It’s 4:30 pm on Monday afternoon, and a resident is called to the surgical ICU to see an elderly postoperative patient who underwent an open right radical nephrectomy (removal of a kidney) earlier in the day. The ICU nurse reports that the patient’s oxygen saturation has decreased moderately and that the patient has been “fighting the ventilator.” The resident examines the patient briefly and orders intravenous sedation with morphine and diazepam, and continued mechanical ventilation. Thirty minutes later, the patient experiences profound hypotension and hypoxemia, and the nurse calls the code team. A brief but careful examination reveals absent breath sounds throughout the right hemithorax. The team inserts a large-bore catheter into the second intercostal space, which is accompanied by the sudden egress of air and followed rapidly by improved oxygenation and blood pressure. A chest radiograph confirms a right pneumothorax; a chest tube is inserted, and the patient recovers uneventfully.
In a former era, this event would have been discussed at a departmental case conference, and likely, the resident would have been rebuked in some manner, either by public embarrassment or by the whispered comments of attendings and fellow residents. This would have been viewed as an active error, and the resident, the final common pathway in this care process, would have been blamed for poor knowledge, judgment, or skills. Today, this event would trigger a root-cause analysis performed by an investigative team of fellow residents, attendings, nurses, and other relevant personnel. The team would look not only at the immediate event (the active error) but also at system-related precursors that might have contributed to the event. These precursors, which are often overlooked because they are hidden in behind-the-scenes processes, are deemed “latent errors.” This more thorough analysis would reveal that the responding resident had been in the hospital since 8:00 am the prior morning (32+ hours) and had slept a total of three hours in interrupted segments, none longer than 45 minutes. The team would note that the resident had been both fatigued and hurrying to complete his last new admission of the day before going home, and they would conclude that the system of assignments that had created this situation was a major contributing factor to this near-fatal event.
Examples such as this one, which occurred all too frequently, contributed to the implementation of limits on resident duty hours. During the past decade, the Institute of Medicine and several other national organizations have brought a new emphasis to the issues of quality and safety in health care, including error reduction. Eliminating errors requires a focus on the systems and processes of care delivery, because such processes can affect human performance. The resulting public awareness and governmental concern about both active and latent medical errors, including the detrimental effects of fatigue on performance, led to increased scrutiny of the work assignments of residents. Educators noted that fatigue also affected participation in formal education (e.g., teaching conferences). Further, there was increasing resident dissatisfaction with the workloads that were required in some training programs, and that some programs were seeing fewer applicants, whereas other programs with more controlled work schedules were seeing more applicants. Thus, there were increasing pressures from several directions, both within health care and externally, that culminated in the duty hours restrictions that were implemented by the Accreditation Council for Graduate Medical Education (ACGME) in July 2003.
The implementation of duty hours restrictions was accompanied by both high expectations and great concerns, depending on the viewpoint of the beholder. Many anticipated that resident duty hours restrictions would improve the quality and safety of care by minimizing the detrimental effects of fatigue on human performance. Their reasoning was supported by evidence that fatigue contributes to errors both in health care1 and in personal activities (such as auto accidents).2 Others expected that eliminating sleep deprivation and excessive work hours would lead to increased resident satisfaction and improved attitudes, which would be reflected in enhanced quality of relationships with others3 and improved professionalism among trainees. Education also would be enhanced by providing more time for reading and self-study, fostering the concept that clinicians must be committed to life-long study if they are to remain current in their knowledge base and thus capable of practicing current evidence based medicine.
There were doubters as well. Some residents and teaching physicians were concerned about the educational consequences of reduced clinical contact hours. Others were concerned that the fundamental clinical and educational principle of continuity of care would be lost or at least eroded, and that more frequent “hand-offs” might result in more clinical errors. Hospital or practice administrators were concerned about the potential for increased costs resulting from the need to “backfill” lost resident services. Attendings were concerned about many aspects of these changes, including the potential for increases in their own workloads. Undoubtedly, there was the subtle but real feeling that “what was good enough for me is good enough for them.” Some lamented the loss of the total-emersion residency experience that would serve as a forging process to temper the mind and body to create a finely honed clinician. As one who has served as a department chair and program director and, subsequently, as senior vice-president and corporate chief medical officer for a large urban academic health system, I’ve had firsthand experience with the various viewpoints and with the challenges and consequences of implementing these rules.
