Share this article on:

A Gender-Based Analysis of Work Patterns, Fatigue, and Work/Life Balance Among Physicians in Postgraduate Training

Gander, Philippa PhD; Briar, Celia PhD; Garden, Alexander MB, ChB, PhD; Purnell, Heather MSc (Hons); Woodward, Alistair MBBS, PhD

doi: 10.1097/ACM.0b013e3181eabd06
Gender Issues

Purpose: To document fatigue in New Zealand junior doctors in hospital-based clinical training positions and identify work patterns associated with work/life balance difficulties. This workforce has had a duty limitation of 72 hours/week since 1985. The authors chose a gender-based analytical approach because of the increasing proportion of female medical graduates.

Method: The authors mailed a confidential questionnaire to all 2,154 eligible junior doctors in 2003. The 1,412 respondents were working ≥40 hours/week (complete questionnaires from 1,366: response rate: 63%; 49% women). For each participant, the authors calculated a multidimensional fatigue risk score based on sleep and work patterns.

Results: Women were more likely to report never/rarely getting enough sleep (P < .05), never/rarely waking refreshed (P < .001), and excessive sleepiness (P < .05) and were less likely to live with children up to 12 years old (P < .001). Fatigue risk scores differed by specialty but not by gender.

Fatigue risk scores in the highest tertile were an independent risk factor for reporting problems in social life (odds ratio: 3.83; 95% CI: 2.79–5.28), home life (3.37; 2.43–4.67), personal relationships (2.12; 1.57–2.86), and other commitments (3.06; 2.23–4.19).

Qualitative analyses indicated a common desire among men and women for better work/life balance and for part-time work, particularly in relation to parenthood.

Conclusions: Limitation of duty hours alone is insufficient to manage fatigue risk and difficulties in maintaining work/life balance. These findings have implications for schedule design, professional training, and workforce planning.

Dr. Gander is director, Sleep/Wake Research Centre, Massey University, Wellington, New Zealand.

Dr. Briar was, at the time of this study, senior lecturer, Department of Sociology, Social Policy, and Social Work, Massey University, Palmerston North, New Zealand, and is a freelance writer and researcher in New Zealand and the United Kingdom.

Dr. Garden was, at the time of this study, clinical associate director, Sleep/Wake Research Centre, Massey University, Wellington, New Zealand, and is currently clinical scientific leader, Simulation and Skills Centre, Capital and Coast District Health Board, Wellington, New Zealand.

Ms. Purnell was, at the time of this study, junior research officer, Sleep/Wake Research Centre, Massey University, Wellington, New Zealand, and is currently clinical research associate comonitor, Biogen Idec Ltd., London, United Kingdom.

Dr. Woodward is head, School of Population Health, University of Auckland, Auckland, New Zealand.

Correspondence should be addressed to Dr. Gander, Sleep/Wake Research Centre, Massey University, Private Box 756, Wellington, New Zealand; telephone: (644) 380-0633; fax: (644) 380-0629; e-mail:

Numerous studies have documented adverse effects of the work patterns of physicians in postgraduate training on their own health, safety, and well-being, as well as on patient safety.1–12 Initiatives to improve work patterns—such as the U.S. Accreditation Council of Graduate Medical Education (ACGME) standards,13 which were introduced in July 2003, and the provisions of the European Working Time Directive,14 which became applicable to doctors working in European Union member states in August 2004—have typically focused on reducing duty hours.

In New Zealand, junior doctors in hospital-based clinical training positions have been subject to a contractual maximum of 72 hours per week since 1985.15 However, we have previously reported that they are twice as likely as the general adult population to score as excessively sleepy on the Epworth Sleepiness Scale (30% versus 15%); moreover, 24% can recall falling asleep at the wheel while driving home from work (since becoming doctors), and 42% can recall a fatigue-related clinical error in the past six months.16 (The term fatigue is here used as a catchall term for impairment that occurs when people continue working even though they have not fully recovered from the demands of prior work and other waking activities.)

