In response to concerns that medical errors may cause as many as tens of thousands of deaths among patients in American hospitals each year1 as well as calls for the medical profession to either self-regulate or face federal legislation, the Accreditation Council for Graduate Medical Education (ACGME) implemented regulations that were effective July 1, 2003 and that affect duty hours for ACGME-accredited residency programs in all specialties.2,3 These regulations represent the largest national effort ever undertaken to reduce errors in teaching hospitals. The duty hours regulations require that when averaged over four weeks, residents must work no more than 80 hours per week and have one day in seven free. Residents may not work more than 24 continuous hours, with an additional 6 hours for education and transfer of care, and they must have inhouse call no more frequently than every third night and have at least 10 hours of rest between duty periods.3
These duty hours regulations have been controversial. Although there is evidence that sleep deprivation affects cognitive performance4–8 and that reducing work hours among interns lowers the rate of serious medical errors9 and attentional failures,10 a necessary by-product of the reform has been compromised continuity of care attributable to the increased number of patient hand-offs between residents.11–14 It is not clear that the duration of work shifts has been reduced sufficiently to reduce sleep deprivation and improve quality of care.15 Implementation has been costly.16 There are reports that some residents feel that they are being deprived of patient care and educational opportunities.17
To our knowledge, efforts to assess residents’ attitudes about duty hours reform have largely been limited to surveys of single sites or small groups of surgical program residents. Little is known about the broad effects of duty hours regulations on quality of care, the quality of resident education, or resident quality of life. Proponents of duty hours reform have argued that improvements in patient safety will result from reduced resident fatigue and improved systems of care,18 although there is little evidence to support this assertion.19 Opponents of duty hours reform have claimed that limiting duty hours will disrupt the continuity of patient care, which may negatively affect patient safety and lead to a downward trend in physician professionalism as a result of reduced responsibility and accountability.20
To better understand the perspectives of residents on the effects of the ACGME duty hours regulations, we undertook a multisite survey of geographically diverse medical and surgical residency programs. The survey focused on residents who were in training both before and after duty hours reform and solicited opinions about the impact of the duty hours rules on quality of care and medical errors, quality of resident education, and resident quality of life. We hypothesized that residents’ perceptions of medical errors related to fatigue would decrease, perceptions of medical errors related to discontinuity of care would increase, the perceived quality and frequency of resident educational experiences would decrease, and that residents’ perceptions of quality of life would improve. We also anticipated that the magnitude of these changes would be greater for the surgical resident respondents than for the medicine resident respondents because of a greater overall reduction in duty hours for surgical residents as a result of the ACGME regulations.
Nearly two years after duty hours regulations were implemented, we surveyed 137 third-year internal medicine residents at the Hospital of the University of Pennsylvania, Brigham and Women’s Hospital, and the Johns Hopkins University Hospital and 63 third- through fifth-year general surgery residents at the Hospital of the University of Pennsylvania, Case Western Reserve University, and the University of Texas at Houston. These programs constituted a convenience sample of geographically diverse and relatively large academic programs. Our study population was purposely limited to residents in the latter portion of their training to capture the perspectives of trainees who had experienced residency both before and after implementation of the duty hours regulations.
We designed a survey instrument based on literature review, the purported effects of the ACGME duty hours rules, and input from a focus group consisting of medicine and surgery residents at one of the five study sites. The instrument was reviewed by the focus group after the session, and no substantive changes were made. The instrument consisted of 19 Likert-scale questions designed to assess residents’ opinions in three overall content areas: quality of patient care and safety, residency education, and quality of resident life. Survey responses were rated on a scale of 1 to 5, where 1 indicates “decreased a lot” or “strongly disagree,” and 5 indicates “increased a lot” or “strongly agree.”
We distributed the survey in the spring of 2005 by e-mail. In the survey introduction, residents were informed of the purpose of the study and given a summary of the ACGME duty hours regulations. They were informed that completion of the survey was voluntary and that their responses would not be linked to their name or training program. All participants were informed that respondents to the survey would be eligible for entry into a lottery drawing for a $100 restaurant gift certificate at each study site. Nonrespondents received three subsequent e-mail requests, each at one-week intervals after the initial survey-distribution date. The institutional review boards at all five institutions approved the study. There were no external funding sources for the study.
