implement a system that
controls costs in order to fulfi ll their budgetary obligations.
14.6 New/innovative technologies
14.6.1 Steps required prior to introduction in hospitals
In Germany, most medical innovations are fi rst introduced in the inpatient
sector, because inpatient facilities may employ any technology that has not
264 Diagnosis-Related Groups in Europe
been excluded explicitly by the G-BA. The G-DRG system was designed, at least
in theory, to be always current, and classifi cation and reimbursement rates are
updated each year. However, as already outlined, a certain time-lag – and thus
a fi nancing gap – is nonetheless inherent in the system, because both the
G-DRG classifi cation and the reimbursement rates are based on retrospective
data. The time-lag may represent an important hurdle in the uptake of
new technologies. To address this defi cit, legislators introduced the so-called
New Diagnostic and Treatment Methods Regulation (NUB) as part of the
2005 KHEntG. The NUB Regulation has two key objectives: fi rst, to bridge the
above-mentioned fi nancing gap by providing for extrabudgetary, non-DRG
payments for new technologies and, second, to use the data generated during
this time-lag period to expedite the process for including these technologies in
the regular system of G-DRG reimbursement. The NUB Regulation sets up three
important regulatory hurdles that a new technology must clear before it can be
included in the regular system of G-DRG reimbursement: (1) a hospital wishing
to employ – and receive appropriate reimbursement for – a new medical
technology must fi rst apply to the InEK; (2) if the hospital’s application is
accepted, it must successfully negotiate with the sickness funds to receive NUB
reimbursement for its use of the technology; and (3) the technology must ulti-
mately be included in the regular system of G-DRG reimbursement (Henschke
et al., 2010).
Applying to the InEK
A hospital wishing to employ and receive NUB reimbursement for a new
medical technology must apply to the InEK for permission to enter into
contractual negotiations with the sickness funds. The technology does not
need to have an OPS code. The hospital’s application is assessed based on the
following criteria: (1) benefi ts to patients; (2) groups of patients who will be
treated using the new technology; (3) any additional labour and material costs
associated with the new technology; and (4) the reason why the costs of the
new technology are not adequately covered by the current G-DRG system.
Successfully negotiating NUB reimbursement with the sickness funds
An accepted application does not guarantee that a hospital will be reimbursed
for the use of a new technology. Before NUB reimbursement (E1 in Figure 14.6)
can take place, the hospital must negotiate a contractual agreement with the
sickness funds concerning the size of the payments to be made. If the technology
in question does not have an OPS code, the hospital may negotiate contracts
for two types of NUB reimbursement: additional payments, or full payments.
NUB reimbursement for a technology without an OPS code represents a prelimi-
nary step towards inclusion in the regular system of G-DRG reimbursement and
is represented in Figure 14.9 as the box labelled ‘Accepted NUB application
(without OPS)’. The arrows show prototypical pathways towards complete
integration in the system.
Germany: Understanding G-DRGs 265
Inclusion in the regular system of G-DRG reimbursement
The lowest stage of integration within the regular system of G-DRG reimburse-
ment is the so-called local valuated supplementary fee (D2 in Figure 14.6).
These payments are made in addition to DRG payments if the use of a certain
technology does not yet justify creating a unique DRG or a national valuated
supplementary fee (C1 in Figure 14.6). The decision to include a technology in
this category is made by the InEK. The local valuated supplementary fee has an
important advantage over NUB reimbursement: once a technology has been
included in the category of local valuated supplementary fees, any hospital in
Germany may enter into negotiations with the sickness funds to determine the
exact level of this payment. In contrast, when InEK accepts an application for
NUB reimbursement, only the hospital that applied may enter into negotiations
with the sickness funds; all other hospitals must apply with the InEK separately.
Finally, the last stage of integration into the regular system of G-DRG reim-
bursement is the formation of a unique DRG.
Figure 14.9 Prototypical regulatory pathways for introducing new technologies into
the regular system of G-DRG reimbursement
Source: Henschke et al., 2010.
266 Diagnosis-Related Groups in Europe
14.6.2 (Dis-)incentives for hospitals to use new technologies
Hospitals will use new and innovative technologies if they are adequately
reimbursed or are of major research interest. The NUB methodology enables
hospitals to use, and be reimbursed for, new technologies that are generally
more expensive than those included in the regular Case Fee Catalogue. As such,
being accepted for NUB reimbursement represents a preliminary step towards
the full inclusion of a new technology in the regular G-DRG system. However,
a recent study found that most German hospitals do not receive any revenue
via NUB payments, while those receiving NUB payments only generate 0.3 per
cent of revenue through this short-term payment instrument (DKI, 2009).
Moreover, the negotiation process between the hospitals and sickness funds is
tedious and does not guarantee a minimum payment in the event of unsuccessful
negotiations (Henschke et al., 2010).
14.7 Evaluation of the DRG system in Germany
14.7.1 Offi cial evaluation
The corporatist partners (Federal Association of Sickness Funds, Association of
Private Health Insurance, German Hospital Federation) are obliged by law to
ensure adequate research is undertaken to evaluate the impact of DRGs on the
provision of, as well as the quality of care (§17 KHG, para. 8). The research also
addresses DRGs’ effects on other supply sectors, such as rehabilitation or long-
term care (transfer of services out of the hospital). To introduce evaluation
activities, the corporatist institutions invite tenders for research assignments.
They also assign responsibility to the InEK for evaluating hospital-related
structural and case-related performance data (§21 datasets, see subsection
14.2.3). The fi rst results of the evaluation were intended to be published in
2005, but the corporatist partners have yet to meet their legal obligations. To
date, only the InEK has reported its analysis of the §21 data annually. As a fi rst
step in December 2008, the corporatist partners appointed a private institute
(IGES Institute) to conduct the mandatory evaluation. Preliminary results of
this evaluation indicate that the intended aims of the G-DRG system intro-
duction will be achieved and that most of the negative consequences of
prospective payment systems have not occurred (IGES, 2010). In addition, to
obtain a preliminary short-term evaluation, the Federal Ministry of Health
developed a qualitative questionnaire for the corporatist institutions and other
important stakeholders in 2007. Results indicate a broad acceptance of the
G-DRG system. However, the increased documentation effort and the increased
system complexity were criticized.
In addition, several research groups and institutions have examined the
effects of the G-DRG system on hospital reimbursement and service quality.
During the introduction, the adequacy of reimbursement for inpatient services
was evaluated (in particular by the DRG Research Group, University Hospital
Münster). Through this process, shortfalls were identifi ed in reimbursement
relative to the resource consumption of medical services (delivered in certain
Germany: Understanding G-DRGs 267
departments, for example oncology, rheumatology or dermatology), which led
to an increased number of DRGs and supplementary fees (see Table 14.2,
subsection 14.2.2). Furthermore, the effects of DRGs on quality were examined
in a study published in 2009 by the Centre for Quality and Management in
Health Care, which is a facility of the physicians’ chamber in Lower Saxony.
The study found no evidence of adverse effects, such as cream-skimming or
inappropriate early discharge. Other studies suggest that quality of care im-
proved or was not substantially affected, due to better organized care since the
introduction of DRGs (Sens et al., 2009).
14.7.2 Authors’ assessment
As with every case payment system, the G-DRG system has strengths and
weaknesses – the main ones are summarized in Table 14.4.
The increased transparency due to more precise documentation of hospital
services is one of the main strengths that has been identifi ed. Based on the
annually summarized §21 datasets, a structured summary of services delivered
and patient characteristics in German hospitals is undertaken.1 Another advan-
tage is the (increased) compliance of hospitals in supporting the G-DRG system,
which involves an accurate mapping of resource consumption and a stepwise
introduction process (see subsection 14.2.2). Indeed, hospitals have been
obliged to use G-DRGs since 2004, but weak cooperation on the part of the
hospitals is likely to have extended the introduction process. With the
incorporation of cost data from universities and other large hospitals in 2005,
even more complex services were available for consideration by the InEK for
developing the Case Fee Catalogue. Because of larger proportions of hospitals
delivering cost data, the system is now widely accepted. The use of G-DRGs for
reimbursement must also be highlighted. As every coded case is equivalent to
an invoice, the hospitals are strongly incentivized to code correctly in order to
avoid a review of their invoices by the sickness funds (see section 14.3.3). This
improves the coding quality and leads to a more accurate characterization of
delivered hospital services in Germany.
Despite these strengths, there are also some weaknesses and areas in need of
improvement. First, indicators of the quality of inpatient treatment are not
incorporated. Therefore, the level of reimbursement is unrelated to the quality
of service provision. Different approaches to incorporating quality of care
Table 14.4 Strengths and weaknesses of the G-DRG system
Strengths Weaknesses
Transparency and documentation No quality adjustments for reimbursement
Compliance of hospitals No refl ection of different input prices
Reimbursement tool Uniform accounting system but no full sample of
hospitals
Precision Increasing complexity with number of DRGs
268 Diagnosis-Related Groups in Europe
aspects in reimbursement – such as pay for performance (P4P) – have been
discussed in Germany, but due to a lack of evidence on effectiveness and
cost–effectiveness from other countries that have introduced P4P systems
in recent years, there is skepticism about its appropriateness in the German
context (Lüngen et al., 2008). Moreover, the development of quality indi-
cators that can easily be collected as part of routine data is still in progress
(Busse et al., 2009). Therefore, the connection between quality and reimburse-
ment will be one of the major topics for the further development of the G-DRG
system.
Second, the InEK calculates the Case Fee Catalogue with the assumption that
hospital input prices do not differ across Germany and all hospitals are working
under the same conditions. All cases are summarized and handled as if they
were treated in the same hospital. This ‘one hospital’ approach prevents the
consideration of hospital-specifi c (structural) costs within the reimbursement
system. Yet, current research shows that structural differences which are not
controllable by the hospitals affect their costs (Busse et al., 2008). Hence,
hospitals with higher costs due to structural differences are at risk of inadequate
reimbursement.
Third, the sample size of the hospitals delivering cost data could be extended
in order to increase the statistical power of the cost weight calculation. With
the cost-accounting scheme of the InEK as a standard (see subsection 14.4.2),
more hospitals and datasets can easily be incorporated. The resulting uniform
accounting system across Germany would simplify effi ciency comparisons and
benchmarking projects.
A known threat of DRG systems is increasing complexity with an increasing
number of DRGs. With the current G-DRG system incorporating 1200 groups
and several additional payments, every hospital needs to employ specialized
staff for coding purposes (see subsection 14.4.1). This additional effort must be
weighed against the advantages for the individual hospital and the whole
system.
14.8 Outlook: Future developments and reform
14.8.1 Trends in hospital service or general delivery
There is a general trend towards concentration on selected specialties, which is
an indirect result of the introduction of the G-DRGs. This has been associated
with increasing hospital market penetration by (profi t-making) hospital chains,
and the reduction of overall capacities, which forced hospitals to specialize or
to accept across-the-board cuts in resources (Leclerque & Robra, 2009). Moreover,
regulatory reform of the SGB V (sections §115b, §116b, §140) provides hospitals
with more freedom to offer outpatient services and to shift the boundaries
between inpatient and outpatient care. A general trend is therefore the
establishment of so-called ‘Medizinische Versorgungszentren’ (Care Centres),
which try to achieve clinical as well as economic benefi ts through integrated
care models and economies of scale (Neubauer & Minartz, 2009).
Germany: Understanding G-DRGs 269
14.8.2 Trends in DRG application/coverage
In recent years the G-DRG has been characterized by two trends with regard to
patient classifi cation:
1. refi nement of the grouping algorithm inherited from the AR-DRG system,
especially the development of a hierarchy of sub-groups below the level of
the MDCs;
2. refl ection of complex treatments and repetitive surgical procedures in DRG
weights; this implied greater use of procedures for defi ning DRGs and
weighting them (Roeder et al., 2008).
Another trend is preparation for the introduction of case payments for
psychiatric services/care in Germany. The latter will build on the experiences of
the G-DRG system, but will most likely be an independent system that will
operate totally separately. We therefore do not discuss this in any further detail
here.
14.8.3 Future developments and reform
The main future development activity can be distinguished in two fi elds:
fi nancing and regulation, and the design implications of the G-DRG system.
Financing and regulation
There is a long-standing debate in Germany about hospital fi nancing. Critics
argue that the dualistic hospital fi nancing structure leads to ineffi cient
investment decisions (Felder et al., 2008). While this claim is controversial, it is
widely accepted that the level of public investment in hospitals is no longer
appropriate to meet infrastructural needs. Between 1993 and 2005, public
investment in hospitals declined by 3 percent while adjusting for infl ation
(Augurzky et al., 2007). During the same period, economic pressures, documen-
tation and performance requirements increased due to the introduction of the
G-DRG system. Competitive pressures will further increase and hospitals will be
even more dependent on adequate investment. Many policy-makers and
researchers therefore argue that German hospital fi nancing should follow the
principle of monistic fi nancing, that is, sickness funds should cover operating
costs as well as investment in infrastructure (capital costs). Often this proposal
is linked to demands to liberalize the regulation of prices and the benefi ts
catalogue for the inpatient sector, which are currently strictly defi ned by
collective decision-making. Large sickness funds argue that regulators should
defi ne benefi ts and prices only for acute and emergency services, while for
elective procedures provision and prices should be negotiated between hospitals
and payers (AOK BV, 2009).
G-DRG system design implications
As outlined in this case study, the G-DRG system is characterized by increasing
differentiation, as the grouping of hospital services by diagnosis and procedures
270 Diagnosis-Related Groups in Europe
is constantly refi ned to ensure adequate resource allocation. This constant
refi nement nevertheless also has ambiguous consequences, such as the emer-
gence of DRGs with a very low number of cases, decreasing stability of the pay-
ment regime as parameters constantly change, as well as increasing complexity
(Roeder et al., 2008). In addition, DRGs are often no longer homogeneous in a
medical sense. As a consequence, their use is increasingly limited to reimburse-
ment purposes, as their application in quality monitoring, treatment pathways
and so on is no longer appropriate (Roeder et al., 2008, p. 37). The G-DRG
system may therefore need to fi nd adequate solutions for fi nancing specialized
treatments that are as yet not adequately represented in specifi c DRGs. One way
to achieve this may be to increase reliance on extrabudgetary, non-DRG
payments for new technologies (namely, the ‘NUB’ approach).
14.9 Note
1 The summary for the latest available data year (currently 2008) is published on the
InEK web site via an Access database and is publicly accessible (albeit in German only)
(www.g-drg.de, accessed 10 July 2011).
14.10 References
AOK BV (2009). AOK-Positionen zur Gesundheitspolitik nach der Bundestagswahl 2009.
Berlin: AOK-Bundesverband (http://www.aok-bv.de/politik/reformaktuell/index_
01540.html, accessed 14 November 2009).
Augurzky, B., Engel, D., Krolop, S. et al. (2007). Krankenhaus Rating Report 2007 – Die Streu
trennt sich vom Weizen. Essen: RWI.
Bölt, U. (2010). Statistische Krankenhausdaten: Grund- und Kostendaten der
Krankenhäuser 2007, in J. Klauber, M. Geraedts, J. Friedrich, eds. Krankenhaus-Report
2010. Stuttgart: Schattauer.
Braun, T., Rau, F., Tuschen, K.H. (2007). Die DRG-Einführung aus gesundheitspolitischer
Sicht. Eine Zwischenbilanz, in J. Klauber, B.P. Robra, H. Schellschmidt, eds.
Krankenhaus-Report 2007. Stuttgart: Schattauer.
Busse, R., Riesberg, A. (2004). Health Care Systems in Transition: Germany. Copenhagen:
WHO Regional Offi ce for Europe on behalf of the European Observatory on Health
Systems and Policies.
Busse, R., Nimptsch, U., Mansky, T. (2009). Measuring, monitoring, and managing quality
in Germany’s hospitals. Health Affairs, 28(2):w294–w304.
Busse, R., Schreyögg, J., Smith, P.C. (2008). Variability in healthcare treatment costs
amongst nine EU countries – results from the HealthBASKET project. Health Economics,
17(1 Suppl.):1–8.
DESTATIS (2009). Statistisches Jahrbuch 2009. Wiesbaden: Statistisches Bundesamt
Deutschland.
DKI (2009). Anspruch und Realität von Budgetverhandlungen zur Umsetzung medizintechnischer
Innovationen. Gutachten des Deutschen Krankenhausinstituts (DKI) im Auftrag des
Bundesverbandes Medizintechnologie (BVMed). Düsseldorf: Deutsches Krankenhausinstitut.
Felder, S., Fetzer, S., Wasem, J. (2008). ‘Was vorbei ist, ist vorbei’: Zum Übergang in die
monistische Krankenhausfi nanzierung, in J. Klauber, B.P. Robra, H. Schellschmidt,
eds. Krankenhaus-Report 2007. Stuttgart: Schattauer.
Germany: Understanding G-DRGs 271
Friedrich, J., Leclerque, G., Paschen, K. (2008). Die Krankenhausbudgets 2004 bis 2006
unter dem Einfl uss der Konvergenz, in J. Klauber, B.P. Robra, H. Schellschmidt, eds.
Krankenhaus-Report 2007. Stuttgart: Schattauer.
Henschke, C., Bäumler, M., Weid, S., Gaskins, M., Busse, R. (2010). Extrabudgetary (‘NUB’)
payments – a gateway for introducing new medical devices into the German inpatient
reimbursement system? Journal of Management and Marketing in Healthcare, 3(2):
119–33.
IGES (2010). DRG Impact Evaluation According To Section 17b Paragraph 8 Hospital Financing
Act. Berlin: IGES Institute.
InEK (2004). Abschlussbericht zur Weiterentwicklung des G-DRG-Systems für das Jahr 2005.
Siegburg: Institut für das Entgeltsystem im Krankenhaus gGmbH.
InEK (2007). Handbuch zur Kalkulation von Fallkosten Version 3.0. Siegburg: Institut für das
Entgeltsystem im Krankenhaus gGmbH.
InEK (2009a). Abschlussbericht zur Weiterentwicklung des G-DRG-Systems für das Jahr 2010.
Siegburg: Institut für das Entgeltsystem im Krankenhaus gGmbH.
InEK (2009b). Verfahrenseckpunkte: Anfragen nach § 6 Abs. 2 KHEntgG (Neue Untersuchungs-
und Behandlungsmethoden) für 2009. Siegburg: Institut für das Entgeltsystem im
Krankenhaus gGmbH.
Leclerque, C., Robra, B.R. (2009). Einführung, in J. Klauber, B.R. Robra, H. Schellschmidt,
eds. Krankenhaus-Report 2008/2009. Stuttgart: Schattauer.
Lüngen, M., Gerber, A., Lauterbach, K.W. (2008). Pay for Performance: Neue Impulse für
den Wettbewerb zwischen Krankenhäusern? in J. Klauber, B.P. Robra, H. Schellschmidt,
eds. Krankenhaus-Report 2007. Stuttgart: Schattauer.
MDS (2011). Abrechnungsprüfungen der MDK in Krankenhäusern sind angemessen, wirtschaftlich
und zielführend. Zahlen und Fakten der MDK-Gemeinschaft. Essen: Medizinischer Dienst
des Spitzenverbandes Bund der Krankenkassen e.V.
Neubauer, G., Minartz, C. (2009). Zentrierte Versorgung – Ziele und Optionen, in J.
Klauber, B.R. Robra, H. Schellschmidt, eds. Krankenhaus-Report 2008/2009. Stuttgart:
Schattauer.
Neubauer, G., Pfi ster, F. (2008). DRGs in Germany: introduction of a comprehensive,
prospective DRG payment system by 2009, in J.R. Kimberly, G. de Pouvourville, T.
D’Aunno, eds. The Globalization of Managerial Innovation in Health Care. Cambridge:
Cambridge University Press.
Roeder, N., Bunzmeier, H., Fiori, W. (2008). Ein lernendes Vergütungssystem – Vom
Budgetierungsinstrument zum deutschen Preissystem, in J. Klauber, B.P. Robra, H.
Schellschmidt, eds. Krankenhaus-Report 2007. Stuttgart: Schattauer.
Schreyögg, J., Tiemann, O., Busse, R. (2006). Cost accounting to determine prices: how
well do prices refl ect costs in the German DRG system? Health Care Management
Science, 9:269–80.
Sens, B., Wenzlaff, P., Pommer, G., von der Hardt, H. (2009). DRG-induzierte Veränderungen
und ihre Auswirkungen auf die Organisationen, Professionals, Patienten und Qualität.
Hanover: Zentrum für Qualität und Management im Gesundheitswesen, Einrichtung
der Ärztekammer Niedersachsen.
Tuschen, K.H., Trefz, U. (2004). Krankenhausentgeltgesetz: Kommentar. Stuttgart: Kohlhammer.
Velasco-Garrido, M., Busse, R. (2004). Förderung der Qualität in deutschen
Krankenhäusern? Eine kritische Diskussion der ersten Mindestmengenvereinbarung.
Gesundheits- und Sozialpolitik, 58(5/6):10–20.
chapter f i f t e e n
Ireland: A review of casemix
applications within the acute
public hospital system
Jacqueline O’Reilly, Brian McCarthy
and Miriam Wiley1
15.1 Hospital services and the role of DRGs in Ireland
15.1.1 The Irish health care system
Health care expenditure in Ireland experienced unprecedented growth during
the late 1990s, increasing by almost 80 per cent in real terms between 1997 and
2002 (Nolan, 2005; Wiley, 2005; McDaid et al., 2009). Growth in expenditure
has remained strong in subsequent years and total current health expenditure
was in excess of €19 billion in 2008 (McDaid et al., 2009; Brick et al., 2010).
About four fi fths of total (current and capital) health expenditure was publicly
funded in 2007, up from three quarters in 1997 (McDaid et al., 2009; OECD,
2009; Brick et al., 2010). Private health expenditure from out-of-pocket
payments and private health insurance accounted for the remainder (Brick
et al., 2010). In 2007, public health expenditure accounted for 7.2 per cent of
gross domestic product (GDP) (8.5 per cent of gross national product, GNP)
(McDaid et al., 2009).2
This period of expenditure growth was followed by fundamental structural
reform. The health system was previously organized according to a regional
structure. However, this decentralized structure created tensions between
national health policy objectives and local service delivery (Brennan, 2003;
Prospectus, 2003; McDaid et al., 2009). Consequently, the Health Service Execu-
tive (HSE) was established in January 2005 to focus on service delivery and
management at national level, while the Department of Health and Children
(DoHC) was charged with devising policy and strategy, thereby effecting a
separation of operation and management from policy-making. Since October
274 Diagnosis-Related Groups in Europe
2009 the Integrated Services Directorate within the HSE oversees the primary
care and acute hospital sectors – a task that was previously divided between
the Primary, Community and Continuing Care Directorate and the National
Hospitals Offi ce (HSE, 2010d).
The unusual public/private interaction in Irish health care means that acute
public hospitals can provide private services. Private practice within public
hospitals is generally constrained to beds designated for private patients, which
amount to approximately 20 per cent of all acute public hospital beds
nationally.3 Consultants in acute public hospitals may – depending on their
employment contract – be permitted to treat private patients up to a maximum
of 20–30 per cent of their complexity-adjusted workload (Brick et al., 2010).
Public hospitals and consultants face different payment mechanisms for public
and private patients, which have been criticized for incentivizing the treatment
of private patients (Nolan & Wiley, 2000; Colombo & Tapay, 2004; Brick et al.,
2010; O’Reilly & Wiley, 2010; Ruane, 2010). Given the relatively low number of
private hospitals and the paucity of readily available private sector data, what
follows focuses on the acute public hospital sector.4
There are three main categories of entitlement to access health care services
in Ireland (see Table 15.1). Eligibility for a medical card is largely determined on
the basis of income (McDaid et al., 2009).5 GP visit cards are also allocated on
the basis of income, with the income threshold being 50 per cent higher than
that for medical card holders (Brick et al., 2010). In 2009, medical card holders
and GP visit card holders comprised 33.2 per cent and 2.2 per cent of the
population, respectively (DoHC, 2010a). The remainder of the population
(approximately 65 per cent in 2009; Brick et al., 2010) do not qualify for a
medical card or a GP visit card. About 46 per cent of the population in 2009
held supplementary private health insurance, which mainly covers acute
hospital services (Brick et al., 2010). A further 5 per cent held both a medical
card and private health insurance (Brick et al., 2010).
15.1.2 Hospital services in Ireland
In 2010, 52 acute public hospitals provided day-case, inpatient, outpatient and
emergency department (ED) services to public and private patients (Brick et al.,
2010; HSE, 2010c). All of these public hospitals receive funding from the HSE
on a global budget basis (McDaid et al., 2009; Brick et al., 2010). Annual global
budgets are determined on an historic basis, with some adjustment for, inter
alia, infl ation, pay adjustments and one-off funding. A subset of these public
hospitals (39 in 2010; HSE, 2010a) participate in the National Casemix
Programme under which their global budgets are prospectively adjusted using
diagnosis-related groups (DRGs) (see subsection 15.5).
In addition to funding public hospitals, the HSE also directly operates 34
acute public hospitals (Brick et al., 2010; ESRI-HRID, 2010). The remaining
public hospitals are typically owned and operated by voluntary organizations
(for example, religious orders) (Robbins & Lapsley, 2008; McDaid et al., 2009).
A small number of acute public hospitals provide specialist services (such as
maternity, paediatric and orthopaedic care). Table 15.2 provides a breakdown of
Ireland: A review of casemix applications 275
Table 15.1 Health care charges for public and private patients
Public patient Private patient
Medical card holder Non-medical card holder
(including GP visit card
holder)
GP visits Nil Charge determined by
GP
Nil for GP visit card
holder
Charge determined
by GP
Prescription
medicines
50c charge per
prescription item
up to maximum of
€10 per family per
montha
Free above €120 out-of-pocket payment per
month
No charge for certain long-term illnesses/
conditions
Public hospitals
ED Nil €100 unless referred by GP or subsequently
admitted to hospitalc
Outpatient
department
Nil €100 unless referred by GP or subsequently
admitted to hospitalc
No charge for repeat attendances
Day case/
inpatient
Nilb Daily hospital charge of
€75 (up to a maximum
of €750 in any 12
consecutive months)b
Daily hospital charge
as per public patients
plus a hospital
maintenance charge
and consultant feesd
Sources: Adapted from Brick et al., 2010; Citizens Information Board, 2011.
Notes: aFrom 1 October 2010; bAdditional charges may be levied on long-stay patients; cRates
effective from 1 January 2009; dThe hospital maintenance charge is a per diem charge, which
varies according to the type of treatment (inpatient or day case), accommodation (private or
semi-private bed) and hospital (DoHC, 2009b); For 2011, this charge ranges from €193 for
day care in district hospitals to €1017 for private accommodation in certain hospitals, such as
regional hospitals (Citizens Information Board, 2010).
public hospitals by their size, ownership and type and Table 15.3 reports the
changes in hospital beds and activity between 2000 and 2008.6
15.1.3 Purpose of the DRG system
The DRG system in Ireland has a number of national applications. First, follow-
ing a recommendation of the Commission on Health Funding (1989), the DRG
system has been used since 1993 to adjust acute public hospitals’ budgetary
allocations for the complexity of their casemix and their relative performance.
Second, under a renegotiated contract (effective from September 2008), the
outputs of the national casemix models are used to adjust hospital consultants’
day-case and inpatient activity for complexity and to take account of their
involvement in each case (HSE, 2008). Third, DRG data feed into HealthStat, a
276 Diagnosis-Related Groups in Europe
Table 15.3 Summary of hospital activity and beds, 2000 and 2008
2000 2008 % change
2000–
2008Number % Rate per
1000
population
Number % Rate per
1000
population
Hospital beds
Day-case beds 721 5.8 0.2 1 697 12.2 0.4 135.4
Inpatient beds 11 704 94.2 3.1 12 182 87.8 2.8 4.1
Total beds 12 425 100 3.3 13 879 100 3.1 11.7
Discharges
Day-case
discharges 273 677 34.3 72.2 771 145 56.3 174.4 181.8
Inpatient
discharges 525 181 65.7 138.6 597 449 43.7 135.1 13.8
Total discharges 798 858 100 210.8 1 368 594 100 309.5 71.3
Beds days
Day case 273 677 7.5 72.2 771 145 17.2 174.4 181.8
Inpatient 3 371 089 92.5 889.6 3 700 959 82.8 836.9 9.8
Total bed days 3 644 766 100 961.8 4 472 104 100 1 011.3 22.7
Acute inpatient
ALOS (days)a 5.0 – – 4.6 – – –8.0
ED attendances 1 211 279 – 319.6 1 150 674 – 260.2 –5.0
Outpatient
attendances 1 996 474 – 526.8 3 288 917 – 743.7 64.7
Sources: ESRI-HPID, 2007; DoHC, 2009a, 2010a; ESRI-HRID, 2010.
Notes: These data relate to hospitals that participated in the Hospital In-Patient Enquiry (HIPE);
All acute public hospitals (apart from one) and a small number of long stay hospitals participated
in HIPE in 2008; HIPE data collection has changed over the period (see, for example, ESRI-HRID,
2008b); aAcute inpatients are inpatients with a length of stay of 30 days or less.
Table 15.2 Distribution of hospitals by number of beds, ownership and type, 2008
Number
of beds
HSE Public voluntary Total
(HSE and public voluntary)
General Special Total General Special Total General Special Total
<100 5 4 9 2 1 3 7 5 12
100–<200 5 1 6 2 4 6 7 5 12
200–<300 10 0 10 2 2 4 12 2 14
300–<400 5 0 5 0 0 0 5 0 5
400–<500 0 0 0 0 0 0 0 0 0
500–<600 3 0 3 1 0 1 4 0 4
≥600 1 0 1 4 0 4 5 0 5
Total 29 5 34 11 7 18 40 12 52
Source: Adapted from Brick et al., 2010.
Notes: Data on psychiatric beds were not available for four hospitals; Bed data relate to the
average number of beds available and include both day-case and inpatient beds.
Ireland: A review of casemix applications 277
HSE initiative to monitor and assess performance within the acute public
hospital sector (HSE, 2010b). In addition, the DRG system has been (and con-
tinues to be) used to inform the planning and reconfi guration of acute hospital
services.
15.2 Development and updates of the DRG system
15.2.1 The current DRG system at a glance
The Australian Refi ned Diagnosis-Related Group (AR-DRG) system was adopted
in Ireland in 2005 for classifying day-case and inpatient activity (excluding
non-acute psychiatry, geriatric care and rehabilitation) (ESRI-HRID, 2008a). The
current version of AR-DRGs (Version 6.0) was introduced in Ireland in January
2009. For the 2010 casemix budgetary adjustment, a patient classifi cation
system specifi cally designed for the HSE – Treatment-Related Groups (TRGs) –
was used for the fi rst time to group outpatient attendances. There is currently
no grouper for ED attendances in Ireland. The remainder of this chapter will
focus on the day-case and inpatient casemix models. The HSE’s National
Casemix Programme is responsible for developing these models and combining
the required cost and activity data from hospitals. The AR-DRG algorithm is
outlined in section 15.3.2.
15.2.2 Development of the DRG system
Since DRGs were introduced in Ireland in the early 1990s, 10 classifi cation
systems have been used for inpatient and day-case activity (see Table 15.4).
These classifi cation systems are discussed in greater detail in the remainder of
this section.
Inpatient classifi cation systems
Between 1992 and 2002, the United States Health Care Financing Administration
(HCFA)-DRGs, together with Maryland cost/service weights, were used to group
Table 15.4 Overview of inpatient and day-case classifi cation systems used in Ireland
Data yeara Inpatient classifi cation system Day-case classifi cation system
System Version System Version
1992–1993 HCFA 9 No grouper
1994–1998 12 DPG procedural grouper
1999–2001 16
2002 DG procedural grouper
2003–2004 AR-DRG 5.0
2005–2008 5.1 ADRG 5.1
2009 to date 6.0 6.0
Note: aData year refers to the year(s) in which the discharge activity took place.
278 Diagnosis-Related Groups in Europe
inpatients in Ireland (see Table 15.4). During this period, updates to the grouper
were driven by revisions to the clinical coding scheme (International Classifi -
cation of Diseases 9th revision – Clinical Modifi cation, ICD-9-CM) in Ireland.
The HCFA classifi cation had a maximum of two severity levels – with or without
complications.
Recognizing the need to update the clinical coding scheme to ICD-10, the
DoHC commissioned a review of alternative grouping schemes in 2003 (Aisbett
et al., 2007). As part of this review, several groupers were evaluated against a
range of criteria, including vendor support, and international use and recog-
nition. The review recommended the adoption of an unmodifi ed version of the
AR-DRG Version 5.0 grouper.
For 2003 and 2004, it was necessary to use a coding map to convert clinical
data coded in ICD-9-CM to ICD-10-AM (10th revision, Australian Modifi ca-
tion) for diagnoses and to the Australian Classifi cation of Health Interven-
tions (ACHI) for procedures.7 However, since 2005, all Hospital In-Patient
Enquiry (HIPE) data have been coded using unmodifi ed versions of ICD-10-AM
and ACHI (see Murphy and colleagues (2004) on adopting these coding
schemes).
Table 15.5 demonstrates key differences between the four DRG systems used
to group inpatients in Ireland since 1999. Compared to previous classifi cations,
AR-DRGs have a slightly higher number of groups, more severity levels, and can
be applied to day cases as well as inpatients.
Table 15.5 Key facts on DRG systems used in Ireland since 1999
HCFA-DRG
Version 16
AR-DRG
Version 5.0 Version 5.1 Version 6.0
Year of introduction 1999 2003 2005 2009
Data year 1999–2002 2003–2004 2005–2008 2009 to date
Number of groups 511 665 665 698
Sub-classifi cation None ADRG (399) ADRG (399) ADRG (399)
Maximum severity
levels 2 4 4 4
Birth weight Not used Required
Type of cases included Inpatient only,
excluding non-
acute psychiatric
care, geriatric care
and rehabilitation
Inpatients and day cases, excluding
non-acute psychiatric care, geriatric
care and rehabilitation
Number of MDCs 25 24a 24a 24a
Number of
participating public
hospitals 32 (2002) 32 (2004) 39 (2008) 39 (2009)
Sources: Based on information from HCFA (undated); see also Commonwealth Department of
Health and Ageing, 2002, 2004, 2008; National Casemix Programme, 2010a.
Note: aIncludes Pre-MDC.
Ireland: A review of casemix applications 279
Day-case classifi cation systems
In Ireland, a day case is characterized by a patient being admitted electively and
discharged on the same day, as planned. The National Casemix Programme was
extended to include day-case activity in the mid-1990s. The fi rst day-case
classifi cation system – the Day Patient Grouper (DPG) – was developed by the
DoHC, based on similar United States groupers (Table 15.4). The DPG defi ned
73 groups, differentiated mainly on the basis of principal procedure coded in
ICD-9-CM. Cases without procedures were grouped into a single medical day-
case group. In 2002, the DPG was replaced by the Day Grouper (DG), which was
also developed by the DoHC and comprised 169 groups. Like its predecessor,
the DG was predominantly driven by the principal procedure, although
diagnoses were also used in some cases to determine the group.
With the introduction of AR-DRGs in Ireland, day cases and inpatients could
be included in the same classifi cation system. Under the AR-DRG classifi cation
scheme, Adjacent DRGs (ADRGs) are used to group day cases (see subsection
15.3.2).8 Unlike the procedural-based DPGs and DGs, the AR-DRG classifi cation
for day cases incorporates more information than simply the principal procedure.
15.2.3 Data used for development and updates
of the DRG system
The AR-DRG classifi cation system from Australia, which was adopted in Ireland,
was developed in the late 1990s by the Australian Casemix Clinical Committee
(ACCC). To update the AR-DRG system, the ACCC relies on Australian clinical
and cost data, as well as on input from Australian health professionals (Com-
monwealth Department of Health and Ageing, 2008).
Irish cost/service weights are determined by modifying Australian cost/
service weights using Irish data from the hospital costing fi le. The costing fi le is
based on hospitals’ Annual Financial Statements (AFSs) and consists of specialty-
level hospital cost data for inpatients and day cases.9
Data on day-case and inpatient activity for the DRG system and the National
Casemix Programme are obtained from the HIPE system, which is the only
national source of administrative, demographic and clinical information on
discharges from acute public hospitals in Ireland. Since 1990, the Economic
and Social Research Institute (ESRI) has been responsible for all aspects of
managing this database.
15.2.4 Regularity and method of system updates
The AR-DRG grouper classifi cation is updated every two years in Australia in
conjunction with the update of the ICD-10-AM classifi cation. Since their
adoption in Ireland, the clinical coding and AR-DRG classifi cations have been
updated every four years (only twice so far – 1 January 2005 and 1 January
2009).10 To date, therefore, Ireland has adopted every second version of the AR-
DRG classifi cations.
280 Diagnosis-Related Groups in Europe
15.3 The current patient classifi cation system
15.3.1 Information used to classify patients
In Ireland, administrative, demographic and clinical data from the HIPE
database are used to group hospital discharges into AR-DRGs (see subsection
15.2.3).11
Administrative and demographic data
Relevant administrative data include admission and discharge dates, discharge
destination (for example, home, transfer to rehabilitation), and length of stay.
Demographic variables consist of date of birth, age in years/days, and gender.
Three further variables (non-acute length of stay, leave days, and mental health
legal status) are required by the grouper, but are not available in Ireland and,
consequently, are set to their default values.
Clinical data
Since January 2009, clinical data are coded in the HIPE using the sixth edition
of ICD-10-AM, which contains the ACHI classifi cation for procedures. Two sets
of coding standards apply: the Australian coding standards and the comple-
mentary Irish coding standards. The grouper requires a principal diagnosis and
can accept up to 29 additional diagnoses and up to 30 procedures.12 Resource-
consumption information – such as length of stay in an intensive care
environment – is implicit in the coded procedures used in the grouper.
15.3.2 Classifi cation algorithm
Grouping occurs in the hospital after the case has been coded. Each discharge
is allocated to one group only. The AR-DRG grouping algorithm consists of four
steps (see Figure 15.1).13 In the fi rst step, data on each discharge undergo a series
of checks for completeness, validity and consistency. Failing a check may result
in a discharge being assigned to an Error AR-DRG. In addition to these checks,
HIPE data are subject to rigorous validation checks during the collection process
(see subsection 15.3.3).
In the second step, the principal diagnosis determines the allocation to one
of 24 major diagnostic categories (MDCs).14 For certain specialist, high-cost
conditions (such as transplants), however, the initial MDC assignment may be
altered, with some discharges being reclassifi ed to a Pre-MDC.
The presence of a procedure and its type are used to classify discharges into
one of three partitions, which constitutes the third step. A discharge is assigned
to the ‘surgical’ partition if there was at least one operating room (OR) proce-
dure; to the ‘other’ partition if there was at least one relevant non-OR procedure;
or to the ‘medical’ partition if there were no relevant procedures (that is, a proce-
dure may have been performed, but it was not relevant to the MDC). After the
partition is assigned, the discharge is allocated to one of 399 ADRGs, depending
Ireland: A review of casemix applications 281
on diagnoses, procedures and administrative data. The ADRG assignment
completes the grouping process for day cases.
Finally, the discharge is allocated to one of 698 AR-DRGs according to severity
level, which is determined by characteristics considered to be drivers of resource
use (for example, presence of complications and co-morbidities (CCs), age,
gender).15 Usually, CCs are considered in combination to provide a composite
score (the Patient Clinical Complexity Level, PCCL) that indicates the patient’s
overall complexity and determines the fi nal AR-DRG.16
Each AR-DRG consists of four alphanumeric characters (Commonwealth
Department of Health and Ageing, 2008). The fi rst character is a letter between
A and Z indicating the MDC. The second and third characters are numbers,
which (together with the MDC letter) defi ne the three-character ADRG. The
fi nal character is a letter and indicates whether the ADRG was not split (denoted
by ‘Z’) or the level of severity/resource use (with four categories starting with
‘A’, indicating the highest resource use).
15.3.3 Data quality and plausibility checks
In addition to the data checks described in subsection 15.3.2, HIPE data are also
subjected to several plausibility and validity checks at local (hospital) and national
Figure 15.1 Assignment to DRGs using the AR-DRG classifi cation scheme
Source: Adapted from Commonwealth Department of Health and Ageing, 2008.
282 Diagnosis-Related Groups in Europe
levels. The fi rst set of plausibility checks occurs in hospitals during the data-entry
phase and these have been built into the standardized HIPE data-collection
software, which has been developed by the Health Research and Information
Division (HRID) within the ESRI and is provided free of charge to all participating
hospitals. The software has been designed to validate information as it is entered,
so that coders can readily reference patient charts if necessary. Such plausibil-
ity checks include: validation of diagnosis and procedure codes against patient
administrative and demographic details; examining the ordering of diagnoses to
ensure that certain diagnosis codes are never used as the principal diagnosis; and
notifying coders of certain unusual diagnosis/procedure combinations.
A further set of checks is conducted by the HRID on the national HIPE fi le.
First, year-on-year changes in casemix-adjusted discharges (calculated on the
same basis for all years) for a particular hospital are examined. Second, casemix-
adjusted discharges are compared to costs in order to identify where activity has
increased without a corresponding increase in costs.
Analysis of the national fi le may highlight specifi c areas at specialty or
AR-DRG level where patient-level audits are required. These audits are carried
out by trained coders, external to the hospital. Local audits by hospitals are also
supported by the HIPE system and the HRID has developed the HIPE Coding
Audit Toolkit (HCAT) to assist in this process.17
The National Casemix Programme may revise the number of allocated
casemix-adjusted discharges if there is evidence from these checks and external
audits that inaccurate data have been submitted (such as the inclusion of
outpatient activity in the day-case model). Where costs or activity data are
found to be incomplete or inaccurate, this would be expected to result in a
lower casemix budgetary adjustment.
15.4 Cost accounting within hospitals
15.4.1 Regulation
Hospitals participating in the National Casemix Programme must submit their
cost data to the hospital costing dataset in a standardized format, as outlined in
the casemix specialty costing manual. These data are then submitted to the
National Casemix Programme, which performs regular data quality audits. In
the fi rst step of the audit process, all costs submitted in the costing fi le are
reconciled with the AFS to check for omissions or errors. Hospitals’ costs and
patient activity data are then linked using the national casemix models. The
overall audit process typically involves discussions between the National
Casemix Programme and the hospitals. Where issues remain unresolved, the
National Casemix Programme can amend the hospital costing fi le where the
costing manual rules have been breached.
15.4.2 Main characteristics of the cost-accounting system
The cost-accounting system for the casemix models starts by adjusting hospitals’
AFSs (for example, excluding capital expenditures which are not fi nanced
Ireland: A review of casemix applications 283
through the national casemix models; National Casemix Programme, 2010b).
In line with the HSE’s costing manual, the adjusted AFS costs are allocated on a
top-down basis to specialty level for inpatients and day cases, as well as to
outpatient clinics, primarily using direct allocation.
Once the hospital costing fi les are submitted and the auditing process by the
National Casemix Programme is complete, the national costing fi le is prepared.
For compatibility with the national casemix models, the costing fi le is arranged
into the following 13 cost centres: allied health; critical care; coronary care
unit; emergency; imaging; pathology; medical pay; prosthesis; nursing; phar-
macy; theatre operating procedures; theatre non-operating procedures; and
blood.
When the national casemix models are complete, the combined national
patient activity and costing fi le is used to produce national statistics (for
example, average cost per casemix-adjusted discharge).
15.5 DRGs for hospital fi nancing
15.5.1 Range of services and costs included in
casemix-adjusted budgets
Public hospitals are invited to participate in the National Casemix Programme
by the HSE. The number of participating acute public hospitals increased
from 15 in 1993 to 39 in 2009 (HSE, 2010a; National Casemix Programme,
2010a). To ensure comparability, hospitals are assigned to one of four peer
groups in the Programme. In the 2009 casemix models, Group I included eight
major academic teaching hospitals; Group II contained 26 hospitals; and three
maternity and two paediatric hospitals constituted the remaining two groups.
Casemix budgetary adjustments are calculated separately for each hospital
group.
There are separate national casemix models for day cases and inpatients;
non-acute psychiatry, geriatric care and rehabilitation are excluded. All costs
are included in the models, apart from those related to capital, depreciation,
pensions, bad debts and transfers from other hospitals. Teaching costs are
included but research costs are fi nanced separately. Less than 80 per cent of all
acute public hospital costs were included in the most recent casemix models
(run in 2009; National Casemix Programme, 2010d). As part of the cost-
reconciliation process, all participating hospitals can make submissions on a
case-by-case basis to the National Casemix Programme for additional funding
for unusual and/or high-cost activity.
15.5.2 Service weights and relative values
In Ireland, service weights (locally referred to as cost weights) differ from
relative values (RVs).18 Service weights indicate the share of specialty costs
allocated to each AR-DRG and are defi ned for each of the 13 cost centres listed
in subsection 15.4.2. Service weights are based on the patient cost data from
284 Diagnosis-Related Groups in Europe
Australia, adapted to the Irish health care system by the inclusion of Irish cost
data. In contrast, RVs indicate the relative resource intensity of each AR-DRG.19
RVs are calculated separately for day cases and inpatients as part of the casemix
models using the national cost and activity fi les. A separate set of RVs is
determined for paediatric hospitals, given their specialist nature.
The process for determining the inpatient RVs is described here.
1. Inpatient discharge data are initially trimmed at three standard deviations
from the mean length of stay for each AR-DRG to remove cases with
extremely short or long lengths of stay from the calculation of the upper and
lower length-of-stay thresholds, which are set at 1.96 standard deviations
above and below the mean, respectively.20
2. The average cost per cost centre and per hospital is determined using the
hospital’s inpatient cost and activity data, together with inpatient service
weights.
3. The inpatient cost and discharge activity data are analysed per AR-DRG to
calculate the national average costs per DRG and RVs.
15.5.3 DRG-based budgetary adjustments
The following subsections describe how hospitals’ casemix budgetary adjust-
ments are determined.
Calculation of hospital casemix-adjusted activity/casemix units
The casemix unit (CMU) is determined by the RV of the AR-DRG to which
the patient is assigned, the patient’s length of stay, and the upper and lower
length-of-stay thresholds of the AR-DRG. The casemix-adjusted activity of
each hospital is the sum of the CMUs across all discharges. Since RVs are
calculated using different models and cost data for inpatients and day cases, it
is not possible to combine casemix-adjusted inpatient activity with that for day
cases.
Figure 15.2 illustrates how the CMU is determined for inpatient discharges.
The CMU for a same-day, one-day or inlier case can be determined directly,
since same-day, one-day and inlier RVs are calculated separately for each
AR-DRG. If the patient is a low outlier, the CMU is determined using the one-
day RV with a per diem adjustment. Conversely, the CMU for high outliers is
based on the inlier RV of the patient’s AR-DRG plus a per diem AR-DRG
adjustment for each day in hospital above the upper threshold.
Casemix-adjusted day cases are simply the RV of the applicable ADRG.
Determining hospital casemix budgetary adjustment
Figure 15.3 shows the calculation of a casemix-adjusted budget for inpatients.
The casemix-adjusted budget of a hospital depends on the hospital’s CMUs and
base rate, a peer-group base rate and a blend rate. The hospital base rate is
calculated by dividing the total costs of the hospital by the hospital’s CMUs.
Ireland: A review of casemix applications 285
Figure 15.2 Distribution of CMUs for inpatient discharges
Notes: A same-day case is admitted and discharged on the same day. Unlike a day case,
which is an elective episode, a same-day case may be admitted as an emergency. A one-day
case has admission and discharge dates on consecutive days, thereby spending one night in
hospital. An inlier has an LOS between the upper and lower bounds of the AR-DRG. A short-
stay outlier is admitted for more than one day and is discharged before their LOS exceeds
the lower bound. The RV for this case is a per diem weight. A long-stay outlier case has an
LOS which exceeds the upper bound. The RV for this case is a per diem weight.
Figure 15.3 Calculation of the hospital inpatient casemix-adjusted budget
286 Diagnosis-Related Groups in Europe
The peer-group base rate is calculated by dividing total costs of treating patients
in a particular hospital group by the total CMUs for that group. The blend rate
is used to mitigate any potentially destabilizing effects of the casemix budgetary
adjustments on hospitals.21 Since the early 2000s, the blend rate has progressively
increased – a trend which is likely to continue (see Table 15.6). An ADRG-based
budget is similarly calculated for day cases.
There is a lag between the time period to which the activity and cost data
relate and the application of the casemix budgetary adjustment. For example,
the 2009 casemix models used activity and cost data from 2008 and from the
fi rst half of 2009 to calculate the casemix budgetary adjustments that applied to
hospitals’ 2010 budgets.
The difference between the hospital’s casemix-adjusted budget and their his-
toric allocation is the casemix budgetary adjustment. A hospital will receive a
positive budget adjustment if the calculated casemix-adjusted budget is greater
than the historical budgets. The casemix models are revenue-neutral within
each peer group, so, where a hospital gains a positive adjustment, another
hospital(s) in the same peer group will incur a negative adjustment.
15.5.4 Quality-related adjustments
The national casemix models do not currently incorporate quality-related
adjustments.
Table 15.6 Inpatient and day-case blend rates, 2000–2010 casemix models
Model dataa Model Budgetary adjustment Blend rate
Inpatient (%) Day case (%)
1999/2000 2000 2001 15 5
2000/2001 2001 2002 15 5
2001/2002 2002 2003 20 10
2002/2003 2003 2004 20 10
2003/2004 2004 2005 20 20
2004/2005 2005 2006 30 30
2005/2006 2006 2007 40 40
2006/2007 2007 2008 50 50
2007/2008 2008 2009 60 60
2008/2009 2009 2010 70 70
2009/2010 2010 2011 80 80
Source: National Casemix Programme, 2010a.
Notes: A blend rate of more than 50 per cent implies that hospital budgets are determined
more by the costs of treating patients in the hospital’s peer group than based on the hospital’s
own costs. For example, a blend rate of 70 per cent implies that the calculated cost for a
casemix-adjusted discharge will be based on 70 per cent of the peer-group costs and 30 per
cent of the hospital’s own costs. aThe casemix models include a workload adjustment which is
calculated on the basis of activity in the latter half of year t and the fi rst half of year t+1. Thus,
a blend rate of 70 per cent is applied to the inpatient/day-case adjustments (68 per cent) and
the workload adjustment (2 per cent).
Ireland: A review of casemix applications 287
15.5.5 Main fi nancial incentives for hospitals
The explicit link between activity and cost data under the National Casemix
Programme inevitably means that hospitals face a number of potential fi nancial
incentives to infl uence measurement of activity and costs. They may attempt to
modify their coding practices to classify patients into AR-DRGs with higher
weights (that is, up-coding or ‘DRG creep’); submit cases to the incorrect
casemix model; or transfer patients as soon as possible to non-participating
hospitals or other institutions. However, most of these unintended consequences
can be monitored and controlled through regular data audits by the National
Casemix Programme and the ESRI.
Participation in the National Casemix Programme may be expected ex ante to
encourage hospitals to improve their relative effi ciency. However, this incentive
may be weakened to some extent by the time-lag between when the activity
was undertaken and the budgetary adjustment. The inclusion of a workload
adjustment in the casemix models reduces this lag to some extent (see subsection
15.5.3).
15.6 New/innovative technologies
Participating hospitals can make submissions on a case-by-case basis to the
National Casemix Programme for additional funding for high-cost drugs and
innovative treatments. This mechanism is not considered to have a signifi cant
impact on the introduction of new technologies. Rather, the availability of
capital investment funding would be considered to have a greater infl uence on
the adoption of new technologies.
15.7 Evaluation of the DRG system in Ireland
The application of DRGs in casemix funding has been assessed as part of a
number of national reviews within the Irish health sector. The 2001 National
Health Strategy considered the National Casemix Programme then in place to
be ‘the most developed system for assessing comparative effi ciency and for
creating incentives for good performance’ and committed to support it at
national and regional level (DoHC, 2001, p. 114). In 2004, the DoHC undertook
a ‘root and branch’ review of the National Casemix Programme, which
committed the Department to expanding the number of participating hospitals,
increasing the blend rate, and incorporating ‘sub-acute’ and ‘non-acute’ care.
Progress on each of these commitments is ongoing. The review also confi rmed
the adoption of ICD-10-AM as the national clinical coding standards in Ireland,
thereby confi rming the recommendations arising from assessments of the best
options available internationally for updating clinical coding and the DRG
system (Murphy et al., 2004; Aisbett et al., 2007).22
The parameters derived from the national casemix models can provide a useful
insight into how the operation of the National Casemix Programme infl uences
hospital behaviour. Figure 15.4 shows the variability in casemix-adjusted average
288 Diagnosis-Related Groups in Europe
costs for major teaching hospitals from the 2001–2010 inpatient casemix models.
Interestingly, there has been convergence in Group I hospitals’ average costs over
time and across hospitals, towards the mean group cost. This convergence may
be associated with the increasing blend rate, which may have encouraged
hospitals (particularly those with above-average costs) to change their cost
profi les in line with those of their peers.
15.8 Outlook: Future developments and reform
The application of DRGs in the Irish acute public hospital system has travelled
an interesting route. The uses of the DRG system have extended beyond hospital
reimbursement to informing the planning and reconfi guration of acute public
hospital services; facilitating hospital benchmarking; and adjusting and moni-
toring the hospital consultants’ workloads. Most recently, a review of charges
associated with private and semi-private treatment in public hospitals has
recognized the advantages of a DRG-based case payment system and recom-
mended the piloting of such a system (DoHC, 2010b).
The continued role for DRGs in hospital reimbursement has been recom-
mended in two reports. The fi rst – entitled Promoting performance-related services
Figure 15.4 Casemix-adjusted inpatient average cost and blend rate for Group I
hospitals, 2001–2010
Source: Based on data contained in Circulars prepared by the National Casemix Programme.
Notes: Each horizontal bar represents an observation on a hospital. The larger horizontal bar
represents the average for the hospital group. Data for 2001 were converted from Irish pounds
to Euros using an exchange rate of €1 = IR£0.787564.
Ireland: A review of casemix applications 289
and commissioned by the HSE – suggested implementing cost and volume
contracts over a period of ten years. This prompted the HSE to establish a
project team, which recognized the advancement of patient-level costing as a
key issue. Consequently, the HSE initiated a pilot project in several hospitals in
2009 to assess potential approaches to patient-level costing (National Casemix
Programme, 2010c). The second report by an Expert Group established by the
Minister for Health and Children in 2009 recommended a mix of hospital
payment mechanisms, incorporating DRG-based case payment and lump-sum
payments (Ruane, 2010). However, it remains to be seen how this will be
translated into policy.
Alongside the continued commitment to the use of DRGs in Ireland has been
the simultaneous achievement of improvements in quality, technical factors
and the system’s scope. As data systems improve and the technical capability to
implement increasingly sophisticated measurement systems increases, new and
better DRG systems are expected to be developed in the future. However, the
core objective of using DRGs in Ireland remains the same: achieving greater
transparency in relating resource use to outputs within the acute hospital sector.
15.9 Notes
1 Assistance provided by Brian Donovan and Mark O’Connor is gratefully
acknowledged. The authors are also obliged to Wilm Quentin and Conrad Kobel for
comments on earlier drafts.
2 Due to profi t repatriation by multinational companies located in Ireland, GNP is
considered to be a more appropriate indicator of economic performance than GDP
(Nolan, 2005; McDaid et al., 2009).
3 A private patient may occupy a public bed only if admitted as an emergency when a
designated private bed is not available (Government of Ireland, 1991a, b, c). The
Minister for Health and Children approves public/private bed designation (DoHC,
1999).
4 It has been estimated that there are 19 private hospitals operating in Ireland (Brick
et al., 2010).
5 Between 2001 and 2008, all those aged 70 years and over, irrespective of income,
received a medical card (McDaid et al., 2009; Citizens Information Board, 2011).
6 Some changes in data collection partly account for increased activity reported
(see, for example, ESRI-HRID, 2008b).
7 The procedure classifi cation system used in Ireland (namely, ICD-10-AM) is more
commonly referred to as the Australian Classifi cation of Health Interventions (ACHI).
8 ADRGs are similar to ‘base-DRGs’ used in other countries.
9 A hospital’s AFSs are subject to audit, typically by the Comptroller and Auditor
General (National Casemix Programme, 2010b).
10 Updates to the classifi cation system are covered within a contract between Ireland
and Australia.
11 For details of the variables required for grouping, see Commonwealth Department of
Health and Ageing (2008).
12 Until 2010, the HIPE scheme collected up to 19 additional diagnoses and up to
20 procedures.
13 Full details and examples of the grouping algorithm are contained in the AR-DRG
defi nitions manual (Commonwealth Department of Health and Ageing, 2008).
290 Diagnosis-Related Groups in Europe
14 A discharge may be categorized as ‘unassignable to MDC’ or to an ‘error DRG’ where
it cannot be directly assigned to one of the 24 MDCs.
15 An ADRG is the fi rst three characters of an AR-DRG (for example, B78 is the ADRG
associated with AR-DRG B78B).
16 The PCCL ranges from 0 (lowest complexity level) to 4 (highest). A high PCCL
indicates the presence of additional, unrelated CCs that are expected to affect length
of stay and cost.
17 For further information on HCAT, see the information leafl et available from the ESRI
web site (http://www.esri.ie/health_information/hipe/clinical_coding/data_quality/
HCAT_Info_Leafl et_2007 , accessed 10 July 2011).
18 What are termed ‘relative values’ in Ireland may be referred to as cost weights in
other countries.
19 An average AR-DRG has an RV of 1 and more costly AR-DRGs would have an RV
greater than 1.
20 For these calculations, length-of-stay values are transformed into natural logarithms.
21 For example, a blend rate of 70 per cent implies that the calculated cost for a casemix-
adjusted discharge will be based on 70 per cent of the peer-group costs and 30 per
cent of the hospital’s own cost.
22 Also, a review of the clinical coder training programme and data quality and audit
procedures within the HIPE was commissioned in 2004 by the ESRI and undertaken
by Michelle Bramley and Beth Reid, from the University of Sydney.
15.10 References
Aisbett, C., Wiley, M., McCarthy, B., Mulligan, A. (2007). Measuring Hospital Casemix:
Evaluation of Alternative Approaches for the Irish Hospital System. Dublin: Economic and
Social Research Institute (ESRI Working Paper No. 192).
Brennan, N. (2003). Report of the Commission on Financial Management and Control Systems
in the Health Service (The Brennan Report). Dublin: The Stationery Offi ce.
Brick, A., Nolan, A., O’Reilly, J., Smith, S. (2010). Resource Allocation, Financing and Sustain-
ability in Health Care: Evidence for the Expert Group on Resource Allocation and Financing
in the Health Sector. Dublin: Department of Health and Children and Economic and
Social Research Institute.
Citizens Information Board (2010). Charges for Hospital Services. Dublin: Citizens
Information Board (http://www.citizensinformation.ie/en/health/hospital_services/
hospital_charges.html, accessed 11 April 2011).
Citizens Information Board (2011). Medical Cards. Dublin: Citizens Information Board
(http://www.citizensinformation.ie/en/health/entitlement_to_health_services/
medical_card.html, accessed 5 April 2011).
Colombo, F., Tapay, N. (2004). Private Health Insurance in Ireland: A Case Study. Paris:
Organisation for Economic Co-operation and Development (OECD Health Working
Paper No. 10).
Commission on Health Funding (1989). Report of the Commission on Health Funding.
Dublin: The Stationery Offi ce.
Commonwealth Department of Health and Ageing (2002). Australian Refi ned Diagnosis
Related Groups Version 5.0 Defi nitions Manual. Canberra: Commonwealth Department
of Health and Ageing.
Commonwealth Department of Health and Ageing (2004). Australian Refi ned Diagnosis
Related Groups Version 5.1 Defi nitions Manual. Canberra: Commonwealth Department
of Health and Ageing.
Ireland: A review of casemix applications 291
Commonwealth Department of Health and Ageing (2008). Australian Refi ned Diagnosis-
Related Groups Version 6.0 Defi nitions Manuals. Canberra: Commonwealth Department
of Health and Ageing.
DoHC (1999). White Paper – Private Health Insurance. Dublin: The Stationery Offi ce.
DoHC (2001). Quality and Fairness: A Health System for You. Dublin: The Stationery Offi ce.
DoHC (2009a). Health in Ireland: Key Trends 2009. Dublin: Department of Health and
Children.
DoHC (2009b). Value for Money and Policy Review of the Economic Cost and Charges Associated
with Private and Semi-Private Treatment Services in Public Hospitals – Interim Report.
Dublin: Department of Health and Children (http://www.dohc.ie/publications/pdf/
vfm_review_economic_cost_interim_report ?direct=1, accessed 31 May 2010).
DoHC (2010a). Health in Ireland: Key Trends 2010. Dublin: Department of Health and
Children.
DoHC (2010b). Value for Money and Policy Review of the Economic Cost and Charges Associated
with Private and Semi-Private Treatment Services in Public Hospitals. Dublin: Department
of Health and Children (http://www.dohc.ie/publications/pdf/vfm_review_economic_
cost_fi nal_report , accessed 17 December 2010).
ESRI-HPID (2007). Activity in Acute Public Hospitals in Ireland: Annual Report 2004. Dublin:
Economic and Social Research Institute, Health Policy and Information Division.
ESRI-HRID (2008a). Activity in Acute Public Hospitals in Ireland: Annual Report 2005. Dublin:
Economic and Social Research Institute, Health Research and Information Division.
ESRI-HRID (2008b). Activity in Acute Public Hospitals in Ireland: Annual Report 2006. Dublin:
Economic and Social Research Institute, Health Research and Information Division.
ESRI-HRID (2010). Activity in Acute Public Hospitals in Ireland: Annual Report 2008. Dublin:
Economic and Social Research Institute, Health Research and Information Division.
Government of Ireland (1991a). Health (Amendment) Act, 1991. Dublin: Offi ce of the
Attorney General (www.irishstatutebook.ie/1991/en/act/pub/0015/index.html,
accessed 26 May 2010).
Government of Ireland (1991b). Health Services (In-Patient) Regulations, 1991 (S.I. No.
135/1991). Dublin: Offi ce of the Attorney General (www.irishstatutebook.ie/1991/
en/si/0135.html, accessed 26 May 2010).
Government of Ireland (1991c). Health Services (Out-Patient) Regulations, 1991 (S.I. No.
136/1991). Dublin: Offi ce of the Attorney General (www.irishstatutebook.ie/1991/
en/si/0136.html, accessed 26 May 2010).
HCFA (undated). Diagnosis Related Groups Defi nitions Manual, Version 16.0. St. Paul, MN:
3M Health Information Systems.
HSE (2008). Consultant Contract 2008 – Measurement of Public–Private Mix. Kildare: Health
Service Executive (http://www.hse.ie/eng/staff/HR/tandc/Consultant_Contract_2008_
_Measurement_of_Public_Private_Mix , accessed 4 August 2010).
HSE (2010a). Casemix Budget Outturns for 2009. Kildare: Health Service Executive (http://
www.hse.ie/eng/services/Publications/corporate/2009casemix.html, accessed 12 May
2010).
HSE (2010b). HealthStat – Supporting High Performance. Kildare: Health Service Executive
(http://www.hse.ie/eng/staff/Healthstat/about/, accessed 4 August 2010).
HSE (2010c). Hospitals and Cancer Control. Kildare: Health Service Executive (http://www.
hse.ie/eng/services/Find_a_Service/HospsCancer/, accessed 3 August 2010).
HSE (2010d). Integrated Services. Kildare: Health Service Executive (http://www.hse.ie/eng/
about/Who/is/Integrated_Services.html, accessed 26 April 2010).
McDaid, D., Wiley, M., Maresso, A., Mossialos, E. (2009). Ireland: Health system review.
Health Systems in Transition, 11(4):1–267.
Murphy, D., Wiley, M.M., Clifton, A., McDonagh, D. (2004). Updating Clinical Coding in
Ireland: Options and Opportunities. Dublin: Economic and Social Research Institute (ESRI).
292 Diagnosis-Related Groups in Europe
National Casemix Programme (2010a). Casemix Manual: Casemix Annual (Technical) Report
2010 – Part 4. Dublin: National Casemix Programme.
National Casemix Programme (2010b). Costing Manual: Casemix Annual (Technical) Report
2010 – Part 5. Dublin: National Casemix Programme.
National Casemix Programme (2010c). Technical Aspects of the National Casemix Programme
– Part II. Dublin: National Casemix Programme.
National Casemix Programme (2010d). C1-10 Casemix Circular Tables. Dublin: National
Casemix Programme.
Nolan, A. (2005). Health: funding, access and effi ciency, in J. O’Hagan, C. Newman, eds.
The Economy of Ireland: National and Sectoral Policy Issues. Dublin: Gill and Macmillan.
Nolan, B., Wiley, M.M. (2000). Private Practice in Irish Public Hospitals. Dublin: Oak Tree
Press in association with the Economic and Social Research Institute.
OECD (2009). OECD Health Data 2011. How does Ireland Compare. Paris: Organisation for
Economic Co-operation and Development (http://www.oecd.org/dataoecd/45/53/
43216301 , accessed 10 July 2011).
O’Reilly, J., Wiley, M. (2010). Who’s that sleeping in my bed? Potential and actual
utilization of public and private in-patient beds in Irish acute public hospitals. Journal
of Health Services Research and Policy, 15:210–14.
Prospectus (2003). Audit of Structures and Functions in the Health System. Dublin: The
Stationery Offi ce.
Robbins, G., Lapsley, I. (2008). Irish voluntary hospitals: an examination of a theory of
voluntary failure. Accounting, Business & Financial History, 18:61–80.
Ruane, F. (2010). Report of the Expert Group on Resource Allocation and Financing in the Health
Sector. Dublin: Department of Health and Children.
Wiley, M.M. (2005). The Irish health system: developments in strategy, structure, funding
and delivery since 1980. Health Economics, 14:S169–S86.
chapter s i x t e e n
NordDRG: The benefi ts
of coordination
Miika Linna and Martti Virtanen
16.1 Introduction
Since the early 1990s the Nordic countries (Finland, Sweden, Denmark, Norway
and Iceland) have been experimenting with patient classifi cation systems. This
led to a common Nordic patient classifi cation system known as ‘NordDRG’ in
the mid-1990s. NordDRG is a diagnosis-related group (DRG) grouper which
emulates Health Care Financing Administration (HCFA)-DRG Version 12, using
defi nitions based on the WHO International Classifi cation of Diseases 10th
revision (ICD-10) and the NOMESCO (Nordic Medico-Statistical Committee)
Classifi cation of Surgical Procedures (NCSP). The fi rst grouper was fi nished in
1996. The grouper is updated yearly, according to the NordDRG maintenance
process. This chapter explores the methods the Nordic countries have used
to establish a grouping system which is unique on the DRG landscape and
functions across the different countries. The implementation of the NordDRGs
in Sweden, Finland and Estonia will be examined more closely in the chapters
that follow.
During the early 1990s Finland developed a ‘FinDRG’ grouper based on the
HCFA grouping system, which automatically converted Finnish ICD-9 diagnosis
and procedure codes into HCFA-DRGs (Salonen et al., 1995; Linnakko, 2001).
At the same time, other Nordic countries were also using DRGs, albeit somewhat
unsystematically. The groupers used were mainly the 3M™ All Patient (AP)-
DRG grouper or the United States HCFA-DRG grouper. DRG application in the
Nordic countries was not at this time directly linked to hospital payment, but
DRGs were used for benchmarking hospitals, health system evaluation or
statistical reporting purposes. However, some of the Swedish county councils
used the 3M™ AP-DRGs and the HCFA-DRG grouper in order to reimburse
hospitals or inpatient care in some specialties (Håkansson & Gavelin, 2001).
In Finland, FinDRGs were mainly used for managerial purposes rather than for
hospital reimbursement (Linna, 1997).
294 Diagnosis-Related Groups in Europe
Since the beginning of 1996 all Nordic countries decided to start using the
ICD-10 classifi cation for clinical diagnoses. However, it was impossible to
employ a satisfactory conversion from ICD-9 (used for example, by the HCFA
and FinDRG groupers) to ICD-10 and, in addition, the Nordic countries were
in the process of implementing the new NCSP. Futhermore, the use of DRGs
for hospital reimbursement in Sweden was increasing and several Finnish
municipalities had expressed their interest in using hospital service defi nitions
based on DRGs. Thus, the NOMESCO assigned the WHO Collaborating Centre
for the Classifi cation of Diseases in the Nordic countries to design a new cross-
country DRG system, namely the NordDRG.
The national health authorities and associations in Finland (the Finnish
Association of Local and Regional Authorities), Sweden (the National Board of
Welfare and Health) and Iceland (the Ministry of Health and Social Insurance)
started the NordDRG project, later accompanied by the Danish and Norwegian
ministries of health. These organizations established the Nordic Casemix
Centre,1 which is responsible for the distribution, maintenance and development
of the NordDRG grouper.
With the introduction of the system in 1996, separate country versions for
Sweden and Finland were released, based on common defi nitions. Later,
Denmark (2000), Iceland and Norway (2002) were also included in the set of
country versions. In 2002 Denmark decided to leave the joint project in order to
build its own system of DRGs (DkDRG), based on different primary classifi ca-
tions (procedures) and major revisions to the NordDRG grouping defi nitions
(Hansen & Nielsen 2001) (Figure 16.1). In 2003 Estonia joined the NordDRG
consortium. However, until 2009 they used the 2003 NordDRG grouper, particu-
larly because the Estonian procedure classifi cation (based on the NCSP) was not
updated during this period. There is also an agreement with the Ministry of
Health of the Republic of Latvia regarding the right to use NordDRGs in Latvia.
Aside from the national NordDRG versions there is also a NordDRG version
based on the common Nordic components of ICD-10 and the NCSP. This
version is used in Iceland. In 2003 an extended NordDRG version was developed
that also includes national modifi cations to the grouping logic, using national
codes that are unique to some of the countries (NCSP+).
16.2 Development and updates of the NordDRG system
The annual maintenance and updating of the NordDRG system is carried out
according to a specifi ed protocol and a fi xed timetable. This updating process is
intended to meet the emerging needs of the main stakeholders of the system
within the Nordic countries. Modifi cations are validated with clinical and cost
data to ensure that both economic and medical (clinical) homogeneity are
retained or improved. Each country tests and implements these modifi cations
separately, using their own data.
Suggestions for annual system updates are administered through an expert
network, which consists of nominated experts from each participating country.
The network is the main advisory group and platform for discussions relating to
the maintenance, performance evaluation and development of the NordDRG
NordDRG: The benefi ts of coordination 295
system. Meetings of the expert network are arranged twice per year. Based on
the experts’ suggestions, the NordDRG steering group makes the fi nal decision
(by consensus) regarding any changes to be made to the system.
The common NordDRG grouper is the reference grouper for inpatient care.
Some modifi cations have been incrementally incorporated to better refl ect the
Nordic health care system context. Except for Denmark, all of the country-
specifi c modifi cations represent only minor adaptations in the grouping algo-
rithm for inpatient care. However, the incorporation of outpatient and day care
differs markedly. From 2004 to 2007 the grouping of outpatient care was fairly
similar in the Finnish and Swedish versions. In 2007 the Swedish version incor-
porated large revisions to the conservative cases in outpatient care. The opera-
tive cases/groups remain quite similar in the Swedish, Finnish and Norwegian
versions. Norway integrated DRGs for outpatient care into its NordDRG coun-
try version in 2010 and Estonia did so during 2011.
General system information is available through an edited and open Nord-
DRG discussion forum.2 In addition, the forum is used for proposing updates
and answering questions relating to the NordDRG system, and changes made
to the NordDRG system each year are also documented.
