Blumenthal, David MD, MPP; Ferris, Timothy G. MD, MPH
Improving patient safety is one of many rallying cries that can be heard these days above the drumbeat of normal business in the corridors of the nation’s academic medical centers (i.e., teaching hospitals with significant research activity). Regulators, payers, purchasers, and the public want many things from academic medical centers (AMCs)—reduced costs, responsiveness to local community needs, more transparency with performance data, continued progress in biomedical research, and more rapid transfer of research results into practice. But no demand is more insistent than the pressure to measure and improve the safety of patients.
Few AMC leaders would disagree with this perspective in principle. The problem, however, is in execution. AMCs face the same challenges that other health care institutions face in improving patient safety, and some unique ones as well. The first challenge is to understand the meaning of patient safety, and the conceptual basis for improving it. A second challenge is to understand the special circumstances of AMCs—their comparative strengths and weaknesses in tackling safety issues. And a third challenge is to understand how AMCs can manage their special circumstances to make their patients safer. In this article, we review these three topics, prefacing the discussion with a very brief review of what is known about the state of patient safety in AMCs at the current time.
Safety in the Academic Medical Center
To our knowledge, no published studies have focused specifically on measuring the safety of care in AMCs as a class of institutions.1 Data on safety in AMCs generally result from subanalyses within studies designed to investigate the safety of care in hospitals generally. Other studies are now outdated. No studies separate AMCs from teaching hospitals more generally. For the most part, this scant, imperfect literature reveals a mixed picture concerning how safety of care in AMCs compares to safety in other inpatient settings.
Data from the well-known, if controversial, Harvard Medical Practice Study suggested that, after adjusting for patient characteristics, teaching hospitals in New York State during the mid-1980s tended to have fewer safety problems— measured as preventable adverse events (PAEs)—than did nonteaching hospitals.2 A subsequent study using the same methodology in Utah and Colorado hospitals showed no difference.3 A more recent study by Romano and colleagues4 examined administrative data from a national sample of hospitals over the period 1995–2000 using a modified version of the so-called Patient Safety Indicators (PSIs) developed by the Agency for Health Research and Quality (AHRQ) for the specific purpose of identifying possible safety problems using administrative data. This work was not encouraging for AMCs: it found that several types of PSIs were higher in urban teaching hospitals than in urban nonteaching hospitals or nonurban institutions. Given the imperfections of administrative data and the experimental nature of PSIs, these results can hardly be taken as definitive.
Other sources of information about safety in AMCs should also motivate self-examination. To their credit, AMCs have led the way in studying patient safety concerns. One reason that safety is an issue now is that these AMC investigators found plenty to worry about right in their own institutions. For example, Bates and colleagues5 documented the frequency of errors in medication administration at the Massachusetts General Hospital (MGH) and the Brigham and Women’s Hospital (BWH) (both routinely listed among the top ten U.S. hospitals in U.S. News and World Report). These same institutions were included in a study that found 1% of ICU admissions were caused by errors and over a third of these were preventable.6 Pronovost and colleagues at Johns Hopkins Hospital documented opportunities to improve safety of care within the Hopkins intensive care unit.7 Gawande interviewed 38 AMC surgeons and obtained reports of 146 errors.8 One third of these errors had resulted in either permanent disability or death.
Residents provide much of the front-line care at AMCs and, notwithstanding recent changes in work hour regulations, their performance has come under increased scrutiny. Impaired performance of sleep deprived residents has been equated with specific blood alcohol levels9 as well as directly implicated in errors in an ICU.10 On the other hand, fewer hours in the hospital require more handoffs, and these too have been associated with safety problems.11
Much work in safety outside of health care has linked the frequency of adverse events to what is described as the “culture” of safety. Surveys of patient safety culture developed by the AHRQ and others12 have been used to assess medical personnel’s perceptions of workplace safety. At the request of some AMCs, in 2004 and 2005, we asked a broad cross-section of nurses and physicians about safety issues in confidential surveys (our findings are still unpublished). We found large differences between AMC units in (1) attitudes regarding safe care; (2) the likelihood that an error would be reported; and (3) the level of comfort with discussing potential safety issues openly. While such unit-to-unit variation may be common to all hospitals, we found specific problems with nurse-resident and attending-resident communication that would only exist in teaching hospitals.