Following on from initiatives in the transport sector, advocacy is growing for a more comprehensive approach to fatigue risk management among physicians in postgraduate training.3,5,12,16–18 The Australian Medical Association has developed a guide to fatigue risk assessment that addresses 10 aspects of work patterns and sleep in the previous week.18 We adapted this guide into a multidimensional matrix for assessing fatigue in junior doctors in a hospital-based shift system, and we previously reported that risk scores on the matrix have consistent, dose-dependent relationships to reports of sleepiness and fatigue-related clinical errors, after control for a range of demographic variables.16

Our aim in the present project was to see whether aspects of work patterns, either singly or as combined in the multidimensional fatigue risk matrix, were also associated with reports of difficulties with work/life balance, which is increasingly important to physicians in postgraduate training.15,19,20 We identified a particular need for better understanding of the needs and aspirations of female doctors with regard to work/life balance and flexible work patterns, because women now are the majority of medical graduates in a number of countries, including New Zealand.20–22 Junior doctors in the present survey included physicians in the first two years of postgraduate training (“house surgeons”), those in the third year of postgraduate training (“senior house officers”—a position designed to lead to specialist training), and doctors who typically are in specialist training positions (“registrars”). All worked in public hospitals.

Back to Top | Article Outline


We sent a four-page confidential questionnaire (available from the corresponding author) to all junior doctors in hospital-based clinical training positions in New Zealand in 2003. It sought information on demographics and professional activities, work patterns in each of the previous two weeks,18 sleep habits23 and sleepiness in daily life in general24 and while driving,25 fatigue-related errors in clinical practice,26 and the availability of different types of support.18 Participants also completed the Social and Domestic Survey module of the Standard Shiftwork Index,27 which asks, “[I]n general, to what extent does your work pattern cause you problems with your social life? [H]ome life? [P]ersonal relationships? [A]nd other commitments?” The choices ranged from 0 (not at all) to 4 (very much). Scores of 3 or 4 were considered to indicate problems due to work patterns. At four points in the questionnaire, participants were prompted to write comments on schedule change, adequacy of supervision, the impact of work patterns on life outside work, and any other aspects of hours of work, shift work, or scheduling that they would like to address.

Study packages with a paid return envelope addressed to the Sleep/Wake Research Centre were initially distributed by the Resident Doctors' Association (the union representing the potential participants). To ensure confidentiality, the Sleep/Wake Research Centre had no knowledge of names or addresses, and the only shared information was a unique study identification number for each participant. The first mailing (August 2003) was followed by a reminder postcard to nonresponders three weeks later, and then a new study package to any remaining nonresponders four weeks after the postcard. In December 2003, feedback indicated that some intended recipients had not received study packages, probably as a result of having moved between hospitals. We therefore distributed new study packages directly through unit managers responsible for junior doctors in every public hospital (more than 99% of these doctors work in public hospitals). Participants who had already completed the questionnaire were asked not to respond a second time, but whether they complied with this request could not be verified.

We offered participants the option to enter a drawing for a two-day “mystery break” for two adults (including airfare, meals, and accommodation within New Zealand). The Wellington regional ethics committee approved the study. Participation was voluntary, and confidentiality of data was ensured.

Back to Top | Article Outline

Data management and analysis

Questionnaires were entered separately by two people, and all discrepancies were resolved. We conducted statistical analyses in SAS (version 8.02; SAS Institute, Inc., Cary, North Carolina) and SPSS (version 12.0.1; SPSS, Inc., Chicago, Illinois) software programs. We calculated a fatigue risk score out of 20 for each participant, based on 10 aspects of their work pattern and sleep in the preceding week—namely, total number of hours worked, having worked any shifts of ≥14 hours, having worked any shifts longer than scheduled and/or ≥24 hours, breaks of <10 hours, breaks of ≥24 hours, the number of days on call, the number of night shifts, schedule predictability, the number of opportunities for sleep between 11:00 pm and 7:00 am, and the number of nights on which enough sleep was obtained.16,18

We used logistic regression to identify independent risk factors for problems with social life, home life, personal relationships, and other commitments. Table 1 provides information on the independent variables considered. Independent variables significantly associated with a dependent variable at the univariate level (P < .10) were included in the respective logistic multiple regression models. We ran two models for each dependent variable: Model 1 included demographic variables and separate work-related variables, and model 2 included demographic variables and the fatigue risk score.

