Search tips
Search criteria 


Logo of oenvmedOccupational and Environmental MedicineVisit this articleSubmit a manuscriptReceive email alertsContact usBMJ
Occup Environ Med. 2007 November; 64(11): 719–720.
PMCID: PMC2078423

Physicians' work hours: desperately seeking evidence

Short abstract

Commentary on the paper by Gander et al (see page 733)

Evidence‐based medicine is the cornerstone for excellence in healthcare delivery. However, as a group, physicians have been reluctant to advance the necessary science and to examine in depth related science to formulate evidence‐based safe work schedules.

A case in point is the history of work hour regulation for physicians in training in the USA. The first statewide attempt to regulate work hours occurred over 20 years ago in response to the death of a young woman in a New York City hospital. The woman's father was a New York Times reporter and former federal prosecutor, who pressed for criminal prosecution. A grand jury heard the case and concluded that the prolonged work hours of the house staff (junior doctors) caring for the woman contributed, in part, to a medication mistake that resulted in the young woman's death. An ad hoc advisory committee of prominent physicians, without expertise in sleep and circadian disorders or neurobiology (the Bell Commission), reviewed the grand jury's report and agreed that excessive work hours might have resulted in fatigue‐related errors. The committee developed a work hour policy based simply on the mathematical midpoint between the 40‐hour workweek schedule for the majority of the American workforce and the 120‐hour workweek that many house staff in New York City were working at that time. Without a well‐substantiated scientific basis, this costly regulation was not enforced. Eventually, medical students in the American Medical Student Association began to fight for better work hours prior to their impending residency programmes. In a brilliant move, they petitioned the Occupational Safety and Health Administration (OSHA), asking that the OSHA investigate the safety of work hours for residents. The Accreditation Council for Graduate Medical Education (AGCME) was informed of this petition and conveyed to the OSHA they would address this issue carefully. In part wanting to do the right thing and in part concerned that they would lose control over the residency work hours, the ACGME settled with the medical students agreeing to strictly enforce work hours. In a less brilliant move, the medical students asked for the Bell Commission work hours. The ACGME agreed, promising tightly controlled regulation of the new work hours, and the OSHA was avoided. Consequently, the present work hours so vigorously regulated in the USA for residents are very close to the original Bell Commission work hours: 80 h/week, with on‐call duty no more frequent than every third night; on‐call shifts no longer than 24 h, but up to 30 h for transfer of care to the next resident, and on average one full day off every seven days.

Importantly, it is never too late to advance the science of safe work hours. Indeed, there are many unresolved issues regarding physicians' work hours. For example, does the new 80‐hour workweek reduce medical error? What is the optimal number of workweek hours that residents should work to balance patient safety and learning? What factors, aside from workweek hours, contribute to impaired physician performance?

Presently, there is no evidence that the 80‐hour workweek increases patient safety. In the early 1990s, there were three retrospective studies in New York examining the effect of the 80‐hour workweek on either medical error or patient mortality. All three studies are limited by the lack of documentation of actual work hours and by retrospective study design.1,2,3 Nonetheless, the studies found no improvement in patient outcomes, and one study showed increased mortality and greater delays in diagnostic testing after the imposition of an 80‐hour workweek.3 Findings are mixed on whether the 80‐hour workweek hinders education.4,5,6,7 The most consistent effect of the 80‐hour workweek is improvement in quality of life for the residents.8 Thus, two decades after the beginning of work hour regulations for house staff in the USA, we do not know how many total hours a physician or physicians in training can work effectively, efficiently and safely, or whether total hours per week is an important factor in fatigue.

Returning to carefully controlled studies, there are important factors beyond total hours of wakefulness or work that contribute to fatigue and error. Alertness requires both the fulfilment of the homeostatic drive for sleep and circadian arousal timing.9 For example, healthcare providers who work only 40 h/week are not immune to fatigue and cognitive impairment if they experience extended (>10 h) time on the job or must work early morning hours without adjusting circadian timing. The Australian Medical Association has taken a lead in organising extensive discussions with experts in alertness and shift work, hospital administrators, nurses and physicians and integrating data from many sources to develop a list of the most likely contributors to fatigue‐related error.10 In addition to total hours, the factors include night duty, shift schedule change frequency, short breaks between work periods, commuting time, age, medical experience, and supervision. Of note, based on this checklist, many American residency programmes adhering completely to the ACGME work hour regulations would still fall into the “high risk” category for fatigue‐related error.

