DHA 8044 CU Safety Culture Improvement Presentation

Hey Tutor, you will produce 20 powerpoint slides
with speaker notes for the assignment below:
Topic: Creating a Culture of Safety and
Compliance
Introduction
Creating an organization-wide culture of safety is vital to any health care
organization. For this assignment, assume that you are working as a health care
executive for a large health care system that includes both in-patient hospitals
and a wide array of outpatient services. Your role includes ensuring all clinical
department heads have resources necessary to create and ensure a culture of
compliance. You have decided to create training for top managers to help them
develop a culture of safety for their staff. The training is designed for nursing
managers, intake managers, clinical supervisors and other leaders that oversee
staff with direct patient care.
Instructions
For this assignment, develop a 15–20 slide PowerPoint presentation that
outlines your culture of safety and compliance for your department training. Your
training presentation must include the following concepts:
1. Examination of the role the quality professional has when working with
federal, state, or local agencies.
2. Description of how leaders develop and establish protocols to identify and
monitor patients who qualify for being at risk.
3. Analysis of best practices, current trends, and proven approaches for creating
a safety culture with various departments of a health care organization.
4. Analysis and recommendation of evidence-Âbased best practices that help
improve an organization’s health care safety culture.
Submission Requirements
Your presentation should meet the following requirements:
•
•
Length: 15–20 slide PowerPoint presentation, excluding the cover
slide and references list. Include slide numbers, headings, and running
headers.
References: 3–5 current peer-reviewed references.
Format: Use current APA style and formatting, for citations and references.
Font and font size: Fonts and styles used should be consistent throughout
the presentation, including headings.
Review the scoring guide to ensure you understand the grading criteria for this
assignment. Submit your PowerPoint presentation as an attachment to the
assignment area.
•
•
Note: Your instructor may also use the Writing Feedback Tool to provide
feedback on your writing. In the tool, click the linked resources for helpful writing
information.
•
Rozovsky, F. A., & Woods, J. R., Jr. (Ed.). (2011). The handbook of patient safety
compliance: A practical guide for health care organizations. San Francisco, CA:
Jossey-Bass. ISBN: 9781118086995.
• Read Chapter 10, “Patient Safety in Human Research,” pages 160–182.
HAPTER TEN
PATIENT SAFETY IN HUMAN RESEARCH
Rodney K. Adams
Since the Second World War, autonomy and safety have been primary considerations in human
research. The Nuremberg Code, The Declaration of Helsinki, and the Belmont Report are all important
statements on protecting participants in human research. Similarly, federal human research regulations
are focused on the informed consent and safety of research participants. The Common Rule, used by
many federal agencies (see, for example, 45 C.F.R. 46, 2003), is a detailed road map of how a research
protocol is to be created and implemented. Institutional review boards (IRBs) are charged with
protecting research participants. Unfortunately, federal regulations do not apply to all human research
in the United States and even the most detailed regulations do not preclude inadvertent lack of
attention to detail. Zeal for knowledge and advancement may cloud a researcher’s judgment. Therefore
a health care provider engaged in research must have a consciousness of safety when drafting,
implementing, and monitoring a research protocol.
Well-publicized events at many academic centers have pointed out that errors in safety are of serious
concern for participants in human research. More than a year prior to the 1999 completion of the
Institute of Medicine report To Err Is Human, the deputy inspector general of the U.S. Department of
Health and Human Services (HHS), during testimony before Congress, in essence challenged IRBs to do a
better job to ensure the rights and safety of human research participants. This testimony is
encapsulated in the June 1998 Office of Inspector General (OIG) report Institutional Review Boards: A
Time for Reform (Department of Health and Human Services, OIG, 1998). The OIG called for data safety
monitoring boards (DSMBs) that would provide summary assessments of adverse event reports to IRBs
so that the IRBs could better assess the ongoing safety of research trials. Further, the OIG recommended
elimination of the forum shopping done when sponsors unhappy with the reviews carried out by one
IRB take their protocols to another IRB without telling the second board about the earlier one’s
involvement or determinations. The OIG viewed this as denying the second IRB important information
and short-circuiting the opportunity for human subject protection. In keeping with the theme of human
subject protection, the OIG recommended that IRBs become more involved with what transpires during
research trials. For particularly high-risk trials, this might involve the use of intermediaries, counselors,
or other third parties to observe the consent process. IRB members might conduct random
unannounced visits to review relevant documentation and to oversee the consent process.
The OIG reported in April 2000 that a number of changes had been made in human research programs
but that many recommendations remained unheeded (HHS, OIG, 2000). Some of these
recommendations involved patient safety. During this time frame, major structural changes in federal
oversight were initiated involving the Office for Protection from Research Risks (OPRR), a part of the
National Institutes of Health (NIH) with authority over NIH-funded research. In June 2000, the secretary
of Health and Human Services replaced OPRR with the Office for Human Research Protections (OHRP),
which reports directly to the secretary. Positioned at the departmental level, the new agency was
designed to provide “leadership” for the federal agencies funding human research. Since its inception
OHRP has moved to develop strategies for educating IRB members and clinical investigators and to
improve regulatory compliance.
This chapter describes a practical approach to enhancing the safety of human research as well as to
responding to adverse events. The work of DSMBs is examined. Research strategies that lessen the risk
of adverse events and unanticipated outcomes are also described. Management of concurrent
situations, incident reporting, and sentinel event reporting are discussed in the context of human
research.
Creating a Patient Safety Environment for Clinical Trials
A number of practical steps can be implemented to create a safer environment for human research.
These steps include employing effective screening tools to identify participants who are at risk of
experiencing an adverse event, having appropriate consent procedures, having an effective on-call
system, using signage and checklists, supplying participants with MedicAlert-style bracelets, and
educating all involved. Making vital information available to a participant’s other health care providers
as well as responding to a participant’s concerns in a timely manner is important. A related concept is
having a contingency plan for adverse events. Each of these categories is discussed in the following
sections.
Screening Tools
Critical to successful management of a clinical trial is the proper selection of research participants. The
participants must meet the study design requirements; however, they may also present with a
constellation of risk factors. These risk factors may have nothing to do with the clinical study, yet in
some cases their presence should bar the individual from inclusion in the investigation. For example, a
trial might involve investigation of a new diabetes medication. Screening information obtained from an
individual may clearly indicate that he or she is eligible for the study. However, additional scrutiny may
reveal a history of noncompliance with structured medication schedules. No familial caregiver or friend
who might assist the individual with staying on the required regimen is available. Because it is critical for
participants to stay within the rigid parameters of the study, this is a person at risk, who probably should
not be enrolled in the study.
Consent Tools
Consent is a pivotal component in human research and clinical trials. It is equally important for a
participant’s safety. Consent is more than securing a written authorization for inclusion in a study; it is a
communication tool that enables the principal investigator and study personnel to impart important
information for safe participation in a research protocol. Disclosure of possible anticipated side effects
should always be part of the discussion. However, from a safety perspective, what is equally important is
to use the consent process as a forum for schooling participants on what steps to take if such side
effects occur. Additionally, research consent procedures usually include some discussion about
“unforeseeable” risk factors. Participants may have unusual or unanticipated responses to a study
intervention or an investigational drug. Rather than having these responses dismissed as “nothing to
worry about,” participants should be encouraged to report them promptly. A practice that is gaining
more acceptance is to involve a family member of the participant in the consent process. This can be
done only with the permission of the research participant. The rationale is that the family member may
have a perspective different from the participant’s on key issues such as predisposing risk factors, ability
to adhere to the protocol, and early detection of adverse events. At the same time, the family member
can school others about what to look for in terms of reactions or adverse events and the appropriate
response to such situations. For example, a participant may be enrolled in a study that involves an
implanted investigational cardiac pacemaker device. He collapses at home on the living room floor, and
911 is called. The emergency medical services response team has no idea that the individual is enrolled
in a clinical trial. No one in the immediate family is aware of the individual’s role in the study, and so
they are unable to shed light on what might have caused the person to collapse. The unfortunate result
is a therapeutic intervention that is incompatible with the study, resulting in harm to the subjectpatient. The chance of avoiding such real-life risk situations is increased if one or more family members
are involved from the outset in the communication process called consent. As a communication tool the
consent process can be used to reinforce the importance of alerting study personnel to risk-prone
situations and aid those assisting a subject-patient who requires a therapeutic intervention.
Checklists
A practical approach to maintaining research participant safety involves the use of checklists. When
used as guidelines or pathways, these tools are not “cookbook” clinical research. Instead, they serve as
reminders or helpful algorithms for reducing unwanted or risk-prone variations in the proper
management of a clinical protocol. Some of the checklists can be for the use of clinical research staff.
Others can be useful guides for inpatient, emergency department, and urgent care personnel who
encounter research participants as patients. Finally, yet other checklists can provide important
reminders to research participants and their family members about what to do when reactions or
adverse effects occur.
On-Call Systems and Interactive Web Sites
Other measures are also important in maintaining excellent communication. One involves having an
effective on-call protocol so that when research participants call with questions or to report what they
believe is an adverse event, someone can respond quickly. This is especially important in high-risk trials
and also in those protocols in which the consent document “assures” participants that if they call,
someone will get back to them promptly. An on-call system may forward calls to someone carrying a cell
phone or a pager. In practical terms this means there must be timely follow-through in returning calls.
Another technique is to provide participants with the Internet address for a secure Web site. (This
presupposes that research participants have access to the Internet.) Using their protocol identification
number or a similar code process, research participants could visit a Web site that includes frequently
asked questions (FAQs) and answers to common concerns stemming from involvement in the research
trial. It might also have a monitored chat room or “e-mail us” function through which participants could
receive prompt answers to such questions as, “Am I experiencing an adverse reaction to the study
drug?” or, “I missed two of my medications yesterday and this morning. What should I do?” For the Web
site approach to meet participants’ needs, someone on the study staff must monitor the Web and email traffic and ensure that a timely response is made. (Some investigators have complained that
research participants are organizing their own chat rooms and causing problems such as unblinding
studies. This may lead to the participants’ reaching incorrect conclusions about the study drug or to the
participants’ being influenced by the group discussion. An official Web site that addresses the primary
concerns of the participants with reliable responses is a better course.) The Internet provides a powerful
tool for overseeing the participants’ concerns and questions and for averting possible harm to
participants and reducing the likelihood that they will drop out of the investigation. In this way Web site
information might contribute to enhancements in the study design.