Only participants with complete data for all variables in the model were included (n = 1,109 for problems with social life, n = 979 for problems with home life, n = 1,072 for problems with relationships, and n = 988 for problems with other commitments).

Two of us (H.P. and P.G.) undertook, separately, a thematic analysis of the written comments and then made comparisons and resolved differences, to arrive at a final set of themes. Another one of us (C.B.) reviewed all of the comments focusing on themes relating to life outside of work and also reviewed areas where the quantitative data indicated differences between men and women.

Back to Top | Article Outline



The 1,412 returned questionnaires represent 66% of the 2,154 junior doctors registered in the 2002 Medical Council Workforce Survey (that item of data was selected as the denominator because it contained demographic information for the entire workforce, thus enabling an evaluation of response bias).28 We included in the analyses 1,366 doctors (63% response rate) who worked at least 40 hours/week and whose average (SD) age was 30.7 (5.8) years. Of this group, 49% were women compared with 43% in the 2002 Medical Council Workforce Survey (χ2 = 11.9, P < .001). Among respondents, 26% were house surgeons, 14% were senior house officers (categorized together with house surgeons as house officers), and 61% were registrars. Women constituted the majority of house officers (57%) and registrars in obstetrics–gynecology (72%), pediatrics (65%), and pathology (59%), and they were most markedly underrepresented in surgery (28%). The mean fatigue risk scores of women and men did not differ significantly (women: 7.60, 95% CI: 7.31, 7.88; men: 7.80, 95% CI: 7.52, 8.08; Wilcoxon test: P(W) = 0.226).

Women were less likely than men to report living with dependents (20.0% and 34.1%, respectively; χ2 = 31.9, P < .0001), and this discrepancy was evident for each quartile of age (Table 2). Overall, the average age of men and women who reported having dependents did not differ significantly (t = −0.48, P = .635). Men were more likely than women to live with children aged 0 to 5 years (23.0% and 9.4%, respectively; χ2 = 45.6, P < .0001) and children aged 6 to 12 years (11.8% and 5.9%, respectively; χ2 = 14.3, P < .001). Women were more likely than men to live with people more than 60 years old (4.6% and 2.3%, respectively; χ2 = 5.2, P < .05). Table 3 summarizes the percentages of men and women reporting problems with sleep and sleepiness and problems in life outside work that were due to work patterns.

Back to Top | Article Outline

Work patterns and problems in life outside work

The Appendix summarizes results of the logistic regression analyses identifying independent risk factors for reporting problems with aspects of life outside work. Additional logistic multiple regression models, including the fatigue risk score as a continuous variable and the same demographic factors, confirmed a dose-dependent relationship between the fatigue risk score and each of the outcome measures: problems with social life (OR = 1.96; 95% CI: 1.67, 2.29; P < .0001), problems with home life (1.84; 1.56, 2.16; P < .0001), problems with personal relationships (1.46; 1.25, 1.69; P < .0001), and problems with other commitments (1.75; 1.50, 2.05; P < .0001). This trend was also evident in the unadjusted comparisons across the tertiles of the fatigue risk score (Table 4).

Back to Top | Article Outline

Written comments

Written comments (n = 1,671) were provided by 871 participants (63.8% of the total sample of 1,366). The effects of work patterns on life away from work attracted the most comments—663. In these 633 comments, five themes were mentioned by at least 5% of the total sample: difficulty in being able to commit to regular activities away from work (n = 146; 10.5%), being too tired to maintain activities away from work (n = 121; 8.9%), the disruptive nature of shift work for life outside work, particularly work at night and on weekends (n = 119; 8.7%), having insufficient time to spend with partners, children, or family (n = 109; 8.0%), and particular problems resulting from having to both work and study (n = 83; 6.1%). When we compared the responses of men and women, we found more similarities than differences.

