Setting a Course for Patient Safety Research
Although patient safety research has made great strides over the past 15 years, the science base is still evolving. As the lead United States federal agency for patient safety research, the Agency for Healthcare Research and Quality (AHRQ) works with multiple stakeholders to guide and fund research aimed at improving "the quality, safety, efficiency, and effectiveness of health care."(1) In carrying out its mission, AHRQ program officials consider input from and disseminate information to many groups, including other federal agency partners, clinicians, research experts, payers, patients, and the general public. This exchange helps identify research topics and content that lead to improvements in patient safety.
Patient safety research occurs in several stages. Since 1999, the agency's Patient Safety Research Portfolio has focused on the most fundamental of these stages—the identification of risks and hazards inherent to health care. Projects at more advanced research stages have examined the design and development of approaches to preventing patient harm, the successful implementation of these safety practices, and, ultimately, the most effective strategies for spreading these practices. During the past 8 years, this latter area—implementation and spread—has seen some significant progress. In this area of research, investigators consider not only whether a project improved our understanding of patient safety, but also how this knowledge can enhance clinical and organizational practices and support safer care. Here, matters of context, change management, and opportunities for successful dissemination become as important as the more technical aspects such as clinical circumstances and patient risk factors.
AHRQ strives to establish a wider knowledge base that will generate opportunities and resources (2) for patient safety improvements across all parts of the health care system. The aim is to translate research findings into usable information that can improve the delivery of clinical care and ultimately save lives. This ambitious goal requires a dynamic research agenda. Current observations suggest there are select patient safety topic areas that warrant further attention. In this paper, we highlight a few of them: research in health care simulation, diagnostic error and performance, patient safety in ambulatory care settings, and clinical performance measurement. An overview of these examples will illustrate AHRQ's efforts to address prevailing research issues in patient safety, topic-specific challenges, and the funding approaches (3) that are used to convey research priorities to the field.
Health Care Simulation
Some of most promising patient safety research methods and tools have historically originated outside of health care, and we expect this to continue. This is the case for health care simulation, which has been widely applied in aviation training and gained the attention of anesthesiologists looking to improve the safety of surgical patients beginning in the late 1980s.(4) Some think of this tool exclusively as the technology it uses; however, health care simulation has a broader scope, and it has been successfully applied without complex technical or electronic resources.(4) For example, health care simulation offers high-value learning experiences for both individual clinicians and groups of providers. Simulation includes techniques such as the use of standardized patients and team-based exercises to improve communication between providers and patients. Whether high-tech or not, simulation has been used by students and clinicians to practice and strengthen technical and interpersonal competencies and reduce the risks of patient harm. Health care simulation has not reached its full potential, and, as more applications are developed and tested, simulation is poised to help maintain the momentum of safety and quality improvements.
AHRQ has identified health care simulation as a research priority through the development and funded support of a Program Announcement (PA) titled, "Advances in Patient Safety through Simulation Research."(5) Research Program Announcements describe long-term program interests for which grant applications are invited.(6) AHRQ may also convey interest in funding research in a particular subject through publication of a Special Emphasis Notice (SEN). SENs often target more focused research topic areas that have been recognized by the agency, whereas PAs have broad aims.(6) Grant applications received from investigators in the field in response to these funding announcements undergo a rigorous peer review, and successful proposals receive further assessment by administrative and program staff at AHRQ to make funding decisions.
The current PA for simulation research targets novel implementation research to "develop, test and evaluate various simulation approaches for the purpose of improving the safe delivery of health care." The agency initially solicited simulation research proposals by publishing Requests for Applications (RFAs) in 2006.(7) RFAs are another vehicle AHRQ uses to communicate research priorities to the field and typically precede PAs as an initial long-term research investment. The content of these solicitations, whether PA, SEN, or RFA, reflect some of AHRQ's prevalent research priorities. Over the past 7 years, AHRQ has funded 39 patient safety simulation grants.(8) While many of these grants are still in progress, the lessons learned from successfully completed research projects have been integrated into several AHRQ-sponsored contracts, a funding vehicle that can also be used for other topic areas. A research contract allows AHRQ staff to design prescriptive research projects and activities in support of the agency's goals.