Signage
Signage is also an important step toward creating a culture of safety for clinical research trials. It
involves well-displayed notices in languages used by members of the community in which the facility is
situated. Signage is particularly important in hospitals, freestanding urgent care units, ambulatory
patient care units, and retail pharmacies. The message to be communicated is simple:
Are You Taking Part in a Clinical Trial or Research?
If so, please tell us when you register for treatment.
These signs are a friendly reminder, a prompt to disclose an important consideration that caregivers
need to take into account when evaluating a patient’s needs, performing therapeutic interventions, or
prescribing medication. Why place the same type of signage at multiple locations? The rationale is
simple: these messages are a communication safety net to catch research participants who might
otherwise miss an opportunity to provide salient information about participation in a clinical trial that
has or could have an adverse impact on proposed treatment. For example, a research subject-patient
might have missed the prompt at the free-standing urgent care center she visited. However, seeing it at
the retail pharmacy she visits next, the participant says to the pharmacy technician, “Yes, I am in a
study. Why is that important? After all, the doctor only prescribed some antibiotics and over-thecounter pain medication.” It is important because this may be the last opportunity to prevent an
untoward synergistic effect between an investigational drug and routine medical treatment. When told
about her participation in the study, the pharmacy technician should inform the pharmacist. In turn the
pharmacist should contact the prescribing doctor, who should discuss the treatment with a study
investigator or coordinator. This conversation may turn out to be an early warning of an adverse effect
or it may be about a routine infection that has nothing to do with the study. At the very least the
communication provides peace of mind for the patient, prescribing physician, and investigator and
allows a decision to be made about the next steps, such as clearing the proposed prescription or
suggesting an alternative. The root of this safety process is the signage in the retail pharmacy that
prompted the research participant to notify staff about her involvement in a research protocol.
Research Bracelets
Many individuals wear bracelets or necklaces emblazoned with medical warnings or alerts. Some include
information about allergies, underlying medical conditions, medication requirements, or organ donor
status. Still others provide instructions on resuscitation requests. The information is designed to alert
first responders and emergency department personnel about what to do when assisting the person
wearing the bracelet or necklace. It is a very practical idea for clinical trial participants and can increase
their safety. Study participants often travel miles away from the location in which they participate in an
investigation. If a participant then requires medical attention, there is no way in which his caregivers
would know that he was enrolled in a study unless he were to tell them. He might need treatment as the
result of an adverse reaction to a study drug or injuries sustained in an unrelated event. Without the
benefit of knowing that the patient is taking part in a research protocol, care providers may make a
misdiagnosis or provide inappropriate treatment given the medications or devices involved in the clinical
study. Providing participants with warning bracelets or necklaces can greatly reduce this risk to
participant safety. Bracelet or necklace information should include the following:
The name of the protocol
The patient’s clinical research identification number
A toll-free number or e-mail address to contact for urgent information
In addition the information should include a warning statement, as follows:
Warning!
I am enrolled in a clinical research study. Please contact the study coordinator before performing
treatment!
Taking this additional step helps to ensure the safety of research participants. It also may help to
maintain the integrity of the study, so that participants are not lost to follow-up as a result of an
unfortunate response to a need for treatment.
Even though individually identifiable information is involved, it is possible to create a HIPAA (Health
Insurance Portability and Accountability Act) compliant, firewall-protected Web site for posting
important clinical trial information needed for the treatment of a research participant who takes ill miles
away from the clinical trial location. A secure Web site that is password protected is much more useful
to a treating physician than are records locked in a file cabinet at night. The information is available to a
treating physician regardless of where in the world the research participant might travel. Tracking access
to the Web site is no more difficult than for any electronic medical record.
Education
Everyone involved in clinical research trials can benefit from safety education. Providing staff with
practical training on safety practices can prevent them and the participants from experiencing harm. The
same is true for research participants and their families: understanding the protocol, knowing what to
look for in terms of early warning signs of adverse effects, and knowing how to interact with the
research staff and treatment facilities can all contribute to positive outcomes. Offering such education
provides a useful way of addressing issues that may become life- or health-threatening events, and it
may improve the quality of clinical data. Rather than letting situations evolve into crises, the prudent
approach is to anticipate possible adverse events through cogent education that leads to successful
strategies for avoiding or managing problems.
These measures are illustrative of the steps to take to create a culture of safety in studies that use
human participants. Other steps may be added that are consistent with ongoing patient safety strategies
in a health care organization or a contract research organization.
Continuous Quality Assurance
Vitally important in all aspects of health care, continuous quality assurance is equally important in
human research. It may take the shape of random auditing, skills checklists, participant interviews, and
other tools that can ensure that the research protocol is being followed. A certain amount of
redundancy, in the form of double-checks, increases in importance as the risk of harm to a participant
increases in a study. When the line between a medication’s therapeutic dosage and fatal dosage is very
small, who draws up the drug, how and where the drug is administered, and how the event is
documented are all part of the quality assurance process.
Responding to Adverse Events
The first and foremost objective in health care is to provide care to the patient. This is equally true when
responding to a potentially adverse event in a research protocol. Treatment of the patient is the first
priority. Then comes the paperwork. It is imperative that IRBs receive prompt notification about adverse
events stemming from research studies. These events include situations in which participants
experience serious harm or unforeseen accidents or problems. When an adverse event such as a serious
injury or a death occurs, a thorough review is warranted. A failure analysis or root cause analysis is
warranted to ascertain why the event occurred. The results may be surprising.
Clinical Contingency Plan for Adverse Events
Along with properly trained personnel to field questions and the use of warning bracelets or necklaces, a
plan should be in place to manage the care of participants who experience an adverse event. Such a plan
should include
Having clinical management algorithms for differentiating adverse reactions to a device or drug under
investigation in the study and the results of drug-drug or drug-food interactions or other factors
Having a procedure for notifying the clinical care team about study-related adverse events and
successful treatment plans from other episodes when the adverse events occurred.
Providing instruction sheets to participants at the time of their enrollment in the study reminding them
to alert their primary care provider, family, and urgent care or emergency department staff that they are
involved in a clinical trial of an investigational drug or device
Providing instruction sheets to on-call personnel and those monitoring e-mail traffic on how to manage
the situation when a participant is believed to be experiencing an adverse event
Following-up once care is completed to determine what worked and what did not work in managing the
condition suffered by the participant who has experienced an adverse event related to the drug or
device under investigation
Analysis of Adverse Events
Although many observers may be ready to ascribe the adverse event that causes injury or death to the
investigational device or drug, it may instead prove to be the culmination of a series of system failures.
Delving into the issue is worthwhile as it could mean correcting system failures rather than suspending a
valid research protocol. Models exist for this type of investigation, including those with which accredited
hospitals are familiar (such as the Framework for Conducting a Root Cause Analysis, available from the
Joint Commission on Accreditation of Healthcare Organizations [ JCAHO]) and those with which many
manufacturing industries are conversant (such as PROACT, root cause failure analysis software from the
Reliability Center). It must be appreciated, however, that even if a timely and thorough investigation of
an adverse event occurs, it may not preclude the possibility of litigation stemming from an injury to or
the death of a participant, investigation by a coroner or medical examiner, criminal probes by law
enforcement, or compliance oversight investigations by pertinent federal or state regulatory bodies such
as the Food and Drug Administration (FDA), the Office for Human Research Protections (OHRP), or a
state Board of Medicine.
A Systematic Approach to Managing Adverse Events
With such an array of possible regulatory investigations, it is important that health care facilities or
contract research organizations have in place a systematic approach to investigating, managing, and
promptly reporting the findings on adverse events. Such an approach would include the following
considerations:
Investigation. Once an adverse event occurs an investigation should be started immediately to
determine what gave rise to it. Care should be taken not to jump to conclusions but rather to look at
processes that should have occurred and to compare this information with what is learned about what
did occur. For example, a research participant may have received ten times the dose of radiation
expected in the research protocol. The investigators should not jump to the conclusion that “somebody
did something wrong.” Rather, they should look at documentation, communication, and equipment
calibration processes as well as staff education. Taking this approach it may turn out that several
processes or procedures failed and coalesced into a systems failure. Alternatively, human factors may be
involved. It may be found that a staffer “took it for granted” that the radiation machine was always set
at the proper parameters and did not bother to double-check the calibrations per study protocol. In this
situation the adverse event involves a human failure as well as a systems failure, implicating a lack of
fail-safes to prevent such an adverse event.
Sequestration of Documentation. As part of an investigation all pertinent documentation should be
obtained and copies sequestered pending the disposition of the matter. This does not mean, however,
that the original documents must be sequestered. Rather, verified copies can be maintained for
investigative purposes. The copies should be date and time stamped to indicate when they were
prepared, along with the name and position of the person doing the copying. Most qualified health
information specialists are conversant with applicable state laws that address the “usual and customary”
business practices that may be used to prepare a credible, authenticated copy of such records. Taking
these measures does not impede the use of the original record for purposes of ongoing research work
or the treatment of the research participant.
Sequestration of Ancillary Record Information. Paper documentation alone is not sufficient for purposes
of an adverse event investigation. As with hard copy information, verified copies should be made of
electronic record data. These copies should be preserved in a format that is timed and dated and
impervious to modification. This format should also include the name and title of the individual who
makes the reproduction. A recognized process should be used so that the reproduction will be
considered authentic.
Sequestration of Equipment. It is important to secure any devices involved in an adverse event,
including investigational devices. Care should be taken in the handling of such equipment too. Specific
procedures should be in place that address who may examine the device and whether the equipment
may be removed from the premises for this purpose. Manufacturers’ warranty language often carries
warnings that any tampering with or removal of the factory seal on a device breaches the warranty.