Each year in which changes are necessary, two evolution versions are pro-
duced. The fi rst (planning) version includes changes of the grouping logic of
some DRGs for the next year. These changes may include entirely new groups,
Figure 16.1 The development of the NordDRGs country versions
296 Diagnosis-Related Groups in Europe
splitting of DRGs into sub-groups or changes in the assignment rules (patient
case being entered into different DRGs). The fi rst version is based on the primary
classifi cations of the concurrent year and can therefore be used to test the
effect(s) of the changes. The second (production) version is based on the pri-
mary classifi cations for the next year, but there are no new changes to the
assignment logic of the NordDRG. If no errors are revealed, the second version
will be accepted as the offi cial production version for the next year.
In Finland all proposed alterations are discussed at the Finnish DRG Centre
in cooperation with an expert physician from each medical specialty. They
select which of the suggestions are relevant for the system update. If necessary,
the suggestions are further discussed with the NordDRG expert network.
In Sweden the National Board of Health and Welfare is responsible for
maintaining and developing the Swedish version of the NordDRG system. The
National Board works together with the Swedish Association of Local Authorities
and Regions (SALAR), representing the 21 county councils and regions in
Sweden. There are also a number of expert groups in which the hospitals are
represented. The test data for the updating process originate from the National
Patient Register. For validating the homogeneity of DRGs, patient-level cost
data are supplied by the National Cost Database. This database contains cost-
accounting information from several hospitals.
Responsibility for developing and updating the DRG system in Estonia relies
on the Estonian Health Insurance Fund (EHIF) and these tasks are carried out
according to the DRG development plan.
The calculation of cost weights is accomplished separately in each country,
using the NordDRGs and the respective cost data. In Finland the cost weights
for the Finnish grouper have been calculated annually, based on patient-level
cost data from the hospitals of the largest hospital district in Finland, Helsinki
and Uusimaa, covering approximately 30 per cent of the specialized care setting
in Finland (Mikkola & Linna, 2002). In Estonia the calculation of cost weights
and DRG prices is based on the prices of the health services, not on the actual
resource consumption. The main source used to develop the Estonian version is
the case costing database of the EHIF. Expertise is also drawn upon from medical
professionals’ associations and health care providers. In Sweden the Centre for
Patient Classifi cation is responsible for the updating process and for calculat-
ing the cost weights. It is not mandatory to use the national weight sets in
Sweden; there are also local weights in use in some of the counties. National
weights are based on the national case costing database (using ‘bottom-up’
costing approach), which comprises 62 per cent of all inpatients in Sweden.
Case costing data are also collected for outpatient care, and weights for day
surgery and visits are also based on ‘bottom-up’ costing data.
The Finnish versions of NordDRGs only included groups for inpatient care
until 2004. From then on, the hospital districts have gradually moved away
from using the classic (only inpatient and day-care) grouper to using the
outpatient DRGs.
Between 2003 and 2009, there were two versions of NordDRGs in effect in
Sweden: a full version, which handled both inpatient care and day surgery
(including intraluminal endoscopies); and a classic version for inpatient care
only. Aside from these, a separate NordDRG grouper that included outpatient
NordDRG: The benefi ts of coordination 297
care also existed. From 2010, there will be just one version of NordDRGs for
handling inpatient care and outpatient care. Estonia has been using the same
version of NordDRGs since the year 2000. This is partly because the procedure
classifi cation has not been updated during this time. In 2011 DRGs for ‘short-
term therapy’ (day care) were introduced.
16.3 Differences in diagnosis and procedure classifi cation
The responsibility for updating the NCSP lies with the Nordic Centre, which
maintains the aforementioned electronic discussion forum for the discussion
of changes (see section 16.2). The reference group makes recommendations
on changes to the Board of the Nordic Centre, which makes any formal
decisions.
To update the NCSP, the reference group requires certain information for tak-
ing update proposals into consideration. Proposals should be approved and
submitted by a responsible national classifi cation body. The proposal can be
prepared by the relevant medical professionals’ association. The proposal
should include a description of the new method and an account of its indica-
tions. In addition, it should include a rough estimate of how many procedures
are performed per year in the country submitting the proposal, or the expected
annual number in the near future. References should be included to the rele-
vant literature (preferably accessible on the Internet), documenting the fact
that proposed new codes represent established procedures and are not purely
experimental in nature, giving a broad overview of the indications and tech-
niques of any proposed new code.
However, the classifi cations of diagnoses and procedures are all slightly
different, varying across the countries using NordDRGs. Finland has added a
number of 5th-character codes to the ICD-10, but mostly there are no confl icts
with the original WHO version. However, the country has made a number of
updates to the NCSP (for example, diagnostic radiology, including ultrasound
examination, therapeutic radiology and rehabilitation interventions were
added). The Finnish Full DRG version (grouping both inpatients and outpatients)
has been expanded to outpatient care, with a number of groups for hospital
outpatient visits. This is based on the common Nordic model, in which so-called
day surgery has its own groups per visit. The national additions refer mostly to
the expensive radiology procedures, usually coded in the patient administration
systems in Finland. The Finnish model still uses the original concept, from the
common Full NordDRGs system, of ‘short-term therapy’ (in and out during the
same calendar day) instead of an outpatient approach. Expensive medication
has been another issue for consideration, but – due to the lack of systematic
data collection – this has produced only temporary solutions.
The Swedish ICD-10 includes a number of 5th-character additions to the orig-
inal WHO ICD-10. Similar to the Finnish version, these are mostly compatible
with the WHO original. Sweden is the only Nordic country that fully applies
the external cause coding of the ICD-10 classifi cation to 5th-character level
(~25 000 external cause codes). Sweden uses the NSCP, along with a national
classifi cation of conservative interventions that are especially important for
298 Diagnosis-Related Groups in Europe
DRGs in outpatient settings. Diagnostic radiology is much less developed in
Sweden than its Finnish counterpart but minor (short-stay) procedures and
rehabilitation, for example, are more detailed in the procedure classifi cation
than in Finland. The Swedish Full NordDRG version is expanded to include all
specialized outpatient care (conservative and surgical), along with psychiatry
and rehabilitation.
Iceland is using a version of the expanded NCSP called NCSP+ for procedure
classifi cation. NCSP+ is based on the different additions to the NCSP in different
Nordic countries. NCSP+ was developed for the NordDRG process as a tool that
links the different national versions of procedure classifi cations together, so
that the rules for different counties can be defi ned together and applied through
the NCSP+ mapping to all national versions. The ICD-10 in Iceland, similarly to
the Estonian version, is a direct translation of the WHO ICD-10.
Norway has modifi ed the NCSP only when the common version has been
updated. However, there is a separate classifi cation of ‘non-surgical’ interventions
that can be used together with the NCSP. It covers important areas, even for the
classic NordDRG, and thus also includes important features for the outpatient
groups. Norway has developed an expanded version of the Full NordDRG
system, which is closely related to the Swedish version.
Denmark has revised the ICD-10 coding by replacing the ‘dagger-asterisk’
system with a large number of predefi ned combination codes. External cause
codes were also replaced by the NOMESCO Classifi cation of External Causes of
Injury (NCECI). The procedure classifi cation is mostly taken from the NCSP.
Some countries have added a system of ‘nursing interventions’ (comprising
a number of codes) to the NordDRG system. For example, the Finnish and
Swedish Full NordDRG version includes a number of DRG groups for (outpatient-
based) nursing procedures.
16.4 Grouping process
The information required by the NordDRG grouper includes the following
items: main diagnosis, secondary diagnoses (a list of diagnoses), procedures
(a list of procedures), age, gender, mode of discharge and length of stay.
Internally, the grouping algorithm uses various predefi ned sets and sub-
groupings of diagnoses and procedures to determine the properties which affect
the grouping of each case (for example, the complication properties for diag-
noses and procedures). This information is available in the defi nition tables for
the NordDRG logic. Diagnoses and procedures that have an effect on the group-
ing are clustered into intermediate groups called ‘categories’ and ‘properties’.
Each code belongs to only one category, but it may have several properties.
Properties relating to co-morbidities and/or complications (CCs) are binary,
that is, having only two levels. About 75 per cent of the inpatient DRGs are
divided into ‘non-complicated’ DRGs (without CCs) and ‘complicated’ DRGs
(with CCs). Complicated cases are defi ned based on secondary diagnoses or in
some cases procedures undertaken because of complication(s).
Operating room (OR) properties are binary in the Classic grouper, but in the
Full grouper can have three values. Values 1 and 0 indicate whether a surgical
NordDRG: The benefi ts of coordination 299
procedure has been undertaken or not. According to this information, cases are
assigned into ‘surgical’ and ‘medical’ DRGs, respectively. Procedures that are
important in outpatient setting but do not affect DRG assignment of hospital
inpatients have OR-property 2 (OR=2). In the case of hospital inpatients, OR=2
has no impact on the DRG assignment. The grouper algorithm returns three
codes: the major diagnostic category (MDC) code, the NordDRG group code
and a separate return code which indicates the outcome of the grouping,
consistency checks and the reason for unsuccessful grouping.
16.5 Reimbursement via DRGs
The Finnish version of NordDRGs was initially introduced at the Helsinki
University Central Hospital at the beginning of 1998. In 2001, fi ve Finnish
districts were employing NordDRGs to some extent in their pricing of hospital
treatment. Today, 13 out of 21 districts have incorporated DRG-based pricing,
but the methodology still varies greatly because regulations or even guidelines
for hospital reimbursement are lacking at national level in Finland. Therefore,
each district may determine the hospital payment method autonomously
(Häkkinen & Linna, 2006).
In Sweden, different DRG systems (but mainly the HCFA-DRGs) have been
used since the beginning of the 1990s. Stockholm County Council imple-
mented DRGs as a payment system for inpatient care in 1992. The DRG system
was developed as a process of cooperation between the National Board of
Health and Welfare and the county councils, and the adoption of the DRGs has
mainly been the concern of the county councils. This background – together
with a tradition of a high degree of local autonomy – resulted in a situation
in which central coordination on DRG-related issues was relatively weak
during the 1990s. Since 1999 the National Board of Health and Welfare has
coordinated DRG matters. All county councils and regions now use DRGs to
some extent.
The DRG system introduced in Estonia in 2003 replaced the previous fee-for-
service and per diem payments for hospital reimbursement. In 2011 the update
of the current DRG grouping version took place, with the aim of increasing
clinical relevance and resource homogeneity in DRGs by introducing them for
‘short-term therapy’ (day care), taking into account the different costs according
to the patient’s length of stay.
In Norway the funding of hospital care has largely comprised a mixture of
global budgeting and activity-based funding (DRGs) since 1997. The implemen-
tation of the 1997 reform changed the format of hospital fi nancing from block
grants to a combination of block grants and activity-based reimbursement
using NordDRGs.
In Denmark, the Ministry of the Interior and Health introduced casemix rates
for the reimbursement of patients who received basic-level treatment outside
of their home county (‘free-choice patients’) in 2000. At the same time, the
voluntary 90/10 payment model was introduced in the counties. In the 90/10
model, 90 per cent of the predicted health care delivery/production costs were
allocated to hospitals and the remaining 10 per cent of the hospital funding
300 Diagnosis-Related Groups in Europe
was to be allocated based on health care activities carried out, as measured by
DRGs.
16.6 Notes
1 For more details see the Nordic Casemix Centre web site (www.nordcase.org, accessed
10 July 2011).
2 See the Nordic Centre for Classifi cations in Health Care dedicated forum web site for
further details (www.norddrg.net/norddrgforum, accessed 10 July 2011).
16.7 References
Håkansson, S., Gavelin, C. (2001). Casemix in Sweden. Experiences of DRGs 1985–2001,
in F. France, I. Mersents, M. Closon, J. Hofdijk. Casemix: Global Views, Local Actions.
Evolution in Twenty Countries. Amsterdam: IOS Press.
Häkkinen, U., Linna, M. (2006). DRGs in Finnish health care. Euro Observer, 7(4):7–8.
Hansen, P., Nielsen, S. (2001). Casemix in Denmark: Status on the development and use
of the Nord ‘DK’ DRG, in F. France, I. Mersents, M. Closon, J. Hofdijk. Casemix: Global
Views, Local Actions. Evolution in Twenty Countries. Amsterdam: IOS Press.
Linna M. (1997). DRG:n käyttö sairaaloiden tuottavuus- ja tehokkuusvertailuissa [The use
of DRGs in hospital productivity comparisons], in M. Neonen. DRG suomalaisessa
terveydenhuollossa [DRGs in Finnish Health Care]. Helsinki: National Institute for
Health and Welfare (STAKES).
Linnakko E. (2001). Casemix in Finland, in F. France, I. Mersents, M. Closon, J. Hofdijk.
Casemix: Global Views, Local Actions. Evolution in Twenty Countries. Amsterdam: IOS
Press.
Mikkola, H., Linna, M. (2002). Utilisation des DRGs dans le système hospitalier fi nlandais.
Revue Médicale de l’Assurance Maladie, 33(1):37–43.
Salonen, M., Häkkinen, U., Keskimäki, I., Linna, M. (1995). DRG-ryhmien kustannuspainot
suomalaisella aineistolla [DRG Cost Weights Based on Finnish Cost-Accounting Data].
Helsinki: National Institute of Health and Welfare (STAKES) (Aiheita 35).
chapter s e v e n t e e n
Estonia: Developing
NordDRGs within social
health insurance
Kristiina Kahur, Tõnis Allik,
Ain Aaviksoo, Heli Laarmann
and Gerli Paat
17.1 Hospital services and the role of DRGs in Estonia
17.1.1 The Estonian health care system
The Estonian health care system is built on a platform of compulsory, solidarity-
based insurance and universal access to health services made available by
providers that operate under private law (Koppel et al., 2008).
In 2008 Estonia had one of the lowest shares (6.1 per cent) of expenditure on
health care relative to gross domestic product (GDP) in Europe (NIHD, 2008).
Estonian health expenditure has remained stable over time, with only small
variations due to changes in the economic environment. As its main system-
level input and output characteristics are comparable with more affl uent coun-
tries, Estonia is often described as a country with a very cost-effi cient health
care system (Björnberg et al., 2009).
Since the country’s independence in 1991, the Estonian health system has
undergone two major shifts: fi rst, from a centralized, state-controlled system to
a decentralized one; and second, from a system funded by the state budget to
one funded through social health insurance (SHI) contributions (Koppel et al.,
2008). In 1992, following the introduction of health insurance and the estab-
lishment of autonomous providers, health care professionals ceased to be pub-
lic employees, lost their civil service status and began to work under private
labour regulations. The restructuring of the health system has taken place in
several phases. The current organizational and management principles were
302 Diagnosis-Related Groups in Europe
established between 1999 and 2002 by legislation intended to re-centralize
some health system functions.
The Ministry of Social Affairs and its agencies – the State Agency of Medicines
(SAM), the Health Board and the National Institute for Health Development
(NIHD) – are responsible for the general stewardship and management of the
health care system, as well as for health policy development.
The State budget contributes about 11.5 per cent of total health expenditure,
mainly for the fi nancing and management of public health services, emergency
medical care of uninsured people and emergency ambulance services. Local
municipalities have a minor, somewhat voluntary role in organizing and
fi nancing health services (Koppel et al., 2008). This means that local munici-
palities have no defi ned responsibility to cover health care expenditure and,
therefore, fi nancing practices vary widely. Some local governments provide pri-
mary care providers with fi nancial support, while some partially reimburse
pharmaceutical expenses and nursing care costs for low-income households
and for the elderly. In addition, health care providers that treat uninsured
people might receive some reimbursement from local municipalities for certain
expenditures, to varying degrees depending on the municipality. The majority
share of fi nancing for health care services is contributed by the public indepen-
dent legal body, the Estonian Health Insurance Fund (EHIF), which contributes
about 64.8 per cent of total health expenditure (NIHD, 2008). EHIF revenues are
pooled from earmarked payroll taxes. Being effectively a single purchaser of
care for most providers, the EHIF has gradually become one of the main actors
driving developments in the health system. Private spending comprises about
20 per cent of total health expenditure, mostly in the form of co-payments for
pharmaceuticals and dental care. Private insurance is almost non-existent in
Estonia (0.3 per cent) (NIHD, 2008).
All actors in the Estonian health care market are public or private organiza-
tions operating under public or private law, which indicates that direct respon-
sibility for provider performance has been delegated by the Ministry of Social
Affairs and the municipalities to the hospital supervisory boards. With regard to
purchasers of health care, the main actors are public organizations, such as the
EHIF, the Health Board, and the NIHD. The latter two bodies are agencies of the
Ministry of Social Affairs. The Health Board acts as a public purchaser of ambu-
lance service providers and ensures suffi cient national coverage. The NIHD is
the main purchaser of public health services and is responsible for the imple-
mentation of all national public health programmes and strategies. However,
the planning and coordination of the programmes is carried out by the Minis-
try of Social Affairs.
The EHIF is the main purchaser of health services. EHIF contracts evolved
over a decade of well-established relationships on an equal footing with the
service providers. At the beginning of the 1990s the contract content was rather
unsophisticated and only the capped total costs were agreed. Currently, the
contracts include agreements on rights and obligations of the parties con-
cerned, service quality and access, as well as fi nancial reporting requirements
and a detailed cost- and volume-based fi nancial component.
Estonia: Developing NordDRGs within social health insurance 303
17.1.2 Hospital services in Estonia
An important characteristic of the Estonian hospital system is that since
2001 all hospitals operate under private law in the form of limited liability
companies or foundations (Koppel et al., 2008). All hospitals own their capital
assets and they are independent in their management decisions. Personnel
who work in hospital-based departments have contracts with the hospital and
are therefore salaried employees. Between 1991 and 2000 the number of doc-
tors fell by 24 per cent, from 5500 to 4190, and the number of nurses by 14 per
cent, from 9900 to 8500 (Jesse et al., 2004). Although the number of doctors
and nurses continued to decrease after 1998, the ratio per 1000 inhabitants
slightly increased, due to a parallel reduction in the size of the population
(Koppel et al., 2008).
However, most hospitals are owned (or founded) by the state, local govern-
ments or public legal bodies, and thus effectively act as public hospitals. Estonia
has therefore preserved public ownership of the hospital network, but has
introduced management concepts specifi c to the private sector. This has created
a framework in which public hospitals are run as networks or integrated pro-
viders and as true business entities, with management incentives geared at
effi cient fi nancial performance.
In many instances, the hospital has multiple owners, for example a number
of municipalities, or the state and municipalities jointly owning one hos-
pital. A few hospitals owned by private entities provide specifi c services (such as
gynaecology, obstetrics, rehabilitation, plastic surgery, and so on). The rela-
tionship between the EHIF and all hospitals is based on contracts. Owners
(including public ones) can infl uence hospital activity through supervisory
bodies or capital investment decisions.
In 2003 the Government approved the Hospital Network Development
Plan (HNDP), which drew up a list of hospitals that serve the public interest and
are therefore eligible for state aid. The Plan stipulated that hospitals are divided
into regional, central, general, local, special, rehabilitation care and nursing
care hospitals. Regional, central, general and local hospitals are acute care hos-
pitals providing treatment for acute diseases requiring active medical interven-
tion. Special care hospitals provide inpatient services in orthopaedics, vascular
surgery, plastic surgery, psychiatry, obstetrics, gynaecology and otorhinolaryn-
gology.
All hospitals need to be licensed by the Health Board. Differences in
requirements according to hospital levels are mainly in the form of a minimum
set of medical specialties that certain levels of hospitals must represent.
Each acute care hospital covers a certain area or region. The location has been
chosen so that acute care services are available to everyone at a distance of
70 km or 60 minutes’ drive; the Government approved the HNDP based on this
principle. In order to ensure equal availability of specialist medical services,
the HNDP foresees the existence of 19 acute care hospitals, including 11 general,
4 central, 3 regional and 1 local hospital (see Figure 17.1).
In the period 2000–2006 the number of hospital beds decreased by 20 per
cent, from 9828 to 7588 beds, and the structure of beds by specialty changed
304 Diagnosis-Related Groups in Europe
signifi cantly; that is, the proportion of nursing care beds increased remarkably,
while the proportion of acute care beds decreased (see Table 17.1).
At the end of 2006 there were 55 hospitals in Estonia: 3 regional, 4 central,
12 general, 6 local, 7 special, 3 rehabilitation care and 20 nursing care hospitals,
with a total of 7588 beds.
The reduction of acute beds has been related to the establishment of a
hospital licensing system. As a result, small hospitals – hitherto predominantly
providing long-term care – have lost their licence to provide acute care and
have been turned into nursing homes. Other hospitals have been transformed
into outpatient care centres providing specialist ambulatory care (Koppel et al.,
2008).
In Estonia the range of activities and services in the hospital sector includes
specialized outpatient care (including day care and day surgery) and inpatient
care (including acute care, follow-up care, nursing care and rehabilitation).
The total number of discharges, outpatient contacts, and insured individuals
that used specialized medical care services during 2006 and 2009 are shown in
Table 17.2.
17.1.3 Purpose of the DRG system
In 2001 the EHIF decided to introduce a DRG-based payment system. Stringent
fi nancial constraints exist for Estonian hospitals and the EHIF is not entitled to
spend more than its budget (including reserves), since it is not able to raise health
insurance contributions to cover the defi cit. Therefore, the main motivation
Figure 17.1 Overview of areas served by the Estonian hospital network
Source: Koppel et al., 2008.
Estonia: Developing NordDRGs within social health insurance 305
for introducing the DRG system was fi nancial in nature, bearing in mind the
particularly strict fi nancial constraints of the EHIF budget. The consequences of
these expense limits were particularly serious following the Russian economic
crisis that affected the Estonian economy in 1999, driving the EHIF’s reserves to
zero. During the years that followed, the EHIF’s budget revenues slowly increased,
while pharmaceutical expenditure also increased rapidly, resulting in serious cost
pressure on the EHIF. Thus, the DRG system was mainly seen as a tool to increase
productivity and effi ciency, rather than to increase the transparency of hospital
output.1 Another motivation for introducing a case-based payment system was
Tab le 17.1 Hospital indicators, 1998–2008 (selected years)
1998 2000 2002 2004 2006 2008
Structure of hospital beds
by specialty (%)
Acute care beds n/a 77.3 74.2 73.2 69.7 n/a
Psychiatric beds n/a 11.0 10.7 9.0 9.8 n/a
Beds for TB n/a 3.2 3.5 3.8 3.6 n/a
Nursing care beds n/a 8.4 11.6 13.9 17.0 n/a
General indicators of
hospital beds
Number of hospital beds rate
per 1000 7.62 n/a n/a 5.8 5.65 5.7
Acute care beds rate per 1000 5.84 n/a n/a 4.26 3.94 3.85
Psychiatric beds rate per 1000 0.89 n/a n/a 0.52 0.55 0.56
Hospital admissions per 1000 204.1 n/a n/a n/a 188.3 n/a
ALOS 10.3 9.2 8.4 8.0 7.8 7.9
Bed turnover 26.8 27.7 29.5 33.2 34.6 34.6
Bed occupancy rate (%) 74.6 69.9 67.7 72.6 74.1 74.3
Number of treatment cases
Ambulatory care per 1000
insured n/a n/a 1876 1845 2000 n/a
Day care per 1000 insured n/a n/a n/a 26 35 n/a
Inpatient care per 1000 insured n/a n/a 200 197 195 n/a
Source: Koppel et al., 2008.
Note: n/a: not available.
Tab le 17.2 Overview of discharges, outpatient contacts and insured individuals used
specialized medical care services
Indicator 2006 2007 2008 2009
Discharges 249 398 248 711 249 784 240 227
Outpatient (excluding family
practitioner) contacts 3 481 857 3 624 744 3 722 259 3 573 286
Number of insured individuals
that used health care services
during one year 796 815 810 834 819 055 800 578
Sources: EHIF, 2008, 2010b.
306 Diagnosis-Related Groups in Europe
that the previous fee-for-service and per diem payment systems had led to infl a-
tion in the average reimbursement rate per case: infl ation reached about 30 per
cent between January 2000 and September 2002, while the offi cial price increase
was only 13 per cent (Koppel et al., 2008).
The importance of the DRG system has increased gradually with the increase
(from 10 per cent up to 70 per cent since July 2009) in the share of the DRG
payment system since the introduction of the system. In addition, over time
the DRG system has became a tool for benchmarking and analysis. Since 2005,
the EHIF provides hospitals with regular information updates regarding average
length of stay, casemix index (CMI) (since 2008), use of some DRGs, share of
outliers and so on, in order to give them the opportunity to compare with other
hospitals, as well as to follow the trend of certain indicators across time.
17.2 Development and updates of the DRG system
17.2.1 The current DRG system at a glance
One DRG system has been in place in Estonia since 2003. No differentiation
by region, purpose, or health care provider is applied. The decision to use one
DRG system for the whole country was made early in the implementation
planning process, and the question of whether to implement more than one
system was not under consideration. Before the implementation of the current
DRG system, several DRG systems were compared in order to fi nd the best
option for Estonia. The fi nal decision was made in favour of the Nordic patient
classifi cation system (NordDRG).
The NordDRG system was adopted in 2003, along with the system’s DRG
grouping logic. In 2011 the grouping logic was updated and the NordDRG 2010
version was implemented. The total number of DRGs in the NordDRG Estonian
2010 version is 786 (496 in the 2003 version), 655 of which (489 in 2003) are
used for reimbursement (see Table 17.3). The rest of the DRGs are ‘empty’; that
is, no cases are assigned to them.
The assignment of cases is based on diagnoses, procedures performed, age,
gender, length of stay and status at discharge. DRGs apply only to inpatient care
and day surgery, with the exception of long-term care, such as psychiatry,
tuberculosis (TB) and nursing care, as well as expensive drugs and inpatient
cases which include treatment with cytostatics (see subsection 17.5.1).
17.2.2 Development of the DRG system
The DRG implementation plan in Estonia was prepared in 2001 by the EHIF.
It was initially planned for DRG-based reimbursement of hospitals to start
in 2002. However, during the preparatory process it became clear that the
plan was unrealistic and more time was needed for technical preparation.
It was therefore decided that in 2003 the DRG system would be used only as
a grouping tool and in 2004 the DRG system would start to be used as a pay-
ment tool.
Estonia: Developing NordDRGs within social health insurance 307
Before the implementation of the DRG system in Estonia, several DRG sys-
tems were compared to fi nd the best option. The alternatives under consider-
ation were the Australian Refi ned (AR)-DRG system, the Nordic NordDRG
system and the Estonian case-based system. Various criteria were used to evalu-
ate the available systems, such as other clinical classifi cations in use, clinical
practice, clinical cases, cost of implementation, and technical support. Once the
NordDRG system was chosen, work on adaptation began. The Nordic Casemix
Centre produced an Estonian NordDRG version that was implemented in 2003.
For DRG weight calculation, two alternatives were considered. First, Estonia
would calculate its own DRG weights according to the available historical
billing information based on fee-for-service payments. The second alternative
was to carry over Health Care Financing Administration (HCFA) weights and
the DRG prices would be calculated based on the average reimbursement rate of
each case. It was evident that hospitals would not be able to provide DRG-based
cost information to use as an input for DRG weights calculation. The fact that
Finland had tended to use HCFA weights from the outset without any problems
encouraged the EHIF to choose this option. It was thought that starting with
the United States HCFA weights system would provide a good basis for further
development. In any case, the weight proportions tend to be analogous in
different countries (EHIF, 2009). However, health care providers were more
supportive of the Estonian national weights idea, as these were seen to better
refl ect the Estonian context. The decision was therefore made to use a ‘home-
made’ mix of Estonian data and HCFA weights. In 2006 the project of developing
Estonian national cost weights began and since 2008 the Estonian cost weights
are used in the DRG price calculation. The adjustment of cost weights is in line
with the recalculation of the prices of health care services.
Responsibility for developing and updating the DRG system in Estonia lies
with the EHIF and it is carried out according to the DRG development plan.
Tab le 17.3 Summary description of the DRG system
Date of introduction 2003
(Main) purpose 2003 as a grouping tool since April 2004 as a
reimbursement tool
DRG system NordDRG
Data used for development Database of EHIF
Number of DRGs 496 (until 2010)
786 (since 2011)
Applied to Health care providers contracted with EHIF,
acute inpatient cases and those outpatient
cases involving surgical procedure(s)
Proportion of DRG/fee-for-service
payments
2004: 10/90
2005: 50/50
2009 July: 70/30
Introduction of NCSP 2003
NCSP update 2010
Introduction of Estonian cost weights 2008
Update of DRG version 2011, NordDRG 2010 Full version
308 Diagnosis-Related Groups in Europe
17.2.3 Data used for development and updates of
the DRG system
For developing the DRG system (including the grouping logic, cost weights,
prices, and so on), the data mainly originate from the EHIF’s electronic billing
system. The data used for DRG grouping consist of different patient charac-
teristics, such as age, gender, diagnoses, surgical procedures, the way patients
arrived at the hospital, their status at discharge, and so on. For the develop-
ment of the DRG system, other characteristics are used, such as the level of
the hospital (regional, central, general hospital), average length of stay, CMI,
average cost per case, and so on. Resource-consumption data are used for calcu-
lation of cost weights and DRG prices.
According to their contractual obligations with the EHIF, every health care
provider must transfer the patient-level data (services provided, length of stay,
diagnosis, and so on) to the EHIF database in order to be reimbursed. Thus, the
main source used to develop the DRG system is the EHIF database. Expertise
from medical professionals’ associations and health care providers is also used
as an input for system development, but the contribution of these actors to
developing the DRG system has remained relatively modest.
17.2.4 Regularity and method of system updates
Since Estonia incorporated the NordDRG system, the regularity and methodol-
ogy of system updates is steered by the Nordic Casemix Centre (see Chapter 16).
NCSP update
The fi rst NOMESCO (Nordic Medico-Statistical Committee) Classifi cation of
Surgical Procedures (NCSP) version in Estonia was the generic classifi cation
Version 1.6 that was introduced in 2003. In practice, the NCSP has generic and
country-specifi c versions that can be updated on an annual basis to introduce/
change coding. In Estonia, a new updated NCSP version was introduced at the
beginning of 2010.
Countries using the NCSP can further develop their own national versions of
the classifi cation. However, before proposing updates, local capacities need to
be developed at country level and panels convened to facilitate discussions
between administrators and medical doctors. Until recently this has not been
implemented well in Estonia.
Cost weights and DRG prices update
Updating cost weights and DRG prices is the responsibility of the EHIF. As the
calculation of cost weights and DRG prices is based on prices for fee-for-service
health services, not on the actual resource need, the updating process is carried
out as often as the prices of health services are updated. This update occurs
annually (see subsection 17.5.2).
Estonia: Developing NordDRGs within social health insurance 309
17.3 The current patient classifi cation system
17.3.1 Information used to classify patients
Data used to classify patients (cases) are transmitted electronically by health
care providers to the EHIF database. The regularity and frequency of data trans-
mission, content of data and so on are regulated by legislation and by contracts.
The information needed for grouping consists of the following information:
principal diagnoses (in some cases diagnoses of co-morbidities and complica-
tions (CCs)), procedures performed, age, gender, length of stay and status
at discharge. Resource-consumption data are not used for grouping. The pri-
mary classifi cations used in the NordDRG system are International Classifi ca-
tion of Diseases 10th revision (ICD-10) for diagnoses and the NCSP for surgical
procedures.
17.3.2 Classifi cation algorithm
The overview of the DRG assignment process is depicted in Figure 17.2. It starts
from a set of ‘Pre-MDC’ (major diagnostic category) assignment rules. Pre-MDC
defi nitions refer to the group of DRG assignment rules that ignore the MDC indi-
cated by the principal diagnosis of the patient. This includes DRGs for highly
Figure 17.2 Main logic of the NordDRG system
Source: Nordic Casemix Centre, 2003.
310 Diagnosis-Related Groups in Europe
specialized and expensive care. Examples of these are lung transplantation and
bone marrow transplantation. Second, the Pre-MDC category is needed to avoid
misclassifi cation due to some differences between ICD-10 and ICD-9-CM (9th
revision, Clinical Modifi cation). Patients with multiple traumas, HIV-related
problems, obstetrics and neonatology might entail a principal diagnosis that
originally indicates specifi c MDCs. However, subsequent information may
require the cases to be reallocated.
After the Pre-MDC is interpreted, the principal diagnosis of the case indicates
the MDC. MDCs are mainly determined by organ systems, such as nervous
system, digestive system, ear, nose, mouth and throat, and so on. Some MDCs
are related to the etiology of the disease, for example infectious and parasitic
diseases, injuries, poisonings and toxic effects of drugs, burns, and so on. There
are 26 MDCs in the NordDRG 2010 version used in Estonia (there were 27
in 2003).
For each MDC a decision tree is designed with branching nodes, requiring
information relating to the surgical procedure (which needs an operating room
(OR)),2 diagnoses of CCs, age, length of stay, status at discharge, and whether
the patient is discharged to home or to another institution, whether the patient
died or discharged her/himself against medical advice. At each branching node,
the condition at the node is processed by identifying the information needed
and comparing it to lists of codes or rules, determining which of the alternative
routes to follow. The process is continued until the case ends up in one appro-
priate DRG.
17.3.3 Data quality and plausibility checks
Different means are used by the EHIF for assessing and improving data quality.
All information in the EHIF database is gathered by using the electronic data
transmission system. Health care providers complete their medical bills by
inputting different patient and provider characteristics, as well as details of
services carried out according to the fee-for-service health care service list.
Completed bills are gathered together as ‘electronic packages’ and transmitted
to the EHIF.
In the initial stage of data quality checks, format controls are carried out in
the electronic system before the electronic packages enter the EHIF database.
During format controls, different patient and provider characteristics are
checked to determine whether they meet certain requirements, for example
those set by legislation or under contract. The format of diagnoses and proce-
dure codes described on medical bills is also checked. All medical bills with
mistakes are returned to health care providers, giving them the opportunity to
correct the inaccurate information. The health care provider can then transmit
the electronic package again. Those bills that contain no mistakes are passed to
the provider for fi nal acceptance before payment.
In addition to format controls in the electronic system, some other methods
are used for checking data quality. For instance, randomized controls of medical
records are carried out by teams of EHIF ‘trustee doctors’. These checks aim to
compare the ICD-10 and NCSP codes described in the medical records with
Estonia: Developing NordDRGs within social health insurance 311
those on the reimbursement claims and to detect inappropriate use of primary
classifi cations which could lead to the change in assignment of cases into DRGs.
On a randomized basis the trustee doctors are sent to the hospitals to check the
medical records or the medical records are brought to the EHIF offi ce upon
request. The percentage of the medical records checked in order to verify the
coding quality amounts to about 4–5 per cent of the total number of records
collected.
For most of the cases in which errors are found no fi nancial sanctions are
applied, unless fraud or abuse is detected. However, in detecting problems in
coding quality, the trustee doctors provide feedback to health care providers,
informing them of any inappropriate coding.
17.3.4 Incentives for up- or wrong-coding
There is no clear evidence of up- or wrong-coding. Instead, the results of
randomized controls carried out by the EHIF show under-coding by health care
providers, mainly due to the lack of accurate reporting of relevant information
among medical doctors.
17.4 Cost accounting within hospitals
Cost accounting within hospitals in Estonia can be described as operating on
two levels. For reimbursement purposes, hospitals must carry out service
volume accounting per patient for all the services listed in the EHIF price list,
and they must issue a relevant invoice per patient (volume of services delivered
multipled by service prices). In the case of inpatient care and day care, the
majority of invoices are recalculated by the EHIF and 70 per cent of the value is
replaced by the relevant DRG price (see subsection 17.5.1).
This chapter deals with another level of cost accounting in hospitals, relating
to the cost of providing services. This information is an input for fee-for-service
pricing, and billing information relating to fee-for-service pricing is an input for
DRG pricing (on an annual basis) (see subsection 17.5.2).
Cost accounting in hospitals in Estonia is not regulated in a specifi c way, as
there is no requirement to report costs to health care authorities. Hospital
steering and fi nancial control is carried out by hospital supervisory board, and
the Ministry of Social Affairs receives hospitals’ annual reports (including
fi nancial reports). The only cost item monitored by the Ministry of Social Affairs
is the average salaries of medical professionals, such as doctors, nurses and
assistant nurses. That said, in the case of approving new health care services or
updating prices of existing services, there is a regulated process for presenting
relevant information to the EHIF (see subsection 17.4.1).
17.4.1 Regulation
The service costing process is regulated by ministerial decree and is an integral
part of the benefi ts package (service list) update process, which is regulated
312 Diagnosis-Related Groups in Europe
by law (Health Insurance Act) and by governmental decree. According to the
content of the regulation, the Government approves services and the DRG
pricelist, the EHIF is responsible for expertise relating to cost-effi ciency analyses
of services within the benefi ts package and all applications for new services.
Updating of existing services or elimination of services is processed in collabo-
ration with the EHIF and professional associations or providers’ associations
(see subsection 17.6.1).
According to regulation, applications for new or updating existing services
must include relevant cost information. In order to process applications, the
EHIF needs to receive actual cost data from at least one hospital from each
category of hospitals (regional, central and general). The regulation of costing
(and pricing) of hospital services in Estonia can therefore be described as
centralized and ‘top-down’ in approach.
17.4.2 Main characteristics of the cost-accounting system(s)
For the costing process of services in specialized medical care, the EHIF has set
up a standard costing model which – according to regulations – comprises an
activity-based costing methodology, whereby each service is described through
certain activities and those activities are related to the costs of resources.
Recourses are allocated to direct and indirect resource categories , including,
for example, drugs; single-use medical devices; multi-use medical devices;
labour, including training and administration,; infrastructure-related costs,
including investment (loan interest not included); and auxiliary services. From
July 2003, infrastructure costs have been included in the prices paid to pro-
viders by the EHIF, in order to ensure geographical consistency and fairness in
infrastructure development. The infrastructure costs in health service prices
include the facilities’ depreciation costs based on the market price of buildings,
and a 36-year depreciation period. The mark-up has been calculated according
to providers’ optimal capacity per bed (which includes a standard number
of square meters per bed that will produce an optimal occupancy rate). Since
2008, infrastructure cost expenditures were covered by the state budget as an
earmarked allocation to the EHIF’s budget, and will still be allocated to providers
through the service prices. In 2009 the state stopped allocating infrastructure
costs to the EHIF due to the economic downturn and therefore the EHIF must
cover these from the regular health insurance budget.
Within all resource categories (except drugs and single-use medical devices),
annual costs and effective utilization of resource units (in minutes or usage
frequency) are determined, along with unit costs per utilization unit. Annual
costs of resources are established by regulating degree (for doctors’, nurses’ and
assistant nurses’ salaries), expert opinion (for infrastructure investment costs)
and all other resources are determined based on actual cost data presented by
hospitals. The level of effective utilization is determined by the EHIF; usually
8 hours per working day (minimum one shift effective utilization).
Utilization of resource units by activity and by service is based on the
expert opinion of professional associations, but this is checked by the EHIF
Estonia: Developing NordDRGs within social health insurance 313
against actual unit data from hospitals and often negotiated in the event of
discrepancies.
Cost and resource-utilization data presented by hospitals should represent
the total cost of the previous year’s audited fi nancial statements. As costs in
hospitals are recorded mostly at the department or hospital level, and not at the
service level (top-down approach), cost data presented by hospitals are aggre-
gated only by medical specialty or at hospital level. This creates some uncer-
tainty and results in an averaging approach within the EHIF cost model (costs
defi ned at service level) in terms of checking pre-calculated costs against actual
data presented.
Although there are several hospitals where costs are recorded at the level of
service(s) delivered to the patient, the generated information is not comparable
due to a lack of standardization of hospital information systems.
17.5 DRGs for hospital payment
17.5.1 Range of services and costs included
in DRG-based hospital payments
How applicable the DRG system is to health care providers depends on the
existence of a contract with the EHIF, regardless of the ownership, geographical
location, teaching status, size, and so on, of health care providers. It means
that the DRG system is applied to all specialized medical care providers con-
tracting with the EHIF. The health care providers working without an EHIF
contract are mainly fi nanced on the basis of fee-for-service payments paid out
of pocket.
The DRG system is used in combination with the fee-for-service payment
method. To minimize any fi nancial risk in the new system, the share of DRG
payment applied upon submitting a reimbursement bill was initially (in 2004)
set as low as 10 per cent. In 2005 the share of DRG payment was raised to 50 per
cent and since July 2007 it amounts to 70 per cent. The share of fee-for-service
payment applied upon submitting a reimbursement bill has decreased accord-
ingly, from 90 per cent in 2004 to 30 per cent in 2009.
In practice, every bill lists the health care services delivered to the patient
during their hospital stay. The bill is calculated by adding together the fee-for-
service prices of each of the services. In addition, every bill is assigned to one
DRG with its respective price. The combination of DRG and fee-for-service
reimbursement means that the total sum of the medical bill is calculated as
follows: (1) the fee-for-service element of the bill is multiplied by 0.3 (since July
2009); (2) the corresponding DRG price is multiplied by 0.7; (3) the latter is
added to the fee-for-service sum.
DRGs are used for reimbursement in acute inpatient cases and for those
outpatient cases involving surgical procedure(s). However, the DRG payment
system does not apply to all assigned cases. A system of DRG outliers (rules to
detect cases that do not come under the DRG-based reimbursement system)
exists in Estonia. The outliers can be divided into two groups, as detailed here.
314 Diagnosis-Related Groups in Europe
1. Cases with certain characteristics (types of care), such as psychiatry, rehabili-
tation, TB and follow-up cases. This same group of outliers also contains
cases determined according to their principal diagnosis (for example, Z51.1
and Z51.2 – chemotherapy; and Z76.3 – healthy person accompanying
ill person), as well as some referred cases (for example, while referring the
patient from a higher level hospital to a lower level one, the patient is con-
sidered an outlier).
2. Cost-outliers; that is, cases that are too low cost or too high cost.
All above-mentioned cases are treated as DRG outliers and are reimbursed
fully through fee-for-service payment(s). Pricing of DRGs is based on prices and
casemix (according to reimbursement information) of the health services
provided, rather than on the explicit cost information (see subsection 17.5.2).
DRG prices – along with health service prices – are equal for all providers and
there are no higher rates for teaching hospitals or for other higher level or
specialized hospitals. Health service prices cover all costs related to providing
services, except those related to scientifi c and teaching activities, which are
funded separately. All prices approved are maximum prices and providers and
the EHIF can agree on lower prices for specifi c contracts.
17.5.2 Calculation of DRG prices/cost weights
DRG price calculation is conducted by the EHIF and alongside updating the
prices of health services; that is, when the prices of health services are going to
change, current DRG prices need to be recalculated and the cost weights
adjusted accordingly. The DRG price calculation is carried out at patient level
and is based on fee-for-service billing information, taking into account the
latest available data from all health care providers. DRG prices are equal for
all health care providers.
The calculation of DRG prices can be divided into separate steps, as detailed
here.
• The process starts with data quality analysis, in order to detect and eliminate
data of poor quality.
• Second, the coeffi cient of volume infl ation is calculated, in order to take into
account the changes in the structure of health services by comparing specifi ed
periods of time.
• Third, the impact of change of health service prices is calculated by comparing
current DRG prices with the new average price per DRG, calculated on the
basis of new fee-for-service prices. For calculation of the average price per
DRG, two-phase trimming is used in order to eliminate the impact of outliers
(see Figure 17.3). In the 1st phase of trimming, the outlier cases with costs
outside 3 standard deviation are excluded and in the 2nd phase, outlier cases
with costs outside 2 standard deviation are excluded.
• The correction of the average price per DRG is carried out by applying to the
current DRG price the above-mentioned coeffi cients of volume infl ation and
impact of change of health service prices.
Estonia: Developing NordDRGs within social health insurance 315
• For calculation of the base rate, all bills remaining after two phases of trim-
ming are taken into account. The base rate is calculated by dividing the total
sum of fee-for-service prices by the total number of bills.
• Cost weights per DRG are calculated by dividing the corrected DRG price by
the base rate. Cost weights are compared with current ones and adjusted if
necessary;
• The limits of each DRG are set up – the upper and lower limits are identifi ed
according to the last trimming points (see Figure 17.3). In many cases the
lower limit is a negative value and therefore the lower limit is set equal to the
lowest per diem rate.
• Finally, the calculation of the price for each DRG is carried out by multiplying
the base rate by the cost weight of the corresponding DRG.
17.5.3 Use of DRGs in hospital payment
As already mentioned, three different payment methods are used for inpatient
care – DRGs, fee-for-service payments and per diem payments. As the share of
DRG payments has increased throughout the years (see subsection 17.5.1), the
utilization of the DRG-based payment method has increased accordingly and in
2009 it accounted for 39 per cent of total hospital expenditure for inpatient
care (see Table 17.4).
Figure 17.3 Trimming method used in the calculation of the DRG price
Source: Compiled by Jorma Lauharanta.
Tab le 17.4 Proportion of different payment mechanisms for inpatient care in acute
care hospitals, 2006–2009
Payment method 2006 2007 2008 2009
Fee-for-service (%) 35 36 37 33
DRG (%) 36 34 33 39
Per diem (%) 29 30 30 28
Source: EHIF, 2010a.
316 Diagnosis-Related Groups in Europe
17.5.4 Quality-related adjustments
Although introducing the pay for performance (P4P) initiative has been consid-
ered, no quality-related adjustment reimbursement mechanism is applied to
hospitals in Estonia thus far.
The implementation of new payment methods and any changes in payment
methods – together with the processes by which care is commissioned – should
be undertaken carefully and with emphasis on making the most of available
evidence and contributing to the body of evidence on how trading incentives
affect the effi ciency of health care delivery (Maynard, 2008).
17.5.5 Main incentives for hospitals
Not many incentives exist to set up DRG systems for hospitals. The main
argument of the EHIF in favour of setting up such a system was that hospitals
could control the increase of services in the casemix. However, in the current
approach to DRG pricing, the change in the average reimbursement rate for the
casemix is taken into account. The only incentive for hospitals is to maximize
outpatient and day-care services, for which DRGs are not applicable.
17.6 New/innovative technologies
17.6.1 Steps required prior to the introduction of
new/innovative technologies in hospitals
Estonia has no systematic programme for health technology assessment (HTA),
mainly due to a lack of interest on the part of policy-makers and a lack of
trained human resources. The main activities in this fi eld include assessing new
services to be added to the benefi ts package and prescription drugs to the posi-
tive list; evaluating the need for high-cost technologies; and ensuring the safety
of medical equipment. These activities are carried out at national level and
there is no evidence on the use of HTA at the organizational level. However,
hospitals conduct some cost-analysis studies when high-cost technologies are
purchased (such as magnetic resonance imaging (MRI) or computerized tomog-
raphy (CT) scanners) (Koppel et al., 2008).
During the 1990s, the inclusion and exclusion of services from the benefi ts
package was decided by the Ministry of Social Affairs, following evaluation by a
ministry committee made up of provider and sickness fund representatives.
Evaluations were based on treatment effectiveness criteria and, where possible,
proposals for adding new treatments were weighed against existing treatments.
Since 2002, there have been clearer and more explicit rules for adding new
services to the benefi ts package and establishing the appropriate level of cost-
sharing. In 2002, when the EHIF was established as an independent public
body, it was tasked with the responsibility for defi ning the benefi ts package in
collaboration with other stakeholders. The benefi ts package is agreed by the
EHIF and the Ministry of Social Affairs, and a fi nal decision is made by the
Estonia: Developing NordDRGs within social health insurance 317
Government, which endorses the price list. The procedure for amending the
benefi ts package is presented in Figure 17.4.
17.6.2 Payment mechanisms
The funding of new technologies included in the benefi ts package does not
differ from the funding of any other services within the package and there is
no specifi c or separate funding for innovative technologies before they are
included.
17.6.3 (Dis-)incentives for hospitals to use
new/innovative technologies
No straightforward system of incentives exists for the utilization of new or
innovative technologies in Estonia. The driving force is the interest and initia-
tive of clinicians and leadership of medical groups in certain hospitals or at
country level. However, due to some historical decisions, utilization of some
technologies (high-end radiology, interventional radiology and cardiac surgery,
Figure 17.4 Procedure for amending the benefi ts package
Source: EHIF, 2008, 2010b.
318 Diagnosis-Related Groups in Europe
for example) is increasing, as these are overpriced relative to the actual costs of
providing the service(s).
17.7 Evaluation of the DRG system in Estonia
17.7.1 Offi cial evaluation
Evaluation of the NordDRG system relies on the Nordic Casemix Centre. The
NordDRG expert network is the main advisory group and platform for discus-
sions relating to the maintenance, performance evaluation and development
of the NordDRG system. The suggestions for annual updates of the common
NordDRG are based on expert network recommendations.
In addition, the Nordic Casemix Centre carries out evaluation, testing and
certifi cation of DRG groupers to ensure compatibility with the NordDRG defi ni-
tions. The Centre also cooperates with the main NordDRG software provider(s)
in delivering the common NordDRG version (as well as national versions based
on national codes) to the NordDRG users, as appropriate.
In terms of evaluation at national level, there has not been any offi cial evalu-
ation of the DRG system. However, the EHIF has conducted various analyses
in order to assess the data quality (mainly the use of primary classifi cations),
DRG pricing methodology, preparedness for and impact of shifting to new
NCSP and DRG grouper versions, as well as the impact of increasing the
share of DRG-based payment in reimbursement, and so on. The results of the
analyses have been used to further develop and fi ne-tune the DRG system
in Estonia.
17.7.2 Authors’ assessment of successes and problems
One of the central arguments for the introduction of the DRG system was to
increase effi ciency and contain the health insurance expenditure. From the
purchaser point of view, during the initial years of the adoption of the DRG
payment system in Estonia the set objective was met; namely, the DRG system
has contained the average cost per case compared to the situation that would
have arisen if only a fee-for-service and per diem-based payment system had
been used (as it was before the implementation of the DRG system).3 However,
the results show the differences between various hospitals. Therefore, bearing
in mind the strategic goal of the DRG system, the further development and
fi ne-tuning of the system is carried out according to the four-year DRG develop-
ment plan.
The use of the DRG system as a benchmarking tool began in 2005 and has
developed over time. Since 2009 the range of indicators has broadened and
hospitals listed in the HNDP are provided the data via a web page of EHIF. This
provides the opportunity to compare and assess different performance indica-
tors. Until recently, hospitals in Estonia were benchmarked (length of stay, use
of some DRGs, percentage of outliers, and so on) mostly at the specialty level,
without standardizing for case structure and severity. The introduction of the
Estonia: Developing NordDRGs within social health insurance 319
CMI since 2008 affords hospitals (and other authorities) additional dimensions
for more objective comparison of performance results.
Comparisons are a powerful way of driving performance improvement. How-
ever, there is a great deal of potential in Estonia to use this information in order
to locate and pinpoint the strengths and weaknesses of hospital performance
and to use the data to support decision-making processes.
17.8 Outlook: Future developments and reform
17.8.1 Trends in hospital service
(or general health care) delivery
The most fundamental changes for building a functioning health system in
Estonia were made in the early stages of reform, during the early 1990s. The
incremental arrangements that followed were implemented to support the pub-
lic health, primary health care and hospital sector reforms and to strengthen
the EHIF’s purchasing function. Therefore, the attention has shifted to improv-
ing and monitoring performance of the system as a whole.
In terms of service delivery, the main challenge is presented in the need to
optimize the system. The strength of the current delivery system is in family
medicine-centred primary health care. This system covers a wide range of ser-
vices, without co-payments and with minimal waiting times. It is comple-
mented by the ambulance (emergency) services for care outside normal working
hours. The challenge lies in making the delivery system more patient-centred
and coordinating care at the primary care level, with the development of addi-
tional nursing and rehabilitation services.
17.8.2 Trends in DRG application/coverage
The EHIF is responsible for the development of the DRG system in Estonia. This
is carried out in line with the four-year DRG development plan approved by the
management board of the EHIF.
The strategic goal of the DRG system is to contribute to increased effi ciency
in the use of health insurance resources. In order to achieve this goal, the EHIF
will implement the following measures:
• improve the quality of coding
• develop DRG-based analysis and feedback
• develop transparent pricing and funding principles.
Improving coding quality involves, inter alia, the correct and unifi ed use of
primary classifi cations (ICD and NCSP) by health care providers, and the
improvement of coding quality assessment. The development of methods of
analysis and benchmarking is related to improvement in technical resources,
together with the respective competences of the staff of the EHIF. Actions to
develop transparent methodology of DRG pricing and funding principles
include, inter alia, the regular updating of the current DRG grouping version.
320 Diagnosis-Related Groups in Europe
17.9 Notes
1 Due to the former fee-for-service payment structure and well-developed electronic
data transmission systems, Estonia already had a relatively transparent overview of
hospital output.
2 Cases with an OR procedure are assigned to the ‘surgical’ DRGs and those without an
OR procedure are assigned to the ‘medical’ DRGs.
3 More details are available at the EHIF web site (www.haigekassa.ee, accessed 1 August
2011).
17.10 References
Björnberg, A., Garrofé, B.C., Lindblad, S. (2009). Euro Health Consumer Index 2009 Report.
Brussels: Health Consumer Powerhouse.
EHIF (2008). Estonian Health Insurance Fund Annual Report 2008. Tallinn: Estonian Health
Insurance Fund.
EHIF (2009). Overview of Estonian Experiences with DRG System. Tallinn: Estonian Health
Insurance Fund (http://www.haigekassa.ee/uploads/userfi les/Implementation_DRG_
EST_291209_cover(1) , accessed 21 June 2010).
EHIF (2010a). DRG tagasiside aruanne [DRG Feedback Report]. Tallinn: Estonian Health
Insurance Fund (http://www.haigekassa.ee/uploads/userfi les/DRG_tagasiside_aruanne_
2009 , accessed 21 June 2010).
EHIF (2010b). Estonian Health Insurance Fund Annual Report 2010. Tallinn: Estonian Health
Insurance Fund.
EHIF (2011). Estonian Health Insurance Fund. Tallinn: Estonian Health Insurance Fund
(www.haigekassa.ee, accessed 1 August 2011).
Jesse, M., Habicht, J., Aaviksoo, A. et al. (2004). Health Care Systems in Transition: Estonia.
Copenhagen: WHO Regional Offi ce for Europe on behalf of the European Observatory
on Health Systems and Policies.
Koppel, A., Kahur, K., Habicht, T. et al. (2008). Estonia: health system review. Health
Systems in Transition, 10(1):1–230.
Maynard, A. (2008). Payment for Performance (P4P): International Experience and a Cautionary
Proposal for Estonia. Copenhagen: WHO Regional Offi ce for Europe (Health Financing
Policy Paper, Division of Country Health Systems).
NIHD (2008). Department of Health Statistics, Health Statistics and Health Research Database
[online database]. Tallinn: National Institute for Health Development (http://pxweb.
tai.ee/esf/pxweb2008/dialog/statfi le2.asp, accessed 20 January 2010).
Nordic Casemix Centre (2003). NordDRG System. Helsinki: Nordic Casemix Centre (http://
www.nordcase.org, accessed 10 July 2011).
chapter e i g h t e e n
Finland: DRGs in a
decentralized health
care system
Kirsi Kautiainen, Unto Häkkinen
and Jorma Lauharanta
18.1 Hospital services and the role of DRGs in Finland
18.1.1 The Finnish health system
In its institutional structure, fi nancing and goals, the Finnish health care system
is closest to those of other Nordic countries and the United Kingdom, to the
extent that it covers the whole population and its services are mainly delivered
by the public sector and fi nanced through general taxation (for more details,
see Häkkinen, 2005, 2009; Häkkinen & Lehto, 2005; OECD, 2005; Vuorenkoski,
2008). However, compared to the other Nordic countries, the Finnish system is
more decentralized (Magnussen et al., 2009); in fact it can be described as one
of the most decentralized in the world. Even the smallest of the 342 munici-
palities (local government authorities) are responsible for arranging and taking
fi nancial responsibility for a whole range of ‘municipal health and social
services’. From an international perspective, another unique characteristic of
the system is the existence of a secondary public fi nancing scheme (the National
Health Insurance scheme) also covering the whole population, which partly
reimburses the same services as the tax-based system, but only services which
are provided by the private sector. The National Health Insurance also partly
reimburses the use of private hospital care.
Municipally provided services include primary and specialized health care.
In addition, municipalities are responsible for other basic services, such as nurs-
ing homes and other social services for the elderly, child day-care, social assis-
tance and basic education. Municipal health services are fi nanced through
municipal taxes, state subsidies and user charges. Primary health care is mainly
322 Diagnosis-Related Groups in Europe
provided at health centres, which are owned by municipalities or federations
of municipalities. Preventive care for communicable and non-com municable
diseases, ambulatory, medi cal and dental care, an inc reasing number of out-
patient specialized services, and various public health programmes (such as
maternity and school health care) are provided by the health centres. They
also provide occupational health services and services for specifi c patient groups
(for example, clinics for diabetes and hypertension patients). Health centres
include also inpatient departments. The majority of patients in these depart-
ments are elderly and chronically ill people, but in some municipalities, health
centres also provide acute short-term curative inpatient services. In addition
to the inpatient departments of the health centres, long-term care is provided
at homes for the elderly that in administrative terms come under municipal
social services.
Specialized care (psychiatric and acute non-psychiatric) is provided by
hospital districts, which correspond to the federations of municipalities. Each
municipality is obligated to be a member of a hospital district. In addition to
services provided through health centres and hospital districts, municipalities
may purchase services from a private provider.
18.1.2. Hospital services in Finland
Acute somatic hospital care is mostly publicly provided by hospital districts
and, to a lesser extent, some health centres are supplemented by care provided
in private hospitals. In 2007, specialized care comprised in total 33 per cent of
the total health care expenditure, of which the share of private service provision
was only 1 per cent (THL, 2010).
There are 21 hospital districts in the country. Most hospital districts have a
central hospital and in some districts, care is supplemented by small regional
hospitals. There are 14 regional hospitals in the country. Tertiary care is pro-
vided in fi ve university hospitals, which also act as central hospitals for their
hospital district. All of the 34 hospitals owned by hospital districts provide both
inpatient and outpatient services; in 2007, on average 59 per cent of resources
were allocated to inpatient care, 7 per cent to day-case surgical care, and 34 per
cent to ambulatory care. In addition, some acute somatic care is provided in
health centres owned by municipalities or federations of municipalities, as well
as in private hospitals (Table 18.1).
Hospital districts are managed and funded by the member municipalities.
Funding is mainly based on municipalities’ payments to hospital districts,
according to the services used (see Figure 18.1). In 2008, the share of municipal
payments of all costs of somatic care was 89 per cent. A total of 2 per cent of
the total funding of hospitals comes from state subsidies for research and
teaching, and 4 per cent from user charges. The Government defi nes the
maximum fees that hospitals can charge. In practice, every hospital applies the
maximum fees. In 2010 these were €32.50 for an inpatient day, €27.40 for
outpatient visits and €89.90 for day surgery. User charges within public sector
health care have an annual ceiling (€633), after which patients receive services
free of charge.
T
a
b
le
1
8
.1
A
cu
te
h
o
sp
it
al
k
ey
fi
g
u
re
s
b
y
t
y
p
e
an
d
o
w
n
er
sh
ip
,
2
0
0
9
T
yp
e
a
n
d
ow
n
er
sh
ip
N
u
m
b
er
o
f
h
os
pi
ta
ls
P
a
ti
en
ts
(
in
pa
ti
en
t
a
n
d
d
a
y-
ca
se
su
rg
ic
a
l
ca
re
)
D
is
ch
a
rg
es
(i
n
pa
ti
en
t
a
n
d
d
a
y-
ca
se
s
u
rg
ic
a
l
ca
re
)
B
ed
d
a
ys
(
in
pa
ti
en
t
a
n
d
d
a
y-
ca
se
s
u
rg
ic
a
l
ca
re
)
A
ve
ra
ge
le
n
gt
h
o
f
st
a
y,
d
a
ys
E
m
er
ge
n
cy
d
is
ch
a
rg
es
D
a
y-
ca
se
s
u
rg
er
y
d
is
ch
a
rg
es
N
u
m
b
er
Sh
a
re
N
u
m
b
er
Sh
a
re
N
u
m
b
er
Sh
a
re
N
u
m
b
er
Sh
a
re
N
u
m
b
er
Sh
a
re
N
u
m
b
er
Sh
a
re
U
n
iv
er
si
ty
h
o
sp
it
al
s
5
5
.4
2
7
1
1
0
4
3
9
.7
3
9
4
9
7
5
4
1
.0
1
3
9
6
5
7
0
4
0
.7
6
3
.5
1
6
2
9
6
8
3
9
.4
5
8
8
9
2
3
2
.4
C
en
tr
al
h
o
sp
it
al
s
1
5
1
6
.1
2
4
5
2
7
8
3
5
.9
3
5
5
7
0
3
3
6
.9
1
2
4
0
6
0
6
3
6
.2
0
3
.5
1
8
0
5
6
6
4
3
.6
5
9
0
7
8
3
2
.5
R
eg
io
n
al
h
o
sp
it
al
s
1
4
1
5
.1
7
9
0
9
4
1
1
.6
1
0
3
5
2
0
1
0
.8
3
5
3
8
0
4
1
0
.3
3
3
.4
4
8
3
5
9
1
1
.7
2
1
1
0
3
1
1
.6
H
ea
lt
h
c
en
tr
es
(s
p
ec
ia
li
se
d
)
1
5
1
6
.1
3
7
8
6
8
5
.5
5
1
1
8
6
5
.3
2
8
1
3
3
1
8
.2
1
5
.5
2
0
8
6
6
5
.0
9
3
1
7
5
.1
P
u
b
li
c
h
o
sp
it
al
s
4
9
5
2
.7
6
3
3
3
4
4
9
2
.7
9
0
5
3
8
4
9
4
.0
3
2
7
2
3
1
1
9
5
.5
0
3
.6
4
1
2
7
5
9
9
9
.8
1
4
8
3
9
0
8
1
.5
P
ri
v
at
e
h
o
sp
it
al
s
4
4
4
7
.3
4
9
9
5
4
7
.3
5
7
5
7
9
6
.0
1
5
4
3
2
1
4
.5
0
2
.7
9
8
4
0
.2
3
3
6
0
5
1
8
.5
T
o
ta
l
9
3
1
0
0
.0
6
8
3
2
9
8
1
0
0
.0
9
6
2
9
6
3
1
0
0
.0
3
4
2
6
6
3
2
1
0
0
.0
0
3
.6
4
1
3
7
4
3
1
0
0
.0
1
8
1
9
9
5
1
0
0
.0
So
u
rc
e:
A
u
th
o
rs
’
o
w
n
c
o
m
p
il
at
io
n
b
as
ed
o
n
t
h
e
n
at
io
n
al
h
o
sp
it
al
d
is
ch
ar
g
e
d
at
a.
N
ot
e:
F
in
la
n
d
n
o
l
o
n
g
er
c
o
m
p
il
es
d
at
a
o
n
h
o
sp
it
al
b
ed
n
u
m
b
er
s
si
n
ce
t
h
is
i
s
re
co
g
n
iz
ed
a
s
an
i
n
ac
cu
ra
te
m
ea
su
re
o
f
h
o
sp
it
al
c
ap
ac
it
y.
324 Diagnosis-Related Groups in Europe
As purchasers, municipalities negotiate annually the provision of services
with their hospital district. There are different contractual or negotiation mech-
anisms between hospital districts and municipalities for agreeing target vol-
umes and payments. Both the volumes and costs are planned based on the
previous year. In many cases, the municipalities and the hospital districts differ
in opinion regarding the correct size of the resource allocations. There is a ten-
dency for budgets to be set too low, which means that agreements sometimes
need to be revised during the year, according to the actual amount and type of
hospital services delivered. Usually, no explicit sanctions are applied if agreed
plans and targets are deviated from, and municipalities cover any defi cits and
retain any savings in their accounts. The negotiation mechanisms have been
under continuous change and development.
The budget of each hospital district is formally decided by a council, the
members of which are appointed by each municipality. The council also
approves the fi nancial statements (such as payment methods and levels of
payments (prices)) and makes decisions regarding major investments. The
hospital payments from municipalities are based on the total budget and the
predicted use of services. If the hospital budget is exceeded, the municipalities
Figure 18.1 Funding of hospital care in Finland
Note: NHI: National Health Insurance
Finland: DRGs in a decentralized health care system 325
must cover the defi cit from their own revenues, usually by paying higher prices
for services. In the case of budgetary surplus, the prices per service can be
lowered. Thus, the major purpose of hospital pricing systems has been to cover
the costs of production and to allocate hospital costs fairly between the
municipalities that fi nance the provision of services within a hospital district.
In the absence of nationally set regulations or even guidelines, each hospital
district determines the payment methods used to reimburse its hospitals.
Payment methods are organized according to district and as such they vary
from district to district. The trend of pricing has been consistently moving away
from a price-per-bed-day approach towards case-based pricing. Currently, 13
out of 21 districts are using DRG-based payment methods. The principles and
rules for DRG usage vary greatly between hospital districts.
In order to diminish municipalities’ fi nancial risk from expensive patients, a
compensation system has been created. With this system, treatment costs per
patient above a particular limit are shared between all the member municipalities
of a hospital district. In most hospital districts this limit is between €50 000
and €60 000. Variation exists, and in some districts the limit is even higher
than €80 000. The payment share for each municipality is defi ned based on the
number of citizens living in the municipality.
18.1.3 Purpose of the DRG system
Within the hospital districts, the DRG system (as well as other prevailing pay-
ment systems) is not used for resource allocation but mainly as a method of
collecting payments from municipalities; that is, as a billing instrument. Hospi-
tal districts use DRG-based prices to charge municipalities for the services they
have delivered. The Finnish payment system does not create similar incentives
to the prospective payment system used in many countries. This is because
hospitals do not bear any responsibility for fi nancial loss, as municipalities
cover their defi cits.
In addition to the collection of payments, DRGs are used for patient
classifi cation in the planning, evaluation and management of hospital services.
The motivation behind using DRGs was to simplify the hospital product defi -
nitions, in order to assess hospital performance, develop hospital operating
processes, monitor the quality of care and develop performance-based budgets.
At national level, DRGs are used for hospital benchmarking. In 1996 the then
National Research and Development Centre for Health and Welfare (STAKES,
now the National Institute for Health and Welfare) launched a project, called
the Hospital Benchmarking Project, in cooperation with the hospital districts
(Linna & Häkkinen, 2008). Its main purpose was to provide hospital managers
with benchmarking data for improving and directing activities in hospitals.
The project designed and implemented an Internet-based information system
supporting continuous data gathering and processing, as well as displaying
benchmark measures at the desired level of aggregation. The project has
taken advantage of the existing information systems in hospitals (the patient
administration systems, cost-accounting and pricing/reimbursement data, and
cost-administration processes) to collect patient-level data on delivered services
326 Diagnosis-Related Groups in Europe
and their costs. Now, annual data are collected routinely. Productivity and
effi ciency calculations are made with traditional activity measures, such as DRG
admissions and outpatient visits, with a more advanced DRG-weighted measure
for episodes of care.
The quality as well as effi ciency of specialized care has been evaluated since
2004 in the PERFECT project (PERFormance, Effectiveness and Cost of Treatment
episodes).1 Within the framework of this project, protocols have been devel-
oped for eight diseases/health problems (acute myocardial infarction; revascular
procedures (percutanous transluminal coronary angioplasty, coronary artery
bypass grafting); hip fracture; breast cancer; hip and knee joint replacements;
very low birth weight infants; schizophrenia; and stroke) (Häkkinen, 2011). The
development has been undertaken in seven separate expert groups, the members
of which are leading clinical experts on the aforementioned diseases. DRGs are
used for calculating the costs of diseases. At present, register-based indicators
(at both the regional and hospital levels) relating to the content of care, as well
as costs and outcomes between 1998 and 2008 are available for seven health
problems. The indicators are freely available on the Internet, and they are to be
routinely updated using more recent information. They have been widely
used in local decision-making and have also been discussed in the media. The
Ministry of Social Affairs and Health uses the information in strategic planning:
the indicators developed in the project will be used to evaluate the develop-
ment of regional differences in the effectiveness of specialized care. The Minis-
try has also used the information in its recommendation concerning the
centralization of certain services (such as care of low-birth-weight infants) to
university hospitals with adequate resources.
18.2 Development and updates of the DRG system
18.2.1 The current DRG systems at a glance
Currently, two different DRG systems are in use: the NordDRG Classic and the
NordDRG Full systems (see Chapter 16). The main difference between these two
groupers is that the Classic system covers only hospital inpatient and day-case
surgical activities, whereas the Full system extends the coverage to hospital
outpatient activities, that is, to scheduled and emergency visits. Because of the
lack of outpatient groups, the Classic DRGs were supplemented in productivity
analysis with a separate outpatient grouping based on visit types by specialty.
This grouping has been developed as part of the Finnish Hospital Benchmarking
Project and is not used as a pricing method.
In the Classic DRG system, inpatient and surgical day cases with the same
diagnosis and procedures are grouped to the same DRG. The Full DRG system
takes lengths of stay into account and classifi es one-day cases into so-called
O-groups. As such, the O-groups contain both surgical and medical day cases.
The O-groups are equivalent to inpatient groups, except that they have a lower
cost weight indicating the lower cost structure of day cases.
Finland: DRGs in a decentralized health care system 327
18.2.2 Development of the DRG systems
The development of the NordDRG system is described in the NordDRG chapter
of this volume (see Chapter 16). In Finland the growing signifi cance of hospital
outpatient services during the late 1990s created demand for a more advanced
grouping structure for these services. In 2004 the Nordic Centre for Classifi ca-
tions in Health Care implemented a NordDRG Full version for both hospital
in- and outpatients. The fi rst Full grouper comprised 831 groups, of which
91 were ambulatory care groups. The ambulatory care groups consist of series
of groups for endoscopies (so-called ‘700-series’), non-extensive procedures
(‘800-series’), and ‘short therapies’ (short-stay treatment) without signifi cant
procedures (‘900-series’). Since the fi rst version, the ambulatory care grouping
has been developed markedly; the 2010 Full grouper comprises 370 ambulatory
care groups and in total 1020 groups. The Full grouper applies the same rules
that are used in inpatient settings in assigning patients to specifi c outpatient
groups. The Finnish Full DRG version has been in use in the Helsinki and
Uusimaa hospital district (HUS) since 2004. At the beginning of 2008 the sys-
tem was also introduced in one central hospital. Since 2008, the Finnish Hospi-
tal Benchmarking Project has used the Full grouper.
A summary of all the DRG systems that have been used in Finland is pre-
sented in Table 18.2. The information represents the offi cial national DRG
groupers, which are not used in every hospital district. In the absence of
national guidelines for DRG usage, hospital districts are free to change the
national groupers; for example, splitting groups further if they fi nd it necessary
for their own purposes. The actual number of DRG groups used in different
hospitals may therefore vary.
Tab le 18.2 Main facts relating to the different DRG versions
FinDRG NordDRG
Classic
version*
NordDRG
Classic
version
NordDRG
Full version*
NordDRG
Full version
Year of
introduction
1995 1996 2010
(current)
2004 2010
(current)
(Main) Purpose Research Billing Billing
DRG system HCFA Version 3
modifi ed
HCFA Version 12
modifi ed
HCFA Version 12
modifi ed
Data used for
development
Cost data from
3 university
hospitals
Cost data from Helsinki
and Uusimaa hospital
district
Cost data from Helsinki
and Uusimaa hospital
district
Number of DRGs 470 495 650 831 1020
- of which
scheduled and
emergency visits
– – – 91 370
Applied to Some public
hospitals,
inpatients
Some public hospitals,
inpatients
Some public hospitals,
in- and outpatients
*Updated annually.
328 Diagnosis-Related Groups in Europe
18.2.3 Data used for development and
update of the DRG systems
The Finnish version of NordDRGs is based on patient-level data from the HUS
district hospitals. These hospitals provide about 30 per cent of specialized care
in the country. In HUS hospitals a patient information system is used, which
collects all relevant information needed in the DRG grouping. In 2010, about
2 million cases were used to calculate DRG cost weights for the NordDRG Full
grouper. Data are now available from all university hospitals (about 370 000
cases) for calculating cost weights for the NordDRG Classic version.