Moving from data to anecdotes, which are often more powerful, two dramatic examples show that AMCs are hardly immune to medical accidents. The first is the celebrated case of the death of Betsy Lehman, a health care reporter who received a massive chemotherapy overdose at the Dana Farber Cancer Institute in the 1990s. The second was the episode at Duke University Medical Center in which the transplantation of a mismatched heart led to the recipient’s death. The literature on safety in other complex, risky, human institutions suggest that such dramatic incidents are invariably the tip of the iceberg: the unfortunate but telling cases that rise to wide-spread awareness. Personally, our collective 40-person-year experience working as clinicians at the Massachusetts General Hospital suggests that safety problems are frequent, and that only the talent and dedication of physicians and nurses—and the increasing vigilance of patients—prevent these problems from exacting a greater human toll.
Conceptualizing the Safety Problem
The Institute of Medicine’s (IOM’s) landmark report, To Err is Human, provides the most authoritative primer on the definition and origins of threats to safety in health care, and at the risk of distracting or boring some readers, it is worth summarizing some fundamental lessons of that document.13
Following Reason,14–17 the IOM defined an error to be “a failure of a planned action to be completed as intended (i.e., error of execution) or the use of a wrong plan to achieve an aim (i.e., error of planning).” Reason asserts that errors result from two types of failures: active and latent. Active failures are unsafe acts (slips, mistakes, and violations) by individuals working at the point of production, the so-called “sharp end.” The nurses and oncology fellows who gave Betsy Lehman her overdose of chemotherapy and the surgeon at Duke who implanted the wrong heart were operating at the sharp end. Active failures are more visible, and traditionally, the blame and the remedy for addressing safety threats have focused on the individuals who were unfortunate enough to have their hands on the controls when the accident occurred.
But safety problems, the IOM points out, are usually the product of deeper problems, known as latent failures. Latent failures are the consequences of decisions of designers, builders, managers, regulators, and policy-makers who operate at a distance from the point of production (the blunt end) and directly or indirectly shape the systems in which humans operate. This perspective highlights the role of the medication administration process and its flawed design in the Betsey Lehman case, or the process of assuring the correct matching of organs and donors in the Duke case. Efforts to reduce errors are most effective when they focus on preventing latent rather than active failures.17,18
This analysis suggests that error reduction efforts in health care should concentrate on changing systems both to improve their resistance to human error and to identify and repair flaws in the systems themselves that produce errors.19,20 Reducing errors requires designing processes that are reliable, ones that are resistant to human errors in execution. Reducing error also requires addressing other systemic factors that often underlie poorly designed processes: policies, procedures, practices, and conditions that may permeate organizations and their subunits, and reduce the reliability of processes. Examples of the latter factors include lack of organizational leadership for safety, absence of a suitable organizational culture, and generalized failure to incorporate human factors knowledge into the design of processes.13
Consistent with these observations, the IOM outlined a number of actions that are essential to error reduction.13 These include the creation of national leadership for error reduction, respect for human limits in process design, the promotion of effective team functioning, and anticipating the unexpected by creating processes that are error-resistant. This focus on improving systems of care —reducing latent failures—is consistent with arguments for seeing problems of safety in the larger context of efforts to improve quality of care generally.21 The emphasis on safety has lead in some venues to artificial distinctions between addressing safety problems and making quality better. In fact, threats to safety are often indistinguishable—conceptually and practically—from threats to quality of care, as are the methods for improving safety and improving quality. The IOM’s report notes that error reduction is, in fact, a species of quality improvement, and that errors are most likely to fall within the category of quality problem known as “misuse” of care.22 A subsequent IOM report, Crossing the Quality Chasm, made this connection clear by defining safety as one of the six dimension of quality of care.23
Preventing accidents such as the overdose at Dana Farber or the transplant catastrophe at Duke requires that we improve the quality of medication administration at the former institution and transplant surgery at the latter. This in turn requires understanding the systems of medication administration and organ matching at work in these respective institutions so that we may identify the active and latent failures that may occur in these systems and eliminate these failures, thereby making those systems more reliable, more resistant to human failures, and higher in quality.