However, when the manufacturer’s representative and a biomedical engineer from a reputable third
party examine the device, this does not void the warranty. The idea of a third party viewing a medical
device can prove very sensitive. A confidentiality agreement may be necessary to allay fears that
intellectual property may be stolen. Then, although the representative from the third-party biomedical
engineering firm may examine the device and document or testify about the findings, he or she will be
precluded by the confidentiality agreement from sharing or using the information for any other purpose.
Interviews. An important part of the adverse event investigation is to get information from firsthand
observers. The interviews to accomplish this should be carefully constructed to avoid biases or
presuppositions about what transpired. All those involved in the event, from those who prepared the
protocol, test article, or equipment to the person carrying out the protocol should be interviewed.
Interviews with the study participant are equally important. Documentation for the interviews should
include the following:
The subject (topic) of the interview
The interviewer’s name and title
The interview time, date, and location
A list of the questions posed
The responses to the questions
The assistance of an attorney will be needed to determine whether these interviews can be considered
privileged and thus protected from disclosure and what steps will be necessary to ensure that the
privilege is applicable.
Process Indicators. An adverse event investigation can be time consuming and expensive. Because the
results may mean suspension of a trial or a regulatory sanction, it is important to take a consistent,
meticulous approach to examining adverse events. A series of process indicators, or prompts, can be
used to guarantee a consistent and thorough approach to adverse event investigations:
Identify Processes Involved in the Event
Consent
Confidentiality
Medication
Surgery
Radiation
Chemotherapy
Physical therapy
Psychotherapy
Behavior management or modification
Investigational device
Investigational drug
Combination investigational device and conventional treatment (describe treatment)
Combination investigational drug and conventional treatment (describe treatment)
Clinical trial staff education and training
Education and training for research participants and family members
Identification of onset of adverse event
Communication linked to identification of onset of adverse event
Reporting of adverse event
Treatment for adverse event
Answer These Questions
Have authenticated copies of relevant documentation been obtained?
Have authenticated copies of electronic files been obtained?
Has equipment been sequestered?
Have interviews been completed? For each interview supply
Name
Title (if applicable)
Date of interview
Interviewer
For Each Process Identified, Answer These Questions
Did a process variance occur?
If yes, describe the variance.
List Identified Opportunities for Process Improvement
A Visual Algorithm for Adverse Event Investigations. Those who must manage an adverse event may find
it very useful to construct a flowchart. It should depict the steps to take in identifying, managing, and
reporting an adverse event in a research protocol.
Communication About an Adverse Event
Communicating about an adverse event involves determining what reports are needed, to whom the
reports should be made, and how study participants will be involved. In most cases, an adverse event
will need to be reported to several entities in a very short time period.
Determining Whether the Event Requires Reporting. One important aspect of the investigation is
determining what types of reports must be completed subsequent to an adverse event. The threshold
consideration, however, is determining whether an individual’s outcome constitutes an adverse event.
The definition of adverse event may vary among interested constituencies. For example, an end-user,
device-related serious injury may trigger reporting under the Safe Medical Devices Act and its
accompanying regulations. If an individual is receiving both accepted treatment (patient) and an
experimental protocol (research participant), there may be no obligation to report the situation to the
IRB. In practical terms, however, it may be prudent to notify the IRB even when a nonresearch aspect of
the participant’s treatment leads to an adverse outcome. Moreover, even if federal regulations do not
trigger a reporting obligation in such situations, a contract with a private sponsor may do so. For
example, the sponsor of an experimental drug under investigation for treatment of cancer might want
to know that the serious injury to a study participant came about not as a result of the study but for
totally different reasons. The point is that an investigation should yield sufficient information with which
to make a reporting determination. Unfortunately, the time limit within which a report must be made
often gives investigators little opportunity to determine the cause of the adverse event.
Determining Who Must Receive a Report. When an adverse event occurs, it may trigger reporting
obligations under any of several federal, state, and contractual requirements. A number of questions
should be asked in determining report recipients:
Did the adverse event stem from a clinical protocol, or was the individual a participant in a clinical trial
and a patient receiving accepted treatment?
Did the adverse event involve an approved or an investigational device?
Did the adverse event involve an approved drug or a drug under investigation?
Did the adverse event involve an approved drug that was being tested for a new use?
Did the adverse event result in the unexpected death of the participant?
Did the adverse event occur in a JCAHO-accredited facility? If so, did it constitute a reviewable event as
defined by JCAHO?
Has the adverse event triggered a mandatory reporting obligation under applicable state law?
Does the adverse event require a report to the sponsor under the terms of the agreement with the
sponsor?
Does the adverse event require a report to the IRB under the policies and procedures of the IRB?
Does the adverse event require a report to other facilities carrying out the same or similar studies,
under the terms of the protocol, the policies and procedures of the IRB, or the agreement with the
sponsor?
Does the adverse event require the notification of either the quality assurance or risk management
officer under the policies and procedures of the health care facility or contract research organization?
If the facility in which the adverse event took place is obliged to follow the Conditions of Participation in
Medicare or Medicaid for hospitals, is there a requirement to notify senior management of participant
complaints or grievances linked to a research-related adverse event?
If the health care organization has a corporate compliance plan, is there a responsibility to notify the
compliance officer about a research-related adverse event?
Is there a responsibility to notify liability insurance carriers about a research-related adverse event? If
the health care organization has an insurance captive, is there an obligation to report to the captive
manager?
If the adverse event is related to competency, misrepresentation, fraud, deceit, or unprofessional
conduct, is there an obligation to report to state licensing bodies or the federal Office of Research
Integrity?
Depending on the circumstances of the specific case, the answers to these questions may identify
multiple reporting obligations. Some of these reporting obligations may compete with or conflict with
one another in terms of order of priority and the amount of detail to be provided in each instance.
To be certain that all the necessary reports are made, it is suggested that the IRB and research office
personnel work together to develop a reporting matrix. Developed by a team of key stakeholders, this
chart can summarize all the necessary tracking information: when reports should be made, to whom,
and in what priority. Any concerns about adverse event reporting should be referred to the risk
management department or legal counsel. The multidisciplinary approach is important for capturing all
the information about the need to report. Thus the design team may include representatives from the
IRB, legal counsel, and research office and from the compliance, risk management, and accreditation
functions of the health care organization. When state and federal laws respecting reporting are
changed, the reporting matrix should be updated accordingly. To be certain that adverse events do not
go unreported, the matrix should be provided to principal investigators and their key administrative
staff. Protocol approval should be contingent upon meeting the IRB’s policy on adverse event reporting.
Training on the importance of reporting should also be included in the staff education process.
Managing Multiple Reporting Obligations. There is always the chance that one adverse event report may
appear to contradict another report about the same adverse event. What investigators and others
involved in preparing reports must focus on, however, is the information required for meeting the
reporting obligations set by each report recipient. For example, a sponsor’s agreement may oblige the
principal investigator to report the deaths of all study participants. It may, however, go on to specify,
“Please provide a detailed report only if the participant’s death was known to be or believed to be
related to the investigational device.” Therefore, if the investigator found that a medication was
involved in the participant’s demise, the sponsor would not receive a detailed report on that cause. At
the same time, detailed information may be part of a root cause analysis completed by the JCAHOaccredited hospital in which the person was a participant in the research trial. On their face, the reports
might appear contradictory. However, each report is made for a specific purpose, and in each instance
the information supplied meets the reporting requirements.
Difficulties do arise when those required to receive essentially the same information end up with
contradictory reports. This is not acceptable, and all reports should be reviewed carefully to be certain
that the information is accurate and complete for the recipient’s purpose. Inconsistencies should trigger
a further review of the draft reports.
Federal regulations set out a process for addressing adverse events affecting human research
participants. The Common Rule (for example, 45 C.F.R. 46, 2003) is used by most federal agencies that
fund or oversee human research. The FDA has a separate set of regulations (21 C.F.R. 312, 2003) that
are similar but not identical to the common rule for investigating the efficacy and safety of new drugs
and devices. For example, under the Common Rule, an IRB must have a written process in place for
prompt reporting by an investigator of unanticipated problems involving risks to research participants.
Under FDA investigational new drug (IND) requirements the sponsor of the IND has to notify the FDA
and participating investigators, in a written safety report, of any adverse experience associated with the
use of the drug that is both serious and unexpected. The FDA also may request that the sponsor
telephone or fax reports of unexpected fatal or life-threatening experiences associated with the use of
the drug as soon as possible. The requirement must be met no later than seven calendar days after the
day that the sponsor received the information. This means that the IRB and clinical investigators must be
cognizant of the different reporting triggers for research conducted under HHS regulations and that
conducted under the FDA’s IND protocols.
Under the Conditions of Participation in Medicare and Medicaid for hospitals, a patient is entitled to file
a grievance under the patient’s rights standards. This will trigger a detailed probe and report. It may be
quite a challenge to sort out the conventional treatment-related events from those that took place
under the aegis of the clinical research protocol in which the individual is enrolled. This multiplicity of
investigations and fact finding, all done in the name of creating a safe environment, could prove
expensive and time consuming and may generate contradictory results owing to the differences in the
regulations. Final analysis may generate a confirmation that the patient has a genuine point of
grievance, but on the research side there may be no basis for considering the occurrence an adverse
event.
From the point of view of accreditors, an event may constitute an unanticipated outcome, triggering yet
a different type of reporting to the patient, as described in the patient safety standards published by the
Joint Commission on Accreditation of Healthcare Organizations. This is an important consideration when
trying to put in place the infrastructure for a patient-subject safety-oriented protocol.
A health care provider is constantly challenged to determine when an adverse event has actually
occurred and to make timely reports to the appropriate entities. Failure to fulfill reporting obligations
can be embarrassing and very costly in many ways.