The early results, as indicated in the seven articles about resident duty hours in this issue of the journal, are mixed. The accompanying articles examine resident opinions, attending opinions, and residency director opinions, as well as various approaches to accomplishing compliance with the work rules, the implications for undergraduate medical education, and the regulations and trends in other countries. Both individually and collectively, they contain information that advances our understanding of the current situation, and they guide us toward future explorations that will inform and enhance the evolving approaches to duty hours.
Resident perceptions, at least in some programs, are perhaps less positive than many might have expected. In a multicenter survey of both medical and surgical residents (six training programs in five geographically distributed academic medical centers), Myers et al4 report the opinions of residents who were in training programs both before and after the implementation of resident duty hours restrictions. Residents perceived that their quality of life had improved and that their training program was more appealing after duty hours had been reduced, but they expressed concerns about the resulting quality of care and educational opportunities. They estimated that there were fewer errors resulting from fatigue, but that there were more attributable to the loss of continuity of care and more frequent hand-offs. Further, they opined that bedside teaching and opportunities for mentoring by faculty had decreased after the implementation of duty hours reform. They also observed that duty hours had led to a shift-work mentality among trainees.
Jagsi et al5 report more positive resident perceptions in their institutions (two large teaching hospitals in Boston). In general, the residents in those internal medicine and surgery programs that had previously experienced particularly heavy workloads responded more positively to the duty hours interventions than did their colleagues who had been less affected by the new rules. Indeed, the residents in formerly heavy-workload programs noted that fatigue was now considerably less likely to negatively impact their quality of care, their ability to provide psychosocial support for patients, their ability to learn, their interactions with coworkers, the safety of their patients, or their own personal satisfaction and well-being. However, these benefits were not without potential consequences. For example, direct patient-care hours decreased, but the residents still spent approximately one third of their time in ancillary work (so-called “scut work”), and even with the new approaches, approximately one in six residents reported having worked more than 80 hours in the past week (recall, though, that this does not necessarily imply noncompliance with the ACGME rules, because the 80-hour limitation is averaged over four consecutive weeks). Further, the changes had come at the combined cost of $3 to $4 million annually, to provide additional personnel (residents, midlevel practitioners, fellows, and attendings).
Attending views differ significantly from resident views on the overall impact of resident duty hours reform. For example, 69% of otolaryngology faculty reported that implementation of resident duty hours restrictions had had a negative effect on resident education, whereas only 31% of residents felt their education had been negatively affected.6
Family medicine program directors also expressed significant concerns about the consequences of resident duty hours reform (see Peterson et al7). These program directors expressed concerns about decreases in professionalism among residents because of a shift mentality, the loss of continuity of care, a decrease in opportunities in formal education, and the creation of expectations that might be inconsistent with the realities of practice. Indeed, some felt the residency should be lengthened to compensate for the decrease in educational opportunities. Only 4% of program directors felt the residents fared better overall under the new duty hours rules. Clearly, many programs found it hard to comply with the new rules, and approximately half reported increased patient-care duties for attendings, suggesting that teaching physicians were inheriting some of the duties that previously had been handled by house officers. Half also reported decreased formal educational opportunities and decreased exposure to specialty rotations and continuity clinics. The latter is particularly ironic because continuity clinics are intended to facilitate the family medicine residents’ exposure to long-term longitudinal care.
There are, however, encouraging signs in that some programs have implemented approaches that are attractive to both residents and attendings. Ogden et al8 describe the consequences of initial efforts to meet the 80-hour work rule, efforts that negatively affected resident education and patient care without improving overall resident well-being or fully meeting the ACGME requirements in a few cases. This led to a thoughtful, comprehensive review of those initial results and a commitment to redesign those efforts to achieve better results. Teams of educational leaders and residents designed new processes that were intended to meet specific educational and clinical goals while decreasing the potential for clinical errors, and then implemented these processes and evaluated their results (a process that conceptually follows the “Plan–Do–Check–Act” approach, which is recommended by the Joint Commission on Accreditation of Health care Organizations for revising other health care processes). The results of this redesign have been considerably more positive and encouraging. However, the revised approach required the hiring of four additional hospitalists, and the authors observed a greater-than-expected turnover among hospitalists, implying that the result had a negative impact on attendings.