Back to Top | Article Outline

Social isolation.

Many participants reported feeling socially isolated from friends, family, and community outside the hospital, and this was a source of frustration. Many commented that they were unable to commit to regular social and cultural activities, such as evening classes, music groups, sports training, or attending church. In addition, they perceived long hours of work as precluding enjoyment of life outside work. One woman commented on “constantly having to curtail activities in order to get enough sleep.” Some comments indicated that opportunities for a social life within the hospital were also limited by long hours and varying shifts.

Back to Top | Article Outline


Both men and women commented that tiredness sapped their motivation to do a range of things that would improve their health and well-being. For example, “[I'm] too exhausted from work to do anything” (man). Many respondents wrote comments about not having the time or energy for regular exercise. Typical comments were “[I have] no time or energy for exercise” (woman) and “[W]ork takes up most of my time, and, when I'm free, I'm tired” (man). With regard to diet, comments addressed the lack of time or energy for preparing healthy meals and indicated poor eating habits. Three-quarters (76.3%) of participants indicated that they did not have access to healthy food at all hours while at work.

Some participants expressed concern that their health was suffering as a result of constant work-related fatigue and a lack of recreational activities or a proper home life. One woman commented, “Shift work is too hard for me in the long term. I have become low in iron and intermittently depressed, and [I] rarely go out in the evenings and [on] weekends.” Some men also reported that their health was suffering. One commented, “[E]xtended periods of work with minimal breaks have tended to affect my health, lots of viral illnesses and chronic fatigue.”

Back to Top | Article Outline

Difficulties with personal relationships.

Both men and women reported difficulty in maintaining personal relationships because of work demands. Typical comments included, “[R]elationships would suffer if I had any,” “[I have a] failed marriage, social isolation, depression,” and “I don't have my own personal life; I live for the institution [hospital] I work for.” However, one man noted that, since he completed his exams, his relationships had improved.

Some participants, particularly men, commented that their relationship continued only because of a very understanding partner. One woman commented, “[M]y husband, a GP with better working hours, is understanding, but I miss friends.” However, for some participants, having a partner who was also in medicine made it difficult to spend time together.

Back to Top | Article Outline


Many of the participants did not live with children, and some said that their current working life was not suited to having children. This was especially the case when both partners were doctors. Among participants who did have children, time pressure was a significant problem. For some, the combination of work and family precluded almost any other activity.

Not having enough time to spend with their children was also a common concern. An additional worry was that short staffing meant not always being able to take leave if a child became ill. Mothers of young children found the combination of full-time work and parenting difficult. The mother of a 20-month-old wrote, “[O]pportunities for job sharing would be great!”

Some men who had partners and children also acknowledged that their jobs put pressure on their partners, because the male physicians tended to be unavailable to assist at home. As one man put it, “I often arrive home late, meaning I am often unable to help with feeding, bathing, and putting our daughter to bed.” In other cases, participants reported sacrificing sleep time to spend time with their children.

Back to Top | Article Outline

Issues related to medical training.

Participants reported studying as much as 30 hours per week. Study hours were longest when examinations were approaching, but this circumstance was not reflected in a reduction in their scheduled work time. As one woman noted, “If I'm not tired, I have to study; if I'm tired, I still have to study, [so I] have little time free.” One man commented, “[M]y incentive is to become a consultant and to have less calls, no exams, and more personal time.”

Of the 1,366 participants, 17% provided a total of 232 written comments on the adequacy of supervision at work. A number of participants regularly had to make treatment decisions without being able to obtain advice from a more senior colleague. Although they usually had access to supervision during daytime hours, access was more difficult during night duty, at weekends, or at the end of a long shift. As one respondent pointed out, the periods of a lack of supervision generally coincided with the times of day when fatigue was more likely to cloud the clinical judgment of junior doctors. Some stated that they relied on peer input at the times when a senior colleague's supervision was unavailable.