18.3 The current patient classifi cation system
18.3.1 Information used to classify patients
In the Classic DRG system, the grouping algorithm used to assign a patient to
a DRG is based on the inpatient hospital discharge dataset, which consist of:
major diagnosis, secondary diagnoses, procedures, patient characteristics
(gender, age, weight of neonates) and discharge status (death, transferred to
other institution, left against medical advice). In the Full DRG system the
grouping algorithm is similar to the Classic one, except that it uses both
in- and outpatient data to assign patients to a specifi c DRG. Moreover, it uses
the length of stay as a grouping criterion alongside the aforementioned criteria.
18.3.2 Classifi cation algorithm
In the grouping process, patient discharge data are fed into a special software
tool, the so-called ‘grouper’. The process is described in detail in the NordDRG
Users’ Manual (Nordic Centre for Classifi cation of Health Care, 2009). In
NordDRGs the grouping rules are presented as a series of ten tables.
Diagnoses and procedures that have an effect on the grouping are clustered
into larger subsets called ‘categories’ and ‘properties’. Each code belongs to only
one category, but it may have several properties. A CC property (co-morbidities
and complications) is binary; that is, it has only two levels. An OR property
(operating room procedure) is binary in the Classic DRG grouper, but in the Full
grouper it can have three values. Values 1 and 0 indicate whether a surgical
procedure has been carried out or not. According to this information, cases are
assigned into ‘surgical’ and ‘medical’ DRGs, respectively. Procedures that are
important in the outpatient setting but do not affect the DRG assignment of
hospital inpatients have OR property 2 (OR=2). In the case of hospital inpatients
OR=2 has no impact on the DRG assignment (see Figure 18.2).
Because of the complexity of the decision process, as well as for logistical
reasons, one specifi c DRG can be represented by several rows in the DRG logic
table. The complexity of the table is a refl ection of the detailed nature of the
original assignment rules. The rows in the table follow the hierarchy of the
original assignment rules. Therefore, when allocating patient cases, each row
F
ig
u
re
1
8
.2
G
ro
u
p
in
g
p
ro
ce
ss
i
n
t
h
e
N
o
rd
D
R
G
s
y
st
em
So
u
rc
e:
N
o
rd
ic
C
en
tr
e
fo
r
C
la
ss
ifi
c
at
io
n
s
in
H
ea
lt
h
C
ar
e,
2
0
0
9
.
N
ot
e:
T
h
e
C
la
ss
ic
s
y
st
em
i
s
d
ep
ic
te
d
b
y
c
o
n
ti
n
u
o
u
s
li
n
es
,
th
e
Fu
ll
s
y
st
em
a
ls
o
i
n
cl
u
d
es
t
h
e
p
ar
ts
p
re
se
n
te
d
w
it
h
d
as
h
ed
l
in
es
.
330 Diagnosis-Related Groups in Europe
has to be checked in ascending order until a match is found. The order in which
variables are checked on each row does not affect the allocation, while preceding
through the logic rows in the correct order is essential to obtain a correct
grouping result.
18.3.3 Data quality and plausibility checks
Due to the lack of national guidelines for the use of DRGs, no offi cial data
quality and plausibility checks are undertaken. Hospitals themselves are
responsible for data quality assurance. Hospitals use DRGs in billing munici-
palities for the services they have provided and it is therefore important for
the hospitals that patient cases are correctly assigned to DRG groups. Incorrect
DRG assignment leads to a failure in the billing process. After the patient is
discharged, the DRG grouping system performs the grouping automatically,
using information on diagnoses; procedures; and patients’ age, gender and
discharge status. Special attention has been paid to the coding of diagnoses and
procedures. One problem has been the insuffi cient coding of secondary diagno-
ses and additional procedure codes – so-called ‘Z-codes’ which indicate, for
example, long lengths of stay, bilateral operation and emergency status. The
lack of these codes leads to a ‘down-coding’ and therefore a lower billing price.
To enable correct billing, a manual check is performed to ensure that each
patient is assigned to a correct DRG group.
18.3.4 Incentives for up- or wrong-coding
Since DRGs are not used as a prospective payment system, there are no strong
incentives for up- or wrong-coding. Hospitals only need to cover their operat-
ing costs and therefore they do not have any profi t-based incentives for up-
coding. The use of DRGs as a payment system has increased the accuracy of
coding secondary diagnoses and procedure codes. Information from the STAKES
hospital benchmarking database2 shows that the coding quality is much higher
in those hospital districts that use DRGs as a payment method, compared to the
hospitals without DRG payment.
18.4 Cost accounting within hospitals
18.4.1 Regulation
In Finland there are no national guidelines for cost accounting. Hospital
districts or individual hospitals are therefore allowed to choose their own level
of cost accounting and the cost-accounting system used. For DRG purposes,
however, a particular standard is required. Advanced patient-level cost-
accounting systems were originally only used in the HUS hospitals. Today,
some other large hospitals have also developed patent-level cost-accounting
data systems that fulfi l the requirements for calculating cost weights. In 2010,
Finland: DRGs in a decentralized health care system 331
national weights for the Classic DRG version were calculated using data from all
fi ve university hospitals.
18.4.2 Main characteristics of the cost-accounting system(s)
In the HUS hospitals, an advanced cost-accounting system (Ecomed IC) is in
use.3 Cost accounting starts with the calculation of the overhead costs, which
are then allocated to lower organizational levels, using a top-down approach.
All overhead costs (such as administrative costs) are allocated to the organiza-
tional level relevant for hospitalization days, outpatient visits, operations
or ambulatory procedures. After this stage, the ‘bottom-up’ cost-analysis phase
begins. For each treated patient the following costs are defi ned: nursing (basic
care – ‘price of the hospital day’), procedures undertaken in OR and ambulatory
care settings, radiology, laboratory tests, expensive drugs, blood products, and
pathological services (see Table 18.3). These costs include both staffi ng and
devices. The bottom-up cost accounting is undertaken in each of the fi ve hos-
pitals, at department level. ‘Controllers’ bear the main responsibility for cost
accounting, and nurses and doctors are used as experts in the process.
18.5 DRGs for reimbursement
18.5.1 Range of DRGs used for reimbursement
In Finland, DRGs are not used as a prospective payment system, as in many
other countries, but rather as a fi nancing instrument in hospital districts,
Tab le 18.3 Distribution of cost by categories used in cost accounting in the HUS, 2009
Cost category Inpatient and day
surgery DRGs
(Classic)
Scheduled and
emergency visit
DRGs (O-groups) (%)
Classic and
O-groups
total (%)
Basic (inpatient) care 50.5 14.1 45.8
Basic (outpatient) care 0.2 14.2 2.0
Inpatient consultations 0.4 0.0 0.3
Laboratory tests 4.2 1.4 3.9
Blood products 1.9 0.2 1.7
Pathological services 1.1 1.4 1.2
Physiological services 0.1 0.0 0.1
Radiology 3.7 2.7 3.6
OR procedures 26.9 49.8 29.8
Procedures in outpatient
departments
9.1 15.4 9.9
Expensive drugs 1.6 0.4 1.4
Expensive products 0.3 0.5 0.4
Total 100.0 100.0 100.0
Share of total cost 87.1 12.9 100.0
Source: Compiled by the authors on the basis of information provided in a personal
communication from Virpi Alander (HUS).
332 Diagnosis-Related Groups in Europe
used to collect payments related to services use by municipalities. As explained
in section 18.1.1 the total budget (including capital, administration) for hospital
districts is decided fi rst and prices are then set in such a way that they fi t the
budget.
Not all hospitals use the system, and those that do use it do so in different
ways. For example, some hospitals have split the DRGs further where necessary
for their own purposes, which is acceptable due to the lack of national guidelines
relating to the use of DRGs.
There are no national guidelines that obligate hospitals to use DRG. The
Finnish National DRG Centre – which is a part of the private FCG Finnish
Consulting Group OY – maintains and develops the groupers, as well as provid-
ing recommendations for their use. Currently 13 out of 21 hospital districts use
DRG billing, but the extent to which it is used varies a lot between the hospital
districts. Moreover, the type of services covered by DRG billing varies; in all
hospitals, psychiatric patients and patients requiring long-term intensive treat-
ment (such as patients with respiratory arrest) are excluded and in some hospi-
tals dermatological and cancer patients, for example, are also excluded. DRG
billing covers outpatient visits (completely or partially) in four hospital dis-
tricts. The pricing and billing of services excluded from the DRG system is based
on bed days or treatment packages in inpatient settings and on visit types by
specialty in outpatient settings.
18.5.2 Calculation of DRG prices/cost weights
The national cost weights calculated by the National DRG Centre are based on
patient-level costing data from the HUS and (since 2010) university hospitals
(the costing data are described in more detail in subsection 18.4.2). The
National DRG Centre also calculates cost weights for individual hospitals based
on their patient-level cost data. Trimming is used in defi ning the average cost
of a DRG group, and because of skewed distribution (SD) of cost. The trimming
process is depicted in Figure 18.3. The trimming is undertaken in two phases: in
the 1st phase, patient cases for which treatment costs are ± 3 SD from the mean
cost of all patient cases are excluded; in the 2nd phase, patient cases for which
the treatment costs are ± 2 SD from the mean cost of the 1st phase patient
population are excluded. In 2010 the centre also calculated the outliers using a
method based on variation coeffi cients. The National DRG Centre recommends
that university hospitals should use outlier methods in their pricing, since a
considerable proportion of the high-cost patients come from municipalities not
belonging to their own hospital districts. The centre does not recommend
applying an outlier approach in central and regional hospitals, in which
almost all patients came from municipalities within their own districts.
18.5.3 DRGs in actual hospital payment
Five out of thirteen hospital districts (using DRG as payment method) have cal-
culated cost weights based on their own patient-level costing data. Other hospi-
tal districts use the national relative cost weights calculated by the National
Finland: DRGs in a decentralized health care system 333
DRG Centre, but have calculated the price for a DRG point (that is, base rate),
based on their own cost accounting. The price of a DRG point represents the
costs of an average DRG group in the whole system; the billing price of each
individual DRG group is obtained by multiplying the price of a DRG point by
the respective cost weight. In most hospital districts, irrespective of whether
they use their own or national cost weights, the price for a DRG point is defi ned
separately for each hospital, and in many hospitals, separately for each depart-
ment. This is because the casemix complexity – and hence the average treatment
cost – varies between hospital and department types; this variation is not cap-
tured perfectly by the cost weights. For example, large university hospitals treat
on average more complex and expensive patients than small local hospitals, and
the current DRG system is too rudimentary to take this into account.
Many hospitals have a defi ned upper outlier limit based on the number of
bed days above which DRG billing is not used, but the billing is based instead
on bed days. The outlier limit varies markedly between hospital districts and
this creates great variation in terms of the billing prices for the outlier patients.
For these patients the billing price consists of the DRG price and the sum of
outlier bed-day prices.
As already explained, the use of DRGs varies considerably between hospital
districts. In the HUS, where DRGs have been in use since 1998, about 65 per
cent of service charges paid by the municipalities were based on NordDRGs in
2008; the rest of the billing was based on bed days and outpatient visits (HUS,
2009). Similar data are not available from other hospitals.
18.5.4 Quality-related adjustments
As explained in subsection 18.1.2, the budgetary and payment system used
in Finland does not create similar incentives for hospitals to use DRGs to
those that a prospective payment system is known to create. There are no
Figure 18.3 Defi nition process of the average cost of a DRG group
Source: Compiled by Jorma Lauharanta.
334 Diagnosis-Related Groups in Europe
quality-related adjustments in pricing and, thus, there are no fi nancial incen-
tives for effi ciency or quality. In Finland these aspects are thought to have been
taken into account by involving municipalities as purchasers as well as owners
in decision-making within the hospital districts. The aim of comparative infor-
mation on productivity, effi ciency and outcomes is to help the local decision-
makers to improve their performance, but the use of this information in
decision-making varies considerably between hospital districts (Junnila, 2004;
Linna & Häkkinen, 2008).
18.6 New/innovative technologies
18.6.1 Steps required prior to introduction in hospitals
Funding decisions are made at hospital district or department/clinic level,
based on the total budget approved by the hospital district. There is no national
regulation. It is assumed that the introduction of new technologies is based
on health technology assessment. It is generally accepted that before the intro-
duction of new equipment, treatment practices or drugs can take place, evi-
dence relating to cost–effectiveness is required. University hospitals and some
large central hospitals carry out these kinds of research activities on their own.
The Finnish Offi ce for Health Technology Assessment (Finohta) – in coopera-
tion with hospital districts – provides information on the effectiveness and
safety of new technologies for national use. Systematic literature reviews are
undertaken in order to gather all relevant information. Finohta does not issue
any guidelines, but it provides hospitals with information needed for decision-
making. However, there is no exact information available as to how decisions
are made in each hospital district and to what extent they are based on scien-
tifi c evidence.
18.6.2 Payment mechanisms
Hospital districts have different practices for funding new technologies. The
funding can be based on surpluses from previous years, internal fi nancing or
loans. The costs of new technologies are at least partly transferred to DRG prices
via depreciation and interest on loans. Exact information is not available
regarding the principles of funding new technologies in hospital districts.
18.6.3 (Dis-)incentives for hospitals to use new technologies
No direct fi nancial incentives or disincentives exist for the use of new
technologies. The system is based on the idea that municipalities – as providers
of specialized care for their citizens and as fi nanciers of hospital districts – are
interested on the one hand in receiving the best possible care for their popula-
tions and on the other hand in controlling for the how the money they have
paid to hospital districts has been used. With this dual role in mind, municipalities
Finland: DRGs in a decentralized health care system 335
are likely to be interested in making sure that the money is used for effective
technologies, rather than it being wasted on something less effective.
18.7 Evaluation of the DRG system(s) in Finland
18.7.1 Offi cial evaluation
In Finland the right to make decisions regarding specialized care is afforded
to hospital districts, which can decide independently their own method of
charging municipalities for the services that their citizens use. It has therefore
not – so far – been of interest to the central Government to evaluate how the
DRG system (or other prevailing payment systems) works in practice.
18.7.2 Authors’ assessment
In Finland the main problem concerning DRGs is that there are no national
guidelines on how to use the system. Not every hospital uses the system and
those that do are free to use it in different ways. The main purpose of the DRG
system in Finland is to make hospital billing transparent by encouraging hospi-
tals to introduce the same billing system. However, as long as hospital districts
keep modifying the system in order to make it perfectly suitable for each indi-
vidual hospital, using different rules (for example, differing outlier limits and
department-level DRG point prices), comparison of billing prices between hos-
pitals is impossible. Each municipality is obligated to be a member of one hos-
pital district and without being able to compare the prices, municipalities are
not able to make rational choices. This prevents competition, which, in turn,
does not incentivize hospitals to improve their operating effi ciency. In order to
make the DRG billing system function properly and to enhance effi ciency, the
Ministry of Social Affairs and Health should introduce national rules and obli-
gate all hospital districts to use the DRG billing, as has been implemented in
many other countries. This is also important because private and non-profi t-
making – and even multinational – fi rms are now entering the health care
market. In addition, a new Health Care Act (accepted by the Finnish Parliament
in December 2010) introduces patient choice of hospital. However, so far, it has
not been decided (or indeed proposed) how municipalities would pay hospitals
within the new framework. If the central Government is to take a more active
role in developing the pricing rules, the development of the DRG system should
also be carried out by a public authority, not by a private fi rm (National DRG
Centre) as is currently the case in Finland.
In individual hospitals the introduction of DRG billing has brought about
improvements. In these hospitals it has been necessary to pay more attention
to the coding of diagnoses and procedures, and indeed the coding has been
improved compared to hospitals not using the DRG billing system. This has
made hospital billing more accurate, but also improved hospital management.
It is therefore possible that the introduction of a DRG-based pricing system
could lead to much improvement in the management and provision of hospital
336 Diagnosis-Related Groups in Europe
services, such as greater transparency and more accurate cost information
(Häkkinen & Linna, 2005). However, if the system moves in the direction of a
general prospective payment system, the potential for incentives/bias should be
considered (see Chapter 6 of this volume). Given the current structure of the
Finnish health care system, the choice of pricing method is not the most crucial
one to be made. Efforts should be directed towards more important questions,
such as the development of contracts between municipalities and hospitals,
the management and control of care chains (total episodes of care), quality of
services, or (even more generally) the governance role of central Government,
as well as the issue of centralization – that is, increasing the size/impact of the
purchasing and providing functions. All these elements are currently under
considerable scrutiny in Finland.
18.8 Outlook: Future developments and reform
18.8.1 Trends in hospital service (or general
health care) delivery
Finland places a strong emphasis on the public provision of health services.
However, the importance of private service provision has been growing rapidly
in recent years. Simultaneously, the boundary between public and private
service provision has become blurred. Public hospitals order services from
private producers if they are not able to produce the services themselves within
the required time scale. Municipalities can also order private services directly.
This trend is expected to continue.
Finnish public hospitals have not traditionally been highly specialized. Hos-
pital districts have tried to be as self-suffi cient as possible in treating their
patients. In recent years it has been realized that it is not effi cient, or even con-
ducive to delivering high-quality health care, to provide all services in each
hospital district. A trend has been developing towards more specialized units.
At the same time, an increasing number of patients are treated as outpatients,
in day-care and ambulatory care settings. This has led to the reduction of
inpatient capacity.
Since the early 2000s, several local reforms have been implemented to inte-
grate municipal service provision into a single organization. The purpose of
these reforms is to enhance cooperation between primary and secondary health
care and social welfare services (Vuorenkoski, 2008). The reforms include merg-
ing health centres and local hospitals into one organization, creating new
regional self-regulating administrative bodies for all municipal services (includ-
ing health and social services, upper secondary schools, and vocational ser-
vices) with their own regional councils, and hospital districts also taking
responsibility for primary health care. In 2008 about 10 per cent of the Finnish
population lived in areas in which most primary and secondary care is provided
by the same organization. The most recent initiative from the Ministry of Social
Affairs and Health is to create 40–60 health and social regions (federations of
municipalities or large municipalities) that are responsible for social services as
well as primary and (most) specialist care services, along with fi ve districts with
Finland: DRGs in a decentralized health care system 337
special responsibility, which would be responsible for the most expensive ter-
tiary care. If the trend continues and the proposal is to be implemented, the
contract and payment systems for hospital care should also be reconsidered.
18.8.2 Trends in DRG application/coverage
In 2005−2007 the National DRG Centre organized a project related to the usage
of DRGs in Finnish hospitals, in cooperation with Finnish DRG experts and
hospital districts. In the fi nal report (Kuntaliitto, 2007) the project team offers
suggestions and a schedule relating to how to proceed in the implementation
and development of a DRG system in hospitals in the near future. The main
targets are as follows.
• The DRG billing would be implemented in all Finnish public hospitals by
2010, covering at least inpatient and day-care activities.
• The coverage of DRG billing would be extended to outpatient services by
2011 and at the same time the grouping would be developed to correspond
better to outpatient and psychiatric services.
• By 2011 all the hospitals would have advanced patient-level cost-accounting
systems in place, in order to calculate their own DRG cost weights.
• Hospitals would stop using the department-level pricing and use the same
DRG prices across departments within a hospital.
• Hospitals would use a national handbook in order to ensure that all hospitals
apply the same principles for cost accounting and coding for diagnoses and
procedures; the coding handbook should be available on the Internet and it
should be updated constantly.
• A certifi cation system should be created for hospitals, which would obligate
them to maintain their own system of internal quality standards in terms of
coding, and to submit to regular external auditing.
Most hospitals that do not currently use the DRG billing system have already
launched a DRG implementation project and are planning to introduce the
system as soon as possible. For some hospitals, however, the target time frame
for introducing the system by 2010 was too tight. Similarly, many hospitals
currently using the Classic DRG system are preparing to implement the Full
DRG system. Currently, four of the thirteen hospital districts (using DRGs) use
the Full DRG system, at least in part.
There has been a trend towards outpatient production in Finnish hospitals
since the early 2000s and the importance of developing a grouper which is able
to take into account treatment episodes in ambulatory care is therefore grow-
ing. The Full DRG grouper contains outpatient groups, but it still functions
mostly on a fee-for-service basis. The challenge will be to develop the Full DRG
system so that it will be able to capture the whole treatment pathway, instead
of separate visits. This will not be straightforward, as patient treatment in an
ambulatory care setting is not as homogeneous as in an inpatient setting. Fur-
ther development of inpatient groups is also needed, in order to ensure that
they better take into account patient casemix. Currently, the DRG system
underestimates the complexity level of patients treated in university hospitals.
338 Diagnosis-Related Groups in Europe
18.9 Notes
1 More information available on the National Institute for Health and Welfare web site
(http://www.thl.fi /fi -FI/web/fi /tutkimus/hankkeet/perfect, accessed 10 July 2011).
2 Outdated database available at the STAKES web site (http://info.stakes.fi /benchmarking/
EN/benchmarking.htm, accessed 1 August 2011).
3 The system was developed by a private fi rm, Datawell. It was fi rst introduced in the
HUS but is now used in many other hospitals.
18.10 References
Häkkinen, U. (2005). The impact of changes in Finland’s health care system. Health
Economics, 1(September):101–18.
Häkkinen, U. (2009). Finland, in E. Jonsson. Cost Containments in National Health Systems.
Weinheim: Wiley International Science.
Häkkinen, U. (2011). The PERFECT project: measuring performance of health care
episodes. Annals of Medicine, 43(S1):1–3.
Häkkinen, U., Lehto, J. (2005). Reform, change and continuity in Finnish health care.
Journal of Health Politics, Policy and Law, 30(1–2):76–96.
Häkkinen, U., Linna, M. (2005). DRGs in Finnish health care. Euro Observer, 7(4):7–8
(http://www2.lse.ac.uk/LSEHealthAndSocialCare/LSEHealth/pdf/euroObserver.
Obsval7n04 , accessed 10 July 2011).
HUS (2009). Tilinpäätös ja toimintakertomus 2008 [Financial Statement and Annual Report
2008]. Helsinki: Hospital District of Helsinki and Uusimaa.
Junnila, M. (2004). Sairaaloiden tuottavuus: Benchmarking-tietojen käyttö erikoissairaanhoidon
toiminnan suunnittelussa, seurannassa ja arvioinnissa [Productivity of Hospitals: the Use of
Benchmarking Information on the Planning, Follow-up and Evaluation of Specialized Care].
Helsinki: National Institute for Health and Welfare (STAKES) (Reports 280).
Kuntaliitto (2007). Erikoissairaanhoidon palvelujen tuotteistus Suomessa [Product Defi nition in
Specialized Care in Finland]. Helsinki: Kuntaliitto.
Linna, M., Häkkinen, U. (1998). A comparative application of econometric and DEA
methods for assessing cost-effi ciency of Finnish hospitals, in P. Zweifel, ed. Health, the
Medical Profession and Regulation. Boston: Kluwer Academic Publishers.
Linna, M., Häkkinen, U. (2008). Benchmarking Finnish hospitals, in J. Blank, V. Valdmanis,
eds. Evaluating Hospital Policy and Performance: Contributions from Hospital Policy and
Productivity Research. Oxford: JAI Press.
Magnussen, J., Vrangbæk, K., Saltman, R.B. (2009). Nordic Health Care Systems: Recent
Reforms and Current Policy Challenges. Maidenhead: Open University Press.
Nordic Centre for Classifi cations in Health Care (2009). NordDRG Users’ Manual (Version
2009). Oslo: Nordic Centre for Classifi cations in Health Care (http://www.norddrg.
net/norddrgmanual/NordDRG_2009_NC/index.htm, accessed 23 October 2009).
OECD (2005). OECD Reviews of Health Systems – Finland. Paris: Organisation for Economic
Co-operation and Development.
THL (2010). Terveydenhuollon menot ja rahoitus vuonna 2008 [Health Expenditure and
Financing 2008]. Statistical Report 12/2010. Helsinki: National Institute for Health and
Welfare (STAKES).
Vuorenkoski, L. (2008). Finland: Health system review. Health Systems in Transition,
10(4):1–170.
chapter n i n e t e e n
Sweden: The history,
development and
current use of DRGs
Lisbeth Serdén and Mona Heurgren
19.1 Hospital services and the role of DRGs in Sweden
19.1.1 The Swedish health care system
Sweden has a decentralized health care system. There are three political and
administrative levels; central Government, county councils and local muni-
cipalities. All are involved in fi nancing, providing and evaluating health care
activities. The central Government has a legislative supervisory role and par-
tially fi nances health care, while the county councils and municipalities are
responsible for both fi nancing and providing health services. The municipalities
and county councils are also politically accountable through their directly elect-
ed assemblies. The 21 county councils/regions are responsible for most health
care services, except long-term care of the elderly and disabled people (including
mentally ill people), for whom the 290 municipalities are responsible.
The Swedish health care system is mainly fi nanced by taxes. The county coun-
cils and municipalities are entitled to collect direct income tax revenues as their
major fi nancial source; council tax amounts to about 10 per cent of the residents’
income. There is also a grant from the Government, which amounts to about
9 per cent of the counties’ revenue. The individual patient’s co-payment is low –
fees account for about 2.5 per cent of the total county revenues. In total, Sweden
spent about 9 per cent of its gross domestic product (GDP) on health care in 2009.
19.1.2 Hospital services in Sweden
In total there are 81 hospitals in Sweden (as of 2009). Of these hospitals, seven
are university hospitals (with 7300 disposable beds) (for more details relating
340 Diagnosis-Related Groups in Europe
specifi cally to acute care hospitals, see Table 19.1). The vast majority of Swedish
hospitals are publicly funded. There are only three private profi t-making
hospitals and some smaller private non-profi t-making hospitals (SALAR, 2010).
Annually, there are about 1.5 million inpatient care cases and 10 million
visits by physicians for specialized outpatient care carried out at Swedish
hospitals. In addition, there are 3 million private specialist visits, outside of the
hospitals. Acute cases account for 75 per cent of inpatient care and elective care
for 25 per cent. Outpatient care is distributed as 25 per cent acute care and
75 per cent elective care (Forsberg et al., 2009).
The hospital services encompass inpatient care, day surgery, day medicine
and specialized outpatient care. Inpatient care is divided into specialized care,
psychiatry, rehabilitation and geriatrics. Rehabilitation care is mostly carried
out as a hospital treatment, but there are also units that offer rehabilitation
services as care delivered outside of hospitals. GPs in ambulatory care refer
patients to specialists at hospitals. All elective patients at hospitals are referred
from GPs in their role as gatekeepers.
Each of the 21 counties/regions decides independently how their health care
should be organized and reimbursed.
19.1.3 Purpose of the DRG system
In the early 1990s, the Swedish health care system needed save money. There
was also a strong movement towards a more patient-oriented system. This
signalled the emergence of incentives to start using diagnosis-related groups
(DRGs); the main motive for introducing DRG-based payment schemes was to
increase productivity and thereby to save (or make better use of) the money
used for health care. Long waiting lists for elective surgery were another reason
for this change. A third important reason for introducing DRG-based payment
systems was to allow the patient freedom of choice to select a hospital for
treatment. The idea was that, by giving freedom of choice to the patients – and
if the money follows the patient – a degree of competition could be introduced
among the hospitals. By providing good services and thereby attracting patients,
the hospitals would secure higher revenues. There was also a need for higher
quality information and greater transparency in health care. The global budget
Table 19.1 Hospital beds in acute care, psychiatric and
long-term care, and beds per 1000 people, 2009
2009
Acute care hospital beds 18 944
per 1000 people 2.0
Psychiatric hospital beds 4 449
per 1000 people 0.5
Long-term care beds 2 167
per 1000 people 0.2
Source: SALAR, 2010.
Sweden: The history, development and current use of DRGs 341
encouraged neither productivity nor patient-oriented care, so the move to a
DRG-based funding system was initiated. However, the need for cost control
was also important, and budget ceilings were introduced to prevent oversupply
and overuse.
In accordance with the counties’ right to self-determination in health
care activities, use of the Nordic patient classifi cation system (NordDRG)
in Sweden is voluntary. The counties decide for themselves independently
how to use DRGs within their own payment systems and what comple-
mentary rules should be applied (such as reimbursement of outliers, cost ceil-
ings, and so on). The counties are also responsible for the follow-up of fraudulent
activity and any other misuse of the system. The availability of health care
represents another issue for the county councils. The most common method
used for controlling the supply of health care activities has been to limit
availability.
Today, aside from their application as a payment mechanism, DRGs are used
for managerial purposes, benchmarking, health statistics, measuring hospital
performance and calculating productivity (and effi ciency) at all levels of health
care. The National Board of Health and Welfare has started working to fi nd a
method for calculating effi ciency in Swedish health care. In order to do so,
DRGs were used to describe performance, as well as process costs.
19.2 Development and updates of the DRG system
19.2.1 The current DRG system at a glance
The NordDRG system is currently the only DRG system in Sweden. Thus far
there are only two different licensed software suppliers for NordDRGs. Each of
them provides groupers, either available as interactive single cases, or as a
‘batch’ grouper.
The Full version of NordDRG 2009 embraces a total of 983 DRGs (see Table
19.2). Of these groups there are 216 outpatient groups designed for day surgery,
day medicine and endoscopies. There are also 190 groups for specialized
Table 19.2 Number of NordDRG codes in different settings in 2009
Setting Number
Inpatient care 577
– Specialized care 514
– Psychiatry 30
– Rehabilitation 33
Day surgery 162
Day medicine 34
Endoscopy 20
Outpatient specialized care 190
Total 983
Source: Nordic Centre for Classifi cations in Health Care, 2009.
342 Diagnosis-Related Groups in Europe
outpatient care visits. The groups for other day-treatment visits and outpatient
care carry an ‘O’ or a ‘P’ at the end of the DRG code. Day surgery is allocated the
same number as the corresponding inpatient group, but with an ‘O’ in the
DRG-code (Nordic Casemix Centre, 2011).
The counties in Sweden can be divided into three categories with regard
to their usage of DRGs. The fi rst category uses DRGs for reimbursement to hos-
pitals for a large range of care (both in- and outpatient care, to some extent).
The eight counties/regions in this category represent more than half of the
Swedish health care system (calculated by health care expenditure). Psychiatry
is included in the payment system of one of the counties. The second category
of counties use DRGs only as a tool for analysis, to calculate casemix, for hospi-
tal budgeting or for reimbursement of patients across county borders. The third
category of counties uses DRGs as a component in the reimbursement system
for a smaller component of health care; for example, for patients across county
borders, or for a single hospital.
In total, about 90 per cent of inpatients are grouped into DRGs, and 65 per
cent are fi nanced by DRGs. In outpatient care, 80 per cent are grouped into
DRGs, and 30 per cent are fi nanced by them.
19.2.2 Development of the DRG system
The National Board of Health and Welfare is responsible for developing and
maintaining the Swedish version of the NordDRG system. In validating the
resource homogeneity process in DRGs, the Board cooperates with the Swedish
Association of Local Authorities and Regions (Sveriges Kommuner och Landsting,
SALAR) which is responsible for the Swedish National Case Costing Database.
All cost data in use with respect to DRG maintenance are calculated using a
‘bottom-up’ approach.
The Swedish NordDRG version has been developed to comprise both in- and
outpatient care, as well as psychiatry and rehabilitation (see Table 19.3). The
NordDRG system can be implemented in any type of hospital. In 2011, the
15th version of NordDRG was introduced.
Between 2003 and 2009, two versions of NordDRG were operating in
Sweden; a Full version, which handled both inpatient care and day surgery
(including intraluminal endoscopies), along with outpatient specialized care,
and a Classic version for inpatient care only. Since 2010, just one version of
NordDRG is in effect for handling all in- and outpatient care using the same
logic.
19.2.3 Data used for development
and updates of the DRG system
Test data from the National Patient Register (NPR) (except the personal identi-
fi cation number) are used to inform the update process. The National Case
Costing Database is used to validate the resource homogeneity in the DRGs.
T
a
b
le
1
9
.3
V
ar
io
u
s
N
o
rd
D
R
G
v
er
si
o
n
s
in
S
w
ed
en
,
1
9
9
5
–2
0
0
8
(
se
le
ct
ed
y
ea
rs
)
1
st
D
R
G
v
er
si
on
6
th
D
R
G
v
er
si
on
8
th
D
R
G
v
er
si
on
1
0
th
D
R
G
v
er
si
on
1
1
th
D
R
G
v
er
si
on
1
2
th
D
R
G
v
er
si
on
D
a
te
o
f
in
tr
od
u
ct
io
n
1
9
9
5
2
0
0
1
2
0
0
3
2
0
0
5
2
0
0
6
2
0
0
8
(M
ai
n
)
P
u
rp
o
se
R
ei
m
b
u
rs
em
en
t,
an
d
t
o
d
es
cr
ib
e
p
er
fo
rm
an
ce
R
ei
m
b
u
rs
em
en
t,
an
d
t
o
d
es
cr
ib
e
p
er
fo
rm
an
ce
R
ei
m
b
u
rs
em
en
t,
an
d
t
o
d
es
cr
ib
e
p
er
fo
rm
an
ce
R
ei
m
b
u
rs
em
en
t,
an
d
t
o
d
es
cr
ib
e
p
er
fo
rm
an
ce
R
ei
m
b
u
rs
em
en
t,
an
d
t
o
d
es
cr
ib
e
p
er
fo
rm
an
ce
B
en
ch
m
ar
k
in
g
P
ro
d
u
ct
iv
it
y
M
ea
su
re
m
en
t
R
ei
m
b
u
rs
em
en
t,
an
d
t
o
d
es
cr
ib
e
p
er
fo
rm
an
ce
B
en
ch
m
ar
k
in
g
P
ro
d
u
ct
iv
it
y
M
ea
su
re
m
en
t
D
R
G
s
y
st
em
N
o
rd
D
R
G
N
o
rd
D
R
G
N
o
rd
D
R
G
N
o
rd
D
R
G
N
o
rd
D
R
G
N
o
rd
D
R
G
D
at
a
u
se
d
f
o
r
d
ev
el
o
p
m
en
t
C
o
st
w
ei
g
h
ts
,
U
SA
N
at
io
n
al
C
o
st
D
at
ab
as
e,
N
P
R
N
at
io
n
al
C
o
st
D
at
ab
as
e,
N
P
R
N
at
io
n
al
C
o
st
d
at
ab
as
e,
N
P
R
N
at
io
n
al
C
o
st
D
at
ab
as
e,
N
P
R
N
at
io
n
al
C
o
st
D
at
ab
as
e,
N
P
R
N
u
m
b
er
o
f
D
R
G
s
5
0
0
4
9
8
7
2
2
7
4
0
9
2
9
9
7
6
A
p
p
li
ed
t
o
A
ll
h
o
sp
it
al
s,
o
n
ly
i
n
p
at
ie
n
ts
A
ll
h
o
sp
it
al
s,
o
n
ly
i
n
p
at
ie
n
ts
in
cl
u
d
in
g
c
h
il
d
re
n
/
n
eo
n
at
o
lo
g
y
A
ll
h
o
sp
it
al
s,
in
p
at
ie
n
ts
a
n
d
d
ay
s
u
rg
er
y
A
ll
h
o
sp
it
al
s,
in
p
at
ie
n
ts
,
p
lu
s
p
sy
ch
ia
tr
y
a
n
d
d
ay
su
rg
er
y
A
ll
h
o
sp
it
al
s,
in
p
at
ie
n
ts
a
n
d
a
ll
o
u
tp
at
ie
n
ts
A
ll
h
o
sp
it
al
s,
in
-
an
d
o
u
tp
at
ie
n
ts
,
in
cl
u
d
in
g
re
h
ab
il
it
at
io
n
So
u
rc
e:
D
es
ig
n
ed
f
o
r
th
is
r
ep
o
rt
b
y
L
is
b
et
h
S
er
d
én
a
t
th
e
N
at
io
n
al
B
o
ar
d
o
f
H
ea
lt
h
a
n
d
W
el
fa
re
2
i
n
2
0
1
1
.
344 Diagnosis-Related Groups in Europe
This database contains bottom-up cost data collected directly from hospitals.
The hospitals join the database on a voluntary basis. National cost data have
been available since 1997 and the quality of the data has improved signifi cantly
over time (see section 19.4) (Ludvigsson et al., 2011).
The tradition of collecting data at the individual patient level is strong in
Sweden; the NPR has been in use since the 1960s and contains all individual
inpatient records in Sweden. All data can be linked to the individual patient by
the personal identifi cation number that is given to all citizens at birth. The
register has traditionally been used mainly for research purposes. The use of the
register for measuring productivity and various types of follow-ups in health
care is gradually increasing over time.
19.2.4 Regularity and method of system updates
The NordDRG system is updated yearly. The original development work carried
out in advance of major changes is normally conducted in each country that
uses the system. Some major changes can be implemented as joint projects
between the countries concerned.
Changes in DRGs may be initiated by problems with either cost heterogeneity
or clinical relevance, according to the basic concept that applies to all DRG
systems: patient cases are to be assigned to clinically relevant groups with the
least possible variance in cost.
For all types of changes (splitting DRGs, merging DRGs, partial or total
reassignment), there are specifi ed statistical criteria that must be evaluated
with cost-per-case data from at least one of the Nordic countries. Changes can
sometimes be made even if not all criteria are met, but in those cases a clear
rationale must be put forward (Lindqvist, 2008).
19.3 The current patient classifi cation system
19.3.1 Information used to classify patients
The Nordic countries have a long tradition of collaborating on classifi cation
systems – as manifested in the Nordic Centre for Classifi cations in Health Care
(which is a WHO collaborating centre). The Nordic countries collaborate
concerning the basic classifi cations, but are also obligated to maintain national
versions of the classifi cations in their national languages.
For coding diagnosis, the Nordic countries use a national version of the
International Classifi cation of Diseases 10th revision (ICD-10) and, for surgery
procedures, the common Nordic classifi cation of surgery is used (NOMESCO
Classifi cation of Surgical Procedures, NCSP). In Sweden, a new national clas-
sifi cation system for non-surgical procedures was introduced in 2006 (KMÅ).
Combined, the classifi cation of surgery and non-surgical procedures is called
KVÅ. The surgical procedures in KVÅ are in general the same as the procedures
in the NCSP, but the medical procedures are national in scope.
Sweden: The history, development and current use of DRGs 345
19.3.2 Grouping algorithm
NordDRG is a system for classifying inpatient cases and outpatient visits into
categories with similar resource use (see Figure 19.1). The grouping is based on
diagnoses, procedures performed, age, birth weight, gender and status at dis-
charge. The history, design and classifi cation rules of the DRG system – as well
as its application in terms of patient discharge data and updating procedures –
are presented in the DRG Defi nitions Manual (Nordic Casemix Centre, 2011).
19.3.3 Data quality and plausibility checks
The NPR – managed by the National Board of Health and Welfare – is quality
checked on an annual basis. For each record reported to the NPR, a data con-
trol is performed to check that compulsory variables are reported, such as the
patient’s personal identifi cation number, the hospital and the main diagnosis.
Codes for different variables and dates are also checked. Some obviously incor-
rect data are corrected in the quality controls, while other data are sent back to
the hospital for correction. In the same way, the cost data in the National Case
Costing Database are checked annually by the SALAR.
Many performance indicators can be deducted from Swedish national health
data registers; for example, registers have been used for analysing the differ-
ences in case fatality within 28 days after acute myocardial infarction or stroke.
Signifi cant efforts are now being made, at both the national and local levels, to
fi nd valid and accepted quality indicators for following up health care perfor-
mance, and also for productivity/effi ciency measurements. Some counties
already have models in use for the follow-up of performance indicators. Since
2006, Sweden has used national performance indicators to measure perfor-
mance at county level on an annual basis. (National Board of Health and
Welfare, 2010). For the year 2010, 134 quality indicators were published. The
National Board of Health and Welfare also publishes reports on coding activity
and quality on a yearly basis.
The county councils are responsible for checking the quality of DRG-grouped
data by means case record audits. Some of the county councils carry out audits
of case records on a regular basis in order to identify incorrect coding. The
process in place in the event that fraud is identifi ed in the records differs from
county to county. In most cases the hospital or private clinic will be obligated
to pay back the discrepancy. Coding quality has improved and continues to
improve in Sweden. As such, there are attempts to introduce more time for
coding issues in physicians’ education programmes, and many county councils
are educating their medical secretaries in coding and encouraging them to play
a larger role in this fi eld.
19.3.4 Incentives for up- or wrong-coding
Very few cases of up-coding occur because of the small number of private
hospitals in Sweden. However, a problem still exists in terms of ‘down-coding’
F
ig
u
re
1
9
.1
G
ro
u
p
in
g
a
lg
o
ri
th
m
i
n
t
h
e
Sw
ed
is
h
v
er
si
o
n
o
f
N
o
rd
D
R
G
s
y
st
em
So
u
rc
e:
D
es
ig
n
ed
f
o
r
th
is
p
u
b
li
ca
ti
o
n
b
y
M
at
s
Fe
rn
st
rö
m
a
t
th
e
N
at
io
n
al
B
o
ar
d
o
f
H
ea
lt
h
a
n
d
W
el
fa
re
i
n
2
0
1
1
.
N
ot
es
:
1
O
u
tp
a
ti
e
n
t
P
re
-M
D
C
D
R
G
s:
e
x
tr
em
el
y
r
es
o
u
rc
e-
in
te
n
si
v
e
p
ro
ce
d
u
re
s
ar
e
se
ld
o
m
p
er
fo
rm
ed
o
n
o
u
tp
at
ie
n
ts
,
so
t
h
er
e
ar
e
fe
w
c
as
es
i
n
th
es
e
D
R
G
s.
2
O
u
tp
a
ti
e
n
t
S
u
rg
e
ry
D
R
G
s
(D
a
y
S
u
rg
e
ry
):
t
h
es
e
D
R
G
s
ar
e
al
so
c
al
le
d
D
ay
S
u
rg
er
y
a
n
d
t
h
ey
h
av
e
a
g
ro
u
p
in
g
l
o
g
ic
v
er
y
s
im
il
ar
t
o
th
e
In
p
at
ie
n
t
Su
rg
er
y
D
R
G
s,
b
u
t
th
er
e
is
n
o
a
g
e
sp
li
t
o
r
C
C
s
p
li
t.
3
I
n
p
a
ti
e
n
t
S
u
rg
e
ry
D
R
G
s:
a
b
o
u
t
5
0
p
er
c
en
t
o
f
th
es
e
D
R
G
s
ar
e
sp
li
t
b
as
ed
o
n
a
g
e
an
d
/o
r
C
C
.
T
h
e
ag
e
sp
li
t
is
l
es
s
th
an
1
8
y
ea
rs
,
b
u
t
fo
r
M
D
C
1
5
t
h
e
p
at
ie
n
t
m
u
st
b
e
le
ss
t
h
an
1
y
ea
r
o
ld
.
4
O
u
tp
a
ti
e
n
t
M
e
d
ic
a
l
D
R
G
s:
s
o
m
e
o
f
th
es
e
D
R
G
s
ar
e
fo
r
lo
n
g
er
v
is
it
s
(f
o
r
ex
am
p
le
f
o
r
so
m
e
h
o
u
rs
o
f
o
b
se
rv
at
io
n
)
an
d
a
re
c
al
le
d
D
ay
M
ed
ic
in
e.
5
I
n
p
a
ti
e
n
t
M
e
d
ic
a
l
D
R
G
s:
l
ik
e
th
e
In
p
at
ie
n
t
Su
rg
er
y
D
R
G
s,
a
b
o
u
t
5
0
p
er
c
en
t
o
f
th
e
g
ro
u
p
s
ar
e
sp
li
t
b
as
ed
o
n
a
g
e
an
d
/o
r
C
C
.
T
h
e
co
m
m
o
n
a
g
e
sp
li
t
is
i
d
en
ti
ca
l
to
t
h
e
sp
li
t
fo
r
In
p
at
ie
n
t
Su
rg
er
y
D
R
G
s
b
u
t
th
e
D
R
G
s
fo
r
D
ia
b
et
es
a
re
d
iv
id
ed
i
n
to
>
3
5
o
r
<
3
5
y
ea
rs
o
f
ag
e.
6
O
u
tp
a
ti
e
n
t
M
in
o
r
P
ro
c
e
d
u
re
D
R
G
s:
t
h
es
e
D
R
G
s
ar
e
fo
r
m
in
o
r
p
ro
ce
d
u
re
s,
e
x
ce
p
t
en
d
o
sc
o
p
ie
s
an
d
m
in
o
r
su
rg
er
y.
T
h
es
e
p
ro
ce
d
u
re
s
ar
e
o
ft
en
p
er
fo
rm
ed
b
y
p
er
so
n
n
el
o
th
er
t
h
an
d
o
ct
o
rs
a
n
d
t
h
er
ef
o
re
a
p
ri
n
ci
p
al
d
ia
g
n
o
si
s
is
n
o
t
m
an
d
at
o
ry
.
7
M
e
d
ic
a
l
V
is
it
D
R
G
s
fo
r
p
e
rs
o
n
n
e
l
o
th
e
r
th
a
n
d
o
c
to
rs
:
ac
co
rd
in
g
t
o
S
w
ed
is
h
l
aw
,
o
n
ly
d
o
ct
o
rs
a
re
o
b
li
g
ed
t
o
r
ep
o
rt
d
ia
g
n
o
se
s,
s
o
th
e
g
ro
u
p
in
g
l
o
g
ic
f
o
r
th
es
e
D
R
G
s
is
b
as
ed
o
n
p
ro
fe
ss
io
n
a
n
d
t
h
e
ty
p
e
o
f
v
is
it
(
si
n
g
le
,
te
am
o
r
g
ro
u
p
).
348 Diagnosis-Related Groups in Europe
(due not to failings of the fi nancial system, but rather the tradition of entering
only few codes into the system). At national level, the authorities encourage
hospitals to operate better coding practices, which has often led to a greater
number of registered secondary diagnoses per case. Systematic selection of
patients for fi nancial reasons (cherry-picking or cream-skimming) has not
occurred in public hospitals, but has occurred to some degree among private
providers in Stockholm.
Several record audits in Sweden (2300 medical records altogether) show that
abuse of secondary diagnosis coding can create an increase, but also (at the
same time) a decrease in DRG weights compared with accurate coding. Audits
can lead to adjustments in reimbursement to hospitals and other providers of
health care (National Board of Health and Welfare, 2006). Most wrong-coding
is not in fact a sign of abuse of the system, but rather a matter of ignorance.
19.4 Cost accounting within hospitals
19.4.1 Regulation
It is not mandatory to implement case-costing databases within a hospital. The
incentive for the hospitals to do so is that they will achieve not only a greater
degree of cost control within the hospital management, but also an infl uence
over the national DRG weights. National guidelines have been developed for
cost-per-case calculations. About 65 per cent of inpatient cases and 36 per cent
of outpatient visits were individually calculated in 2009.
Case costing is a costing method that uses a bottom-up approach. All costs,
including indirect costs, should be incorporated. There are some exceptions,
such as costs for research and teaching, external projects, ambulances, and the
counties’ politicians and their staff (SALAR, 2011).
19.4.2 Main characteristics of the cost-accounting system
Case costing data have been collected from hospitals with case costing sys-
tems since the mid-1990s and added to the National Case Costing Database
(Heurgren, et al., 2003; SALAR, 2011). The data are held in a common database
for the calculation of Swedish DRG weights, managed by the SALAR. The
information in the National Case Costing Database is almost the same as in the
NPR, except that the cost data are added and the personal identifi cation
numbers are missing.
Case costing model
The case costing model comprises four steps: (1) accurately identifying the total
cost of the hospital; (2) allocating indirect costs to the cost centres (that are
absorbing the cost); (3) identifying intermediate products and calculating their
costs; (4) distributing products and costs to the patients. Figure 19.2 provides
more details.
Sweden: The history, development and current use of DRGs 349
In the case costing process, all costs are distributed to the individual cases by
the unique personal identifi cation number. A patient-specifi c service mix is tied
together with patient characteristics. The case costing system provides informa-
tion about activities such as surgery, laboratory tests, intensive care, and nurs-
ing care. Their costs are calculated uniquely for each patient. The system also
holds information on diagnoses, procedures, DRG, and so on, for each patient.
National guidelines have been implemented to ensure data quality and com-
parability (SALAR, 2009).
The most common IT-structure for case costing is the integration model, in
which data are collected from various databases in the hospital into a ‘data
warehouse’. The case costing system uses these data to link the relevant case
and patient by means of the personal identifi cation number for each patient
and the date of their stay or visit (see Figure 19.3). It is also possible to collect
data for a case costing system directly from medical records, but in general this
does not take place in Sweden (except for a few hospital departments), since cur-
rent medical records are not designed for this purpose.
The IT structure is important when implementing case costing systems.
Access to information must be highly automated, and an important principle
to apply is to use existing data as much as possible in order to minimize manual
work and obtain higher data quality.
The use of case costing is important from many perspectives. In Sweden, case
costing data are used in the following areas:
• management support for hospital departments and hospitals – process
management;
• support for buyers of health care;
• benchmarking studies of costs and medical praxis;
• development of the DRG system and calculating prices in health care;
• calculating relative cost weights in the NordDRG system;
• calculating productivity and effi ciency.
Figure 19.2 The Swedish case costing model
Source: SALAR 2009.
350 Diagnosis-Related Groups in Europe
19.5 DRGs for hospital payment
19.5.1 Range of services and costs included
in DRG-based hospital payments
In the public health care sector, the decision to use a DRG for reimbursement is
made at the county council level.
Swedish hospitals are traditionally fi nanced via global budgeting. This is due
to the fact that Sweden has a tradition of publicly owned hospitals, and there-
fore cost control has always been an important issue. Moreover, before the intro-
duction of DRGs, there was no general, accepted system in use for describing
performance. The counties’ knowledge of hospital activity and productivity was
poor, and therefore a great need existed to fi nd ways to measure productivity.
The main reasons for introducing the DRG system as a fi nancing tool for
hospitals included to improve productivity and effi ciency, to increase trans-
parency in the hospital sector, and to create a ‘market’, with purchaser and pro-
viders sharing the fi nancial risk.
In Sweden there are differences in the way DRGs are used for fi nancing in
different types of hospitals (regional, teaching hospitals, acute care, psychiatric,
rehabilitation, and so on). Until recently, DRGs were used only for somatic
care (inpatient and outpatient) in Sweden. In 2010, one county used DRGs for
fi nancing psychiatry. Rehabilitation is another new fi eld for DRG use in
Sweden, introduced in 2008, but only a few counties have adopted DRGs for
fi nancing purposes. Others simply use this part of the DRG system to describe
performance.
On the other hand, the use of DRGs has been quite similar within hospitals,
regardless of whether they are teaching hospitals or rural hospitals. Both acute
and planned care have been included, but the teaching hospitals have
Figure 19.3 An example of a patient in the Swedish National Case Costing
Database: femur fracture
Source: SALAR, 2009.
Sweden: The history, development and current use of DRGs 351
incorporated more exceptions from the DRG list, such as fee-for-service prices
for unusual and costly treatments. There are no differences in the way DRGs are
used in terms of the legal status of the hospital (that is, whether it is public,
private profi t-making or non-profi t-making).
In Sweden, the cost all health professionals’ (for example doctors’, nurses’)
fees are included in the DRG weights and prices. The vast majority of health
care professionals are employed by the hospitals/counties. The costs of infra-
structure, important medical equipment and installations, communication sys-
tems or informatics are also included.
Costs for outliers are not included in the DRG weights. Burn injuries are also
not included. In some counties, specifi c regional care and rehabilitation are
excluded. Some unusual and expensive drugs/materials might also be excluded.
All these exceptions are reimbursed separately, and the exclusion list varies
widely between counties.
The cost of education, and of research and development (R&D) are not
covered by DRGs. The majority of R&D costs are covered by grants from the
Government. Most counties also offer local grants to their hospitals for R&D
activities. Other activities of general interest (such as accreditation, incentives
to hospital personnel, participation in social or other projects, and so on) are
covered by specially designated project budgets.
The most used model for reimbursement of all areas of the health care sector
is a mixed model, with global budgets, prospective payment systems, retrospec-
tive payment systems and payment for performance (P4P) in use within the
same system.
19.5.2 Calculation of DRG prices/cost weights
There is no ‘national price’ per DRG in Sweden, but there are reference cost
weights. Sums vary by county and by hospital, resulting in different prices
per DRG. The county councils are the payers and purchasers of hospital ser-
vices and the DRG prices are set according to the budget and regulation of the
council.
The National Board of Health and Welfare develops and publishes national
prospective weights for NordDRGs (both in- and outpatients) on an annual
basis. It is not mandatory to use the national weight sets; local weights are also
in use in two counties/regions. A goal for the future is that all counties will use
the same weights.
The average real cost in the cost database from last year is used to calculate
the national weights. The average cost is adjusted by the budgeted cost increases
and decreases for the next year and sometimes also by an estimated increase
in productivity (about 1 per cent). The most common method variant is that
hospitals use the same weight set within and between counties, but that the
prices per DRG are different for each hospital/council.
Items that are reimbursed on a ‘fee-for-service’ basis (rather than by cost
weights) represent very unusual and expensive treatments that cannot be
properly described in the DRG system, such as burns or special treatments
delivered at teaching hospitals.
352 Diagnosis-Related Groups in Europe
The national DRG weights are based on individual patient costs, as are the
outlier limits. In addition, also outlier limits are also calculated based on length
of stay by those hospitals that do not yet calculate individual patient-related
costs. Outliers are reimbursed outside the DRG system, with money from the
global budget of the county councils.
The cost trim-point is calculated using the quartile (Q) method, given 5 per
cent outliers by the following formula: Q3 + cost constant x (Q3 – Q1). The cost
constant is chosen to give 5 per cent outliers.
19.5.3 DRGs in actual hospital payment
The councils decide independently how to pay for health care. The reimburse-
ment model is set up in negotiation between purchaser and provider, within
the councils. The councils are free to use the DRGs or part of the DRG system,
or other models such as capitation or fi xed reimbursement.
Over 65 per cent of all discharges from Swedish acute somatic care are
reimbursed by NordDRGs to some extent. Outpatient care is reimbursed by the
NordDRG system to a more modest degree, and psychiatry and rehabilitation
even less so.
Various methods are used in different counties for keeping within the budget.
The county’s purchases and their volume are set, in negotiation with the hospi-
tals. In some counties a ceiling is in place for expenditures and the hospital
faces making a loss if it treats too many patients. In other counties, this could
be a gradual shift of responsibility, with a shared risk for when the negotiated
volume is exceeded.
19.5.4 Quality-related adjustments
The counties use quality indicators to describe performance and to some
extent for reimbursement purposes, in addition to the use of DRGs. Most
hospitals contribute to the national quality registers and also report to the
national waiting-time database.1 Sanctions are decided upon in each indi-
vidual county.
There is no quality assurance tool attached to the DRG-based health care
production in Sweden. Every county must decide themselves how to monitor
quality. Most counties use the national quality indicators, among other tools.
19.5.5 Main incentives for hospitals
Following the introduction of DRGs in Swedish health care, there has been a
signifi cant increase in coding diagnoses; in 1998, there were 1.8 diagnoses per
case and in 2009 there were 2.7 diagnoses coded per case in inpatient care
(Serdén at al., 2003). Quality has been improved to some extent in the national
registers, by educating the medical secretaries who input diagnosis and
procedure codes into the administrated systems.
Sweden: The history, development and current use of DRGs 353
19.6 New/innovative technologies
19.6.1 Steps required prior to introduction in hospitals
County councils and hospitals should take the initiative to develop health care
by introducing new technologies. The adoption of innovations into the DRG
system is decided in the updating process, which itself is published in reports.
At national level, the DRG system is administrated and developed by the
National Board of Health and Welfare. Within the updating process, innovations
are discussed at Nordic level in the NordDRG expert group(s) and fi nally in a
steering group in which decisions are made.
19.6.2 Payment mechanisms
During the fi rst two years, new and innovative technologies are funded either
separately, outside of the DRG system, or via the DRG system (through addi-
tional payments for high-cost outliers), depending on the regulations in each
county council. Most hospitals negotiate with the county regarding separate
prices for new technologies. When the innovations are adopted into the DRG
system, their funding is embraced within the DRGs.
19.6.3 Incentives for hospitals to use
new/innovative technologies
There is a delay in the process, from the decision to use new technologies until
those technologies are incorporated into the reimbursement system. The whole
process usually takes about two years.
19.7 Evaluation of the DRG system in Sweden
19.7.1 Offi cial evaluation(s)
County councils are responsible for the primary coding of and registration of
DRGs at the hospital. The councils are also responsible for evaluating the DRG
results. Unfortunately, only a few counties carry out audits to check the DRG
results. As already mentioned, Sweden does not have a signifi cant problem in
terms of up-coding, but a problem does exist relating to too few diagnoses
and procedures being coded (in some counties). In counties in which this is a
problem, it is characteristic for them to only use DRGs to a minor extent.
19.7.2 Authors’ assessment
The original goals that were set out before the introduction of the DRG sys-
tem have been reached: a rise in productivity and transparency in hospital
354 Diagnosis-Related Groups in Europe
activities, creating a common ‘language’ between professionals and administra-
tors, resulting in a fi nancing system focused on hospital activities instead of
organization, along with better describing performance, and a tool for bench-
marking and productivity calculation.
In the early 1990s, many physicians were opposed to DRGs. To start with
there was very little knowledge relating to DRGs, in terms of how they worked
and how the system could be utilized. This lack of knowledge was a problem.
Many actors had also unrealistic expectations of the benefi ts of the system; for
example, that it would save a lot of money and solve the issue of quality
monitoring. In addition, many politicians disliked activity-based funding. This
has now changed, and most are in favour of activity-based funding to some
degree.
As time passed, users learnt more and the expectations became more realistic,
along with the ability to see the good and bad aspects of the system. In addition,
the introduction and use of cost-outliers achieved better acceptance levels –
today, most professionals accept the system. Extending the system to encompass
both outpatient care and psychiatry has also been a positive development.
In psychiatry, DRGs were not accepted until 2005, when 26 new groups for
psychiatry were incorporated. There is just one county using DRGs for fi nancing
in psychiatric care; predominantly, it is used as a tool to describe performance
(in eight counties).
There were not many technical problems in implementing the grouper
system. The period of time needed for technical implementation of the system
was different in each hospital. The cost of developing and implementing DRGs
(for the Government, hospitals, taxpayers, and so on) remained fairly low and
did not exceed expected levels.
The DRG impact
After introducing DRGs in Sweden, there was an increase in productivity and
service delivery increased. At the start, the Stockholm County Council had a
problem with the use of DRGs to control total costs; when the system was
introduced, the increase in volume resulted in the costs exceeding the global
budget. Within a few years, this could be controlled. In general, the hospitals
that are using DRGs have better control (with some exceptions) over their
activities and have a lower cost per DRG point than hospitals that do not use
the system.
In terms of the impact on the patient, the introduction of the DRG system
has shortened waiting times, due to the increase in productivity (more services
carried out) (Charpentier & Samuelson, 1998).
Sweden has had a major reduction in length of stay since the early 2000s – for
a number of reasons, but partly because of the use of DRGs. Counties that use
DRGs tend to have shorter lengths of stay than others. Whether the present
length of stay is too short or not is a matter for debate, but most will agree that
it is good for the patients if the length of stay is short. A short length of stay
shows that the process works and that the patient is well informed. The
argument against short lengths of stay is mainly that elderly people are sent
back to their homes too early in the health care process, but there is no evidence
Sweden: The history, development and current use of DRGs 355
of an increased level of readmission when introducing DRG-based reimbursement
systems.
The introduction of the DRG system has not had a direct impact on the way
inpatient and outpatient care is organized on a daily basis, although it may
have had an indirect effect. The DRG system has exerted no infl uence on
hospital organization as a whole.
DRGs are not a ‘miracle cure’
The most important experience gained from working with DRGs for reimburse-
ment is that the introduction of payment systems does not solve all the prob-
lems that health care systems are facing (Lindqvist, 2008). When DRGs for
reimbursement were introduced in some counties in the early 1990s, there was
a strong notion that this was a ‘miracle cure’. The few that were opposed to the
transformation, on the other hand, saw the change as the end of the Swedish
health care model as we know it. Both of these expectations have been proven
wrong. Other political decisions and changes, economic conditions and the
general public’s expectations have had more of an impact on health care than
the introduction of DRG-based payment systems.
One of the most signifi cant problems with using DRGs for reimbursement, at
least from a Swedish perspective, has been the mechanisms of cost control. In
the case of Stockholm County, productivity rose quite dramatically during the
years following its introduction, but the increased production also led to higher
total expenditures. To secure cost control, budget ceilings were introduced –
which led to a reduction in the increased rate of productivity. Finding a balance
between the desire to increase productivity and the need to control cost (given
limited resources) has been the biggest challenge in the introduction of payment
systems.
A casemix reimbursement system improves productivity
It is quite simple – when a funding system based on recorded activity is intro-
duced, the activity increases – or, to be more precise, the recorded activity in-
creases (Lindqvist, 2008). The fi rst problem is to determine whether the increase
is an effect of better or changed recording, or of an actual increase in volume.
The experience in Sweden, especially in the outpatient care setting (where there
was no tradition of good recording), is that the initial increase seen following
the introduction of DRG-based payment systems was to a great extent due to
changes in recording. However, the number of inpatient admissions also in-
creased, and this effect is better documented, since the medical recording of
admissions was of good quality in Sweden.
Good information systems and good data are crucial
When shifting to a system in which clinical data are the basis for reimbursement,
the increasing need for data and information system is of great importance
(Lindqvist, 2008). In Sweden, the tradition of collecting clinical data and the
356 Diagnosis-Related Groups in Europe
use of a personal identifi cation number have been benefi cial, but a new
information system for follow-up and analysis needed to be developed. This
development was regrettably slow in terms of tools for analysing production at
hospital and department levels. This was quite ironic, considering that the
responsibility for the hospitals’ economy was to a large extent moved to the
department heads.
The data quality must also be considered. In spite of a long tradition of data
collection, the quality of data was poorly analysed in Sweden. This necessitated
efforts to improve the quality of basic clinical data. There is a trend towards
‘going back to the basic data quality’. More efforts are being directed towards
correct registration and regular revisions of coding. One key question concerns
access to and quality of data. Working with prospective payment systems based
on DRGs means dealing with core health care data, and the performance of the
systems is heavily dependent on the quality of the basic data. To implement the
systems, access to individual patient data is required – to both reimburse and
assess performance more accurately.
19.8 Outlook: Future developments and reform
19.8.1 Trends in hospital service
(or general health care) delivery
The National Board of Health and Welfare has received a government commis-
sion to improve the reimbursement system in primary care; specifi cally, a sys-
tem promoting health care activities and results. Uniform classifi cation of
diagnoses and procedures in primary care is necessary when creating high-
quality squared systems to describe performance, which form the basis of reim-
bursement and high-quality follow-up in primary care. Uniform classifi cation
systems are also necessary to compare health care within primary care. It is
benefi cial if the classifi cation is comparable to other settings.
A Swedish classifi cation system for diagnoses exists for Swedish primary care,
but it is not generally used and, when it is, it is not used properly. Since there is
no classifi cation system for procedures in primary care, the need exists to develop
such a system or to improve existing classifi cation relating to procedures. There
is much to be done before a new secondary patient classifi cation system in
primary care can be established.
19.8.2 Trends in DRG application/coverage
An extensive amount of work in exchanging and improving the system has
taken place during the 2000s. Sweden has just fi nished the development of a
new grouper, which will be available in Sweden from 2012. The purpose of this
grouping system is to divide DRGs into three severity sub-group levels. The role
model is the grouper of the Centers for Medicare & Medicaid Services (CMS) in
the United States (3M, 2011).
Sweden: The history, development and current use of DRGs 357
19.9 Notes
1 More information is available at the relevant web site of the Swedish Association of
Local Authorities and Regions (www.vantetider.se, accessed 1 August 2011).
2 More details available at the National Board of Health and Welfare web site (http://
www.socialstyrelsen.se/klassifi ceringochkoder/norddrg/logikenidrg, accessed 1 August
2011).
19.10 References
3M (2011). Defi nitions Manuals. St Paul, MN: 3M Health Information Systems (http://
solutions.3m.com/wps/portal/3M/en_US/3M_Health_Information_Systems/HIS/
Products/Defi nition_Manuals/, accessed 1 August 2011),
Charpentier, C., Samuelson, L.A. (1998). Effekter av en sjukvårdsreform – En analys av
Stockholmsmodellen [The Effects of the Stockholm Model]. Stockholm: Nerenius &
Santérus Förlag.
Forsberg, L., Rydh, H., Jacobson, A., Nygvist, K., Heurgren, M. (2009). Kvalitet och innehåll
i patientregistret. Discharge from inpatient treatment 1964–2007 and visits to specialist
outpatient care (excluding primary care visits) 1997–2007 [Quality and Content of the
Patient Register]. Stockholm: National Board of Health and Welfare.
Heurgren, M., Nilsson, H., Erlö, C., Sjöli, P. (2003). What does the individual patient cost?
CPP – the Cost Per Patient method – is the answer. Lakartidningen, 100(42):3312–15.
Lindqvist, R. (2008). From naïve hope to realistic conviction: DRGs in Sweden, in J.R.
Kimberly, G. de Pouvourville, T. D’Aunno, eds. The Globalization of Managerial
Innovation in Health Care. Cambridge: Cambridge University Press.
Ludvigsson, J.F., Andersson, E., Ekbom, E. et al. (2011). External review and validation of
the Swedish national inpatient register. BMC Public Health, 11(450):1–16.
National Board of Health and Welfare (2006). Diagnosgranskningar utförda i Sverige 1997–
2005 samt råd inför granskning [Audits Performed in Swedish Health Care 1997–2005].
Stockholm: National Board of Health and Welfare (http://www.socialstyrelsen.se/
publikationer2006/2006-131-30, accessed 10 July 2011).
National Board of Health and Welfare (2010). Quality and Effi ciency in Swedish Health Care –
Regional Comparisons 2009 [English version]. Stockholm: Swedish Association of Local
Authorities and Regions (SALAR) (http://www.socialstyrelsen.se/Lists/Artikelkatalog/
Attachments/18023/2010-4-37 , accessed 10 July 2011).
Nordic Casemix Centre (2011). Swedish Version. Helsinki: Nordic Casemix Centre
(http://www.nordcase.org/eng/norddrg_manuals/versions/swedish, accessed 1 August
2011).
SALAR (2009). Nationella KPP-principer, version 2. [National Case Costing Principles, Version
2]. Stockholm: Swedish Association of Local Authorities and Regions.
SALAR (2010). Statistik om hälso- och sjukvård samt regional utveckling 2009 [Health Care
Statistics and Regional Development in 2009]. Stockholm: Swedish Association of Local
Authorities and Regions.
SALAR (2011). Swedish National Case Costing Database. Stockholm: Swedish Association of
Local Authorities and Regions (www.skl.se/vi_arbetar _med/statistik/sjukvård/KPP,
accessed 1 August 2011).
Serdén, L., Lindqvist, R., Rosén, M. (2003). Have DRG-based prospective payment systems
infl uenced the number of secondary diagnoses in health care administrative data?
Health Policy, 65(2):101–7.
chapter t w e n t y
Poland: The Jednorodne
Grupy Pacjentów – Polish
experiences with DRGs
Katarzyna Czach,
Katarzyna Klonowska,
Maria Świderek and
Katarzyna Wiktorzak
20.1 Hospital services and the role of DRGs in Poland
20.1.1 The Polish health care system
Poland has a mixed system of public and private health care fi nancing. Total
health expenditure amounts to about €667 per capita per year, which corre-
sponds to 6.6 per cent of gross domestic product (GDP) (European Commis-
sion, 2009). Public expenditure accounts for roughly 72 per cent of total health
expenditure and is mostly based on mandatory social health insurance contri-
butions. Private expenditure accounted for about 24 per cent of total health
expenditure in 2008, and predominantly took the form of out-of-pocket pay-
ments and co-payments from members of social health insurance schemes, for
example for food and accommodation at rehabilitative care facilities, or for a
certain percentage of the costs of medicines and diagnostic examinations
(Kuszewski & Gericke, 2005; European Commission, 2009).
The three most important actors in the system are: (1) the Ministry of Health,
(2) the territorial governments, and (3) the National Health Fund (NFZ)
(Kuszewski & Gericke, 2005). The Ministry of Health is responsible for policy-
making and regulation. As such, it designs national health policies, fi nances
major capital investments and oversees medical science and medical education.
The territorial governments (local, county and municipality levels) manage the
majority of public hospitals; they develop strategies and health plans for their
populations, as well as engaging in health promotion activities. The NFZ, which
360 Diagnosis-Related Groups in Europe
was established in 2003 is the purchaser of health care services for all members
of the social health insurance system.
About 98 per cent of the population are members of a social health insurance
scheme and contributions currently amount to 9 per cent of most individuals’
taxable income. Contributions are paid either to the Social Insurance Institution
(ZUS) or to the Agricultural Social Insurance Fund (KRUS), which forward
collected contributions to the NFZ. Health care benefi ts for uninsured people,
the unemployed population and individuals requiring complex and expensive
medical care are fi nanced directly from tax-funded state budgets or the budgets
of local governments. State budgets also contribute to capital expenditures of
health care providers, while recurrent costs are paid from health insurance
contributions managed by the NFZ (Ministry of Health, 2008).
The NFZ is composed of 16 regional branch offi ces, plus one central offi ce.
It is supervised by the NFZ Council, consisting of nine members appointed by
the Prime Minister for a fi ve-year term. The NFZ pools and manages all revenue
received through contributions from social health insurance members. As the
purchaser of health care, the NFZ operates within a budget that is fi xed for a
given year. A ‘Universal Catalogue of Services’ is defi ned at the national level
and the regional branches of the NFZ negotiate contracts with providers
competing for contracts in the form of a competitive bid. There are strict
regulations prohibiting the NFZ from engaging in the direct provision of health
care services and from undertaking income-generating activities.
20.1.2 Hospital services in Poland
Historically, there has been a relatively strict separation between outpatient
care and inpatient care in Poland (Kuszewski & Gericke, 2005). Outpatient care
(both delivered by general practitioners (GPs) and specialists) is mostly provided
in private medical practices or in independent health care institutions. Hospital
care is provided either in general (county) hospitals, specialized (province
(voivodship))-level hospitals, or highly specialized university hospitals.
Throughout most of the 1990s, hospitals in Poland had the status of budgetary
units that received funds from the Ministry of Health or from territorial
governments. This changed fundamentally with the Law on Universal Health
Insurance that came into effect in 1999. The law introduced a split between the
purchasers and providers of health care and all public hospitals were obligated
to change their status into independent institutions that must generate revenue
through health service delivery. As a result of the reform, hospitals can incur
defi cits and make profi ts. However, most hospitals are still public and are owned
by territorial governments (Kozierkiewicz, 2008).
At the end of 2007, there were 578 public hospitals and 170 non-public
hospitals in Poland (Table 20.1). The share of non-public hospitals increased
from 4 per cent in the year 2000 to almost 23 per cent in 2007. Changes in the
total number of hospitals have been diffi cult to identify. However, the number
of hospital beds has continued to decrease since the year 2000. In absolute
numbers, there were 8 per cent fewer hospital beds in 2007 than in 2000, a
decrease that is even more pronounced when looking at the number of hospital
Poland: The Jednorodne Grupy Pacjentów – Polish experiences with DRGs 361
beds per 100 000 inhabitants. That said, some departments have seen increases
in the number of beds over the same period of time (for example, the number
of beds in cardiology, oncology, psychiatry, and intensive care units (ICUs)
increased by more than 10 per cent (Świderek, 2009)). Although the number of
hospital beds decreased from 2000 to 2007, the total number of patients treated
in hospitals increased by 14 per cent.
In recent years, the number of day-care patients in hospitals has increased
dramatically, and during the course of 2007, hospitals treated more than 1 mil-
lion patients as day cases.