We used Student’s t-test to compare survey responses between internal medicine and surgical residents, junior and senior surgical trainees, male and female respondents, and residents who reported sleeping less than, greater than, or equal to 42 hours in the preceding week. We also stratified responses by the various study sites and assessed the robustness of our results by excluding respondents who had not done clinical rotations since July 2003. Analyses were performed using SAS software version 6.12 (SAS institute, Cary, NC).
The overall response rate was 80%, with 159 out of 200 residents completing the survey. The response rates for medicine and surgery residents were 81% and 76%, respectively (Table 1). Each individual program site response rate was 59% or higher. Fifty-two percent of respondents were male and 48% were female. Ninety-three percent of residents were in a clinical year of training (as opposed to a research year) at the time of survey completion. Six respondents (4%) reported not having clinical responsibilities in a hospital after July 1, 2003. The mean number of sleep hours in one week before survey completion was 45.9, with slightly less self-reported sleep among surgical residents (44.1 hours) than medical residents (46.6 hours) (P = .08).
In the paragraphs that follow, we report all results as means of residents’ total responses per survey question. Medical and surgical residents’ feelings about quality of care and medical errors after implementation of the duty hours requirements were similar (Table 2). On a Likert scale where 1 indicated “decreased a lot,” 3 indicated “stayed the same,” and 5 indicated “increased a lot,” medical as well as surgical residents reported that quality of care had decreased a little (medicine = 2.5, surgery = 2.6) and that continuity of care had decreased a lot (medicine = 1.6, surgery = 1.9). Residents reported their perception that errors attributable to continuity of care had increased (medicine = 3.7, surgery = 3.5), but that errors related to resident fatigue had decreased (medicine = 2.5, surgery = 2.7) as a result of reduced duty hours.
Attitudes were mixed regarding various aspects of residency education after duty hours regulations (Table 3). Medical residents but not surgical residents reported that exposure to formal educational experiences such as conferences had decreased (medicine = 2.3, surgery = 3.3, P = .00), but assessments of opportunities for operative and nonoperative procedures were quite similar among medical and surgical residents (medicine = 2.6, surgery = 2.6, P = .84). Medical residents reported greater decreases in the available opportunities for bedside learning and teaching than surgical residents (medicine = 2.1, surgery = 2.5, P = .002). Residents’ reported performance of ancillary duties such as phlebotomy decreased slightly for both types of residents with the advent of duty hours reform (medicine = 2.5, surgery = 2.4). On a rating scale where 1 indicated “strongly disagree” and 5 indicated “strongly agree,” residents agreed with the statement that “work-hour rules have created a shift-work mentality among the housestaff” (medicine = 3.7, surgery = 3.9, P = .10). Neither medical nor surgical residents felt that the quality of program graduates had appreciably changed after the duty hours regulations (medicine = 3.2, surgery = 3.2, P = .77).
We also examined the effects of duty hours reform on relationships between residents and attendings (Table 3). Both types of residents felt that the reduction in duty hours had slightly reduced their opportunities for mentoring from individual attendings (medicine = 2.5, surgery = 2.7). They did not feel that duty hours regulations had increased or decreased their attendings’ workload (medicine = 3.1, surgery = 3.0), but many felt that their attendings had slightly negative attitudes about duty hours regulations (medicine = 2.6, surgery = 2.5, P = .50).
Residents did report that quality of life had improved substantially since implementation of duty hours regulations (Table 4), where 3 was a neutral response and 5 indicated “increased a lot”; reported improvements in quality of life were particularly notable among surgical trainees (medicine = 3.4, surgery = 4.2, P < .001). On a scale where 1 indicated “decreased a lot” and 3 was a neutral response, there were larger reported decreases in burnout among surgical than medical residents (medicine = 2.8, surgery = 2.4, P = .006). Surgical residents also experienced larger increases in the appeal of training in their specialty than medical residents (medicine = 2.8, surgery = 3.8, P < .0001). Overall, residents did not express any strong desire to have had the duty hours regulations in place when they were interns (medicine = 2.5, surgery = 3.1, P < .001).
