Twenty-five years ago, the death of a young woman, Libby Zion, in a New York hospital raised the issue of sleep deprivation, fatigue, and supervision of physicians-in-training (residents and fellows) to a national (if not international) stage. Concern about sleep deprivation and fatigue of trainees on clinical assignments has been long-standing, even pre-dating the Libby Zion case. In 2003, the Accreditation Council for Graduate Medical Education (ACGME) implemented detailed regulations to address some of these concerns. In addition to the requirement that all trainees and teaching-program faculty be educated about monitoring for evidence of sleep deprivation and fatigue (and intervening when necessary), the ACGME currently defines limits on trainee shift length, days off, and work hours per week (see Table below).1
Implementation of the 2003 ACGME regulations forced many training programs to redesign their curricula, clinical rotation structure, and approach to providing medical care to patients around the clock. To comply with these regulations, many training programs have developed “day float” and “night float” rotations for their trainees, created non-covered patient care units, and used hospitalists, nocturnists (hospitalists that work the overnight hours), and physician extenders (i.e., physician assistants and nurse practitioners). Although not typically a direct problem for hematology-related fellowship programs, the implemented changes in the delivery of patient care often have led to indirect effects on our hematology-related training programs. In addition, compliance with these duty- hour restrictions has been difficult for some programs and for some specialties.
As a result of implementation of these guidelines and despite the above-noted interventions, concern has been raised about the effects of these duty-hour policies on patient safety, patient care, and trainee education. Although intended to improve patient safety, implementation of these regulations has led to the identification of “new” risks to patient safety that occur with the resulting increase in the number of “sign-outs” involved with the transfer of patient-care responsibilities to the caregiver for the next shift.2,3 Fletcher and colleagues found in a systemic review of the available literature that the “evidence on patient safety is insufficient to inform the process of reducing resident work hours.”4 There are other issues related to more trainees being involved with a given patient’s care, including the impact on the trainees’ sense of responsibility and “ownership” for a given patient’s care, the ability of trainees to observe the course of a patient’s disease and response to therapy and the medical matters that arise over time, and the impact of trainees not being available to attend educational sessions.
Table. Comparison of IOM Committee Adjustments to Current ACGME Duty-Hour Limits
| . | 2003 ACGME Duty-Hour Limits . | IOM Recommendation . |
|---|---|---|
| Maximum hours of work per week | 80 hours, averaged over 4 weeks | No change |
| Maximum shift length | 30 hours (admitting patients up to 24 hours then 6 additional hours for transitional and educational activities) | 30 hours (admitting patients for up to 16 hours, plus 5-hour protected sleep period between 10 p.m. and 8 a.m. with the remaining hours for transition and educational activities)16 hours with no protected sleep period |
| Maximum in-hospital on-call frequency | Every third night, on average | Every third night, no averaging |
| Minimum time off between scheduled shifts | 10 hours after shift length | 10 hours after day shift12 hours after night shift14 hours after any extended duty period of 30 hours and not return until 6 a.m. of next day |
| Maximum frequency of in-hospital night shifts | Not addressed | 4 night maximum; 48 hours off after 3 or 4 nights of consecutive duty |
| Mandatory time off duty | 4 days off per month1 day (24 hours) off per week, averaged over 4 weeks | 5 days off per month1 day (24 hours) off per week, no averagingOne 48-hour period off per month |
| Moonlighting | Internal moonlighting is counted against 80-hour weekly limit | Internal and external moonlighting is counted against 80-hour weekly limitAll other duty-hour limits apply to moonlighting in combination with scheduled work |
| Limit on hours for exceptions | 88 hours for select programs with a sound educational rationale | No change |
| Emergency room limits | 12-hour shift limit, at least an equivalent period of time off between shifts; 60-hour workweek with additional 12 hours for education | No change |