In this month's issue Gander et al take advantage of this well thought out risk assessment tool to identify relative risks for fatigue in physicians and medical error and to validate the Australian risk assessment tool (see page 733).11 They performed a national survey of physicians in training in New Zealand that included all items on the Australian Medical Association National Code of Practice checklist to determine individual risk. Total hours worked per week, >70, 50–70 or <50, was not an independent risk factor for fatigue‐related medical error. In contrast, the frequency of night shift duty and being a physician in training were independent risk factors for fatigue‐related medical error. Night duty, schedule change and commuting distance were all independent risk factors for sleepiness while driving. Interestingly, the presence of supervision at work reduced the risk for both fatigue related medical error and sleepiness while driving. These findings support the concept that reducing work hours alone will not deliver alert doctors to the bedside. Further, these findings support the concept that night shift schedules remain a major risk for medical error. To begin to address this important healthcare cancern, we must identify effective means of adjusting the circadian rhythm in all healthcare workers working night shifts and we must be able to monitor entrainment in night shift workers. At the same time, there will be emergencies when physicians must work across circadian nadirs and, for these times, we must identify optimal short sleep schedules and truly effective fatigue counter measures.

One important factor left out of the Australian Medical Association's risk assessment checklist is how rested physicians are at the start of duty. Sleep loss has cumulative effects on performance, and the intern or resident who is out late at night is far more likely to have lapses in judgment and performance than the well‐rested intern or resident. It is essential that the interns/residents do their part to minimise fatigue‐related error. This particular point cannot be regulated, and therefore requires a cultural change. As such, physician role models must reiterate this particular point frequently.

In conclusion, while we await improved science upon which to base work hour schedules for healthcare providers, the Australian Medical Association's National Code of Practice for work hours, combined with the cultural change of delivering rested physicians to the bedside, is an excellent foundation on which to build evidence‐based practice of safer work hours.


Competing interests: None declared.


1. Daigler G E, Welliver R C, Stapleton F B. New York regulation of residents' working conditions. 1 year's experience. Am J Dis Child 1990. 144799–802.802 [PubMed]
2. Howard D L, Silber J H, Jobes D R. Do regulations limiting residents' work hours affect patient mortality? J Gen Intern Med 2004. 191–7.7 [PMC free article] [PubMed]
3. Laine C, Goldman L, Soukup J R. et al The impact of a regulation restricting medical house staff working hours on the quality of patient care. JAMA 1993. 269374–378.378 [PubMed]
4. Jagsi R, Shapiro J, Weissman J S. et al The educational impact of ACGME limits on resident and fellow duty hours: a pre‐post survey study. Acad Med 2006. 811059–1068.1068 [PubMed]
5. Brunworth J D, Sindwani R. Impact of duty hour restrictions on otolaryngology training: divergent resident and faculty perspectives. Laryngoscope 2006. 1161127–1130.1130 [PubMed]
6. Hutter M M, Kellogg K C, Ferguson C M. et al The impact of the 80‐hour resident workweek on surgical residents and attending surgeons. Ann Surg. 2006;243:864–71; discussion 871–5.
7. Karamanoukian R L, Ku J K, DeLaRosa J. et al The effects of restricted work hours on clinical training. Am Surg 2006. 7219–21.21 [PubMed]
8. Lockley S W, Cronin J W, Evans E E. et al Effect of reducing interns' weekly work hours on sleep and attentional failures. N Engl J Med 2004. 3511829–1837.1837 [PubMed]
9. Van Dongen H P, Dinges D F. Investigating the interaction between the homeostatic and circadian processes of sleep‐wake regulation for the prediction of waking neurobehavioural performance. J Sleep Res 2003. 12181–187.187 [PubMed]
10. Australian Medical Association National Code of Practice: Hours of Work, Shiftwork and Rostering for Hospital Doctors. AMA 2005
11. Gander P, Purnell H, Garden A. et al Work patterns and fatigue‐related risk among junior doctors. Occup Environ Med. 2007;64:733–8

Articles from Occupational and Environmental Medicine are provided here courtesy of BMJ Publishing Group