The Special Circumstances of AMCs
What does all this mean concretely? There is no mystery to the job of quality and safety improvement. It is mostly about elbow grease—applying tried and true quality and safety improvement techniques to the daily work of health care. The shelves of libraries groan under the weight of texts, many dusty with age, that describe these techniques in minute detail.24–27 For the most part, they suggest methods for applying the scientific method to diagnosing the failures of processes, testing ways to make them more streamlined and reliable, and putting those methods in place. Continuous quality improvement requires first the measurement of the outcomes of current processes of care, then the analysis of the steps in those processes and how they relate to one another, followed by experiments with changing those steps, measuring the effects on quality and safety, and either incorporating the changes or rejecting them based on the data collected – and starting again. Anyone with even passing acquaintance with the scientific method will find these techniques instantly recognizable, understandable, and appealing.24
This makes it all the more puzzling that AMCs—given their leadership in applying science to improving medicine—have not been leaders in the adoption of quality and safety improvement techniques. To understand why they have not been, and to understand how AMCs should approach safety issues in the future, we need to examine the inherent strengths and weaknesses of that unique U.S. health care enterprise, the academic medical center. The observations that follow are based on an unpublished study conducted by the Institute for Health Policy at MGH for the Robert Wood Johnson Foundation in 2003 on the subject of the problems and opportunities experienced by AMCs trying to improve quality of care.28 In the process, we interviewed over 50 AMC leaders and national quality experts to try to understand the special opportunities and challenges that AMCs face in the quality and safety improvement realm. We also held a half-day meeting with AMC and quality leaders to vet our observations and conclusions.
We have already alluded to one of the many strengths of AMCs: their involvement in research. A corollary of this is that AMCs teem with smart, well-trained personnel, both physicians and nonphysicians, who have the skills to undertake the analytic tasks that are fundamental to science-based process improvement. No one should underestimate the importance of this reservoir of talent. In many small health care institutions around the United States, the basic understanding of statistics and experimental design that form the foundation of continuous quality improvement must be developed from scratch or imported.
AMCs also benefit from their role in teaching. A critical task in quality and safety improvement is to communicate the results of quality and safety improvement projects to colleagues in the health care institution, so that they will accept the required changes in the way their organizations conduct care. This communication process is analogous to teaching, a skill that should be much more abundant in AMCs than in other health care institutions.
AMCs’ weaknesses and challenges
Balanced against these and other strengths, however, are several weaknesses that AMCs must overcome if they are to make fundamental advances in patient safety through quality improvement. The first is the diversity of the missions they pursue, which is in some ways the flip side of their above-mentioned skills in teaching and research. The great majority of U.S. health care institutions have a single focus on patient care. AMC managers must divide their attention between patient care, teaching, and research—all of which are central tasks in many AMCs. This diversity of missions makes it harder to capture the attention of AMC faculty than the attention of clinicians in other facilities. Several interviewees in our study commented on the limited intellectual “space” for health professionals at AMCs. “You are very lucky to harness even one-third of the ‘mind-space’ of people at AMCs. They are chronically overwhelmed.” Every action physicians and their managers take must be assessed not only for its impact on clinical care, but also for its impact on research and training. What is more, patient care is not always accorded the top priority by AMC personnel. Perhaps with some exaggeration, one commentator noted: “It’s like we are riding a tricycle but with the wheels in a line. The front wheel is research, the wheel connected to the drive shaft. The middle wheel is teaching. It’s small and missing a few spokes. Dragging up the rear is the patient care wheel and it’s got a flat tire!”