Disclosing Information to Research Participants. Study participants should be told when it is determined
that they have experienced a research-related adverse event. This report is part of the ongoing
communication that is so important in restoring public confidence in the safety of human research
programs. That an adverse event has occurred does not mean, however, that there must be an
admission of liability or culpability. In fact the disclosure may be made before an investigation
determines the cause or causes of the adverse event. What participants want to know is quite
understandable:
What does this mean to me?
Will I recover?
How long will my recovery take?
Does this episode mean I am out of the study?
Who pays for my continued care?
I want to complain about what happened. Whom do I contact?
A related issue is what should be disclosed to the participant’s family or the legally authorized
representative. If the participant has authorized the involvement of family members or surrogates, there
should be no difficulty in disclosing information about the adverse event. However, if a participant has
specifically asked that information not be revealed to family members or a legally authorized
representative, that request should be respected. Some exceptions may be permitted under the terms
of state law or the HIPAA regulations dealing with health information.
What may be more problematic for some investigators is revealing to unaffected research participants
that other participants have experienced adverse events. Some argue that by disclosing this information,
investigators could bias the study by planting the suggestion that the currently unaffected participant
might experience similar adverse events. Others argue that it is important to be open and honest with
participants and to let them know when adverse events have become a documented part of the study.
(Some argue that this is required as part of the ongoing informed consent process.) In between these
viewpoints is a third perspective. This middle stance looks to the facts and circumstances of the adverse
event. If it is one that is likely to be experienced throughout the study population, it is information that
participants should receive so that they can pinpoint the event if they experience it and take appropriate
action. However, if it is a rare event, the IRB may not require the investigator to disclose this finding if
doing so may jeopardize the integrity of the study or prevent the research from going forward or if the
risk of harm to other participants is minimal. At the very least, once the adverse event is known to occur
the IRB and sponsor should be notified so that an ethical and legal decision can be made about
disseminating information to the study population.
Giving Participants Directions for Adverse Event Reporting. Research participants need clear,
understandable information about what is considered an adverse event. Some reactions may be
anticipated whereas others may not. To be certain that participants understand what an adverse event
is and what they should do when such an event occurs, it is prudent for research coordinators to do the
following:
Develop standard information to be provided to all research participants.
Test the comprehensibility of this information by using it with a sample of individuals who may become
research participants and those who may act as their legally authorized representatives.
Adjust the language to meet the reading level of the potential study population.
Develop in-service training for investigators and their staff that gives them practical ways to teach
participants to identify and report adverse events.
Obtain input from and the approval of the IRB for the standard information format.
Educating Participants About Emergency Treatment for Adverse Events. Because many research studies
are based in hospitals or near such facilities, participants are often encouraged to seek assistance at the
hospital emergency department. Some research consent documents or instruction sheets state, “Should
you experience a problem and you are unable to contact us, please proceed to our [or a named]
emergency department and they will be able to assist you.” This type of language often alarms health
care attorneys and risk managers as it creates the impression that there is a legal relationship between
the study and the hospital emergency department. Moreover, such instructions, implying that the
emergency department is informed about the study and its participants, are seen as creating a false
sense of security. The emergency department personnel may not have any idea that the individual is
enrolled in a study. Even when that information is known, they may not know what treatment the study
involves and if in fact it relates to the presenting ailment. A related concern is the notation on some
instruction sheets that directs participants to “call or e-mail our office night or day if you have an
adverse event or problem.” Unless the phone line or e-mail system is monitored around the clock, a
research participant could wait hours for a response. What might have been a manageable adverse
event at the outset may become impossible to control over that time. In addition, giving participants the
expectation that they have access to a constantly monitored phone line or e-mail system when they do
not could constitute a liability risk exposure.
Researchers can explore several practical options for avoiding needless delay in responding to an
adverse event. For example:
Provide a phone rollover service; calls that are not picked up after five or six rings are forwarded to a
staffed call center.
Make certain the call center can always reach a research staff member who can respond to the
participant within a defined time period.
Make certain the pager system is staffed twenty-four hours a day.
Have the e-mail message system linked to a pager, cell phone, or personal digital assistant (PDA) to alert
the on-call research staff member to respond to the participant.
Perform quality audits to be certain that voice and e-mail messages are answered within a designated
time period.
If a decision is made to encourage participants to seek emergency services at a designated facility or to
telephone an on-call medical group, it is important that the facility or group have sufficient background
information on the study to plan appropriate treatment. As was suggested earlier, this is why there is
merit in providing participants with a MedicAlert-style bracelet or necklace. With pertinent information
such as the study number, participant identifier, and a staffed toll-free number to call or Internet Web
address to contact, the MedicAlert-style notification gives caregivers a way to access important details
needed in treating a participant who presents with an adverse event.
Using Data and Safety Monitoring Boards
When constructing a research protocol, an investigator must ensure that ongoing evaluation of study
data occurs as these data are accumulated. In an FDA-regulated study, data monitoring is also the
responsibility of the sponsor. An IRB must ensure that this critical element is part of a research protocol.
Using an independent group, often called a data and safety monitoring board (DSMB), to perform this
function is an excellent way to achieve objective analysis of study information, says the Office for
Human Research Protections (HHS, OHRP, 2002). (DSMBs are occasionally referred to as data monitoring
committees [DMCs], particularly in FDA publications.)
A DSMB can analyze interim data, often from multiple sites, as well as external data from other studies
and reports. A DSMB should be evaluating the efficacy of the therapy under study, its relative merit
compared to other treatments (including no treatment), and any adverse events causally related to the
therapy. Often the DSMB is the first to be in a position to make a sound judgment that a study should be
terminated because the investigational therapy is clearly superior or inferior to the control therapy or
patient safety is clearly in jeopardy. More often, a DSMB will recommend that a study continue, as a
conclusion cannot yet be supported statistically. A DSMB may also recommend alterations to the
protocol, the populations enrolled, or the information that should be supplied to participants.
The OIG believes that an IRB alone cannot provide a sufficient degree of protection to research
participants and that DSMBs can play a vital role when trials pose significant risks to patients (HHS, OIG,
2000, p. 4). The OIG has recommended that DSMBs be required for certain high-risk and multisite trials,
believing that “DSMBs are independent assessment bodies that provide medical, scientific and other
expertise that is not typically available on IRBs, thereby serving an invaluable function in protecting
human subjects” (HHS, OIG, 2000, p. 12). However, an IRB does not absolve itself of responsibility for
protecting research participants when it employs a DSMB. A DSMB is merely a useful tool for an IRB to
use in fulfilling its responsibility. A DSMB should provide summary assessments of a study’s adverse
event reports to the overseeing IRB(s). “IRBs are swamped with individual adverse event reports from
multi-site trials, but these reports lack the essential context to confer meaning about the relative safety
of the trial. DSMBs can provide this context and thereby enhance the IRB’s capacity to assess the
ongoing safety of a trial” (HHS, OIG, 2000, p. 12). In addition, a DSMB decides how missing or suspect
data are to be handled in overall analyses of a study.
Indications for Use
DSMBs have gained favor in large, randomized, multisite studies that evaluate interventions intended to
prolong life or reduce the risk of a major adverse health outcome such as a cardiac event or recurrence
of cancer (FDA, 2001). They are generally not practical in short-term studies where a DSMB would not
have the opportunity to analyze the data in a timely fashion. The FDA has suggested that the following
criteria should be considered when determining whether to employ a DSMB:
Risk to study participants
Practicality of DSMB review
Assurance of scientific validity
A study sponsor, especially in sponsor-designed pharmaceutical and medical device studies, often
considers whether use of a DSMB is indicated. A DSMB can help a sponsor meet its obligation to monitor
studies evaluating new drugs, biologics, and devices and report adverse events to the FDA (21 C.F.R. §§
312.50 and 312.56, 2003, for drugs and biologics; 21 C.F.R. §§ 812.40 and 812.46, 2003, for devices).
Many sponsors now routinely use a DSMB as part of clinical trials. Prudence would counsel use of a
DSMB in any study where the ability to monitor data is beyond the resources of the local IRB. However,
FDA regulations do not require the use of a DSMB, except for studies in emergency settings (under 21
C.F.R. § 50.24[a][7][iv], 2003) where informed consent may be waived.
An IRB should determine at the time of its initial assessment whether a DSMB is required. What is the
complexity of the study? Is mortality or major morbidity a primary or secondary end point? What is the
size and duration of the study? What is the degree of risk? Are the subjects members of a fragile
population, such as children or the elderly? A DSMB is intended to review unblended interim data. An
IRB can rely on the recommendations of a DSMB. A DSMB usually does not monitor a single test site for
protocol compliance or appropriate data entry.
The NIH (1998) gives excellent guidance on the use of DSMBs in NIH-funded or NIH-conducted trials.
Phase 3 trials generally are required to have DSMBs, and most DSMB engagements involve site
monitoring as part of the quality assurance program of the funding institute or center. The NIH (2000)
also recommends the use of a DSMB or similar arrangement for certain Phase 1 and Phase 2 studies,
especially when large numbers of participants are enrolled or multiple sites are used. DSMBs associated
with NIH trials are expected to forward summary reports of adverse events to IRBs (NIH, 1999).
Composition
Unfortunately, federal regulations and policy statements give little guidance about the quantity of
individuals who should serve on a DSMB or about those individuals’ attributes. The FDA (2001) has
noted that a DSMB may have as few as three members, but it encourages a broader group than this.
Ideally, some members will have scientific backgrounds useful in evaluating the data being collected
from an investigation. A biostatistician needs to undertake the analysis of the data. Whether the
biostatistician should be a member of the DSMB is debated. Specialists in pharmacokinetics, toxicology,
epidemiology, and the medical specialty of the study may be needed. The FDA (2001) has also suggested
that adequate representation of gender, ethnicity, and the disease process under study is important.
Equally important are the independence and the influence of the DSMB. Thus the members must have
standing in the scientific community sufficient to make them effective in persuading an investigator, an
IRB, and the sponsor to modify or terminate a study when that course is indicated by data analysis.