Lundberg et al9 report their experiences with an approach that is conceptually similar, although it differs in the details. Like Ogden et al,8 they implemented an approach that was intended to meet the goals of clinical care, resident education, and resident well-being, and they seem to have accomplished this by creating clinical teams that include nurse practitioners who work in concert with residents and attendings on each service. Faculty as well as residents have expressed satisfaction with the nurse practitioner team model, and they have high regard for their nursing colleagues. However, again, the solution was achieved by hiring additional personnel, thus incurring increased overall costs to the delivery system.
Most of the attention surrounding resident duty hours has been focused on resident well-being, graduate medical education, clinical care, and costs. However, alterations in resident duties inevitably affect undergraduate medical education as well. Kogan et al10 surveyed internal medicine clerkship directors to assess this impact. Two thirds of program directors disagreed that resident duty hours reform had positively impacted the educational experience in the core clerkship, whereas only 2% agreed with this assertion. Two thirds disagreed with the assertion that there was now more time for bedside teaching. Only 1% felt the residents had more time to teach, and only 6% felt residents had improved attitudes toward teaching. Finally, over 60% of clerkship directors felt that medical students were now learning that the number of hours worked takes precedence over patient care, again indicating a decline in this component of professionalism. Clerkship directors again noted an increase in a “shift mentality” among trainees and students, and a decrease in the continuity of care. In all, the learning environment for students seems to be more fragmented after implementation of duty hours restrictions.
Reform of the former graduate medical educational environment was widely perceived as necessary; few could justify the workloads and duty hours that residents had experienced in some programs, and clearly neither the public, the residents, nor the government would or could allow the situation to persist. However, the observations reported here, when viewed collectively, suggest that far more than simple control of duty hours will be required to achieve the goals of clinical excellence, educational excellence, resident well-being, and professionalism.
Collectively, the observations made in these seven articles are a “mixed bag”; a number of positives are balanced by at least an equal number of negatives, when measured against the four goals noted above. A few general conclusions seem evident: residents observe that their quality of life and satisfaction with their programs have improved, but many residents and attendings believe that quality of care has not improved and may have suffered from the loss of continuity of care and the increased frequency of hand-offs. Similarly, many residents and teaching physicians opine that the quality of graduate education has not improved, and likely has suffered, as has the quality of undergraduate medical education. Further, there seems to be a widespread perception that the resultant shift-work mentality has negatively affected professionalism, although this may be balanced to some (unknown) extent by the benefits that improved resident attitudes bring to their interpersonal relations with patients and others. Finally, it seems that those who have met these challenges have been unable to do this without adding additional clinical personnel and their associated costs.
Together, these results imply that there is considerably more work to be done to meet the current ACGME requirements while achieving the goals outlined above. Further, the observations of Woodrow et al11 suggest that the current resident duty hours limitations may be just the initial step in a continuum of attention to both duty hours and other factors (such as stress and fragmented workflows) that may influence human performance.
Are we there yet? Based on the reports in this issue of the journal, the answer is a clear “no.” It’s too early to tell whether the changes implemented here, even the successful ones, will have a sustained impact, or whether they simply reflect the enthusiasm associated with change and the expected beneficial results (sometimes referred to as the Pygmalion effect). Much work and research remain to be completed before we reach a new steady state that addresses the overarching goals of clinical excellence, educational excellence, resident well-being, and professionalism. Additional research is required to address a host of remaining questions, among them: What is the impact of resident duty hours restrictions on objective measures of education (e.g., in-training examination scores, specialty board pass rates, etc.)? What are the measured impacts on medical error rates? What are the most effective approaches to foster effective hand-offs? How can workflows be streamlined and ancillary duties be reduced, to increase the ratio of educational and clinical time to overall work time? What is the role for new approaches to learning in graduate medical education (e.g., podcasts, computer-assisted education, simulation, team-based learning)? What are the net financial impacts of resident duty hours? (The latter will require sophisticated financial approaches that consider the increased costs of additional personnel if required, and also factors such as error rates, length of stay, or redundant and/or unnecessary diagnostic testing, any of which might be either increased or decreased depending on how resident duty hours limitations are implemented in each institution.)