| . | 2003 ACGME Duty-Hour Limits . | IOM Recommendation . |
|---|---|---|
| Maximum hours of work per week | 80 hours, averaged over 4 weeks | No change |
| Maximum shift length | 30 hours (admitting patients up to 24 hours then 6 additional hours for transitional and educational activities) | 30 hours (admitting patients for up to 16 hours, plus 5-hour protected sleep period between 10 p.m. and 8 a.m. with the remaining hours for transition and educational activities)16 hours with no protected sleep period |
| Maximum in-hospital on-call frequency | Every third night, on average | Every third night, no averaging |
| Minimum time off between scheduled shifts | 10 hours after shift length | 10 hours after day shift12 hours after night shift14 hours after any extended duty period of 30 hours and not return until 6 a.m. of next day |
| Maximum frequency of in-hospital night shifts | Not addressed | 4 night maximum; 48 hours off after 3 or 4 nights of consecutive duty |
| Mandatory time off duty | 4 days off per month1 day (24 hours) off per week, averaged over 4 weeks | 5 days off per month1 day (24 hours) off per week, no averagingOne 48-hour period off per month |
| Moonlighting | Internal moonlighting is counted against 80-hour weekly limit | Internal and external moonlighting is counted against 80-hour weekly limitAll other duty-hour limits apply to moonlighting in combination with scheduled work |
| Limit on hours for exceptions | 88 hours for select programs with a sound educational rationale | No change |
| Emergency room limits | 12-hour shift limit, at least an equivalent period of time off between shifts; 60-hour workweek with additional 12 hours for education | No change |
Table reprinted with permission from the National Academies Press, Copyright 2008, National Academy of Sciences.
In the face of increasing concerns voiced by the public and by health-care workers, as well as increased media attention regarding patient safety and the role of fatigue and sleep deprivation, in 2007 the U.S. Congress asked the Institute of Medicine (IOM) “to evaluate current evidence on the topic and to develop strategies to optimize work schedules and other activities.”5 The IOM report, released in December 2008, makes several recommendations that expand on the current ACGME regulations (summarized in Table).5
Per the IOM, the available relevant data were reviewed to look at five parameters: 1) duty hours; 2) impact on trainee quality of life and safety; 3) impact on trainee workload; 4) impact on education; and 5) impact on patient safety. Recommendations were provided for each of these five subject areas.
Overall, it was felt that restricting duty hours was not enough. In addition to more clearly defining the structure and application of their duty-hour recommendations, an increase in senior level supervision was felt to be necessary. The IOM recognized that implementation and compliance with their recommendations might be costly, estimated at $1.7 billion; this is thought by many to be a gross underestimate.
The IOM’s recommendations maintain an 80-hour workweek limit, but the recommendations include further definition of what is appropriate for the specific shift components. These recommendations are currently being reviewed and discussed by the ACGME. Training programs are not obligated to implement the IOM’s recommendations until the ACGME gives programs direction as to what, if any, changes are to be implemented to meet their accreditation standards. Both the ACGME’s 2003 duty-hour policies and the IOM’s new recommendations lead to a number of concerns for training programs and teaching hospitals.
There is question about the quality and relevance of the published data that were reviewed. There has been disagreement as to the selection, interpretation, and quality of the studies reviewed that attempt to address the effects of sleep and fatigue.
Program directors (and others) have consistently expressed concern that the current (and assumed to be modified) duty-hour regulations have been implemented without any assessment of the possible negative (and positive) effects of these changes on all five of the above categories. There is acknowledgment that “unintended consequences” have occurred, and will continue to arise. The IOM does recommend that flexibility be included in duty-hour policies in an attempt to minimize the negative impact that might occur as a result of duty-hour restrictions, but they also admit that they were not charged to look into the downstream effects of their recommendations.