A second challenge facing AMCs is their complexity as institutions, which flows directly from the diversity of their missions, as well as from their size. In modern parlance, AMCs have three product lines, not one—patient care, new health care knowledge, and trained health professionals. And three product lines require at least three explicit or implicit production facilities. Of course, the products overlap, are co-produced and are often inseparable. But there are real organizational consequences to the diverse missions of AMCs, and this makes these institutions bigger and more far flung, with more people to manage, more issues coming to the attention of AMC executives, more crises to respond to, and more revenue and cost streams to balance. It requires greater persistence and managerial talent to maintain a focus on patient safety and quality in the midst of such varied crises than would be required in a less complex institution.
Still a third challenge facing AMCs wishing to improve safety and quality relates to the values and reward systems that operate within these organizations. This might be loosely called the academic culture. In our study of AMCs, we found the following central attributes of AMC culture, which we discuss in the following paragraphs.
* The placement of a higher value on individual autonomy than on commitment to the well-being and goals of the enterprise.
* Resistance to process standardization.
* Low regard for systems thinking.
Though many of these characteristics are not confined to AMCs, most are thought to be stronger in AMCs than in other health care organizations. For example, the high premium that physicians place on individual autonomy is a pervasive feature of health care organizations, both academic and nonacademic. This cultural commitment to individual interests above enterprise interests makes it difficult to unify physicians around the larger goals of their organizations. Often even a few rebellious physicians can frustrate the coordinated activity required to improve systems of care.
Additional factors reinforce the culture of autonomy. Academic promotion generally requires the demonstration of individual excellence in research, teaching and/or clinical care. Promotion committees frequently demand evidence documenting candidates’ independent contribution to their field and sometimes (e.g., pyramidal promotion systems), a candidate must demonstrate that his or her contribution is superior to that of potential competitors within the candidate’s department. It is hard to imagine an incentive structure that is more toxic to teamwork, more likely to build a culture that reinforces individualism at the expense of cooperation. Faculty who succeed in this process progress to become tenured in their institutions and some rise through the ranks to become department chairs, deans, and CEOs of hospitals and health systems. Thus, the culture of individualism and autonomy becomes pervasive.
Experts described strong resistance to and ignorance of the advantages of process standardization as a second cultural barrier to improvement in AMCs. Applied appropriately, standardization of processes using tools such as pathways, guidelines and forcing functions can reduce the probability of human error, patient harm, and work stress. Yet aversion to reducing wasteful practice variation is common in clinical settings of all types, particularly AMCs. In many ways, this cultural attribute is a corollary of zealous commitment to individual autonomy.
Here again, in AMCs unique aspects of culture and social function reinforce this resistance to standardization. A core value and social responsibility of academia is innovation. Innovation by definition departs from standard practice, and there is a legitimate concern that efforts to reduce wasteful or harmful variation could inhibit innovation. A counterargument is that reducing variation in routine care makes it easier to conduct rigorous controlled experiments in clinical settings. However logical, this argument is counterintuitive to academics and often falls on deaf ears.
A second objection to standardization stems from the prevailing approach to medical education. A critical function of health professions education is to teach critical thinking, a prerequisite for life-long learning in a rapidly changing discipline. If faculty teach and role model the acceptance of standardized practice, questions arise about whether students will learn how to critique standards when they become outdated or when they are incorrectly implemented, and also about whether future physicians will exhibit the skepticism that is so important to learning and innovation. It should be possible to instruct students on how to both question and accept process standardization, when and where appropriate. Current faculty are not trained, however, to understand or convey such a complex set of attitudes.
