AMA Journal of Ethics®

Illuminating the art of medicine

Journal of Ethics Header

AMA Journal of Ethics®

Illuminating the art of medicine

Virtual Mentor. March 2003, Volume 5, Number 3.

Journal Discussion

  • Print

Debate Over the 80-Hour Work Week

Various organizations are trying to limit the number of hours medical residents can work per week, in hopes of fostering better patient care.

Jeremy Spevick

In residency, the final stage in a physician's formal medical training, long hours and altered sleep schedules are the norm, as residents take on the full responsibilities of a medical professional. With workweeks regularly exceeding 80 hours, resident fatigue is a common problem. Work hour regulations are set and enforced by the Accreditation Council for Graduate Medical Education (ACGME), which, as its name indicates, accredits all US medical residency programs. Current regulations stipulate that residents must average (over a month) 1 day per week free of patient care and cannot be on-call more frequently than every third night.

The issue of overworked residents came to national prominence in 1999, when the National Labor Relations Board overturned a 23-year precedent by ruling that residents at private institutions could unionize and enter into collective bargaining. Although many thought this would lead to widespread unionization and large-scale changes in resident work hours, this has not been the case [1]. Organized medicine's resistance to being considered a trade group rather than a profession exerted a restrictive influence on those who wished to unionize.

In April of 2001, several groups, including the Public Citizen Health Research Group, the Committee of Interns and Residents, and the AMA, petitioned the Occupational Safety and Health Administration (OSHA) to establish and enforce a federal work hour standard for residents. They petitioned to limit the residents' workweek to 80 hours, restrict consecutive hours of on-call duty to 24, and allow for minimal periods of 10 hours rest between shifts. As of March 2003, the OSHA has not formally responded to this petition [2].

In June 2002, the ACGME announced that new guidelines for resident work hours would become effective for all residency programs on July 1, 2003. The new guidelines specify an 80-hour workweek, averaged over a four-week period. Failure to comply with regulations could result in the non-compliant program's loss of accreditation. Implementing the new changes will be challenging and expensive [3]. Residents' salaries are lower than those of staff physicians, and reducing residents' work hours will create a need for more expensive medical help.

A debate has emerged as to whether the ACGME's new policy will be beneficial to the medical profession. Will it provide patients with better care from doctors who are more rested, or will it cut short valuable time that residents need to mature into fully qualified physicians? This and related questions are discussed in a New England Journal of Medicine article by David M. Gaba and Steven K. Howard [4]. First, the authors review the challenges to current work hours that I have just mentioned. Then they turn to the arguments for and against limiting work hours and discuss the changes in hospital environment that will be needed to implement the policy changes.

Those in favor of restructuring the environment of graduate medical education argue that the medical profession, like the aviation and transportation industries, should limit its work hours to reduce employees' fatigue. Gaba and Howard cite many studies that link fatigue to higher levels of depression, anxiety, confusion, and anger [5]. With more rest, the inference is, residents would have better interactions with patients and be better equipped to learn.

It is not entirely clear to Gaba and Howard that sleep loss leads to poor clinical performance, which, in turn, may cause harm to patients. Most of the studies that draw that conclusion have methodological flaws such as inconsistent definitions of fatigued and rested subjects, invalidated measures of clinical performance, and failure to account for circadian effects. Despite this uncertainty, Gaba and Howard feel that it is necessary to limit excessive work hours. In fact, they believe that the upcoming ACGME policy may not go far enough. For one thing, it will allow individual programs to apply for a 10 percent increase in the hour limit (to 88 hours per week) if the program can provide a sound educational rationale for doing so. And, they say, the guidelines are not as strict as those of other Western countries.

On the other side of the limited hours debate are physicians who think that the current working environment is essential for the learning and maturation of residents. Residents must learn that being a medical professional means having obligations to patients that go beyond personal schedules and conveniences. Only by spending long periods of time with patients can residents observe disease progress and how it affects patients and their families [6]. Those hesitant about work-hour changes are also concerned that shorter work weeks will result in more frequent changes in who is caring for each patient. Less continuity in care could possibly lead to more miscommunication between practitioners and delays in treatment for patients.

The only precedent to limiting residents' workweeks occurred in 1989 when New York state enacted legislation limiting work to 80 hours per week in response to the death of Libby Zion, in which physician fatigue was suspected as a contributing factor. The New York legislation has been expensive and difficult to enforce and has illustrated the difficulty in instituting widespread changes in the residency system. In June 2002, the New York State Health Department announced that 54 of 82 teaching hospitals inspected were cited for violations of the law [7].

The ACGME may confront a conflict of interest in enforcing the regulations. Shortly after the announcement of the new regulations, a the New York Times editorial [8] noted that the Council's board is dominated by the trade associations of hospitals, doctors, and medical schools, all of which benefit from the inexpensive labor performed by residents. The Council may find it difficult to enforce regulations that go against the interests of its constituency organizations.

As the New York experience demonstrates, the cost of implementing the new regulations will pose problems. The New York state government allocated funds to teaching hospitals to cover some of the costs of reforming the system. The ACGME has no funds to give to hospitals to ease the cost of the transition, and most doctors fear that accepting state funds for residency programs would give the state a reason to think it could regulate resident work hours. Self-regulation is one of the most highly guarded rights of the medical profession (of most professions for that matter).

As this debate continues, the irony is that both proponents and critics of the changes share similar views on quality of care and the goals of residency training, although they differ in their means for achieving these goals [9]. In the end, the goals of improving residency training will not be achieved merely by limiting work time to 80 hours a week. This task will require a broader restructuring of training programs to achieve balance between the duties devoted to educating residents and those devoted to caring for patients.

Questions for Discussion

  1. Will the new ACGME policy improve patient care by allowing residents to be more alert and rested, or will it hinder patient care by limiting the time needed for resident development?
  2. If you are a resident, do you think that the number of hours you work negatively affects your functioning as a doctor?
  3. Gaba and Howard argue that changing the behavior of clinicians goes beyond the limiting of work hours. What other changes are needed in the culture of medicine in which exhaustion is viewed more as a sign of dedication than as an unacceptable risk?
  4. Should the government be involved in enforcing regulations on resident work hours?


  1. Gaba DM, Howard SK. Fatigue among clinicians and the safety of patients. N Eng J Med. 2002;347:1251.
  2. Steinbrook R. The debate over residents' work hours. N Eng J Med. 2002;347:1296.
  3. Steinbrook, 1301.
  4. Gaba, 1249-55.
  5. Gaba, 1249.
  6. Steinbrook, 1297.
  7. Steinbrook, 1299
  8. Sleep-deprived doctors. New York Times. June 14, 2002:A34.
  9. Drazen JM. Epstein AM. Rethinking medical training–the critical work ahead. N Eng J Med. 2002;347:1272.

The viewpoints expressed on this site are those of the authors and do not necessarily reflect the views and policies of the AMA.