It is generally accepted that the true financial costs of complying with the IOM’s recommendations will be many orders of magnitude greater than initial estimates. It has been suggested that having more trainees in a given program might be a desired solution that provides the necessary patient care while allowing for compliance with duty-hour requirements. However, to do this will require readjusting the current “caps” on training positions and expanding the funding provided for these trainees by the Centers for Medicare and Medicaid Services and other third-party payers. In addition, with anticipated physician workforce shortages, it is not clear that there will be enough qualified physicians to fill these needs even if there were funding for them. We are already seeing a shift to using physician extenders in the delivery of health care in the United States without a clear understanding of the long-term implications on patient care and safety.
Another unexpected consequence that continues to be discussed is whether some training programs will need to extend their length of training in order to make up for the lost clinical and educational opportunities that will occur as a result of shorter work schedules and the need to meet ACGME and/or specialty board certification requirements. This would have obvious financial and institutional consequences and would likely provide further disincentive to trainees who are considering the pursuit of research-based careers.
There is significant concern that implementation and enforcement of duty-hour requirements are already leading to a decline in professional behavior and attitudes among our trainees and graduates. “Teaching” is more than just delivering lectures and providing reading assignments; it is also about setting examples and expectations.
As mentioned above, the ACGME is beginning to evaluate the IOM’s recommendations. It is not clear what opportunity will be provided for the academic training program community to comment on any proposed changes to the current ACGME requirements. However, several institutions, training programs, professional societies (including the American Society of Hematology), and other organizations are watching this matter closely and will be providing feedback to the ACGME as soon as possible.
What is clear is that organizations such as the IOM and ACGME continue to develop policies that often appear to be in response to political and social pressures and that have significant impact on the care of patients and the education of physicians without any apparent regard to the unintended consequences of their actions. Although these groups will claim that they have “representatives” from some different relevant constituencies, these organizations typically identify these individuals themselves, and there is no assurance that these representatives have the ability to independently review or have a broad understanding of the matters at hand. Typically, there is limited opportunity for the affected constituencies to fully understand how recommendations or policies are developed or to provide input that may have significant relevance to the matter. For example, they have not been asked to assist in identifying and addressing the unintended consequences that are anticipated to result from implementation of these new recommendations or policies.
We are all interested and motivated to provide excellent care for our patients and effective, high-quality training for our residents, fellows, and students. The matters of patient safety and trainees’ sleep and fatigue are certainly deserving of attention, as patients’ well-being and trainee education is crucial. Although there should always be efforts to make improvements, it would be best if significant changes to patient care and training environments be thought out in advance and with collaborative communication with and involvement of the many stakeholders, including program directors, trainees, and patients. These should not be precipitous actions made in response to emotionally charged comments from politicians or news media. There also needs to be thought given to how any mandated changes will be paid for, both monetarily and with human resources. Much needs to be considered before any further regulations are imposed upon trainees and their training programs. Despite the presumed good intentions of the IOM and the ACGME, these organizations would benefit from more collaborative contact with those of us actually living daily with the consequences of their actions.
The environment for training physicians and for delivering patient care is changing for a number of reasons. Let’s hope that the changes that have occurred and will occur will truly be for the improved well-being of all and that there will be consideration of unintended consequences before they occur. This can best be pursued with the design and implementation of “good” studies that ask specific questions and not by politicians looking for sound bites or by those trying to sell newspapers.
1.Accreditation Council for Graduate Medical Education. Resident Duty Hours. Duty Hours Language. Resident duty hours in the learning and working environment. February 2007. [PDF]
2. Laine C, Goldman L, Soukup JR, et al. The impact of a regulation restricting medical house staff working hours on the quality of patient care. JAMA. 1993;269:374-8.
3. Petersen LA, Brennan TA, O’Neill AC, et al. Does housestaff discontinuity of care increase the risk for preventable adverse events? Ann Intern Med. 1994;121:866-72.
4. Fletcher KE, Davis SQ, Underwood W, et al. Systematic review: effects of resident work hours on patient safety. Ann Intern Med. 2004;141:851-7.
5. Institute of Medicine. Reports. Resident duty hours: enhancing sleep, supervision, and safety. December 2008.