One project that follows this approach is titled, "The Design and Development of an Obstetric and Perinatal Improvement System," which draws from the strengths and outcomes of multiple AHRQ simulation grants. The project team, whose goal is to develop implementation tools and a framework for the reduction of obstetric and perinatal harm, has included the use of in situ simulation to model best clinical practices within the inpatient setting. This type of simulation methodology embeds the simulated training experience or scenario in actual care delivery setting in order to teach and improve their competencies. It is one of the many examples that demonstrate how AHRQ's grant-funded research often informs subsequent activities in everyday clinical settings. Projects that are contracted by AHRQ have drawn on findings from existing AHRQ research and the field in order to develop resources and tools and to extend their dissemination.
While the adoption of health care simulation is booming due to its high face value, there remain many important questions about its actual impact on safety and patient outcomes. There is a persistent need for agreed-upon metrics, nomenclatures, skill taxonomies, and tools to guide research and development in different specialty areas. Through targeted research, AHRQ strives to expand understanding about the optimal use of health care simulation and channel the enthusiasm that surrounds it to maximize its potential to improve safety and quality.
Diagnostic Error and Performance
Most of these challenges are not unique to health care simulation research, but are sorely evident in many areas that are struggling to gain wider traction. The field of diagnostic error and performance is another example of one of these emerging subject areas. The Institute of Medicine's landmark patient safety publication, To Err is Human: Building a Safer Health System, covered many important issues that warranted immediate attention. However, the issue of diagnostic error and performance was not thoroughly addressed in this report, nor has it been firmly established in the safety field.(9) This gap is due in part to the complexity of this area and the relative absence of a common research framework to inform scope and direction. A diagnostic error is a patient safety event that results from missed, delayed, or inaccurate diagnosis of a patient during the course of clinical care.(10) Diagnostic performance examines both the individual performance (cognitive) and system-related factors associated with providing patient diagnosis accurately and precisely.(11) AHRQ recognizes that diagnostic error research is still finding its identity in the health services research landscape. As work in this area progresses, a better understanding of both systems and individual performance factors that relate to diagnosis offers significant prospects for improving quality.
The agency has taken active steps to contribute further knowledge about diagnostic error, its causes, and ways to reduce it. In 2013, AHRQ issued a SEN for grant funding to accelerate research progress in diagnostic performance. The SEN's call for "Research to Improve Diagnostic Performance in Ambulatory Care Settings" was designed to gain "a better understanding of the incidence, cost, determinants, and strategies for preventing or mitigating diagnostic errors" in ambulatory care settings.(12) This solicitation prompted the development of several resources and tools and has also complemented and informed other AHRQ initiatives.
One example of this linkage is demonstrated through the research program, "Proactive Risk Assessment during Clinical Laboratory Testing." Output from this project has significant promise to address the patient safety risks associated with clinical laboratory processes by identifying themes in the current literature and also offering vetted risk assessment models for use by health care delivery organizations. While these efforts represent progress, major gaps still remain. In 2013, JAMA Internal Medicine published the findings of an AHRQ-funded study in this area. Research investigators determined that the most common forms of harm were associated with provider–patient interactions, and that the potential remedies will not come from a discrete and well-defined area of research or medicine, alone.(13) The prevention of diagnostic error and sustained improvement of diagnostic performance will not only require a closer look at clinical processes, but also providers' cognitive perceptions and processes as they deliver care. It will also require stronger engagement by providers and systems, both of whom must partner to most effectively address diagnostic performance and error prevention.(11)
Ambulatory Patient Safety Research
Today, the number of patients who receive care in ambulatory environments is expanding, and research in this setting continues to grow in importance. Much of the early focus in the patient safety field concentrated on the inpatient setting, and that effort has resulted in some major steps forward. Hospital-associated patient safety issues have been relatively easier to measure and improve, because of this environment's controlled (albeit complex) structure. In comparison to the hospitals, ambulatory settings often have fewer resources and less infrastructure specifically focused on safety and quality. Although many challenges remain in the inpatient safety field, AHRQ recognizes the need to examine and keep up with shifting trends in care delivery. These movements are influenced by many sources, including AHRQ stakeholder input and patient safety research outcomes.
Ambulatory care includes medical services provided in an outpatient basis (i.e., non-hospitalized, non–long-term care). These services may include diagnosis, treatment, surgery, and rehabilitation and span medical offices, community pharmacies, diagnostic laboratories, emergency departments, outpatient surgery centers, home health care, and others.(14) Although these environments may be less technologically complex than the inpatient setting, they often face more logistical complexities than inpatient care. Challenges with coordination and communication—some similar to those in hospital care and others unique to ambulatory settings—can interfere with information exchange among providers and between patients and providers. The dispersed nature of outpatient care can result in greater reliance on patients to actively manage their care. This creates different types of opportunities for patient safety events to occur and is further complicated for patients with limited English proficiency, low health literacy, and low income.