A typical treatment episode starts when a patient visits a primary health care
physician who issues a referral to a specialist physician or a hospital. There is a
group of specialist physicians (dermatologists, oculists, gynaecologists, oncolo-
gists, psychiatrists), for which no referral is required. In an emergency, a patient
is admitted directly to a hospital. Treatment is completed when the patient is
discharged from the hospital. If further treatment is required, the patient is
referred to another hospital, a primary health care physician or a specialist phy-
sician (see Figure 20.1).
Since the introduction of social health insurance in 1999, hospitals must
raise the majority of their revenues through the provision of health services
to social health insurance members. In 2007, this is thought to have accounted
for more than 60 per cent of total revenues for hospitals. In order to im-
prove purchasing for social health insurance members, the NFZ introduced a
Table 20.1 Number of hospitals, hospital beds and patients in Poland, 2000, 2005
and 2007
Year 2000 2005 2007 Change in %
Number
(%)
Number
(%)
Number
(%)
2000 to
2007
Hospitals by ownership
Total 716 781 748 4
Public 686 611 578 –16
(96) (78) (77)
Non-public 30 170 170 467
(4) (22) (23)
Number of hospital beds
Total 190 952 179 493 175 023 –8
Beds per 100 000 inhabitants 515 469 459 –11
Number of patients treated
in hospitals
Number of inpatients (in
thousands) 6 007 6 739 6 850 14
Number of day cases in hospital
wards (in thousands)a – 895 1 014 –
Sources: Central Statistical Offi ce, 2009b; aCentral Statistical Offi ce, 2007 (p. 124), 2008
(p. 135), 2009a (p. 145).
F
ig
u
re
2
0
.1
T
y
p
ic
al
e
p
is
o
d
e
o
f
ca
re
a
cr
o
ss
s
ec
to
ra
l
b
o
rd
er
s
So
u
rc
e:
N
FZ
,
2
0
1
0
.
Poland: The Jednorodne Grupy Pacjentów – Polish experiences with DRGs 363
diagnosis-related group (DRG)-based hospital payment system in 2009, which
means that the majority of hospital revenues are now related to DRG-based
hospital payment. However, in addition to these revenues, hospitals still receive
fi nancial contributions from national and territorial governments for capital
investments, for teaching and research, and for highly specialized services. In
addition, hospitals also receive funds from private sources (benefactors).
20.1.3 Purpose of the DRG system
Prior to the introduction of DRG-based hospital payment in Poland, hospital
payment was based on a ‘Catalogue of Health Care Products’ that was annually
updated by the NFZ (Kozierkiewicz et al., 2006). Similar to DRGs, health care
products were defi ned through specifi c diagnoses or procedures, and hospitals
received a fl at payment per admission based on the point value of a given
product. However, in contrast to DRGs, the system was not based on systematic
coding of all the diagnoses and procedures of the patients. Consequently,
information was available only for the specifi c services defi ned by the Catalogue
of Health Care Products. Since the defi nitions of products changed every year
and the number of items in the catalogue continued to increase, the system
lacked transparency. In addition, because of the constantly changing product
defi nitions, hospital performance could not be assessed across time. Furthermore,
the NFZ used its position of power to negotiate hospital payment rates that
were often below the costs of service provision (Kozierkiewicz et al., 2006). This
led to a deterioration of service quality and compromised access to hospital care
through the emergence of waiting lists.
The main goals of introducing DRG-based hospital payment in Poland were:
(1) to improve resource allocation to hospitals, and (2) to increase transparency
of service provision.
DRG-based hospital payment was considered to be better able to provide
adequate (fair) reimbursement to hospitals for delivered services, which was
thought to increase the availability of services and to improve quality. Further-
more, DRG-based hospital payment was assumed to promote cost-accounting
practices within hospitals, which would enable effective auditing of provider
accounts, and would – ultimately – restrict unjustifi ed increases in health care
costs.
Transparency of hospital services was expected to improve because DRG-
based hospital payment requires the collection of detailed data on every
patient admitted to hospital (including primary diagnosis, secondary diag-
noses, procedures, length of stay, gender and age). Given the need for data col-
lection, hospitals would be encouraged to develop information technology
(IT) systems that could facilitate the fl ow of information between the regional
NFZ branch offi ces and service providers. Consequently, the NFZ would have
better data regarding patients treated by contracted providers, and provider
performance.
Last but not least, the international success of DRG-based hospital payment
systems infl uenced the decision to introduce a similar system in Poland.
364 Diagnosis-Related Groups in Europe
20.2 Development and updates of the DRG system in Poland
20.2.1 The current DRG system at a glance:
the Jednorodne Grupy Pacjentów
In July 2008, a national DRG system was introduced in Poland, entitled Jedno-
rodne Grupy Pacjentów (JGP), which can be translated as ‘homogeneous groups
of patients’. The British Healthcare Resource Groups (HRGs) (Version 3.5) served
as the starting point for the JGP system, resulting in similarities between the
two systems. Each JGP represents a distinct group of patients with similar char-
acteristics (for example, diagnoses, procedures, patient age) and similar
resource-consumption patterns or costs. Table 20.2 summarizes some of the main
facts regarding the fi rst national DRG system in Poland (Schreyögg et al., 2006).
Since July 2008, all hospitals (public and private) that have contracts with the
NFZ must classify their patients using JGPs in order to receive DRG-based
hospital payment for services they deliver. The system covers only hospital
inpatient services and (similar to the British system of HRGs) differentiates
between emergency admissions, planned admissions and day-care treatment
episodes. Rehabilitation is only partly included, and psychiatry is not included
in the JGP system. Rehabilitative care is mostly fi nanced using fee-for-service
payments; psychiatric services are paid for by means of per diem payments.
However, plans are being developed to extend DRG-based hospital payment to
include these areas of care (see section 20.8).
The NFZ enters into contracts with hospitals, specifying which JGPs hospitals
are permitted to provide. In order to receive payments under the JGP system,
hospitals must group each patient into a specifi c JGP and report such data to
the regional branch of the NFZ. Each JGP has a predetermined score between
5 points (for example, ‘minor procedure on eye’s protective apparatus’) and
Table 20.2 Main facts relating to the fi rst national DRG version in Poland: the
JGP system
Date of introduction Patient classifi cation: July 2008
Hospital payment: January 2009 (voluntarily since July
2008)
(Main) purpose DRG-based hospital payment
DRG system Homogeneous Groups of Patients (JGP) (based on British
Healthcare Resource Groups (HRGs) Version 3.5)
Data used for development Expert consultations, data on length of stay
Number of DRGs (as of 2010) 518
Applied to All hospitals (public and private) that have contracts
with the NFZ
Range of included services All hospital inpatients and day cases except psychiatric
and rehabilitative care. Since October 2010, also
including neurological and cardiological rehabilitation
Range of included cost
categories
Capital and recurrent costs, excluding major
investments
Update of JGP Scheduled for 2011
Source: Compiled by the authors based on grey literature from the NFZ.
Poland: The Jednorodne Grupy Pacjentów – Polish experiences with DRGs 365
4706 points (for example, ‘transplantation of hematopoietic cells’), specifi ed in
the JGP catalogue. Depending on the score, hospitals receive a fi xed sum of
money, which is the same for all hospitals contracted by the NFZ.
Four documents are essential for the JGP system: (1) the JGP catalogue,
which contains a full list of all JGPs, their scores and some further speci-
fi cations; (2) the JGP characteristics fi le, which specifi es the variables that
defi ne each JGP; (3) the grouper algorithm, which describes how to develop
a grouper software tool by outlining all steps necessary in order to select the
correct JGP; and (4) a parameterization fi le that is a functional form of the
grouper algorithm (Gilewski, 2010).
The JGP catalogue is divided into 16 sections (or major diagnostic categories,
MDCs) that correspond to anatomic or physiological systems of the body or to
a specifi c clinical specialty (see Table 20.3). The 16 sections contain a total of
518 JGPs. Within each section, JGPs are arranged from highest to lowest scores.
There are a total of 283 procedural JGPs and 235 medical JGPs.
20.2.2 Development of the JGP system
When the fi rst national JGP system was introduced in July 2008, it was the
result of several years of preparation and experimentation with different DRG
systems in Poland: at the end of the 1990s, regional sickness funds had used
DRGs for hospital payment in the Łódzkie, Dolnoślą skie and Podkarpackie
voivodships for several years (see Figure 20.2). After 2003, when sickness funds
had been replaced by the NFZ, interest in DRGs remained strong.
Table 20.3 Sections of the JGP system
Section Section name Number of DRG groups
A Diseases of the nervous system 36
B Eye diseases 31
C Diseases of the face, oral cavity, throat, larynx,
nose and ears
27
D Diseases of the respiratory system 29
E Heart diseases 57
F Diseases of the digestive system 39
G Diseases of the liver, bile ducts, pancreas and
spleen
24
H Diseases of the musculoskeletal system 47
J Diseases of breasts and skin, and burns 30
K Diseases of the hormonal system 27
L Diseases of the genitourinary system 45
M Female genital diseases 22
N Obstetrics and care of neonates 22
P Paediatrics 27
Q Vascular diseases 30
S Diseases of blood-forming organs, poisoning and
infectious diseases
25
Total 518
Source: Regulation of the President of the NFZ No. 69/2009.5
F
ig
u
re
2
0
.2
T
im
el
in
e:
i
n
tr
o
d
u
ct
io
n
o
f
th
e
n
at
io
n
al
J
G
P
s
y
st
em
i
n
P
o
la
n
d
a
n
d
p
ri
o
r
su
b
-n
at
io
n
al
u
se
s
o
f
D
R
G
s
So
u
rc
e:
N
FZ
,
2
0
1
0
.
Poland: The Jednorodne Grupy Pacjentów – Polish experiences with DRGs 367
Between 2004 and 2006, a European Union (EU)-funded research project on
hospital costs in Poland (VITAPOL) contributed to the establishment of a close
working relationship between British experts and Polish NFZ managers. The
project generated interest in British HRGs and, in early 2007, the fi nal decision
was made by the President of the NFZ to introduce DRG-based hospital payment
in Poland on the basis of British HRGs.
The process of developing the Polish JGP hospital payment system stretched
over a period of about two years. In a fi rst step, the NFZ adapted British HRGs
to the Polish context. In order to do so, the British Classifi cation of Interventions
and Procedures (OPCS) was mapped to ICD-9-PL codes (WHO International
Classifi cation of Diseases 9th revision, Polish Clinical Modifi cation). In a second
step, hospital services from the previously existing Catalogue of Health Care
Products were matched to HRGs. Since the number of hospital products in the
old system was greater than the number of ‘homogeneous groups of patients’
(there were 1500 hospital products in 2005 but only 518 JGPs), the new groups
of patients sometimes contained different types of hospital services. Therefore,
in a third step, the (old) prices for the different types of services within one
JGP were assessed, and – if possible – homogeneity of prices was increased by
reassigning cases to different JGPs.
Once the draft version of the patient classifi cation system had been prepared,
a score (or price) per JGP was calculated by the NFZ: the price of each of the old
hospital products within a JGP was weighted by the relative frequency of the
service within the JGP, in order to calculate a (weighted) average price for the
services grouped into one JGP. This calculated score per JGP was then compared
with price ratios in the British HRG system. Furthermore, length-of-stay
thresholds were determined for certain JGPs in order to defi ne outlier cases, for
which hospitals would receive supplementary payments.
Finally, the fi nancial impact of the new DRG-based hospital payment
system was estimated, based on data relating to the payment rate per JGP
and the anticipated number of patients per JGP. The estimated expenditures
(for the NFZ and regional branches of the NFZ) and estimated revenues (for
selected providers) were compared to expenditures and revenues under the
old system, in order to assess the fi nancial impact of the hospital payment
reform.
A draft version of the JGP system was published in August 2007 and discussed
in a process of broad consultations with medical professionals and hospital
managers. During numerous meetings between the NFZ and national consul-
tants (recognized experts in a specifi c medical specialty, appointed by the
Minister of Health), a number of modifi cations were agreed upon that were
introduced into the President of the NFZ’s draft ordinance. In March 2008, an
early JGP version was tested as a pilot project in 44 selected hospitals. On
the basis of the information gathered, the NFZ further modifi ed the ICD-9-PL
classifi cation of procedures and the defi nitions and payment rates of JGPs
(Kozierkiewicz, 2009).
As part of a broader process of extending the JGP system to other areas of care
(see section 20.8), the NFZ started to introduce JPGs for neurological and car-
diological rehabilitation patients treated at hospitals in October 2010. Rehabili-
tation patients are grouped on the basis of their primary diagnosis, medical
368 Diagnosis-Related Groups in Europe
procedures, secondary diagnoses and the assessment of each patient’s health
status according to the Barthel Activity of Daily Living scale, the American
Spinal Injury Association scale and the Gross Motor Function Classifi cation
System. A total of 14 DRGs were created within neurological rehabilitation and
three within cardiological rehabilitation.
During most of the process of introducing JGPs, Poland experienced a period
of sustained high economic growth, which resulted in increased revenue for the
NFZ. As a result, the NFZ had suffi cient funds to increase total expenditure for
hospital care and to raise payment rates for previously underfunded services.
These additional revenues were an important positive infl uence in the process
of introducing DRG-based hospital payment in Poland, since they helped to
assure support from providers for the new payment system.
20.2.3 Sources of information used for developing
and updating the JGP system
As already described, the JGP system is mostly based on imported British HRGs.
However, Polish data were used to assess the adequacy of HRGs in the Polish
context and to calculate JGP scores: fi rst, information about the prices of hospi-
tal products under the previous payment system was used to assess the homo-
geneity of JGPs in Poland. Second, national hospital statistics from 2006/2007
were used to estimate the relative frequency of services bundled within each
JGP, in order in turn to estimate payment rates. Third, data on hospital patients’
lengths of stay were used to determine length-of-stay thresholds that delimit
the number of days for which hospital payments are calculated on the basis of
JGPs. Furthermore, information about innovative medical technologies is used
to update the system. Unfortunately, data on costs of hospital services are not
systematically collected by the NFZ and are not used to develop the JGP system.
However, cost data from specifi c hospitals were used (albeit in a non-systematic
way) to inform decisions during the process of setting the payment rate(s) un-
der the old hospital payment system (Kozierkiewicz et al., 2006), and such data
are still being used under the new system.
20.2.4 Regularity and methods of system updates
The JGP system was introduced by the NFZ in mid-2008. Since then, a number
of minor updates have been introduced into the system (mostly in the second
half of 2008). In most cases, these were motivated by suggestions from medical
consultants or health care providers, but also by economic analyses conducted
by the NFZ. In some cases, these suggestions have resulted in the creation of
new JGPs. However, in order to introduce new JGPs, it must be demonstrated
that the proposed group would comprise more than 300 cases, or that total
payments for patients in the JGP would amount to more than PLN (Polish
Złoty) 1.5 million per year (about €370 000).
At present, JGP scores are updated annually and are the same for all health
care providers in Poland. In the years to follow, the principle of universal
Poland: The Jednorodne Grupy Pacjentów – Polish experiences with DRGs 369
applicability of scores may be subject to change. The point value, used to
convert scores per JGP into PLN depends on the resources available in the NFZ’s
annual fi nancial plan.
Decisions relating to the introduction of new medical technologies into the
Polish hospital sector are made by the Ministry of Health on the basis of
recommendations from the Health Technology Assessment Agency (Agencja
Oceny Technologii Medycznych, AOTM) (see section 20.6).
Major updates to the JGP system are planned for the year 2011. Among other
things, the NFZ plans to introduce a new process of regular updates to the
patient classifi cation system and to the payment rates (see section 20.8).
20.3 The current patient classifi cation system
20.3.1 Information used to classify patients
Each JGP is defi ned on the basis of data available from the common hospital
discharge dataset. This contains information about the diagnoses of patients
(primary diagnosis and secondary diagnoses coded using ICD-10; procedures
coded using ICD-9-PL; demographic variables (age and gender); reason for
hospital admission/discharge; and length of stay). After hospital discharge, one
JGP is selected for the entire hospital stay by a specialized software program
called a ‘grouper’, which uses information about diagnoses, procedures, age,
type of admission, type of discharge and length of stay to classify patients into
the appropriate group of patients.
20.3.2 Classifi cation algorithm
The grouping algorithm is illustrated in Figure 20.3. In a fi rst step, the grouper
checks whether any services were provided that are reimbursed without being
assigned to a specifi c JGP. For example, very costly procedures such as trans-
plantations, treatment of drug-resistant epilepsy, and chemotherapy are ex-
cluded from the further grouping process (see Figure 20.4 for reimbursement
components besides those for JGPs). The next step of the grouping algorithm is
to check whether any signifi cant procedures were performed. If this is the case,
the grouper determines for each procedure a rank between 0 and 6. The highest
ranked procedure is then indicated as the dominant procedure for the hos-
pital stay. However, unlike in the British HRG system, the grouper does not auto-
matically select the dominant procedure. Instead, health care providers can
manually select the procedure that was the most important during the hospital
stay in question.
Subsequently, this procedure determines the section of the JGP system. If the
rank of the procedure is > 2, which is the case for most operating room (OR)
procedures, the JGP is determined directly (94 basic ‘surgical’ procedural JGPs).
If the procedure rank is ≤2, the grouper checks whether additional conditions
concerning secondary diagnoses, secondary procedures, age, gender, and so on
are met, in order to determine the JGP (191 procedural JGPs).
370 Diagnosis-Related Groups in Europe
Cases without any procedures – or those cases for which a JGP could not be
determined by the grouping algorithm described so far – are assigned to the
relevant section of the JGP system on the basis of the primary diagnoses of the
hospital stay. Subsequently, the primary diagnosis determines a base group.
Depending on the primary diagnosis, the JGP can be determined directly (for
108 medical JGPs), or additional conditions may have to be met (for 125 medi-
cal JGPs).
Whether a specifi c secondary diagnosis is considered a to be a complication
or co-morbidity (CC) in the grouping process depends on the section of the JGP
system concerned. For every section, a list of ICD-10 codes exists that defi nes
Figure 20.3 JGP grouping algorithm
Source: Compiled by the authors based on grey literature from the NFZ.
Poland: The Jednorodne Grupy Pacjentów – Polish experiences with DRGs 371
relevant CCs for the section. In addition, a global list of secondary diagnoses
exists, which are considered CCs in all sections.
Each JGP is characterized by a three-digit code. The fi rst digit is a letter indi-
cating the section of the JGP system, for example ‘F’ is used for diseases of the
digestive system (see Table 20.4). The second and third digits are numbers that
specify the JGP group, for example ‘F83’ for appendectomy in uncomplicated
cases of appendicitis. In general, lower numbers indicate more complex groups
of patients, while higher numbers indicate less-complicated (less-costly) groups
of patients. However, there are several important differences between the JGP
system and the British HRG system. The most important differences are sum-
marized in Table 20.4.
20.3.3 Data quality and plausibility checks
Hospitals submit all data relevant for reimbursement to the regional branch
offi ces of the NFZ. The NFZ withholds payments in the event that the required
information is not complete. Data quality and plausibility are regularly verifi ed:
fi rst, health care providers verify their data before sending them to regional
branch offi ces of the NFZ, in order to avoid external auditing. Second, the NFZ
verifi es that data are complete and checks for plausibility of combinations of
diagnoses and procedures. Finally, the NFZ carries out coordinated monitoring
by sending out review teams to hospitals, to check patients’ medical documen-
tation. If any irregularities are detected, the health care provider must correct its
reports and fi nancial penalties are applied.
In October 2009, the Section for Verifi cation and Validation Standards of
the NFZ launched a central process of validation and verifi cation of reported
data. The aim is to harmonize and automate most processes of validation and
verifi cation and to contribute to improved data quality. However, at present,
coding problems detected during the process of validation and verifi cation
indicate that hospitals still need to improve medical monitoring and controlling
in order to improve data quality.
Table 20.4 Differences between British HRGs and Polish JGPs
British HRGs (Version 3.5) Polish JGPs
Procedure classifi cation
system
OPCS ICD-9-PL
Selection of dominant
procedure
Grouper selects highest
ranked procedure
Provider selects most relevant
procedure of hospital stay
Poly-trauma cases Included in ‘Pre-MDC’-
like group
No specifi c groups. JGP score
is infl ated depending on the
number of affected organs
Paediatric cases Specifi c section only for
neonates
Separate section for
paediatrics
Number of chapters/sections 19 16
Number of groups 610 518
Source: Compiled by the authors based on grey literature from the NFZ.
372 Diagnosis-Related Groups in Europe
20.3.4 Incentives for up-coding or wrong-coding
Since hospital payment is determined to a large extent by patients’ JGPs,
hospitals have strong incentives to ‘optimize’ their coding practices in order to
achieve higher payments. However, during data quality checks and controls of
patient records at hospitals, the NFZ regularly checks for up- and wrong-coding.
If fraudulent coding practices are detected, hospitals may be punished by means
of penalties (high fi nes), or even termination of the contract.
20.4 Cost accounting within hospitals
20.4.1 Regulation
Health care facilities are obliged to produce cost-accounting statements
according to the rules set out in an Ordinance of the Minister of Health and
Social Policy.1 However, this document does not specify in detail how health
care providers should carry out their cost accounting. Consequently, signifi cant
discrepancies exist in the methods of calculating costs between particular
service providers.
Given the lack of consistent cost-accounting data, the JGP system is currently
not directly related to the costs of hospital services. The only available informa-
tion on costs of hospital services in Poland is selective. Some hospitals collect
cost information on particular medical care episodes and voluntarily submit it
to the NFZ. The NFZ may use this information in the process of setting payment
rate(s), but does not use it in a systematic way.
20.4.2 Main characteristics of the cost-accounting system
Recently, the NFZ has launched an initiative to establish a cost database. The
idea is that about 15 hospitals will collect data on the costs of selected treat-
ment episodes included in the JGP system. The initiative will allow hospitals
to know more about their cost structures in comparison to other hospitals. Hos-
pitals that want to participate in the project must fulfi l certain requirements.
They must:
• comply with the aforementioned Ordinance of the Minister of Health and
Social Policy (Dz.U.98.164.1194);
• allocate overhead costs to direct cost centres through a step-down cost-
accounting approach;
• estimate costs of sub-ward cost centres (ORs, doctors’ rooms for the provision
of services for numerous wards, diagnostic laboratories, and so on);
• estimate total costs at direct cost centres; that is, costs at medical departments
engaged in the provision of services to patients;
• collect patient-level data on certain consumed resources (drugs, high-cost
materials, diagnostic tests, and so on);
• disaggregate costs according to defi ned cost groups: labour, drugs, diagnostic
tests, medical materials, and overheads.
Poland: The Jednorodne Grupy Pacjentów – Polish experiences with DRGs 373
As a result, hospitals should be able to provide information about (1) per
diem costs (both ‘hotel’ costs and care costs), (2) average costs per patient, (3)
costs of medical procedures, and (4) costs of services provided by hospital wards
for other wards or outside the hospital. In order to standardize the methods of
gathering the data from selected service providers, the NFZ has prepared a web-
based application which will facilitate the sharing of information.
20.5 JGPs for hospital fi nancing
20.5.1 Role of JGP-based hospital payment
in the overall fi nancing of hospitals
All hospitals in Poland that have contracts with the NFZ are fi nanced through
the JGP system. The same conditions apply to all hospitals, irrespective of own-
ership status, hospital type, or regional differences. The system applies to all
patients, including day cases, except for psychiatric and most rehabilitative care
patients. Payments under the JGP system are supposed to cover the full costs
(capital, personnel, overheads, and so on, except costs of major investments)
of all services provided by hospitals between admission and discharge of the
patient. However, hospitals may receive additional funds for investments from
regional governments, for teaching from the Ministry of Science and Higher
Education, and from patients for add-on services, such as for a sole-occupancy
treatment room.
No information is available on the budget structure of hospitals. However, it
is assumed that public funds contracted from the NFZ constitute more than
60 per cent of hospitals’ total budgets.
20.5.2 Calculation of JGP scores and trimming
As described in subsection 20.2.2, calculation of JGP scores was based mainly
on information about prices of services from the old catalogue of hospital prod-
ucts and the assumed relative frequency of these services within one JGP
(in the years 2006–2007). In addition, the ratio of JGP scores was compared
with the price ratio of comparable HRGs, while taking into account particulari-
ties of the Polish health care system. In particular, the costs of intensive care
treatment were included in the most complex groups of patients and costs of
medical equipment were considered.
Scores per JGP differ according to the type of admission of the patient. For
example, in general, JGP scores are lower for planned hospital admissions or
day-care treatment episodes than for unplanned hospital admissions. Table
20.5 shows an example of different scores, according to the type of admission,
for a selected group of JGPs.
In order to adequately remunerate hospitals for treating cases with very high
costs (outliers), the JGP system provides supplementary payments for cases
with a ‘very long’ lengths of stay (LOS outliers). These cases are identifi ed using
an upper LOS threshold, beyond which cases are to be considered to be outliers.
374 Diagnosis-Related Groups in Europe
The threshold is defi ned through a non-parametric trimming method based on
the interquartile range, and is applied only to certain JGPs:
upper LOS threshold = Q3+1,5*(Q3–Q1),
where:
Q1 is the LOS of the fi rst quartile of patients within a particular JGP and
Q3 is the LOS of the third quartile of patients within a particular JGP
Beyond this threshold, the JGP score is increased by a per diem-based
supplementary point value that amounts to 80 per cent of the average per diem
value per day below the upper LOS threshold. This is because it is assumed that
beyond the upper LOS threshold, the intensity of care is lower. However, upper
LOS thresholds are not calculated for planned hospitalizations and day-care
treatment episodes.
For certain JGPs the system identifi es short-stay outliers; namely, cases in
which the patient should usually stay in hospital for more than one day (in
accordance with standard medical practice). If these cases are discharged after
only one day, hospitals do not receive the full JGP-based payment. Instead, the
JGP score is reduced for these lower length-of-stay outliers to 20 per cent of the
full JGP score (except in the case of death of a patient during the fi rst day of
hospital stay). Table 20.5 shows an example of a JGP score for lower length-of-
stay outliers within a particular JGP.
20.5.3 JGP-based hospital payment
Before hospital payment takes place, the regional branch offi ces of the NFZ
check whether JGPs reported by the provider are consistent with the scope of
their contracts. For example, if a hospital has grouped a patient into a surgical
JGP but does not have the right to provide the procedure, the patient is
reclassifi ed by the NFZ into a JGP that was specifi ed in the contract.
Figure 20.4 illustrates the calculation of hospital payment under the JGP-
based hospital payment system in Poland. Hospital payment is determined,
on the one hand, by basic score points for the JGP (1a) or basic scores for
unbundled services (1b), chemotherapy (1e) or therapeutic programmes (for
Table 20.5 Example of JGP scores and supplementary points for selected JGP groups
JGP Name JGP score Upper
LOS
threshold
< 2-day
stay
score
Per diem
surcharge
General
admission
Planned
admission
Day
care
G24 Cholecystectomy
with CCs 71 70 69 – – –
G25 Cholecystecomy 63 60 57 – – –
N34 Minor surgical
intervention on
infants and babies 57 – – 10 11 5
Source: President of the National Health Fund, Order No. 69/2009/DSOZ on defi ning conditions
of concluding and executing such contracts as hospital treatment, 3 November 2009.
F
ig
u
re
2
0
.4
C
al
cu
la
ti
o
n
o
f
h
o
sp
it
al
p
ay
m
en
t
in
t
h
e
er
a
o
f
JG
P
s
So
u
rc
e:
N
FZ
,
2
0
1
0
.
376 Diagnosis-Related Groups in Europe
example, treatment of chronic hepatitis) (1g). On the other hand, additional
points are added to the score as supplementary payments for specifi c services
(1c, 1d, 1f, 1h–j). For chemotherapy (1e) or therapeutic programmes (1g), the
basic score points are supposed to cover the costs of the stay in hospital,
whereas the supplementary payment points 1f or 1h pay for specifi c medicines
or diagnostic evaluations. There are 429 such supplementary payments for
a range of specifi c high-cost drugs or procedures. The sum of points for the
JGP score, plus supplementary payment points, is multiplied by a base rate
(point value) in order to calculate hospital payment for a specifi c patient.
The base rate was PLN 51 (about €14.5) in 2009 and 2010. The rate depends
on the total available NFZ budget and is determined through negotiations
between the NFZ, the Ministry of Health and representatives of associations of
medical professionals.
20.5.4 Quality-related adjustments
The JGP system does not adjust the hospital payment for specifi c quality
indicators. All hospitals receive the same amount of money (fi xed) for each
JGP, which means that hospitals face strong incentives to lower their costs.
This could potentially compromise the quality of care. However, the quality of
hospitals is taken into account by the NFZ prior to the conclusion of contracts
with hospitals. In particular, the NFZ considers two types of quality standards
– these are implemented by health care providers of their own accord and are
not legally binding:2
1. International Organization for Standardization (ISO) certifi cates (ISO 14001,
9001) that certify organizational quality standards of management processes,
but not medical standards;
2. accreditation by the National Center for Quality Assessment, which assesses
quality on the basis of measurable (structure, process and outcome) indica-
tors. The indicators are dynamic and subject to periodical modifi cations.
In addition, further structural quality standards are specifi ed in the ‘Acts and
regulations of the Minister of Health’.3 For example, personnel must have
certain qualifi cations, equipment must undergo regular controls, and fl oor
space must comply with certain criteria. Furthermore, the NFZ verifi es that
Ministry of Health requirements for the provision of specifi c services are met
before determining the scope of contracts with hospitals, that is, before deciding
which services the hospital will be allowed to provide.
20.6 New/innovative technologies
As a result of the ‘Act on health care benefi ts fi nanced from public funds’4 the
AOTM was established. Hospitals, pharmaceutical companies and manufactures
of medical equipment can apply to the AOTM with a proposal for a new
technology. The agency assesses applications and makes recommendations
regarding whether or not a specifi c technological innovation should be included
Poland: The Jednorodne Grupy Pacjentów – Polish experiences with DRGs 377
in the list of public health care benefi ts that is published by the Minister of
Health. As part of the assessment, the AOTM establishes the level of superiority
of the new technology compared to existing treatment options, and assesses
possible ways of fi nancing it, along with conditions for implementation. In the
process, the AOTM holds meetings with a Consultative Council composed
of 12 members who are experts in the assessment of health care technologies.
The Council meets once a month and can approve a new technology through
majority vote. Subsequently, the president of the AOTM must consult with
legal experts and with the NFZ President in order to ensure that the fi nancial
impact of introducing a specifi c technology does not compromise the fi nancial
stability of the NFZ. The fi nal decision regarding whether or not to incorporate
a new technology into the list of health care benefi ts is made by the Ministry
of Health.
Unlike in other countries (see Chapter 9), there is no system to specifi cally
encourage the adoption of innovations, for example through additional pay-
ments for the use of certain technologies. If a specifi c technology is included in
the list of public health care benefi ts, it is either fi nanced through an existing
JGP, or a new JGP is created to account for the higher costs of the new technol-
ogy. Alternatively, high-cost drugs or procedures can be added to one of the lists
for supplementary payments, for example 1c, 1d, 1f, 1h–j (see Figure 20.4).
20.7 Evaluation of the JGP system
20.7.1 Offi cial evaluations
The pilot testing of JGPs in early 2008 has been the only offi cial evaluation of
the JGP system thus far. However, the main objectives at the time were to test
coding practices, the practical functioning of the JGP grouper algorithm and
the possibility of paying providers using JGPs (Paszkiewicz, 2008). In addition,
the pilot study generated information that facilitated the improvement of the
classifi cation of medical procedures and JGP characteristics. The effects of the
introduction of JGPs on quality or effi ciency of health care service provision
were not assessed during the pilot study and still remain unknown.
20.7.2 Authors’ assessment
The main advantage of the Polish JGP system is that it contributes to
transparency in the hospital sector. Grouping algorithms are publicly available
and the payment system is based on transparent procedures. In addition, the
JGP system contributes to increased availability of data relating to hospital
activity, by enforcing coding of diagnoses and procedures and grouping this
information into JGPs.
However, important shortcomings persist: cost accounting and collection
of cost data are not standardized in Poland, and cost information is not
systematically used to determine payment rates. This means that payment rates
do not necessarily refl ect the costs of service provision. Hospital payments for
378 Diagnosis-Related Groups in Europe
certain services may be too generous, while other services are not fi nanced
suffi ciently. Consequently, hospital payments may be considered unfair; they
may provide the wrong incentives and could lead to unintended behaviour
patterns on the part of providers (see Chapter 6 of this volume). Furthermore,
the system does not take into account structural differences between providers,
even though these differences may have a signifi cant impact on providers’ costs
of service provision (Busse et al., 2008).
20.8 Outlook: Future developments and reform
In 2011 the NFZ plans to implement a new change-management system. This
will be the most signifi cant alteration of the JGP system since its introduction
in 2008. The change-management system is intended to provide a regulatory
framework for the process of updating the JGP system. Three main characteristics
of the change-management system are as follows: (1) updates to the system will
occur no more than twice a year; (2) the JGP patient classifi cation system will
be updated on the basis of statistical analyses of length of stay and cost data;
and (3) cost data will be used to determine payment rates.
An interesting feature of the proposed change-management process is that it
is intended to create two lists: one detailing the most frequently performed
JGPs, and another itemizing the JGPs with the highest expenditures. Updates to
the JGP patient classifi cation system (for example, splitting of groups, creating
new groups, and so on) will focus on the most important JGPs from the two
lists. In terms of the payment rate-setting procedure, various projects are cur-
rently in the process of improving cost-accounting practices in health facilities,
supported by the Ministry of Health and the European Commission. If cost-
accounting standards can be suffi ciently improved, payment rates will be more
closely related to the costs of service provision in Polish hospitals. Furthermore,
plans exist to depart from the uniform applicability of a national base rate and
to introduce structural and quality adjustments for certain hospitals.
Another major development is that the NFZ plans to gradually extend the
JGP system to other areas of care, such as rehabilitation and psychiatry. As
mentioned in subsection 20.2.2, the introduction of JGPs for cardiological and
neurological rehabilitation in October 2010 forms part of this broader process.
In conclusion, the JGP system is still at a relatively early stage in Poland. The
planned change-management system aims to continuously improve the JGP
system to better refl ect Polish health care patterns and costs of service provision.
It seems very likely that JGP-based hospital payment will provide adequate
reimbursement to hospitals in the long term, and will thus contribute to
improved quality and effi ciency of hospital care in Poland.
20.9 Notes
1 Ordinance of the Minister of Health and Social Policy on special principles of cost
accounting in public health care facilities, 22 December 1998 (Dz. U. of 1998: No. 164,
item 1194 as amended); and Ordinance of the Minister of Health and Social Policy on
Poland: The Jednorodne Grupy Pacjentów – Polish experiences with DRGs 379
guaranteed hospital treatment benefi ts, 29 August 2009 (Dz. U. of 2009: No. 140, item
1143 as amended).
2 President of the National Health Fund, Order No. 73/2009/DSOZ on defi ning
assessment criteria for offers, 13 November 2009.
3 • Ordinance of the Minister of Health of 3 March 2004 sets forth the requirements
which need to be met by the premises and equipment of a medical diagnostic
laboratory with special focus on the sanitary condition of the premises and
equipment, as well as technical and substantive requirements for the staff and the
manager of the laboratory. (Dz.U. No. 43, item 408 of 2004 as amended)
• Ordinance of the Minister of Health of 23 March 2006 sets forth the quality
standards for medical diagnostic and microbiological laboratories as regards the
operations of medical laboratory diagnostics, assessment of their quality and
diagnostic value and laboratory interpretation and authorization of test results.
(Dz.U. No. 61, item 435 of 2006 as amended)
• Ordinance of the Minister of Health of 27 March 2008 concerning minimum
requirements for health care units which provide medical services consisting of
X-ray diagnostics, interventional radiology and radioisotope diagnostics and
treatment of non-cancerous diseases (Dz. U. 2008, No. 59, item 365)
• Ordinance of the Minister of Health of 29 March 1999 concerning the qualifi cations
of personnel at various positions in public health care institutions. (Dz. U. 1999, No.
30, item 300).
4 Act on health care benefi ts fi nanced from public funds, 27 August 2004 (Dz. U.
of 2008, No 164, item 1027, as amended).
5 President of the National Health Fund, Order No 69/2009/DSOZ on defi ning conditions
of concluding and executing such contracts as hospital treatment, 3 November 2009.
20.10 References
Busse, R., Schreyögg, J., Smith, P.C. (2008). Variability in health care treatment costs
among nine EU countries – results from the HealthBASKET project. Health Economics,
17(Suppl. 1):1–8.
Central Statistical Offi ce (2007). Basic Data on Health Care, Day Places and Outpatients in
Hospital Wards. Warsaw: Central Statistical Offi ce.
Central Statistical Offi ce (2008). Basic Data on Health Care, Day Places and Outpatients in
Hospital Wards. Warsaw: Central Statistical Offi ce.
Central Statistical Offi ce (2009a). Basic Data on Health Care, Day Places and Outpatients in
Hospital Wards. Warsaw: Central Statistical Offi ce.
Central Statistical Offi ce (2009b). Statistical Yearbook of the Republic of Poland 2009.
Warsaw: Central Statistical Offi ce (http://www.stat.gov.pl/cps/rde/xbcr/gus/PUBL_rs_
rocznik_statystyczny_rp_2009 , accessed 7 January 2010).
European Commission (2009). Eurostat: Health Care Expenditure (Data in focus 26/2008).
Luxembourg: Statistical Offi ce of the European Commission (http://epp.eurostat.ec.
europa.eu/statistics_explained/index.php/Healthcare_expenditure#Further_Eurostat_
information, accessed 15 November 2010).
Gilewski, D. (2010). Jednorodne Grupy Pacjentów. Podstawy systemu [Diagnosis-Related
Groups: Basics]. Warsaw: National Health Fund (NFZ).
Kozierkiewicz, A. (2008). Koło ratunkowe dla szpitali. Od doświadczeń do modelu [A Lifeline
for Hospitals. From Experience to a Recipe for Restructuring]. Poznan: Termedia Publishing
House.
Kozierkiewicz, A. (2009). Jednorodne Grupy Pacjentów. Przewodnik po systemie [Diagnosis-
Related Groups: Guide Through the System]. Warsaw: National Health Fund (NFZ).
380 Diagnosis-Related Groups in Europe
Kozierkiewicz, A., Stamirski, M., Stylo, W., Trabka, W. (2006). The defi nition of prices for
inpatient care in Poland in the absence of cost data. Health Care Management Science,
9(3):281–6.
Kuszewski, K., Gericke, C. (2005). Health Systems in Transition: Poland. Copenhagen: WHO
Regional Offi ce for Europe on behalf of the European Observatory on Health Systems
and Policies.
Ministry of Health (2008). Health Care Financing in Poland [in Polish]. Green Paper II,
version 3. Warsaw: Ministry of Health of the Republic of Poland (http://www.mz.gov.
pl/wwwfi les/ma_struktura/docs/zielona_ksiega_06012009 , accessed 12 January
2010).
NFZ (2010). Annual Business Reports 2009. Warsaw: National Health Fund (NFZ) (http://
www.nfz.gov.pl/new/index.php?katnr=3&dzialnr=10&artnr=4126, accessed 25 October
2010).
Paszkiewicz, J. (2008). Efekt JGP [DRG effect]. Menedz
.
er zdrowia, 10:42–9.
Schreyögg, J., Stargardt, T., Tiemann, O., Busse, R. (2006). Methods to determine reim-
bursement rates for diagnosis related groups (DRG): a comparison of nine European
countries. Health Care Management Science, 9(3):215–23.
Świderek, M. (2009). Comparative analysis of DRG systems in the EU countries. Acta
Universitatis Lodziensis, Folia Oeconomica: 303–15.
chapter t w e n t y o n e
Portugal: Results of 25 years
of experience with DRGs
Céu Mateus
21.1 Hospital services and the role of DRGs in Portugal
21.1.1 The Portuguese health system
Since 1979 the Portuguese health care system has been based on a National
Health Service (NHS) structure fi nanced by general taxation, characterized by
universal coverage and access to care that is mostly free at the point of use. The
state is committed to achieving equity, and to promoting effi ciency, quality and
accountability in the Portuguese health care system (Assembleia da República,
1990). However, the Portuguese NHS has never conformed to the general char-
acteristics of the Beveridge model of health care, mainly due to the incomplete
transition from a previously fragmented social insurance system. Occupation-
based insurance schemes that existed in 1979 are yet to be integrated into the
NHS (Barros & de Almeida Simões, 2007). These schemes benefi t from addi-
tional public funding and provide additional coverage to around 25 per cent of
the population, who enjoy double coverage.
The Portuguese health care system is mainly fi nanced through the state bud-
get (around 75 per cent; Barros & de Almeida Simões, 2007), that is, through
taxes. Since the early 1980s, total health expenditure has increased steadily and
Portugal is at present among the highest spenders in the European Union (EU)
in terms of health care expenditure as a percentage of gross domestic product
(GDP) (above 10 per cent), in comparison with other countries that have NHS-
based systems (OECD, 2008). Private fi nancing accounted for around 25 per
cent of total expenditure in 2006 (OECD, 2008), corresponding mainly to
out-of-pocket payments for specialty visits, pharmaceuticals, dental care and
physiotherapy.
All patients are assigned to an NHS general practitioner (GP) within their area
of residence. Primary care GPs are expected to act as gatekeepers and refer
patients to secondary care provided by medical specialists. However, access to
382 Diagnosis-Related Groups in Europe
emergency services is not restricted, contributing to an imperfect gatekeeping
system. A large private sector co-exists with the NHS and its role was explicitly
recognized in the 1990 NHS law that instituted a mixed health care system
(Assembleia da República, 1990). Consequently, the public and private sectors
are both involved in the delivery of health care, with the private sector mainly
responsible for carrying out specialist visits, elective surgery, ancillary tests and
kidney dialysis.
The central government level still exerts most powers according to the tradi-
tion of centralized management. Regional Health Administrations manage the
provision of primary care and are responsible for state reimbursement of pre-
scribed drugs to the Pharmacies Association. The Central Administration of the
Health System (ACSS) is an agency of the Ministry of Health, principally respon-
sible for managing NHS fi nancial resources for primary and hospital care. It also
produces statistical information and regulates information technology (IT)
both in hospitals and health care centres.
A detailed description of the Portuguese health system – including an over-
view of key institutions, relationships between the public and private sectors,
modes of payment used for different providers and services – is beyond the
scope of this chapter but can be found in Barros & de Almeida Simões (2007).
21.1.2 Hospital services in Portugal
Hospital services are provided by both the public and private sectors. Statistics
on the number of beds in the private sector are not available for the whole
country. The number of public and private hospitals (according to the last fi ve
years of data available) is presented in Table 21.1.
In 2007 the number of patients discharged per bed was 37 and (in NHS hos-
pitals only, including mental health care facilities) the average length of stay
(ALOS) corresponded to 7.7 days (DGS, 2008). The occupancy rate was close to
79 per cent, but higher in medical specialties than in surgical ones, despite
existing waiting lists for surgeries.
NHS hospitals are owned by the state but are managed as independent
institutions that are allowed to make profi ts and run defi cits. However, defi cits
are generally compensated by the Ministry of Health if they threaten the
fi nancial viability of hospitals. Hospitals provide elective and non-elective care,
ambulatory surgery, maternity services, diagnostic procedures, ancillary tests,
Table 21.1 Number of NHS (public) and private
hospitals, 2003–2007
Year NHS Private Total
2003 114 90 204
2004 116 93 209
2005 111 93 204
2006 107 93 200
2007 99 99 198
Source: INE, 2009.
Portugal: Results of 25 years of experience with DRGs 383
and accident and emergency services. Non-acute psychiatric inpatient and
outpatient services are mostly provided by psychiatric hospitals.
Private hospitals provide care to private patients, whether or not they are
covered by occupational schemes or private insurance. Private hospitals charge
patients a fee for each service according to costs incurred but can also negotiate
fees with occupational schemes or insurance plans.
Inpatient services and ambulatory surgery provided at NHS hospitals, except
psychiatric hospitals, are paid on the basis of diagnosis-related groups (DRGs).
On the one hand, DRGs have been used since the late 1980s to determine DRG-
based case payments from occupation-based and other insurance schemes to
hospitals. On the other hand, the NHS has used DRGs since 1997 for DRG-
based hospital budget allocation. Other hospital services (day care, specialist
consultations, emergency services, high-cost drugs and so on) are paid on the
basis of fee-for-service with a volume cap that is negotiated between the hospital
and the ACSS.
21.1.3 Purpose of the DRG system
When DRGs were fi rst introduced in Portuguese hospitals through a pilot study
in 1984, the Ministry of Health had two main objectives. Urbano and colleagues
(1993) who were leading the introduction process at the Ministry of Health
have recalled that:
[T]he fi rst objective of the project was to create an integrated information
system for hospital management based on a set of necessary and uniform
data, which would allow all levels of management to measure and control
their productivity, support their decision-making, make plans and budgets,
and establish equitable fi nancing criteria. The second objective was to
develop an information system that could effi ciently collect, treat, analyse,
and transmit information within hospitals, between hospitals and central
departments, and among central departments.
In summary, the two objectives were to (1) improve resource allocation, and
(2) increase transparency. However, the main goal of introducing DRGs was to
rationalize the allocation of resources to NHS hospitals by more closely linking
resources for inpatient care to hospital output (as measured through DRGs).
According to Dismuke & Sena (2001), the Portuguese Ministry of Health sought
to encourage a more effi cient utilization of resources in public hospitals in
order to increase productivity and to curb the uncontrolled growth of public
expenditure in the health care sector.
21.2 Development and updates of the DRG system
21.2.1 The current DRG system at a glance
A non-modifi ed version of All Patient (AP-)DRGs (Version 21.0) was imported
in 2006. There is only one DRG system in Portugal that applies to all NHS
384 Diagnosis-Related Groups in Europe
hospitals and all patients (inpatients and ambulatory surgery), except out-
patients and patients treated in psychiatric and rehabilitation care settings.
Private hospitals are not included in the system.
The current AP-DRG system defi nes 669 DRGs within 25 Major Diagnostic
Categories (MDCs), each corresponding to one organ or physiological system,
and one Pre-MDC (including high-cost cases such as transplantations). The
DRG system is supervised and maintained by the ACSS within the Ministry
of Health.
DRGs are used for DRG-based hospital budget allocation from the NHS to
hospitals and for DRG-based case payment from third-party payers. DRG-based
hospital budget allocations amount to about 75–85 per cent of total hospital
inpatient budgets. The rest corresponds to DRG-based hospital payments from
third-party payers. In order to control overall spending, the national base rate
can be adjusted to ensure that total hospital payment does not exceed the
available budget.
21.2.2 Development of the DRG system
In 1984 the Portuguese Ministry of Health started a pilot project to study the
feasibility of implementing United States Health Care Financing Administra-
tion (HCFA-)DRGs as a measure of hospital output. The results of the pilot study
were seen to be encouraging, and a decision was made to extend the system to
all public acute care hospitals. The implementation process followed a central-
ized top-down approach (from the Ministry of Health to the hospitals). Given
the centralization of the Portuguese political system, and the fact that the Min-
istry of Health owns the majority of hospitals, hospitals had to comply with the
decision to introduce DRGs.
Implementation originally started in the Ministry of Health and was spread
to NHS hospitals through the involvement of physicians and hospital managers
in selected hospitals. A small team led by hospital managers working at the
Ministry of Health (initially João Urbano and, after his departure, Margarida
Bentes) worked in close cooperation with the Secretary of State for Health.
Margarida Bentes was the most infl uential person regarding the implementation
of DRGs in Portugal.
After the pilot study it was decided to adopt HCFA-DRGs to the Portuguese
setting, without any adaptation. Several versions of HCFA-DRGs have been
used (see Table 21.2). In 2006, the Ministry of Health decided to switch to
Version 21.0 of the AP-DRG system, which was developed for use in hospitals
in the United States for the calendar year 2004.
The fi rst release of the AP-DRGs was Version 5.0, developed by 3M™ Health
Information Systems in the 1980s. In the United States it was effective for the
1988 calendar year. As described in the All Patient DRG Defi nitions Manual (3M,
2003, p. 12) ‘the process of forming the DRGs was highly iterative, involving a
combination of statistical results from test data with clinical judgement’.
AP-DRG Version 21.0 was imported to Portugal without modifi cations.
The most signifi cant Portuguese development concerning the use of DRGs
was that in 1996 a process was initiated to adapt DRGs for the classifi cation of
T
a
b
le
2
1
.2
V
er
si
o
n
s
o
f
D
R
G
g
ro
u
p
er
s
u
se
d
i
n
P
o
rt
u
g
al
D
R
G
s
y
st
e
m
H
C
FA
-D
R
G
V
er
si
o
n
4
.0
H
C
FA
-D
R
G
V
er
si
o
n
6
.0
H
C
FA
-D
R
G
V
er
si
o
n
1
0
.0
H
C
FA
-D
R
G
V
er
si
o
n
1
5
.0
H
C
FA
-D
R
G
V
er
si
o
n
1
6
.0
A
P
-D
R
G
V
er
si
o
n
2
1
.0
D
a
te
1
9
8
4
–1
9
8
9
1
9
9
0
–1
9
9
3
1
9
9
4
–1
9
9
8
1
9
9
8
–2
0
0
0
2
0
0
1
–2
0
0
6
2
0
0
6
–2
0
1
1
P
u
rp
o
se
P
il
o
t
st
u
d
y
In
fo
rm
at
io
n
s
y
st
em
,
D
R
G
-b
as
ed
c
as
e
p
ay
m
en
ts
f
ro
m
t
h
ir
d
p
ar
ti
es
–
D
R
G
-b
as
ed
b
u
d
g
et
a
ll
o
ca
ti
o
n
D
a
ta
u
se
d
f
o
r
d
e
v
e
o
p
m
e
n
t
N
o
n
e:
t
h
e
g
ro
u
p
er
i
s
n
o
t
d
ev
el
o
p
ed
i
n
P
o
rt
u
g
al
.
C
o
st
w
ei
g
h
ts
h
av
e
b
ee
n
a
d
ap
te
d
b
as
ed
o
n
M
ar
y
la
n
d
c
o
st
w
ei
g
h
ts
a
n
d
t
h
e
P
o
rt
u
g
u
es
e
h
o
sp
it
al
c
o
st
d
at
ab
as
e
N
u
m
b
e
r
o
f
D
R
G
s
4
7
0
4
7
7
4
9
1
5
0
3
5
1
1
6
6
9
In
c
lu
d
e
d
s
e
rv
ic
e
s
A
ll
i
n
p
at
ie
n
t
ca
re
A
ll
i
n
p
at
ie
n
t
ca
re
a
n
d
a
m
b
u
la
to
ry
s
u
rg
er
y
E
x
cl
u
d
in
g
:
o
u
tp
at
ie
n
t
ca
re
,
an
d
p
sy
ch
ia
tr
ic
a
n
d
r
eh
ab
il
it
at
io
n
c
ar
e
In
c
lu
d
e
d
c
o
st
s
N
o
n
e
Fu
ll
c
o
st
s:
i
n
cl
u
d
in
g
r
ec
u
rr
en
t
an
d
c
ap
it
al
c
o
st
s,
a
n
d
c
o
st
s
o
f
re
se
ar
ch
a
n
d
t
ea
ch
in
g
i
n
r
el
ev
an
t
h
o
sp
it
al
;
ex
cl
u
d
in
g
c
er
ta
in
h
ig
h
-c
o
st
d
ru
g
s
A
p
p
li
e
d
t
o
P
ar
ti
ci
p
at
in
g
h
o
sp
it
al
s
A
ll
N
H
S
h
o
sp
it
al
s
386 Diagnosis-Related Groups in Europe
ambulatory surgery procedures (Bentes et al., 1996; Mateus & Valente, 2000).
The rationale for this lay in the growing trend to shift care from inpatient to
ambulatory settings and an ever-growing number of ambulatory surgery proce-
dures being fi nanced through DRGs as short-stay outlier or inlier admissions,
depending on the low trim-point of the relevant DRG. From the viewpoint of
the payer (the Ministry of Health itself), this was a clear distortion of the in-
patient DRG system, considering that it was neither designed nor intended to
classify ambulatory surgery procedures.
Based on a list developed by the Irish Department of Health and a survey
of 56 selected hospitals, a set of 33 DRGs was selected from existing HCFA-
DRGs as being eligible to classify ambulatory surgery procedures based on
four criteria: physician’s responses; homogeneity of the DRG’s content; reported
volume of zero day stays for the most common selected procedure code
above 30 per cent; and, in addition, the low trim-point for the DRG in question
had to be less than or equal to two days, to preserve face validity. To ensure
validity and acceptability of results, panels of physicians were assembled and
through consensus techniques a fi nal list of 38 HCFA-DRGs were selected as
‘ambulatory surgery DRGs’ (fi ve other DRGs were included with the original
set of 33).
Subsequently, a price was computed for each ambulatory surgery DRG accord-
ing to the Portuguese DRG cost/weight model (see subsection 21.5.2). Hospital
costs were separated into those that could be assumed to vary with length of
stay (e.g. physician, hotel) and those which were likely to be similar for each
inpatient admission in the same DRG (e.g. laboratory, pharmacy). The price of
each ambulatory surgery DRG was established according to the following price
components of the corresponding inpatient DRG: 100 per cent of operating
room (OR) cost; 100 per cent of physician cost for one day; 100 per cent of hotel
and nursing costs for one day; 100 per cent of administration cost for one day;
80 per cent of the cost of supplies; 25 per cent of imaging and laboratory costs;
25 per cent of the cost of the relevant drugs for the procedure; zero per cent of
intensive care unit (ICU) and other ancillary costs.
The fi rst prices were published in 1998 and used for funding in the same year.
More recently, a similar methodology has been used to select a list of specifi c
therapeutic medical procedures that are eligible to be carried out in day-care
settings.
DRGs were introduced for hospital payment from third-party payers in 1988.
Starting in 1997, DRGs were also progressively introduced for the calculation of
hospital budget allocations from the NHS (see subsection 21.5.3). Since 2002,
the total NHS inpatient budget has been allocated through DRG-based hospital
budget allocations.
21.2.3 Data used for development and
updates of the DRG system
All DRG systems that have been in use in Portugal were purchased from abroad,
and no Portuguese data were used to develop these systems. For the selection of
Portugal: Results of 25 years of experience with DRGs 387
ambulatory surgery DRGs only, information about Portuguese ambulatory care
patterns was considered.
Table 21.3 presents sources of information that are considered for the calcula-
tion of Portuguese cost weights and for setting the national base rate. Portu-
guese DRG cost weights are calculated using Maryland cost weights (see
subsection 21.5.2) and data from the national hospital costs database at ACSS,
which contains information about treatment costs in all NHS hospitals (see
section 21.4). Cost information is forwarded electronically to the national
database, which is maintained and updated by the data unit at ACSS. Cost data
from the database are also used to calculate Portuguese cost weights for ambula-
tory surgery DRGs.
Another important database used for the calculation of Portuguese cost
weights is the DRG database at ACCS. The database contains all information
from the uniform minimum basic datasets (UMBDS) of all NHS hospitals in the
country. The national base rate is set by the Ministry of Health but the decision
is based on information relating to the average costs of the average patient for
a given year.
21.2.4 Regularity and method
of system updates
The 3M™ Health Information System regularly updates the AP-DRG system
in order to account for changes of the ICD-9-CM (WHO International Classi-
fi cation of Diseases 9th revision, Clinical Modifi cation), or to adjust the sys-
tem to new developments in medical technology or to changing practice
patterns. A variety of statistical techniques are used to ensure optimal rede-
sign and to maintain and improve quality and statistical coherence of the
grouper.
Imported DRG systems have always been implemented in Portugal without
any changes to the grouping algorithms developed in the United States.
No system of regular updates of the DRG system exists in Portugal. DRG cost
weights are recalculated at irregular intervals. The base rate is usually revised
every 18 months. However, there is always a time lag of at least two years
between the year of the data and the year of application of new DRG cost
weights in hospitals. For example, hospital cost data from the year 2009 were
analysed during the year 2010 in order to defi ne the national base rate used for
hospital payment in 2011.
Table 21.3 Sources of data for the calculation of Portuguese prices by DRG
Variable Country Source
Costs Portugal Hospitals cost database (ACSS)
Relative weights United States Maryland cost weights
Inpatient discharges Portugal DRG database (ACSS)
National base rate Portugal Ministry of Health
388 Diagnosis-Related Groups in Europe
21.3 The current patient classifi cation system
21.3.1 Information used to classify patients
After a patient is discharged from hospital, the information on her/his medical
record is abstracted to the UMBDS according to the coding rules. All patients are
classifi ed into AP-DRGs on the basis of the principal diagnosis, secondary diag-
noses, procedures, age, sex and discharge status (3M, 2003).
Since the beginning of the fi rst pilot project, diagnoses and procedures have
always been coded through ICD-9-CM and coding activities have been carried
out by trained physicians within hospitals.
21.3.2 Classifi cation algorithm
Hospitals group every discharged patient into exactly one DRG using a comput-
erized grouping software. The AP-DRG system defi nes 669 DRGs within 25
MDCs and one Pre-MDC. Figure 21.1 illustrates the grouping algorithm. In a fi rst
step, the grouping algorithm checks an exception hierarchy that specifi es certain
cases, which are separated during the Pre-MDC process (see Table 21.4). The Pre-
MDC process defi nes specifi c high-cost procedures (such as transplants) that
lead to direct classifi cation of patients into certain DRGs within the Pre-MDC, or
it assigns cases to MDCs on the basis of certain criteria other than their principal
diagnosis. All cases that do not have conditions specifi ed in the exception hier-
archy are classifi ed into MDCs on the basis of their principal diagnosis.
Within MDCs, the algorithm groups cases into a ‘surgical’ partition or a
‘medical’ partition according to whether an OR procedure was performed
during the hospital stay. At the partition level, cases with secondary diagnoses
Table 21.4 AP-DRG hierarchy in the pre-MDC process
Exception Hierarchy MDC / AP-DRG Assignment
Liver transplant Assign to AP-DRG 480
Lung transplant Assign to AP-DRG 795
Simultaneous kidney/pancreas transplant Assign to AP-DRG 805
Pancreas transplant Assign to AP-DRG 829
Heart transplant Assign to AP-DRG 103
Kidney transplant Assign to AP-DRG 302
Allogenic bone marrow transplant Assign to AP-DRG 803
Autologous bone marrow transplant Assign to AP-DRG 804
Age less than 29 days Assign to MDC 15
Principal diagnosis of HIV or secondary diagnoses of
HIV and principal diagnosis of HIV-related condition Assign to MDC 24
ECMO or tracheostomy Assign to AP-DRG 482 or 483
Principal diagnosis of trauma and at least two
signifi cant traumas from different body sites Assign to MDC 25
Principal diagnosis Assign to MDCs 1–14, 16–23
Source: 3M, 2003.
Note: ECMO: extracorporeal membrane oxygen.
Portugal: Results of 25 years of experience with DRGs 389
that are considered a ‘major CC’ (major complication and/or co-morbidity) are
separated and combined into a small number of medical or surgical ‘major CC’
groups. All other surgical cases are grouped into a surgical ‘class’ based on the
highest ranked OR procedure or into the ‘unrelated surgery’ class in the event
that the procedure is unrelated to the principal diagnosis. ‘Classes’ are similar
to base-DRGs in other DRG systems in that they comprise patients with similar
characteristics (see Chapter 4). Medical cases are grouped into medical classes
(e.g. neoplasm, or symptoms and special conditions relating to the organ sys-
tem). Both partitions also include a residual class for infrequent cases or those
that are not well defi ned. In the last step of the grouping algorithm, classes can
be split in order to increase economic homogeneity of the fi nal DRGs based
on age, discharge type, body weight (neonates) and secondary diagnoses that
are considered to be ‘CCs’. Both the ‘CC list’ and the ‘major CC list’ are globally
defi ned with exclusions and adjustments at the principal diagnosis level
(3M, 2003). There are 303 DRGs in the surgical partition and 364 DRGs in the
medical partition.
21.3.3 Data quality and plausibility checks
High-quality data are essential for any DRG system. In Portugal, coding is car-
ried out by physicians who are voluntarily trained as coders. A hospital-wide
Figure 21.1 AP-DRG (Version 21.0) classifi cation algorithm
Source: 3M, 2003.
390 Diagnosis-Related Groups in Europe
framework for data evaluation was put in place early in the process of intro-
ducing DRGs (Bentes et al., 1997). Since 1995, internal and external hospital
clinical coding audits have been carried out on a regular basis. Each hospital has
assigned an internal auditor who coordinates the data-collection process and
supervises the clinical coding of physician coders within hospitals. An external
clinical coding auditing team – composed of eight physicians and a senior man-
ager from ACSS – has been appointed to promote, support and monitor clinical
coding audit activity at the hospitals. In addition, the external auditing team
conducts site visits and verifi es patient records in order to assess whether the
classifi cation of patients into DRGs has been carried out correctly.
Clinical coding audits are supported by computer software, which identifi es
the main data errors and inconsistencies in samples of hospital records. The
software returns information about the average number of codes per record,
the percentage of invalid codes for diagnoses, procedures and administrative
data, and coding errors (e.g. opposite codes), as well as alerting the users to
coding problems or defi cient information within the medical records (e.g.
diagnosis and procedure duplication, unspecifi ed principal diagnosis) and
atypical lengths of stay.
By the end of 2009 a web portal completely devoted to coding issues was
made available.1
21.3.4 Incentives for up- or wrong-coding
As more complex DRGs have a higher cost weight, hospitals are incentivized to
code all existing secondary diagnoses of their patients. However, there is also an
incentive to engage in up-coding (see Chapter 6), although the periodic coding
audits that are carried out strongly disincentivize this behaviour. During
the external clinical coding audits that took place between 2006 and 2008, one
third of the records presented critical non-conformities. Nevertheless, only
11 per cent had errors leading to a change of the original DRG, and the change
was not always for a less complex group.
It is worth noting that as part of the fi nancing criteria for 1998, the fi nal
quality coding score of each individual hospital has been considered for
adjustments to the preliminary budget (in terms of premiums/penalties), thus
creating incentives for data-quality improvement. The impact of the premiums/
penalties was below 1 per cent of the inpatient budget of the hospital, and this
was a one-off adjustment.
21.4 Cost accounting within hospitals
21.4.1 Regulation
It is mandatory for all NHS hospitals to report their activity and costs annually
to ACSS. Since 1995 an NHS Costing Manual has been in place and sets out the
mandatory practice of costing to be applied in NHS hospitals (IGIF, 2007). With
the implementation of the Costing Manual, the collection and production of
costing information presents a greater degree of consistency.
Portugal: Results of 25 years of experience with DRGs 391
Clinical costing standards cover acute inpatient care, consultations, day-case
treatments, emergency visits and ancillary tests. In January 2010 more detail
was introduced for mental care services and psychiatric hospitals also had to
collect costing information according to the Costing Manual regulations.
21.4.2 Main characteristics of the cost-accounting system
The range of costs included in accounting terms corresponds to the full cost of
the provision of all services borne by the hospital. Therefore, all operating
expenses, staff costs and capital costs are included. As usual, total costs should be
reconciled to the fi nancial costs of the provider for the previous fi nancial year.
Costs are calculated using a top-down approach because information on
itemized resource use by individual patients is not collected at NHS hospitals.
The Costing Manual (IGIF, 2007) specifi es that hospitals should group their
costs into fi ve homogenous sections (see Table 21.5), which are cost centres
within hospitals, created to absorb direct costs and to allocate indirect costs
(IGIF, 2007).
Costs are allocated to cost centres in the principal section, following a step-
down approach that includes four steps (see Figure 21.2), as detailed here.
1. First step: imputation of direct costs into principal, auxiliary and administra-
tive sections.
2. Second step: allocation of total costs of the administrative sections to the
auxiliary and principal sections.
3. Third step: allocation of total costs of general support auxiliary sections to
the sections that benefi t from their activities.
4. Fourth step: allocation of total costs of clinical support auxiliary sections to
the sections that benefi t from their activities.
Ta ble 21.5 Homogenous sections and cost centres
Homog enous sections Relevant cost centres
1. Principal sections Clinical inpatient services, e.g. medical specialties, surgical
specialties, obstetrics, radiotherapy, ICU, transplant unit
Clinical ambulatory services, e.g. day-case treatments,
ambulatory surgery, outpatients, emergency care
2. Auxiliary sections
of clinical support
Diagnostic and therapeutic tests
Anaesthesiology
OR
Other clinical support services
3. Auxiliary sections
of general support
Buildings and equipment services
Hotel services
4. Administrative
sections
Administration and board, e.g. accountancy, management
Technical and administrative services
5. Non-imputable Costs not associated with activities of other sections, e.g. tests
ordered by other hospitals
Source: IGIF, 2007.
392 Diagnosis-Related Groups in Europe
Auxiliary sections
Principal
sections
Clinical
support
General
support
Administrative
sections
Non-
imputable
Non-
imputable
Direct
costs
0 0 0 0 0
Direct costs Direct costs
Auxiliary sections
Principal
sections
Clinical
support
General
support
Administrative
sections
Direct costs Direct costs
Indirect
costs
Indirect
costs
Indirect
costs
Total
costs
Auxiliary sections
General
support
Administrative
sections
1st
step
2ndstep
3rd step
4thstep
Direct
costs
Direct
costs
Direct
costs
Direct
costs
Direct
costs
Indirect
costs
Indirect
costs
Indirect
costs
Direct
costs
Indirect
costs
Indirect
costs
Indirect
costs
Indirect
costs
Direct
costs
Principal
sections
Direct
costs
Direct
costs
Direct
costs
Direct
costs
Non-
imputable
Clinical
support
Total
costs
Total
costs
Total
costs
Total
costs
Total
costs
Principal
sections
Indirect
costs
Direct
costs
Direct
costs
General
support
Clinical
support
Indirect
costs
Indirect
costs
Administrative
sections
Direct
costs
Indirect
costs
Indirect
costs
Direct
costs
Non-
imputable
Figure 21.2 Structure/rules for cost allocation
Source: IGIF, 2007.
Portugal: Results of 25 years of experience with DRGs 393
First-level costs – direct costs – are directly associated to a homogenous sec-
tion because they are a direct result of its activity. Second-, third-, and fourth-
level costs – indirect costs – have to be apportioned to the principal sections
that use the services they provide according to the reciprocal distribution
method (IGIF, 2007).
Hospitals report their costs for all sections (principal, auxiliary and adminis-
trative) to the national hospital cost database that is used for the DRG cost/
weight calculation model. A working group was established in 1997 in order to
improve the linkage between the hospitals’ cost accounting and the DRG cost/
weight calculation model. For all hospitals, items to be included in each cost
centre were reviewed in order to ensure greater comparability within costs
included in the model. However, the hospitals’ cost-accounting systems still use
different reporting criteria, which impacts negatively on the DRG cost/weight
calculation model.
21.5 DRG-based hospital payment in Portugal
21.5.1 Range of services and costs
included in DRG-based hospital payment
For payment of hospital activities, DRG use in Portugal is threefold: (1)
calculation of a prospective global budget for inpatient care (accounting for
about 75–85 per cent of total NHS hospital inpatient revenues); (2) DRG-based
case payment for care provided to benefi ciaries of occupation-based schemes
and insurance companies (mainly for traffi c accidents and occupational
injuries); (3) DRG-based case payment for surgical procedures included in the
waiting list recovery programme. Outpatient care, rehabilitation and psychiatric
care are not included in the system. Furthermore, hospital payment of private
hospitals is not based on DRGs.
DRG-based budgets or DRG-based case payments cover the full costs of
treatment of a patient in a particular DRG, including recurrent costs such as
salaries of medical doctors, and capital costs. There are no specifi c adjustments
for research and teaching activities, which are supposed to be fi nanced through
additional budgets from other ministries. Nevertheless, a proportion of the
higher costs of teaching and research are supposed to be taken into account
through the adjustments that are in place for different groups of hospitals (see
subsection 21.5.3).
21.5.2 Calculation of DRG cost weights
Calculation of national Portuguese cost weights suffers from the inexistence
of data relating to per-patient costs in NHS hospitals. However, a Portuguese
DRG cost/weight calculation model is in place that adjusts original AP-DRG
cost weights to the Portuguese context using Maryland cost weights and
information contained in the national hospital cost database and the national
DRG database (see section 21.2.3). Maryland cost weights are developed by the
394 Diagnosis-Related Groups in Europe
Maryland Cost Review Program and provide a set of internal cost weights that
refl ect the costs of one service relative to other services within each DRG. By
assuming that Portuguese hospitals have the same pattern of service use as hos-
pitals in Maryland, but at different levels, it is possible to determine the relative
costs of each service that comprise total hospital costs by DRG (see also Chapter
22 of this volume).
The shortcoming of this methodology lies in assuming an identical profi le of
treatment in Portugal and in Maryland in the United States. Since 1994, panels
of physicians (by MDC) have convened at the ACSS when necessary to validate
Portuguese cost weights estimated through the DRG cost/weight calculation
model.
In order to account for cases with extremely long or short length of stay,
calculated cost weights apply only to cases falling within specifi ed length-of-
stay thresholds. These thresholds are calculated on the basis of an inter-quartile
method using length-of-stay data from the national DRG database.
The national base rate is determined by the Ministry of Health on the basis of
the ‘calculated base rate’ for the previous year (e.g. 2009), and the available
budget for the next year (e.g. 2011). The calculated base rate is the quotient
between the total costs of a given year and the total number of discharges for
the same year, and it expresses the average costs for the average patient for that
year. When setting the value of the national base rate, the Ministry of Health
bears in mind its impact on third-party payers’ budgets and on the expenditure
of the NHS. The base rate is used to calculate the hospitals’ budgets and the
DRG tariff for third-party payers (see subsection 21.5.3). After changes in cost
weights or base rates, the model is recalibrated in order to ensure a national
casemix index (CMI) of one.
21.5.3 DRGs for hospital payment
DRG-based budget allocation from the NHS
Key components of the DRG-based budget allocation model are the hospital
casemix indices, hospital adjustment rates and the total number of discharges.
However, it should also be kept in mind that the amount spent under the DRG
funding system cannot exceed the national budget for inpatient care and, fre-
quently, it is necessary to adjust the national base rate to conform to that con-
straint. The hospital inpatient budget (H
i
) is calculated by multiplying the
number of standardized NHS inpatients the hospital is expected to treat during
the budget year (t) (predicted equivalent discharges)2 multiplied by its CMI in
the year before last (t-2) and multiplied by the group base rate. Hospital groups
are defi ned normatively based on the number and diversity of specialties they
can provide, among other criteria. The group base rate is determined to a cer-
tain percentage based on the hospitals’ base rates within the group (that is, the
average costs of the average patient treated in each hospital), with the remain-
ing proportion based on the national base rate.
H
i
= CMI
(t–2)
� ∑ predicted equivalent
discharges � group base rate
Portugal: Results of 25 years of experience with DRGs 395
Hospital budgets for the year 2011 were defi ned on the basis of the hospitals’
CMIs in the year 2009.
Since 2003, DRGs have been used to set the total amount of each NHS hos-
pital’s inpatient budget. However, between 1997 and 2002 they were introduced
gradually with a progressively increasing share of the budget being determined
through DRGs (Table 21.6). For some years, the Ministry of Health decided to
limit losses to zero; that is, it was decided not to reduce the budgets from one
year to another.
DRG-based case payments from third-party payers
In order to receive DRG-based hospital payments for a given month, hospitals
must provide information to third-party payers about the number and type of
DRGs that were provided to their patients. The payment for every DRG is based
on the offi cial tariff, which is determined by multiplying the applicable DRG
cost weight with the current national base rate. Payments are adjusted to
account for both long-stay and short-stay outliers. Long-stay outliers are paid,
after the high trim-point (according to the relevant DRG), at a daily rate that is
identical for all DRGs. For short-stay outliers, the payment corresponds to the
number of days multiplied by the day price for the DRG. The payment is the
same to all NHS hospitals, regardless of the type of hospital (specialty hospitals
or general hospitals).