Some sponsors create DSMBs from their own upper scientific management teams. Such a DSMB has the
benefits of being organized easily, efficient in gathering the members for a meeting, and usually
persuasive in its recommendations. However, it may suffer from being pressured explicitly or implicitly
to minimize the significance of adverse events or to accept dubious clinical data as showing the
effectiveness of the investigational drug. The FDA (2001) points out that sponsors may feel pressured to
make a premature disclosure of results because of SEC reporting requirements, fiduciary responsibilities,
and other business considerations and that an independent DSMB protects the sponsor from such poor
decisions. Thus a better approach than the DSMB made up entirely of members of internal scientific
management is a DSMB with a sufficient number of independent members to ensure that data are
analyzed critically and appropriate responses are made. Clearly, a study investigator or participant
should not be a member of the study’s DSMB. Scientists who have outspoken views on the merits of the
therapy under investigation should also not be on the DSMB.
Relationship to an IRB
A DSMB will collect and interpret data from all of a study’s clinical sites. Often each IRB will approve and
give continued approval to only one research site. Therefore the DSMB may be able to spot significant
data earlier by seeing data from more than one location and from a larger population. “IRBs conducting
continuing review of research may rely on a current statement from the DSMB indicating that it has
evaluated study-wide adverse events, interim findings, and any recent literature that may be relevant to
the research, in lieu of requiring that this information be submitted directly to the IRB. Of course, the
IRB still must receive and review reports of local, on-site unanticipated problems involving risks to
participants or others and any other information needed to ensure that its continuing review is
substantive and meaningful” (HHS, OHRP, 2002). The IRB always maintains the responsibility of
protecting patients at its site. Although it may rely on reports from the DSMB, it has the responsibility of
taking whatever actions are necessary to protect the participants.
Activities
The activities of a DSMB should be structured like those of an IRB. Participant safety, public relations,
and scientific validity require that the DSMB operate in a logical and predictable manner. Policies and
procedures should be written and followed. These should include a schedule of meetings and the format
for each meeting, the format for presenting data to the DSMB, the format for reporting interim findings,
and the statistical methods to be used. Minutes of every meeting should be written in detail and filed.
The board’s recommendations should be communicated clearly to the sponsor. A DSMB usually does
not report directly to the FDA unless its doing so is part of the study’s design. However, the sponsor has
an obligation to report the DSMB’s recommendations in a timely manner to the FDA and the responsible
IRB(s). The DSMB’s activities are reviewed as part of the FDA’s consideration of the study data.
The NIH Model for Data and Safety Monitoring
As mentioned previously, the National Institutes of Health require data and safety monitoring, generally
through a DSMB, for Phase 3 clinical trials sponsored by NIH. It also recommends a DSMB for Phase 1
and Phase 2 clinical trials when these studies (1) have multiple clinical sites, (2) are blind (masked), or (3)
employ particularly high-risk interventions or vulnerable populations. Conversely, an independent DSMB
may not be necessary or appropriate when the proposed intervention is low risk. Monitoring by the
investigator or by an independent individual may be sufficient. When a DSMB is employed, an IRB can
rely on the summary reports of adverse events from the DSMB (NIH, 2000).
Conclusion
Research participant safety will remain a core issue for regulators and participants. Steps can be taken
to minimize needless risk exposure for research participants. Communicating with all involved in a study,
documenting activities, training staff, and managing adverse events are important considerations in
developing a culture of safety for clinical trials. Strong investigatory tools, methodologies, and reporting
practices help to develop the necessary infrastructure for a safe research environment for clinical trial
participants. Knowing to whom to report adverse events and knowing when to report are also important
factors in developing a culture of safety in human research.
References
Department of Health and Human Services, Office for Protection from Research Risks. Institutional
Review Board Guidebook. (Chap. III, Sec. E., “Monitoring and Observation.”) Washington, D.C.: U.S.
Government Printing Office, 1993.
Department of Health and Human Services, Office of Inspector General. Institutional Review Boards: A
Time for Reform. OEI-01-97-00193. Rockville, Md.: Department of Health and Human Services, June
1998.
Department of Health and Human Services, Office of Inspector General. Protecting Human Research
Subjects: Status of Recommendations. OEI-01-97-00197. [http://oig.hhs.gov/oei/reports/oei-01-9700197.pdf]. Apr. 2000.
Department of Health and Human Services, Office for Human Research Protections. Guidance on
Continuing Review. [http://www.hhs.gov/ohrp/humansubjects/guidance/contrev2002.htm]. July 11,
2002.
Food and Drug Administration. Guidance for Clinical Trial Sponsors on the Establishment and Operation
of Clinical Trial Data Monitoring Committees. Draft Guidance. Rockville, Md.: Food and Drug
Administration, Nov. 2001.
National Institutes of Health. Policy for Data and Safety Monitoring.
[http://www.nih.gov/grants/guide/notice-files/not98-084.html]. June 10, 1998.
National Institutes of Health. Guidance on Reporting Adverse Events to Institutional Review Boards for
NIH-Supported Multicenter Clinical Trials. [http://grants.nih.gov/grants/guide/notice-files/not99107.html]. June 11, 1999.
National Institutes of Health. Further Guidance on Data and Safety Monitoring for Phase I and Phase II
Trials. OD-00-038. [http://grants1.nih.gov/grants/guide/notice-files/not98-084.html]. June 5, 2000.
REVIEW ARTICLE
The Relationship Between Patient Safety Culture and Patient
Outcomes: A Systematic Review
Margaret Hardt DiCuccio, RN, MSN
Context: In the past 13 years since the Institute of Medicine report,
To Err is Human, was published, considerable attention was placed on
the relationship between patient safety culture and patient outcomes.
Research to understand this relationship has been conducted; however,
now, it is important to systematically review these studies to determine
if there are tools, levels of measure and outcomes that have been shown
to result in significant correlations.
Objective: The purpose of this review is to evaluate the state of research connecting patient safety culture and patient outcomes to determine nurse-sensitive patient outcomes that have been significantly
correlated to culture of safety and commonly used tools to measure culture of safety in the studies with significant correlations.
Data Sources: Published English only research articles were considered for the review. Only studies that directly measured patient outcomes
in relationship to patient safety culture in hospitals involving registered
nurses as a participant were included.
Results: Evidence of relationships between patient safety culture and
patient outcomes exist at the hospital and nursing unit level of analysis;
however, the number of studies finding statistically significant correlations particularly using nurse-sensitive outcomes is limited.
Conclusions: The findings from this review suggest that there are
emerging trends indicating that the specific patient safety culture measurement tools, the level of analysis, and selection of outcome measures are important considerations in study design. More research is
needed to determine interventions that improve patient safety culture
and outcomes.
Key Words: safety culture, safety climate, patient outcomes
(J Patient Saf 2015;11: 135Y142)
t has been over a decade since To Err Is Human1 was published by the Institute of Medicine (IOM). This groundbreaking report emphasized the responsibility of health-care
providers to examine flawed systems within their organizations
with the aim to improve the clinical outcomes of patients. Also
included was the need to develop a culture that encourages all
staff members to raise concerns regarding practices that place
patients at risk, or said a different way, to engage in advocacy
activities to keep patients safe. To promote staff engagement
in patient advocacy, there is a need to improve psychological
safety. Psychological safety is defined as a staff member’s comfort level to challenge someone more powerful and know that
there will be no retribution.2 The impetus for improving hospital
systems and psychological safety is the unnecessary patient
deaths resulting from preventable errors. Nurses have patient
advocacy as one of their core responsibilities3 yet all too often
I
From the University of Pittsburgh Medical Center Mercy, Pittsburgh,
Pennsylvania.
Correspondence: Margaret Hardt DiCuccio, RN, MSN, University of
Pittsburgh Medical Center Mercy, 1400 Locust St. Pittsburgh PA 15219
(email: dicucciome@upmc.edu).
The author discloses no conflict of interest.
Copyright * 2014 Wolters Kluwer Health, Inc. All rights reserved.
J Patient Saf
& Volume 11, Number 3, September 2015
they do not feel safe and culturally supported to speak up when
a patient is at risk.2
Since the time of the original IOM report, there has been
significant attention given to the following activities:
& defining the terms patient safety culture and climate,
& developing tools to measure these concepts, and
& conducting research to establish the relationship between
safety culture/climate and patient outcomes.
The purpose of this review is to evaluate the state of research connecting patient safety culture and nurse-sensitive
patient outcomes. The review includes study designs, measurement tools, and an examination of outcomes that did and
did not have significant correlations to patient safety culture.
Gaps in knowledge and next steps for research on this topic
are noted.
STATE OF RESEARCH
The inclusion criteria for selection of the research articles
will be outlined as well as search strategies used to find the
data sources. The measurement of patient safety culture and patient outcomes in the studies will be described. Also, the current state of research outlining the relationship between patient
safety culture and patient outcomes is examined.
Inclusion Criteria
In this review, patient safety culture is defined as ‘‘the
values shared among organization members about what is
important, their beliefs about how things operate in the organization, and the interaction of these with work unit and organizational structures and systems, which together produce behavioral
norms in the organization that promote safety’’4 (p.400).
Colla et al5 defines patient safety climate as the measureable
components of patient safety culture. Therefore, for the purpose
of this review, the term patient safety culture will refer to both
culture and climate as it is all encompassing.
Studies measuring patient/family satisfaction or direct patient outcome measures (falls, hospital-acquired conditions, readmission rates, hospital compliance to best practice guidelines,
medication errors, and mortality) were included in this review.
Studies using health-care professional’s perceptions of patient
safety outcomes were not included because of the indirect nature
of these measures.
The electronic databases used to locate the research articles were EBSCO host for Hospitals and Medical Institutions,
OVID, and ProQuest. These hosts include multiple data sources
such as CINAHL, MEDLINE, psychology, sociology, healthcare leadership databases, and dissertation abstracts. A manual
search of references from the selected studies was also conducted.