Surgical residents from three postgraduate training levels were represented in this study. We compared responses from third-year residents with the combination of fourth- and fifth-year residents. Third-year residents trained as interns before duty hours reform and then for two subsequent years as residents under duty hours reform. Combined fourth- and fifth-year residents had more training experience (an intern year and one or two resident years) before duty hours reform. Among surgical residents, the combined group reported greater decreases in the quality of patient care than did third-year residents (2.4 versus 2.9, P = .02) and did not perceive that medical errors related to fatigue were decreasing to the same degree that their junior colleagues did (2.7 versus 2.3, P = .007). Otherwise, the responses given by fourth- and fifth-year residents were statistically similar to those given by third-year residents.
We also tested whether those residents who met criteria for chronic sleep deprivation responded differently from those who did not. The American Academy of Sleep Medicine defines sleep amounts of less than 42 hours per week as chronic sleep deprivation.7 Forty-four (27%) residents in our study reported having slept less than 42 hours in the preceding week, and 115 (72%) reported having slept more than 42 hours. When we analyzed data comparing responses among residents who reported having slept less than 42 hours versus more than 42 hours in the week before the survey, minimal differences emerged in residents’ attitudes and opinions on our survey items. We also compared responses between residents in the bottom quartile of self-reported sleep (Q1 ≤ 40 hours of sleep per week) with responses from the uppermost quartile (Q4 ≥ 50 hours of sleep per week) and found that responses were not significantly different for 16 of the 18 questions asked. The two questions for which there were significant differences in responses were “Attendings’ attitudes have been positive” (Q1 group = 2.3, Q4 group = 2.8; P = .01), and “The ACGME has been effective in enforcing compliance with these work-hour rules” (Q1 group 2.8, Q4 group 3.4; P = .02).
With respect to compliance, residents neither agreed nor disagreed with the assertion that the ACGME has been effective in monitoring compliance (medicine = 3.2, surgery = 3.2, where 3 equals a neutral response). Residents were slightly opposed to the notion of the federal government taking over enforcement of these regulations from the ACGME (medicine = 2.4, surgery = 2.5).
We found no differences in responses by study site. The results were robust to exclusion of the 4% of respondents who had not done clinical rotations since July 2003.
The findings of this multisite survey suggest that residents who worked in teaching hospitals both before and after the institution of the ACGME duty hours regulations believe that these regulations have improved quality of care because of reduced resident fatigue, have worsened quality of care because of reduced continuity, and have had some substantial benefits in improving residents’ quality of life, particularly among surgical trainees. Notably, neither surgical nor medical residents believed that the quality of physicians trained under duty hours reform has suffered, although they did report that opportunities for educational experiences have diminished.
We believe that this study reports on the largest and most diverse group of residents to comment on this topic thus far, because other studies to date have been single-center or single-specialty research surveys focusing on surgical residents.21–29 This survey achieved a response rate of 80%, making this a highly representative sample. The timing of our report is important because current internal medicine residents and the majority of current surgical residents did not experience residency training before duty hours reform, making within-subject comparisons increasingly infeasible.
Given the recent public attention to human errors in medicine and the national campaigns to improve the safety of patients within our hospitals, our findings are important. Although we did not measure clinical markers of patient safety, the collective opinions about medical errors from a representative sample of residents in academic medical centers provide the medical community and the public with some insight into the possible effects of the work-hour rules on patient care. The residents’ opinions we collected suggest that errors related to fatigue may have been replaced with errors related to discontinuity of care as a result of duty hours reform. Although residents may be less fatigued at work and thus perform at a higher level, the redesign of residency training models after duty hours reform has forced teaching hospitals to care for the same number of patients with fewer resident physician-hours. Alternative solutions to the reduction in resident hours have been found through increased use of float models and physician extenders,30 both of which introduce another practitioner to the care of a single patient and create a need for more patient hand-offs. Residents themselves have highlighted continuity of care as the area that has suffered the most as a result of the ACGME duty hours rules. Given the general lack of good systems for sign-outs,31 this is a major concern for patient safety. The effects of hand-offs on patient safety need to be studied further, and novel methods to reduce these discontinuities in care should be pursued. The residents’ opinions that we report—an increased error rate attributable to increased hand-offs and a decreased error rate attributable to less fatigue—are congruent with previous smaller survey studies and raise the important question of whether the ACGME duty hours regulations will actually accomplish the goal of reducing medical errors.