Still a third cultural characteristic that can inhibit improvement in AMCs is what interviewees described as lack of “systems thinking.” This can be defined as an inability or unwillingness to understand or value the way in which the actions of individuals or organizational subunits contribute to the success of the whole enterprise, and, conversely, the way in which enterprise success benefits individuals and subunits. A unique problem in AMCs with regard to systems thinking is that the mission to advance knowledge and learning through research and teaching encourages faculty to develop deep allegiances to their fields of professional investigation or specialty. Thus, the academic’s understanding of and loyalty to fields such as human genetics or gastroenterology may exceed, and at times compete with, understanding of and loyalty to that individual’s home institution. Since faculty cannot excel in their own AMC without cultivating such external bonds, AMC leaders must not only tolerate them, but encourage them.
A structural attribute of AMCs that discourages systems thinking is the “soft” funding base for core activities, including research and some teaching. AMCs are not hierarchical organizations in any traditional sense, but rather loose alliances of individual academic entrepreneurs organized into somewhat artificial groups: units, departments and schools. The structure of the AMC is more of a federation of individuals, emergent like a coral reef, than it is a coordinated enterprise with unifying values and goals. Indeed, faculty members often feel that they give more to their institutions (through overhead and dean’s taxes) than they get back. They may have trouble identifying how AMCs as systems add to their core product: new knowledge, improved learning or even, excellent, innovative care.
Making Care Safer at AMCs
Despite the many challenges facing quality and safety improvement at AMCs, it would be erroneous to conclude that AMCs are uninterested in such improvement, or unmotivated to accomplish it. It seems more accurate to conclude that interest and commitment to quality and safety improvement are wide-spread among AMCs, but not necessarily wide-spread within them. Almost all AMCs have groups of managers and faculty who understand the vision of the IOM’s recent reports on safety and quality, who have experimented with particular projects, who attend national meetings at which examples of success are shared, and who would strongly support greater organizational investment in these goals. Recent developments have increased somewhat the leverage of these individuals within AMCs. Notable in this regard have been efforts by purchasers such as the Leapfrog Group, private insurers, and the Center for Medicare and Medicaid Services (CMS) to increase market pressures and rewards for quality and safety improvement. AMCs are certainly not immune to such market pressures and rewards. It is fair to say that senior managers in AMCs are more aware of quality and safety concerns now than ever before. Rather than ignoring these as in the past, many now regard them with at least skeptical curiosity and a few have become enthusiastic supporters and champions.
The question, of course, is what these potential leaders of safety and quality efforts can and should do to make progress within the unique environment of the AMC. While the answers to this question are far from clear at this time some strategies are apparent. Regardless of the specific strategy adopted, AMCs have the opportunity, and we would suggest obligation, to be (1) the source of new knowledge on health care safety; and (2) the transmitter of new skills in safe patient care for the health care providers of the future.
First, to make safety improvement more appealing to AMCs and their key staff, the empirical basis for alternative approaches to quality and safety improvement needs to be improved. In our interviews, several experts referred to a need to dramatically expand research into effective organizational change and performance improvement. This recommendation was endorsed by a recent publication in Health Affairs laying out a research agenda for implementing the Quality Chasm report.29 AMCs and their personnel are well suited to the tasks of scientifically based innovation and should be the source of new information for the nation on how best to keep patients safe. Examples include the rapidly growing literature on computerized provider order entry and training using medical simulators.
Second, AMC leaders may benefit from positioning efforts to improve safety and quality within more comprehensive organizational-change approaches. Some AMCs may find safety and quality more appealing if they are presented in the context of improving overall organizational performance. This last point mirrors the growing interest in the Baldrige approach to organizational transformation. This approach is a public domain strategy for comprehensive change based upon periodic assessments using the criteria for the Malcolm Baldrige National Quality Award (MBNQA). In other industries and in health care, award applications represent a small subset of those using the assessment process as the foundation for organizational maturation. In 2002, there were significantly more applications for the MBNQA from health care organizations than from other sectors of the U.S. economy: 17 health care applicants compared to 19 total from education, manufacturing, and small business sectors. In 2003, 19 health care organizations applied for the MBNQA out of a total of 68 organizations across the U.S. economy, indicating a dramatic increase in interest from other sectors and a small increase from the health care sector.