Another supervision issue raised was that some participants had to cover multiple wards at night, and in small hospitals, sometimes a participant was the only doctor on duty or on call in the entire hospital. On the other hand, participants did acknowledge and appreciate receiving good-quality supervision and advice from specialists, even when the specialist was available only by telephone—a circumstance that meant, at times, that the specialist made an assessment without seeing the patient.

Back to Top | Article Outline


These data from a New Zealand national survey show that limiting work to 72 hours per week has not prevented many junior doctors in hospital-based clinical training positions from experiencing fatigue and problems with work/life balance. The written comments added a compelling human perspective to these issues. The findings support previous studies indicating that expectations concerning work/life balance are changing among physicians in postgraduate training,15,17,19,20 a change that has major implications for postgraduate medical education and for workforce planning.

In contrast to the general population,29 female participants were more likely than male participants to report inadequate sleep and excessive sleepiness (Epworth Sleepiness Score >10). The female doctors also were less likely to be living with dependents of any age and, in particular, with children up to 12 years old. Medical training normally spans the most fertile years of a woman's life, and delaying having children increases the risk of fertility problems to a greater extent for women than for men in this age range. A number of women commented that they did not think they could cope with both having children and working full-time in medicine. Among those who already had children, both men and women commented on the difficulty of finding sufficient time to fulfill their roles as parents.

The logistic regression analyses confirmed that multiple aspects of work patterns, not just long working hours, are associated with problems of work/life balance. Independent risk factors for reporting problems with social life, home life, personal relationships, and other commitments included longer total duty hours, increasing amounts of night duty, and schedule changes. The total fatigue risk score was a stronger predictor of problems in life outside work than was any single aspect of the work pattern.

Quite modest commuting times independently increased the risk of reporting that work patterns caused problems in life outside work. Physicians in postgraduate training should be alerted to the effect of commuting times as part of their introduction to shift work.

Living with dependents was an independent risk factor for reporting problems with home life (OR: 2.35; 95% CI: 1.59, 3.48), and the written comments highlight some of the particular difficulties for participants who were parents. These issues are among the reasons given for the strong desire for part-time training and expanded work options on the part of both male and female medical graduates.20

Spending more than 20 hours a week studying was also an independent risk factor for reporting problems with home life (OR: 1.85; 95% CI: 1.15, 2.96). An obvious solution would be to reduce working hours in the lead-up to critical examinations, but the feasibility of such an adjustment would depend on staffing levels, which were highlighted as a problem in a number of the written comments.

Often or always having adequate supervision at work was a risk factor that independently reduced the likelihood of reporting problems with personal relationships, compared with having adequate supervision less often (OR: 0.57; 95% CI: 0.43, 0.75). This finding suggests a carryover of work-related stress into relationships outside work. In the written comments, respondents highlighted supervision at night (43 comments) as a particular concern. Because all participants were working in accredited teaching hospitals, the levels of supervision presumably met the training requirements of their respective specialist medical colleges, but this issue merits further investigation.

Only 14.7% of participants had ever had training or guidance on personal strategies for coping with shift work and extended hours. Online journals offer advice, encouraging physicians in postgraduate training to maintain friendships and outside interests, to be vigilant about their health, to get enough catch-up sleep, and to undertake stress reduction activities such as exercise and meditation.30 However, many written comments suggested that work demands precluded such activities.

Back to Top | Article Outline

Study limitations

It is possible that the findings of this study are influenced by response bias, given the overall response rate of 66%. However, it is not possible to establish whether doctors experiencing high levels of fatigue were more or less likely to respond. The participants included a slightly higher proportion of women than did the 2002 Medical Council Workforce Survey (48.5% versus 42.5%; χ2 = 11.9, P < .001). However, any biases introduced by the overrepresentation of women in this sample would be expected to underestimate the current situation (56% of house officers and 46% of registrars in the 2008 Medical Council Workforce Survey were women).

An important study limitation is that all of the information is retrospective and self-reported. The factors included in the fatigue risk score and the definitions of risk categories are somewhat arbitrary, and the score does not consider workload.16 Nevertheless, the dose-dependent relationships between the fatigue risk score and the outcome measures suggest that the score captures important aspects of the work patterns.