Over the past decade, AHRQ has spurred progress in this area by addressing the technological limitations as well as the behavioral and logistical complexities of ambulatory care teams. In 2007, the agency's Health Information Technology program launched an Ambulatory Safety and Quality (ASQ) grant initiative, which examined how technology such as electronic health records can improve safety and quality.(15) Subsequent contracts have further promoted work in this area.
Improving patient safety depends on fostering constructive changes within the organizational work culture environment of health care facilities—a problem that is not unique to ambulatory care settings. As ambulatory providers increasingly adopt a patient-centered perspective, this will require a significant change in the way providers deliver care in teams. Training opportunities and resources in this area are available through the agency's Team Strategies and Tools to Enhance Performance and Patient Safety (TeamSTEPPS) program, an "evidence-based teamwork system to improve communication and teamwork skills among health care professionals" that is currently being implemented in hospitals across the nation.(16) This program has also produced instructional models for teams that work in primary care offices and provide care to patients with limited English proficiency. Further implementation of TeamSTEPPS in medical offices is underway and demonstrates another example of a targeted research project to address a specific improvement goal.
AHRQ projects are also supporting improving and spreading tools, and providing technical assistance to users in support of the agency's mission. The Consumer Assessment of Healthcare Providers and Systems (CAHPS) program and the Survey on Patient Safety Culture for the Medical Office are examples of such publicly available tools. Survey on Patient Safety Culture instruments are staff surveys designed to help health care organizations assess the culture of safety in their institutions.(17) CAHPS surveys "ask consumers and patients to report on and evaluate their experiences with health care."(18) Both tools have specific instruments focused on the ambulatory setting, and their marked success has prompted further development of a Survey on Patient Safety Culture and CAHPS resources for other ambulatory environments.
It is critical that researchers test the effectiveness of prospective tools and resources in real world circumstances—something hard to do in a fast-paced and busy ambulatory environment. The agency's primary care Practice-Based Research Network (PBRN) program has provided health care laboratories in which to study and address patient safety issues. In 2012, there were more than 160 active registered networks representing thousands of ambulatory medical practices that focus on improving the delivery of patient care.(19) Recent AHRQ work with primary care PBRNs has focused on developing and implementing strategies to reduce inappropriate prescribing in primary care settings.
Clinical Performance Measurement
Today's health care delivery system is replete with many different forms of measurement, many of them automated to varying degrees. While most providers complain about the ever-increasing burden of "paperwork" (including required reporting by hospitals of performance measures), there are few systems in place that provide truly valuable, comprehensive information about providers' performance. There are even fewer that allow benchmarking against other similar delivery sites (e.g., hospitals, physician offices, etc.), trending over time, and aggregation of compatible clinical information on a national basis. There are notable exceptions, most particularly including hospital measures derived from billing data (AHRQ's Patient Safety Indicators and Quality Indicators), specific performance measures required by accrediting or insurance organizations (Centers for Medicare and Medicaid Services' Inpatient Quality Reporting or IQR), and manually entered clinical data on health care–associated infections (the Centers for Disease Control and Prevention's National Healthcare Safety Network or NHSN). A hospital using all of these measures would, however, only scratch the surface of evaluating the quality and safety of all of its clinical services. Recognizing this, many hospitals (or health systems) mount their own specialized studies to answer important clinical questions. Depending on the skill of those developing such studies, they may yield high-quality information and improved care, or they may not. In either case, these studies remain locked within the silos of their developers and are not "interoperable" with other measurement systems, electronic health records, etc.—either clinically or electronically. The time spent, lessons learned, and software developed for localized projects do not transfer easily, if at all, to other settings.