Third, customizing safety and quality improvement to the special missions of AMCs may be valuable. There are opportunities to contribute to patient safety and quality of care not only in the clinical areas of AMCs, but through initiatives in teaching and research as well. In the latter case, patient safety during clinical trials has particular salience and has become an issue for many academic centers. In addition, recent moves by the ACGME to incorporate systems improvement into residency curricula may serve to reinforce nascent efforts to promote safety consciousness through the formal and informal teaching occurring daily on the wards and units. Increasingly common links between operational and teaching missions include executive safety teaching rounds.30,31 This customization of safety interventions to AMCs would respond both to the need for performance enhancement in these other areas and to the felt need of senior managers to take into account the effects of improvements in one mission on the operation of others.
Fourth, managers face challenges with prioritizing investments in safety interventions. Again, this problem is not unique to AMCs, but the process of prioritization and investment, from smart pumps to bar codes to additional staff, must include a wider array of constituents, interests, and priorities. The complexity of this task alone is itself a barrier to progress, let alone to obtaining the capital required for some of this work
Finally, as pressures for AMCs to improve safety and quality increase, it may be necessary for leaders to begin a conscious effort to reconcile the academic culture with the organizational requirements for improving patient safety and quality of care. This is almost certainly the most fundamental and important challenge that AMCs must confront in this area, and will require a multipronged strategy over many years. One element in this strategy must be enlisting the boards of AMCs more in the patient safety and quality issue. Boards can provide valuable assistance to leaders wishing to make change by encouraging them during tough times, and supporting them when dissident faculty resist efforts at reform. Another critical element in reconciling safety and quality with the academic culture is creating routes for promotion that build on leadership in safety and quality. Some AMCs have identified promotion pathways based on excellence in clinical care and teaching in addition to traditional research. Clinical excellence should arguably encompass extraordinary accomplishments in improving clinical care through identifying and implementing breakthrough changes in organizational processes. A third element in reconciling academic culture and safety and quality improvement will be developing a cadre of faculty role models who can provide bedside (and computer-side) instruction to housestaff, fellows and students in problems of safety and quality. The opportunities for such instruction are legion in every AMC—quality and safety problems are visible daily to anyone skilled in identifying them. Changing the culture of AMCs will require creating current and future faculty who understand systems of care and have the interest and capability to conduct teaching and investigation related to safety and quality.
Other approaches can and should be identified. Suffice it to say, once again, that among the many opportunities and problems knocking on the doors of AMC leaders, none is pounding louder and more insistently than the need to address safety problems. To respond, AMC leaders will need to implement strategies that are consistent with the theory and practice of safety and quality improvement, and adapted to the special circumstances of their unique institutions. Centers dedicated to innovation in medical science can and should be able to effectively innovate in medical care delivery and spread those innovations widely. If they are successful in this regard, AMC leaders will accomplish much more than making another managerial headache go away: they will save lives, and lay the groundwork for a more effective, informed, and professional health care workforce for generations to come.
1 Kupersmith J. Quality of care in teaching hospitals: a literature review. Acad Med. 2005;80:458–66.
2 Localio AR, Lawthers AG, Brennan TA, et al. Relation between malpractice claims and adverse events due to negligence. Results of the Harvard Medical Practice Study III. N Engl J Med. 1991;325:245–51.
3 Thomas EJ, Studdert DM, Burstin HR, et al. Incidence and types of adverse events and negligent care in Utah and Colorado. Medical Care. 2000;38:261–71.
4 Romano PS, Geppert JJ, Davies S, Miller MR, Elixhauser A, McDonald KM. A national profile of patient safety in U.S. hospitals. Health Aff (Millwood). 2003;22:154–66.
5 Bates DW, Cullen DJ, Laird NM. Incidence of Adverse Drug Events and potential adverse drug events: implications for prevention. JAMA. 1995;274:29–34.