21.5.4 Quality-related adjustments
According to the goals set in the contracts between the ACSS and hospitals, the
hospitals can receive a bonus if the percentage of readmissions in the fi rst fi ve
days after discharge remains under a defi ned threshold. The share of ambulatory
surgery procedures as a percentage of the total programmed surgical procedures
and the ALOS are also considered for the calculation of bonuses.
21.5.5 Main incentives for hospitals
One of the main goals related to the introduction of DRGs for hospital payment
was to improve effi ciency through increased activity and shorter ALOS and,
therefore, to reduce waiting times. The other goal was to control the growth of
Table 21.6 Percentage of funding based on DRGs
Year DRG (%) Previous year’s budget (%) Limit of losses
1997 10 90 Zero
1998 20 80 No limit
1999 30 70 No limit
2000 30 70 Zero
2001 40 60 Zero
2002 50 50 n/a
Note: n/a: not available.
396 Diagnosis-Related Groups in Europe
public expenditure in the health care sector. As hospitals are paid according to
the number and type of DRGs that they provide, DRGs encourage hospitals to
cut costs, to reduce length of stay and to treat more patients. Analysis of the
evolution of the number of patients treated and of the ALOS seems to confi rm
these expectations (see section 21.7).
21.6 New/innovative technologies
Innovative drugs and devices not included in any DRG are treated on an ad-hoc
basis. Physicians remain responsible for decisions related to need. However,
physicians must justify their decisions to the hospital’s board or to the Drugs
and Therapeutics Committee. Usually, clinical criteria prevail. There are no
disincentives for hospitals to use new or innovative technologies, as hospital
budgets are not fi xed and there are no penalties if hospitals incur a defi cit.
21.7 Evaluation of the DRG system in Portugal
21.7.1 Existing evaluations
Since the introduction of DRGs, the ALOS has been decreasing in Portuguese
NHS hospitals, while the number of patients discharged shows the opposite
trend (Table 21.7). This could indicate increasing effi ciency in the treatment
process, especially as occupancy rates have been constant at around 75 per cent
(Bentes et al., 2004).
It should be noted that data for the last two years originate from a different
source, which might explain the increase observed in the inpatients’ ALOS.
Furthermore, the number of patients being treated in ambulatory settings has
been rising, which certainly indicates that the complexity/severity of cases
being treated in inpatient settings has also been increasing.
In Table 21.7, the cost weights from 1998 were used for the computation of
the CMI in each year presented. This explains why the CMI is 1 for the year
1998 and higher for all subsequent years. With that information in mind it can
be ascertained that the complexity of cases being treated in Portuguese NHS
hospitals has been increasing.
As was the case in the United States, after the fi rst ten years with a DRG-based
hospital payment system, in Portugal some of the interest in evaluations of the
system has also vanished. Mateus (2008) provides a review of previous
evaluations. In summary, the results found in Portugal were similar to those in
the United States or Australia; namely, a decrement in the ALOS, and an increase
in the CMI and in the number of discharged patients. More patients are being
treated, which could indicate effi ciency gains.
21.7.2 Author’s assessment
After 25 years we can now say that the implementation of DRGs in Portugal has
been a success. Not only have NHS hospitals since the late 1980s been billing
T
a
b
l e
2
1
.7
C
as
es
t
re
at
ed
,
A
L
O
S
an
d
C
M
I
in
t
h
e
p
er
io
d
1
9
9
7
–2
0
0
6
1
9
9
7
1
9
9
8
1
9
9
9
2
0
0
0
2
0
0
1
2
0
0
2
2
0
0
3
2
0
0
4
2
0
0
5
*
2
0
0
6
*
In
p
at
ie
n
t
d
is
ch
ar
g
es
8
1
0
9
7
9
8
1
8
5
1
3
8
3
9
3
9
3
8
9
9
9
3
5
8
9
5
8
3
6
8
9
2
6
0
7
9
0
0
4
1
5
8
7
6
3
8
5
9
7
0
1
4
6
9
5
8
6
0
6
A
m
b
u
la
to
ry
s
u
rg
er
ie
s
2
0
2
3
7
2
0
3
3
4
2
4
8
5
0
2
6
8
5
7
3
7
8
5
6
4
7
7
3
5
6
3
3
6
6
7
7
8
2
1
8
0
4
1
7
9
0
4
8
7
T
o
ta
l
8
3
1
2
1
6
8
3
8
8
4
7
8
6
4
2
4
3
9
2
6
7
9
2
9
3
3
6
9
2
9
4
0
3
4
2
9
6
3
7
8
1
9
5
4
2
0
6
1
0
5
0
5
6
3
1
0
4
9
0
9
3
A
L
O
S
(i
n
p
at
ie
n
t
o
n
ly
)
7
.5
7
.5
7
.3
7
.1
7
.1
7
.0
6
.8
6
.8
7
.2
7
.2
C
M
I
0
.9
9
1
.0
0
1
.0
2
1
.0
4
1
.0
7
1
.0
8
1
.0
8
1
.1
0
1
.1
2
1
.1
3
So
u
rc
es
:
IG
IF
,
2
0
0
5
;
*
A
C
SS
,
2
0
0
7
.
398 Diagnosis-Related Groups in Europe
third-party payers based on DRGs, but also, since the late 1990s, DRG-based
hospital budget allocation has been used for the funding of NHS inpatient
activity. More patients are being treated in a shorter time, and DRGs have
proved to be a helpful cost-control mechanism (Mateus, 2008; Barros & de
Almeida Simões, 2007).
Moreover, the creation of the DRG database with morbidity information and
treatment profi les of the Portuguese population has great potential in terms of
informing decision-makers. Yet, it has remained unused, apart from in three
major fi elds: hospital funding, hospital comparisons, and the setting of national
tariffs for inpatient and ambulatory surgery care. In the author’s opinion, infor-
mation from the DRG database could be useful in the design of national health
plans and health care policy: to profi le morbidity characteristics of the Portu-
guese population at national and regional levels; to organize provision of care
according to need; to target areas for utilization review and quality assurance;
and to control the achievement of goals set in existing national health plans for
different pathologies. Furthermore, data collected in patient registries could be
used – as pointed out by Noe and colleagues (2005) – to support health econom-
ics research in certain areas, such as identifying of practice patterns and evaluat-
ing variations based on setting, examining regional differences, conducting
population sub-group analyses, determining the characteristics of high- or low-
cost patients, and so on.
It would be a major achievement to develop national cost weights, and a
sample of representative hospitals should be sought. Encouraging private hos-
pitals to code their inpatient activity with the DRGs would be another worth-
while step. The private sector is becoming increasingly more important in the
provision of care and there is no justifi cation for the non-reporting of their
activity.
21.8 Outlook: Future developments and reform
21.8.1 Trends in hospital service (or general
health care) delivery
According to the latest available data, hospital inpatient care is still very
important in Portugal and ambulatory surgery represents no more than 10 per
cent of inpatient activity. Services provided in non-inpatient settings are funded
on a fee-for-service basis. The number of hospitals being built (both public and
private) sends a signal that, at the least at the planning level, inpatient care is
and will continue to be important in the provision of health care in Portugal.
21.8.2 Trends in DRG application and coverage
The use of DRGs for hospital payment and budget allocation has been evolving
since its implementation. In the mid-1990s two pilot studies were carried out
concerning the adaptation of Ambulatory Patient Groups (APGs) to the Portu-
guese context. The results obtained from both studies were very promising;
Portugal: Results of 25 years of experience with DRGs 399
however, APGs were never implemented due to a lack of political will. Cur-
rently, ambulatory activity (consultations and ancillary services) is fi nanced on
a fee-for-service basis, which is the preferred method of payment of hospital
managers and physicians. Therefore, there is strong opposition from hospitals
to the implementation of APGs.
Studies have been carried out regarding the feasibility of using a patient clas-
sifi cation system suited for mental health care settings. No decision has been
made regarding the grouper to be used, but both the system implemented in
the United States and those being developed in Canada (System for Classifi ca-
tion of In-Patient Psychiatry)3 and Australia (Mental Health Classifi cation And
Service Costs) 4 are being closely analysed.
Due to the development of a National Network of Nursing Homes in Portugal,
interest is also being raised in the implementation of a patient classifi cation
system for rehabilitation care. At present, care is fi nanced by a per diem payment
that refl ects neither the characteristics of the patients nor of the facilities.
It can be expected that increasingly more fi elds of care will be covered by
DRG-like systems for funding and for information purposes. Portugal was one
of the earliest countries in Europe to adopt DRGs for inpatient care and it is
likely that it will also be one of the pioneers in the use of patient classifi cation
systems for other types of care.
21.9 Notes
1 Available only in Portuguese (http://portalcodgdh.min-saude.pt, accessed 10 July
2011).
2 Equivalent discharges correspond to the total number of inpatient episodes obtained
after standardizing outlier lengths of stay (below or above the low and high trim-
points) in each DRG with equivalent lengths of stay in terms of the ‘normal’ episodes
(those with lengths of stay within the trim-points). Patients with short-stay admissions
are accounted for as less than one; patients with long-stay admissions are accounted
for as more than one; and those with episodes involving lengths of stay within the
trim-points are accounted for as one.
3 See the Canadian Institute for Health Information web site (http://secure.cihi.ca,
accessed 10 July 2011).
4 See the Australian Government Department of Health and Ageing web site (http://
www.health.gov.au, accessed 10 July 2011).
21.10 References
3M (2003). All Patient DRG Defi nitions Manual (Version 21.0). St. Paul, MN: 3M Health
Information Systems.
ACSS (2007). Sistema Classifi cação de Doentes em Grupos de Diagnósticos Homogéneos (GDH).
Informação de Retorno: Ano 2006 [Patient Classifi cation System into Diagnosis-Related
Groups (DRGs). Feedback Reports: Year 2006]. Lisbon: Administração Central do Sistema
de Saúde.
Assembleia da República (1990). Lei 48/90: Lei de Bases da Saúde [Law 48/90: Basic Law
on Health]. Diário República, 195:3452–9.
400 Diagnosis-Related Groups in Europe
Barros, P., de Almeida Simões, J. (2007). Portugal: Health system review. Health Systems in
Transition, 9(5):1–140.
Bentes, M., Dias, C.M., Sakellarides, C., Bankauskaite, V. (2004). Health Care Systems in
Transition: Portugal. Copenhagen, WHO Regional Offi ce for Europe on behalf of the
European Observatory on Health Systems and Policies.
Bentes, M., Valente, M.C., Mateus, C., Estevens, S. (1997). Feedback and audit: ingredients
for quality improvement. Proceedings of the 13th PCS/E International Working Conference,
Florence, 1–3 October.
Bentes, M., Mateus, C., Estevens, S., Valente, M.C., Veertres, J. (1996). Towards a more
comprehensive fi nancing system for the Portuguese NHS hospitals. Proceedings of the
12th International PCS/E Working Conference, Sidney, 19–21 September.
DGS (2008). Centros de Saúde e Hospitais: Recursos e Produção do SNS: Ano de 2007 [Health
Care Centres and Hospitals: NHS Resources and Results: Year 2007]. Lisbon: Direcção-
Geral da Saúde.
Dismuke, C.E., Sena, V. (2001). Is there a trade-off between quality and productivity? The
case of diagnostic technologies in Portugal. Annals of Operations Research, 107(1–4):
101–16.
IGIF (2005). National DRG Database. Lisbon: Instituto de Gestão Informática e Financeira
& Departamento de Desenvolvimento de Sistemas de Financiamento e de Gestão
(DDSFG)
IGIF (2007). Plano de Contabilidade Analítica dos Hospitais (3rd edition) [Hospitals’ Accounts
Plan]. Lisbon: Instituto de Gestão Informática e Financeira.
INE (2009). Indicadores Sociais 2008 [Social Indicators 2008]. Lisbon: Instituto Nacional de
Estatística.
Mateus, C. (2008). Casemix implementation in Portugal, in J.R. Kimberly, G. de
Pouvourville, T. D’Aunno, eds. The Globalization of Managerial Innovation in Health
Care. Cambridge: Cambridge University Press.
Mateus, C., Valente, M.C. (2000). The impact of ambulatory surgery DRGs. Proceedings of
the 16th PCS/E International Working Conference, Groningen, 27–30 September.
Noe, L., Larson, L., Trotter, J. (2005). Utilizing patient registries to support health
economics research: integrating observational data with economic analyses, models,
and other applications. ISPOR CONNECTIONS, 11(5):6–8.
OECD (2008). OECD Health Data 2008. Paris: Organisation for Economic Co-operation
and Development.
Urbano, J., Bentes, M., Vertrees, J.C. (1993). Portugal: national commitment and the
implementation of DRGs, in J.R. Kimberly, G. de Pouvourville et al., eds. The Migration
of Managerial Innovation. San Francisco, CA: Jossey-Bass Inc.
chapter t w e n t y t w o
Spain: A case study on
diversity of DRG use –
The Catalan experience
Francesc Cots, Xavier Salvador,
Pietro Chiarello, Montse Bustins
and Xavier Castells
22.1 Hospital services and the role of DRGs in Spain
22.1.1 The Spanish health system
The Spanish 1978 Constitution granted all citizens the right to health protection
and care, and this was confi rmed by the 1986 General Health Care Act (GHCA).
The GHCA specifi ed the basic features of the Spanish health care system, such
as public fi nancing and universal access to public health care services free of
charge at the point of use. Furthermore, it recognized the devolution of health
care responsibilities to the Autonomous Communities (ACs), that is, to the
Spanish regions, which is an important characteristic of the Spanish health care
system today.
In 2007, total health expenditures amounted to €1980 per capita per year,
which corresponds to 8.4 per cent of gross domestic product (GDP) (European
Commission, 2011). Public expenditures (mostly fi nanced through general
taxation) are the most signifi cant source of fi nance in the Spanish health system
as they account for roughly 72 per cent of total health expenditures. Household
out-of-pocket expenditures account for about 22 per cent of total health
expenditures and are mostly spent on services not covered by the public system
(for example, dental care and services provided by private specialists). In
addition, an increasing share of the population (25 per cent in 2007) holds
private health insurance coverage (López Casasnovas, 2008), which pays for
care provided in the private sector.
402 Diagnosis-Related Groups in Europe
On the one hand, the main responsibilities of the central Government still
include setting the general framework for coordination and fi nancing of the
National Health Service (NHS), defi ning the basic NHS benefi ts package,
regulating pharmaceuticals, and coordinating medical education (Durán et al.,
2006). On the other hand, each of the 17 ACs has a Regional Health Service that
is responsible for purchasing and provision of health care. In addition, ACs
develop public health policies, and are entitled to extend the basic NHS benefi ts
package (Health Information Institute, 2010).
Most importantly, the central Government collects income taxes and value-
added tax (VAT) and allocates health budgets to the Health Service of each AC
on a simple per capita allocation basis which includes criteria for adjusting the
allocation, such as the proportion of the elderly population and insularity. In
addition, each AC is free to collect additional resources through marginal
add-on taxes on income.
Besides the general per capita allocations, the central Government also
fi nances the Health Cohesion Fund, which was created in 2002 and accounts
for less than 1 per cent of public health expenditures. The Cohesion Fund is
managed by the NHS and aims to assure equal access to health care for the
entire Spanish population. In order to do so, the Fund allocates resources to ACs
that provide care to patients from ACs in which certain services defi ned by the
Ministry of Health are not available (mostly high-technology services).
Each AC has developed its own structures and fi nancing mechanisms. This
chapter focuses on Catalonia in particular, as it was the fi rst AC to adopt
diagnosis-related groups (DRGs) for casemix analyses, management and hospital
fi nancing (HOPE, 2006). In Catalonia, in 2007, total health expenditures were at
7.4 per cent of GDP (CatSalut, 2010), of which 67.2 per cent were paid from
public sources.
22.1.2 Hospital services in Spain and Catalonia
Spain
About 40 per cent of total health expenditures in Spain are spent on hospital
care, almost exclusively (93 per cent) from public sources (European Com-
mission, 2011). There are about 770 hospitals in Spain, of which 591 are acute
care hospitals. Some 42 per cent of acute care hospitals are public (247). They
represent 72 per cent of all acute care beds, with their average being greater
than that of private hospitals (380 beds and 105 beds, respectively).
In general, the private sector offers services which are excluded from the
public benefi ts package. It has specialized in areas such as plastic surgery and
certain elective procedures for which waiting lists exist in the public sector.
As the responsibility for purchasing and provision of health care lies with the
ACs, the central Government is not directly involved in the fi nancing of
hospital care. Instead, all resources (per capita allocations and Cohesion Fund
resources) are channelled through the ACs’ Regional Health Services, which
have set up different organizational structures and management tools for the
purchasing and provision of hospital care.
Spain: A case study on diversity of DRG use – The Catalan experience 403
Catalonia
ACs differed greatly in terms of the availability of public health care infrastruc-
tures at the time of devolution. In some ACs the existing number of public
hospitals (and their capacity) was adequate for the task; in others less so. In
Catalonia, public health centres and hospitals were mostly concentrated in
major cities. Therefore, in order to ensure universal availability of services a
Public Hospital Network (Xarxa Hospitàlaria d’Utilització Pública, XHUP) was
created by incorporating hospitals from a wide range of owners, including
several town councils, the Red Cross, the Catholic Church and private charity
societies.
In the XHUP there are 68 hospitals with an average of 237 beds (Table 22.1).
Just 10 of the hospitals are directly owned by the Health Care Department and
they constitute the Catalan Health Care Institute (ICS). The remaining hospitals
have different owners (some of them private non-profi t-making entities) and
they are represented by two hospital associations: the Consorci de Salut i Social
de Catalunya and the Unió Catalana d’Hospitals.
The private sector is relatively important as it represents about 20 per cent of
discharges and 15 per cent of beds. Private care tends to be primarily used for
obstetric services (perceived to offer higher comfort and room quality), for
elective surgery (in order to avoid waiting lists), and for specialties with no
public coverage, such as cosmetic surgery and dental care.
Hospitals do not only provide inpatient care: In 2007, over 10 million ambula-
tory care specialist visits took place at public hospitals. Primary care is provided
through a public network organized on a territorial basis. Within the primary
care network, each person is assigned to a general practitioner (GP), who has a
coordinating role with the XHUP and who refers patients to hospital specialist
Table 22.1 Acute hospitals in Catalonia
Acute hospital activity
XHUP 2007
XHUP % Private
network
% Total
Hospitals 68 63.0 40 37.0 108
Beds 16 119 85.5 2 813 14.5 18 932
Occupancy rate (%) 83.8 – 63.8 – 80.7
Staff 42 624 90.4 4.512 9.6 47 136
Staff/Number of beds 2.6 – 1.5 – 2.5
Discharges 725 108 79.7 184 864 20.3 909 972
Total bed days 4 932 360 87.4 709 004 12.6 5 641 364
Ambulatory visits 10 061 109 89.1 1 232 137 10.9 11 293 246
Emergencies 3 923 380 85.2 679 619 14.8 4 602 999
Publicly fi nanced discharges 665 755 99.9 475 0.1 666 230
Publicly fi nanced discharges/
Total discharges (%)
92.7 – 0.3 – 83.9
Expenditures 4 960 000 90.9 495 000* 9.1 5 455 000
Revenues 4 917 000 91.6 450 000* 8.4 5 367 000
Sources: Authors’ own compilation based on EESRI, 2007 and CatSalut, 2007.
* Includes mixed acute, social and mental health care centres.
404 Diagnosis-Related Groups in Europe
ambulatory care. In addition, major ambulatory surgery (MAS) has increased
signifi cantly since the mid-1990s and accounted for 15 per cent of all hospital
discharges and 40 per cent of total surgical activity in 2007 (CatSalut, 2007).
The current hospital payment system in Catalonia has been in place since
1997 and is the same for the entire XHUP, independent of hospital ownership.
Global inpatient budgets are set on the basis of DRGs and structural indicators.
Additional budgets are distributed to hospitals for specifi c health programmes,
capital investments, research and education. Inpatient hospital treatment
accounts for the largest share of total hospital revenues. Outpatient consultations
are paid through a fl at fee per visit that differs according to the structural
characteristics of the hospital. Emergency care is paid for by means of a fee-for-
service system, whereby the fee is adjusted according to the structural
characteristics of the hospital. ‘Specifi c techniques’ are fi nanced through
additional payments (see subsection 22.5.1 for further details).
22.1.3 Purpose of the DRG systems in Spain and Catalonia
Spain
National use of All-Patient (AP)-DRGs in Spain has two main purposes: (1)
performance assessments and benchmarking, and (2) enabling DRG-based case
payments from the Cohesion Fund to the ACs.
The casemix index (CMI) and the length of stay per DRG are the basic indica-
tors of hospital scorecards used for performance assessments and benchmarking.
Most ACs and the national Government give hospitals feedback in terms of
national and regional DRG norms. The use of DRGs to evaluate effi ciency is
quite popular, and in some cases it is related to the evaluation of contract
programmes. However, the most extensive and important uses of benchmarking
come from private companies (IASIST, 2008).
The Cohesion Fund uses AP-DRGs to compensate ACs for care provided
within their hospitals to patients from other ACs. As already mentioned, the
Cohesion Fund was introduced in 2002 to assure equal access to public sector
hospital services for the entire Spanish population. However, prior to receiving
treatment in another AC, patients must seek authorization from their home
AC. The Cohesion Fund does not compensate ACs for emergency care provided
in their hospitals to patients from other ACs.
Catalonia
In Catalonia, DRGs have been used since 1997 to adjust hospital payments.
Since the year 2000, 35 per cent of hospital inpatient budgets have been related
to DRGs. Before the introduction of DRGs as a tool for hospital payment, the
Catalan Health Service used the UBA (Basic Care Unit) model, which paid
hospitals an equal amount of money per equivalent hospital stay (Brosa &
Agusti, 2009). The purpose of the introduction of DRGs for hospital payment
was to ensure an enhanced measure of hospital activity that would contribute
to making hospital payment more closely related to performance (Cots &
Spain: A case study on diversity of DRG use – The Catalan experience 405
Castells, 2001). Using DRGs was thought to encourage effi ciency, improve data
quality, and facilitate hospital management.
22.2 Development and updates of the DRG systems
22.2.1 The current DRG systems at a glance
All ACs provide to the Ministry of Health their minimum basic datasets (MBDS)
detailing hospital activity, grouped using 3M AP-DRGs (current version 25), as
is the standard defi ned at national level. National data from hospital patients
grouped with AP-DRGs are used to analyse the casemix and for benchmarking.
There are no national modifi cations of the imported AP-DRG system to the
number of DRGs, nor to the algorithms used.
In Catalonia, all XHUP hospitals are required to group their discharges
(inpatient and MAS) using Centers for Medicare and Medicaid Services (CMS)-
DRGs. The CHS uses CMS-DRGs in order to adjust hospital budgets (see section
22.5). However, all Catalan hospitals also have to report on AP-DRGs, which are
used to analyse effi ciency, and to compare Catalan hospitals with the rest of
Spain.
22.2.2 Development of the DRG systems
Spain has not developed a national DRG system but has relied on different DRG
systems imported from abroad. In the 1990s, Spanish authorities decided to use
imported DRG systems as there was no reliable cost-accounting information
available in Spain that would have allowed a Spanish national DRG system to
be developed. Only very few Spanish hospitals have a complete bottom-up
cost-accounting system (as explained in section 22.4).
Spain
Table 22.2 provides an overview of the main facts relating to the national-level
use of DRGs in Spain. AP-DRGs were introduced in 1999 for the benchmarking
of hospitals. Since then, copies of the MBDS of all Spanish hospitals are
transmitted to the national Ministry of Health, which uses the information to
group discharges into DRGs. Every year, hospital activity data for all Spanish
hospitals are reported by the Ministry of Health using AP-DRGs.
Since 2002, imported AP-DRG cost weights have been adjusted to the Spanish
context by using cost-accounting information from an increasingly large
sample of Spanish hospitals to calculate national tariffs. The original sample for
the calculation of national AP-DRG tariffs included 19 hospitals, increasing to
30 in 2008. Hospitals were deliberately selected in order to be representative of
all national public hospitals. However, the sample only includes one Catalan
hospital, which means that wage and price differences between ACs are not
adequately refl ected in the estimated national AP-DRG tariffs. Hospitals
406 Diagnosis-Related Groups in Europe
participating in the data sample must follow a standardized cost-accounting
methodology (see section 22.4 for more details).
Catalonia
Since 1985 some initiatives introduced United States Health Care Financing
Administration (HCFA)-DRGs in the CHS (Ibern, 1991). However, it was not
until 1997 that the Catalan Health Authority decided to offi cially introduce
CMS-DRGs to the CHS (see Table 22.3).
The CMS-DRG system is not modifi ed for use in Catalan hospitals, and
unadjusted CMS-DRG cost weights are used for hospital payment. However,
CMS-DRGs and AP-DRGs are used in Catalonia not only for inpatient care but
also to group high-profi le emergencies and MAS.
Since 1999, when data from all Spanish hospitals started being transmitted to
the national Ministry of Health, data from Catalan hospitals were included in
the national dataset. However, only in 2006 did the CHS start using AP-DRGs
for casemix analysis at the regional level. Since then, the CHS has grouped
patients discharged from XHUP hospitals using AP-DRGs. The CHS produces
annual AP-DRG reports for every hospital to facilitate performance comparisons
and to benchmark hospitals.
Table 22.2 Main facts about the use of DRGs in Spain
National/ Intercommunities
DRG system AP-DRG v. 14.1 AP-DRG v.
18.0
AP-DRG
v. 21.0
AP-DRG
v. 23.0
AP-DRG
v. 25.0
Date of
introduction
1999 2002 2006 2008 2010
(Main) Purpose Analysis/benchmarking of hospital data at national level
– DRG-based case payments for compensation
of intercommunity activity through the
Cohesion Fund
Data used for
development
Grouping algorithm:
Completely imported.
National DRG tariffs: Data
at cost centre level form a
sample of 18 hospitals
Grouping algorithm:
Completely imported.
National DRG tariffs: Data
at cost centre level form a
sample of 30 hospitals
Number of DRGs 644 656 670 676 684
Applied to MBDS of all hospitals of the NHS (inpatient and MAS care)
– ACs (to compensate patient mobility)
Included services All inpatient
care, excluding
psychiatric and
long-term care
Benchmarking: all inpatient care (excluding
psychiatric and long-term care)
DRG-based case payments: elective high-
complexity patients treated in non-
resident ACs
Included costs – Capital and recurrent costs
T
a
b
le
2
2
.3
M
ai
n
f
ac
ts
a
b
o
u
t
th
e
u
se
o
f
D
R
G
s
in
C
at
al
o
n
ia
C
a
ta
lo
n
ia
1
st
s
et
2
n
d
s
et
D
R
G
s
y
st
em
H
C
FA
-D
R
G
s
(n
o
w
C
M
S-
D
R
G
)
v.
1
3
.0
C
M
S
v.
1
6
.0
C
M
S
v.
1
8
.0
C
M
S
v.
2
0
.0
C
M
S
v.
2
2
.0
C
M
S
v.
2
4
.0
A
P
-D
R
G
v
.
2
1
.0
A
P
-D
R
G
v.
2
3
.0
A
P
-D
R
G
v.
2
5
.0
D
at
e
o
f
in
tr
o
d
u
ct
io
n
1
9
9
7
2
0
0
0
2
0
0
2
2
0
0
4
2
0
0
6
2
0
0
8
–2
0
1
0
2
0
0
6
2
0
0
8
2
0
1
0
(M
ai
n
)
P
u
rp
o
se
A
d
ju
st
t
h
e
al
lo
ca
ti
o
n
o
f
g
lo
b
al
b
u
d
g
et
s
H
o
sp
it
al
b
en
ch
m
ar
k
in
g
D
at
a
u
se
d
f
o
r
d
ev
el
o
p
m
en
t
N
o
n
e
(c
o
m
p
le
te
ly
i
m
p
o
rt
ed
)
N
u
m
b
er
o
f
D
R
G
s
4
9
2
4
9
9
4
9
9
5
1
0
5
2
0
5
3
8
6
7
0
6
7
6
6
8
4
A
p
p
li
ed
t
o
X
H
U
P
s
X
H
U
P
s
In
cl
u
d
ed
s
er
v
ic
es
In
p
at
ie
n
ts
,
M
A
S
an
d
h
ig
h
-p
ro
fi
le
e
m
er
g
en
ci
es
,
ex
cl
u
d
in
g
p
sy
ch
ia
tr
ic
a
n
d
lo
n
g
-t
er
m
c
ar
e
A
s
in
t
h
e
re
st
o
f
Sp
ai
n
In
cl
u
d
ed
c
o
st
s
D
R
G
s
d
et
er
m
in
e
3
5
%
o
f
h
o
sp
it
al
i
n
p
at
ie
n
t
b
u
d
g
et
s
n
/a
N
ot
e:
n
/a
:
n
o
t
ap
p
li
ca
b
le
.
408 Diagnosis-Related Groups in Europe
22.2.3 Data used for development and updates
of the DRG systems
All DRG versions in use in Spain and Catalonia have been imported from the
United States. Their algorithms were not developed in Spain.
National use of AP-DRGs for benchmarking and performance comparison
relies on imported cost weights. However, in order to adapt AP-DRGs for
intercommunity hospital payment, national tariffs have been calculated based
on United States cost weights and cost data from an increasingly large sample
of Spanish hospitals.
Over the last few years, cost data used for updates were always two years old:
tariffs in use in 2008 and 2009 were based on cost data from the year 2006;
tariffs in use since 2010 are based on cost data from the year 2008.
22.2.4 Regularity and method of system updates
In Spain as a whole (all ACs), DRG systems are updated every other year; that is,
every two years a new version of AP-DRGs (CMS-DRGs in Catalonia) is imported
from the United States. A new AP-DRG version is purchased every other year
by the Ministry of Health from 3M Health Information Systems. In addition,
ACs and individual hospitals purchase AP-DRGs according to their needs. In
Catalonia, CMS-DRGs are also purchased from 3M Health Information Systems.
For the last update of national AP-DRG tariffs, a new version of AP-DRGs
(version 25) was imported at the end of 2008. Then cost data from the hospital
sample were grouped using AP-DRGs during the year 2009 in order to calculate
tariffs for the year 2010. Consequently, in 2010 a version of AP-DRGs was used
that had been introduced in the United States two years earlier. In Catalonia,
CMS-DRGs are always two years old as they are introduced in the United States
two years prior to their import to Catalonia.
Both AP- and CMS- (in Catalonia) DRG systems require that information
about diagnoses and procedures is coded using the WHO International
Classifi cation of Diseases (ICD) 9th revision – clinical modifi cation (ICD-9-CM),
which is a United States-modifi ed version of the ICD. As each version of the
DRG systems is based on a specifi c ICD-9-CM version, a new ICD-9-CM version
is always imported, together with the new DRG systems.
22.3 The current patient classifi cation systems
22.3.1 Information used to classify patients
Every hospital in Spain produces a standardized minimum basic dataset
(CMBD), which provides information on demographic characteristics of each
patient, length of stay, type of admission, discharge destination, discharging
department, and diagnoses and procedures coded using the ICD-9-CM. This
information is transmitted to the Regional Health Authority (for example, the
Conselleria de Salut in Catalonia), which forwards the data to the national
Spain: A case study on diversity of DRG use – The Catalan experience 409
Ministry of Health. The national Ministry of Health extracts the necessary
information from the national dataset in order to group patients into
AP-DRGs.
In Catalonia, the Conselleria de Salut uses the same data from the CMBD to
group patients into AP- and CMS-DRGs.
22.3.2 Classifi cation algorithm
Since the general grouping algorithm of AP-DRGs is presented in the Portuguese
chapter of this volume (see Chapter 21), this section will focus on CMS-DRGs.
Figure 22.1 illustrates the grouping process of the CMS-DRG system: in the
fi rst step, the grouper checks for invalid or implausible data and classifi es patients
into Error DRGs. Subsequently, cases are assigned into one of 25 major diagnostic
categories (MDCs) on the basis of their principal diagnosis, with each MDC
corresponding to a single organ system or etiology. However, there are a certain
number of high-cost treatments (such as transplantations), which are reclassifi ed
into a Pre-MDC DRG on the basis of the performed procedure, without
considering the principal diagnosis. Within each MDC, the presence of a surgical
intervention assigns patients into the surgical ‘partition’, and its corresponding
class (for example, Major Surgery), according to the procedure.
If no procedure was performed, cases are assigned to the medical ‘partition’
and to one of four classes (such as Neoplasms), according to the principal
diagnosis. In the last step, the presence of certain secondary diagnoses that are
considered to be complications and co-morbidities (CCs) is checked, and the
fi nal DRG is determined based on CCs, age of the patient, weight of the newborn
(where relevant), and discharge status. This is similar to the process in the
AP-DRG system, but the CMS-DRG system does not differentiate between major
CCs and other CCs (see Chapter 4 of this volume).
There are a total of 281 DRGs in the medical partition and 255 in the surgical
partition. Each DRG is characterized by a three-digit number, for example DRG
167 (appendectomy without complicated diagnosis, without CCs). The numbers
are counted from DRG 001 to DRG 578 and do not indicate the MDC or the
partition.
22.3.3 Data quality and plausibility checks
Spain
The Spanish Ministry of Health audits the cost-accounting information pro-
vided by the hospitals for the calculation of national AP-DRG tariffs (Spanish
Ministry of Health and Consumption, 2008). For these hospitals a systematic
auditing process is undertaken when the CMBD is submitted to the national
Ministry of Health. The Ministry of Health verifi es the plausibility of clinical
data and the cost information provided. In addition, the Ministry performs site
visits and checks patient records within hospitals, to ensure that the information
provided is correct.
F
ig
u
re
2
2
.1
C
M
S-
D
R
G
v
er
si
o
n
2
4
g
ro
u
p
in
g
a
lg
o
ri
th
m
Spain: A case study on diversity of DRG use – The Catalan experience 411
22.3.4 Incentives for up- or wrong-coding
Catalonia
Since hospital payment depends partly on classifi cation of patients into DRGs,
hospitals have moderately strong incentives to ‘up-code’ their patients. In order
to ensure that fraudulent practices are avoided, the CHS regularly carries out
random auditing of hospital records to verify consistency between internal
patient records and those reported. However, even when unusual coding prac-
tices are detected, auditing is oriented towards improving the quality of coding,
rather than towards imposing sanctions.
22.4 Cost accounting within hospitals
22.4.1 Regulation
There are no national regulations mandating hospitals to use a specifi c cost-
accounting system. However, given that the responsibility for the public health
care system lies with the ACs, some ACs have defi ned minimum cost-accounting
standards for their hospitals, requiring them to generate patient-level cost data
(for example Catalonia and Cantabria).
In Catalonia, the Central de Balanços, which is a department of the CHS,
provides hospitals with regulations relating to the production of fi nancial state-
ments (CatSalut, 1992). Furthermore, some (mostly private) hospitals collect
patient-level cost information in order to improve hospital management.
As already mentioned, in 2008 a sample of 30 hospitals collected cost data for
the calculation of Spanish AP-DRG tariffs. Hospitals participating in the sample
were required to have at least a top-down cost-accounting system.
22.4.2 Main characteristics of cost-accounting
systems in Spain
There are only few hospitals which have complete patient-level cost information.
Different cost-accounting models are in use in Spain.
Before 2002 there was the GECLIF model (Financial and Clinical Manage-
ment) of INSALUD (the former national centre of the Ministry of Health). The
model was developed at national level within some projects that aimed to
calculate costs per department (SIGNO I and SIGNO II – still used in a few
hospitals) (González Pérez, 2008) in order to determine costs per DRG (INSALUD,
2001a, b).
At present, GESCOT™ – developed by a private consulting fi rm (SAVAC S.L.)
and based on the GECLIF model – is one of the most common and most consis-
tently used accounting systems in Spanish hospitals. It can determine costs per
patient if hospitals have a fully functioning and high-quality information
system, allocating costs to each patient according to each care service received.
It means that the Health Information System must register each service, patient
412 Diagnosis-Related Groups in Europe
and cost centre at which the service is provided. The advantage of GESCOT™
is that it has a strong and valid imputation system, based on matricial distri-
bution between cost centres, namely the Structural, Intermediate and Final cost
centres.
In addition to these systems, some ACs have started introducing and develop-
ing their own cost-accounting systems, the most important of which are listed
here.
• COANh, by the Andalusian Health Service, extended to the XHUP since
1995, uses a full costing system that employs reciprocal imputation in
attributing costs between different types of cost centres.
• ALDABIDE, by the Basque Health Service, implemented in 1994 and updated
in 1996 and 1998, aims to calculate costs at the department level.
• SIE has been implemented by the AC of Valencia since 1992.
Table 22.4 summarizes the four main cost-accounting systems in Spain, the
methodology of cost imputation to fi nal cost centres (that is, whether reciprocal
imputation or mixed imputation with iterations (loop or matricial imputation)
is used); and whether the model is able to provide patient costs through direct
allocation to patients (that is, whether the system includes a bottom-up cost-
accounting module).
Table 22.4 shows that relevant differences exist in imputation methodologies
between different cost-accounting systems. GESCOT™ has the same structure
within all hospitals for the fi rst step of the cost-accounting process – that is, the
primary cost-distribution method (top-down approach) – and each of these
hospitals has developed a different fi nal attribution structure, according to the
level of detail of the activity information system.
Monge (2003b) surveyed hospitals employing the presented accounting
systems and found that almost 40 per cent of polled hospitals declared using
their own accounting systems, and that they were characterized by imprecise
Table 22.4 Main Spanish cost-accounting systems
Accounting system Cost categories Methodology Level of aggregation
GESCOT™ Staff costs
Goods and services
Structural services
Secondary services
Reciprocal imputations Bottom up
COANh Staff costs
Goods and services
Structural services
Secondary services
Reciprocal imputations Top down
ALDABIDE Staff costs
Goods and services
Structural services
Secondary services
Mixed, with iterations
between cost centres of
the same level
Top down
SIE Staff costs
Goods and services
Mixed, with iterations
between cost centres of
the same level
Top down
Spain: A case study on diversity of DRG use – The Catalan experience 413
imputing methods and parameters, as well as slow information processing with
manual dataset capture and management.
SAVAC Consultants S.L., provider of GESCOT™, estimate that – out of the
almost 130 Spanish (public and private) hospitals using the GECLIF-derived
cost-accounting system (GESCOT™) – only a few of them (around 15) can
calculate costs per patient in accordance with a secondary cost distribution
(bottom-up) process from fi nal cost centres to patients.
22.5 DRGs for hospital payment
22.5.1 Range of services and costs included in DRG-based
hospital payment
Spain
In general, AP-DRGs are only used for determining payments from the Cohesion
Fund to compensate ACs for treating inpatients from other ACs. Payments from
the Cohesion Fund are made mostly on behalf of small ACs that do not have
the capacity to treat highly complex cases. In these cases, patients are treated
electively in hospitals of other ACs, after authorization has been obtained from
the Regional Health Authority where the patient lives. Payments from the
Cohesion Fund are supposed to include all costs categories, that is, capital costs
(for example, buildings and equipment) and running costs (for example,
personnel and supplies).
Catalonia
In Catalonia, CMS-DRGs are used to determine DRG-based budgets for all
hospitals within the public network, which includes many non-profi t-making
organizations. DRGs are used not only for inpatient activity (hospitalization),
but also for MAS and high-profi le emergencies (stays of longer than 12 hours,
deaths or transfers to other hospitals). In general terms, DRGs include only
acute care and do not cover psychiatric and long-term care. The inclusion of
MAS was a political decision designed to provide a powerful incentive to set
surgery in an outpatient setting, as one of the measures to reduce waiting lists,
as well as to reduce costs.
About 15–20 per cent of total revenues in Catalan hospitals are related to the
DRG-based CMI (relative resource intensity, RRI), which means that incentives
to use DRGs are only moderate or weak. A much larger share of hospital revenues
is determined by the hospitals’ structural characteristics – namely its equipment,
size, and so on – which infl uence payments not only for inpatient care but also
for outpatient care and emergency care.
Non-surgical day cases are fi nanced through fee-for-service prices, adjusted at
the hospital level. Outpatient consultations are paid by means of a fl at fee per
visit, which is supposed to cover all possible following visits, and differs accord-
ing to the structural characteristics of the hospital. Emergency care is fi nanced
by a fi xed price that is adjusted according to the structural characteristics of the
hospital.
414 Diagnosis-Related Groups in Europe
Hospitals receive additional funding for teaching and research. Furthermore,
certain specifi c techniques are paid for on a fee-for-service basis (such as radio-
therapy sessions), while others are fi nanced through specifi c budget allocations
(such as breast cancer screening programmes). Furthermore, a specifi c fund
exists that fi nances surgical activity related to waiting list reduction and a
programme to cope with emergency pressure in the winter. There are also
additional payments for high-complexity treatments and diagnostic tests, such
as radiotherapy, neuroradiology, catheterization and dialysis.
22.5.2 Calculation of DRG prices/cost weights
Spain
As already mentioned, Spanish national tariffs are calculated by adapting
American AP-DRG cost weights on the basis of cost information from 30 Spanish
hospitals. To elaborate these datasets and relative hospital-level costs, fi rst a top-
down cost allocation is realized in order to estimate costs of 11 ‘partial cost
centres’ (Operating Room, Radiology, Laboratory, Pharmacy, Medical Services,
Intensive Care, Other Hospitalization Costs, Other Intermediate Hospitalization
Costs, Medical Staff, Functional Costs, and Overheads).
Hospital cost-accounting systems perform a top-down process using a limited
amount of clinical data. Once the top-down distribution to the 11 partial cost
centres is completed, American DRG weights are used to value the cost of each
patient and calculate an average cost per DRG (Falguera Martínez-Alarcón,
2001). The main weakness of this system is that it calculates an estimated,
rather than real, cost per patient (Table 22.5).
Table 22.5 Evaluation of unit cost using internal DRG weights
CC
1
. . . CC
i
. . . CC
11
DRG
1
N
1
* W
1-1
N
1
* W
i-1
N
1
* W
11-1
. . .
DRG
j
N
j
* W
1-j
N
j
* W
i-j
N
j
* W
11-j
. . .
DRG
886
N
886
* W
1-886
N
886
* W
i-886
N
886
* W
11-886
Total weighted activity TW
1
=Σ(N
j
* W
i=j
) TW
i
=Σ(N
j
* W
1-j
) TW
11
=Σ(N
j
* W
11-j
)
UC
11
= TCOST
11
/
Unit cost 1 to 11 UC
1
= TCOST
1
/ TW
1
UC
i
= TCOST
i
/ TW
i
TW
11
Cost per DRG
j
CDRG
j
= Σ(UC
i
* W
i-j
)
CC
i
is a partial cost centre
W
i-j
is the internal (partial) DRG weight for DRG
j
and the partial cost centre CC
i
N
j
is the total number of patients classifi ed into DRG
j
TCOST
i
is the total cost for the partial cost centre CC
i
UC
i
is the unit cost for the internal (partial) cost weight W
i
CDRG
j
is the cost in Euros for DRG
j
Spain: A case study on diversity of DRG use – The Catalan experience 415
The imputation is realized for each discharge (by its length of stay) and for
each ‘partial cost centre’. The weights are based on information about costs for
these partial cost centres from a large number of American datasets, which are
supposed to be statistically representative (Spanish Ministry of Health and
Consumption, 2008).
By multiplying the number of cases (N
j
) in each DRG (DRG
j
) with an internal
AP-DRG cost weight (W
i-j
), the total weighted activity is calculated for each
partial cost centre. Subsequently, the total costs of each partial cost centre (W
i
)
(derived from the top-down cost allocation) are divided by its total weighted
activity to calculate a partial unit cost and then assigned to each discharge, for
which a total cost can be calculated. Consequently, average costs for each
AP-DRG can be calculated, which are used for setting national AP-DRG tariffs.
Catalonia
In Catalonia, the original CMS-DRG cost weights are used. However, in the
process of determining hospital budgets, CMS-DRG cost weights are used only
as an indicator of the RRI of cases within one hospital compared to the RRI of
cases in the entire XHUP.
22.5.3 DRGs in actual hospital payment
Spain
The process before payment from the Cohesion Fund takes place can be
described as consisting of several steps: (1) authorization must be obtained from
the AC in which the patient is living; (2) the patient is transferred to another
AC for treatment; (3) after treatment has been completed, the hospital is paid
on the basis of the normal system of payment applicable in the hospital’s own
AC; (4) the Cohesion Fund compensates the AC in which the patient was
treated for the provided services.
In order to determine payment to a specifi c AC, the MBDS of all non-resident
patients treated in the AC are submitted to the Cohesion Fund at the end of the
year. The Cohesion Fund groups the patient information from the CMBD into
AP-DRGs, and pays hospitals on the basis of the national tariff, which is the
same for all ACs.
Catalonia
The current Catalan hospital payment system relies on two types of information
in order to determine global hospital budgets: (1) the RRI of cases treated by the
hospital (measured through CMS-DRGs), and (2) each hospital’s structural
characteristics. Based on these two types of information, hospitals are paid for
the number of discharges contracted by the CHS.
Figure 22.2 shows how the RRI of cases treated by a hospital is accounted for
in the payment system: fi rst, the CMI is calculated for each hospital by dividing
the sum of all CMS-DRG cost weights of all patients treated by the hospital in
F
ig
u
re
2
2
.2
T
h
e
h
o
sp
it
al
p
ay
m
en
t
sy
st
em
i
n
C
at
al
o
n
ia
So
u
rc
e:
A
d
ap
te
d
f
ro
m
S
án
ch
ez
-M
ar
tí
n
ez
e
t
al
.,
2
0
0
6
.
Spain: A case study on diversity of DRG use – The Catalan experience 417
the previous year by the number of total patients treated. Accordingly, the CMI
of the entire XHUP is calculated (Sánchez-Martínez et a l., 2006). Second, each
hospital’s CMI is divided by the casemix of the XHUP in order to determine the
RRI of patients treated in each hospital. Third, the complexity of all discharges
within the XHUP is determined by summing the product of the RRI of each
hospital (RRI
h
), multiplied by the number of discharges in the hospital. Finally,
the price per RRI (RRI price) is calculated by dividing the predetermined global
hospitalization budget by the sum of complexity-weighted discharges of the
XHUP.
In order to account for the structural characteristics of each hospital, a struc-
tural relative index (SRI) is calculated by the CHS for each hospital every four
years. First, different structural groups are defi ned (for example, university hos-
pitals, specialized hospitals), and structural weights are assigned to each group.
Then, each hospital is classifi ed into one or more of these structural groups, and
its Grade of Membership to the group is determined through regression analysis.
Finally, the SRI is computed for each hospital by applying the Grade of Mem-
bership proportions to the mean structural weight of each group.
In order to calculate DRG-based hospital budgets, the RRI and the SRI are
multiplied by the RRI price and the SRI price, which are established by the
CHS and updated every year. Hospital budgets for the year 2010 were determined
on the basis of hospital activity data for the period from July 2008 to June
2009.
When CMS-DRGs were introduced in 1997, the CHS planned to gradually
reduce the weight of the SRI and to increase the weight of the RRI. However, the
weight of the RRI has been increased only once (in the year 2000), from 30 per
cent to 35 per cent. Consequently, the hospital structure (the SRI) still determines
65 per cent of hospital payment. Apparently, the current weight attached to the
RRI is not signifi cant enough to motivate hospitals to attract more complex
patients, as the complexity of patients (measured through CMS-DRGs) is respon-
sible for only 35 per cent of hospital payment. Figure 22.3 shows the distribution
of RRI values for each hospital within the network of public hospitals in
Catalonia.
22.5.4 Quality-related adjustments
There are no quality-related adjustments to hospital payments on the basis
of DRGs. In general, it is assumed that certain structural characteristics of
hospitals – such as teaching status – imply higher quality and higher costs.
However, these costs are not refl ected in the DRG weight but are accounted for
in the structural payment components of the Catalan hospital fi nancing system.
22.5.5 Main incentives for hospitals
Since hospital payment is only partly based on DRGs, hospitals have only
moderate incentives to up-code their patients (especially to increase the number
and severity of secondary diagnoses).
F
ig
u
re
2
2
.3
R
el
at
iv
e
re
so
u
rc
e
in
te
n
si
ty
v
al
u
es
f
o
r
C
at
al
an
h
o
sp
it
al
s,
2
0
0
9
So
u
rc
e:
S
ec
re
ta
ri
a
T
èc
n
ic
a
U
C
H
,
2
0
0
9
.
Spain: A case study on diversity of DRG use – The Catalan experience 419
‘Present on admission’ fl ags are not used in the Spanish coding system, and the
DRG system is unable to discriminate between secondary diagnoses representing
genuine co-morbidities and secondary diagnoses, refl ecting complications as a
result of medical errors or negligence. However, it remains unclear whether this
inability to discriminate between different types of secondary diagnoses has a
negative effect on treatment quality. The fact that hospitals have received the
same payment for surgical procedures performed in an ambulatory setting (MAS)
as for those surgical patients treated as inpatients has been a strong incentive to
substitute inpatient surgery with MAS.
22.6 New/innovative technologies
Providers (for example, teaching hospitals) are entitled to make use of any
health technology on the market. However, in order to receive additional pay-
ments for a specifi c new technology that is more costly than an existing one,
hospitals need to apply to the CHS for funding. Unfortunately, there is not a
clearly specifi ed process of steps required in order for hospitals to receive
additional payment. Applications by hospitals proposing the introduction of a
specifi c technology can be either accepted by the CHS or rejected, depending
on the available evidence on effectiveness and costs of the new technology.
The costs of initial applications of new technologies are usually borne by
providers (such as teaching hospitals) and ‘the industry’ in general (namely,
pharmaceutical companies or medical device manufacturers). Eventually, the
technology is either included in the general benefi ts basket and paid for as with
any other activity, or it is added to a list of certain approved innovative services
to be fi nanced from a specifi c fund for certain techniques and specifi c procedures
(Brosa & Agusti, 2009).
Treatments and procedures that are fi nanced through additional (fee-for-
service) payments are generally delivered in day-care settings, and include high-
complexity treatments and diagnostic tests, such as radiotherapy, stereotactic
radiosurgery and neuroradiology, cardiac or hepatic catheterism, as well as
highly technical care procedures such as those in urology departments and care
of patients with renal failure.
In general, as hospital payment is only partially based on DRGs, the DRG-
based incentives and disincentives relating to introducing new technologies
(see Chapter 9) are thought to be of only moderate strength in Catalonia.
22.7 Evaluation of DRG systems in Spain
22.7.1 Offi cial evaluations
Spain
The Spanish Ministry of Health oversees the development of AP-DRGs and, re-
cently, has also been pilot testing International Refi ned (IR)-DRGs. The Ministry
publishes data and information about casemix and costs based on AP-DRGs
(Spanish Ministry of Health and Consumption, 2008).
420 Diagnosis-Related Groups in Europe
Recently, the potential to change to the IR-DRG system has been discussed,
as this would provide the opportunity to include in one DRG-based system also
non-surgical outpatient activity and emergency care.
Catalonia
The CHS has used HCFA- and later CMS-DRGs since the fi rst introduction of
DRGs in Catalonia in 1997. It did not make the change to the AP-DRG system
used in the rest of the country because the change would have brought about
signifi cant changes in the distribution of resources among hospitals.
22.7.2 Authors’ assessment of successes and problems
Spain
Spain has adopted a foreign DRG system based on the notion that it was too
small to develop its own system. Several articles have been published by the
authors of this chapter criticizing the adoption of a foreign DRG system. How-
ever, given the diffi culty of developing and updating a national DRG system,
and given the increasing interest in cross-border comparisons of hospital perfor-
mance, the adoption of a foreign system does not seem to be such a bad choice.
If Spanish authorities seek to adopt a DRG system that explicitly considers
outpatient care, Spain can migrate to the IR-DRG system without any signifi cant
cost. Currently, a project is under way, which aims to estimate national Spanish
cost weights for IR-DRGs, using detailed patient-level cost data from a sample
of hospitals following a common bottom-up cost-accounting methodology.
Therefore, once these national cost weights are available, a change to IR-DRGs
would also have the advantage that IR-DRGs would better refl ect Spanish
practice patterns.
Yet, any DRG system has the limitation that it only partially refl ects the entire
patient health care process. DRGs are always related to only one hospital stay
and ignore care provided prior to admission or after discharge.
Catalonia
Almost 85 per cent of hospital fi nancing remains related to structural indicators
(SRI or prices for medical day cases, outpatient visits and emergencies based on
structural levels) (Brosa & Agusti, 2009). Only 15–20 per cent of hospital
revenues are related to the DRG-based RRI index. Consequently, the Catalan
system of adjusting hospital payment on the basis of DRGs carries only moderate
incentives, whether these are intended (to increase effi ciency) or unintended
(to engage in up-coding). Since hospital revenues are mostly determined by
their SRI, hospitals are more likely to focus on introducing new and advanced
technologies in order to increase their SRI, rather than focusing on improving
performance as measured by DRGs.
The importance of structural fi nancing components can be partly explained
by the fact that outliers are not accounted for in the Catalan hospital fi nancing
Spain: A case study on diversity of DRG use – The Catalan experience 421
system. Teaching hospitals tend to have a higher rate of outliers (Cots et al.,
2003). Increasing the share of DRG-based payment in total hospital revenues
would mean that hospitals would not receive adequate payments if they have a
high rate of outliers. If the CHS wanted to increase the importance of DRG-
based fi nancing, it would need to fi nd a way of paying hospitals for treating
outliers. Until then, the SRI ensures that those hospitals that are likely to treat
a large share of outlier patients (for example, teaching hospitals) receive suf-
fi cient funding to cover their associated costs.
22.8 Outlook: Future developments and reform
The most signifi cant trend in health care delivery in Spain (including Catalonia)
is the increasing importance of outpatient care in hospital activity. Consequently,
there is a need for the health care system to develop and use management tools
that better fi t with the new patterns of service delivery. In this context, adoption
of the IR-DRG system would be a step forward. IR-DRGs have been specifi cally
developed to better integrate outpatient activity with inpatient activity. Unfor-
tunately, the benefi ts of the process of moving towards IR-DRGs at the national
level are not being suffi ciently promoted in Spain.
Furthermore, there is an increasing awareness of the need to coordinate
hospital-level care with primary care and long-term care. Consequently, there
has been a lot of interest in per capita grouping algorithms, such as Adjusted
Clinical Groups (Sicras-Mainar & Navarro-Artieda, 2009) or Clinical Risk Groups
(Inoriza et al., 2009). Currently, the CHS is focusing on a project for the
development of per capita grouping mechanisms (Brosa & Agusti, 2009).
Until now, suffi ciently detailed information about treatment costs in Spanish
hospitals – to inform decisions relating to DRG system development – has
remained unavailable. Ten Spanish hospitals have now built a cost database1 for
per-patient cost information, which has been used to provide information for
the EuroDRG project. The intention is to regularly update the database with
new cost information, with consistent input from a greater number of partici-
pating hospitals, and to make it accessible for use by other hospitals, researchers
and benchmarking projects.
In the Spanish context, which includes the specifi c case of Catalonia, there is
marked stagnation in the development of patient classifi cation systems. The
current hospital fi nancing system does not require refi ned per-patient cost
information. The division of regulating, fi nancing, purchasing and supplying
functions of the health care system has not been consolidated, even though
this separation was clearly specifi ed in national legislation. DRGs are and will
probably continue to be used as tools to generate quality indicators (effi ciency
and effectiveness), but they also continue to have a minor impact on the
management of the health care system and its transformation.
The inability to make changes is a signifi cant aspect of this stagnation.
Although the 17 Spanish ACs could independently introduce changes in patient
classifi cation systems, the overall health system requires homogeneity in its
health care model. An active process of improvement is hampered by the dupli-
cation of responsibilities among the different levels of government, namely the
422 Diagnosis-Related Groups in Europe
ACs and central Government. A joint national and regional effort would be
required in order to progress to a better DRG system and to use it more
consistently for hospital payment.
22.9 Note
1 See the Red Española de Costes Hospitalarios (Spanish Network of Hospital Costs) web
site (www.rechosp.org, accessed 10 July 2011).
22.10 References
Brosa, F., Agusti, E. (2009). Public Health Care Payment Systems in Catalonia, 1981–2009
(Historical Evolution and Future Perspectives). Barcelona: Servei Català de la Salut,
Generalitat de Catalunya Departament de Salut (http://www10.gencat.net/catsalut/
archivos/publicacions/planif_sanit/sistemes_pagament_angles , accessed 10 July
2011).
CatSalut (1992). La Central de Balanços del Servei Català de la Salut [Central Institute of Financial
Statements of the Catalan Health Service]. Barcelona: Servei Català de la Salut, Departament
de Sanitat i Seguretat Social (http://www10.gencat.cat/catsalut/archivos/central_
balancos/informe_3_cat , accessed 10 July 2011).
CatSalut (2007). Activitat asistencial de la xarxa sanitària de Catalunya. Any 2007 [Health
Care Activity of the Catalan Health Network]. Barcelona: Servei Català de la Salut,
Generalitat de Catalunya Departament de Salut (http://www10.gencat.cat/catsalut/
archivos/cmbd/cmbd_07 , accessed 27 July 2011).
CatSalut (2010). La Salut com a sector econòmic de Catalunya [Health Care as an Economic
Sector of Catalonia. Catalan Health Service]. Barcelona: Servei Català de la Salut,
Generalitat de Catalunya Departament de Salut (http://www10.gencat.cat/catsalut/
archivos/publicacions/econo_sanitaria/informes/informe_10_cat , accessed 10
July 2011).
Cots, F., Castells, X. (2001). Cómo pagamos a nuestros hospitales. La referencia de
Cataluña y el contrapunto desde Andalucía [How we pay our hospitals. The point of
reference of Catalonia and the counterpoint from Anadalusia]. Gaceta Sanitaria,
15(2):172–81.
Cots, F., Elvira, D., Castells, X., Dalmau, E. (2000). Medicare’s DRG-weights in a European
environment: the Spanish experience. Health Policy, 51:31–47.
Cots, F., Elvira, D., Castells, X., Saez, M. (2003). Relevance of outlier cases in casemix systems
and evaluation of trimming methods. Health Care Management Science, 6(1):27–35.
Durán, A., Lara, J.L., van Waveren, M. (2006). Spain: health system review. Health Systems
in Transition, 8(4):1–208 (http://www.euro.who.int/__data/assets/pdf_fi le/0005/
96386/E89491 , accessed 10 July 2011).
EESRI (Servei d’Informació i Estudis) (2007). Informació estadística de l’assistència
Hospitalària. Catalunya 2007 [Statistical Information on Hospital Care. Catalonia 2007].
Barcelona: Generalitat de Catalunya Departament de Salut (http://www.gencat.cat/
salut/depsalut/html/ca/xifres/eesri07 , accessed 10 July 2011).
European Commission (2011). Eurostat: Database on Health Care Expenditure (February
update). Luxembourg: European Commission (http://epp.eurostat.ec.europa.eu/
portal/page/portal/statistics/search_database, accessed 27 July 2011).
Falguera Martínez-Alarcón, J. (2001). La contabilidad de gestión en los centros sanitarios
[Management Accounting in Health Care Centres] (Thesis). Barcelona: Universitat
Pompeu Fabra.
Spain: A case study on diversity of DRG use – The Catalan experience 423
González Pérez, J.G. (2008). Informe SEIS: La gestión de los medicamentos en los Servicios de
Salud. Sistemas de gestión de costes, benefi cios y oportunidades de desarrollo con las TIC
[Informe SEIS: Drug Management in Health Care Services: Cost-Accounting Systems.
Advantages and Development Opportunities with ICT Tools]. Madrid: Sociedad Española
de Informática de la Salud.
Health Information Institute (2010). National Health System Spain 2010. Madrid: Ministry
of Health and Social Policy, Health Information Institute (http://www.msps.es/
organizacion/sns/docs/sns2010/Main , accessed 10 July 2011).
HOPE (2006). DRGs as a Financing Tool. Brussels: European Hospital and Healthcare
Federation (http://www.hope.be/05eventsandpublications/docpublications/77_drg_
report/77_drg_report_2006 , accessed 10 July 2011).
IASIST (2008). Top 20. What is it? Barcelona: IASIST S.A, UBM Medica (http://www.iasist.
com/en/top--20, accessed 10 July 2011).
Ibern, P. (1991). The development of cost information by DRG – experience in a Barcelona
hospital. Health Policy, 17:179–94.
Inoriza, J.M., Coderch, J., Carreras, M. et al. (2009). Measuring the severity and resource
utilization of hospitalized and ambulatory patients. Gaceta Sanitaria, 23(1):29–37.
INSALUD (2001a). Gestión clínico-fi nanciera y coste por proceso [Clinical-Financial Manage-
ment and Cost Per Process]. Madrid: Instituto Nacional de la Salud.
INSALUD (2001b). Resultados de la Gestión Analítica en los hospitales del INSALUD [Results
of Analytical Accounting Management in INSALUD Hospitals]. Madrid: Instituto Nacional
de la Salud.
López Casasnovas, G. (2008). El paper de l’assegurament sanitari i de la medicina privada en
els sistemes públics de salut. Refl exions per a l’elaboració d’una política sanitària nacional
[The Role of Health Insurance and Private Medicine in Public Health Systems. Considerations
for National Health Policy-Making]. Barcelona: Universitat Pompeu Fabra Centre de
Recerca en Economia i Salut.
Monge, P. (2003a). Estudio comparativo de los diferentes sistemas o modelos de costes
implantados en los hospitales públicos españoles [Comparative study of different
cost-accounting systems implemented in Spanish public hospitals]. Revista
Iberoamericana de Contabilidad de Gestión, 2:13–42.
Monge, P. (2003b). Ventajas e inconvenientes de los diversos sistemas de costes
implantados en los hospitales españoles [Advantages and disadvantages of the
different cost-accounting systems implemented in Spanish public hospitals]. Boletín
Económico de Información Comercial Española, 2764:17–25.
Sánchez-Martínez, F., Abellán-Perpiñán, J.M., Martínez-Pérez, J.E., Puig-Junoy, J. (2006).
Cost accounting and public reimbursement schemes in Spanish hospitals. Health
Care Management Science, 9:225–32.
Secretaria Tècnica UCH (2009). Análisi de l’evolució del preu alta i dels valors IRR i IRE període
1999–2009 en el conjunt de centres de la XHUP [Analysis of the Evolution of SRI and RRI
Prices and SRI and RRI Values for the Hospitals of the Public Hospitals Network of
Catalonia]. Barcelona: Unió Catalana d’Hospitals.
Sicras-Mainar, A., Navarro-Artieda, R. (2009). Validating the adjusted clinical groups
[ACG] casemix system in a Spanish population setting: a multicenter study. Gaceta
Sanitaria, 23(3):228–31.
Spanish Ministry of Health and Consumption (2008). Análisis y desarrollo de los GDR en el
Sistema Nacional de Salud [Analysis and Development of DRGs in the National Health
System]. Madrid: Spanish Ministry of Health and Consumption (http://www.msc.es/
en/estadEstudios/estadisticas/inforRecopilaciones/anaDesarrolloGDR.htm, accessed
10 July 2011).
chapter t w e n t y t h r e e
The Netherlands: The
Diagnose Behandeling
Combinaties
Siok Swan Tan, Martin van Ineveld,
Ken Redekop and
Leona Hakkaart-van Roijen
23.1 Hospital services and the role of Diagnose
Behandeling Combinaties in the Netherlands
23.1.1 The Dutch health care system
The Dutch health care system is mostly health insurance based and is divided
into three compartments (Stolk & Rutten, 2005; Schäfer et al., 2010; Enthoven
& van de Ven, 2007). The fi rst compartment consists of a compulsory social
health insurance scheme, which provides continuous long-term care for those
with chronic conditions and short-term home nursing care for acute condi-
tions. This social health insurance scheme is regulated in the Exceptional
Medical Expenses Act (Algemene Wet Bijzondere Ziektekosten, AWBZ). The AWBZ
is mainly fi nanced through income-dependent contributions. Care is provided
after needs assessment has taken place, and is subject to a complicated system
of cost-sharing.
The second compartment consists of a social health insurance scheme
covering the whole population for ‘basic health insurance’. Since January 2006,
previously existing public sickness funds and private health insurance schemes
have been integrated into one compulsory scheme, which is regulated by the
Health Insurance Act (Zorgverzekeringswet, ZVW) (Schut & Hassink, 2002).
Health insurers must offer a standard benefi ts package including most curative
medical care (general practitioners (GPs), medical specialists, short-term
hospital care). All Dutch citizens contribute to this scheme in two ways. First,
they pay a fl at-rate premium directly to the health insurer of their choice.
Second, an income-dependent employer contribution is deducted through
426 Diagnosis-Related Groups in Europe
their payroll and transferred to the Health Insurance Fund. The resources
from this Fund are then allocated among the health insurers according to a
risk-adjustment system. A ‘health care allowance’ should partly compensate
lower-income individuals for their health insurance costs.
The third compartment consists of complementary voluntary health insur-
ance (VHI), which may cover health services that are not covered under the
AWBZ and the ZVW. Prevention and social support are not part of the compul-
sory social health insurance or VHI, but are mainly fi nanced through general
taxation.
Three independent institutions under the Ministry of Health, Welfare
and Sport (Volksgezondheid, Welzijn en Sport, VWS) are central actors in terms
of supervision and regulation of the Dutch health care system. The fi rst is the
Healthcare Inspectorate (Inspectie voor de Gezondheidszorg; IGZ), which monitors
and controls the quality of health care services, prevention measures and medi-
cal products. The second is the Dutch Healthcare Authority (Nederlandse Zorg
autoriteit; NZa), which determines the fi nancial framework, budgets and tariffs,
as well as advising the VWS on setting the conditions for regulated competi-
tion. The third institution is the Healthcare Insurance Board (College Voor
Zorgverzekeringen; CVZ), which advises the VWS on benefi ts package issues and
monitors compliance with the AWBZ and the ZVW.
In 2005, total health care expenditure amounted to about €68 billion, which
is equal to about 12 per cent of the country’s gross domestic product (GDP).
The Dutch health care system is predominantly fi nanced by the AWBZ (about
27 per cent) and the ZVW (41 per cent). Only 4 per cent is fi nanced by VHI.
Other sources of fi nancing include out-of-pocket expenses (10 per cent), the
VWS (13 per cent) and health care-related profi t-making and non-profi t-
making organizations (5 per cent).
In general, the Dutch health care delivery system is divided into 11 sectors.
The hospital sector is the most signifi cant sector in terms of expenditure (26 per
cent in 2005). Other important sectors include elderly care institutions (19 per
cent), social service institutions (12 per cent) and suppliers of pharmaceuticals
and medical aids (12 per cent). The ‘other health care providers’ sector (3.4 per
cent) comprises, amongst others, Independent Treatment Centres (Zelfstandige
Behandel Centra; ZBCs) and private clinics (Poos et al., 2008).
23.1.2 Hospital services in the Netherlands
Inpatient care and day care are only provided by hospitals. In 2009, there
were 8 university hospitals, 85 general hospitals, 32 specialized hospitals and
23 rehabilitation centres in the Netherlands. The specialized hospitals com-
prised 1 abortion clinic, 4 audiology centres, 3 dialysis centres, 2 epilepsy
centres, 10 integral cancer centres, 4 radiotherapy centres, 3 asthma centres and
5 other specialized hospitals. All hospitals work on a non-profi t basis but
may provide services excluded from the standard benefi ts package, which are
reimbursed by VHI.
Table 23.1 presents some key fi gures for university and general hospitals in
2009 (Kiwa Prismant, 2010). The number of inpatient days in 2009 amounted
The Netherlands: The Diagnose Behandeling Combinaties 427
to 1 709 000 at university hospitals and 9 125 000 at general hospitals, with
average length of stay (ALOS) durations of 7.3 and 5.5 days, respectively. The
number of hospital admissions increased while the ALOS has decreased in
recent years. This is largely due to an increase in the number of day-care
admissions (38 per cent between 2005 and 2009). The number of fi rst outpatient
visits increased by 10 per cent between 2005 and 2009.
Although day care and outpatient visits were traditionally only provided by
hospitals, competition between health care providers is now encouraged by
allowing ZBCs free access to the hospital care market. Figure 23.1 depicts the
role of hospitals and ZBCs in the delivery of hospital services. Whereas hospitals
Table 23.1 Key fi gures for university and general hospitals in 2009
University hospitals General hospitals
Number of hospitals 8 85
< 200 beds 0 9
200–300 beds 0 15
300–400 beds 0 22
400–600 beds 0 18
> 600 beds 8 21
Inpatient admissions * 1000 235 1 653
Inpatient days * 1000 1 709 9 125
Inpatient stay duration 7.3 5.5
Day-care admissions * 1000 226 1 627
Outpatient visits * 1000 3 142 24 257
Source: Kiwa Prismant, 2010.
Figure 23.1 Role of hospitals and ZBCs in the delivery
of hospital services
428 Diagnosis-Related Groups in Europe
provide the whole spectrum of day care and outpatient care, ZBCs usually only
provide straightforward non-acute day care and outpatient care, which requires
cooperation between at least two medical specialists working on a non-profi t-
making basis. ZBCs deliver care included in the basic health insurance (ZVW),
but also engage in services covered by VHI. In order to remain competitive over
the years, many hospitals have established ZBCs. These treatment centres
only account for less than 3 per cent of expenditure in the hospital sector, but
the number of ZBCs has rapidly increased from 79 in 2005 to 195 in 2009 (Kiwa
Prismant, 2010). Private clinics provide private medical specialist care, which is
not covered by the social health insurance scheme. They are not included in
the scope of this chapter.
Structural reforms of the health care sector in recent years have entailed
substantial changes in the fi nancing and budgeting of health care providers.
Hospitals in the Netherlands are independent and are contracted by health
insurers through either collective or selective contracts. Before 2005, budgeting
and fi nancing systems were mainly targeted towards controlling health care
expenditure. Incentives to increase production or to produce health services
more effi ciently were mainly absent (Oostenbrink & Rutten, 2006). In order to
provide stronger incentives for effi ciency and quality, a new system for the
payment of hospitals and ZBCs was introduced in February 2005. The new
system relies on a self-developed system of diagnosis–treatment combinations
(Diagnose Behandeling Combinaties; DBCs) as the basis of payment for care
provided by medical specialists and hospitals.
23.1.3 Purpose of the DBC system
The main purpose of the introduction of the DBC system was to reform
hospital payment to facilitate negotiations (in particular on quality) between
purchasers and providers by defi ning the products of hospitals (that is, DBCs)
(van Ineveld et al., 2006; van de Ven & Schut, 2009). DBCs were believed to
provide a concise defi nition of hospital products as the basis for selective con-
tracts. However, only a small selection of DBCs (list B DBCs) were freely nego-
tiable when the new system was introduced. For the majority of DBCs (list A
DBCs), hospitals received a fi xed amount per treated case within the framework
of a collective contract. In the future, the Government aims to gradually
increase the share of list B DBCs to about 70 per cent, as it wishes to increase the
share of hospital services for which hospitals and providers can negotiate
regarding quality.
Since the introduction of DBCs in the Netherlands, benchmarking has
become increasingly important. Average resource-use profi les are calculated for
list A DBCs on the basis of resource-use and cost-accounting data collected in
Dutch hospitals. These resource-use profi les have become an important external
benchmark for individual hospitals. In addition, other benchmarking tools
have been developed; for example, the Association of Dutch Health Insurers
annually publishes a guide containing hospital performance indicators relating
to list B DBCs, to support its members.