The residents in our sample reported sleeping an average of 45 hours per week—just slightly greater than 42 hours, which is the cutoff for chronic sleep deprivation. Other industries suggest that for every one-hour reduction in time at work, employees will sleep perhaps one fourth to one third of this time. This might explain why duty hours reform has not resulted in significantly more hours of sleep per week. It is notable that even after duty hours reform, 27% of residents in our study reported sleeping less than 42 hours per week and were therefore, by definition, still chronically sleep deprived. This finding challenges the notion that fatigue-related errors will decrease as a result of duty hours reform.
Our results corroborate the findings of other studies indicating that senior surgical residents have more negative attitudes about the duty hours regulations than junior residents, both in terms of likely negative effects on quality of care and in terms of educational opportunities.21,25–27 Although we did not objectively quantify weekly duty hours in our study, it is possible that unexpected increases in workload as a result of duty hours reform during senior residency years could lead to negative attitudes about duty hours reform among senior residents. In our study, junior and senior surgical residents agreed that their quality of life and the appeal of training in their specialty improved after duty hours reform. We were unable to determine whether these improvements offset the perceived decline in patient care and continuity in the opinions of surgical trainees.
The effects of duty hours reform on residents’ educational experiences have been substantial. Medical residents in this survey noted a large reduction (overall 2.1; range 1.9 to 2.5) in bedside teaching, perhaps because their prior training experiences had emphasized this patient-centered approach to learning. Opportunities for mentoring from attendings were also reported by medical as well as surgical residents to be reduced, presumably because of the effects of reduced time in the hospital, where these relationships are fostered. Both of these findings highlight the unintended educational consequences of duty hours reform that will need to be addressed as residency programs adapt their educational programs to meet the regulatory requirements. Few would argue with the statement that direct contact with attendings plays a pivotal role in the clinical training and professional development of residents. Creative ways to preserve this training model should be emphasized in future efforts to enhance duty hours reform.
The opinions of surgical and medical residents about the effects of duty hours reform on education and quality of care were surprisingly similar. We hypothesized that surgical residents would describe much greater declines in education compared with medical residents, because of a greater reduction in hours spent in the hospital, where procedures and other educational activities occur. In fact, medicine residents reported significantly greater declines in two aspects of education—exposure to formal lectures and opportunities for bedside learning—even though the Internal Medicine Residency Review Committee has recommended an 80-hour workweek since July 2003. One possible explanation for this discrepancy is that surgical programs have done a better job of eliminating the nonphysician tasks previously done by residents, such that the net time in the operating room, which is a cornerstone of surgical education, has not decreased. This finding also challenges whether or not internal medicine programs were truly compliant with work-hour restrictions before 2003. Finally, this finding confirms the belief that both groups of residents had valuable educational opportunities in the perceived “excessive” hours worked before duty hours reform. Both cohorts reported large declines in the quality and continuity of patient care after duty hours reform, highlighting how medical and surgical residents implicitly believe that their long work hours before duty hours reform had contributed positively to quality of care.
