NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Institute of Medicine (US) Committee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedule to Improve Patient Safety; Ulmer C, Miller Wolman D, Johns MME, editors. Resident Duty Hours: Enhancing Sleep, Supervision, and Safety. Washington (DC): National Academies Press (US); 2009.

Cover of Resident Duty Hours

Resident Duty Hours: Enhancing Sleep, Supervision, and Safety.

Show details

7Strategies to Reduce Fatigue Risk in Resident Work Schedules

The scientific literature makes clear that risks of fatigue-related errors and accidents derive from multiple interacting variables of work and sleep. This chapter discusses the literature on sleep and human performance and recommends specific adjustments to the current Accreditation Council for Graduate Medical Education (ACGME) resident duty hours to enhance the prevention and mitigation of resident fatigue as an unsafe condition, thereby improving performance and the safety of both patients and resi dents. The major rationales for the recommendations are the following: (1) work duration should be limited because human performance degrades after 16 hours of wakefulness whether one is working or not; (2) suffi cient time for sleep needs to be incorporated into daily and weekly work schedules to prevent acute and chronic sleep deprivation, respectively, and to allow recovery from accumulated sleep debt; and (3) when extended duty periods are considered an essential aspect of resident training and continuity of care, a protected sleep period should be provided during that period to reduce the effects of acute sleep loss and to enhance performance. Because of the diversity of specialty and hospital needs, the committee leaves some flexibility for programs, but urges that adequate protected sleep periods be maintained, and that fatigue prevention and mitigation be a matter of professionalism that requires attention by residents, attending physicians and all those charged with maintaining patient safety.

The consensus committee was charged by the Agency for Healthcare Research and Quality (AHRQ) to “1) synthesize current evidence on medical resident schedules and healthcare safety and 2) develop strategies to enable optimization of work schedules to improve safety in the healthcare work environment.” This chapter deals with scientific evidence that fatigue is an unsafe condition that can occur relative to the timing and duration of work and sleep opportunities, which are fundamental components of residents’ schedules. The chapter reviews scientific literature on fatigue, its consequences, and its prevention, and provides recommendations and evidenced-based justifications for ways to reduce fatigue as a safety risk while residents are training intensively by working long hours.

Many of the recommendations focus on ensuring residents obtain adequate sleep, which research has shown is among the most fundamental biological needs, to counter fatigue and promote learning and memory. The focus on resident fatigue prevention in the recommendations for duty hour adjustments in the latter sections of this chapter is a response to AHRQ’s charge that the committee develop strategies to enable optimization of work schedules to improve safety in the healthcare work environment. Other chapters in this report recommend additional ways in which safety can be enhanced through supervision, appropriate workload, teamwork, and system changes. This chapter takes an evidenced-based approach to developing recommended changes in only those aspects of resident duty hours that are most likely to result in fatigue as an unsafe condition that can pose risks to both patients and residents. Thus, to retain the training value and flexibility in scheduling required by different specialties and rotations within specialties, while preventing and mitigating sleep loss that contributes to fatigue-related errors and accidents, the recommendations derived from this chapter relative to duty hours are focused more on providing predictable and protected time for sleep and recovery sleep than on limiting total work hours.


In healthy individuals, fatigue is a general term used to describe feelings of tiredness, reduced energy, and the increased effort needed to perform tasks effectively and avoid errors. It occurs as performance demands increase because of work intensity and work duration, but it is also a product of the quantity and quality of sleep and the time of day work occurs (Dinges, 2001). All of these factors are relevant aspects of residency training duty hours (Buysse et al., 2003). Some current aspects of resident duty hours can interfere with normal sleep patterns and lead to sleep deprivation, with the extent of this deprivation differing according to the workload demands and schedule of each residency specialty.

Risks of fatigue-related errors and accidents in relation to work schedules derive not from the single factor of the total hours of work in a week, but from multiple interrelated and interacting aspects of work, rest, and sleep. These include but are not limited to (1) the duration of work peri ods within a single day and over time, (2) the time of day at which work occurs, (3) variation in the timing of work within and between weeks, (4) the duration of sleep obtained on work days and on non-work days, (5) the frequency and duration of days off from work, (6) the different vulnerabilities of workers to fatigue from these factors, and (7) the volume and intensity of work (Dinges, 1995; Drake et al., 2004; Folkard et al., 2005; Rosa, 2001; Van Dongen, 2006). It is not surprising that concern about the negative effects of sleep deprivation on residents is one of the primary reasons duty hour restrictions have been implemented by the Accreditation Council for Graduate Medical Education (ACGME, 2003).