The Netherlands: The Diagnose Behandeling Combinaties 429
23.2 Developing and updating the DBC system
23.2.1 The DBC system at a glance
There is only one national DBC system in the Netherlands, which is centrally
regulated and monitored by ‘DBC onderhoud’ (DBC-O), a governmental
institution specifi cally set up for that task. The system is used to enable DRG-
type payment of all hospitals and ZBCs in the country, including payment of
psychiatric care services since 2008 and rehabilitation care at hospitals and
rehabilitation centres since 2009.
In contrast to DRGs in other countries, most DBCs stretch from the fi rst
contact with a medical specialist to treatment completion (Steinbusch et al.,
2007). These DBCs, referred to as ‘regular care’ DBCs, could include one or more
inpatient admissions in addition to several outpatient visits and post-discharge
follow-up care during the same year.
Next to ‘regular care’ DBCs, two other important types of DBC exist (Figure
23.2). The fi rst type, referred to as ‘continuation of regular care’ DBCs, is opened
to replace a ‘regular care’ DBC when treatment exceeds 365 days. The second
type, referred to as ‘inpatient without days’ DBCs, is opened in addition to a ‘regu-
lar care’ DBC when a patient requires treatment which is medically not related
to the ‘regular care’ DBC for which they are initially admitted. For example, a
patient admitted for chronic non-specifi c lung disease could require an appen-
dectomy. In this case, a ‘regular care’ DBC is opened for lung disease and an
‘inpatient without days’ DBC for appendectomy. ‘Inpatient without days’ DBCs
narrowly defi ne specifi c hospital stays similar to those defi ned by other DRG
systems.
DBCs belong to one of two lists: currently, about 67 per cent of DBCs belong
to list A and 33 per cent to list B. List B DBCs are supposed to comprise high-
incidence cases with suffi ciently homogeneous resource-consumption patterns,
such as hip and knee replacement, diabetes mellitus, cataract and inguinal
hernia repair. Hospital payment is different for list A DBCs and list B DBCs (see
subsection 23.5.2).
Figure 23.2 Fictional overview of types of DBCs
430 Diagnosis-Related Groups in Europe
Table 23.2 Traditional DRGs versus DBCs
Difference Typical DRG systems DBC system until 2010
Defi ned hospital product One hospital admission or
outpatient contact
One diagnosis–treatment
combination (may include
several hospital admissions
or outpatient contacts)
Number of DRGs/DBCs
per patient
One per patient (but
exceptions exist)
Several per patient
Level of detail/precision
of the system
Aggregated system < 3000
DRGs
Detailed system > 30 000 DBCs
Selection of DRG/DBC Assigned by computerized
grouping algorithm after
hospital discharge
Medical specialist opens DBC
upon fi rst diagnosis
Table 23.2 presents some of the main differences between typical DRG
systems and the DBC system. While DRG systems generally defi ne different
types of hospital admissions or outpatient contacts, the DBC system defi nes
different types of diagnosis–treatment combinations. Consequently, all hospi-
tal services related to this diagnosis–treatment combination during one year are
included in only one DBC. While DRG systems typically assign one DRG per
patient according to the most important diagnosis or procedure, the DBC sys-
tem uses several DBCs per patient if several diagnoses require treatment. As
opposed to DRG-based systems, which often consist of between 600 and 2000
DRGs, the DBC system currently comprises about 30 000 DBCs.
In addition, the current DBC system does not entail a computerized grouping
algorithm. The medical specialist decides which DBC is applicable and manually
opens this DBC upon fi rst diagnosis. It is possible to change the DBC registration
during the treatment process. However, a new generation of DBCs is forthcoming
in which some aspects of traditional DRG systems are covered (see subsection
23.8).
23.2.2 Development of the DBC system
In the late 1990s, a simplifi ed version of the All Patient (AP)-DRG system was
tested at six pilot hospitals in the Netherlands to examine the extent to which
the system was able to refl ect Dutch medical specialist and hospital care patterns.
Given the growing importance of outpatient care in the Dutch health care sys-
tem, the inability of the AP-DRG system to adequately account for outpatient
cases was seen as a major defi cit of the system. Furthermore, since patients were
grouped by administrative staff members after hospital discharge, rather than by
medical specialists, interpretational differences and mistakes were perceived to
be problematic (Custers et al., 2007; Zuurbier & Krabbe-Alkemade, 2007). There-
fore, health insurers and hospitals initiated the development of DBCs. Medical
specialists’ associations defi ned DBCs for each medical specialty. A representative
sample of 23 ‘frontrunner’ hospitals registered detailed resource-use and cost data
for all inpatient and outpatient hospital services according to the DBC system.
The Netherlands: The Diagnose Behandeling Combinaties 431
DBC tariffs comprise two separate components (Beersen et al., 2005; Zuurbier
& Krabbe-Alkemade, 2007): (1) the honorarium component for the payment of
specialists; and (2) the hospital cost component for the payment of all relevant
hospital services. For the calculation of the honorarium component, the ‘norm-
time’ was determined for each DBC. The ‘norm-time’ is supposed to refl ect the
time requirements of medical specialists to perform all relevant tasks related to
a DBC. The time was estimated from hospitals’ administrative databases and
validated by expert opinion. The ‘norm-time’ was then multiplied with a fi xed
fee per hour of €135.50 to calculate the honorarium component (Folpmers &
de Bruijn, 2004). With respect to list A DBCs, the hospital cost component was
determined based on the resource-use and cost data of the hospital services at
the 23 ‘frontrunner’ hospitals; average resource-use profi les were multiplied
with national unit costs (see subsection 23.4.2). Hospital services were cat-
egorized into 15 resource-use categories, as presented in Table 23.3. National
unit costs for these hospital services included wages, equipment, overheads
and – since 2009 – capital costs (see subsection 23.4.2). With respect to list B
DBCs, the hospital cost component results from negotiations between health
insurers and hospitals (see subsection 23.5.2).
Since February 2005, the DBC system has been continuously updated through
revisions and additions that are implemented without the defi nition of new
versions of the system. Table 23.4 shows some main facts relating to the DBC
system upon fi rst introduction (2005) and the current version (2010). At the
introduction of the DBC system, each diagnosis and treatment combination
was appointed one DBC for the fi rst outpatient visit only, and one DBC for all
related hospital services with the exception of the fi rst outpatient visit. The
number of DBCs amounted to about 100 000, of which about 90 per cent were
list A DBCs. List A DBC tariffs excluded capital costs.
In the current version of the DBC system, the classifi cation of patients has
been simplifi ed. The number of DBCs has been substantially reduced from
about 100 000 to 30 000, of which about 67 per cent relate to list A DBCs. Each
Table 23.3 Hospital services resource-use categories
Inpatient days
Intensive care days
Day-care hours
Outpatient and emergency room visits
Laboratory services
Medical imaging services
Medical devices
Surgical procedures
Diagnostic activities
Microbiological and parasitological services
Pathological services
Blood products
Paramedical and supportive services
Rehabilitation services
Other services
Source: DBC-O, 2011.
432 Diagnosis-Related Groups in Europe
diagnosis and treatment combination is now appointed one single DBC cover-
ing all related hospital services, including the fi rst outpatient visit. In addition,
DBCs were rearranged, for example, by reducing the number of categories to
describe the ‘type of care’ and ‘treatment’ dimensions (see subsection 23.3.2).
The hospital cost component for list A DBC tariffs is currently determined from
detailed resource-use profi les of all hospitals and cost data derived from 15–25
‘frontrunner’ hospitals. In addition, the hospital cost component now includes
capital costs.
An increasing share of DBCs is progressively being moved from list A to list
B, which is in line with the original purpose of the DBC system (see subsection
23.1.3). There are six main criteria which must be met by a list A DBC in order
for it to be transferred (DBC-O, 2009). The DBC must: (1) be is characterized by
suffi ciently homogeneous levels of resource consumption; (2) have a suffi ciently
high volume of cases; (3) be suffi ciently spread amongst health care providers;
(4) involve predictable non-acute care. In addition, (5) the transfer must be
supported by medical specialists and hospitals; and (6) all list A DBCs defi ned
on the basis of the same diagnosis must meet these criteria.
23.2.3 Data used for development and
updates of the DBC system
Regarding the aforementioned honorarium component, the ‘norm time’ is
updated based on time studies and validated by expert opinion (Oostenbrink &
Rutten, 2006). The fi xed fee per hour is set by the NZa.
The hospital cost component of list A DBCs is determined and updated by
Table 23.4 The main facts relating to the DBC versions, at its introduction (2005) and
the current version
1st DBC version Present DBC version
Date of
introduction
2005 2010
(Main) Purpose Hospital payment Hospital payment, benchmarking
Source Self-developed Self-developed
Data used for
development
Resource use and unit costs of
23 ‘frontrunner’ hospitals
Resource use of all hospitals; unit
costs of 15–25 ‘frontrunner’
hospitals
Services included Whole spectrum of inpatient
and outpatient care, excluding
psychiatric and rehabilitation
care
Whole spectrum of inpatient
and outpatient care, including
psychiatric and rehabilitation
care
Cost categories
included
Recurrent costs, excluding costs
of education, teaching,
research and commercial
exploitation
Recurrent costs and capital costs,
excluding costs of education,
teaching, research and
commercial exploitation
Number of DBCs ± 100 000; list A: 90%; list B:
10%
± 30 000; list A: 67%; list B: 33%
Applied to All hospitals and ZBCs All hospitals and ZBCs
The Netherlands: The Diagnose Behandeling Combinaties 433
DBC-O on the basis of a database that is maintained by a subdivision of DBC-O,
called the ‘DBC information system’ (DBC-DIS). The database contains two data-
sets: (1) resource-use information from the minimum basic datasets (MBDS)
collected by all hospitals; and (2) unit cost information from a varying number
of 15–25 ‘frontrunner’ hospitals (see subsection 23.4.2). Figure 23.3 depicts the
data-collection process from medical specialists to the national database at the
DBC-DIS. From the opening of a DBC by a medical specialist, resource use per
DBC and per treated case is collected and integrated into one hospital database.
The registration system also records the DBC for which a hospital service is
performed. After integration of the data at DBC-DIS, technical feedback is pro-
vided to medical specialists to assure high-quality data.
23.2.4 Regularity and method of system updates
Regularity and method of updating the DBC classifi cation system
Medical specialists’ associations notify DBC-O when problems arise in classify-
ing DBCs, as DBC-O is responsible for the irregular but continuous updating of
the DBC classifi cation system. DBC-O is also the gatekeeper for innovation in
the DBC system (see section 23.6). DBCs may be merged, split or created.
Examples include the recent reduction in the number of DBCs (see subsection
23.2.2) and the introduction of a new generation of DBCs (see section 23.8).
Updating is based on feedback from medical specialists’ associations and
information from the national DBC-DIS database.
Regularity and method of updating tariffs
As already mentioned, the norm-time relating to the honorarium component is
updated at irregular intervals. The fi xed fee per hour is re-examined annually
and updated when necessary. The hospital cost component of list A DBCs is
recalculated annually, or as necessary, by multiplying the average resource-use
profi le and national unit costs. An example of a fi ctional resource-use profi le
for a specifi c DBC (‘surgery/ regular care// arthrosis knee/ surgery with clinical
episode’) is provided in Table 23.5. The calculation of unit costs per service is
described in section 23.4.
There is always a time-lag of at least two years between the year of the data
and the year of application of tariffs in hospitals. For example, hospital resource-
use and cost data from the year 2009 will be analysed during the years 2010 and
2011 in order to defi ne the DBC hospital cost component that will be used for
hospital payment in 2012.
23.3 The current patient classifi cation system
23.3.1 Information used to classify patients
A medical specialist is consulted to decide which DBC is applicable and (s)he
manually opens the DBC upon fi rst diagnosis by specifying fi ve types of
F
ig
u
re
2
3
.3
D
at
a-
co
ll
ec
ti
o
n
p
ro
ce
ss
f
o
r
th
e
n
at
io
n
al
d
at
ab
as
e
The Netherlands: The Diagnose Behandeling Combinaties 435
information called ‘dimensions’ of a DBC (van Beek et al., 2005): (1) medical
specialty; (2) type of care; (3) demand for care; (4) diagnosis; (5) treatment axis
(setting and nature).
Thus, the information used to classify patients includes clinical and resource-
use data. The DBC system does not distinguish between principal and sec-
ondary diagnoses. If a patient has a second diagnosis that requires treatment,
this second diagnosis will be classifi ed into a separate DBC (see subsection
23.2.1).
23.3.2 Classifi cation algorithm
Classifi cation of patients follows the order of the fi ve dimensions: (1) the medical
specialty is specifi ed through a four-digit code; (2) a two-digit code for the type
of care is added to the fi rst four digits; (3) the demand for care is indicated for
certain medical specialties; (4) the diagnosis is specifi ed by adding another three-
digit code; and (5) the treatment axis is defi ned by the last three digits of the
DBC. An illustrative example of the patient classifi cation for patients with
appendicitis treated in a surgery department is provided in Table 23.6.
Medical specialty
Patients can be classifi ed into one of 27 medical specialties (codes 0301 to
1900). For patients with appendicitis treated in a surgery department, the ‘medi-
cal specialty’ code would be 0303//// (surgery////).
Table 23.5 Fictional average resource-use profi le for the DBC ‘surgery regular care/
arthrosis knee surgery with clinical episode’
Hospital services % of patients
receiving this
hospital service
Average resource-use
for patients receiving
this hospital service
Average
resource-use for
all patients
Inpatient days 100 6.0 6.0
Outpatient visits 100 8.0 8.0
Laboratory services 100 1.0 1.0
Medical imaging services
X-ray thorax 50 1.0 0.5
X-ray knee/lower leg 100 2.0 2.0
X-ray hip joint 50 1.0 0.5
MRI hip/lower leg 10 1.0 0.1
Surgical procedures
Surgery dislocation 100 1.0 1.0
Paramedical and supportive
services
Physiotherapy 100 2.0 2.0
Source: Zorgverzekeraars Nederland, 2004.
Notes: For instance, 50 per cent of the patients received X-ray thorax examinations; the
average number of X-ray thorax examinations for these patients was 1.0; the average number
of X-ray thorax examinations for all patients was 0.5.
436 Diagnosis-Related Groups in Europe
Type of care
Currently, two categories are used to describe the ‘type of care’ dimension:
‘regular care’ (code 11) and ‘continuation of regular care’ (code 21). For patients
with appendicitis, the code would be 0303/11/// (surgery/ regular care///).
Demand for care
The ‘demand for care’ dimension is only used for a limited number of medical
specialties (namely, plastic surgery, urology, gastroenterology and radiotherapy).
The dimension specifi es demand for care which is expected to result in higher
than average resource consumption. For the medical specialty ‘plastic surgery’,
the ‘demand for care’ dimension distinguishes ‘≥ two procedures in the same
surgical area’, ‘extensive crush injury within the surgical area’, ‘congenital im-
pediments within the surgical area’, ‘requirement of a second surgeon’ and
‘children ≤ 10 years of age’.
Diagnosis
The ‘diagnosis’ dimension describes the diagnosis of the patient in medical
terms. The classifi cation of diagnoses is based on the International Classifi cation
of Diseases 10th revision (ICD-10) coding, even though the ICD-10 codes are not
used in the codifi cation of DBCs. For patients with appendicitis, the ‘diagnosis’
code would be 0303/11//113/ (surgery/ regular care// appendicitis/).
Table 23.6 The patient classifi cation system logic: Surgery example
Medical
specialty
Type of care Demand
for care
Diagnosis Treatment axis
0303
Surgery
11 Regular care Not applicable 113
Appendicitis
201 Open-surgery
outpatient
21 Continuation
of regular care
202 Open-surgery in day
care
203 Open-surgery with
clinical episode(s)
204 Single outpatient
with procedure
206 Inpatient without
days Open-surgery
with clinical episode(s)
301 Endo-surgery
outpatient
302 Endo-surgery in day
care
303 Endo-surgery with
clinical episode(s)
306 Inpatient without
days Endo-surgery
with clinical episode(s)
The Netherlands: The Diagnose Behandeling Combinaties 437
Treatment axis
The ‘treatment axis’ dimension expresses the ‘treatment setting’ and ‘treat-
ment nature’. The ‘treatment setting’ is either ‘outpatient’, ‘in day care’ or ‘with
clinical episode(s)’. The subdivision of ‘treatment nature’ varies by medical
specialty and may, for instance, specify whether treatment concerns an ‘open-
surgery’ or a laparoscopic procedure. The number of treatment axes varies from
6 for the medical specialties ‘gastroenterology’ and ‘paediatrics’ to over
60 for the medical specialty ‘internal medicine’. For patients presenting with
appendicitis, the ‘treatment axis’ code could, for example, be:
0303/11//113/201 (surgery/ regular care// appendicitis/ open-surgery out-
patient); 0303/11//113/202 (surgery/ regular care// appendicitis/ open-surgery
in day care); or 0303/11//113/203 (surgery/ regular care// appendicitis/ open-
surgery with clinical episode(s)).
23.3.3 Data quality and plausibility checks
The DBC-DIS performs data quality and plausibility checks relating to develop-
ing and updating the DBC system. These annual checks take place at the
national level and comprise the technical validation of the information from
the MBDS in the national database (technical correctness, comprehensiveness
and functional correctness). There is no system of external data audits.
23.3.4 Incentives for up- or wrong-coding
Although up-coding has been described as a potential threat to the DBC system,
the Dutch system seems to be less sensitive to up-coding compared to DRG
systems in the United States and Australia (Steinbusch, 2007). The relative
strength of the Dutch system is related to the use of classifi cation criteria that
are aligned with clinical practice, the fact that DBCs are opened upon diagnosis,
and the fact that hospitals generally operate as non-profi t-making institutions.
23.4 Cost accounting within hospitals
23.4.1 Regulations
Cost accounting is not mandatory for the majority of Dutch hospitals, which
only provide their MBDS to the DBC-DIS. However, the 15–25 ‘frontrunner’ hos-
pitals must follow a uniform product costing model, which was developed dur-
ing the DBC system’s introductory period (Zuurbier & Krabbe-Alkemade, 2007).
23.4.2 Main characteristics of the cost-accounting system
All frontrunner hospitals have to allocate all relevant hospital costs to individual
hospital services. Relevant hospital costs include wages, equipment, overheads
438 Diagnosis-Related Groups in Europe
and capital costs (see subsection 23.2.2). Hospital costs relating to education,
teaching, research and commercial exploitation are not considered relevant
because they are not fi nanced by the DBC system.
Allocating relevant hospital costs from support cost centres to fi nal cost centres
Hospital departments producing hospital services are called ‘fi nal cost centres’.
These include, among others: inpatient and outpatient clinics, laboratories,
operating rooms (ORs) and radiology departments. Departments not providing
patient care are called ‘support cost centres’. These include, among others:
departments for administration, personnel, billing, communications, fi nance,
security and availability in case of emergencies. Costs of support cost centres
may also be referred to as overheads.
In the fi rst step, relevant hospital costs are allocated from support cost centres
to fi nal cost centres. Hospitals are free to choose the allocation method for the
assignment of hospital costs from support cost centres to fi nal cost centres. As
the allocation method was found to have only a minor impact on individual
patient’s costs (Zuurbier & Krabbe-Alkemade, 2007), hospitals commonly use
simple direct allocation, in which the costs of support cost centres are assigned
to the fi nal cost centres without interaction between support cost centres
(Finkler et al., 2007; Horngren et al., 2005). The product costing model contains
specifi cations regarding the allocation base to be used for each cost centre; for
example, the area (m2) to allocate costs of accommodation, or the number of
full-time equivalents to allocate the costs of administration.
Allocating relevant hospital costs from fi nal cost centres to hospital services
Once the costs of support cost centres are assigned to fi nal cost centres, the total
costs of each fi nal cost centre can be assigned to individual hospital services,
such as inpatient days, intensive care days, laboratory services, medical imaging
services and surgical procedures (see Table 23.3). Weighting statistics are used to
assign relevant hospital costs from fi nal cost centres to hospital services. They
differ between fi nal cost centres according to the type of service they produce.
An example of such a weighting statistic is the average time of surgical inter-
ventions to distribute the cost of the fi nal ‘OR’ cost centre to these interven-
tions. The NZa determines the national unit costs of about 4500 hospital
services from the weighted average across the 15–25 ‘frontrunner’ hospitals.
National unit costs are determined with a lag-time of at least two years. The
tariffs for 2012 will be based on the national unit costs of 2009.
23.5 DBCs for hospital payment
23.5.1 Range of services and costs included in
DBC-type hospital payment
Inpatient and outpatient hospital care of all hospitals and ZBCs (including
psychiatric and rehabilitation care) is fully fi nanced according to the DBC
system logic. One exception concerns some very expensive and orphan drugs
The Netherlands: The Diagnose Behandeling Combinaties 439
for which the NZa provides hospitals with additional funding (80 per cent of
the purchase price for expensive drugs and 100 per cent of the purchase price
for orphan drugs) (Rodenburg-van Dieten, 2005). Other relevant sources of
fi nancing for hospitals exist but do not relate to hospital care, such as education,
teaching, research and commercial exploitation. These sources accounted for
15.9 per cent of total hospital revenues in 2009 (Kiwa Prismant, 2010).
23.5.2 Calculation of DBC tariffs
DBC tariffs consist of two parts: (1) the honorarium component and (2) the hospi-
tal cost component. The honorarium component is calculated on the basis of a
‘norm-time’ and a fi xed fee per hour both for list A and for list B DBCs (see
subsections 23.2.2 to 23.2.4). For list A DBCs, the hospital cost component is
calculated on the basis of average resource-use profi les from all hospitals and
unit costs calculated through the product costing model described in subsec-
tion 23.4.2. A fi ctional example to illustrate the cost calculation of the hospital
cost component for the DBC ‘surgery/ regular care// appendicitis/ surgery with
clinical episode’ is provided in Table 23.7.
The tariff for the hospital cost component of list B DBCs is negotiated between
hospitals and insurers. Insurers are not obliged to contract all hospitals for list
B DBCs, and may employ different DBC prices for different hospitals. Like-
wise, hospitals may negotiate different prices for the same DBC with different
insurers. Health insurers and hospitals determine the frequency and terms of
agreements. Current practice suggests that negotiations take place annually,
but that either party can reopen negotiations if required by the circum-
stances (van Ineveld et al., 2006). Examples of such circumstances include long
waiting lists, increased public attention to a specifi c health problem or the
introduction of very expensive/orphan drugs or medical devices.
23.5.3 DBCs in actual hospital payment
All hospitals in the Netherlands receive a nationally uniform payment per list A
DBC and a negotiated hospital-specifi c payment for list B DBCs. In order to re-
ceive payments under the DBC system, hospitals classify all patients into the
appropriate DBCs. After treatment is completed, a bill is sent to the patients’
health insurer indicating all relevant DBCs. Subsequently, the insurer pays hospi-
tals on the basis of the fi xed list A DBC tariffs or the negotiated list B DBC tariffs.
For list A DBCs, prospective budgets determine the total fi nancial volume
which hospitals can earn through the provision of DBCs. Budgets are estab-
lished annually by the NZa based on fi xed and variable costs and a variety of
parameters, including the hospital’s adherent population, the type of facilities,
the number of beds and production parameters (such as the number of in-
patient days and outpatient visits) (Nederlandse Zorgautoriteit, 2009). Hos-
pitals are fully compensated for the difference between the prospective budget
and DBC payments (yield). Consequently, higher production may result in
higher costs without additional yield, while lower production results in lower
costs but not in lower yield.
440 Diagnosis-Related Groups in Europe
For list B DBCs, insurers may limit the maximum volume of list B DBCs that a
hospital is allowed to produce. That aside, insurers and hospitals may agree upon
a lower or higher DBC price if production exceeds a predetermined fi gure. The
hospital’s and medical specialists’ yield only depends on DBC payments.
Consequently, higher production may result in higher costs and additional yield,
while a lower level of production directly results in lower costs and lower yield.
The DBC system also applies to ‘non-contracted care’; that is, care provided
to foreign patients, uninsured patients or patients whose health insurer does
not have a contract with the hospital. In these situations, the foreign insurer
or the patient must pay the DBC tariff. The tariffs for the honorarium com-
ponent of list A and B DBCs and for the hospital cost component of list A
DBCs are the same both for non-contracted and contracted care. The tariffs
for the hospital cost component of list B are determined by the hospital and
may differ between contracted and non-contracted care. Hospitals do not have
Table 23.7 Fictional cost calculation of the hospital cost component
Hospital services Total resource
use for all
patients
National
unit
costs (€)
Total
costs (€)
Average
costs per
patient (€)a
Inpatient days 1 250 296 370 083 1 341
Outpatient visits 864 43 37 147 135
Day-care hours 1 029 34 35 002 127
Laboratory services
Urine screening 560 2 1 121 4
Ureum 836 1 836 3
Creatinine 974 2 1 949 7
Leucocytes 781 1 781 3
Medical imaging services
X-ray thorax 615 52 32 005 116
X-ray abdomen 781 52 40 616 147
CT abdomen 144 228 32 723 119
Echo abdomen 281 83 23 320 84
Surgical procedures
Appendectomy 276 548 151 248 548
Colon resection 8 1 595 13 207 48
Small intestinal resection 11 1 056 11 658 42
Resection appendicular abscess 6 761 4 201 15
Diagnostic activities
Diagnostic laparoscopy 41 484 20 038 73
Diagnostic duodenoscopy 14 408 5 630 20
Cysto-/urethrography 6 479 2 644 10
Microbiological and parasitological
services
856 33 28 235 102
Paramedical and supportive services
Physiotherapy 500 31 15 486 56
TOTAL 827 929 3 000
Source: Zuurbier & Krabbe-Alkemade, 2007.
Notes: For the list A DBC: ‘surgery/ regular care// appendicitis/ open-surgery with clinical
episode’; aThe average costs per patient add up to the DBC tariff ‘~ number of closed DBCs: 276’.
The Netherlands: The Diagnose Behandeling Combinaties 441
to publish tariffs for contracted care, whereas they are obliged to publish tariffs
for non-contracted care.
23.5.4 Quality-related adjustments
For list A DBCs, no quality related adjustments exist. The tariff is the same for
all hospitals, regardless of quality. Although the negotiations on list B DBCs
were intended to be based on the quality of delivered care, insurers and hospitals
currently predominantly negotiate on price and/or production volume (see
subsection 23.7).
23.5.5 Main incentives for hospitals
Hospitals are incentivized to keep their costs below the national unit costs for
any specifi c list A DBC. For list B DBCs, hospitals are incentivized to keep costs
below negotiated prices. The DBC system therefore offers hospitals an incentive
to improve those quality aspects that lead to lower resource consumption.
For example, it encourages quality improvements that would lead to fewer
unnecessary diagnostic services and to a reduction in the ALOS (Custers et al.,
2007).
Quality improvement aimed at reducing complication rates – such as post-
operative infections and/or readmission rates – are not stimulated by the DBC
system, because the occurrence of complications might lead to a new DBC
(Custers et al., 2007). Hospitals could even be incentivized to accept a price
below the costs of production for a specifi c list B DBC, in order to gain a contract
with an insurer, and could then try to compensate for the losses by providing
profi table list A DBCs to these patients.
23.6 New/innovative technologies
23.6.1 Steps required prior to usage in hospitals
DBC-O is the gatekeeper for innovation in the DBC system. Current regulations
require a process of seven steps following an application (for example, from a
hospital) before a new technology can be included in the DBC system (VWS,
2009), as detailed here.
1. DBC-O assesses the admissibility, completeness, nature, size and complexity
of the application.
2. The CVZ performs a systematic literature review to examine the extent and
level of evidence supporting the specifi c technology.
3. DBC-O assesses the costs, effectiveness, ethical aspects, patient preferences
and system consequences of the application.
4. Based on the information acquired from steps 2 and 3, DBC-O decides upon
the implementation of the technology in the DBC system.
442 Diagnosis-Related Groups in Europe
5. The positive decision by DBC-O is approved by the NZa.
6. The CVZ advises the VWS whether the new technology should be made part
of the insurance benefi ts package.
7. Finally, DBC-O incorporates the new technology into the DBC system.
The seven steps should take no longer than six months from registration of
the new treatments (VWS, 2009). At fi rst introduction of the new technology in
the DBC system, average resource-use profi les are not yet available and DBC
tariffs are based on expert opinion. For the DBC system until 2010, 24 new
technologies have been assessed by DBC-O, the NZa and the CVZ. Seven led to
new DBCs, four have been merged with existing DBCs, and fi ve were not
approved. Eight are still under consideration.
23.6.2 Funding
Currently, new or innovative treatments are introduced into the DBC system
twice a year. Until the new technology is incorporated in the DBC system,
additional payments exist only for innovative drugs. Since 2006, an innovative
drug can be provisionally included on the ‘list of expensive drugs’ or ‘list of
orphan drugs’ for four years, on the conditions that: (1) added therapeutic
value is demonstrated; (2) its expenses account for over 0.5 per cent (‘expensive
drugs’) or 5.0 per cent (‘orphan drugs’) of the annual hospital drugs budget; and
(3) a plan for the assessment of cost–effectiveness in daily clinical practice is
approved by the pharmaceutical advisory committee.
23.7 Evaluation of the DBC system in the Netherlands
The main purpose of introducing DBCs was to enable price and quality
negotiations between insurers and providers. Although these negotiations were
intended to be based on the quality of delivered care, insurers and hospitals
currently predominantly negotiate on price and/or production volume. Since
2006, prices for list B DBCs have increased at a lower rate than those for list A
DBCs and the health insurers increasingly apply pressure to hospitals to charge
even lower prices (van de Ven & Schut, 2009). Table 23.8 depicts the negotiated
tariffs in 2007 compared to those in 2004 for seven list B DBCs at four health
insurers. List B DBC prices had increased by about 8 per cent in 2007, compared
to 2004 tariffs. In general, major price deviations only occurred for a minority
of DBCs. More complex and chronic DBCs seem to be less sensitive to market
competition. Evidence from recent years suggests that hospitals negotiate on
the total budget of the total B segment, rather than on the individual DBC level
(van Ineveld et al., 2006).
Insurers have been reluctant to selectively contract with hospitals and to
offer preferred hospital contracts to their customers. Aside from the problems of
having the right mix of criteria to determine quality, obtaining accurate data,
and doing so in a timely manner, there are several limitations for Dutch health
insurers that limit their interest in negotiating on quality and to selectively
The Netherlands: The Diagnose Behandeling Combinaties 443
contract with higher quality hospitals (Custers et al., 2007; van de Ven & Schut,
2009).
• Health insurers are afraid of acquiring a bad reputation if they restrict
consumer choice to a limited network of preferred hospitals.
• Patients assume that the quality of care in terms of effectiveness and safety is
equal among all hospitals. As a result, insurers have no incentive to negotiate
for higher quality (and to pay higher prices) if patients do not appreciate
higher quality in contracted hospitals.
• Furthermore, a ‘free-rider’ problem exists: hospitals have contracts with
several insurers. If one single insurer motivates a particular hospital to
improve quality, all of this hospital’s patients will benefi t from the quality
improvement, including patients who are insured through other insurers.
• Finally, if an insurer acquires recognition for providing high-quality care, it is
likely to enrol a disproportionate share of patients with chronic medical
problems.
Unfortunately, information necessary to evaluate the DBC system is not easily
accessible. A lot of information is available in the national DBC-DIS database
but, at present, only a limited number of actors have access to the database.
23.8 Outlook: Future developments and reform
A new generation of DBCs – the so-called ‘DBCs towards transparency’ (‘DBCs
Op weg naar Transparantie’) – is forthcoming. In the new system, patients will be
classifi ed according to a computerized grouping algorithm (see Figure 23.4).
The number of DBCs will be substantially reduced from about 30 000 to 4000
by discarding the ‘medical specialty’ dimension. In addition, expensive/orphan
drugs, intensive care and other products are to be accounted for by means
of treatment related ‘add-ons’, each with their own tariff. Other products
Table 23.8 Negotiated tariffs in 2007 compared to those of 2004
N 2004
tariff (€)
Average
2007
tariff (€)
Relative
price
increase (%)
Minimum
2007 price
Maximum
2007 price
Inguinal hernia
repair 407 2 163 2 254 4.2 1 529 3 088
Diabetes 410 409 483 18.1 385 1 027
Tonsillectomy 409 740 800 8.1 433 1 498
Cataract 407 1 317 1 381 4.8 1 044 1 599
Hip replacement 409 8 561 9 097 6.3 7 603 11 370
Knee replacement 404 10 228 10 746 5.1 9 097 13 000
Spinal disc
herniation 354 3 046 3 308 8.6 2 413 5 778
Source: Nederlandse Zorgautoriteit, 2005.
Note: Example of seven list B DBCs at four health insurers.
444 Diagnosis-Related Groups in Europe
may concern transmural/shared care; namely, hospital services provided in
cooperation with medical professionals outside of the hospital (for example,
the GP).
Another important feature of the future grouping algorithm is the possibility
to consider care intensity for the classifi cation of patients. For example, separate
DBCs could be defi ned for an inpatient stay of up to fi ve days and for over fi ve
days. The grouping algorithm is currently being tested, but it is not yet clear
when it will be implemented nationwide.
Another future development concerns the transition to a situation in which
prospective budgets are solely determined based on production parameters,
such as fi rst outpatient visits, fi rst admissions, the number of inpatient days and
day-care hours. The transition phase started early 2010 and is expected to last
at least three years (Nederlandse Zorgautoriteit, 2009).
Figure 23.4 The grouping algorithm for the new generation of DBCs
Notes: EOD: ‘add-on’ expensive/orphan drugs.
The Netherlands: The Diagnose Behandeling Combinaties 445
It is too early to predict the potential effects of the future developments of
the DBC system. Most signifi cantly, the importance of negotiations between
hospitals and insurers is bound to increase.
23.9 References
Beersen, N., Redekop, W.K., de Bruijn, J.H.B. et al. (2005). Quality-based social insurance
coverage and payment of the application of a high-cost medical therapy: the case of
spinal cord stimulation for chronic non-oncologic pain in the Netherlands. Health
Policy, 71:107–15.
Custers, T., Arah, O.A., Klazinga, N.S. (2007). Is there a business case for quality in the
Netherlands? A critical analysis of the recent reforms of the health care system. Health
Policy, 82:226–39.
DBC-O (2009). Scenario’s invulling B-segment 2011 [B-Segment Composition Scenarios].
Utrecht: DBC-Onderhoud.
DBC-O (2011). DBC Onderhoud Diagnose Behandeling Combinatie. Utrecht: DBC-Onderhoud
(www.dbconderhoud.nl, accessed 10 July 2011).
Enthoven, A.C., van de Ven, W.P.M.M. (2007). Going Dutch – managed competition health
insurance in the Netherlands. New England Journal of Medicine, 357(24):2421–3.
Finkler, S.A., Ward, D.M., Baker, J.J. (2007). Essentials of Cost Accounting for Health Care
Organizations. Third edition. New York, NY: Aspen Publishers.
Folpmers, M., de Bruijn, J. (2004). Honorering in de Zorg, Overgang naar de nieuwe
landelijke DBC-systematiek [Remuneration in Health Care, Transition towards the
New National DBC Systematic]. MCA Tijdschrift voor Organisaties in Control, 6:18–25.
Horngren, C.T., Datar, S.M., Foster, G. (2005). Cost Accounting: A Managerial Emphasis.
Twelfth edition. New York, NY: Prentice-Hall, Englewoord Cliffs.
Kiwa Prismant (2010). Kengetallen Nederlandse Ziekenhuizen 2009 [Key Numbers Dutch
Hospitals 2009]. Utrecht: Dutch Hospital Data.
Nederlandse Zorgautoriteit (2005). Oriënterende monitor ziekenhuis zorg: analyse van de
onderhandelingen over het B-segment in 2005 [Exploratory Monitoring of Hospital Care:
Analyses of the Negotiation on the B-Segment in 2005]. Utrecht: Nederlandse Zorgautoriteit.
Nederlandse Zorgautoriteit (2009). Consultatiedocument Prestatiebekostiging binnen de
medisch specialistische zorg [Consultation Document on Remuneration According to
Performance within Medical Specialist Care]. Utrecht: Nederlandse Zorgautoriteit.
Oostenbrink, J.B., Rutten, F.F. (2006). Cost assessment and price setting of inpatient care
in the Netherlands. The DBC casemix system. Health Care Management Science, 9:
287–94.
Poos, M.J.J.C., Smit, J.M., Groen, J. et al. (2008). Kosten van ziekten in Nederland [Cost of
Illnesses in the Netherlands]. Bilthoven: National Institute for Public Health and the
Environment (RIVM).
Rodenburg-van Dieten, H.E.M. (2005). Richtlijnen voor farmaco-economisch onderzoek
[Guidelines for Pharmaco-Economic Research]. Diemen: Healthcare Insurance Board
(CVZ).
Schäfer, W., Kroneman, M., Boerma, W. et al. (2010). The Netherlands: health system
review. Health Systems in Transition, 12(1):1–228.
Schut, F.T., Hassink, W.H. (2002). Managed competition and consumer price sensitivity in
social health insurance. Journal of Health Economics, 21:1009–29.
Steinbusch, P.J.M., Oostenbrink, J.B., Zuurbier, J.J., Schaepkens, F.J.M. (2007). The risk of
up-coding in casemix systems: a comparative study. Health Policy, 81(2–3):289–99.
Stolk, E.A., Rutten, F.F.H. (2005). The ‘health benefi t basket’ in the Netherlands. European
Journal of Health Economics, 6:53–7.
446 Diagnosis-Related Groups in Europe
Van Beek, L., Goossen, W.T., van der Kloot, W.A. (2005). Linking nursing care to medical
diagnoses: heterogeneity of patient groups. International Journal of Medical Informatics,
74:926–36.
Van de Ven, W.P.M.M., Schut, F.T. (2009). Managed competition in the Netherlands: still
work-in-progress. Health Economics, 18:253–5.
Van Ineveld, M., Dohmen, P., Redekop, K. (2006). De startende marktwerking in de
gezondheidszorg [The starting market forces in health care]. Economisch Statistische
Berichten, 91(4494):470–3.
VWS (2009). Innovaties en DBC’s [Innovation and DBCs]. Letter from the Minister of Health,
Welfare and Sport to the Second Chamber, 24 March 2009. The Hague: Minister of
Health, Welfare and Sport.
Zorgverzekeraars Nederland (2004). DBC-inkoopgids 2005, Segment B: Zorgverzekeraars
onderhandelen over zorg [DBC Purchase Guide, Segment B: Health Insurers Negotiate on
Care]. Zeist: Zorgverzekeraars Nederland.
Zuurbier, J., Krabbe-Alkemade, Y. (2007). Onderhandelen over DBC’s [Negotiating on DBCs.
Second edition]. Maarssen: Elsevier Gezondheidszorg.
Index
3M™ Health Information System
Portugal 384, 387
Sweden 293
abbreviations xvii–xx
activity reports
indicator of effi ciency of hospitals 106–9
transparency 150–1
acute care hospitals
DRG-based hospital payments 80–2
Finland 322–5
France 222–3
hospital landscape 15–19
ownership types 19
Spain 403–4
adjustments, structural, DRG-based hospital
payments 78, 79–80
aims, this book’s 6–7
ALOS see average length of stay
asymmetry of information see information
asymmetry
Austria 175–95
acute care hospitals, fi nancing 178
classifi cation algorithm 184–5
cost accounting 186–7
data quality 185–6
DRG system
authors’ assessment 192–3
data used 182
development 179–82
evaluation 191–3
hospital fi nancing 187–91
offi cial evaluation 191–2
purpose 179
updating 179–82
DRGs in actual hospital payment
188–90
evaluation of LKF framework 191–3
future developments and reform 193
grouping algorithm 184–5
health care system 175–7
hospital fi nancing 187–91
hospital services 175–9
hospitals, defi ning 177
hospitals infrastructure 177–8
incentives 186, 190–1
information used to classify patients 183
LDF scores calculation 187–90
LDFs in actual hospital payment 188–90
LKF-PCS 179–93
patient classifi cation system (PCS) 183–6
plausibility checks 185–6
quality-related adjustments 190
range of services and costs 187
role of DRGs 175–9
self-developed DRG-like PCSs 46–8
technological innovation 191
up-coding 186
wrong-coding 186
average length of stay (ALOS)
acute care hospitals 16
quality of care 117, 118–19
448 Diagnosis-Related Groups in Europe
budgets
DRG-based budget allocation model,
Portugal 394–5
DRG-based budgetary adjustments, Ireland
284–6
global budgets
hospital payment models 95–6
hospital payment systems and incentives
14–15
building blocks
DRG-based hospital payments 28–30
DRG systems 24–7
DRGs 24–30
care quality see quality of care
case costing model, Sweden 348–50
Catalonia see Spain
categorization problems, effi ciency of
hospitals 109
CCs see complication and comorbidity
checks on reported cost data, cost accounting
67–8
classifi cation algorithms
Austria 184–5
DRG systems 204–6, 229–32
England 204–6
Estonia 309–10
European DRG-like PCSs 44–9
Finland 328–30
France 229–32
Germany 254–5
Ireland 280–1
Netherlands 435–7
Poland 369–71
Portugal 388–9
Spain 409, 410
Sweden 345, 346–7
classifi cation variables, European DRG-like
PCSs 49–51
coding of diagnoses and procedures
DRG systems 43–4
quality of care 121
transparency 151
up-coding 186
wrong-coding 186
Commissioning for Quality and Innovation
(CQUIN)
quality of care 122, 154
quality-related adjustments 213
common EuroDRG system, future 166–9
common system likelihood, European
DRG-like PCSs 55–6
competition, yard stick competition and
effi ciency 93–4
complication and comorbidity (CCs),
European DRG-like PCSs 51
consequences
DRG-based hospital payments
75–92
unintended 86–90
contracting, quality of care 117
control of expenditure, effi ciency of hospitals
110
coordination benefi ts, Nordic system of
patient classifi cation (NordDRGs)
293–300
cost accounting 59–74
accuracy 70–1
Austria 186–7
characteristics 208–10, 233–4
checks on reported cost data 67–8
cost-accounting systems overview 60–1
costing approaches 64–9
costs covered by DRG-based payments 69
data checks on reported cost data 67–8
data for costing DRG prices 65
direct cost allocation 63–4, 66–7
DRG systems development 69–71
DRGs driving cost accounting 72–3
driving DRGs 72–3
England 207–10
Estonia 311–13
Finland 330–1
France 233–4
Germany 257–8
in healthcare 61–4
hospital cost information 27–8, 159–62
impact 71–2
inaccuracies 70
indirect cost allocation 62–3, 66–7
Ireland 282–3
mandatory systems 64
methods across Europe 65–8
national costing guidelines 64
Netherlands 437–8
overhead allocation 61–2, 66
Poland 372–3
Portugal 390–3
regulation 207–8, 233
relevance, systems 69–70
Spain 411–13
strategies, cost reduction 84–5
Sweden 348–50
transparency 151
cost-based reimbursement, hospital payment
models 95
cost-containment, quality of care 119
cost information, hospital 27–8
cost accounting 159–62
differences between countries 27–8
recommendations 159–62
uses 27
costs
indicator of effi ciency of hospitals 106–9
technological innovation 132–6
coverage of services, European DRG-like PCSs
52
CQUIN see Commissioning for Quality and
Innovation
data checks on reported cost data, cost
accounting 67–8
Index 449
data for costing DRG prices 65
data quality
Austria 185–6
England 206–7
Estonia 310–11
Finland 330
France 232–3
Germany 255–6
Ireland 281–2
LKF-PCS, Austria 185–6
Netherlands 437
patient classifi cation systems (PCSs) 185–6,
206–7, 232–3
Poland 371
Portugal 389–90
Spain 409
Sweden 345
data requirements
coding of diagnoses and procedures 43–4
DRG systems 43–4
day-case classifi cation systems, Ireland 279
DBC classifi cation system 48–9
see also Netherlands
diagnosis-related groups see DRGs
dis-incentives, technological innovation
214–15, 266
Estonia 317–18
Finland 334–5
DRG-based budget allocation model,
Portugal 394–5
DRG-based budgetary adjustments, Ireland
284–6
DRG-based case payments from third-party
payers, Portugal 395
DRG-based hospital payments 28–30, 162–6
acute care hospitals 80–2
building blocks 28–30
consequences 75–92
consequences, unintended 86–90
DRG application/coverage, trends 217
effects 30–3
effi ciency of hospitals 30–1, 93–114
Estonia 313–16
fairness 86–8
high cost services 86–8
hospital payment models 96–100
hospital payment systems and incentives
15
impact on effi ciency 102–6
incentives 28–9, 75–6, 82–6, 100–1, 134–7
indicators of effi ciency 106–9
information asymmetry 89, 116–17
innovation, technological 32–3
institutional context 156
integrating quality 121–5, 164–5
legal context 156
linear payments 97
marginal payments 98–100
mixed payments 97–8
models 28, 95–101
modifi cations 86–90
monetary conversion/adjustment rates
29–30, 77, 79–80
outliers 86–8
political situation 155–6
Portugal 393–6
quality of care 121–5, 164–5
quality of hospitals 31–2, 82–6
rate determination 76, 77–80
recommendations 155–66
reducing 90
Spain 413–19
structural adjustments 78, 79–80
technological innovation 32–3, 131–47
theory 82–6
trends 217
understanding 23–35
weights see DRG prices/cost weights
yard stick competition 93–4
DRG-like PCSs see European DRG-like PCSs
DRG prices/cost weights
DRG-based hospital payments 77–9
Estonia 314–15
Finland 332
Germany 261–2
Portugal 393–4
rate determination 77–9
Spain 414–15
Sweden 351–2
DRG systems
authors’ assessments
Austria 192–3
England 215–16
Estonia 318–19
Finland 335–6
France 240
Germany 267–8
Poland 377–8
Portugal 396–8
Spain 420–1
Sweden 353–6
building blocks 24–7
calculation of scores and trimming, Poland
373–4
characteristics 24–5, 37, 40–3
classifi cation algorithm 204–6, 229–32
coding of diagnoses and procedures 43–4
common EuroDRG system 166–9
current
England 200
France 225–6
Poland 364–5
Portugal 383–4
Spain 405
Sweden 341–2
data requirements 43–4
data used
Austria 182
Estonia 308
Finland 328
France 226–8
Germany 251–3
450 Diagnosis-Related Groups in Europe
DRG systems (Continued)
Ireland 279
Netherlands 432–3
Portugal 386–7
Spain 408
Sweden 342–4
design implications, Germany 269–70
development
Austria 179–82
cost accounting 69–71
England 200–4
Estonia 306–8
Finland 326–8
France 225–9
Germany 247–53
Ireland 277–9
Netherlands 429–33
Poland 364–9
Portugal 383–7
Spain 405–8
Sweden 341–4
economically homogeneous groups
25–7
England 215–16
EuroDRG system, common 166–9
evaluations
Austria 191–3
England 215–16
Estonia 318–19
Finland 335–6
France 239–40
Germany 266–8
Ireland 287–8
Netherlands 442–3
Poland 377–8
Portugal 396–8
Spain 419–21
Sweden 353–6
France 225
future 166–9
historical origins 38–40
hospital fi nancing
Austria 187–91
Poland 373–6
hospital payment, Poland 374–6
institutional context 156
legal context 156
Nordic system of patient classifi cation
(NordDRGs) 168, 293–300
offi cial evaluations
Austria 191–2
England 215
Estonia 318
Finland 335
France 239–40
Germany 266–7
Poland 377
Spain 419–20
Sweden 353
options 25–6, 156–7
overview 40–3
patient classifi cation systems (PCSs) 24–7,
44–9
phases, Germany 248–51
political situation 155–6
purposes 158
Austria 179
England 199–200
Estonia 304–6
Finland 325–6
France 225
Germany 247
Ireland 275–7
Netherlands 428
Poland 363
Portugal 383
Spain 404–5
Sweden 340–1
recommendations 155–66
sources of information, Poland
368
updating 159
Austria 179–82
England 200–4
Estonia 306–8
Finland 326–8
France 225–9
Germany 247–53
Ireland 277–9
Netherlands 429–33
Poland 364–9
Portugal 383–7
Spain 405–8
Sweden 341–4
versions
Finland 326–7
France 226–8
Portugal 384–6
Sweden 342, 343
DRGs
homogeneity 158–9
implementation in Europe 4–7
introduction to Europe 9–19
key questions 33–4
objectives 9–19
origins 3–4
purposes 10–13, 156
understanding 23–35
DRGs for hospital payment
England 210–13
Ireland 283–7
Netherlands 438–41
Spain 413–19
Sweden 350–2
DRGs in actual hospital payment
Austria 188–90
Finland 332–3
France 237
Germany 262
Netherlands 439–41
Spain 415–17, 418
Sweden 352
Index 451
drug regulations
Netherlands 143–4
technological innovation 143–4
economically homogeneous groups, DRG
systems 25–7
effi ciency of hospitals
activity reports 106–9
categorization problems 109
control of expenditure 110
costs 106–9
defi ning 94
DRG-based hospital payments 30–1,
93–114
empirical evidence, DRG-based hospital
payments 101–9
improving 155–66
incentives and their consequences 82–6
indicators 106–9
key challenges 109–10
length of stay 106–9
moving towards 152–3
price-setting independence 110
recommendations 155–66
unfair patient selection 109
unfair reimbursement 109
yard stick competition 93–4
empirical evidence, effi ciency, DRG-based
hospital payments 101–9
England
actual hospital payment, DRGs in 211
classifi cation algorithm 204–6
cost accounting 207–10
DRG prices calculation 210–11
DRG system
authors’ assessment 215–16
current 200
development 200–4
evaluation 215–16
offi cial evaluation 215
purpose 199–200
updating 200–4
DRGs for hospital payment 210–13
health care system 197–8
Healthcare Resource Group system
197–220
hospital services 197–200, 217
incentives 213
information used to classify patients 204
outlook 217
patient classifi cation system (PCS) 204–7
quality-related adjustments 213
range of services and costs 210, 212
regulation 207–8
role of DRGs 197–200
self-developed DRG-like PCSs 46–8
technological innovation 213–15
trends 217
Estonia 301–20
authors’ assessment 318–19
classifi cation algorithm 309–10
cost accounting 311–13
data quality 310–11
dis-incentives, technological innovation
317–18
DRG-based hospital payments 313–16
DRG prices/cost weights 314–15
DRG system
authors’ assessment 318–19
data used 308
development 306–8
evaluation 318–19
offi cial evaluation 318
purpose 304–6
updating 306–8
DRGs in hospital payment 315
future developments and reform 319
health care system 301–2
hospital services 301–6
incentives 311, 316
information used to classify patients 309
offi cial evaluation, DRG system 318
patient classifi cation system (PCS) 309–11
payment mechanisms 317
plausibility checks 310–11
quality-related adjustments 316
range of services and costs 313–14
regulation 311–12
role of DRGs 301–6
technological innovation 316–18
trends 319
up-coding 311
wrong-coding 311
EuroDRG system, common 166–9
Europe, introduction of DRGs 9–19
Europe evidence, quality of care 120–1
European DRG-like PCSs 44–55
see also patient classifi cation systems
(PCSs)
classifi cation algorithm 44–9
classifi cation variables 49–51
common system likelihood 55–6
complications and comorbidities (CCs) 51
coverage of services 52
number of groups 52–4
séances 54–5
self-developed DRG-like PCSs 46–8
severity levels 49–51
specifi c high-cost services 54–5
supplementary payments 54–5
trends 51–5
unbundling 54–5
expectations, DRGs 10–13
fairness, DRG-based hospital payments 86–8
fee-for-service reimbursement
Germany 137–9
hospital payment models 95
hospital payment systems and incentives
14
New Diagnostic and Treatment Methods
Regulation (NUB) 137–9
452 Diagnosis-Related Groups in Europe
fi nancing health care, Germany 244, 269
Finland 321–38
acute care hospitals, key fi gures 322–5
classifi cation algorithm 328–30
cost accounting 330–1
data quality 330
dis-incentives, technological innovation
334–5
DRG application/coverage 336–7
DRG prices/cost weights 332
DRG system
authors’ assessment 335–6
data used 328
development 326–8
evaluation 335–6
offi cial evaluation 335
purpose 325–6
updating 326–8
versions 326–7
DRGs for reimbursement 331–4
DRGs in actual hospital payment 332–3
funding of hospital care 322–5
future developments and reform 336–7
health care system 321–2
hospital services 321–6, 336–7
incentives 330
information used to classify patients 328
patient classifi cation system (PCS) 328–30
payment mechanisms 334
plausibility checks 330
quality-related adjustments 333–4
regulation 330–1
reimbursement, DRGs for 331–4
role of DRGs 321–6
technological innovation 334–5
trends 336–7
up-coding 330
wrong-coding 330
France 221–41
acute care hospitals 222–3
authors’ assessment, GHM system 240
classifi cation algorithm 229–32
cost accounting 233–4
costs/prices calculation 235–7
data quality 232–3
DRG system
authors’ assessment 240
current 225–6
data used 226–8
development 225–9
evaluation 239–40
offi cial evaluation 239–40
purposes 225
updating 225–9
versions 226–8
DRGs for reimbursement 234–8
DRGs in actual hospital payment 237
evaluation, GHM system 239–40
future developments and reform 240
health care system 221–2
hospital services 221–5
incentives 233, 238
information used to classify patients 229
offi cial evaluation, GHM system 239–40
overview 222–5
patient classifi cation system (PCS) 229–33
plausibility checks 232–3
public/private activity 222–3
quality-related adjustments 237–8
range of services and costs 234–5
reference costs/prices 235–7
regulation 233
reimbursement, DRGs for 234–8
role of DRGs 221–5
technological innovation 238–9
up-coding 233
wrong-coding 233
future
common EuroDRG system 166–9
developments and reform
Austria 193
Estonia 319
Finland 336–7
France 240
Germany 268–70
Ireland 288–9
Netherlands 443–5
Poland 378
Portugal 398–9
Spain 421–2
Sweden 356
DRG systems, Europe 166–9
LKF-PCS, Austria 193
Germany 243–71
classifi cation algorithm 254–5
clinical data 255–6
cost accounting 257–8
cost data 255
data quality 255–6
DRG prices/cost weights 261–2
DRG system
authors’ assessment 267–8
current 247–8
data used 251–3
design implications 269–70
development 247–53
evaluation 266–8
offi cial evaluation 266–7
phases 248–51
purpose 247
updating 247–53
DRGs for reimbursement 258–63
DRGs in actual hospital payment 262
evaluation, DRG system 266–8
fee-for-service reimbursement 137–9
fi nancing health care 244, 269
future developments and reform 268–70
health care system 243–4
hospital services 243–7
incentives 256–7, 263
information used to classify patients 253–4
Index 453
New Diagnostic and Treatment Methods
Regulation (NUB) 137–9
patient classifi cation system (PCS) 253–7
plausibility checks 255–6
quality-related adjustments 262–3
range of activities and services 246
range of services and costs 246, 258–61
regulation, cost accounting 257, 269
reimbursement, DRGs for 258–63
relationship with third party payers 246–7
role of DRGs 243–7
technological innovation 263–6
trends 268–9
up-coding 256–7
wrong-coding 256–7
global budgets
hospital payment models 95–6
hospital payment systems and incentives
14–15
grouping algorithms
see also classifi cation algorithms
Austria 184–5
LKF-PCS, Austria 184–5
patient classifi cation systems (PCSs) 184–5
Sweden 345, 346–7
health care systems
Austria 175–7
England 197–8
Estonia 301–2
Finland 321–2
France 221–2
Germany 243–4
Ireland 273–4
Netherlands 425–6
Poland 359–60
Portugal 381–2
Spain 401–2
Sweden 339
health technology assessment (HTA)
agencies, technological innovation
144–5
Healthcare Resource Group system, England
197–220
high cost services, DRG-based hospital
payments 86–8
historical origins, DRG systems 38–40
Holland see Netherlands
hospital cost information 27–8
cost accounting 27–8, 159–62
differences between countries 27–8
recommendations 159–62
uses 27
hospital fi nancing, LKF-PCS, Austria 187–91
hospital landscape, acute care hospitals
15–19
hospital payment models 28, 95–101
see also DRG-based hospital payments
cost-based reimbursement 95
fee-for-service reimbursement 95
global budgets 95–6
hospital payment systems and incentives
13–15
see also DRG-based hospital payments
fee-for-service payments 14
global budgets 14–15
outcome-based payment adjustment 122–4
hospital performance comparisons,
transparency 151
hospital services
Austria 175–9
England 197–200, 217
Estonia 301–6
Finland 321–6, 336–7
France 221–5
Germany 243–7
Ireland 273–7
Netherlands 425–8
Poland 359–63
Portugal 381–3, 398–9
Spain 401–5
Sweden 339–41
trends 217, 398–9
hospitals, acute care see acute care hospitals
HTA (health technology assessment) agencies,
technological innovation 144–5
implementation in Europe, DRGs 4–7
inaccuracies
cost accounting 70
patient classifi cation systems (PCSs) 70
incentives
see also hospital payment systems and
incentives
Austria 186, 190–1
consequences 82–6
dis-incentives, technological innovation
214–15, 266, 317–18
DRG-based hospital payments 28–9, 75–6,
82–6, 100–1, 134–7
England 213
Estonia 311, 316
Finland 330
France 233, 238
Germany 256–7, 263
Ireland 287
LKF-PCS, Austria 186, 190–1
Netherlands 437, 441
patient classifi cation systems (PCSs) 186
Poland 372
Portugal 390, 395–6
power 90
Spain 411, 417–19
Sweden 345–8, 352, 353
technological innovation 134–7, 214–15
up-coding 186, 233
wrong-coding 186, 233
indicators of effi ciency, DRG-based hospital
payments 106–9
information asymmetry
DRG-based hospital payments 89, 116–17
quality of care 116–17
454 Diagnosis-Related Groups in Europe
information used to classify patients
Austria 183
England 204
Estonia 309
Finland 328
France 229
Germany 253–4
Ireland 280
LKF-PCS, Austria 183
Netherlands 433–5
Poland 369
Portugal 388
Spain 408–9
Sweden 344
innovation, technological see technological
innovation
inpatient classifi cation systems, Ireland
277–8
institutional context
DRG-based hospital payments 156
DRG systems 156
instruments, technological innovation
137–42
long-term updating mechanisms 140–2
short-term instruments 137–40
integrating quality, DRG-based hospital
payments 121–5, 164–5
introduction to Europe, DRGs 9–19
Ireland 273–92
charges, health care 274, 275
classifi cation algorithm 280–1
cost accounting 282–3
data quality 281–2
day-case classifi cation systems 279
DRG-based budgetary adjustments
284–6
DRG system
data used 279
development 277–9
evaluation 287–8
purpose 275–7
updating 277–9
DRGs for hospital payment 283–7
future developments and reform 288–9
health care system 273–4
hospital services 273–7
incentives for hospitals 287
information used to classify patients 280
inpatient classifi cation systems 277–8
patient classifi cation system (PCS) 280–2
plausibility checks 281–2
quality-related adjustments 286
range of services and costs 283
regulation 282
relative values 283–4
role of DRGs 273–7
service weights 283–4
technological innovation 287
key challenges, effi ciency of hospitals 109–10
key questions, DRGs 33–4
LDF scores calculation
actual hospital payment 188–90
Austria 187–90
day component 188
LKF core area (Kernbereich) 189
LKF-PCS, Austria 187–8
LKF steering area (Steuerungsbereich) 190
outliers adjustments 188
performance component 187
legal context
DRG-based hospital payments 156
DRG systems 156
length of stay, indicator of effi ciency of
hospitals 106–9
linear payments, DRG-based hospital
payments 97
LKF-PCS, Austria 179–93
authors’ assessment 192–3
data quality 185–6
development 179–82
evaluation of LKF framework 191–3
future developments and reform 193
grouping algorithm 184–5
hospital fi nancing 187–91
incentives 186, 190–1
information used to classify patients 183
LDF scores calculation 187–8
offi cial evaluation, DRG system 192
plausibility checks 185–6
quality-related adjustments 190
range of services and costs 187
up-coding 186
updating 179–82
wrong-coding 186
long-term updating mechanisms,
technological innovation 140–2
marginal payments, DRG-based hospital
payments 98–100
mixed payments, DRG-based hospital
payments 97–8
models, DRG-based hospital payments 28,
95–101
modifi cations, DRG-based hospital payments
86–90
monetary conversion/adjustment rates,
DRG-based hospital payments 29–30,
77, 79–80
Netherlands 425–46
classifi cation algorithm 435–7
cost accounting 437–8
data quality 437
DBC classifi cation system 48–9
demand for care 436
diagnosis 436
DRG system
data used 432–3
development 429–33
evaluation 442–3
purpose 428
Index 455
types 429–30
updating 429–33
DRGs for hospital payment 438–41
DRGs in actual hospital payment 439–41
drug regulations 143–4
future developments and reform 443–5
health care system 425–6
hospital services 425–8
incentives 437, 441
information used to classify patients 433–5
medical specialty 435–6
patient classifi cation system (PCS) 433–7
plausibility checks 437
quality of care 154
quality-related adjustments 441
range of services and costs 438–9
regulation 437
tariffs calculation 439, 440
technological innovation 143–4, 441–2
treatment axis 437
type of care 436
up-coding 437
wrong-coding 437
New Diagnostic and Treatment Methods
Regulation (NUB)
fee-for-service reimbursement 137–9
Germany 137–9
Nordic system of patient classifi cation
(NordDRGs)
coordination benefi ts 293–300
development 294–7
diagnosis differences 297–8
DRG systems 168
procedure classifi cation differences 297–8
reimbursement, DRGs for 299–300
updating 294–7
NUB see New Diagnostic and Treatment
Methods Regulation
number of groups, European DRG-like PCSs
52–4
objectives, DRGs 9–19
offi cial evaluations
DRG system
Austria 192
England 215
Estonia 318
Finland 335
France 239–40
Germany 266–7
Poland 377
Spain 419–20
Sweden 353
GHM system, France 239–40
LKF-PCS, Austria 192
options, DRG systems 25–6, 156–7
origins
historical origins, DRG systems 38–40
origins of DRGs, starting point 3–4
outcome-based payment adjustment, quality
of care 122–4
outliers
adjustments, LDF scores calculation 188
defi ning 87
DRG-based hospital payments 86–8
overhead allocation, cost accounting 61–2, 66
ownership types, acute care hospitals 19
patient classifi cation systems (PCSs)
see also European DRG-like PCSs
Austria 183–6
classifi cation algorithm 44–9, 204–6, 229–32
classifi cation variables, European DRG-like
PCSs 49–51
data quality 185–6, 206–7, 232–3
DRG systems 24–7, 44–9
England 204–7
Estonia 309–11
Finland 328–30
France 229–33
Germany 253–7
grouping algorithm 184–5
inaccuracies 70
incentives 186
Ireland 280–2
Netherlands 433–7
Nordic system of patient classifi cation
(NordDRGs) 168
plausibility checks 185–6, 206–7, 232–3
Poland 369–72
Portugal 388–90
self-developed DRG-like PCSs 46–8
severity levels, European DRG-like PCSs
49–51
Spain 408–11
Sweden 344–8
unbundling 54–5, 205–7
payment mechanisms
Estonia 317
Finland 334
Sweden 353
technological innovation 214
payment systems see DRG-based hospital
payments; hospital payment models;
hospital payment systems and
incentives
payment types, technological innovation
137–40
PCSs see patient classifi cation systems
plausibility checks, PCSs
Austria 185–6
England 206–7
Estonia 310–11
Finland 330
France 232–3
Germany 255–6
Ireland 281–2
LKF-PCS, Austria 185–6
Netherlands 437
Poland 371
Portugal 389–90
Spain 409
Sweden 345
456 Diagnosis-Related Groups in Europe
Poland 359–80
classifi cation algorithm 369–71
cost accounting 372–3
data quality 371
DRG system
authors’ assessment 377–8
calculation of scores and trimming
373–4
current 364–5
development 364–9
evaluation 377–8
hospital fi nancing 373–6
hospital payment 374–6
offi cial evaluation 377
purpose 363
sources of information 368
updating 364–9
future developments and reform 378
health care system 359–60
hospital services 359–63
incentives 372
information used to classify patients 369
patient classifi cation system (PCS) 369–72
plausibility checks 371
quality-related adjustments 376
regulation 372
role of DRGs 359–63
self-developed DRG-like PCSs 46–8
technological innovation 376–7
up-coding 372
wrong-coding 372
political situation
DRG-based hospital payments 155–6
DRG systems 155–6
Portugal 381–400
3M™ Health Information System 384, 387
classifi cation algorithm 388–9
cost accounting 390–3
data quality 389–90
DRG application/coverage 398–9
DRG-based budget allocation model 394–5
DRG-based case payments from third-party
payers 395
DRG-based hospital payments 393–6
DRG prices/cost weights 393–4
DRG system
authors’ assessment 396–8
current 383–4
data used 386–7
development 383–7
evaluation 396–8
purpose 383
updating 383–7
versions 384–6
future developments and reform 398–9
health care system 381–2
hospital services 381–3, 398–9
incentives 390, 395–6
information used to classify patients 388
patient classifi cation system (PCS) 388–90
plausibility checks 389–90
quality-related adjustments 395
range of services and costs 393
regulation 390–1
role of DRGs 381–3
technological innovation 396
trends 398–9
up-coding 390
wrong-coding 390
price-setting, quality of care 117–18
price-setting independence, effi ciency of
hospitals 110
public/private activity, France 222–3
quality of care 115–29
average length of stay (ALOS) 117, 118–19
coding of diagnoses and procedures 121
Commissioning for Quality and
Innovation (CQUIN) 122, 154
contracting 117
cost-containment 119
defi ning 116, 133
DRG-based hospital payments 121–5,
164–5
Europe evidence 120–1
evidence 118–21
improving 155–66
information asymmetry 116–17
integrating into payment 121–5
moving towards 153–5
Netherlands 154
outcome-based payment adjustment 122–4
price-setting 117–18
recommendations 155–66
technological innovation 132–4
theory 116–18
United States evidence 118–20
quality of hospitals
DRG-based hospital payments 31–2, 82–6
incentives and their consequences 82–6
quality-related adjustments
Austria 190
Commissioning for Quality and
Innovation (CQUIN) 213
England 213
Estonia 316
Finland 333–4
France 237–8
Germany 262–3
Ireland 286
LKF-PCS, Austria 190
Netherlands 441
Poland 376
Portugal 395
Spain 417
Sweden 352
range of activities and services, Germany 246
range of services and costs
Austria 187
England 210, 212
Estonia 313–14
Index 457
France 234–5
Germany 246, 258–61
Ireland 283
Netherlands 438–9
Portugal 393
Spain 413–14
Sweden 350–1
rate determination
DRG-based hospital payments 76, 77–80
weights 77–9
recommendations
DRG-based hospital payments 155–66
DRG systems 155–66
effi ciency of hospitals 155–66
hospital cost information 159–62
quality of care 155–66
transparency 155–66
regulation
cost accounting 207–8, 233
cost accounting, Germany 257, 269
England 207–8
Estonia 311–12
Finland 330–1
France 233
Ireland 282
Netherlands 437
Poland 372
Portugal 390–1
Spain 411
Sweden 348
relationship with third party payers,
Germany 246–7
relative values, Ireland 283–4
role of DRGs
Austria 175–9
England 197–200
Estonia 301–6
Finland 321–6
France 221–5
Germany 243–7
Ireland 273–7
Poland 359–63
Portugal 381–3
Spain 401–5
Sweden 339–41
séances, European DRG-like PCSs 54–5
self-developed DRG-like PCSs 46–8
service weights, Ireland 283–4
severity levels, European DRG-like PCSs
49–51
short-term instruments, technological
innovation 137–40
Spain 401–23
acute care hospitals 403–4
classifi cation algorithm 409, 410
cost accounting 411–13
data quality 409
DRG-based hospital payments 413–19
DRG prices/cost weights 414–15
DRG system
authors’ assessment 420–1
current 405
data used 408
development 405–8
evaluation 419–21
offi cial evaluation 419–20
purpose 404–5
updating 405–8
DRGs for hospital payment 413–19
DRGs in actual hospital payment 415–17,
418
future developments and reform 421–2
health care system 401–2
hospital services 401–5
incentives 411, 417–19
information used to classify patients 408–9
patient classifi cation system (PCS) 408–11
plausibility checks 409
quality-related adjustments 417
range of services and costs 413–14
regulation 411
role of DRGs 401–5
technological innovation 419
up-coding 411
wrong-coding 411
specifi c high-cost services, European
DRG-like PCSs 54–5
starting point, origins of DRGs 3–4
strategies, cost reduction, cost accounting
84–5
structural adjustments, DRG-based hospital
payments 78, 79–80
structure, this book’s 19–20
supplementary payments, European
DRG-like PCSs 54–5
Sweden 339–57
3M™ Health Information System 293
case costing model 348–50
classifi cation algorithm 345, 346–7
cost accounting 348–50
data quality 345
DRG prices/cost weights 351–2
DRG system
authors’ assessment 353–6
current 341–2
data used 342–4
development 341–4
evaluation 353–6
offi cial evaluation 353
purpose 340–1
updating 341–4
versions 342, 343
DRGs for hospital payment 350–2
DRGs in actual hospital payment 352
future developments and reform 356
grouping algorithm 345, 346–7
health care system 339
hospital services 339–41
incentives 345–8, 352, 353
information used to classify patients 344
patient classifi cation system (PCS) 344–8
458 Diagnosis-Related Groups in Europe
payment mechanisms 353
plausibility checks 345
quality-related adjustments 352
range of services and costs 350–1
regulation 348
role of DRGs 339–41
technological innovation 353
trends 356
up-coding 345–8
wrong-coding 345–8
systems, DRG see DRG systems
technological innovation
Austria 191
comparison, country 136–45
costs 132–6
dis-incentives 214–15, 266
DRG-based hospital payments 32–3,
131–47
drug regulations 143–4
encouraging 136–45
England 213–15
Estonia 316–18
Finland 334–5
France 238–9
Germany 263–6
health technology assessment (HTA)
agencies 144–5
incentives 134–7, 214–15
instruments 137–42
Ireland 287
long-term updating mechanisms 140–2
Netherlands 143–4, 441–2
payment mechanisms 214
payment types 137–40
Poland 376–7
Portugal 396
quality of care 132–4
scope for improvement 143–5
short-term instruments 137–40
Spain 419
Sweden 353
transparency
activity reports 150–1
coding of diagnoses and procedures 151
cost accounting 151
hospital performance comparisons 151
improving 155–66
moving towards 150–1
recommendations 155–66
trends
coverage of services 52
DRG application/coverage 217, 269, 336–7,
356, 398–9
DRG-based hospital payments 217
England 217
Estonia 319
European DRG-like PCSs 51–5
Finland 336–7
Germany 268–9
hospital services 217, 268, 336–7, 356,
398–9
number of groups 52–4
Portugal 398–9
séances 54–5
specifi c high-cost services 54–5
supplementary payments 54–5
Sweden 356
unbundling 54–5
unbundling
European DRG-like PCSs 54–5
patient classifi cation systems (PCSs) 54–5,
205–7
trends 54–5
unfair patient selection, effi ciency of
hospitals 109
unfair reimbursement, effi ciency of hospitals
109
United States evidence, quality of care
118–20
up-coding
Austria 186
Estonia 311
Finland 330
France 233
Germany 256–7
incentives 233
LKF-PCS, Austria 186
Netherlands 437
Poland 372
Portugal 390
Spain 411
Sweden 345–8
weights, DRG-based hospital payments see
DRG prices/cost weights
wrong-coding
Austria 186
Estonia 311
Finland 330
France 233
Germany 256–7
incentives 233
LKF-PCS, Austria 186
Netherlands 437
Poland 372
Portugal 390
Spain 411
Sweden 345–8
yard stick competition
DRG-based hospital payments 93–4
effi ciency of hospitals 93–4
European Observatory on Health Systems and Policies Series
Diagnosis-Related Groups in Europe
Moving towards transparency, efficiency and quality in hospitals
Diagnosis-Related Group (DRG) systems were introduced in Europe to increase
the transparency of services provided by hospitals and to incentivize greater
efficiency in the use of resources invested in acute hospitals. In many countries,
these systems were also designed to contribute to improving – or at least
protecting – the quality of care. After more than a decade of experience with
using DRGs in Europe, this book considers whether the extensive use of DRGs
has contributed towards achieving these objectives.