We believe that our findings are generalizable to other residency programs in academic medical centers across the country. Our response rate of 80% is high, so it is unlikely that the opinions of the nonresponders differed in any significant way from those of the responders. Through the inclusion of both procedural-based (surgery) and nonprocedural-based (medicine) residents, we can generalize our results to other training programs such as obstetrics and gynecology, surgical subspecialties, and neurology. We recognize that these findings may not represent the opinions of residents in smaller community-based residency training programs. Residents in these programs may have different duty hours, ancillary service requirements, and patient volumes than academic programs and, therefore, may not have experienced large changes in their residencies after duty hours reform. Another limitation of our research is that objective measurements of actual hours worked were unknown for our cohort. It is possible that residents who worked more hours in general would have different attitudes than residents who worked less in general because of their personal time-management and organizational skills. Our research is also susceptible to recall bias. In our retrospective survey design, residents were asked to reflect upon many experiences that had occurred two to four years before the survey. Residents with particularly negative or positive experiences before work-hour reform may have been more likely to selectively remember and report these experiences when asked to comment on the effects of duty hours reform. In addition, all medicine residents in our study were interns before duty hours reform, and all surgery residents were either interns or in an early training year (second or third year). We recognize that the attitudes and opinions of a trainee may be inherently different during earlier training years when compared with later training years because of differences in patient-care responsibilities, number of hours worked, educational experiences, and the accumulation of coping and professional skills. It may have been difficult for the survey respondents to distinguish changes that had occurred as a result of their shift in training year from changes that had occurred as a result of duty hours reform. Despite these limitations, we believe that this study has documented several important effects, both positive and negative, of the duty hours regulations for the public, the ACGME, and the profession as a whole.
This study reveals that a large, diverse group of residents who trained before and after duty hours reform have significant concerns about the effects of the reform on patient care and education. These concerns persist in the face of self-reported improvements in quality of life and perceived reductions in errors related to fatigue. The residents’ opinions substantiate the concerns that medical educators have regarding the quality of educational training programs and that many faculty and staff in teaching hospitals have about the quality of patient care under duty hours reform. Future trainees will be unable to provide us with these and other insights into duty hours reform because they will have no basis for direct comparison.
If these findings are corroborated by objective data on outcomes, the academic community will be challenged to strongly consider modification of the existing duty hours limits and the structure of its training programs to preserve the educational experiences of residents and to improve patient care. Some restructuring occurred concurrently with the advent of duty hours reform, but the opinions reported in this study come two years later, thus informing the medical profession that further improvements are needed. The boundaries that exist around traditional modes of graduate medical education need to be challenged. Fragmentation and discontinuity of patient care necessitates intense focus and targeted improvements in hospital care systems and health care communication. The ACGME core competency of systems-based practice requires residents to gain experience and demonstrate proficiency in systems thinking and practice. Further instruction and assessment of this competency will be vital to patient safety. Through systems teaching, residents may begin to recognize and understand the human factors that contribute to medical errors and, therefore, be more inclined to report their errors and suggest constructive changes.32 We must work within the confines of an 80-hour workweek to develop the most effective and efficient care delivery and educational system possible. Duty hours regulations are here to stay in some fashion, and we cannot afford to simply sit back and yearn for more hours to teach our residents and care for our patients.
The ACGME duty hours regulations represent a major effort toward reducing medical errors. From the standpoint of public health, the effects on outcomes and rates of medical errors are the main issues that need to be resolved by empirical studies using secondary data. The improvements in resident quality of life shown by our findings need to be weighed against the perceived reductions in continuity of care and educational opportunities reported by residents. Ultimately, the success and sustainability of such efforts will depend on assessment by the public and by academic medicine of the evidence of whether duty hours reform has achieved its primary goal of improving patient safety.
The authors thank the Veterans Affairs Health Services Research and Development Service (VA HSR&D) for funding support; Judy Shea, PhD, for her contributions to the survey design; and Kotaro Fujita for his technical assistance with the survey website. Dr. Volpp is a VA HSR&D Career Development Awardee and a Doris Duke Foundation Clinical Scientist Development Awardee.
1 Kohn LT, Corrigan JM, Donaldson MS. To Err Is Human: Building a Safer Health System. Washington, DC: National Academy Press; 2000.
2 Philibert I, Friedman P, Williams WT. ACGME work group on resident duty hours, Accreditation Council for Graduate Medical Education. New requirements for resident duty hours. JAMA. 2002;288:1112–1114.
4 Buysse DJ, Barzansky B, Dinges D, et al. Sleep, fatigue, and medical training: setting an agenda for optimal learning and patient care. Sleep. 2003;26:218–225.
5 Rogers NL, Dorrian J, Dinges DF. Sleep, waking and neurobehavioral performance. Front Biosc. 2003;8:s1056–1067.
6 Dinges DF, Pack F, Williams K, et al. Cumulative sleepiness, mood disturbance, and psychomotor vigilance performance decrements during a week of sleep restricted to 4–5 hours per night. Sleep. 1997;28:267–277.