Personally, I’m optimistic that we can achieve educational excellence, clinical excellence, resident well-being, and professionalism in an era of reduced duty hours, but I believe it will take reform that goes considerably beyond what most have done so far, and perhaps beyond the knowledge and skill base of most teaching physicians. Real reform will likely require the careful analysis of resident workflow, akin to that reported by Gabow et al12 in a recent report in this journal. Like those authors, I suspect that most physician–educators will require support and involvement from those who possess expertise in systems engineering, moving beyond small incremental improvements to achieve the kinds of fundamental changes that will be required to achieve our goals. Systems engineering approaches can help us focus on the truly desired endpoints and identify approaches that will eliminate distractions, fragmentation, and lack of will to address change in long-held traditions. The engagement of those outside our disciplines will bring a fresh look at our processes, one that is not constrained by fear of personal consequences (not infrequently, novel ideas are squelched by those who are personally vested in a process and concerned about “what does this mean for me?”). This strategy is also consistent with the conclusions of a recent report from the National Academy of Sciences regarding future directions for health care delivery reform.13 My hunch is that combined approaches with our systems engineering and nursing colleagues will lead to redesign of workflow, including resident workflow, and the formation of relationship-centered care teams. The organization of such teams and their benefits in selected applications are impressive.14 (The report by Lundberg et al9 in this issue of the journal is consistent with this team concept.) I’m confident that continued research, implementation, evaluation, and redesign can achieve our goals; I can’t wait to see it happen!
1 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. 2004;351:1838–1848.
2 Barger LK, Cade BE, Ayas NT, et al. Extended work shifts and the risk of motor vehicle crashes among interns. N Engl J Med. 2005;352:125–134.
3 Shanafelt TD, West C, Zhao X, et al. Relationship between increased personal well-being and enhanced empathy among internal medicine residents. J Gen Intern Med. 2005;20:559–564.
4 Myers JS, Bellini LM, Morris JB, et al. Internal medicine and general surgery residents’ attitudes about the ACGME duty hours regulations: a multicenter study. Acad Med. 2006;81:1052–1058.
5 Jagsi R, Shapiro J, Weissman JS, Dorer DJ, Weinstein DF. The educational impact of ACGME limits on resident and fellow duty hours: a pre–post survey study. Acad Med. 2006;81:1059–1068.
6 Brunworth JD, Sindwani R. Impact of duty hour restrictions on otolaryngology training: divergent resident and faculty perspectives. Laryngoscope. 2006;116:1127–1130.
7 Peterson LE, Johnson H, Pugno PA, Bazemore A, Phillips RL. Training on the clock: family medicine residency directors’ responses to resident duty hours reform. Acad Med. 2006;81:1032–1037.
8 Ogden PE, Sibbitt S, Howell M, et al. Complying with ACGME resident duty hours restrictions: restructuring the 80-hour work week to enhance education and patient safety at Texas A&M/Scott & White Memorial Hospital. Acad Med. 2006;81:1026–1031.
9 Lundberg S, Wali S, Thomas P, Cope D. Attaining resident duty hours compliance: the acute care nurse practitioners program at Olive View–UCLA Medical Center. Acad Med. 2006;81:1021–1025.
10 Kogan JR, Pinto-Powell R, Brown LA, Hemmer P, Bellini LM, Peltier D. The impact of resident duty hour reform on the internal medicine core clerkship: results from the clerkship directors in internal medicine survey. Acad Med. 2006;81:1038–1044.
11 Woodrow SI, Segouin C, Armbruster J, Hamstra SJ, Hodges B. Duty hours reforms in the United States, France, and Canada: is it time to refocus our attention on education? Acad Med. 2006;81:1045–1051.
12 Gabow P, Karkhanis A, Knight A, Dixon P, Eisert S, Albert R. Observations of residents’ work activities for 24 consecutive hours: implications for workflow redesign. Acad Med. 2006;81:766–769.
13 Reid PP, Compton WD, Grossman JH, Fanjiang G, eds. Building a Better Delivery System: A New Entineering/Health Care Partnership. Washington, DC: National Academies Press, 2005.
14 Safran DG, Miller W, Beckman H. Organizational dimensions of relationship-centered care: theory, evidence and practice. J Gen Intern Med. 2006;21:S19–S15.