Brian C. Drolet, M.D., Derrick A. Christopher, M.D., M.B.A., and Staci A. Fischer, M.D.
N Engl J Med 2012; 366:e35June 14, 2012
The role of the resident physician has evolved substantially over the past century. William Stewart Halsted, who is credited with developing the early system of graduate medical education in the United States, required 362 days per year of service from his residents. However, unlike Halsted’s trainees, who lived in the hospitals in which they worked, today’s first-year residents (interns) must adhere to various work restrictions, including spending no longer than 16 consecutive hours in the hospital.
National regulation of resident duty hours has occurred in response to the recognition that fatigue from extended work hours may result in errors and compromise patient care1 and may also lead to diminishing educational returns. Ultimately, the sensitive balance between patient care and education — given that residents are trainees — remains at the forefront of this discussion.
In 2003, the Accreditation Council for Graduate Medical Education (ACGME) implemented the first national regulation of work hours, establishing the 80-hour workweek. In 2008, the Institute of Medicine recommended additional limitations on work hours and an increase in direct supervision of residents to improve patient safety. It also suggested that if the ACGME and sponsoring institutions did not make changes, the Joint Commission or the Occupational Safety and Health Administration (OSHA) should perhaps step in to regulate residents’ hours. U.S. public opinion supported further regulation as well.2
In response to these pressures, the ACGME Duty Hours Task Force implemented the latest Common Program Requirements for Resident Duty Hours and Supervision in July 2011. Before the implementation of these requirements, two large national studies had shown that program directors and residents had mixed feelings about the potential impact of the proposed changes — fearing specifically that increased frequency of handoffs and loss of continuity might have a negative effect on patient care. Furthermore, there was concern that shortened duty hours, particularly for interns, would impair education and leave trainees less prepared for more senior, supervisory roles.3,4 To understand whether these concerns have become a reality during the year after the changes were adopted, we conducted a follow-up national survey of residents.
All 682 sponsoring institutions of ACGME-accredited residency programs in the United States were invited to participate in the survey, which had been approved by the institutional review board at Rhode Island Hospital. For each site, the designated institutional official (DIO) who is responsible for overseeing all ACGME programs at the institution was contacted by e-mail to invite trainees to participate in the survey. A complete list of e-mail addresses for all DIOs was collected from the publicly available ACGME database. We requested responses for agreement or refusal of institutional participation and e-mailed requests three times to encourage maximal participation. Once an institution had agreed, we e-mailed the survey hyperlink to the DIOs, along with an explanation of the survey, for distribution to all their residents between December 2011 and February 2012. Residents were not compelled to participate, and no incentives were offered for participation.
By February 2012, a total of 123 institutions in 41 states had agreed to participate; 33 DIOs had refused, citing survey fatigue as the primary reason. Data were collected over a period of 12 weeks to allow for study approval at each site and to maximize survey return. The resident populations at the participating institutions included 26,581 residents across a broad distribution of specialties. We received 6202 individual responses (23.3% response rate).
The questionnaire focused on the perceived effects of the 2011 regulations on the care of patients and residents’ education, quality of life, and supervision. In addition, we collected demographic data on sex, postgraduate year, program size, and specialty (see Table 1 Demographic Characteristics of Survey Respondents and Comparative ACGME Data for Resident Physicians in the United States.).
A majority of respondents (77.6%) were in their first 3 postgraduate years and in training programs in internal medicine (21.8%), family medicine (14.9%), or pediatrics (10.8%). Surgical fields (including obstetrics) were well represented, with 1316 respondents (21.2%). Overall, the demographic and specialty distribution of the sample paralleled national numbers published by the ACGME.5
To evaluate residents’ perspectives, we asked 12 questions requiring positive, neutral, or negative responses. We used the standard error of proportions to calculate two-sided confidence intervals with an alpha level of 0.001. Statistical significance was established for results with no overlap of the 99.9% confidence intervals (see Table 2 Perceived Effects of New ACGME Regulations.).
For many questions, residents reported no changes after the implementation of the new ACGME regulations. Although twice as many residents reported receiving better supervision as reported receiving worse supervision (17.9% vs. 8.3%), the availability of supervision was overwhelmingly thought to be unchanged (73.8%). This finding is interesting, given that interns are now required to have “immediately available” supervision, an important policy change necessitating the presence of a senior resident or attending physician within the hospital at all times.
Although 42.8% of residents reported no change in the quality of education, a nearly equal proportion (40.9%) reported worsened education — a far greater number than those who saw improvement (16.3%). Similarly, a majority (51.5%) of residents believed that preparation for more senior roles was worse. These perceptions may reflect the effects of the 16-hour-per-day limitation for first-year trainees and the sense that junior-level responsibilities have been shifted to senior residents (65.5%). Scheduling changes with increased “night float” duties may be reducing residents’ exposure to patients, availability for educational conferences, and continuity of care — an effect that is also reflected in a marked increase in transitions of care (72.0%).
Our study of residents last year also showed that half of residents (50.9%) anticipated positive changes in quality of life with the new regulations.3 However, a positive change seems to have been borne out only for interns (61.8%), whereas senior residents’ quality of life has suffered (49.7%) and, overall, residents claimed that their work schedules were worse (43.0%). Similarly, 50.1% of residents said that the amount of rest they obtained was unchanged, and 58.9% said the total number hours they worked was unchanged, despite the substantial limitation on interns.
The new ACGME regulations were proposed to improve three areas — patient care, resident education, and resident quality of life — by changing the quality and quantity of hours worked, as well as by increasing supervision at teaching hospitals. Yet our data show that many residents believe that these benefits have not been borne out in practice. Almost half of residents (48.4%) disapprove of the regulations — twice as many as those who approve of them (22.9%).
The survey results suggest several possible explanations for this dissatisfaction. First, residents are working the same number of hours with no change in the amount of rest they receive and with worse schedules than last year, which diminishes their overall quality of life. Second, residents believe that a chief goal of training — preparedness for transitioning from intern to senior resident, then ultimately to attending physician — is being delayed. Third, there has been no ostensible increase in available supervision or in the benefits for safety and education that would accompany this increased attendance. Finally, the frequency of handoffs has increased, reducing continuity of care and thereby negatively affecting the educational and emotional experience associated with a strong doctor–patient relationship.
We firmly believe that most residents support some form of duty-hour regulation and would not choose to revert to the Halstedian model. However, a one-size-fits-all approach may not be adequate or appropriate for all trainees and training programs. Ultimately, the intended and actual effects of the 2011 ACGME duty-hours requirements may not be aligned. Nevertheless, more study will be needed to quantify how safety and quality of care, as well as resident education, are being affected.
Disclosure forms provided by the authors are available with the full text of this article at NEJM.org.
This article (10.1056/NEJMp1202848) was published on May 30, 2012, at NEJM.org.
From the Departments of Plastic Surgery and General Surgery (B.C.D.) and the Department of Medicine (S.A.F.), Warren Alpert Medical School of Brown University; and Rhode Island Hospital (B.C.D., S.A.F) — both in Providence, RI; and the Department of Surgery, Vanderbilt University Medical Center, Nashville (D.A.C.).