A cross-sectional study such as this provides limited evidence that relationships between work-related factors and outcomes are causal. However, on the basis of sleep restriction studies31–34 it would be expected that aspects of work patterns that restrict sleep will cause increased sleepiness and degraded performance and mood.

Back to Top | Article Outline

Relevance to policy

The finding of a disproportionate impact of the demands of medical training on women is highly relevant in the context of the feminization of the medical workforce, the current globalized competitive market for medical graduates, and the projected increase in demand for medical practitioners in industrialized countries as a result of aging populations and efforts to reduce working hours.21,35–38 For example, in the United Kingdom between 1996 and 2006, growth in female medical staff was twice that in male staff. Among British medical graduates in 2006, women had different career expectations than did their male counterparts. Women were more likely to express a preference for at least some part-time training (34% versus 15%), to expect to take a break from their medical career at some point (81% versus 51%), and to consider a career incorporating teaching, research, or medical management rather than a purely clinical career (74% versus 56%).20 Thus, the feminization of the medical workforce has important implications for workforce planning and postgraduate training.

From the middle of the 20th century, there has been an explosion in scientific understanding about the essential nature of sleep and its role in the maintenance of human performance, health, and well-being, as well as about the endogenous circadian timekeeping system, which determines that nocturnal sleep is optimal.39 One major implication of this new knowledge is that, to maintain productivity, safety, and workforce health in any continuous activity, there must be a shift in focus from limiting work hours to providing adequate opportunities for recovery sleep.12,39–41 The importance of sleep is illustrated in the present study by the finding that commuting, night work, and schedule changes (all of which affect the amount of time available for sleep) are independent risk factors for reporting difficulties with work/life balance, as are long duty hours. Night work and schedule changes also are independent risk factors for reporting excessive sleepiness and fatigue-related clinical errors among the participating doctors.16

Back to Top | Article Outline


In Europe42 and the United States,43,44 the medical profession still focuses primarily on a single fatigue-risk-reduction strategy for physicians in postgraduate training—namely, limitations on duty hours. The 2008 Institute of Medicine report on resident duty hours45 recommends a somewhat more comprehensive approach, but such an approach has yet to be adopted by the ACGME or any other regulatory agency.46 Perhaps paradoxically, progress in applying sleep science to the management of occupational fatigue is occurring most rapidly in the transportation sector, which is leading the way in developing comprehensive fatigue risk management systems (FRMSs).40,41 In contrast to traditional limits on work hours, FRMSs are data driven, monitoring where and when fatigue risk actually occurs and safety may be jeopardized, and having a layered system of defenses to mitigate and manage that risk. The FRMS approach enables tailored solutions to specific issues40,41 and thus offers flexibility that might help address the difficulties being experienced with current efforts to enforce prescriptive limits on the duty hours of physicians in postgraduate training.42,46

Back to Top | Article Outline


The authors wish to thank the participants for their time and commitment in completing questionnaires and providing written commentary. Margo van den Berg and Erich Janssen (formerly of the Australian Medical Association) provided important input in the early stages of this project. Invaluable support was provided by the staff at the Resident Doctors' Association, Te Ohu Rata o Aotearoa (the Maori Doctors Association), the Medical Council of New Zealand, and resident medical officer unit managers nationwide.

Back to Top | Article Outline


This project was funded by Project Grant #02/249 from the Health Research Council of New Zealand. The Medical Assurance Society provided the incentive prize.

Back to Top | Article Outline

Other disclosures:

A.G. is currently employed by the Capital and Coast District Health Board as a specialist anesthetist. H.P. received travel funding from the Maurice and Phyllis Paykel Trust and the Fulbright Foundation to present other aspects of the Junior Doctors Survey at the 19th Annual Meeting of the Associated Professional Sleep Societies, Denver, Colorado, June 18–23, 2005.

Back to Top | Article Outline

Ethical approval:

This project received ethical approval from the Wellington regional ethics committee (reference 03/02/004).