The problem boils down to this: There is no overall scheme today (nor is there likely to be one soon) guiding individual measure developers to assure (i) coherence and comprehensiveness of content areas addressed by measures; (ii) uniformity of assumptions and conventions employed in constructing measures that shape clinical definitions, electronic specifications, and interoperability; and (iii) an evaluation of the cost of the resources necessary to collect and analyze data that drive measures versus the value of the information produced by those measures. The Patient Safety and Quality Improvement Act of 2005 did, however, provide AHRQ with the authority to begin to address the problem of harmonizing measurement. It contained a provision authorizing AHRQ to promulgate "common definitions and reporting formats," which we have simplified to the term "Common Formats," to allow collection of uniform clinical information. Such uniform measures would, in turn, support collection and submission of uniform clinical information to Patient Safety Organizations (PSOs), also authorized by the Act.(20)
Challenges With Prioritizing the Science of Patient Safety and Research Translation
Even though there is growing recognition of the importance of patient safety, the research base remains insufficient in many areas. Patient safety outcomes can improve through the use of well-planned research projects and thoughtfully developed tools and resources for clinicians and patients, but without wider credence and commitments to safety, improvements will be slower and harder to achieve. Although most individuals would agree that safe health care delivery is important, some of the theoretical basis and processes underlying patient safety improvement are esoteric for many audiences. Foundational disciplines important to patient safety are not widely known by or intuitive to some health care stakeholders. The AHRQ Patient Safety Research Portfolio is bridging this exchange and providing new opportunities for interdisciplinary research translation by applying principles from disciplines without historically strong links to health care. Such areas include systems engineering, human factors, organizational and applied psychology, environmental design, education, and public health surveillance. As more health care stakeholders recognize the importance of patient safety research and its application, more interdisciplinary research initiatives will enable and support safer care for all patients.
P. Jeffrey Brady, MD, MPH Associate Director, Center for Quality Improvement and Patient SafetyAgency for Healthcare Research and Quality
William B. Munier, MD, MBA Director, Center for Quality Improvement and Patient SafetyAgency for Healthcare Research and Quality
Irim Azam, MPH Agency for Healthcare Research and Quality
1. AHRQ at a Glance. Rockville, MD: Agency for Healthcare Research and Quality; August 2012. [Available at]
2. AHRQ Patient Safety Tools and Resources. Rockville, MD: Agency for Healthcare Research and Quality; May 2012. AHRQ Publication No. 12-M008. [Available at]
3. Funding and Grants. Rockville, MD: Agency for Healthcare Research and Quality. [Available at]
4. Gaba DM. The future vision of simulation in health care. Qual Saf Health Care. 2004;13(suppl 1):i2-i10. [go to PubMed]
5. Advances in Patient Safety through Simulation Research (R18). Rockville, MD: Agency for Healthcare Research and Quality. PAR-11-024. [Available at]
6. Funding Opportunity Announcement (FOA) Guidance. Rockville, MD: Agency for Healthcare Research and Quality; December 2012. [Available at]
7. Improving Patient Safety through Simulation Research. Rockville, MD: Agency for Healthcare Research and Quality. RFA-HS-06-030. [Available at]
8. Improving Patient Safety through Simulation Research: Funded Projects. Rockville, MD: Agency for Healthcare Research and Quality; September 2011. AHRQ Publication No. 11-P012-EF. [Available at]
9. Wachter RM. Why diagnostic errors don't get any respect—and what can be done about them. Health Aff (Millwood). 2010;29:1605-1610. [go to PubMed]
10. Patient Safety Primer: Diagnostic Errors. AHRQ Patient Safety Network Web site. Updated October 2012. [Available at]
11. Henriksen K, Brady J. The pursuit of better diagnostic performance: a human factors perspective. BMJ Qual Saf. 2013;22(suppl 2):ii1-ii5. [go to PubMed]
12. AHRQ Announces Interest in Research to Improve Diagnostic Performance in Ambulatory Care Settings. Rockville, MD: Agency for Healthcare Research and Quality; May 2013. NOT-HS-13-009. [Available at]
13. Singh H, Giardina TD, Meyer AN, Forjuoh SN, Reis MD, Thomas EJ. Types and origins of diagnostic errors in primary care settings. JAMA Intern Med. 2013;173:418-425. [go to PubMed]
14. National Advisory Council: Meeting Summary, July 13, 2012. Rockville, MD: Agency for Healthcare Research and Quality; July 13, 2012. [Available at]
15. Ambulatory Safety and Quality Program. Rockville, MD: Agency for Healthcare Research and Quality; July 2012. AHRQ Publication No. 12-P012. [Available at]
16. TeamSTEPPS: National Implementation. Rockville, MD: Agency for Healthcare Research and Quality. [Available at]
17. Surveys on Patient Safety Culture. Rockville, MD: Agency for Healthcare Research and Quality; September 2012. [Available at]
18. CAHPS. Rockville, MD: Agency for Healthcare Research and Quality; July 2012. [Available at]
19. Davis MM, Keller S, DeVoe JE, Cohen DJ. Characteristics and lessons learned from practice-based research networks (PBRNs) in the United States. J Health Care Leadersh. 2012;4:107-116. [Available at]
20. Patient Safety Organizations: Common Formats. Rockville, MD: Agency for Healthcare Research and Quality. [Available at]