The Boolean search mode was used to ensure maximal capture. Concepts searched included safety climate, safety culture,
safety environment, patient outcomes, nurse sensitive outcomes,
treatment outcomes, and outcomes research.
A total of 17 research studies were identified that connected the concept of patient safety culture to nurse-sensitive
patient outcomes as defined previously. Many of these studies
www.journalpatientsafety.com
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
135
136
www.journalpatientsafety.com
Sample: 73 hospital
submitting data to the
HCAPS and Hospital
SOPS comparative
data bases in 2008.
Sorra et al13
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Sampling: 4394 staff members
from a convenience sample of
30 ICUs. 47.9% response rate.
Sample: 36,375 employees
within 67 hospitals.
A response rate of 38.5%.
Level of analysis: hospital
Design: cross-sectional
Level of analysis: hospital
Design: cross-sectional
Safety culture (SAQ-ICU version).
ICU LOS and patient mortality
(multisite clinical database).
Hospital Safety Climate
(PSCHO). Readmission
rates for heart failure,
myocardial infarction
and pneumonia (abstracted
from Medicare data).
Learning climate (Error
Orientation Scale)
The selection criterion
was mortality %of AMI
patients within the first 30 days
post event in CMS database.
Medication errors
(incident reporting data)
Safety culture
(AHRQ HSOPSC)
and patient satisfaction
HCAPS.
Safety culture
(AHRQ HSOPSC)
Family satisfaction
with ICU (tool developed
for a previous study)
Variables and
(Measurement Tools)
Study Findings
A negative correlation between
perception of management and
patient mortality (P = 0.02).
A negative correlation between
safety climate and LOS (P = 0.03).
A negative correlation between
hospital safety climate and
readmission rate for heart
failure (P e 0.05). Front line
staff members’ perception of
patient safety climate are more
highly correlated to readmission
rates than senior leaders (P e 0.01)
Six domains were identified
post theme analysis. Three were
related to patient safety culture,
problem solving and learning,
communication at transitions, and
organizational values and goals.
Negative correlation between
medication errors and perceived
learning climate (P G 0.01).
A correlation between %RNs
on unit and less medication
errors when learning climate
is poor (P G 0.05).
Positive correlation between
subscales of HSOPSC
and nurse driven and composite
HCAPS measures.
Positive relationship between
safety culture and family
satisfaction of nonsurvivor
patients who were in the ICU
for Qto 14 days (P e 0.01).
J Patient Saf
Peer-reviewed article
Huang et al8
Peer-reviewed article
Hanson et al16
Design: qualitative,
descriptive
Curry et al22
Sample: 11 hospitals that
ranked in either the top 5% or
bottom 5% of performance
for MI mortality rates.
Level of analysis: nursing unit
Level of analysis: hospital
Design: Cross-sectional
descriptive
Level of analysis:
nursing unit
Design: cross-sectional
Design: cross-sectional survey
Design and Level
of Analysis
Peer-reviewed article
Sample: 4954 RNs
from medical-surgical
units from 146 hospitals.
Response rate of 75%.
Setting: 23 ICUs in Canada.
Peer reviewed article
Peer-reviewed article
Chang and Mark17
Sample: 2374 ICU staff members.
1381 family members
of ICU patients. 54% and
64% response rates.
Sampling, Response
Rate and Setting
Dodek et al10
Reference Newer
Studies Reported First
TABLE 1. Research Summary Including Patient Safety and Patient Outcome Studies by Date of the Study
DiCuccio
& Volume 11, Number 3, September 2015
* 2014 Wolters Kluwer Health, Inc. All rights reserved.
* 2014 Wolters Kluwer Health, Inc. All rights reserved.
Sampling: 6697 health-care
staff members from a convenience
sample of 59 units in
10 community hospitals.
Obrien6
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Mark et al21
Peer-reviewed article
Sampling: random sample
of 278 nursing units in
143 hospitals. 4911 RNs
(response rate 75%
and 2720 patients.
Design: longitudinal cross-sectional
Level of analysis: nursing unit
Design: Cross-sectional,
descriptive, correlational
Level of analysis: nursing unit
Design: cross-sectional,
descriptive, model testing
Level of analysis:
hospital and unit
Design: cross-sectional
Level of analysis: hospital
Design: descriptive,
multi-level cross-sectional
Level of analysis: nursing unit.
Level of analysis: hospital
Design: cross-sectional
Level of analysis: hospital
Level of analysis: nursing unit.
Design: cross-sectional
descriptive
Hospital safety culture
(AHRQ HSOPSC).
Patient outcomes
(State level data
base reported at the
hospital level)
Hospital safety culture
(AHRQ HSOPSC).
Patient outcomes, CAUTI,
CLABSI, SSI, HAPU, falls
and failure to rescue (hospital
data collection systems).
Culture of safety (NDNQI
RN survey subscales
classified into organizational
support (OS) and work unit
support (WS). Patient
outcomes (PSI rates, HAPU,
failure to rescue, HAI,
VTE rates)
Patient safety culture (SAQ),
fall and HAPU rates (NDNQI
database) Hospital failure rate
(CMS sponsored data
collection-including
community-acquired
pneumonia CAP)
Patient safety
culture (HSOPSC)
Patient experience
(Consumers Assessment
of Healthcare Providers
and Systems-Hospital
version HCAPHS)
Safety climate (Error
Orientation Scale and
Zohar Safety Climate
Scale) Organizational
effectiveness (medication
error rates and falls)
Hospital safety culture
(AHRQ HSOPSC). Patient
safety (8 measures from the
AHRQ’s PSIs in total).
www.journalpatientsafety.com
(Continued on next page)
A positive correlation was found
between medication error rate
and safety climate with the
interaction effect higher %RNs
with BSN and % RNs (P =0.01).
A positive correlation between
%RNs and RNs with BSNs and
falls at high levels of safety climate.
No significant relationship
between patient safety climate
and falls or HAPUs. A negative
relationship was noted between
staff perception of support of
manager and failure rate
for the CAP performance measure.
Positive correlations were found
on several subscales of the
HSOPSC with 5/6 measures on
the HCAPHS (P G 0.001).
An unexpected positive
correlation was noted between
Organizational support (OS)
and PSI (P = 0.03).
No significant relationship
between patient safety culture
and patient outcomes.
Safety grade and positive safety
score was negatively correlated
to mortality (P G 0.01).
The HSOPSC composite score
was negatively correlated with
composite PSI scores (P G 0.001).
& Volume 11, Number 3, September 2015
Dissertation
Gearhart11
Dissertation
Sampling: 287 nursing
staff and 216 patients on
3 hospital units in 3
San Francisco Bay hospitals.
Sampling: a convenience sample
of 97 hospitals that participated in
the NDNQI RN survey in 2005.
Kemper20
Dissertation
Dissertation
Sampling: convenience sample
of 34 unit directors and their
711 staff members in a large academic
medical center. Response rate was 90%
Sampling: 21,730 nurses,
1,010,298 patients (mortality, LOS),
3,473,127 patients (HAPU, post op
PE/VTE) from 688 hospitals
Sampling: 56,480 staff members
from a convenience sample
of 179 hospitals in the AHRQ’s database.
Thompson14
Peer-reviewed article
Olds23
Dissertation
Peer-reviewed article
Mardon et al12
J Patient Saf
Patient Safety Culture and Patient Outcomes
137
138
www.journalpatientsafety.com
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
17
Sampling: a convenience sample
of 118 ICUs in the United Kingdom
enrolled in a prior study that collected
APACHE II data. 45% met
inclusion criteria (18,089
ICU patients). 5540 healthcare
professionals at 68% participation.
Sampling: 42 randomly
selected hospitals. Use of 81
nursing units and 1127 nurses
within the hospitals.
Sampling: a convenience
sample of nurses working on
29 units (with 960% response
rate to the SAQ in one large
academic medical center and
28,260 discharged patients’ data.
Sampling: Convenience
sample of 42 hospitals that
participated in both the AHRQ’s
data base in 2002 and the
PSCHO survey in 2004.
Sampling, Response
Rate and Setting
Level of analysis: nursing unit
Level of analysis: nursing unit
Design: cross-sectional
Level of analysis: hospital
Design: cross-sectional
Design: cross-sectional
Design: Cross-sectional
Level of analysis: nursing unit
Design and Level
of Analysis
Risk-adjusted mortality
(APACHE II data base) Staff
perception of safety climate
(SAQ revised by researcher
to be ICU specific).
Perception of safety climate
(Zohar’s measure of safety
climate and The Error
Orientation Scale) Medication
errors and UTIs (hospital data
base) Patient satisfaction and
perception of responsiveness
(researcher developed tool)
Organizational Culture (SAQ)
Patient outcomes, falls and
medication errors (occurrence
reporting system), PE/DVT
and HAPU (hospital
discharge data)
Hospital safety culture
(PSCHO) Patient outcomes
(14 PSIs from AHRQ data
base combined into 3 groups,
postoperative complications,
nurse sensitive, technical
difficulty with procedures.)
Variables and
(Measurement Tools)
Two subscales of the SAQ,
safety climate (P e 0.005)
and perception of management
(P e 0.006) were negatively
correlated to risk-adjusted
ICU mortality. The same
findings were noted in the
RN only analysis of data.
Safety climate was negatively
correlated to medication errors
and UTIs (P G 0.05) and positively
correlated to patient satisfaction
and perception of nurse
responsiveness (P G 0.01).
One subscale of the SAQ,
increasing stress recognition
was positively correlated to patient
falls (P = 0.000). Safety climate
subscale was negatively correlated
to HAPU (P = 0.000).
Fear of blame was positively
correlated to performance on all
PSI’s, postoperative complications
(P G 0.01) and nurse sensitive
outcomes (P G 0.05). Fear of
shame positively correlated to
technical difficulty (P G 0.05).