7 Van Dongen HPA, Maisllin G, Mullington JM, Dinges DF. The cumulative cost of additional wakefulness: dose-response effects on neurobehavioral functions and sleep physiology from chronic sleep restriction and total sleep deprivation. Sleep. 2003;26:117–126.
8 Dongan HPA, Dinges DF. Sleep debt and cumulative excess wakefulness. Sleep. 2003;26:249.
9 Landigran 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.
10 Lockley SW, Cronin JW, Evans EE, et al. Effect of reducing interns’ weekly work hours on sleep and attentional failure. N Engl J Med. 2004;351:1829–1837.
11 Drazen JM. Awake and informed. N Engl J Med. 2004 351:1884.
12 Mukherjee S. A precarious exchange. N Engl J Med. 2004;351 18:1822–1824.
14 Ofri D. Residency regulations—resisting our reflexes. N Engl J Med. 2004;351 18:1824–1826.
15 fsWylie CD. Sleep, science, and policy change. N Engl J Med. 2004;352 2:196–197.
17 Darves B. Work hour rules rile educators—and residents. ACP Observer. 2003;23:1.
18 Skeff KM, Ezeji-Okoye, S Pompeo P, Rockson S. Benefits of resident work hours regulation. Ann Intern Med. 2004;140:816–817.
19 Fletcher KE, Davis SQ, Underwood W, Mangrulkar RS, McMahon LF Jr, Saint S. Systematic review: effects of resident work hours on patient safety. Ann Intern Med. 2004;141:851–857.
20 Charap M. Reducing resident work hours: unproven assumptions and unforeseen outcomes. Ann Intern Med. 2004;140:814–815.
21 Kort KC, Pavone LA, Jensen E, Haque E, Newman N, Kittur D. Resident perceptions of the impact of work-hour restrictions on health care delivery and surgical education: time for transformational change. Surgery. 2004;136:861–871.
22 Zuckerman JD, Kubiak EN, Immerman I, DiCesare P. The early effects of Code 405 work rules on attitudes of orthopaedic residents and attending surgeons. J Bone Joint Surg. 2005;87:903–908.
23 Mendoza KA, Britt LD. Resident operative experience during the transition to work-hour reform. Arch Surg. 2005;140:137–145.
24 Gelfand DV, Podnos YD, Carmichael JC, Saltzman DJ, Wilson SE, Williams RA. Effect of the 80-hour workweek on resident burnout. Arch Surg. 2004;139:933–940.
25 Basu CB, Chen L, Hollier L, Shenaq S. The effect of the Accreditation Council for Graduate Medical Education duty hours policy on plastic surgery resident education and patient care: an outcomes study. Plast Reconstr Surg. 2004;114:1878–1887.
26 Niederee MJ, Knudtson JL, Byrnes MC, Helmer SD, Smith RS. A survey of residents and faculty regarding work hour limitations in surgical training programs. Arch Surg. 2003;138:663–671.
27 Chandra RK. The resident 80-hour work week: how has it affected surgical specilaties? Laryngoscope. 2004;114:1394–1398.
28 Irani JL, Mello MM, Ashley SW, Whang EE, Zinner MJ, Breen E. Surgical residents’ perceptions of the effects of the ACGME duty hour requirements 1 year after implementation. Surgery. 2005;138:246–253.
29 Barden CB, Specht MC, McCarter MD, Daly JM, Fahey TJ 3rd. Effects of limited work hours on surgical training. J Am Coll Surg. 2002;195:531–538.
30 Todd B, Resnick A, Stuhlemmer R, Morris J, Mullen J. Challenges of the 80-hour resident work rules: collaboration between surgeons and nonphysician practitioners. Surg Clin N Am. 2004;84:1573–1586.
31 Volpp K, Grande D. Residents’ suggestions for reducing errors in teaching hospitals. N Engl J Med. 2003;348:851–855.
32 Wu AW, Folkman S, McPhee SJ, Lo B. Do house officers learn from their mistakes? JAMA. 1991;265:2089–2094.