Back to Top | Article Outline

Previous presentations:

Abstracts on aspects of this project were presented at the Annual Scientific Meeting of the Australasian Sleep Association, Sydney, New South Wales, Australia, October 15–17, 2004; the 19th Annual Meeting of the Associated Professional Sleep Societies, Denver, Colorado, June 18–23, 2005; and the International Commission on Occupational Health's 18th International Symposium on Shiftwork and Working Time, Yeppoon, Queensland, Australia, August 28–31, 2007.

Back to Top | Article Outline


1Gaba DM, Howard SK. Fatigue among clinicians and the safety of patients. N Engl J Med. 2002;47:1249–1255.
2Veasey S, Rosen R, Barzansky B, Rosen I, Owens J. Sleep loss and fatigue in residency training: A reappraisal. JAMA. 2002;288:1116–1124.
3Cass HD, Smith I, Unthank C, Starling C, Collins JE. Improving compliance with requirements on junior doctors' hours. BMJ. 2003;327:270–273.
4Baldwin DC Jr, Daugherty SR, Tsai R, Scott MJ Jr. A national survey of residents' self-reported work hours: Thinking beyond specialty. Acad Med. 2003;78:1154–1163.
5Howard SK, Gaba DM, Smith BE, et al. Simulation study of rested versus sleep-deprived anesthesiologists. Anesthesiology. 2003;98:1345–1355.
6Landrigan CP, Rothschild JM, Cronin JW, et al. Effects of reducing interns' work hours on serious medical errors in intensive care units. N Engl J Med. 2004;351:1838–1848.
7Lockley SW, Cronin JW, Evans EE, et al. Effect of reducing interns' weekly work hours on sleep and attentional failures. N Engl J Med. 2004;351:1829–1837.
8Murray A, Pounder R, Mather H, Black DC. Junior doctors' shifts and sleep deprivation—The European Working Time Directive may put doctors' and patients' lives at risk. BMJ. 2005;330:1404.
9Barger LK, Cade BE, Ayas NT, et al. Extended work shifts and risk of motor vehicle crashes among interns. N Engl J Med. 2005;352:125–134.
10Fletcher KE, Underwood W III, Davis SQ, Mangrulkar RS, McMahon LF, Saint S. Effects of work hour reduction in residents lives: A systematic review. JAMA. 2005;294:1088–1100.
11Arnedt JT, Owens J, Crouch M, Stahl J, Carskadon MA. Neurobehavioral performance of residents after heavy night call vs after alcohol ingestion. JAMA. 2005;294:1025–1033.
12Dawson D, Zee PH. Work hours and reducing fatigue-related risk: Good research vs good policy. JAMA. 2005;294:1104–1106.
13Philibert I, Friedmann P, Williams WT; ACGME Work Group on Resident Duty Hours. New requirements for resident duty hours. JAMA. 2002;288:1112–1114.
14United Kingdom Department of Health Healthcare Workforce Portal. Programme management. Available at: Accessed May 18, 2010.
15Powell D. Key issues facing resident medical officers. N Z Med J. 2004;117:1116–1119.
16Gander PH, Purnell HM, Garden AL, Woodward A. Work patterns and fatigue-related risk among junior doctors. Occup Environ Med. 2007;64:733–738.
17Dawson D, McCulloch K. Managing fatigue: It's about sleep. Sleep Med Rev. 2005;9:365–380.
18Australian Medical Association. National code of practice: Hours of work, shiftwork and rostering for hospital doctors. Available at: Accessed May 18, 2010.
19Dorsey RE, Jarjoura D, Rutecki GW. The influence of controllable lifestyle and sex on specialty choices of graduating US medical students, 1996–2003. Acad Med. 2005;80:791–796.
20British Medical Association. Cohort study 2006 medical graduates. First report, June 2007. Available at: Accessed May 18, 2010.
21Brooks PM, Lapsley HM, Butt DB. Medical workforce Australia: “Tomorrow's doctors—too few, too far.” Med J Aust. 2003;179:206–208.