Study Findings
J Patient Saf
Dissertation
Total
Peer-reviewed article
Sexton9
Peer-reviewed article
Hofmann and Mark18
Singer et al4
Taylor7
Dissertation
Reference Newer
Studies Reported First
TABLE 1. (Continued)
DiCuccio
& Volume 11, Number 3, September 2015
* 2014 Wolters Kluwer Health, Inc. All rights reserved.
J Patient Saf
& Volume 11, Number 3, September 2015
Patient Safety Culture and Patient Outcomes
considered both nurse and patient outcomes. Nurse outcomes
such as turnover, injury rates, and RN satisfaction are not discussed in this review as the focus is on patient outcomes only.
This represents 10 peer-reviewed articles4,8,10,12,13,16,17,18,21,22
and 7 dissertations.6,7,9,11,14,20,23 A summary of these studies
is provided in Table 1.
Study Design
The majority of the studies (16) used a cross-sectional descriptive design, with 1 study22 using a qualitative design. The
cross-sectional design used in these studies often involves secondary analysis of previously collected data at a specific point
in time, when the culture of safety tool was administered, and
then linking these results to various patient outcome measures
collected from the participating health-care facilities. Several
of the studies used large convenience databases made available
by a government source (state and federal databases) or by an
organization (Agency for Healthcare Research and Quality
(AHRQ) database or hospital system) and involved large samples
sizes. This design allows the researcher to interpret extensive
datasets with the use of regression analysis.
The authors of the qualitative study selected 11 hospitals
in the United States that either ranked in the top 5% or bottom
5% in performance for acute myocardial infarction (AMI) mortality rates. After participant interviews and a theme analysis, it
was found that the organizations with lower mortality emphasized problem solving and learning, communication at transitions, and organizational values and goals that related to a
positive patient safety culture as compared with those organizations with higher mortality rates. There was no common protocol regarding the care of the AMI patient indicating that the
positive outcome went beyond protocols and into the culture
of the organization. The results lend credence to the effect of
patient safety culture on patient outcomes, in this case, patient
mortality, and the importance of senior leadership engagement
to improve the culture.
Measurement of Patient Safety Culture
Patient safety culture was measured using 8 different tools.
The 2 most frequently used scales were the Safety Attitudes
Questionnaire (SAQ) Hospital6,7 Intensive Care Unit8,9 and the
AHRQ Hospital Survey on Patient Safety Culture (HSOPSC).10Y14,23
The SAQ has 63 items divided into 6 subscales with a Cronbach
alpha between 0.68 and 0.81. 9 The HSOPSC has 42 items,
12 subscales with Cronbach alpha between 0.62 and 0.85.15
Both of these tools are well designed and have large comparative databases for hospital data. The AHRQ tool is nonproprietary and, therefore in most cases, more economical to
administer. The other 6 measurement tools also reported acceptable reliability ratings, however, are less widely utilized and do
TABLE 2. Summary of Patient Outcomes
Patient Outcome
Study Findings
Source
Significant
Family satisfaction
Patient satisfaction
Medication errors
Mortality
Readmission
PSI composite*
PSI nurse† sensitive
Failure rate AMI/HF
Failure rate CAP
Dodek et al10
Gearhart11
Hofmann and Mark18
Sorra et al13
Chang and Mark17
Mark et al21
Taylor7
Hofmann and Mark18
Huang et al8
Sexton9
Olds23
Hanson et al16
Mardon et al12
Singer et al4
Thompson14
Kemper20‡
Obrien6§
Mark et al21‡
Taylor7
Hofmann and Mark18
Olds23
Obrien6§
Obrien6§
Level of Analysis
Nonsignificant
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Nursing unit ICU
Nursing unit med/surg
Nursing unit med/surg
Hospital
Nursing unit med/surg
Nursing unit med/surg
Nursing unit mixed
Nursing unit med/surg
Nursing unit ICU
Nursing unit ICU
Hospital
Hospital
Hospital
Hospital
Nursing unit mixed
Hospital
Nursing unit mixed
Nursing unit med/surg
Nursing unit mixed
Nursing unit med/surg
Hospital
Hospital
Hospital
X
X
X
X
X
X
X
X
X
*Score represents measures both related to nursing care and others.
†
Includes indicators such as falls, HAPU, PE/DVT, and HAI.
‡
Study yielded results that, although significant, were not in the expected direction.
§
Study considered both hospital and unit level analysis.
* 2014 Wolters Kluwer Health, Inc. All rights reserved.
www.journalpatientsafety.com
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
139
J Patient Saf
DiCuccio
not have extensive nationwide comparative databases. These
findings are consistent with a previous comprehensive review
of patient safety culture surveys conducted by Colla et al.5
Patient Outcomes Correlating to Culture
of Safety
The choice of patient outcomes for the most part was
driven by the level of analysis, hospital or nursing unit, and
the type of nursing units included in the study. A summary of
patient outcomes and significance of findings is available in
Table 2.
If the analysis is at the hospital level, then more global
measures such as composite score for AHRQ patient safety
indicators (PSI), mortality, and readmission rates have been
found to yield statistically significant results in the studies.4,6,12,16 In an additional study, measuring outcomes at the
overall hospital level of analysis patient safety culture and patient experience were significantly correlated.13
When the analysis is at the nursing unit level, those patient outcomes that are predominately nurse driven, such as
hospital-acquired pressure ulcers (HAPUs), family satisfaction, and patient satisfaction, have been seen to yield statistically significant results.7,10,11 When studies are conducted in
the intensive care setting, the relationship between patient safety
culture and patient mortality has also been a statistically significant finding.8,9 Previous research has been found that improved teamwork and communication among members of
the care team has significantly correlated with decreased ICU
patient mortality.19 A summary diagram linking tool selection
& Volume 11, Number 3, September 2015
with level of analysis and significant results is presented in
Figure 1.
Studies With Nonsignificant or
Unexpected Results
Much can be learned from studies that found either nonsignificant or unexpected results. It is suspected that additional studies have been conducted that fall in this category,
but the researchers may not have sought publication. In total,
5 studies were noted to fall in this category. Table 3 is a summary of the limitations of the studies that most likely contributed to the results.
There were 2 studies that reported unexpected significant
results. The first reported that at the hospital level, the PSI nurse
indicators (falls, HAPU, and infection rates) increased as patient safety culture improved.20 This finding is most likely the
result of the tool used to measure patient safety culture, the
National Database of Nursing Quality Indicators Registered
Nurse survey (NDNQI RN). The second reported that a more
positive patient safety culture was related to increases in medication errors.21 This finding could be related to willingness to
report errors if the culture is supportive of patient advocacy.
ANALYSIS AND NEXT STEPS
Overall, the 17 studies conducted examining the relationship between patient safety culture and patient outcomes were
well designed, used instruments with adequate psychometric
properties, and had large sample sizes. Many of the studies also
examined nurse outcome variables with significant findings.
FIGURE 1. Analysis of significant patient outcome results in expected direction.
140
www.journalpatientsafety.com
* 2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
J Patient Saf
& Volume 11, Number 3, September 2015
Patient Safety Culture and Patient Outcomes
TABLE 3. Nonsignificant/Unexpected Results Relating Patient Safety Culture to Outcomes
Source
Study Findings Nonsignificant/Unexpected
Level of Analysis
6
AMI/HF at the hospital level non-significant
Post hoc power analysis indicated
insufficient number of hospitals
Data were nurse reported with few
events reported leading to a heavily
skewed distribution
Low reported number of adverse events
per unit leading to a negatively skewed
distribution of patient outcome variables
The measurement tool chosen for culture
of safety was an RN satisfaction survey
with no demonstrated validity
to measure culture of safety.
Positive patient safety culture was found to
increase medication errors potentially because
of the perception of psychological safety.
The AHRQ PSI data are abstracted
from closed medical records.
The methodology removes reporting
bias however results in small numbers
of events and skewed data distribution.
Nursing unit hospital
Obrien
HAPU/Falls at unit level non-significant
Thompson14
PSI Nurse sensitive nonsignificant
Kemper20
PSI nurse sensitive/unexpected result
Mark et al21
Medication errors/unexpected result
Olds23
Falls/HAPU (AHRQ-PSI) nonsignificant
The patient outcomes that are least consistently reported to
be significant are those considered nurse sensitive such as
medication errors, HAPU, falls, and infections. Of the 7 studies
reporting nurse-sensitive outcomes, 2 had findings that were
opposite the hypothesis,20,21 and 3 had nonsignificant findings.6,14,23 Evidence from these studies suggests that the number of adverse events is so small that variation in the dataset is
inadequate to detect a significant correlation. In addition, the
use of medication errors as an outcome variable has the confounding effect of psychological safety and therefore has not been
shown to be consistently effective.
If the researcher is studying patient safety culture at the
hospital level, readmission rates, AHRQ composite rates, mortality, and patient satisfaction were significantly correlated.
When studying patient safety culture at the ICU level, mortality
and family satisfaction had significant correlations. Finally,
if the med/surg unit or mixed units is the level of analysis,
then patient satisfaction and HAPUs have been significantly
correlated.
There are trends emerging related to connections between
patient safety culture and specific patient outcomes. This information could guide researchers in study construction or administrators in validating the importance of a positive patient
safety culture. The results that yielded a significant relationship between patient safety culture and patient outcomes are
outlined in Table 4.
The AHRQ HSOPSC and the SAQ are the 2 dominant
tools used in these studies to measure patient safety culture.