Written by authors with extensive experience of these systems, this book is a
product of the EuroDRG project and constitutes an important resource for health
policy-makers and researchers from Europe and beyond. The book is intended
to contribute to the emergence of a ‘common language’ that will facilitate
communication between researchers and policy-makers interested in
improving the functioning and resourcing of the acute hospital sector. The book
includes:
• A clearly structured introduction to the main ‘building blocks’ of DRG
systems
• An overview of key issues related to DRGs including their impact on efficiency,
quality, unintended effects and technological innovation in health care
• 12 country chapters – Austria, England, Estonia, Finland, France, Germany,
Ireland, the Netherlands, Poland, Portugal, Spain and Sweden
• Clearly structured and detailed information about the most important DRG
system characteristics in each of these countries
• Useful insights for countries and regions in Europe and beyond interested
in introducing, extending and/or optimizing DRG systems within the
hospital sector
Reinhard Busse is Professor and Head of the Department of Healthcare
Management at Berlin University of Technology, Germany.
Alexander Geissler is a research fellow in the Department of Health Care
Management at Berlin University of Technology, Germany.
Wilm Quentin is a research fellow in the Department of Health Care
Management at Berlin University of Technology, Germany.
Miriam M. Wiley is Professor and Head of the Health Research and Information
Division at the Economic and Social Research Institute in Dublin, Ireland.
www.openup.co.uk
Diagnosis-Related
Groups in Europe
Moving towards transparency, efficiency
and quality in hospitals
Edited by
Reinhard Busse
Alexander Geissler
Wilm Quentin
Miriam Wiley
D
ia
gn
o
sis-R
ela
ted
G
ro
u
p
s in
Eu
ro
p
e
Busse, G
eissler, Q
uentin and W
iley
Diagnosis Related Groups…pb24.5_Diagnosis Related Groups…pb24.5 23/10/2011 11:32 Page 1
Health Sector
Transformation Program
Delivery Plan
Delivery plan 2020-20
21
Custodian of the Two Holy Mosques
King Salman bin Abdulaziz Al Saud
My primary objective is for our
country to be a global model of
excellence, on all fronts
The future of the Kingdom is one of great
promise and potential. Our precious country
deserves more than it has achieved. Therefore,
we will expand and develop our talents and
capabilities to craft this future
His Royal Highness
Prince Mohammed bin Salman bin Abdulaziz
Crown Prince, Deputy Prime Minister and Chairman of the
Council of Economic and Development Affairs
Foreword
In a wise and benevolent gesture from the rulers of this nation, may God protect them, the
Basic Law of Governance of the Kingdom of Saudi Arabia guarantees free healthcare and
treatment for citizens, as stipulated in Article 31: “The State shall be solicitous of public
health and provide health care for every citizen in cases of emergency, illness, disability and
old age.” Moreover, in the Royal Decree No. M/
11
, dated 23/3/1423 AH (corresponding to
4/6/2002 AD), Article 2 states that the system aims to ensure the provision of comprehensive,
integrated healthcare to all the population in a manner that is fair, accessible and organized.
The launch of the Health Sector Transformation Program reinforces this human dimension,
emphasizing continuity, development of the health system in general and enhanced quality
of care and services provided. As part of Saudi Vision 2030, the Health Sector
Transformation Program was established to contribute to the realization of the “Vibrant
Society”, one of the main pillars of the Kingdom’s vision. To meet its objectives, the
program seeks to restructure the health sector, consequently enhancing its capabilities as an
effective, integrated ecosystem that sets the health of every member of society at the
forefront of its priorities.
The program will work to achieve improved health and healthcare services in cooperation
with all public & private health sector bodies, and in alignment with the strategic national
objectives for Vision 2030.
The program will also work to complete the implementation of the four strategic objective
s
set out for the Kingdom’s health system, having transitioned from the National
Transformation Program to the Health Sector Transformation Program, as well as initiatives
associated with it. The strategic objectives are: facilitating access to health services,
improving the quality and efficiency of services, promoting prevention of health risks, and
enhancing traffic safety.
Furthermore, the program will develop a national strategy for the transformation of the
health sector as a whole, in cooperation with all associated parties, establishing a
comprehensive plan encompassing all health and healthcare bodies at the national level and
taking into account the responsibilities, specialties and previous successes of each entity.
1
1
1
2
1
5
1
6
1
7
1
8
20
2
3
2
4
2
9
32
34
36
39
43
44
45
50
52
57
70
77
78
85
Contents
Preface
Introduction
Chapter 1
Health Sector Transformation Program scope
About the Health Sector Transformation Program
HSTP strategic objectives
Level 3 direct objectives
Level 3 indirect objectives
Public policies
HSTP commitments and ambitions
Chapter 2
Current conditions
Core
challenges
Key factors contributing to these challenges
Challenges arising from the Covid-
19
pandemic
Quick solutions and measures to combat Covid-19
Chapter 3
Strategic cornerstones of the Health Sector
Transformation Program
Health Sector Transformation Program
Main challenge areas
Strategic cornerstones
HSTP aspirations for the future of the health sector
Mechanism for achieving future aspirations
HSTP strategic branches
Chapter 4
Initiatives portfolio
Priority initiatives
Current initiatives
٩
١١
11
12
15
16
17
18
20
23
24
29
32
34
36
39
43
44
45
50
52
57
70
77
78
85
Saudi Vision 2030 has several strategic objectives and performance indicators built
on the fundamental pillars of our joint commitments. Together, the public, private
and third sectors will work to realize these objectives as mandated by the Council of
Economic and Development Affairs and under the leadership of His Royal Highness
Crown Prince Mohammed bin Salman.
The vision realization programs aim to translate Saudi Vision 2030 into multiple
delivery plans designed to achieve its strategic objectives through innovative
practices aligned with national objectives, following a specified timetable towards
successful completion.
By royal decree, the Health Sector Transformation Program was launched to
restructure the sector into a comprehensive and effective health ecosystem to meet
the requirements and achieve the ambitions of Saudi Vision 2030.
Delivery Plan
In line with Saudi Vision 2030 and its aspirations for achieving the Kingdom’s
potential and untapped capabilities, a new phase of the Health Sector
Transformation Program has been set into motion. The objective of this phase is to
outline an operational plan for the program that will achieve an enhanced level of
health and beneficiary services contributing to a prosperous, sustainable future.
Implementation of the delivery plan will launch efforts in the transformation of the
health sector through a series of initiatives at the national level, clarifying and
assigning responsibilities to achieve health and healthcare excellence in the
Kingdom of Saudi Arabia.
Preface
11
Saudi Vision 2030 was launched as a roadmap for economic and
developmental efforts, setting objectives and defining commitments and
responsibilities. To implement the Kingdom’s ambitions, the Vision will
reinforce the nation’s position as a pioneering model of success globally and
across various fields. Saudi Vision 2030 is built on three fundamental pillars:
a vibrant society, a thriving economy, and an ambitious nation.
They function in harmony to achieve the desired objectives and maximize
the benefits of the vision.
Introduction
Vibrant Society
Thriving Economy Ambitious Nation
Saudi Vision 2030 strives to ensure a vibrant
society by providing a rich and fulfilled life for
all, and in doing so enabling society to be the
driving force behind economic prosperity.
We will achieve these objectives by drawing on the Kingdom’s intrinsic
strengths, its unique location and potential, and by attracting the best talent
and increased global investments.
Kingdom of Saudi Arabia
Heart of the Arab and Islamic worlds
Investment powerhouse
Hub connecting three continents
Saudi Vision 2030 aspires to a
transformation embodied by highly
effective, transparent and
accountable governance, enabling
citizens, the private sector and
non-profit institutions to explore all
available opportunities to achieve
their objectives.
Saudi Vision 2030 endeavors to provide a
supportive and empowering environment
capable of releasing untapped innovation
in the business sector, contributing to
economic expansion and employment
opportunities for all Saudis.
13
Chapter 1:
Health Sector Transformation
Program Scope
About the Health Sector
Transformation Program
The program aims to restructure the Saudi health sector, enhancing its status and
capabilities as an effective, integrated, value-based ecosystem centered on the
health of the patient. The HSTP furthermore strives towards transparency and
financial sustainability by promoting public health and disease prevention and
implementing modern healthcare paradigms.
The program also aims to facilitate citizen’s access to free health and healthcare
services and insurance. This is accomplished by guaranteeing fair and comprehensive
geographical coverage across all regions of the Kingdom, expanding e-health
services and digital solutions, and improving the quality of healthcare. In addition,
the program will focus on increasing levels of beneficiary satisfaction by
implementing value-based healthcare and international best practices and
enhancing community awareness of traffic safety.
Moreover, the Health Sector Transformation Program will work in cooperation and
in coordination with all health sector bodies, Vision Realization Programs and
relevant government entities to align with the strategic national and Vision 2030
objectives throughout the transformation journey.
16
HSTP Strategic Objectives
Fa
cil
ita
tin
g
ac
ce
ss
t
o
he
al
th
ca
re
s
er
vi
ce
s
Enhancing traffic safe
ty
Im
pro
vin
g qu
ality a
nd efficiency of health
services
Prom
oting prevention of health risks
Improve quality of life in S
aud
i ci
tie
s
Imp
rove health services
En
abl
e ful
l and healthy lives
Vibr
ant society
Level 2 objectives
Branch objectives
Level 3 objectives
Strategic objectives
Saudi Vision 2030 pillar
One of the three major pillars
Level 1 objectives
Overarching objectives
17
The program contributes to achieving:
HSTP Strategic Objectives
Fa
cil
ita
tin
g
ac
ce
ss
t
o
he
al
th
ca
re
s
er
vi
ce
s
Enhancing traffic safe
ty
Im
pro
vin
g qu
ality a
nd efficiency of health services
Prom
oting prevention of health risks
Improve quality of life in S
aud
i ci
tie
s
Imp
rove health services
En
abl
e ful
l and healthy lives
Vibr
ant society
Level 2 objectives
Branch objectives
Level 3 objectives
Strategic objectives
Saudi Vision 2030 pillar
One of the three major pillars
Level 1 objectives
Overarching objectives
17
The program contributes to achieving:
Facilitating
access to
healthcare services
Improving
the quality and
efficiency of health
services
Promoting
prevention of
health risks
Enhancing
traffic safety
Level 3 Direct Objectives
Through the creation of a detailed national strategy for the health sector as a
whole, the program strives to achieve a series of direct objectives that have been
transferred from the National Transformation Program to the Health Sector
Transformation Program. Namely:
18
2.1.1 Facilitate access to healthcare services: Four factors
will be addressed to ensure beneficiary access to healthcare
services at the right time and place: expansion of total capacity
(hospital beds and medical staff), appropriate geographical
distribution (distance between beneficiary and healthcare
provider), timely access to relevant services, and the
affordability of healthcare for individuals.
2.1.3 Promote prevention of health risks: This objective
consists of strategies for better health throughout life:
addressing social causes of health and health inequality, better
health and healthcare for an aging population, chronic disease
prevention, effective care for people with mental health and
advancing the health of the general population. The objective
covers preventive public health to reduce exposure to disease,
and the management of health crises pertaining to both
communicable and non-communicable diseases, including
epidemics and natural disasters.
2.3.4 Enhance traffic safety: Traffic safety addresses the
following factors, arranged by the magnitude of resulting
damage: injuries, accidents, and the spread of serious
violations.
2.1.2 Improve quality and efficiency of healthcare services:
This objective focuses on enhancing the quality and efficiency of
healthcare services and free insurance for citizens by supporting
the sector’s response to the health needs and expectations of the
community, and by promoting safe, effective and financially
sustainable health coverage.
19
4.4.3
Attract global talent
4.1.5
Provide quality resources for
distinguished talent in priority fields
3.1.3
Privatize selected
government services
4.1.7
Expand vocational training
3.3.2
Develop digital economy
4.1.6
Ensure alignment of educational
outputs with labor market needs
3.1.6
Attract foreign and
domestic investments
3.1.2
Unlock state-owned assets for
the private sector
Level 3 Indirect Objectives
20
6.3.1
Develop non-profit sector
5.2.5
Improve quality of services
provided to beneficiaries
5.4.1
Ensure developmental
and food security
5.2.4
Develop e-government
5.3.3
Enhance interaction of government
agencies with beneficiary feedback
5.1.2
Diversify government
revenues
5.2.3
Improve productivity of
government employees
6.1.2
Encourage volunteer work
21
Preparedness to
respond to health risks
by strengthening
health security
Quality of care
through effective
strategy and
governance
Healthcare financing
and resource
allocation to ensure
optimal value
Health benefits
based on real
beneficiary needs
Executive system for
the provision of
integrated healthcare
services
Competent,
distinguished and
adequate health
workforce
Healthcare innovation Healthcare system
built on value
Health inclusion across
all policies
Reducing deaths and
injuries caused by
traffic accidents
1 2
3 4
5 6
7 8
9
10
23
Public Policies
The program abides by a comprehensive system of public policies to ensure the
sustainability of reforms and increase spending efficiency. Centered around the four
strategic objectives, HSTP policies cover several areas, including:
National
Health
System
Providing integrated,
people-centered healthcare
Achieving effective,
sustainable financial resource
management
Investing in private sector
integration to create effective
strategic partnerships
Employing digital and virtual
health tools and artificial
intelligence
HSTP Commitments and Ambitions
Through its four strategic objectives for health transformation, the program will
contribute to achieving Saudi Vision 2030 aspirations and make a tangible positive
impact on multiple fronts in the development of a national health system:
• Providing beneficiary-centered, integrated healthcare built on enhanced prevention,
quality, efficiency and beneficiary satisfaction through outcome evaluation
• Achieving effective, sustainable financial resource management with the aim of
improving the quality of outcomes and reducing waste and duplication
• Employing digital and virtual health tools, artificial intelligence, and best practice
documentation to achieve effective quality healthcare
• Investing in private sector integration to create effective strategic partnerships in
the pursuit of national health objectives
24
• A healthy, improved life for the beneficiary – supported
by a modern healthcare model that contributes to enhancing
the health of the beneficiary and preventing chronic diseases.
Following the recent increase in average life expectancy from
74
to 75 years, planned changes will contribute to a further
increase of five years in average life expectancy in the
Kingdom – reaching 80 years – in line with Saudi Vision 2030
and life expectancy rates in G8 countries.
• Improved healthcare in the Kingdom of Saudi Arabia
by setting national standards for quality and governance of
monitored service provision (public and private sectors) by the
health sector regulator and controller (Ministry of Health).
• Improved efficiency of healthcare providers, resulting in
the financial sustainability of national healthcare, as well as a
decrease in the average growth rate of public spending on
healthcare. The private sector will also play a major role in
providing healthcare to beneficiaries.
• A strong and resilient Saudi health system, able to
protect the beneficiary in the face of various health challenges,
including outbreaks of infectious diseases and epidemics,
natural disasters and other emergencies. Among the aims of
self-sufficiency are to secure 70% of the supplies specified as
health security priorities, such as personal protection
equipment, sterilizers and instant examination tools.
• Beneficiary access to high-quality healthcare that meets
their needs while addressing dual eligibility for services and
containing the costs of care. The program will develop health
services based on best practices and successful precedents.
The program aspires to achieve the following:
25
• Coverage for all citizens. The Center for National Health
Insurance is a national entity concerned with paying
healthcare costs, and its mission is to enhance the value and
quality of healthcare. As such, the Center will be responsible
for providing free health insurance to beneficiaries by
purchasing health services from providers. The aim of the
center is to create a sustainable financing mechanism
contingent on output, and which guides the process of
health service purchase from health cluster operators in
alignment with global practices built around beneficiary
satisfaction. This mechanism guarantees the sustainability of
free health services for citizens.
• Provision of healthcare through health clusters, an
integrated health system modeled on international best
practice. This will raise the level of beneficiary care by
integrating primary, secondary and specialized care, and
eliminating duplication of services. Digital technologies,
e-health and information technology will be employed to
provide virtual (remote) healthcare aligned with best practices
and based on the verified implementation of the modern care
model across the Kingdom’s regions.
• A developed infrastructure for revenue management,
across integrated healthcare institutions in coordination with
local companies and all stakeholders.
• A private sector motivated to participate in the
transformation of the health sector and an increase in the
proportion of private sector investments in the health sector.
• An updated and effective system of enhanced traffic
safety and a reduction in damages resulting from
non-compliance with traffic rules, such as deaths, injuries,
violations, and general accidents.
26
Chapter 2:
Current Conditions
31
The Saudi population has seen a remarkable improvement in health since
the launch of Saudi Vision 2030, which has played a significant role in
raising the Kingdom’s readiness to face health crises, including:
• Swift implementation of effective measures to promote and protect
public health, thereby curbing the spread of disease.
• Rapidly establishing national and regional command centers to manage
the healthcare response to the novel coronavirus (Covid-19) pandemic and
to avoid overwhelming the healthcare system.
• Implementing a phased return to normal social and economic conditions
while prioritizing containment of the novel coronavirus (Covid-19).
This section outlines the current status of the four strategic objectives,
around which the program will build a national strategy for health
transformation incorporating all program participants.
2
4
In 2016, the number of people
with diabetes reached 4.6
million. This number is
expected to almost double by
2030, rising to 8.4 million.
Therefore, corrective measures
must be taken to reinforce
preventive health.
In addition to developing
agility in response to
emergencies, there is a need to
strengthen governance systems
that contribute to reducing
challenges to the health of the
population and the quality of
health services provided.
3
1 An average of 90,000 citizens annually suffer premature deaths from chronic diseases.
The average life expectancy is
5.2 years below the global
average. Therefore, preventive
care measures must be taken to
address multiple health
conditions, including heart
disease, stroke, diabetes,
respiratory diseases, mental
health, congenital diseases and
traffic accidents.
In parallel with its renaissance,
the Kingdom is also witnessing
an accelerated level of
population growth. The total
population is expected to
increase from
33
.4 million in
2018 to 39.4 million in 2030,
causing a rise in demand for
health services.
Core Challenges
The four strategic objectives of the health system address eight
challenges facing Saudi healthcare:
32
5
7
The high rate of deaths and
injuries resulting from traffic
accidents causes significant
human, societal and financial
losses. It also places substantial
pressure on the health sector due
to the costs allocated to
providing healthcare and
rehabilitation for the injured.
Traffic accidents cause not only
human loss and reduced life
expectancy in the Kingdom, but
also lifelong suffering as a result
of injuries that change the
quality of life of survivors.
Effort must be made to address
duplication of health service
provision and financing for the
same beneficiary.
8
6 The overlapping roles of the regulatory body and the health or financial service provider
leads to conflicts of interest.
This gives rise to the necessity
for a single regulatory body –
at the level of the health sector
– to enhance coordination
efforts between government
entities, implement monitoring
and accountability mechanisms,
and reduce absenteeism by
clarifying roles and
responsibilities.
The gap between supply and
demand in the health
workforce has led to an
increased dependence on
foreign labor.
33
Key factors contributing to these
challenges
• The current healthcare model tends towards treatment rather than prevention,
expending great efforts to deal with and treat diseases instead of focusing on
avoiding them in the first place through preventive principles. Similarly, the current
health system tends more towards finding solutions related to resources and
personnel than those related to beneficiaries.
• Poor integration between primary, secondary, and specialized care, and their
balanced distribution throughout the Kingdom. In addition, there is a need to
develop extended care services such as rehabilitation, long-term care and home
healthcare, as well as enhanced management of bed occupancy according to the
level of care required.
• The current infrastructure requires development to support electronic, digital and
virtual (remote) healthcare, which has become a necessity in light of current
challenges, especially those related to pandemics and health threats.
• Policies and legal frameworks need updating to allow for enhanced cooperation
and integration between government entities and the private sector as it pertains to
investments and implementation of tasks.
34
• Inaccessibility of required services to beneficiaries, in terms of both quality and
quantity, can be remedied with the implementation of treatment protocols and
mechanisms to measure healthcare outcomes. There is also a need for the sector to
empower its workforce and raise its capabilities through various development
programs, as well as improve efficiencies by reducing duplication of services for
patients who have dual eligibility, and who receive care across multiple healthcare
institutions.
• The Ministry of Health’s position as payer, regulator and service provider may
affect the governance of the health sector, in addition to impacting coordination of
healthcare provision and investments between the ministry and both the public and
private sectors. Separating functions will promote greater governance, transparency
and clarity of the roles and responsibilities through checks and balances to ensure
high-quality care and services to all beneficiaries and an appropriate response to
health security concerns.
• Traffic safety policies, protocols, laws and regulations are inconsistent with best
practices and comprehensive standards. Policy development will contribute to
enhancing traffic safety by setting standards for emergency service and rapid
response to traffic accidents in the Kingdom of Saudi Arabia.
35
Challenges
of the novel
coronavirus
(Covid-19)
In
di
vid
ua
l a
nd
so
cie
ta
l
Governance and
regulations
Health and
prevention
Data and
technology
Research and
developmentI
nf
ras
tru
ctu
re
an
d l
og
ist
ica
l
su
pp
or
t
Challenges arising from the novel
coronavirus (Covid-19) pandemic
The Kingdom of Saudi Arabia has achieved great success in combatting the novel
coronavirus (Covid-19) pandemic, proving its capabilities as one of the most effective
national responses to the virus and avoiding potentially vast losses.
Nonetheless, the Kingdom seeks to reach even greater levels of efficiency and
accomplishment, setting clear objectives to be a global pioneer of crisis
management and prevention.
The novel coronavirus (Covid-19) pandemic has presented many challenges to
nations around the world, affecting various sectors. The challenges facing the
Kingdom can be summarized in the following six branches:
36
38
Quick solutions and measures to combat
the novel coronavirus (Covid-19)
Infrastructure and Logistics:
• Developing and operating existing and unoccupied healthcare
facilities, including hospitals, laboratories and medical cities, to function
at full capacity, thereby bridging the gap and increasing response levels
to disasters and crises
• Utilizing the capabilities and expertise of the private sector to increase
capacity and infrastructure and meet the growing demand for health
services
• Implementing integrated mechanisms to track and monitor the
provision of medical supplies and security stocks
• Facilitating access to healthcare in remote geographical areas
Data and Technology:
• Accelerating the development and activation of the unified health
record and integration with the Absher
system
• Completing and accelerating the full implementation of virtual care
and telemedicine services
• Updating the national digital transformation strategy to include
advanced technology solutions
• Introducing more community interaction applications to follow up and
report cases
• Establishing a unified national database that issues status reports in
real-time
39
Health and Prevention:
• Monitoring the implementation of public health policies in densely populated areas
• Utilizing national health data to identify, address and contain segments of the
beneficiary population most exposed to health risks
• Accelerating the implementation of the modern healthcare model at the national
level to leverage its strength in raising awareness and educating the individual and
society
• Securing personal protection equipment for health practitioners as well as residents
to limit the spread of diseases and epidemics
• Enhancing medical capacity by supporting and activating the role of the non-profit
sector and volunteer programs
• Increasing local security stocks of essential medicines, supplies and medical devices
• Initiating the constituent elements of primary care, telemedicine and virtual
healthcare
Governance and Regulations:
• Establishing a national committee for preparedness and response to health
emergencies (epidemics, health disasters, etc.) under the leadership of the Ministry
of Health, defining committee roles and responsibilities to allow for rapid and
effective response to conditions
• Developing a health security and self-sufficiency strategy to guide all phases of
planning, managing, securing and nationalizing the needs of the Kingdom in terms
of medicines, medical supplies and equipment
• Recommending the necessity to initiate and strictly enforce regulations and laws
related to crowded areas, informal residential areas (slums), and worker housing
• Enforcing compliance of social distancing and limitations on crowding through the
use of modern technologies (cameras, drones, etc.)
• Addressing the phenomenon of illegal residents through a period of voluntary
disclosure
• Carrying out continuous sterilization of gathering places and imposing penalties
on violators
• Enhancing sources of financing and provisions allowing flexibility of procurement
policies in times of crisis, as was successfully accomplished during the Covid-19 pandemic
40
Research and Development:
• Documenting lessons learned from current and previous crises (successes and
challenges)
• Developing a strategy and governance system for research and innovation in the
relevant fields
• Utilizing the existing infrastructure and facilities in the field of research and
development and enhancing them to be ready to operate at full
capacity
• Consolidating research and development efforts among entities across various
fields, including legislation, governance, infrastructure, systems and human
resources
• Benefiting from the budget allocated for research and development and
employing it to serve national priorities
• Providing a centralized platform through which research and development
systems are integrated
• Focusing efforts on diagnosing and developing vaccines and new treatments
• Doubling national expertise and human capital in the field of research and
development and enhancing capabilities to meet growing needs
Individual and Societal:
• Enhancing the quality of awareness toolkits familiarizing citizens and residents
with health guidelines and the importance of regular check-ups
• Diversifying communication channels to reach the largest number of beneficiaries
of different nationalities, and ensuring that awareness-raising content is
standardized and produced in multiple languages
• Developing capabilities, efficiencies and skills related to planning and response to
crises and epidemics
41
Chapter 3:
Strategic Cornerstones of
the Health Sector
Transformation Program
Health Sector Transformation
Program
The Health Sector Transformation Program encompasses all health
entities of the Kingdom’s health ecosystem within a strategic
framework, directed towards key objectives and built on the
fundamental pillars of Saudi Vision 2030. The program consists of a
series of initiatives that aim to facilitate access to health services,
improve the quality and efficiency of care, enhance prevention against
health risks and raise the level of traffic safety.
44
1- Public health
3- Financial
sustainability of the
healthcare system
2- Effective delivery
of quality healthcare
4- Workforce size,
capabilities and
participation
Key areas
facing
challenges
Key areas facing challenges
Despite the many achievements made during previous years, since the launch of
Saudi Vision 2030, the health sector continues to face challenges. These challenges
fall into four general categories:
45
Average age of the individual
Level of non-communicable diseases posing a health threat
46
Challenges to Public Health:
Prevalence of major health risks in the Kingdom of Saudi Arabia
Traffic accident deaths and injuries
National response to health emergencies
47
Challenges to effective delivery of
quality healthcare:
The cost of healthcare is increasing at a faster rate than the increase in
national income, which reflects the challenges facing the current
healthcare model and the huge costs it bears for treating diseases
rather than preventing them in the first place.
Dual service provision and poor coordination between providers,
which leads to high costs of care stemming from the duplication of
services for the same case.
Poor incentives within the system to reduce cost and maximize value.
Limited participation of the private sector in financing and providing
public healthcare.
Challenges to financial sustainability
of the healthcare system:
48
Attracting and retaining doctors, nurses and other skilled
healthcare professionals in sufficient numbers to meet the needs of
beneficiaries.
Effective recruitment consistent with education, especially in light
of the transformation of the welfare system in the Kingdom of
Saudi Arabia.
Inconsistency of policies with employment practices in the health
sector, and an absence of linkage between reward and the
provision of value-based healthcare.
Challenges to workforce size,
capabilities and participation:
49
Better health
Better sustainability
Better care
Better workforce
Strategic Cornerstones
The program’s efforts to transform the health sector are built on four strategic
cornerstones:
50
Raising the average life expectancy in the Kingdom from 75
to 80 years by launching a series of initiatives within the
National Transformation Program aimed at reducing health
risks affecting average life span and quality of life
Focusing on preventive health to address the key health
challenges facing the population to reduce the health
burdens of non-communicable diseases such as diabetes,
cardiovascular disease, respiratory diseases and common
mental illnesses
Health is an integral part of government policies. Policies
are evaluated in terms of their health effects on Saudi
society and their effectiveness in promoting the general
health of beneficiaries.
HSTP aspirations for the future of the
health sector
The program aspires to bring about a qualitative transformation in the
healthcare and service provision model, and will work to provide significant
improvements in both public health and health services across these four strategic
cornerstones by 2030.
Better health Public health improvements will include:
52
Working to reduce the number of road deaths in line
with international figures: 10 per 100,000 people
Focusing on the principle of health security thereby
enhancing the Kingdom’s ability to respond to and
recover from health challenges, including epidemics,
outbreaks of infectious diseases, and negative health
consequences resulting from major natural disasters and
industrial accidents
53
Better care Radically improving beneficiary healthcare services by:
Enabling all beneficiaries to access healthcare services
regardless of place of residence or category of provider.
Restructuring healthcare delivery to ensure compliance with
the modern model of care, in line with international and
local best and good practices, and maximizing results while
minimizing costs.
Enacting a digital revolution in the field of healthcare, with
e-health and virtual care as the core, making it easier for
beneficiaries to access care services. Digital technologies,
including artificial intelligence, will be employed to help
make better clinical decisions, and hospitals will have
instant access to each patient’s comprehensive healthcare
record.
Providing a sufficient number of doctors, nurses and other
healthcare professionals, raising the level of their
capabilities, and ensuring their optimal distribution to
secure high-quality healthcare provision for all beneficiaries.
Providing quality healthcare to a national standard,
integrated across all service providers. In addition, standards
will be regulated and verified, and treatment results will be
measured to ensure caregivers are committed to providing
quality care.
54
Reducing ineffective care, and carefully evaluating
treatments in terms of clinical effectiveness and
affordability, thus enhancing spending efficiency.
Improving coordination between caregivers to ensure
optimized capacity and fully utilized facilities capable of
responding to urgent needs. This will contribute to
reducing duplication of healthcare service provision and
the resulting misuse.
Offering incentives to encourage all healthcare providers to
deliver better value. Financing mechanisms will be
significantly reformed and budgets based on health needs.
The private sector will be a strategic contributor to
healthcare and will play a major role in developing and
providing healthcare services to beneficiaries, including
those availing government health services.
Implementation of the modern care model will reduce the
cost burden on healthcare budgets, and the negative
impact of high-cost, widespread diseases such as diabetes
will be minimized.
Better sustainability Controlling healthcare cost by:
55
To achieve better health, better care, better sustainability and a better
workforce, the Health Sector Transformation Program will complete
implementation of the radical transformation approved for the Ministry of
Health, the largest healthcare provider in the Kingdom.
Better workforce Sufficient and capable health workforce
The workforce strategy will focus on ensuring the
appropriate number of health professionals, such as doctors
and nurses, and creating optimal multidisciplinary teams.
The strategy will also address recruitment and training of
personnel to provide effective, value-aware, quality services.
Effective workforce planning will take into consideration
demography, comorbidities, healthcare use, and GDP
growth, and relies on an understanding of the healthcare
models, skills and professional expertise required to meet
health needs. Planning should incorporate the workforce as
a whole and take into account all factors related to supply
and demand.
Results will reveal a clear correlation between the reward
system and value-based care, so that the income of health
professionals increases according to their contribution to
the provision of value-oriented care and the achievement of
desired objectives.
Mechanism for achieving future
aspirations
56
This is accomplished through strategic reform initiatives designed to
achieve the target conditions:
1. Health sector governance initiative
2. Modern care model initiative
3. Institutional transformation of healthcare facilities initiative
4. National health insurance and new financing models initiative
5. E-health initiative
6. Workforce initiative
7. Private sector participation initiative
Health governance is being radically reformed to advance the objective
of improving the quality and value of care, in line with international best
practices. As such, clear and distinct markers will differentiate the roles of
healthcare regulator and monitor, provider, and financier.
The role of the Ministry of Health will be redefined as the regulator and
monitor of the health sector. On this basis, the Ministry will cease its role
as a care provider in order to perform its future function, which is to
ensure that all beneficiaries receive care that meets the standards set at
the national level for quality and that the health ecosystem works for the
benefit of the beneficiary. The ministry will provide licenses to healthcare
providers, conduct inspections to safeguard national standards of quality
and value, and intervene effectively to protect the health of
beneficiaries, ensuring that the healthcare provider is accountable for
meeting the established standards.
The ministry will also play a role in ensuring that the system’s needs are
understood and acted upon, bringing an end to service duplication, securing
investment in low-access areas, and defining the national health strategy,
especially as it relates to beneficiary health and disease prevention.
Health sector governance
57
58
Health clusters will work to fundamentally change beneficiary care to meet the
standards set by the modern care model in the Kingdom of Saudi Arabia. All hospitals
and primary care centers will be allocated to health clusters across the Kingdom, each
serving approximately one million people. Each cluster will consist of primary care
centers, general hospitals, and specialized services, so that any beneficiary
undergoing examination in the cluster may avail of all the required services through
an integrated administrative system. Cluster administrations are responsible for
ensuring the flexibility of service procedures, the speed of service delivery, and
beneficiary satisfaction.
In line with Saudi Vision 2030, and as part of endeavors to develop a healthcare
system that contributes to empowering beneficiaries, the model of care was designed
based on the following principles:
• Enabling beneficiaries and their families to control their health
• Providing beneficiaries with the necessary information, as part of their treatment,
and raising their awareness to the ways they may have better control of their health
• Seamless integration of the health system on the beneficiary level
• Maintaining a holistic view of the health of the individual by taking a preventive
approach rather than a solely curative approach
• Providing treatment that focuses on both the curative results and the treatment’s
suitability to the beneficiary – neither suboptimal nor excessive
Modern Care Model
Impact on the beneficiary
Care closer to
home
Reduction in
waiting time for
services
Increase in
health system
capacity
Comprehensive healthcare
within one system (cluster)
Healthcare based on end results
and beneficiary satisfaction
٥٨
Institutional Transformation of Healthcare Facilities
Hospital
This will result in enhancing the value and quality of the beneficiary
care experience, improving upon the current state:
Care Systems
Chronic disease care Managing chronic diseases (e.g., diabetes)
Promoting health and disease prevention
Providing care to beneficiaries in the final stages and to their families
Providing care for pregnant women and newborns
Treating emergency and critical cases
Providing elective care for non-critical cases
Preventive care
Palliative care
Child and maternity care
Urgent care
Scheduled care
The modern care model ensures safe, swift, beneficiary-centered care
provided efficiently and effectively
Physical
health
Mental
health
Social health
60
Hospital
Bone surgeonCardiologist
Vascular
surgeon
Endocrinologist
Physiotherapist
Emergency
doctor
Pulmonologist
Nutritionist
Optometrist
Heart
disease
Diabetes
Arthritis
Hypertension
Pharmacy
Primary care
facility
RadiologyLaboratory
61
To the future state:
Unified
health
record
Pleasant care experience in the target system
Meet Sarah, a 70-year-old woman
Current healthcare delivery system (non-integrated)
1- Healthcare focus on
hospital treatment
provided after the
onset of illness
2- Limited efficiency
and quality (access to
costly, repetitive,
uncoordinated care)
3- Absence of
appropriate tools to
enable beneficiary
effective access to the
necessary healthcare
1- Healthcare focus on
prevention and
treatment
2- Efficient, quality
care benefitting from
integration and
coordination to
provide treatment at
the lowest cost
3- Digital tools
enabling the
beneficiary to easily
access care
Home care
PharmacyRadiologyLaboratory
Hospital
Cardiologist
Bone surgeon
Optometrist
Endocrinologist
Nutritionist
Vascular
surgeon
Physiotherapist
Emergency
doctor
Pulmonologist
Primary
care
facility
Home
care
National Health Insurance and new
financing models within the Ministry
of Health’s transformation strategy
Establishing a national body (namely, the Center for National Health Insurance) to
pay for healthcare costs serves to enhance the value and quality of health services.
The entity will be funded by the Ministry of Finance, securing the strategic purchase
of services, and will form contractual agreements with health clusters to provide free,
quality care based on needs, whereby the citizen will enjoy 100% comprehensive
medical coverage of all healthcare requirements, free of charge.
The Center for National Health Insurance will be established as the only national body
covering citizen healthcare costs, in line with international best practice, creating a
clear distinction between the caregiver and the financier. This enables the financier to
focus solely on paying the value of the service, ensuring quality of care while
continuing free treatment for citizens.
In effect, the center will receive funding from the Ministry of Finance to meet the
needs of citizens benefitting from health services. In addition, a benefit package will
be created and made available to all citizens grounded firmly in evidence confirming
clinical and cost effectiveness to ensure the provision of appropriate value care. The
center will also provide a new financing mechanism for the purchase of health
services from service providers. This will contribute to achieving universal health
coverage, whereby all citizens receive the health services they need without charge.
The Center for National Health Insurance will finance health clusters through
customized payment structures according to the needs of each, which will be based
on an accurate national census implementing international best practice. In addition,
the center will create a pace-of-change policy to drive adoption among health
clusters of a spending methodology built around individual targets, ensuring the
continuity of current services while providing strong incentives for development.
For financing the remaining health sector services, the procurement strategy will be
determined based on the health sector transformation strategy.
62
E-health Initiative
1
2
3
4
5
Facilitate the introduction of
significantly different approaches
to care
Facilitate entry into the health
field for new, non-traditional
service
providers
Provide disease forecasting tools
Empower personal care management
Help alert health cadres in the event
of infectious disease outbreaks
E-health is the effective, safe use of information and communication technologies
to support health and its related objectives. This includes healthcare services, health
monitoring, health literature, health education, knowledge and research.
Effective e-health implementation will:
63
Investment in e-health in the Kingdom is expected to offer multiple benefits,
including the following:
E-Health Initiative
A flexible personal experience, and responsible approach. E-health
offers beneficiaries new ways to interact with and improve the health
system. A spatial healthcare system may be further enhanced with virtual
applications, improving access to services at the time and place required
by the beneficiary. E-health also allows individuals more control over their
lifestyle and health, and builds a greater sense of responsibility.
Safer, more effective services. Quality healthcare relies on full
knowledge of the patient’s health status so that critical healthcare
decisions may be made based on available diagnostic and clinical
information, helping to avoid costly medical errors and negative results
due to absent or erroneous information.
Increased productivity for healthcare practitioners. Given the vital
role played by skilled medical practitioners in the Kingdom, e-health will
offer the health workforce the tools necessary to enhance productivity by
automating administrative tasks, supported by digital information and
knowledge bases in the provision of the required healthcare.
Effective, integrated system. Efficient use of healthcare resources is the
key to achieving a sustainable health system. E-health will enable better
integration of the various healthcare systems across facilities and allow the
sharing of data, ensuring access to consistent information for curative care
and reducing duplication of services.
Creation of a new knowledge industry. Implementing digital technologies
is a challenge that requires a trained workforce. Global experience has shown
that e-health programs have created tens of thousands of job opportunities, in
addition to a large number of startups in the field of health informatics.
Moreover, investing in e-health benefits not only the health system, but also
the economy of the Kingdom of Saudi Arabia as a whole.
64
٦
7
8
9
1
2
3
4
5
6
Strengthening compliance and application of quality and
safety standards in health practices
Facilitating the upload of performance reports and the
monitoring of target achievements
Providing residents with personal care tools to assist them
in making better health choices, raise levels of health
education, and allow interaction with the health system
Integrating continuity of care at all levels within the
modern care model and supporting the flow of health
information
Ensuring effectiveness of operations and service provision
results
Monitoring and ensuring optimal use of healthcare
resources such as manpower, assets and services
Allowing access to high-quality data that supports medical
research, policy development and decision-making
Achieving targeted effectiveness in financial and human
resource services, informatics services, supply chain
management, supporting medical services and quality control
Approving service costs, patient accounts, revenue
cycles and claims management
E-health is one of the main enablers of the healthcare system within the health
sector transformation strategy and would achieve the following:
65
E-health Strategy and the role of the Saudi Health
Information Exchange
The Kingdom of Saudi Arabia’s e-health strategy consists of important elements
relevant to the health sector as a whole, incorporating the Ministry of Health,
other government services and the private sector. The Saudi Health Information
Exchange constitutes an essential mechanism for the integration of digital
capabilities across all sectors of the health system.
66
6
7
8
9
10
1
2
3
4
5
The need for more
health professionals
Raising the ratio of
clinical staff
(especially nursing)
to population
Increasing the
productivity of
health staff
Promoting
Saudization in
medical and
nursing roles
Increasing retention
rates of nursing staff
Workforce Initiative
In the coming period, we must focus on several points, including:
Improving data and
statistics on the
Saudi health
workforce
Establishing a process
of regular reviews of
clinical staff
performance and
safety
Establishing a periodic
renewal system for
medical and clinical
staff licenses
Focusing on resources and
expertise in health sector
support professions such as
finance, health economics,
planning and the legal field
Increasing the number
of subsidized medical
training locations for
graduates
67
Understanding workforce capabilities and demand
The Health Sector Transformation Program is currently preparing a national
strategy for the health workforce that would identify and analyze categories of
current workers in the Kingdom. This includes, but is not limited to,
understanding the major specialties and subspecialties in relation to region and
facility for all categories of employees. This measure will support the
development of annual forecasts for workforce demand and determine the gap
between supply and demand annually until 2030. The forecast will take into
account supply and demand drivers, including changes in technology, the
modern care model, and the corporate system.
Building workforce capacity
Workforce challenges necessitate efforts to strengthen the capabilities of national
healthcare practices in all major and subspecialties as this will ensure a fair distribution of
skills throughout the Kingdom based on needs. This includes the ambitious endeavor of
encouraging qualified Saudi health workers (whether currently in Saudi Arabia or
abroad) to return to the health workforce in the Kingdom, a path that may include
(re)education and training. Also under development is an initiative to improve the
attractiveness of nursing and other healthcare professions.
68
1
2
3
4
5
Pharmacy
Extended care
Rehabilitation
Long-term care
Home care
6
7
8
9
Primary healthcare
Radiology
Laboratory services
Service launches for
hospitals and medical cities
Serving the objective of capability building, a national initiative covering
all major and subspecialties will target the needs identified through the
planning phase. The initiative will include providing accreditation to
teaching hospitals that provide training and establishing family medicine
academies across the Kingdom.
Private Sector Participation
Private sector participation falls within the strategic context of detailed
information in nine main areas:
Building workforce capabilities
69
Health sector governance
and regulation
Health sector value and cost
Private sector
participation
Workforce and
education
Integrated care
provision
Economic
contribution
Data and digitalization Public health and health
security
HSTP Strategic branches
Based on the four cornerstones of better health, better care, better
sustainability, and a better workforce, the Health Sector Transformation
Program exists within a framework closely related to the entire health sector.
Accordingly, the program will work with all stakeholders to develop a strategy
across the following eight strategic branches:
70
Health Sector governance and
regulation
Health Sector value and cost
1
2
3
1
2
3
Employing governance in the health sector to achieve the target scenario
by 2030
Regulating the health sector to ensure continuous access to high-quality,
value-based healthcare
Improving healthcare capacity to ensure value and eliminate duplication
of service provision
Need-based financing for government care providers and strong
incentives for cost control and efficiency
Financing and reimbursement, including funding agency model and
financial reforms to raise efficiency while maintaining free treatment for
citizens
Health benefits package based on actual needs of beneficiaries to ensure
value and cost-effective care
71
Private Sector participation
Workforce and Education
1
Private sector roles required for the financing and operation of publicly
funded healthcare services
2
Main areas of opportunity for the private sector
3
Practical models to maximize the role of the private sector
1
Employing data to gain a clearer understanding of the major healthcare
professions required to deliver the modern care
model
2
Drafting a supply and demand strategy for increasing local supply,
recruitment and employee retention
3
Coordinating with academic institutions to secure market needs
4
Linking incentives and compensation to value-based care delivery for
healthcare professionals
72
Integrated Care provision
Economic Contribution,
includes the following:
1
Building service provider models based on the requirements for
integrated care
2
Transitioning all beneficiaries in the Kingdom from traditional healthcare
provision to the modern care model
3
Focusing on public health and prevention services
4
Improving the quality of care by establishing centers of excellence
1
Attracting investment
2
Research and innovation
3
Biotechnology
4
Saudization
73
Data and Digitalization
Public health and health security
1
Unified electronic health records for all
1
Public health (health in all policies) – health impact assessment and
promotion of health quality
2
Crisis preparedness – ensure the Kingdom’s readiness for any health
challenges
3 Roads and traffic – measures needed to reduce death and injury rates
4 Hajj and Umrah – health measures to support a safe pilgrimage
2
Virtual care and e-health models
3
Cybersecurity and health data protection
4
Information systems and information technology infrastructure
74
Chapter 4:
Initiatives Portfolio
Priority Initiatives
A detailed study was conducted to outline the lessons learned from the novel
coronavirus (Covid-19) pandemic, the findings of which necessitate the
following steps are taken:
1. Developing a clear methodology for crisis response and a national plan for
crisis and disaster management incorporating health regions and clusters
2. Developing the capability to establish shelters, hospitals, testing centers and
temporary tents to enhance response to disasters and crises
3. Improving the infrastructure of remote residential areas to improve
healthcare coverage, enhance communication, and enable remote services,
especially health services
4. Investing in artificial intelligence, simulation models and data analysis tools
to support decision-making
5. Investing in and providing various technological solutions to aid response to
emergencies
6. Accelerating the full implementation of virtual care and telemedicine services
7. Surveying and using population health data to identify various beneficiary
segments and focus on the most vulnerable
8. Nationalizing the pharmaceutical, medical devices and supplies, and
vaccination industries
9. Empowering and promoting national industry and local content in terms of
diagnosis, vaccination and treatment
10. Developing a strategy for change management and communication during
crises and disasters, including objectives, indicators, initiatives and
implementation plans
11. Developing a national awareness program to cover cultural and behavioral
practices (distancing, compliance, personal hygiene and combatting the spread
of infection)
78
Based on the lessons learned, the following initiatives have been proposed as
priority initiatives:
• HSTP strategy preparation initiative
• Health innovation initiative
• National e-health initiative
• National health security initiative
79
Update the scope of the current strategic plan to include the entire
health sector in order to improve and promote healthcare services in
the Kingdom. The themes of this strategy are the health of the
population, providing safe, high-quality care in line with international
best practices, financial sustainability and transparency by
implementing value-based healthcare concepts. This is in addition to
the workforce pillar, that aims to secure and qualify the right leaders,
health professionals and administrators, and digital and innovative
services. Among the objectives of the strategy is to reach a flexible
system that enables effective governance to counteract and manage
health challenges.
Develop national innovation for the health sector to create an
ecosystem that accelerates and drives innovation, contributing to
sustainability, broader opportunities and increased products and
solutions with added economic value in the health sector. Innovation
areas include biotechnology, preventive, curative and rehabilitative
services, public health and e-health, to name a few. The innovation
system is essential to achieving transformation in the health sector.
وصف المبادرة Initiative descriptionالجهة القائدة
Designing an integrated plan for virtual and remote healthcare to
enable the transformation of the health sector at the level of service
providers in all sectors. Constituent elements:
• Remote care follow-up services
• Artificial intelligence to support decision-making in the health field
• Saudi Technical Health Academy
• Individual platform for electronic self-care
• Remote monitoring for advanced home care
• Virtual clinic platform
This includes setting standards and methodologies to assess the current
situation and ways to improve the virtual health and infrastructure of
sectors in the Kingdom, studying global and regional best practices,
and then defining the framework and principles for integrated design.
Accordingly, a comprehensive execution plan will be defined for each
of the elements, as well as the functional and technical requirements to
develop a roadmap for the implementation stages. The scope of work
also includes developing a stakeholder participation plan and a
communication plan, in addition to the initiative’s governance
framework to define roles, responsibilities and working relationships
between the concerned authorities.
This will be followed by the adoption of the model by service providers,
its development and launch according to the approved roadmap.
Support will be provided and work progress and performance indicators
monitored through interactive electronic control panels based on the
governance model, which will reveal potential challenges, risks and
support requirements, and address them in a timely manner.
Leading party
81
Improving
sustainability and
financial transparency
Improving the patient
experience
Increased clinical
(medical) effectiveness
Safety in health
facilities
HSTP strategy preparation
Improving
sustainability and
financial transparency
Improving the patient
experience
Increased clinical
(medical) effectiveness
Safety in health
facilities
Health innovation
Strategic pillars Initiative link to
level 3 objective
Initiative name
Increase geographical
coverage and expand
health service
Facilitate access to
emergency medical
care
Accessible health
services
National e-health
2.1.2
Improve quality and
efficiency of healthcare
services
2.1.2
Improve quality and
efficiency of healthcare
services
2.1.1
Facilitate access to
healthcare services
80
Priority Initiatives
Raising the level of
preparedness for
health disasters
National health security
2.1.3
Promote prevention
against health risks
Strategic pillars
Initiative link to
level 3 objective Initiative name
Priority Initiatives
82 83
Preparing a national health security system based on a comprehensive
strategic approach to assess, prevent, prepare, respond to and recover
from health threats, emergencies and disasters (such as disease
outbreaks, epidemics, natural disasters, chemical, biological and
radiological emergencies, etc.). Developing self-sufficiency and
building the foundations for long-term resilience in order to reduce
dependence on other countries in the health and pharmaceutical
fields, reduce exposure of professional health personnel, fill gaps in
critical and qualitative medical services, and develop scientific research
in the field of health security.
وصف المبادرة Initiative descriptionالجهة القائدة Leading party
Program ambitions and commitments, as well as strategic pillars
and considerations, have been translated into a number of
initiatives designed to fulfill its objectives. These initiatives
constitute a fundamental building block in achieving the
aspirations of Saudi Vision 2030 in the following ways:
Current Initiatives
Impact on
population
1
Contribution to
improving the
macroeconomic
index
3
Contribution to
attracting the
private sector
2
Contribution to
level 3 and 4
indicators
4
85
Initiative description Leading party
The initiative aims to ensure access to health services for all
beneficiaries, and to achieve an average number of 4 visits per person
per year (a global indicator) by increasing the number of visitors to
primary health care services. This will be achieved by the following:
• Implementing the initiative to achieve geographical coverage of
primary healthcare services for all beneficiaries and facilitating access
to primary healthcare through the implementation of extended hours
and urgent care for 312 selected health
centers
• Upgrading existing primary healthcare center facilities
• Providing modern, advanced medical equipment for diagnosis and
treatment
• Developing the capabilities of staff specialized in the field of family
medicine and various medical specialties (children, internal medicine,
obstetrics and gynecology etc.) to provide health services to
beneficiaries
• Employing data and implementing the health information system
based and unified medical record
• Facilitating healthcare access for beneficiaries in remote areas
through mobile clinics
• Developing existing centers to become consulting centers offering all
medical specialties (pediatrics, internal medicine, surgery, gynecology
and obstetrics, ENT, dermatology)
• Providing exploratory examinations for early detection of diseases
among school students, and provision of curative and preventive
services through school health programs
Improving the performance of hospitals and specialized centers of the
Ministry of Health by increasing production efficiency and using
resources to reduce waiting time in clinics, in addition to reducing the
waiting period for patients who receive emergency or urgent medical
care.
The initiative was launched to improve the quality of healthcare
services provided to patients, with a focus on bridging gaps through a
set of performance development projects in the following areas:
mental health, public health, home care, long-term care, laboratories,
emergency care, diabetes, kidney disease, heart disease, tumors, and
organ transplantation. Other aims include improving general behavior
and growth in hospitals and specialized centers, contributing to the
development of leaders and their workers through improvement
projects and continuous evaluation and follow-up.
The initiative’s work includes the following:
• Defining the scope of work and performance indicators for health
evaluation in hospitals and specialized centers
• Developing and raising the efficiency of hospitals and specialized
centers in line with best practices in project and quality management
• Training workers in performance indicators and methods of collecting
the data necessary to measure these indicators
• Determining the baseline for key performance indicators
• Transferring knowledge to workers in hospitals and specialized
centers through on-the-job training.
87
Strategic pillars Initiative link to
level 3 objective
Initiative no. Initiative name
Facilitate access to
consultation with the
required medical
specialty
Increase geographical
coverage and
distribution of health
services
Accessible health
services
Upgrade
infrastructure and
health capacity
1-16-027-1207
Reformation and
restructuring of primary
healthcare services
Facilitate access to
consultation with the
required medical
specialty
Upgrade
infrastructure and
health capacity
1-16-027-0948
Improved performance
of Ministry of Health
hospitals and medical
centers
Current Initiatives
2.1.1
Facilitate access to
healthcare services
2.1.1
Facilitate access to
healthcare services
86
Current Initiatives
Upgrade
infrastructure and
health capacity
1-16-027-1209
Improved performance
of Ministry of Health
primary health care
centers
Facilitate access to
consultation with the
required medical
specialty
1-16-027-1208
Saudi Center for
Medical Appointments
and Referrals
Facilitate access to
emergency medical
care
1-16-0
97
-1211
Raised level and
support of emergency
medical services
Strategic pillars Initiative link to
level 3 objective
Initiative no. Initiative name
2.1.1
Facilitate access to
healthcare services
2.1.1
Facilitate access to
healthcare services
2.1.1
Facilitate access to
healthcare services
88
This initiative aims to improve the quality of health service provided in
primary care centers through a set of performance improvement
projects in a number of categories: quality, radiology, information
technology, diabetes, and laboratories for 671 health centers by
studying the current status of the centers and identifying and
implementing development projects. The initiative would also
contribute to the training of leaders and workers through follow-up in
all projects and continuous evaluation.
The initiative’s work includes the following:
• Defining the scope of work and performance indicators for health
evaluation in hospitals and specialized centers
• Developing and raising the efficiency of hospitals and specialized
centers in line with best practices in project and quality management
• Training workers in performance indicators and methods of collecting
the data necessary to measure these indicators
• Determining the baseline for key performance indicators
• Transferring knowledge to workers in hospitals and specialized
centers through on-the-job training
The initiative aims to launch the Saudi Center for Medical Appoint-
ments and Referrals to manage the movement of medical referrals and
make reservations for all patients referred between health facilities
inside and outside the Kingdom at all levels of medical service
provision.
This will be done by designing a unified electronic system for referral
and reverse referral so that appointment reservation becomes depend-
ent on real and transparent information regarding the capabilities of
hospitals (health specialties, medical personnel, family status, etc.) for
all health sectors (the Ministry of Health, other government sectors,
the private sector) in real-time when referring or booking appoint-
ments.
The initiative aims to raise the level of emergency preparedness by
increasing the number of emergency teams and centers to ensure
geographical coverage, in line with regional and global standards,
which in turn will improve the response time to emergency alerts. To
achieve this, the initiative aims to open and operate 140 new aid
centers by building 35 emergency centers, providing 35 aid centers
through the third sector, and renting 70 ambulance centers with the
provision of all operational requirements. This raises the number of
emergency centers in the Kingdom and contributes to improving
ambulance response time, reducing the gap with global standards for
those centers.
Initiative description Leading party
89
Upgrade
infrastructure and
health capacity
1-16-027-08
96
Development of
intensive care and
emergency depart-
ments to ensure service
speed and quality
Facilitate access to
emergency medical
care
Increase geographical
coverage and
distribution of health
services
Upgrade
infrastructure and
capacity
1-16-027-1206
Societal participation
strategy for select
healthcare professions
Facilitate access to
emergency medical
care
1-16-097-1210
Speed of ambulance
response to emergency
calls
Strategic pillars
Initiative link to
level 3 objective Initiative no. Initiative name
2.1.1
Facilitate access to
healthcare services
2.1.1
Facilitate access to
healthcare services
2.1.1
Facilitate access to
healthcare services
90
Current Initiatives
The initiative aims to raise the level of medical service provision in
hospitals through the following scope of work:
• Developing the infrastructure of intensive care departments for 13
hospitals in various regions of the Kingdom. The development includes
17 intensive care departments (adults, children, and newborns), and a
revision of the targeted segments according to modern health
standards. For example, but not limited to, the use of single rooms for
each care bed, the use of anti-bacterial materials to prevent the spread
of infection, the shift from the traditional system to the flexible
system, which will ensure optimum utilization of the capacity in the
departments and reduce the period of hospitalization.
• Medical operation of adult and neonatal intensive care units with 9
intensive care units (adults and newborns) in the following regions:
Riyadh, Mecca, Al-Mukarramah, Eastern Province, Medina, Jazan, and
Al-Qassim, in cooperation with the private sector
The initiative aims to increase the attractiveness of the nursing
profession and encourage citizens to change behavior towards nurses
and the negative societal image of the nursing profession. This is
achieved by revising current policies and regulations, drawing up a
strategy for developing nursing practices, in addition to carrying out
several campaigns aimed at supporting community participation and
awareness.
The initiative aims to increase the field operational technical capacity
of the Red Crescent and raise its readiness through integration and
coordination with the health sectors (public and private) by linking
the technical systems between them. Technical support will be
provided to follow up the emergency situation from the time the
report is received until the arrival of the emergency case at the
hospital in question using an electronic tracking system for ambulance
vehicles. The initiative also covers field ambulance services manage-
ment systems (electronic paramedic) to facilitate the process of
automating and recording reports and diagnosing an emergency case,
as well as training the concerned about these systems in 13 administra-
tive regions.
Initiative description Leading party
91
Safety in health
facilities
Increase clinical
(medical)
effectiveness
Improve the patient
experience
Improve
sustainability and
financial
transparency
1-16-027-0976
Medical insurance and
healthcare service
acquisition initiative
Improve
sustainability and
financial
transparency
1-16-027-0987 E-health
Safety in health
facilities 1-16-027-1216
Ensure provision of
basic safety standards
and emergency reforms
of health facilities
Strategic pillars
Initiative link to
level 3 objective Initiative no. Initiative name
2.1.2
Improve quality and
efficiency of healthcare
services
2.1.2
Improve quality and
efficiency of healthcare
services
2.1.2
Improve quality and
efficiency of healthcare
services
92
Current Initiatives
The initiative aims to establish and operate the Center for National
Health Insurance to create a sustainable financing mechanism so that
the center purchases health services from health cluster operators in
accordance with the global best practices, which includes beneficiary
satisfaction and improvement of indicators. This mechanism ensures
free, sustainable health service provision to beneficiaries while creating
incentive programs for service providers.
The initiative will enable comprehensive transformation, including
free health insurance for beneficiaries, institutional transformation,
and public-private partnerships. The main objective of the initiative is
to improve the effectiveness and efficiency of the healthcare sector
through IT and digital transformation. The initiative will work to
achieve e-health implementation in health facilities, raise the
efficiency of e-health outputs by establishing a secure information
system and, work on linking all systems to create a unified electronic
medical record.
The initiative aims to unify standards and procedures that guarantee
the safety of health facilities, their employees and beneficiaries, as
well as ensure the effectiveness of equipment and support systems to
achieve safety standards. Verification is conducted to ensure the
availability of the necessary systems and rehabilitate the facilities in
line with safety standards in design and engineering infrastructure, in
addition to qualifying and training the workforce to activate these
systems and apply the necessary safety procedures, including:
• Verify the availability and effectiveness of safety and fire protection
systems in health facilities
• Rehabilitation of health facilities’ design and engineering
infrastructure to achieve safety standards
• Plan effective safety and fire protection systems
• Train and qualify workforce
Initiative description Leading party
93
Safety in health
facilities
Increase clinical
(medical) effectiveness
Improve the patient
experience
Improve sustainability
and financial
transparency
1-16-027-0985
Institutional
transformation of
healthcare facilities
Improve sustainability
and financial
transparency
Improve the patient
experience
Increase clinical
(medical) effectiveness
Safety in health
facilities
1-16-027-0908
Modern healthcare
model
Strategic pillars
Initiative link to
level 3 objective Initiative no. Initiative name
2.1.2
Improve quality and
efficiency of healthcare
services
2.1.2
Improve quality and
efficiency of healthcare
services
94
Current Initiatives
The Ministry of Health is currently carrying out the following tasks:
1. Providing healthcare services
2. Financing healthcare services
3. Supervising and regulating the health sector
The initiative aims to provide higher quality healthcare services to
standards of efficiency and productivity, and to separate the provision of
health services from the Ministry through the formation of 22 health
clusters divided into two separate groups. One group of 5 clusters will
be launched by ministerial decree as a test phase to prove the concept of
institutional transformation. The second group consists of 17 clusters, to
be launched and operated through five units across business regions –
which will be established – supervised by the health holding company –
which will be established as a government-owned company – so that the
work units oversee the health clusters through a centralized
decision-making process. The groupings will eventually be integrated
into healthcare organizations (Accountable Care Organizations) to
provide services. Shared functions include: supply chain, information
technology, laboratories and engineering, in addition to the
establishment of centers of excellence to cover human resources,
financial and technological information, and revenue cycle management
for cost coding. The health clusters include a number of service provision
facilities including primary care centers, hospitals, medical cities, and
specialized hospitals. All necessary decisions are taken by the
administration. These include all actions and decisions related to the
cluster, such as employment and maintenance decisions and transferring
beneficiaries between facilities, and transfer of health personnel
between clusters according to specialty requirements, allowing swift
responses to situations without the need to confer with the Ministry.
The initiative aims to map the pathways of healthcare provision
services starting with preventive care – which promotes health and
makes the beneficiary a partner in the responsibility of maintaining
health – and extending to all paths that address various health
challenges. This is done by designing six pathways to healthcare
provision: preventive care, emergency care, voluntary care, maternity
and childcare, extended care, and palliative care.
Initiative description Leading party
95
Improve sustainability
and financial
transparency
1-16-027-1162
Planning, developing
and management of
health sector workforce
Improve sustainability
and financial
transparency
1-16-028-
101
2
National Health
Information Center –
horizontal national
e-health initiatives
Strategic pillars
Initiative link to
level 3 objective Initiative no. Initiative name
2.1.2
Improve quality and
efficiency of healthcare
services
2.1.2
Improve quality and
efficiency of healthcare
services
96
Current Initiatives
The initiative aims to study the current human resources within the
health sector in various disciplines and professions and identify the
gap between supply and demand to build a national labor force
strategy to bridge the gap. This is achieved by linking educational
outputs and the labor market through workforce planning to identify
sector demand in various disciplines of health, technical and human
capital management in cooperation with educational and health
institutions, and develop programs to build the capabilities of health
workers.
There exist many healthcare service providers under various executive
bodies that include university hospitals, hospitals of the ministries of
defense and the interior, as well as hospitals and facilities of the
Ministry of Health. As electronic connectivity between the different
agencies is not available, this causes a doubling of time and financial
effort – such as the duplication of examinations and treatments – as
well as the difficulty of estimating service requirements or their
availability in the absence of electronic connectivity.
Therefore, the initiative aims to build an electronic infrastructure that
connects all service providers in a way that enables digital
transformation in the health sector, allows the sharing of health
information and avoids re-examination and duplication of services.
The scope of this initiative’s work covers developing the health
infrastructure for digital transformation in the Kingdom through
optimal planning with health and relevant authorities to define the
basic components needed by the health sector to build an
infrastructure capable of meeting the aspirations of e-health through
4 systems:
• Unified electronic health record
• Health knowledge management
• Prescription dispensing management system
• National health payments system (DRG)
The initiative will review the relevant existing legislation and the new
legislation needed for the optimal application of e-health in
coordination with the authorities related to digital transformation in
general, and healthcare providers in particular. Implementation of the
horizontal e-health model and policies will follow, in coordination
with all relevant sectors, and launch of the national electronic link
between sectors operating within the Saudi health system. The
initiative will also follow up on quality and sustainability by evaluating
the results of the various horizontal national e-health initiatives in
coordination with the relevant authorities.
Initiative description Leading party
97
Strategic pillars
Initiative link to
level 3 objective Initiative no. Initiative name
Current Initiatives
Improve sustainability
and financial
transparency
1-16-098-1229 Operational
transformation
Improve sustainability
and financial
transparency
1-16-027-1215 Health sector
governance
Improve sustainability
and financial
transparency
1-16-027-1218 Community
participation
2.1.2
Improve quality and
efficiency of healthcare
services
2.1.2
Improve quality and
efficiency of healthcare
services
2.1.2
Improve quality and
efficiency of healthcare
services
98
Initiative description Leading party
The King Faisal Hospital and Research Center is in the process of
transitioning to an independent, not-for-profit institution. The
operational readiness of the institution is a major and fundamental
requirement to provide support of the objectives of the Health Sector
Transformation Program in the Kingdom.
Therefore, the initiative aims to equip the organization’s operational
services, including clinical, financial, electronic and human resource
services to apply international best practices to achieve efficiency and
effectiveness, and enable the hospital to be self-sustaining.
The initiative’s scope of work includes the following:
• Developing medical, clinical and nursing services to raise the quality
and efficiency of patient care
• Establishing the Center for Medical Services and Genomics
• Developing research and academic cadres to continue as a world
leader in research and education
• Developing the information technology infrastructure of the
institution
• Designing and implementing the target operating model
• Improving the operational efficiency of the facilities to rationalize
consumption
• Improving and developing supply chain projects
• Developing operational services, including financial and electronic
services and human resources
The governance initiative aims to clearly define the functions and
responsibilities of the regulatory roles in the health sector according
to their respective competencies and scope in order to reduce
duplication of tasks or functions, and bridge the gaps in the health
system.
The participation of the charitable and non-profit third sector in health
care services is limited in comparison to the sector’s contribution
globally in this field, especially considering the size of charitable giving
in the Kingdom. Therefore, the health sector must strengthen
supportive legislation for donors to enable them to provide healthcare
services. The initiative aims to create an enabling environment that
supports and stimulates social components to participate in the
provision of healthcare services.
The initiative’s scope of work includes creating and developing
incentives and enabling systems and regulations for donations,
endowments, volunteering in the health field; establishing entities at
the national level and at the level of the Ministry of Health; and
designing channels for directing community resources and capabilities
to meet priority health challenges.
99
Strategic pillars
Initiative link to
level 3 objective Initiative no. Initiative name
Current Initiatives
Safety in health
facilities 1-16-027-1214
Saudi Center for Patient
Safety
Improving
sustainability and
financial transparency
1-16-098-1228
Organizational
development of King Faisal
Specialist Hospital and
Research Center Gen. Org.
Increase clinical
(medical) effectiveness 1-16-028-1224
Initiative for the promotion
of organ donation in
intensive care units and
donator hospitals
2.1.2
Improve quality and
efficiency of healthcare
services
2.1.2
Improve quality and
efficiency of healthcare
services
2.1.2
Improve quality and
efficiency of healthcare
services
100
Initiative description Leading party
The Saudi Center for Patient Safety will be fully designed and launched
to promote a national culture of patient safety reporting and increased
awareness on issues related to patient safety as well as improving
capabilities and training.
To keep pace with the requirements of transforming the King Faisal
Specialist Hospital and Research Center into a non-profit institution, a
set of measures must be implemented to ensure success.
The initiative aims to prepare a new strategic plan by conducting the
necessary studies, establishing support departments, equipping work
teams, and preparing workshops and awareness programs to support
and facilitate the process of implementing the new strategic plan.
Goals will also be set and indicators tracked to evaluate the
performance of the corporation’s transformation into a non-profit
institution.
The initiative’s scope of work includes the following:
• Implementing the general strategy of the institution
• Initiating legal and regulatory arrangements to establish an
independent, not-for-profit entity
• Amending the framework and accounting practices in line with
international standards
• Evaluating fixed assets and determining their value
• Establishing a transformation office for supervision until completion
of the transformation into an independent, not-for-profit entity
Initiative aims:
• Developing the performance of health workers in the intensive care
unit and the departments concerned with organ donation and
transplantation
• Electronic notification
• Advanced training for healthcare providers involved in the initiative
• Optimizing the use of donors
The initiative’s scope of work includes:
• Electronic incorporation of the intensive care units and establishing
electronic registration through the website application
• Rehabilitation and training of medical coordinators, administrators,
and initiative workers
101
Improve sustainability
and financial
transparency 1-16-029-1219
Establishment of
unified electronic
system
Increase clinical
(medical)
effectiveness
1-16-097-1230
Improving skills of
ambulatory care
providers
Improve
sustainability and
financial
transparency
1-16-027-1217 Tender for private
sector submissions
2.1.2
Improve value of
healthcare services
2.1.2
Improve value of
healthcare services
2.1.2
Improve value of
healthcare services
102
Strategic pillars
Initiative link to
level 3 objective Initiative no. Initiative name
Current Initiatives
The initiative aims to develop and standardize procedures for investors
and build a unified electronic system that achieves the efficiency and
effectiveness of services that are carried out at the Food and Drug
Authority. This will contribute to improving the speed and quality of
transactions, supporting decisions and optimizing use of resources, as
well as facilitating the process of communication, information
exchange, decision-making and performance measurement.
The initiative’s scope is to unify the process flow, develop a general
perception of the basic requirements, present recommendations to
challenges, develop and analyze the requirements for each technical
sector, and develop the technical infrastructure to build the unified
electronic system.
The initiative aims to work on developing and improving the
ambulatory skills and knowledge required for service providers.
The scope of work:
• Build the capacities and skills of the ambulance staff and establish
five highly qualified training centers that enable the Red Crescent to
increase the annual training capacity
• Reduce the costs of assigning employees to the staff
• Attract trainers from within the authority with experience and
professional qualifications in emergency medical services to enable
trainees to acquire scientific and practical skills in accordance with
international standards
• Improve quality by building capacities and skills in medical ambulance
staff
• Raise community awareness in first aid to deal with emergency cases,
which reduces the ambulance requests in non-emergency cases,
improving the overall value of health services
Several initiatives have been implemented recently aimed at addressing
some of the issues of the current medical referral system to private
sector facilities. For internal referrals, a committee has been
established to purchase services from the private sector to enforce the
guidelines for referral procedures and reduce referrals to the private
sector by improving access to public sector medical facilities.
Furthermore, the Ministry of Health has prepared an approved price
list for treatment in private medical facilities, and contracted with a
local claims management company to manage the purchase of health
services from private providers.
103
Initiative description Leading party
Strategic pillars
Initiative link to
level 3 objective Initiative no. Initiative name
Current Initiatives
Improve sustainability
and financial
transparency
1-16-029-1221 Creation of Track and
Trace system
Safety in health
facilities
2.1.2
Improve value of
healthcare services
1-16-028-1255
Implementation of
initiatives supporting
the healthcare strategy
2.1.2
Improve value of
healthcare services
Increase clinical
(medical)
effectiveness
2.1.2
Improve value of
healthcare services
1-16-028-1422 National Organ Failure Registry
104
Initiative description Leading party
The establishment of the electronic tracking system for
pharmaceuticals to track their distribution and ensure their safety and
availability within the Saudi market. Pharmaceuticals are tracked using
a binary barcode.
The initiative aims to promote the implementation of the healthcare
strategy, under which all initiatives related to the development of the
health sector are included, as this strategy studies all health legislation
and regulations, sets priorities and designs initiatives after meeting
their requirements. More than 60 projects fall under the initiative to
implement the Kingdom’s healthcare strategy. Most notably: the
project overseeing the balanced distribution of health facilities and
services; and partnerships with non-profit organizations such as the
World Health Organization and the World Bank.
The scope of work includes:
• Reviewing existing health policies and regulations
• Identifying areas of duplication, conflicts, or overlaps between
policies and systems
• Proposing a list of existing health policies and regulations to be
modified or supplemented
• Proposing the list of required additional health policies and
regulations
• Proposing ideas that would improve harmony between the existing
health policies and systems
• Completing the draft national plan for inter-health referral
• Completing the national framework for health disaster management
project
The lack of a unified medical record for patients with organ failure and
living organ donors is a challenge that makes it difficult to study the size
of the problem and develop future strategies to deal with patients with
organ failure as required. Therefore, the initiative aims to establish an
electronic system for a National Organ Failure Registry, which includes the
following tasks: • Registration of patients with kidney failure, organ
transplants, and organ donors • Follow up to establish the necessary
procedures electronically
The system targets patients with organ failure, which contributes to
achieving the level 3 strategic goal and strengthening protection against
health risks, especially chronic non-communicable diseases that cause
organic failure. This is achieved by archiving data and conducting analysis
at the Saudi Center for Organ Transplantation, which involves:
• Identifying the size of the problem and the number of new patients
• Improving the survival rate of transplant patients
• Improving the care of patients with terminal organ failure of the kidney,
liver, heart and lung
• Evaluating the quality and performance of services provided to donors
and transplant patients in line with WHO requirements
• Promoting prevention against health risks, especially for chronic
non-communicable diseases that cause organ failure such as renal and
cardiovascular insufficiency due to diabetes, high blood pressure and obesity
105
Reduce the incidence
of infectious diseases
Reduce the prevalence
of risk factors for
non-communicable
diseases
Raise the level of
preparedness for
health disasters
1-16-027-0883
National Center for
Disease Control and
Prevention
Reduce the prevalence
of risk factors for
non-communicable
diseases
1-16-027-0884 Health Research Center
2.1.3
Promote prevention
against health risk
2.1.3
Promote prevention
against health risks
Strategic pillars
Initiative link to
level 3 objective Initiative no. Initiative name
Current Initiatives
106
The initiative aims to activate the functions and operation of the
National Center for Disease Prevention and Control, contribute to the
reduction of communicable and non-communicable diseases, monitor
and prevent their spread, promote health, and conduct research and
studies in the field of prevention and control of communicable and
non-communicable diseases.