6 Lehmann LS, Puopolo AL, Shaykevich S, Brennan TA. Iatrogenic events resulting in intensive care admission: frequency, cause, and disclosure to patients and institutions. Am J Med. Apr. 2005;118:409–13.
7 Berenholtz SM, Pronovost PJ, Lipsett PA, et al. Eliminating catheter-related bloodstream infections in the intensive care unit. Crit Care Med. Oct. 2004;32:2014–20.
8 Gawande AA, Zinner MJ, Studdert DM, Brennan TA. Analysis of errors reported by surgeons at three teaching hospitals. Surgery. Jun. 2003;133:614–21.
9 Arnedt JT, Owens J, Crouch M, Stahl J, Carskadon MA. Neurobehavioral performance of residents after heavy night call vs after alcohol ingestion. JAMA. Sep 7. 2005;294:1025–33.
10 Landrigan CP, Rothschild JM, Cronin JW, et al. Effect of reducing interns’ work hours on serious medical errors in intensive care units. N Engl J Med Oct 28. 2004;351:1838–48.
11 Petersen LA, Brennan TA, O’Neil AC, Cook EF, Lee TH. Does housestaff discontinuity of care increase the risk for preventable adverse events? Ann Intern Med Dec. 1994;121:866–72.
12 Pronovost P, Sexton B. Assessing safety culture: guidelines and recommendations. Qual Saf Health Care. Aug. 2005;14:231–33.
13 Kohn LT, Corrigan JM, Donaldson MS. To Err is Human: Building a Safer Health System. Washington DC: National Academy Press; 2000.
14 Reason J. Human Error. Cambridge, England: Cambridge University Press; 1990.
15 Reason JT. Human error in medicine. In: Bogner MS, ed. Hillsdale, NJ: Lawrence Erlbaum Associates; 1994.
16 Reason JT. Managing the risks of organizational accidents. Brookfield, VT: Ashgate Publishing Company; 1997.
17 Reason J. Human error: models and management. BMJ. 2000;320:768–70.
18 Sagan SD. The limits of safety. Princeton, NJ: Princeton University Press; 1993.
19 Moray N. Error reduction as a systems problem. In: Bogner MS, ed. Human error in medicine. Hillsdale, NJ: Lawrence Erlbaum Associates; 1994.
20 Juran JM. Juran on leadership for quality: an executive handbook. New York: The Free Press; 1989.
21 Brennan TA, Gawande A, Thomas E, Studdert DM. Accidental deaths, saved lives, and improved quality. N Engl J Med. 2005;353:1405–9.
22 Chassin MR, Galvin RW. The urgent need to improve health care quality. Institute of Medicine National Roundtable on Health Care Quality. JAMA. 1998;280:1000–05.
23 Crossing the Quality Chasm: a new health system for the 21st century. Washington, DC: National Academy Press; 2001.
24 Blumenthal D, Scheck A. Improving clinical practice: total quality management and the physician. San Francisco: Jossey-Bass; 1995.
25 McLaughlin CP, Kaluzny AD. Continuous quality improvement in health care: theory, implementation and applications. Gaithersburg, MD: Aspen Publishers; 1994.
26 Lighter DE, Fair DC. Principles and methods of quality management in health care. Gaithersburg, MD.: Aspen Publishers; 2000.
27 Donabedian A. An introduction to quality assurance in health care. New York: Oxford University Press; 2003.
29 Fernandopulle R, Ferris T, Epstein A et al. A research agenda for bridging the “quality chasm.” Health Aff (Millwood). 2003;22:178–90.
30 Frankel A, Graydon-Baker E, Neppl C, Gustafson M, Gandhi TK. Patient Safety Executive WalkRounds. Jt Comm J Qual Impr. 2003;29:16–26.
31 Frankel A, Grillo SP, Graydon-Baker E, et al. Patient Safety Leadership WalkRounds™ at Partners HealthCare: Learning from implementation. Jt Comm J Qual Impr. 2005;31:423–37.