Given the credible psychometric characteristics and nationwide
data bases associated with each tool, it guides the researcher
TABLE 4. Summary of Significant Study Outcomes: Relationship Between Safety Culture and Patient Outcomes
Culture Tool
AHRQ
HSOPSC
Error Orientation Scale
PSCHO
NDNQI RN Survey
SAQ ICU
SAQ Hospital
Zohar Safety Climate Scale
Source
Patient Outcome
Significant Studies
Level of Measure
Dodek et al10
Sorra et al13
Mardon et al12
Gearhart11
Olds23
Chang and Mark17
Hanson et al16
Singer et al4
Kemper20
Huang et al8
Sexton9
Obrien6
Taylor7
Mark et al21
Hofmann and Mark18
Family satisfaction
Patient experience
AHRQ PSIs (composite)
Patient experience
Mortality
Medication errors
Readmission
AHRQ PSIs (composite)
PSI nurse indicators
Patient mortality
Patient mortality
Community-acquired pneumonia
HAPU
Medication errors
Medication errors and UTI
Patient satisfaction
Positive correlation
Positive correlation
Negative correlation
Positive correlation
Negative correlation
Negative correlation
Negative correlation
Positive correlation
Unexpected positive
Negative correlation
Negative correlation
Negative correlation
Negative correlation
Unexpected positive
Negative correlation
Positive correlation
Nursing unit ICU
Hospital
Hospital
Nursing unit med/surg
Hospital
Nursing unit med/surg
Hospital
Hospital
Hospital
Nursing unit ICU
Nursing unit ICU
Hospital
Nursing unit mixed
Nursing unit
Med/surg
Nursing unit med/surg
* 2014 Wolters Kluwer Health, Inc. All rights reserved.
www.journalpatientsafety.com
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
141
J Patient Saf
DiCuccio
in the direction of one of these tools versus the others that were
used in the reviewed studies.
Now that these associations have been demonstrated the
following next steps are recommended:
& Continue to refine the research connecting patient safety culture and patient outcomes both in conducting research using
the current design (cross sectional) and through meta-analysis
of the available studies to strengthen the connection between
specific patient outcomes and patient safety culture.
& Conduct intervention research to determine the most effective
means to improve patient safety culture and therefore improve
patient outcomes.
& Conduct research that connects patient safety culture and other
culturally sensitive variables, such as propensity for patient
advocacy, to guide administrators to avenues for improving
the culture of hospitals.
CONCLUSIONS
& Volume 11, Number 3, September 2015
4. Singer S, Lin S, Falwell A, et al. Relationship of safety climate
and safety performance in hospitals. Health Serv Res. 2009;
44:399Y421.
5. Colla JB, Bracken AC, Kinney LM, et al. Measuring patient safety
climate: a review of surveys. Qual Saf Health Care. 2005;14:364Y366.
6. Obrien RL. Keeping Patients Safe: The Relationship Between Patient
Safety Climate and Patient Outcomes [dissertation]. University of
California, San Francisco; 2009.
7. Taylor JA. Utility of Patient Safety Case Finding Methods and
Associations Among Organizational Safety Climate, Nurse Injuries,
and Errors [dissertation]. Johns Hopkins University; 2008.
8. Huang DT, Clermont G, Kong L, et al. Intensive care unit safety culture
and outcomes: a US multicenter study. Int J Qual Health Care.
2010;22:151Y161.
9. Sexton JB. A Matter of Life or Death: Social Psychological and
Organizational Factors Related to Patient Outcomes in the Intensive
Care Unit [dissertation]. University of Texas, Austin; 2002.
The research studies available have been conducted in the
last 10 years demonstrating that the study of the relationship
between patient safety culture and patient outcomes has occurred after the IOM report in 1998. There are multiple welldesigned cross-sectional studies to document the significance
of the relationship; however, no intervention studies have been
published to date. A foundation has been laid for interventional
research, which would enhance the available research and
provide direction for health-care administrators as they continue
to improve the patient safety culture of their organizations.
This review serves to assist future patient safety culture
researchers in study design in the areas of tools, level of analysis, and outcome selection. Research correlating these variables has been progressing over the last 10 years; however,
additional research is needed to understand the existing correlations and to determine interventions that improve the patient
safety culture in hospitals.
Health-care administrators today more than ever are being
held accountable, financially and socially, for adverse events
that occur within their health-care organizations as well as the
overall patient experience. The federal government and general
public sentiment has changed from accepting human error as
inevitable to challenging organizational leadership to improve
health-care systems that result in error and/or a negative patient experience. These changes have made understanding patient
safety culture and its effect on patient outcomes imperative; however, as seen in this review, there is work to be done concerning
the study of patient safety culture and its connection to patient
outcomes.
10. Dodek PM, Wong H, Heyland DK, et al. The relationship between
organizational culture and family satisfaction in critical care. Crit Care
Med. 2012;40:1506Y1512.
REFERENCES
20. Kemper CA. The Impact of Work Unit and Organization System on
Hospital Patient Safety [dissertation]. University of Kansas; 2009.
1. Kohn L, Corrigan JM, Donaldson MS, eds. To Err is Human.
Washington, DC: National Academics Press, Institute of
Medicine; 2000.
2. Leonard MW, Frankel AS. Role of effective teamwork and
communication in delivering safe, high-quality care. Mount Sinai
School Med. 2011;78:820Y826.
3. American Nurses Association, Code of Ethics [ANA Web site]. 2001.
Available at: http://www.nursingworld.org/codeofethics. Accessed
January 13, 2013.
142
www.journalpatientsafety.com
11. Gearhart SF. The Relationship Between Care Provider Perceptions
of Safety Culture and Patient Perceptions of Care on Three Hospital
Units [dissertation]. University of California, San Francisco; 2008.
12. Mardon RE, Khanna K, Sorra J, et al. Exploring relationships between
hospital patient safety culture and adverse events. J Patient Saf.
2010;6:226Y232.
13. Sorra JK, Kabir D, Naomi M, et al. Exploring relationships between
patient safety culture and patients’ assessments of hospital care.
J Patient Saf. 2012;8:131Y139.
14. Thompson DN. A Multi-level Study of Nurse Leaders, Safety Climate
and Care Outcomes [dissertation]. University of Pittsburgh; 2010.
15. Sorra JS, Dyer N. Multilevel psychometric properties of the AHRQ
hospital survey on patient safety culture. BMC Health Serv Res.
2010;10:199Y212.
16. Hansen LO, Williams MV, Singer SJ. Perceptions of hospital safety
climate and incidence of readmission. Health Serv Res.
2011;46:596Y616.
17. Chang Y, Mark B. Effects of learning climate and registered nurse
staffing on medication errors. J Nurs Admin. 2011;60:32Y39.
18. Hofmann DA, Mark B. An investigation of the relationship between
safety climate and medication errors as well as other nurse and patient
outcomes. Personnel Psychol. 2006;59:847Y869.
19. Shortell SM, Zimmerman JE, Rousseau DM, et al. The performance of
intensive care units: does good management make a difference? Med
Care. 1994;32:508Y525.
21. Mark BA, Hughes LC, Belyea M, et al. Exploring organizational context
and structure as predictors of medication errors and patient falls.
J Patient Saf. 2008;4:66Y77.
22. Curry LA, Spatz E, Cherlin E, et al. What distinguishes top-performing
hospitals in acute myocardial infarction mortality rates? Ann Intern
Med. 2011;154:384Y390.
23. Olds DM. Safety Culture and Nurse Sensitive Outcomes [dissertation].
University of Pennsylvania; 2010.
* 2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Brasaite et al. BMC Res Notes (2016) 9:177
DOI 10.1186/s13104-016-1977-7
BMC Research Notes
RESEARCH ARTICLE
Open Access
Health care professionals’ attitudes
regarding patient safety: cross‑sectional survey
Indre Brasaite1,3*, Marja Kaunonen1,2, Arvydas Martinkenas3 and Tarja Suominen1
Abstract
Background: Patient safety is being seen as an increasingly important topic in the healthcare fields, and the rise in
numbers of patient safety incidents poses a challenge for hospital management. In order to deal with the situation,
it is important to know more about health care professionals’ attitudes regarding patient safety. This study looks to
describe health care professionals’ attitudes regarding patient safety, and whether differences exist based on the background factors of study participants.
Methods: A quantitative study using a questionnaire was conducted in three multi-disciplinary hospitals in Western
Lithuania. Data was collected in 2014 from physicians, nurses and nurse assistants.
Results: The results showed positive safety attitudes, and these were especially related to the respondents’ levels
of job satisfaction. A respondent’s older age was associated with how they evaluated their teamwork climate, safety
climate, job satisfaction, and perception of management. Profession, working unit, length of work experience, information received about patient safety during education, further education, and working shifts were all associated with
several safety attitude areas.
Conclusions: The safety attitudes of respondents were generally found to be positive. Attitudes related to patient
safety issues were positive among health care professionals and opens the door for the open discussion of patient
safety and adverse events. However, in future we also need to investigate the knowledge and skills professionals have
in relation to patient safety, in order to gain a deeper understanding of the present situation.
Keywords: Attitude, Health care professionals, Nurses, Patient safety, Physicians, Nurse assistants
Background
Attitudes regarding safety-related issues are an important part of what is often called a hospital’s safety culture [1, 2]. An organization’s safety culture consists of
components concerning healthcare provider attitudes
about organizational factors such as safety climate and
morale, work environment factors such as staffing levels
and managerial support, team factors such as teamwork
and supervision, and staff factors such as overconfidence and being overly self-assured [3]. Some authors
[4–6] have noticed that a safety culture is a part of the
wider organisational culture, and may be defined as the
attitudes, beliefs, perceptions, competencies and values
*Correspondence: Brasaite.Indre.X@student.uta.fi
3
Faculty of Health Sciences, Klaipeda University, Klaipeda, Lithuania
Full list of author information is available at the end of the article
that determine an organisation’s health and safety management, and are held in common by employees in relation to safety. An understanding of nurses’ perceptions
and expectations regarding adverse events is therefore
essential for the implementation of appropriate strategies
to manage nursing care. In this sense, the beliefs, values
and organizational culture of registered nurses (RNs) are
important aspects to be considered [7].
Ethical issues are integral to the topic of patient safety
because it is known that millions of patients worldwide
suffer injury or death every year as a result of unsafe
medical practices and care, and patients are mostly
harmed due to preventable causes that they receive during health care in hospital settings [8]. Health care professionals may know that their role is important in the
delivery of safe care and that they should have positive
safety attitudes. However, the results of a safety culture
© 2016 Brasaite et al. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License
(http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium,
provided you give appropriate credit to the original author(s) and the sourc…