22Medical Council of New Zealand. The New Zealand Medical Workforce in 2003. Available at: Accessed May 18, 2010.
23Gander PH, Marshall NS, Harris R, Reid P. Sleep and motor vehicle accident involvement: A national survey. Aust N Z J Public Health. 2005;29:16–21.
24Johns MW. Sleepiness in different situations measured by the Epworth Sleepiness Scale. Sleep. 1994;17:703–710.
25Maycock G. Sleepiness and driving: The experience of UK car drivers. J Sleep Res. 1996;5:229–237.
26Gander PH, Merry AF, Millar MM, Weller J. Hours of work and fatigue-related error: A survey of New Zealand anaesthetists. Anaesth Intens Care. 2000;28:178–183.
27Barton J, Spelten E, Totterdell P. The Standard Shiftwork Index: A battery of questionnaires for assessing shiftwork-related problems. Work Stress. 1995;9:4–30.
28Medical Council of New Zealand. The New Zealand Medical Workforce in 2002. Available at: Accessed May 18, 2010.
29Gander PH, Marshall NS, Harris R, Reid P. Epidemiology of the Epworth Sleepiness Score: Ethnicity and socio-economic factors. Sleep. 2005;28:249–253.
30Robinson G, Bernau S, Aldington S, Beasley R. From medical student to junior doctor: Maintaining good health during the “baptism of fire.” Student BMJ. 2006;14:133–176. Available at: Accessed May 18, 2010.
31Dinges 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;20:267–277.
32Harrison Y, Horne JA. The impact of sleep deprivation on decision making: A review. J Exp Psychol Appl. 2000;6:236–249.
33Belenky G, Wesensten NJ, Thorne DR, et al. Patterns of performance degradation and restoration during sleep restriction and subsequent recovery: A sleep dose–response study. J Sleep Res. 2003;12:1–12.
34Van Dongen HP, Maislin G, Mullington JM, Dinges DF. The cumulative cost of additional wakefulness: Dose–response effects on neurobehavioral function and sleep physiology from chronic sleep restriction and total sleep deprivation. Sleep. 2003;26:117–126.
35Cooper RA. Weighing the evidence for expanding the physician supply. Ann Intern Med. 2004;141:705–714.
36Health Workforce Advisory Committee. Fit for Purpose and for Practice: Advice to the Minister of Health on the Issues Concerning the Medical Workforce in New Zealand. Available at: Accessed May 18, 2010.
37Joyce CM, McNeil JJ, Stoelwinder JU. More doctors but not enough: Australian medical workforce supply 2001–2012. Med J Aust. 2006;184:441–446.
38National Health Service Information Centre. NHS Hospital and Community Health Services: Medical and Dental Staff England 1996–2006. Available at:–2006-medical-and-dental. Accessed May 18, 2010.
39Walsh JK, Dement WC, Dinges DF. Sleep medicine, public policy, and public health. In: Kryger MH, Roth T, Dement WC, eds. Principles and Practice of Sleep Medicine. 4th ed. Philadelphia, Pa: Elsevier Saunders; 2005:648–658.
40Gander PH, Hartley L, Powell D, et al. Fatigue risk management I: organizational factors. Accid Anal Prev. In press.
41Gander PH, Graeber RC, Belenky G. Fatigue risk management. In: Kryger MH, Roth T, Dement WC, eds. Principles and Practice of Sleep Medicine. 5th ed. Philadelphia, Pa: Elsevier Saunders; 2009:760–768.
42House J. Calling time on doctors' working hours. Lancet. 2009;373:2011–2012.
43Blanchard MS, Meltser D, Polonsky KS. To nap or not to nap? Residents work hours revisited. N Engl J Med. 2009;360:2242–2244.
44Schwenk TL. Residents' work hours. N Engl J Med. 2009;361:928.
45Ulmer C, Wolman DM, Johns MME, eds. Resident Duty Hours: Enhancing Sleep, Supervision, and Safety. Washington, DC: National Academies Press; 2008.
46Landrigan CP. Update on resident work hour reform. SRS Bull. 2010;15:13–14.
© 2010 Association of American Medical Colleges