Initiatives include:
• Combatting Infectious Diseases Initiative, eg. tuberculosis, AIDS, viral
hepatitis, influenza, emerging diseases such as the novel coronavirus
(Covid-19), hemorrhagic fevers
• Initiative for vector-borne diseases: eg. dengue, malaria, rabies
• Non-communicable disease control initiative: eg. diabetes, obesity,
smoking, cardiovascular disease, asthma, osteoporosis
• Environment-related disease control initiative: all diseases resulting
from the pollution of water, food and air
• Health Promotion Initiative
• Public Health Legislation Initiative which is related to health in all
policies
• International Health Regulations initiative
• Antimicrobial Resistance Initiative (fighting antibiotic-resistant germs)
• Healthy Cities Initiative
• National laboratories with full scope for public health, including
laboratories for genetic diseases, human genetics, epidemiological
disease, common diseases, radioisotope and secondary calibration of
radiometric devices
• Health Surveys Initiative
• Cancer Control Initiative: breast and colon cancer
The initiative aims to raise the level of health research in the Kingdom
based on a coordinated national strategy and a funding framework.
This is achieved through the establishment and operation of the Saudi
National Institute of Health, which will manage the scientific health
research activities on a national level across all sectors involved. This
also involves:
• Identifying and updating health research priorities periodically to
direct health research expenditures and raise efficiency
• Promoting the application of research regulations to protect the
interests of individuals and society in the field
• Encouraging participation, innovation and creativity in research
• Contributing to the exchange and dissemination of knowledge
• Benefiting from the outputs and recommendations of funded
research to develop health policies and strategies to support public
health and prevent health risks
Through the initiative, the Kingdom of Saudi Arabia will be in a better
position to address public health priorities, develop highly qualified
talent, strengthen the infrastructure for health research and
innovation, and contribute to the diversification of the Kingdom’s
economy through research and innovation.
Initiative description Leading party
107
Raise the level of
preparedness for
health disasters
1-16-027-0949 Preparedness for health
emergencies
Reduce the prevalence
of risk factors for
non-communicable
diseases
1-16-029-1231 Interactive awareness
initiatives
Reduce the prevalence
of risk factors for
non-communicable
diseases
1-16-028-1234 Initiatives of the National
Cancer Center
2.1.3
Promote prevention
against health risks
2.1.3
Promote prevention
against health risks
2.1.3
Promote prevention
against health risks
Reduce the prevalence
of risk factors for
non-communicable
diseases
1-16-029-1319 Pesticide Residue Inspection
Centre 2.1.3
Promote prevention
against health risks
Strategic pillars
Initiative link to
level 3 objective Initiative no. Initiative name
Current Initiatives
108
Despite the notable performance of the Saudi health sector in dealing
with the novel coronavirus (Covid-19) pandemic, the Kingdom seeks to
improve local capabilities to respond to disasters and crises and to
double specialized centers. Therefore this initiative aims to prepare and
equip crisis and disaster management centers to face challenges
through the provision of appropriate infrastructure including
mechanical and electrical works, and technical infrastructure such as the
Internet, wall screens, communications equipment, etc. The initiative
targets all health areas and will contribute to achieving the strategic
goal aimed at strengthening the prevention of risks as well as raising
the Kingdom’s readiness level to deal with various crises.
The initiative aims to raise awareness among community groups by
launching specialized awareness campaigns based on the needs of the
following groups: school students 7-12 years old, housewives, patients
with chronic diseases and youths 18-22 years old. The initiative will also
promote the role and regulations of the Food and Drug Authority.
This initiative aims to develop and improve the services provided to
oncology patients in the Kingdom through the development and
application of policies and regulations in line with scientific and medical
best practices in the field. This also involves strengthening the role of
the National Oncology Center to assume greater responsibilities
towards the target group (oncology patients) in the Kingdom, as well
as contributing to the development of policies and measures to
enhance the role of early detection of cancer as well as raise the level of
awareness in the community.
Establishing, equipping and operating laboratories for examining
pesticide residues in vegetables and fruits
Initiative description Leading party
109
Reduce the prevalence
of risk factors for
non-communicable
diseases
1-16-029-1220 Creation of halal center
Promote optimal road
infrastructure 1-16-031-0914
Road safety initiative to
reduce road traffic deaths
Promote the concept
of awareness,
education and
community
participation
12832
Traffic safety for school
students –instilling traffic
culture in the hearts of
youth
2.1.3
Promote prevention
against health risks
2.3.4
Enhance traffic safety
2.3.4
Enhance traffic safety
Strategic pillars
Initiative link to
level 3 objective Initiative no. Initiative name
Current Initiatives
110
Several procedures are involved in issuing halal food certificates, and
granting these numerous permits causes significant delays. Therefore,
the situation requires unifying procedures and centralizing authority.
The initiative aims to establish a center that deals with food and halal
products in all its aspects, and to grant certificates. Scope of work for
the initiative:
• Requesting the approval of the High Commissioner
• Appointing a team to establish the center
• Equipping the temporary headquarters
• Official approval of the center’s operations and procedures, technical
committees, institutional identity, accreditation and launch
This initiative aims to reduce the rate of road deaths in the Kingdom by
addressing road problems and dangerous sites (sites where deaths and
accidents are frequent), which are represented by the high number of
accidents on the road network. The initiative will also strengthen
cooperation with government agencies, promote the sharing of
information and identify hazardous sites, as well as coordinating the
rapid response time among the concerned authorities, raising the level
of awareness by implementing campaigns on road safety, and
adopting the governance of the National Road Safety Center as a
coordinating body that links the parties related to traffic safety.
This initiative aims to enhance the traffic culture of school students
through training by male and female teachers using various tools such
as: training bags and traffic safety boxes containing virtual traffic
environments; educational materials, whether images, videos or
models; and the use of driving simulators and seat belt devices.
Students are also educated about mobile awareness through school
campaigns.
Initiative description Leading party
111
hstp@
hstp.gov.sa
Contact
For more information:
All rights reserved. Health Sector Transformation Program 2021
hstp.gov.sa
Health Sector
Transformation Strategy
V.3
The Custodian of the Two Holy Mosques
Salman bin Abdulaziz Al Saud
His Royal Highness Prince
Mohammad bin Salman bin Abdulaziz Al Saud
The Crown Prince, First Deputy Prime Minister and Minister of Defense
Contents
Contents
5
1.
Introduction
2. The need for transformation
– Why do we want to change?
3. Defining the transformation goals and methods
– What do we want to change?
Using theory and applying it to our strategy.
4. Delivering the transformation
– How and when will things change?
5. Perceived risks and mitigations
– What do we want to avoid?
6. The economic case
– Do the benefits outweigh the costs?
7. Next steps
8.
Annexes
A. Plans by Themes
B. A description and summary analysis of the Harvard and
Oxford Value frameworks
C. Bibliography of key documents (with hyperlinks)
D. Endnotes
7
9
13
24
29
32
34
35
36
59
63
64
Introduction
Introduction
7
Good health is a gift. Its promotion is amongst the greatest responsibilities of any State. We
take that seriously and are now embarking on a fundamental transformation of our health
system. Our country needs it. Our people deserve it.
Our population is growing. It’s also growing older. Living longer is a blessing but it also
brings challenges – as do our more diversified economy and increasingly urban lifestyles.
Every one of us has a role to play and our attitude towards our physical and mental
wellbeing, as well as the system that supports it, must change if we are to achieve our
ambition of living fulfilling lives in a vibrant society.
Future success relies on us being smart, sophisticated and less centralized. We must draw
on what we’ve achieved in the past but not be held back by it. Healthcare staff need to be
empowered to make the right decisions at the right time in the right place. They need the
authority to take action, make choices and accept responsibility. They must also be given
the chance to make honest, well-intentioned mistakes without undue fear of failure.
The scale and complexity of this transformation cannot be overstated. That is why it has
been broken down into themes and will be executed in phases. We have sought to balance
ambition and realism.
This paper gives an overview of the transformation, spelling out the underlying goals as
clearly and coherently as possible. Delivering the strategy will be more complex and will
only work if we can engage and co-ordinate large numbers of highly educated and skilled
people effectively.
Context
The Ministry of Health (MoH) has been tasked with delivering this strategy as part of “Vision
2030” for the Kingdom of Saudi Arabia [1]. This document addresses the Level 2 Strategic
Objective of raising the standards of our health service, the Level 3 goals of easing access
to it, ensuring better value and strengthening prevention against the main threats to our
health. The strategy also aims to make a significant contribution to Level 2 Strategic
Objective 2.2: Promote a healthy lifestyle.
Though led by the Ministry of Health; this plan has been developed in close collaboration
with the National Transformation Program.
The Need for Transformation
– Why do we want to change?
The Need for Transformation
9
Understanding the challenges [2]
We have identified eight major challenges in our health system that need to be addressed
over the coming decade.
1. The population of the Kingdom continues to grow and age. Our population is expected
to rise from 33.5 million in mid 2018 to 39.5 million in mid 2030. The number of elderly
(aged 60 to 79) is expected to grow from 1.96 million in mid 2018 to 4.63 million in mid
2030.
In 2015 residents from overseas amounted to 30% of the population of the Kingdom or
10
million people. These are mainly adults but also with some accompanying young people.
We have many overseas visitors too, particularly for the major religious festivals. Two years
ago, 537,000 of those attending the Hajj were domestic pilgrims. 1,325,000 came here from
abroad and in some years the total number of foreign pilgrims visiting Mecca has been
estimated as high as three million people . [3]
The population of the KSA was 83.3% urban in 2016, which is projected to rise to 85.9% in
2030.
2. Rates of avoidable injury and non-communicable disease remain high by regional and
international standards. The Kingdom has made notable progress in improving the health of
its population over recent decades, particularly in areas of child and maternal mortality and
the reduction of communicable disease. For both males and females, life expectancy at
birth improved from 64 years in 1970 to 75 years in 2015. There are targets set to ensure it
increases to 80 years by 2030. But there is considerable scope to reduce avoidable
mortality and avoidable morbidity in both the working and elderly populations. Particular
areas of concern include heart disease, stroke, diabetes mellitus, respiratory disease,
mental health, road traffic accidents and congenital diseases, all of which are amenable to
reduction.
We need to strengthen the prevention of non-communicable disease and injury thereby
reducing avoidable illness and death. The risk of major outbreaks of communicable disease
also remains substantial, especially at Hajj or following natural or man-made disasters.
The Need for Transformation
10
When other government departments in the Kingdom develop major policy initiatives, the
health and healthcare implications of their actions are not always at the front of their minds.
Similarly, it has often proved difficult for the MoH to develop the inter-ministry dialogue
necessary to resolve some of its own pressing problems.
3. Primary care remains inadequate and inconsistent. Secondary and tertiary hospitals, and
associated resources, are poorly distributed across the Kingdom. [4] There is inadequate
capacity in extended care services such as rehabilitation, long-term care and home care.
These issues are compounded by low levels of productivity.
4. There are significant gaps in the quality of services provided to patients. Much of this is
due to lack of consistent protocols and pathways for treatment, and incomplete
measurement of patient processes and outcomes. The Saudi Central Board for
Accreditation of Healthcare Institutions (CBAHI) Essential Safety Requirements Survey of
2015 has also emphasized key deficits in safety across all categories of hospitals.
5. There is unwarranted variation in provision, access and investment when assessed using
the population served rather than the patients treated. This included over-use as well as
under-use and leads to significant shortfalls in value as well as quality.
6. The system is currently resource and staff centric rather than patient or person centric in
its orientation. It is also institution centric rather than population centric. A health system
needs to be both accessible and responsive to patients’ overall welfare.
7. There are significant gaps in workforce capacity and capability, specifically in relation to
Saudi employees. The health system also currently lacks robust, consistent and integrated
digital information systems to measure and manage resources, activity levels, quality and
efficiency and to build a learning health system, in which all involved can contribute, share,
and analyze data and in which continuous learning cycles encourage the creation of
knowledge that can be used by a variety of health-information
systems.
8. The health system also needs to support the containment of public expenditure, and the
diversification of the Saudi economy. This is needed to address the risk of long term
reductions in the price of crude oil and the impact that will have on public revenues.
The Kingdom therefore needs to:
• Encourage action within and beyond the health system to reduce injuries and the
primary and secondary prevention of non-communicable diseases;
• Introduce comprehensive measures of population needs and health system
performance to improve the allocation of resources and delivering the outcomes that
people need;
The Need for Transformation
11
• Create a culture of stewardship, in which all clinicians take responsibility for the use
of resources, the prevention of waste and the long term sustainability of universal
healthcare.
• Create incentives for staff and the system to be more effective, efficient and to
encourage locally generated innovation;
• Facilitate levels of revenue and capital investment in healthcare facilities and
systems that are proportionate to the Kingdom’s economic status and are
appropriately distributed to reflect population and patient needs;
• Bring transparency into the provision of care through systematic collection and
internal reporting of clinical data, costs and outcomes at episode level, to
understand and achieve better value;
• Ensure the quality and safety of clinical investigations and treatments, through
system and process and the rigorous use of standards, protocols and clinical
pathways and clinical practice guideline together with the continuous professional
development of clinical and non-clinical staff;
• Ensure a broad-based, highly educated, skilled and highly productive healthcare
workforce, staffed increasingly by Saudi citizens;
• Harness technology, the internet and mobile telephony, computational power and
interoperability, big data and analysis to improve patient access, education and
involvement in protecting and promoting their health, to drive quality and efficiency
gains and to build a learning health system; and
• Develop the information systems, distributed governance systems, accounting
systems, and the professional, employment and communication practices that will
enable the MoH and the health system to be more responsive to current and future
pressures.
Defining the Transformation
Goals and Methods
– What do we want to change?
Using theory and applying it to our strategy.
Defining the Transformation Goals and Methods
13
Our goals
The proposed goals of the Transformation are to:
1. Improve health: Increase the length, wellbeing and quality of life of Saudi citizens,
which includes the Vision 2030 goal of increasing the life expectancy of citizens to
80 years by 2030;
2. Improve healthcare: By improving the quality and consistency of services and the
performance and accountability of healthcare organizations and staff to deliver care
that is safe, effective, patient-centered, timely and equitable; and
3. Improve value: by containing costs, improving outcomes, controlling public
healthcare expenditure and guiding new investment.
All three transformation goals conform with, and are enablers of, the Vision 2030 strategic
objectives for health: access, value and public health.
These goals are also informed by international frameworks for the enhancement of health
systems including:
• The World Health Organization health systems framework [5];
• The World Bank Group health systems framework [6];
• The Institute for Healthcare Improvement “Triple Aim” framework [7];
• National Academy of Medicine, “Vital Directions for Health and Healthcare [8] : and
• The NHS England “Five Year Forward View” strategic plan. [9]
Introducing value based healthcare
The transformation goals are consistent with the related, but distinct, frameworks for value
based healthcare. These frameworks have been articulated by researchers at Harvard
Business School. They are also closely associated with work by The World Economic
Forum [10] which focuses on the implications for associated health industries and the
somewhat distinct work at Oxford University, which has now been adopted in England,
Scotland Wales and Italy. [11]
Defining the Transformation Goals and Methods
14
This transformation strategy is deliberately formulated to utilize and combine aspects of all
three value frameworks.
The financial, economic and institutional imperatives, together with the policy intentions of
the KSA Government, that were expressed through Vision 2030, require a high level of
overall control over health services expenditure and use of other resources. These factors
also require a close interest in the health of the whole population. These considerations are
distinctive features of the Oxford framework.
A description and summary analysis of the Harvard and Oxford Value frameworks can be
found at Annex B.
From theory to strategy
The Vision Realization Office (VRO) has organized its work into seven themes:
• The New Models of Care;
• Provider reforms;
• Financing reforms;
• Governance development;
• Private and third sector participation;
• Workforce development; and
• eHealth development.
The first three themes can be understood as enablers of three levels of value:
• The “New Models of Care” theme as a focal point for enhancing personal value
through the improvement of treatment and care modalities at an individual level.
• The provider theme as a focal point for enhancing utilization value at an
intermediate level, whether that is at the clinical micro-system, hospital or local
health system level. (The important principle in a country committed to universal
healthcare is that the definition of the population captures everyone in a particular
jurisdiction such as a region.)
Defining the Transformation Goals and Methods
15
• The financing theme as a focal point for enhancing allocative value through
ensuring those intermediate levels get “optimum” levels of resource, based on
patients’ needs and their ability to benefit.
It can be argued that financing plays a direct role in securing all three types of value. In
addition, patients’ needs are often qualified by other criteria including patients’ merits,
economic objectives to protect the health of the working population, or patients’ ability and
willingness to pay.
Past experience of health transformation strategies [12] suggests that organizational and
financial changes are unlikely to lead to any major improvements in outcomes unless they
are accompanied with supply side improvements, specifically, improvements in the
efficiency, effectiveness, equity and responsiveness of the public health and health services
provided. It is essential to recognize that the three value dimensions are interdependent
and mutually reinforcing. Simply put, successful delivery of all the seven work themes will
be essential to the overall success of our transformation strategy.
Outlining the transformation strategy
Through 2017 the MoH made progress in developing the principal elements of its strategy.
The following section outlines key policies and links these to the challenges they seek to
address. There is a particular emphasis on the first three themes. The subsequent parts of
the paper examine in more detail how the key policies will be realized in practice.
The New Models of Care [13]
There are many challenges within the existing models of care for key patient pathways
within the Kingdom. There are growing hazards within healthcare facilities due to
inadequate medical quality and low safety standards. Waiting times are prolonged and they
vary considerably across healthcare facilities, causing inevitable dissatisfaction. There are a
number of high priority pathways with specific challenges including the maternity pathway,
chronic conditions, planned care, urgent care, and last phase of life.
There are also challenges common to all pathways, including:
! Shortage of medications and available medicines are dispensed inconsistently;
! Lack of standardized clinical guidelines and variations in the quality and delivery of
care;
! Poor pathway management, with inappropriate referrals, and inappropriate
presentation by ill-informed patients disrupting patient flow;
Defining the Transformation Goals and Methods
16
! Lack of out-of-hospital services for diagnostic, preventative, proactive or follow-up
care;
! Poorly coordinated care, particularly between MoH providers and non-governmental
organizations; and
! Poor communication between providers, and between clinicians and patients.
To address these challenges, the MoH has developed a program to design, pilot, and
implement a patient centric New Models of Care Program.
The program has been designed to answer six key questions:
1. How will the system help to keep me well?
2. How will the system support me when I have an urgent problem?
3. How will the system support me to have a great outcome for my planned procedure?
4. How will the system support me to safely deliver a healthy baby?
5. How will the system support me with my chronic conditions?
6. How will the system support me with compassionate care during the last phase of
my life?
The New Models of Care is designed to support people with their health and wellness
needs: physical wellbeing, mental wellbeing and social wellbeing. This aligns with the
principles set out in the Constitution of the World Health Organization: “health is a state of
complete physical, mental and social wellbeing and not merely the absence of disease or
infirmity”.
Virtual care will be a powerful source of health advice. Virtual care in most instances will
serve as people’s first point of contact with medical care providers, improving people’s
access to medical advice and guiding them to navigate the healthcare system and seek
appropriate care.
The New Models of Care program has been designed based on the following principles:
! Empowering people and their families to take control of their health;
! Providing knowledge to people as part of their treatment, and enabling them to be
well-informed and in control of their health;
! Fully integrating the health system from the people’s perspective;
Defining the Transformation Goals and Methods
17
! Keeping people healthy and focusing on the whole population through a preventive
approach, rather than a solely curative approach to health provision; and
! Providing treatment in a patient-friendly and outcome-focused way, without over-
treating or under-treating patients.
Accordingly, it has been developed with patient needs and asks at the center of the whole
system.
The New Models of Care Program will deliver 42 coordinated interventions, across six
‘systems’ of care by the end of 2020.
The prioritized systems of care are: Keep Well, Safe Birth, Planned Care, Urgent Care,
Chronic Conditions and Last Phase.
The 42 initiatives will include defined patient pathways and key performance indicators
(KPIs) including measurement of: safety and quality process metrics, clinical and patient
reported outcomes, and financial performance. These New Models of Care KPIs will be
incorporated into the KPIs for the Essential Benefits Package (EPB) (see Financing below)
to ensure that they are monitored and maintained when achieved. (A full list of the
42
interventions, progress to date and next steps is contained in Annex A).
Defining the Transformation Goals and Methods
18
Provider reform [14]
Key policies include:
• Developing all existing MoH providers, through an initial “cluster” phase, into
approximately 20 geographically defined, vertically integrated “Accountable Care
Organizations (ACOs);”
• Making increased use of private healthcare provision; and
• Developing third sector provision to cover gaps arising from either government or
market failure.
The ACOs will be established as “corporatized” public bodies, with substantial and clearly
defined “decision rights”. Responsibility and accountability for the successful management
and clinical governance of the ACOs will be vested in Boards which will be established in
each ACO. The Chairman of each ACO Board will be appointed by the Minister of Health.
The NHS in England has had a policy of corporatizing public healthcare providers since at
least 2004. This builds on a key textbook on reforming public hospitals published by The
World Bank in 2003. [15] This Strategy is drawing widely on the English experience
including its guidance on board led governance. [16]
ACO Boards will become the focal point for the successful realization of utilization value in
the system. Each ACO will also have to take responsibility for the allocation of resources for
their population.
The precise decision rights of the Saudi ACOs are still being determined. The more
extensive the rights, the more scope that ACOs Board will have to improve use and
personal value within their geographies and institutions.
Confirmed decision rights of ACO Boards include:
• Salaries and salary structures (clinical and non-clinical); and
• Employment rights over staff, including monitoring and evaluation of clinical and
managerial performance.
Defining the Transformation Goals and Methods
19
Other issues that should be considered include:
• Rights to acquire, dispose of, and invest in physical assets;
• Rights to generate income (adjacent to core activities);
• Rights to set up educational courses and establishments (particularly nursing
schools); and
• Financial rights and responsibilities (e.g. handling of surpluses, ability to take on
debt).
Well performing ACO Boards, with appropriate decision rights, will be central to addressing
the identified challenges within their geographies:
• Insufficient and poorly distributed capacity and capabilities in the provision of public
health, primary care, secondary, tertiary care and extended care services;
• Key deficits in the appropriateness, quality, safety, effectiveness and efficiency of
services provided, including workforce productivity;
• Gaps in workforce capacity and capability, and specifically in relation to Saudi
employees; and
• Unwarranted variation either in comparison with other population based ACOs or
within the ACO.
ACO Boards will also play a vital role in the robust local adoption of digital information
systems. These will measure and manage resources, activity levels, quality and efficiency
and build a learning health system.
Finally, ACO Boards will have an important responsibility for ensuring that ACO staff and
institutions become patient or person centric in their orientation. They will do that both by
personal example and through their behavior and decisions as a Board.
Defining the Transformation Goals and Methods
20
Financing reform [17,18]
Key policies, to be achieved in incremental steps, include:
· Confirmed commitment to universal healthcare coverage which will ensure that all
citizens, residents and visitors to the Kingdom can obtain timely access to healthcare
services, via insurance, without the risk of impoverishment;
· Establishing the program for health assurance and purchasing to be a national payor
to ensure free care to beneficiaries through the newly MOH corporatized providers and
other governmental providers.
· Definition of an Essential Benefits Package (EBP) which clearly defines a core
package of treatment and care (with defined quality standards) that all insured patients will
ultimately become entitled to;
· Creating a system of supplementary health insurance (SHI) which will allow most
citizens and residents to add additional benefits to their EBP. This SHI could be used to
enhance the scope, timeliness or comfort of their treatment and care;
· Creating a wider role for private health insurers as the transformation matures,
through the creation of a market of licensed and regulated insurers who will offer SHI
products.
· Payment mechanisms that will support the implementation of the New Models of
Care and ensure value at all levels.
· Analyzing spend by major program budget at a national and regional level to
optimize the allocation process;
The definition of an EBP will improve the allocation of resources and ultimately ensure that
the whole population receive a guaranteed level of timely and accessible treatment and
care. This will include services that that can achieve primary and secondary prevention of
disease.
Defining the Transformation Goals and Methods
21
Governance development
The policy goals of the governance theme are:
1. To define, design and support the establishment of the regulatory and improvement
functions and institutions necessary to secure and sustain value based healthcare.
2. To embed strong, delegated and devolved leadership and governance throughout the
health system. The Kingdom is too large, and the challenges are too many, complex,
dynamic and inter-connected to be addressed successfully solely by a centralized
bureaucracy, however competent.
A detailed description of the development approach of this work theme can be found in
Annex A.
Capacity and capability building
The other three themes of the strategy are all related to aspects of capacity and capability
building:
The policy goals of private sector participation on the supply/service provision side are:
• To reduce pressures on public funds;
• To provide performance benchmarks for corporatized providers; and
• To potentially facilitate partnership opportunities for corporatized providers to divest
and re-invest to enhance service quality or efficiency.
The policy goal of associated third sector participation is:
• To fill any residual gaps in treatment, care or support services, highly valued by
patients and their families, that would otherwise persist in the presence of
government and market failures. (e.g. hospice services).
The policy goal of the eHealth program is:
• To support the information requirements of the other work themes and, in particular,
to enable a value based healthcare approach and build a learning health system.
Defining the Transformation Goals and Methods
22
A value based healthcare information architecture includes the following minimum
components [19] :
• Standardized outcome metrics and accurate measurement of resource costs by
population segment (segmenting the population by need as well as by geography);
• A universal data taxonomy across diseases and population groups (e.g. a consistent
methodology and scale for measuring pain or patient quality of life);
• Inter-operability that allows databases to efficiently communicate with each other;
• Integration of outcomes data into the systems clinicians use in their daily work
(e.g. electronic medical records [EMRs]) with a user-friendly interface that minimizes the
effort required for entering data;
• Mechanisms to link individual patient data across multiple databases (e.g. a unique
personal identifier); and
• Robust governance processes, with comprehensive
rules for data access, agreements about data sharing and guidelines for managing
privacy.
The policy goal of the workforce development program is:
• to support the transformation of the workforce necessary to enable a value based
healthcare approach. This will require major increases in workforce capability and
some increase, re-distribution and diversification of workforce capacity.
Delivering the Transformation
– How and when will things change?
24
Timeline and scope of the transformation
The reforms will be implemented in three broad phases, of increasing scope and
complexity:
Phase 1 – Building capabilities
Phase 1 is underway. It commenced at the beginning of 2018 and will finish at the end of
2020. This phase is primarily concerned with clinical systems improvement, capacity and
capability building and preliminary institution building.
From a financing perspective, Phase 1 envisages the establishment of a government
mandated, Purchasing Program (PP), managed by new department, initially based within
the Ministry of Health. The PP will initially cover persons already covered by MoH services,
who will be offered entitlements defined in the Essential Benefits Package (EBP).
From a delivery perspective, Phase 1 envisages the establishment of a corporatized public
Holding Company with a number of associated regional Corporates. The Holding Company
will be responsible for the creation of approximately 20, vertically integrated, geographically
defined, clusters of providers.
The present plan is that the clusters will be established in “waves.” Wave 1, comprising five
clusters will be initiated during 2018. The first cluster will be in the Eastern Region, the
second and third in Riyadh, the fourth in Makkah. The location of the fifth cluster will be
Qasim. Subsequent waves of clusters will be initiated by the Corporates, at a pace that
they will determine.
The Phase 1 reforms, from both financing and delivery perspectives, will be heavily
informed by ongoing work coordinated by, the MoH-Vision Realization Office (VRO) on New
Models of Care. This includes the 42 initiatives proposed by the MoH during the latest
round of revisions (July to October 2017) to the National Transformation Plan.
Phase 2 – Developing autonomy and value
Phase 2 will start at the beginning of 2021 and finish at the end of 2025. This phase is
primarily concerned with the secure and robust corporatization of key public entities. These
include the ACOs (developed from the clusters), PHAP in Phase 2, and a range of
regulatory and improvement agencies.
Phase 2 will also include the re-purposing of the MoH; the implementation of Health in All
policies; and the implementation of system wide enabling reforms regarding workforce,
digital and systems governance.
Delivering the Transformation
Delivering the Transformation
25
From a financing perspective, Phase 2 envisages the possible extension of the PHAP to
persons presently funded by all other government departments including Other
Government Services (OGS).
During Phase 2 most persons covered by the PP will be given the opportunity to purchase
supplementary health insurance (SHI).
From a delivery perspective, clusters will be developed into approximately 20 corporatized
ACOs across the Kingdom. The ACOs will primarily be resourced from existing MoH
services and facilities. However, the ACOs will also be able to sub-contract services for
patients covered by the PHAP (and SHI) from the private sector. It is also envisaged that
healthcare facilities and staff presently directly managed by other government departments
may be fully or partly integrated into emerging ACOs during Phase 2.
Phase 3 – Strengthening value and choice
Phase 3 will start at the beginning of 2026 and finish at the end of 2030.
From a financing perspective, Phase 3 envisages the extension of the NHI to all citizens,
residents and visitors to the Kingdom. It envisages the simultaneous push of licensed
private health insurance providers to embrace and implemented the principles of value
based healthcare.
Delivering the Transformation
26
Change management
The change proposed in this strategy is transformational. It involves altering the overall
orientation of the MoH and wider health system. It entails changing the purposes of the
system, types of work, roles and ways of working, learning, culture and organizational
processes. It is also: multi-dimensional, multi-level and multi-phase. The transformation as
envisaged is, appropriately, both planned and emergent.
The transformation also takes place in the context of an “open system”. This is a system
that cannot be fully controlled and must also interact with, and react to, a range of
uncertain and sometimes unpredictable and unknown factors beyond its control. As such
no single change management theory or framework will be adequate for all aspects of the
transformation.
Nonetheless, the core governance of the transformation requires a consistent approach to
change management. This approach must be appropriate to deliver a transformation that is
complex, emergent, and open. [20]
It is therefore proposed that two change management approaches will be incorporated at
every level.
The first approach is Edward Deming’s model for quality improvement, with its embedded
continuous improvement cycle. In this approach, process mapping and analysis of progress
are supplemented by qualitative and quantitative analysis.
The second, complementary approach, is that of action research and learning, with its
emphases on learning through action, process development “facilitation” (co-production)
and collaborative learning. This approach is especially appropriate in the highly
professionalized and specialized arena of healthcare, but should also extend to involving
citizens and patients.
There are, nonetheless, broadly two dimensions to the change management outlined
above: internal change management and external change management. Systems leaders
have more, but not complete, control over the former, less over the latter. These levels of
control will have an effect over the style of change management process adopted by
different work streams. This point is illustrated below by outlining the proposed change
management approaches that have been adopted by the corporatization and eHealth work-
streams.
It will be important that the VRO and MoH leadership keeps approaches to change
management adopted by themselves, and their associated Boards, under regular review.
This will ensure that approaches are adjusted, and remain appropriate, effective and
aligned, as the transformation progresses.
Delivering the Transformation
27
Corporatization Delivery Office – Approach to change management
eHealth – Approach to change management
A fuller description of the plans for each theme can be found at Annex A.
Perceived Risks and Mitigations
– What do we want to avoid?
Perceived Risks and Mitigations
29
Identifying Risks
The key tests for any strategy are that:
1. Identified challenges are adequately addressed;
2. Defined goals can be achieved;
3. Enabling policies and related initiatives are SMART (specific, measurable, achievable,
realistic and timely); and
4. That the strategy is coherent; and the individual elements well aligned with each
other.
The risks associated with this transformation strategy can be divided into three broad
categories:
• Design risks – Where the design of the strategy is flawed in certain respects:
• Implementation risks – Where the design is good but implementation is weak;
• Operational risks – Where design is good, implementation is sound but operational
performance of the new system is impeded by anticipated or unanticipated
operational risks.
Design risks are being mitigated within each work theme using experienced leading
management consultants. “Gateway” review processes are already in place.
Increased emphasis is also being placed on the overall coherence and alignment of the
strategy. The leadership and partners of the management consultants are working together
closely on alignment planning. There are frequent and regular internal “Design Authority”
meetings, which also periodically review the continuing overall coherence of the
transformation as detailed plans evolve.
Another, very important design risk mitigation strategy is the systematic use of national and
international advisory boards composed of leading experts to advice and inform the
progress of the strategy. In addition, international organizations are engaged in the review
of the overall strategy and the strategies of individual workstreams. Key organizations
involved include the World Bank Group, Harvard University (School of Public Health) and
the London School of Economics.
Implementation risks are being identified and addressed. Risk registers are being
assembled at work theme and VRO levels.
Perceived Risks and Mitigations
30
Key perceived risks (as of Q2 2018)
Immediate risks
• Present and future gaps in workforce capacity and capability. Mitigation –
Request for proposal tendered to strengthen workforce planning capabilities.
• Evaluate if all value based healthcare eHealth prerequisites have been
included in eHealth plan; and fill any gaps. Mitigation – Evaluation
commenced
• Low and late investment rates (challenges in securing timely and
proportionate investment in the strategy, for example some aspects of New
Models of Care remain thin, particularly mental health services). Mitigation –
Dialogue with the Ministry of Finance and the Ministry of Economy and
Planning though NTP channels.
Future risks
• Uncertainty regarding timing and nature of integration of other government
services into the transformation strategy. Mitigation – Dialogue being
progressed through the Saudi Health Council (SHC).
• If the scope decision rights given to “corporatized” entities (e.g. ACOs and
regulators) are too narrow and low the corporatization will be costly but
ineffective. Mitigation – Risk recognized and appropriate decision rights still
being evaluated.
The Economic Case
– Do the benefits outweigh the costs?
The Economic Case
32
The Economic Case [21]
The economic case for the transformation of the health system was first modeled in March
2017, and revised in May 2017. Both versions were based on needs and investments
agreed in National Transformation Programme 1.
A third, more detailed modeling was conducted in May 2018, based on investments
proposed in National Transformation Programme 2 (July to October 2017).
The May 2017 modeling generated a benefit to cost ratio of 4.6 for the MoH and 10.9 for the
MoH and the wider KSA economy, on the calculation that investment costs in the
transformation would be SAR 65 Bn (SAR 55 Bn NPV discounted)).
In the same May 2017 modeling, the benefit of increasing the scope of reform to system
wide transformation (from Phase 2), were estimated to increase the benefit to cost ratio from
4.6 to 5.2 for the MoH (and from 10.9 to 11.3 for the MoH and the wider KSA economy). The
investment costs of system wide transformation however were estimated to rise to SAR 129
Bn (SAR 111 Bn NPV discounted).
The initial, provisional, output of the May 2018 modeling, suggests a more modest, but still
positive benefit to cost ratio of 1.8 for the MoH.
It is important to note that the total investment level in the May 2018 modeling is 6 times
higher than in Business Case Version 1 at SAR 356 Bn cumulative to 2030 (compared with
SAR 65 Bn).
This in turn reflects the wider range of service gaps that are being addressed by investments
in the May 2018 modeling, and scale of quality improvement envisaged to be delivered.
It is also important to understand that the May 2018 modeling would still envisage reducing
the projected annual operating budget of the MoH in 2030 from SAR 178 Bn (Do Nothing
scenario) to SAR 126 Bn.
Next Steps
Next Steps
34
Next Steps
Key next steps during 2018 include:
Vision Realization Office leadership
• Proactive communication of the transformation strategy to professional and public
audiences.
Reforms to the New Models of Care
• Commencement of the majority of the 42 New Models of Care initiatives by end
2018.
• Procurement pathway for Patient Reported Outcome Measures (PROMs) defined
and agreed by end 2018.
Provider reform
• The establishment of the provider Holding Company and Corporates by end 2018.
• Commencement of cluster rollout in the Eastern and Central Regions by end 2018.
Financing reform
• The establishment of the “purchasing entity”, initially within MoH by mid 2018.
• Enrollment of “members” in order to commence Purchasing Program on 1 January
2019.
Governance development
• Governance work-streams (See Annex A) completed.
• Pilot Leadership Program completed and substantive Executive Leadership Program
commenced by end 2018.
Private sector participation
• New “White Paper” on emergent private sector participation priorities completed and
agreed by mid 2018.
Workforce development
• Initial quantitative evaluation of workforce capacity and capability requirements to
2030 completed and requirements agreed by end 2018.
eHealth development
• Specified 2018 projects (defined on pages 54 to 55) commenced by end 2018.
Annexes
A. Plans By Themes
B. A description and summary analysis of the
Harvard and Oxford Value frameworks
C. Bibliography of key documents (with hyperlinks)
D. Endnotes
36
Annex A. Plans by Themes
New Models of Care
The New Models of Care Program will deliver 42 coordinated interventions, across the six
systems of care. These 42 interventions are listed below:
Keep Well
The system will support people to stay well, and get well again, providing them with
knowledge, empowering them and supporting them with healthy communities. It will
achieve this through:
1. Health coach programs;
2. Community-based wellness programs;
3. Workplace wellness programs;
4. School wellness programs;
5. Healthy food promotion;
6. Health edutainment programs; and
7. Promoting the Saudi Centre for Disease Control.
Planned Care
The system will support patients to receive a great and consistent outcome for planned
procedures, by providing necessary, efficient, and high quality care. It will achieve this
through:
8. One-stop clinics;
9. Pathway optimization;
10. Length of stay reduction initiatives; and
11. Step-down and post-discharge services.
Safe Birth
The system will support women to have a safe delivery and healthy infants. It will provide
continuous support from pre-marriage and pre-conception to post delivery and during early
years development. It will achieve this through:
12. Premarital screening;
13. Preconception care services;
14. Maternity care services;
15. National birth registry;
16. Postnatal care services;
17. Well baby clinics; and
18. Neonatal care services.
Annexes
37
Urgent Care
The system will support people when they have an urgent problem by providing the
necessary treatment in the right place, at the right time, and by supporting patients return
home and to community services. It will achieve this through:
19. A resource control center;
20. Urgent care clinics; and
21. Population-based critical care centers.
Chronic Conditions
The system will support people to manage their chronic conditions by providing integrated
care, facilitate the patient flow between care setting, and by promoting care at the
appropriate setting. It will achieve this through:
22. Chronic disease screening;
23. Case co-ordination; and
24. Continuing care services.
Last Phase
The system will support patients and their families during the last phase of their lives. It will
provide them with compassionate care and empower them to spend their last days at a
place of their choice. It will achieve this through:
25. Patient and family support;
26. Hospice care services; and
27. Multidisciplinary team development.
Cross Cutting interventions
28. Health in all policies;
29. Virtual self-care tools;
30. Virtual education and navigation tools;
31. Health hotline services;
32. Healthy living campaigns;
33. School education programs;
34. Enhanced primary care services;
35. Enhanced home care services;
36. Resource optimization;
37. Integrated personal health records;
38. National referral networks;
39. National guidelines;
40. Outcomes monitoring;
41. Systematic data collection; and
42. Health research programs
Annexes
38
While development of the New Models of Care program is being coordinated nationally, it is
intended that the interventions will be adaptive to different contexts (e.g. city, town, rural).
How these interventions will be adjusted to respond to greater demand during the religious
festivals will need to be considered.
The national team will also need to consider how mental health needs should be taken into
account, as these services are comparatively less developed.
Progress to date
Design of Models of Care (Phase 1 took place between October 2016 and April 2017. The
Vision Realization Office (VRO) New Model of Care team led a national effort including more
than 450 health professionals and patients, with an additional 2,000 involved in virtual
discussions. These stakeholders worked together to shape the New Models of Care.
Regional Pathway Development (Phase 2) ran from April to August 2017. It focused on
working with five pathfinders to develop the national system of care designs into the
following implementable regional pathways:
! King Saud Medical City (KSMC), Riyadh: Women and Child (and Planned Procedure);
! King Fahad Medical City (KFMC), Riyadh: Last Phase (and Planned Procedure);
! King Khalid Eye Specialist Hospital (KKESH), Riyadh: Planned Procedure (in
conjunction with KSMC and KFMC in Riyadh);
! King Abdullah Medical City (KAMC), Makkah: Urgent Problem; and
! King Fahad Specialist Hospital – Dammam (KFSH-D), Dammam: Chronic Conditions
(starting with hypertension and sickle cell disease).
A national implementation team is working with task forces of experts to implement the
Cross-Cutting and Keep Well systems.
Next steps
Implementation (Phase 3) is expected to occur in waves, in regional hubs in tandem with
the development of the Corporates.
Annexes
39
Provider reforms
Initiating provider reform
Provider reform will be initiated by the creation of a parent Holding Company responsible
for managing existing MoH providers.
The next step will require the Holding Company to execute plans to establish five regional
offices, known as Corporates. These will in turn manage the establishment of a total of ~20
geographic clusters as quickly and effectively as possible. The purpose of the Corporates
will be to support the development of the clusters into autonomous, public corporations to
by the end of 2030.
The VRO corporatization work-stream plans to establish the Holding Company by the end
of Quarter 2 2018 and the Five Associated Corporates at the same time. The clusters will
be developed in waves.
Wave 1, comprising five clusters, will be initiated during 2018. The first cluster will be in the
Eastern Region, the second and third in Riyadh, the fourth in Makkah. The location of the
fifth is Qasim. Subsequent waves of clusters will be initiated by the five Corporates, at pace
that they will determine.
The key initial requirements leading to the development of the clusters include:
! Integration of existing providers;
! The development of cluster level leadership, strategic management and governance
skills;
! The development of public engagement capabilities; and
! The development of information, financial and management systems.
An initial risk and readiness assessment for wave 1 clusters has been undertaken that
considered disruption to patient services, staff redeployment and finances. The VRO
provider reform work theme is now setting up corporate delivery offices which will work on
building capacity at a local level. This will ensure each cluster will have the necessary
competencies and level of autonomy expected of a well-functioning health system.
Corporatization and the establishment of Accountable Care Organizations
(ACOs)
It is planned that clusters will establish Advisory Boards at the earliest opportunity. These
Advisory Boards will be confirmed (once legal structures allow) as Cluster Boards, who will
guide and support clusters towards the competencies and capabilities required to become
corporatized Accountable Care Organizations. Clusters will take on the responsibilities of
ACO status when they have the capabilities to do so.
Annexes
40
The role of medical cities
There are currently several medical cities and specialized hospitals in the KSA. Many of
these are being used as pathfinders for New Model of Care initiatives (see below).
It is proposed that King Fahad Medical City and King Saud Medical City are incorporated
within Clusters 1 and 2 of the Central (Region) Corporate. King Abdallah Medical City is
incorporated within Cluster 1 of the Western (Region) Corporate. In the Eastern Region, the
King Fahad Specialist Hospital in Dammam will be incorporated within Cluster 1 of the
Eastern (Region) Corporate.
Plans regarding the other medical cities and specialized hospitals (e.g. King Khalid Eye
Specialist Hospital) are being developed by the corporate delivery offices in collaboration
with the respective medical cities.
Ambulance services
The New Model of Care trauma pilot program is providing an opportunity to consider
priorities for the wider development of ambulance services. Pre-hospital treatment and care
is a vital component of trauma services. The VRO intends to work closely with the Red
Crescent Authority and other stakeholders on this and related matters in 2018 and 2019.
Annexes
41
Financing reform
Purchasing Program coverage
During Phase 1 the PHAP will cover all persons presently eligible for Ministry of Health
(MoH) services. This includes:
Saudi citizens covered by MOH services unless covered by program provided to them by
entities other than MOH;
• Saudi citizens covered by private health insurance under the Council for Cooperative
Health Insurance (CCHI) Scheme (but only for services in the Essential Benefits
Package (EBP) that are not covered through their private health insurance package);
• Expatriates and their dependents working directly for the KSA Government; and
• Any group that are included in the coverage by a legal resolution.
• The EBP will specify:
• Type of services – the scope of clinical services provided;
• Utilization standards – processes to control utilization levels (e.g. referral pathways);
• Amenity levels – the level of amenity of services provided (e.g. when shared or single
hospital rooms will be provided); and
• Provider network – the service providers that will deliver clinical services.
The EBP will include all services presently included in the CCHI benefit package, together
with additional services based on current MoH practices and international benchmarking.
Annexes
42
Establishing the Purchasing Program
The VRO financing team has developed clear, and increasingly detailed, strategy and
organization development plans for the PHAP from mid 2017 up to the end of 2020. It also
has a vision and outline plans covering the period from the beginning of 2021 to the end of
Phase 2 at the end of 2025.
Provider payment during Phase 1
It is proposed that from January 2019 the PP will advise the Ministry of Finance how to
allocate funds according to the following principles.
Capitation will commence with clusters at the beginning of 2019, as quickly as data quality
allows. Other providers will be paid by improved block budgets with target based incentive
payments and penalties. Capitated clusters will receive a mix of:
• Full Capitation – Reflecting the gross cost of the EBP (including administration but
excluding capital costs) for persons entitled to MoH coverage (adjusted to reflect
those covered by OGS or CCHI); and
• Fractional Capitation – Reflecting the use of MoH facilities by OGS beneficiaries to
obtain EBP services.
Capitated payments (prospective payments per person covered) will be the main payment
mechanism. Capitated payments will be risk adjusted. The first basis of risk adjustment will
be age and gender. During Phase 1 other bases for adjustment, including quality of services
provided and deprivation, will be evaluated. Seasonal variations in expenditure may also
need to be considered. It should be noted that it is not planned for the capitated payments
to include provisions for capital assets and investments during Phase 1.
Key Performance Indicator (KPI) payments – Retrospective payments for performance
against KPIs.
Risk sharing arrangements – To be determined, to reflect external risks that impact on
activity and outcome levels (e.g. communicable disease epidemics or major disasters).
Annexes
43
Additional budget funding will be in the form of Structural Adjustment Support to reflect
current MoH system inefficiencies. This support will be tapered to zero by the end of 2030.
A large proportion of this steadily reducing support will be passed on to MoH providers in
the form of Structural Adjustment Grants (SAGs).
MoH providers may also receive budget funding for funded mandates, for services the MoH
wishes to be provided but which are outside the EBP. This funding may be provided directly
to MoH providers or channeled through the PHAP.
Other payments include mechanisms for paying specialist public hospitals not included in
clusters. The PHAP will also make payments to domestic and international private
providers for emergency overflows and selected elective services using funds allocated to it
by the Ministry of Finance for those purposes.
Provider payment during Phases 2 and 3
From 2021 the PHAP will establish regional offices in proximity to provider regional
corporates as needed. At some point during Phase 2 the PHAP will become an “arms-
length” corporatized public body with its own Board, with clearly demarked responsibilities,
authorities and accountabilities. At the same time, the regulatory role of the PHAP will be
separated from other responsibilities of the corporatized public body. It will be taken-up by
a separate regulatory body.
Phase 3 envisages the extension of the National Health Insurance Scheme (NHI) to all
citizens, residents and visitors to the Kingdom. The opportunity to purchase SHI will also be
made available to most citizens, residents and visitors. These insurance policies will be
provided via the introduction of an expanded, regulated, Saudi based, health insurance
market. Providers of these policies, some existing private health insurance companies in
the Kingdom, and new entrants including for profit and not for profit entities, are expected
to include the PHAP.
The progress of the proposed financing reforms to Phases 2 and 3 is heavily contingent of
the timely and successful development of the provider reforms.
Annexes
44
The development of health systems governance
Definition of health governance
Based on a review of international literature, health governance has been defined as:
“System of policies, regulations and structures to instill appropriate behaviors, monitor
performance and optimize realization of health value to the population.” [22]
Governance work-stream approach
Annexes
Healthcare governance framework
The assessment of the health system governance is based on a health system framework.
This framework incorporates both horizontal and vertical pillars. The horizontal pillars
represent the four core governance functions that apply across each of the vertical pillars.
The vertical pillars themselves represent the key pillars that are found in a robust health
system and are composed of multiple roles.
45
Annexes
The horizontal pillars in the framework are:
• Direction setting – Defines the key strategic objectives and priorities to steer the
healthcare system.
• Policy, regulations and standards – Sets, maintains and develops regulatory
framework and tools.
• Monitoring – Monitors compliance and system efficiency and effectiveness.
• Enablement – Provides the incentives, and support in building capabilities for
continuous improvement.
Current state assessment
A three step process was undertaken to assess the implications on the future health
governance in Saudi Arabia:
1. Saudi health system performance – This assessed the performance of the current
health system.
2. Saudi health system governance structure assessment – This step identified the
system-level governance pain points. Learning from that step is summarized in the
slide below.
3. Saudi health system governance roles assessment – This step mapped current
regulators against core functions and identified improvement opportunities. A
summary is shown below.
46
This assessment helped define the pillars of the future governance framework. For each
vertical pillar the following requirements were identified:
Public health
• The need to have a national owner for public health.
• Balance the centralization and localization of public health programs to optimize
delivery.
Provision
• Separate regulatory and provision roles in the MoH.
• Consider having a national regulator for all healthcare provision and harmonize
responsibilities.
• Institutionalize quality improvement mechanisms (i.e. data collection, analysis, and
actions).
• Ensure objective and equal treatment of all health care providers.
System financing and payment
• The need to have a one sponsor or gate keeper for the national healthcare budget.
• Clarify who is regulating the PP.
• Institutionalize financial and clinical outcome analysis in preparation for a value
based system.
• Expand scope and empower the insurance regulator.
Annexes
47
Food, drugs and medical devices
• The need to strengthen the monitoring role of the Saudi Food and Drug Authority to
ensure compliance of its regulations.
• Introduce a medication cost effectiveness role to enable providers to manage
healthcare budgets.
Human capital
• Formalize workforce planning across health sector.
• Streamline the process of medical professional development from education,
training, to joining the workforce.
• Separate setting training standards and licensing health workforce.
Research and development
• The need to centralize national research agenda setting and facilitation.
Data and digitization
• Empower the National Health Information Centre to lead healthcare digitization
programs across the Kingdom.
• Formalize health data interoperability standards.
• Clarify business and clinical requirements coming out of the transformation on digital
systems.
This assessment revealed commonalties and enabled the consolidation of the original pain
points into four systemic points as outlined below:
Annexes
The new regulatory architecture (Work-steam 2.1) is presently being designed. This work
should be completed by mid 2018.
48
Private sector participation
What
In January 2018, the workstream a detailed strategic context for private sector participation
in the Kingdom. [23] It provided detailed information about nine principal areas:
1. Primary Care
2. Radiology
3. Laboratory services
4. The commissioning of hospital services and medical cities (completing, equipping
and making operational partly built facilities; equipping and making operational built
facilities; and making operational built and equipped facilities)
5. Extended care
6. Rehabilitation
7. Long Term Care
8. Home Care
9. Pharmacy
The strategic context includes a clear vision, a mission and value statements, five goals
and 14 strategic objectives with associated five-year quantified targets. It also includes a
theme alignment analysis, identifies key enablers, critical success factors and risks. It
proposes a clear operating model with six business units, six pilot studies and a clear
timeline for delivery over five years.
The business model
Annexes
49
Third sector participation
The MoH launched its Community Participation Program at the beginning of 2017. This was
aligned to the recent trends that have been the relationship between the government and
the community, and was supportive of the aspirations of Vision 2030 and NTP.
Community participation is based on the belief that health development can best be
achieved through combining the efforts of both the government and all concerned parties.
As the community’s demand for health continues to grow and change this cooperation is
essential.
The community has enormous resources and huge potential that can be used for its benefit
and for solving health problems and challenges. However, the community cannot make an
optimal contribution in the absence of a motivating environment and suitable opportunities
to use its resources and potential.
Through the Community Participation Program, the MoH aims to become a partner that
welcomes new and influential health partnerships with various components of the
community, including individuals, civil societies and organizations, companies and
government agencies.
The vision, mission, values, objectives, tracks and initiatives of the Community
Participation Program are outlined below:
Vision
A community aware of its health challenges and actively participating in addressing
them.
Mission
The Community Participation Program is a motivational and targeted platform that
contributes to health development through effective and sustainable partnerships. It
is built on proactivity, rapid response and an innovative approach between the MoH
and all community components.
Values
1. Missional work.
2. Friendly environment.
3. Flexible performance.
4. Proactive workforces.
5. Effective initiatives.
Annexes
50
Objectives
1. Directing the community resources to areas of priority.
2. Facilitating community participation in health development.
3. Maximizing the impact of community participation on health development.
4. Institutionalizing community participation in health development.
Tracks
1. Organizing and promoting charitable health activities.
2. Developing a volunteer system for health practitioners.
3. Building the capabilities of non-profit health entities and promoting their expansion.
4. Promoting health initiatives of community members.
Initiatives
1. Organizing and encouraging donations and endowments in the health sector.
2. Establishing a health charity fund.
3. The Community Partnerships Service Center.
Annexes
51
Workforce Development [24]
The current national health workforce situation is characterized by:
• Low ratios of clinical staff (particularly nurses) to population levels, and health
providers.
• Poor distribution of clinical staff and chronic shortage of staff in some regions.
• Low productivity of employed clinical staff.
• Limited numbers of prospective Saudi clinical staff in education;
• Low conversion rates from completion of clinical education in the KSA to MoH
employment, particularly nursing staff.
• A limited number of medical and healthcare professional bodies (such as the Saudi
Diabetes and Endocrine Association). [25]
• High reliance on expatriate clinical staff for both medical and nursing roles.
• Unquantified turnover of expatriate clinical staff (whether they are changing
employers or leaving KSA.
• Unquantified medical and nursing capacity at specialty levels.
• Unquantified clinical and clinical support capacity in the following professions:
o Dentists
o Psychologists
o Physiotherapists
o Occupational therapists
o Laboratory technicians
• Major resource and expertise gaps in nonclinical healthcare professions including:
o Chairmen and Non-Executive Directors for newly corporatized entities
o General management
o Finance
o Health economics
o Information and technology management
o Human resources
o Planning
o Estates
o Logistics
o Communications
o Legal
Annexes
52
• Low perceived status of the nursing profession among the Saudi population;
• Poor and fragmented data and statistics about the KSA health workforce, for
example the numbers trained or available for employment, employed; and
• Insufficient IT infrastructure to support workforce data management.
Understanding workforce capacity and demand
The VRO is currently working to analyze and allocate the attributes of the current workforce
professionals in the Kingdom. This includes, but is not limited to, understanding location,
specialties and sub-specialties per region and per facility, for all categories of staff. This
work will provide an annual forecast of healthcare workforce demand and quantify the
annual gap between supply and demand until 2030. The forecast will take into account
drivers of supply and demand, include technological changes, changes to the Model of
Care, and changes to the system resulting from corporatization.
Building workforce capacity
Within this work-stream the goal of the workforce strategy is to build national practice
capacity in all specialties and sub-specialties. This will ensure fair distribution across the
Kingdom based on needs. This includes an ambition to encourage qualified Saudi
personnel (presently in the KSA or abroad) to re-enter the KSA health workforce. This may
include (re-) education and (re-) training. There is also a program of work underway to
improve the attractiveness of nursing and other healthcare professions. There may also be
opportunities to recruit suitably qualified non-clinical staff from other sectors. This work is
scheduled for completion by the end of 2018. Further detail on this project is shown below.
Annexes
53
Annexes
Building workforce capability
As well as building workforce capacity there is the related goal to build capability through a
national program covering all specialties and sub-specialties and targeted on the needs
identified through the capacity planning work. This will include the accreditation of 50
hospitals as teaching hospitals offering training, and the establishment of five family
medicine academies across the Kingdom. In 2017 the focus was on immediately
developing training programs for the highest priority specialty and sub-specialty areas.
There will then be a three-phase rollout of a national training program, with each phase
lasting a year from 2018 to 2020.
54
The development of eHealth [26]
Digital health and healthcare, or eHealth, is the cost-effective and secure use of information
and communication technologies in support of health and health-related goals. This
includes healthcare services, health surveillance, health literature, and health education,
knowledge and research. [27]
The effective adoption of eHealth will:
• Facilitate the introduction for radically different approaches to care;
• Ease the entrance into the health space for new, non-traditional players;
• Provide disease prediction tools;
• Enable customer empowered personal management of care; and
• Help alert population health systems of communicable disease outbreaks.
eHealth is a critical enabler of the system reform program described in this strategy. It is
necessary to:
• Enable the adherence to and enforcement of health system quality and safety
practices;
• Provide people with personalized digital tools to promote health management, health
literacy and help navigate and interact with the health system;
• Integrate the continuum of health across all layers of the New Models of Care to
support the flow of personal health information;
• Monitor operational efficiencies as well as the outcomes of services provided;
• Monitor and optimize the use of healthcare resources, such as workforce, assets and
services;
• Provide access to high quality data to support medical research and policy making;
• Establish a Virtual Academy to build knowledge and skills necessary to deliver the
New Models of Care interventions;
• To achieve the substantial efficiency gains across key corporate functions of provider
organizations, including: finance, HR, IT, procurement, clinical services, clinical
support services and quality control;
Annexes
55
• Introduce activity based costing, patient accounting, revenue cycle and claims
management;
• Implement clinical coding practices;
• Implement a data quality program; and
• Enable health system performance reporting.
Investment in eHealth in the Kingdom is expected to deliver multiple benefits. These
include:
• Enhanced personal experience, convenience and responsibility – People will
have new forms of interacting with the health system and consume health services
differently. The physical healthcare system can be augmented with a virtual health
system improving access to healthcare services whenever and wherever people are
in need. eHealth will help people take more personal control over, and responsibility
for, their lifestyle and health.
• Safer and more effective services – Healthcare is an information and knowledge
intense industry. Critical healthcare decisions are made based on diagnostic and
clinical information at hand. Real-time access to accurate information enables better
assessment. This helps to avoid costly medical errors and adverse events caused by
missing or wrong information.
• Productivity increases for healthcare practitioners – Skilled medical practitioners
in the Kingdom are a scarce and expensive resource. eHealth can provide the
medical workforce with the tools and capabilities to enhance their productivity by
automating a majority of administrative tasks. It provides support by augmenting the
decision-making process with digital information and knowledge capabilities that do
not exist in the Kingdom today.
• Efficient and integrated system – More efficient use of healthcare resources is key
for achieving a sustainable health system. eHealth will enable better integration of
the various systems of healthcare from one care setting to another. A common
definition of data will underpin this capability and ensure consistent information can
be made available for treatment care. Making data more readily available will reduce
duplication and inefficiency in its collection.
• Creation of a new knowledge industry – Implementing digital technologies is a
challenge that requires a skilled labor force. Global experiences have shown that
eHealth programs have established tens of thousands of job opportunities, as well
as a significant number of health informatics startups by entrepreneurs. Investing in
eHealth not only provides benefits to the health system, but also to the wider KSA
economy.
Annexes
56
The National Transformation Program 2020 outlines a key target that by 2020 70% of the
Kingdom’s population should have a Unified Health Record. This has been formalized as
“70% of the population will have at least one record in either of the four 2020 National
eHealth Systems.” Activities that contribute to the achievement of that goal are marked on
the graphics below with a target.
Implementing eHealth
Scope of the eHealth Strategy and the role of the Saudi Health Exchange
The KSA eHealth strategy has important elements that are health sector wide, and will link
the MoH, other government services and private segments of the health system. The Saudi
Health Exchange (SeHE) will be an important mechanism for integration of digital
capabilities across all segments of the health system.
Solution classes, portfolios and programs
The implementation strategy presently comprises, three solution classes, four portfolios
and 52 programs.
Solution classes
The three solution classes distinguish:
• National Solutions –Those that will be provided by one provider for all system
segments (including national solutions specifically enabled by SeHE);
• Cluster Solutions – Those that may vary from MoH cluster to cluster (albeit to
common data standards) and are intended to facilitate MoH cluster and then ACO
enablement; and
• Enterprise solutions – Those that may be provided by multiple vendors (again to
common core data standards) to all system segments.
Annexes
The interrelationship of the solution classes and SeHE is illustrated below:
57
Annexes
Portfolios
The portfolios group individual programs by major function: clinical programs; infrastructure
and operations; personal health services; and administrative services.
The apportionment of programs to the portfolios is shown in the graphics below. The tables
also show, the class, timeline and key milestones of each program.
58
Annexes
59
Annexes
Annex B. A description and summary analysis of the
Harvard and Oxford value frameworks
Introducing the frameworks
Value in healthcare can be defined most simply as “outcome divided by cost.”
The fundamental principles of value based healthcare in healthcare is to first align
stakeholders around the shared objective of improving health outcomes delivered at a
given cost. Then to give stakeholders the autonomy, tools and accountability to pursue the
most rational ways of delivering value to all the people in need within the population. [28]
In some cases, people with specific needs generally reach the right service, for example
people with limb fractures or cancers. For most health problems however, the need is not
always so clear. People with conditions like depression, hip pain or asthma only reach the
specialist service if they present to the primary care service and the primary care clinician
recognizes the need for specialist help. In every country this is influenced by many factors
which mean that:
• Those who reach the specialist service are not necessarily those who will benefit
most; and
• Many of those who do not reach the specialist service would get greater benefit than
those who have done so.
The population as a whole will derive more value if these mismatches were prevented.
60
Annexes
The value-based approach to care rests on four foundational elements:
• Measuring systematically the allocation of resources to different sub-groups, for
example people with cancer or people with mental health problems;
• Measuring health outcomes that matter to patients and the costs required to deliver
those outcomes across the full cycle of care;
• Tracking those outcomes and costs for defined population segments on an ongoing
basis, and
• Developing customized interventions to improve value for each population segment.
Five enablers are considered key to accelerating the adoption of value based
healthcare:
• Health informatics, to facilitate the easy collection, analysis and sharing of outcomes
and cost data;
• Benchmarking, research and tools, to leverage data on outcomes and the costs for
clinical practice improvement and innovation;
• Value-based payments, to create incentives for all stakeholders to focus on value
and a move away from activity based payment which encourage low value
interventions;
• Innovations in organizing care delivery, to improve coordination across the health
system with a single specification relevant to the whole Kingdom, with delivery being
delegated to local networks; and
• A new culture that promotes collaboration and stewardship.
The Harvard framework – Value based healthcare delivery [29]
The Harvard framework focuses principally on optimizing value at the level of individual
patients who have presented for treatment. It contains the systematic reduction of
unwarranted variation in treatment selection, outcomes and costs.
The Harvard framework also concerns itself with optimizing clinical micro-systems within
various institutional frameworks; and the optimization of treatment for defined populations
or groups with identical or related diagnoses.
61
Annexes
The Harvard framework reflects the historic market driven context of healthcare funding
and delivery in the USA. This is where there is a highly inefficient distribution of services
(mediated in large part through voluntary private health insurances) and has reflected the
ability to pay more than ability to benefit. The Harvard framework does not address
evidence based optimization of defined or limited resources in the system, and the
associated trade-offs between different services and dissimilar patients. Nor does the
framework address the evidence based optimization in distributing preventive, curative,
care and palliative services for a defined population.
The Oxford framework – Higher value healthcare [30]
In contrast the Oxford framework reflects the historic UK and European context in which
governments have greater control over the distribution healthcare either through public
expenditure or government mandate.
The additional insight of the higher value healthcare framework developed at Oxford
University is that, in systems that are fully and actively committed to the health and
healthcare of their entire population, such as the Kingdom, an emphasis on personal value
needs to be complemented by:
• A focus on allocative efficiency at the intermediate level; and
• The active identification of people who are not receiving care, but who may derive
more value than those have been referred for treatment. The Kingdom, through its
policy of total population coverage, is committed to reducing inequity as well as
increasing efficiency.
To that end the framework defines three dimensions of value:
The higher value healthcare framework is particularly valuable in the Kingdom as it helps to
distinguish not only different types of value, but also three distinct system levels at which
value can be secured:
• At a national level
• At a sub-population level defined not only geographically but also by need
• At an individual level
62
Annexes
The ambition to improve allocative and utilization value at both national and sub-population
levels also links directly to the related concept of population health management (PHM).
This has been defined as “the technical field of endeavor which utilizes a, variety of
individual, organizational and cultural interventions to help improve the morbidity patterns
and the health care use behavior of defined populations”. PHM is distinguished
from disease management by including more chronic conditions and diseases, by use of “a
single point of contact and coordination”, and by “predictive modelling across multiple
clinical conditions” [31]. It introduces a third dimension to healthcare and is defined as
population healthcare defined by being focused on people with a common symptom such
as back pain, or people with the same condition such as breast cancer, or people with a
common characteristic such as people in the last year of life; and not on institutions or
specialties or technologies. This is the third dimension of health care.
63
Annexes
Annex C. Bibliography of key documents
Atun, R., Kruk M. et al, Saudi Arabia Health System Strategy Project, May 2017,
Harvard University, School of Public Health
Dzau, Victor J., Mark B. McClellan, et al (2017) Vital Directions for Health and Health
Care: Priorities from a National Academy of Medicine Initiative. . JAMA. Published
online Mar 21 2017. doi: 10.1001/jama.2017.1964
KSA, MoH, eHealth Strategy, Version: Final for approval, 13 July 2017
KSA, MoH, eHealth Roadmap (draft updated), .ppt, December 2017
KSA MoH, VRO, Business Case for the Health System Reform Component of the
Kingdom of Saudi Arabia’s Vision 2030 Program, Version 2.0, 31 May 2017
KSA, MoH, VRO, Corporatization Update Meeting, 18 October 2017
KSA, MoH, VRO, Model of Care, .ppt presentation, October 2017
KSA, MoH, VRO, Private Sector Participation (PSP) in Healthcare, Commercial
Prospectus, undated (2017)
Ministry of Health, Healthcare system financing strategy for the Kingdom of Saudi
Arabia, (Draft at October 2017)
Ministry of Health, Healthcare workforce strategy (2017-2030), .ppt, May 2017
Saltmann R., J Figuaras, C Sakellarides (1998), Critical Challenges for Healthcare
Reform, Open University Press, UK
Walshe K. and J. Smith (2006), Healthcare Management, Open University Press UK.
64
Annexes
Annex D. Endnotes
1 KSA Vision 2030, Strategic Objectives and Vision Realization Programs, pdf, undated
2 KSA MoH VRO, Ministry of Health Transformation Journey, Supervisory Committee, 12 December
2017
3 https://en.wikipedia.org/wiki/Hajj#Number_of_pilgrims_per_year
4 KSA MoH VRO, Business Case for the Health System Reform Component of the Kingdom of
Saudi Arabia’s Vision 2030 Program, Version 2.0, 31 May 2017
5 http://www.wpro.who.int/health_services/health_systems_framework/en/
6 http://siteresources.worldbank.org/HEALTHNUTRITIONANDPOPULATION/Resources/
281627-1095698140167/HealthSystemsAnalysisForBetterHealthSysStrengthening
7 http://www.ihi.org/Engage/Initiatives/TripleAim/Pages/default.aspx
8 https://nam.edu/initiatives/vital-directions-for-health-and-health-care/
9 https://www.england.nhs.uk/five-year-forward-view/
10 http://www3.weforum.org/docs/WEF_Insight_Report_Value_Healthcare_Laying_Foundation
11 https://www.phc.ox.ac.uk/research/value-based-healthcare
12 Saltmann R., et al (1998)
13 KSA, MoH, VRO, Model of Care, .ppt presentation, October 2017
14 KSA, MoH, VRO, Corporatization Update Meeting, 18 October 2017
15 https://openknowledge.worldbank.org/handle/10986/15145
16 https://www.leadershipacademy.nhs.uk/wp-content/uploads/2013/06/NHSLeadership-
HealthyNHSBoard-2013
17 Ministry of Health, Healthcare system financing strategy for the Kingdom of Saudi Arabia, (Draft
at October 2017)
18 KSA, MoH, VRO, [Financing] Strategy Alignment, May 2017
19 http://www3.weforum.org/docs/WEF_Insight_Report_Value_Healthcare_Laying_Foundation
20 Shacklady Smith A., Chapter 22: Appreciating the challenge of change in Walshe and Smith
(2006)
21 KSA MoH VRO, Business Case for the Health System Reform Component of the Kingdom of
Saudi Arabia’s Vision 2030 Program, Version 2.0, 31 May 2017
22 KSA, MoH, VRO, correspondence with Governance work theme, 26 March 2018
23 KSA, MoH, VRO, Private Sector Participation in Healthcare, Strategic Framework, .ppt, January
2017
24 Ministry of Health, Healthcare workforce strategy (2017-2030), .ppt, May 2017
25 http://sdea.org.sa/
26 KSA, MoH, VRO, eHealth Strategy, v38, .ppt, 7 May 2017 and KSA, MoH, eHealth Strategy,
Final for approval 13 July 2017
27 http://www.who.int/gb/ebwha/pdf_files/WHA58/WHA58_28-en
28 http://www3.weforum.org/docs/WEF_Insight_Report_Value_Healthcare_Laying_Foundation
29 https://www.isc.hbs.edu/health-care/vbhcd/Pages/default.aspx
30 https://www.phc.ox.ac.uk/research/value-based-healthcare
31 https://en.wikipedia.org/wiki/Population_health#Population_health_management_(PHM)
https://en.wikipedia.org/wiki/Hajj%23Number_of_pilgrims_per_year
http://www.wpro.who.int/health_services/health_systems_framework/en/
http://siteresources.worldbank.org/HEALTHNUTRITIONANDPOPULATION/Resources/281627-1095698140167/HealthSystemsAnalysisForBetterHealthSysStrengthening
http://siteresources.worldbank.org/HEALTHNUTRITIONANDPOPULATION/Resources/281627-1095698140167/HealthSystemsAnalysisForBetterHealthSysStrengthening
http://siteresources.worldbank.org/HEALTHNUTRITIONANDPOPULATION/Resources/281627-1095698140167/HealthSystemsAnalysisForBetterHealthSysStrengthening
http://www.ihi.org/Engage/Initiatives/TripleAim/Pages/default.aspx
https://nam.edu/initiatives/vital-directions-for-health-and-health-care/
https://www.england.nhs.uk/five-year-forward-view/
http://www3.weforum.org/docs/WEF_Insight_Report_Value_Healthcare_Laying_Foundation
https://www.phc.ox.ac.uk/research/value-based-healthcare
https://openknowledge.worldbank.org/handle/10986/15145
https://www.leadershipacademy.nhs.uk/wp-content/uploads/2013/06/NHSLeadership-HealthyNHSBoard-2013
https://www.leadershipacademy.nhs.uk/wp-content/uploads/2013/06/NHSLeadership-HealthyNHSBoard-2013
https://www.leadershipacademy.nhs.uk/wp-content/uploads/2013/06/NHSLeadership-HealthyNHSBoard-2013
http://www3.weforum.org/docs/WEF_Insight_Report_Value_Healthcare_Laying_Foundation
http://www.who.int/gb/ebwha/pdf_files/WHA58/WHA58_28-en
http://www3.weforum.org/docs/WEF_Insight_Report_Value_Healthcare_Laying_Foundation
https://www.isc.hbs.edu/health-care/vbhcd/Pages/default.aspx
https://www.phc.ox.ac.uk/research/value-based-healthcare
https://en.wikipedia.org/wiki/Population_health%23Population_health_management_(PHM)
SaudiMOH Saudi_MohSaudiMOH937 MOHPortalwww.moh.gov.sa