This chapter contains a synthesis of the current evidence about fatigue, performance, and safety risks posed by different work-rest-sleep factors applicable to current ACGME duty hour rules and possible adjustments to those rules. First, the chapter focuses on acute sleep deprivation and ways to prevent the development of acute sleep loss (e.g., shorter duty periods) or ways to mitigate the effects of acute sleep loss by use of sleep during extended duty periods that may be required for patient care and education. The chapter then examines ways to prevent and address the accumulation of chronic partial sleep loss in residents. After reviewing this evidence, the committee proposes strategies to enable optimization of resident work schedules to improve safety in the healthcare work environment while taking into account the learning and experience that residents must achieve during their training.


A sizeable scientific literature exists on the extent to which preventing fatigue, and its associated cognitive performance deficits, depends heavily upon the extent to which acute (daily) and chronic (weekly) sleep needs are met. Moreover, reviews of the risks posed by residency duty hours have emphasized that prevention of sleep deprivation in residents is the most important way to reduce fatigue risks to patient and resident safety (Baldwin and Daugherty, 2004; Buysse et al., 2003; Cavallo and Mallory, 2004; Gaba and Howard, 2002; Howard et al., 2002; Landrigan et al., 2007; Lockley et al., 2006; Parshuram, 2006; Veasey et al., 2002; Weinger and Ancoli-Israil, 2002). Below the committee reviews evidence concerning the benefits to human performance, and potentially to patient safety, from increased sleep time.

Before the 2003 duty hour reforms, first-year residents reported sleeping an average of 5.7 (standard deviation [SD] .90) hours per night and second-year residents reported an average of 5.98 (SD .98) hours (Baldwin and Daugherty, 2004). However, self-reported sleep times tend to overestimate actual physiological sleep obtained (Jean-Louis et al., 2000). No national study of resident sleep hours is available post duty hour reform in 2003. Furthermore, the average number of hours reported by residents may mask the actual degree of sleep deprivation because the schedule of work (e.g., night work, extended duty shifts) may significantly influence the amount of uninterrupted sleep possible.

Reducing work hours does not necessarily result in a corresponding increase in sleep hours. An examination of work hours and the amount of sleep obtained by residents shows that there is a statistically significant but only moderate correlation between residents’ work hours and their sleep hours, with approximately 15 percent (Baldwin and Daugherty, 2004; Baldwin et al., 2003) to 33 percent (Lockley et al., 2004) of common variance between work time and sleep time (see Figure 7-1).

FIGURE 7-1. Relationship of residents’ average weekly sleep to average weekly hours of work.


Relationship of residents’ average weekly sleep to average weekly hours of work. NOTE: Figure A is survey data from n = 3,406 residents. Scatterplot of reported average weekly work hours of sleep with reported average weekly work hours, PGY (postgraduate (more...)

The varied elements influencing fatigue interact in complex ways that make it difficult to attribute risk of reduced resident performance to reduced hours of work. For example, the study by Landrigan and colleagues (2004) restricted the work of interns to approximately 16 hours at a time and eliminated extended duty periods (24 hours or more), which resulted in an average of 19.5 hours less work per week than the traditional schedule, but it also resulted in 5.8 hours more sleep per week. A survey conducted by Baldwin and Daugherty (2004) of residents prior to the 2003 duty hour limits also revealed an inverse relationship between average weekly work hours reported by residents and average weekly sleep time (Figure 7-1A). Thus, it is not possible to isolate the distinct effects of shift length, total work hour limitation, increased sleep time, and/or other consequences of adjusting the duty hour limits (e.g., increased staffing) on the reduction of medical errors found by Landrigan and colleagues (2004). Although the separate contribution of increased sleep time to error reduction cannot be measured, there is ample reason (reviewed below) that sleep could be the primary way in which fatigue and its risks were mitigated in the residents studied by Landrigan and colleagues.

Circadian Influence

Acute sleep loss begins when an individual remains awake beyond 16 to 18 hours or into the habitual nocturnal period for sleep (Van Dongen and Dinges, 2005). Extensive research has shown the brain’s circadian system ceases to oppose the physiologic pressure for sleep after 16-18 hours of being awake. This results in steady increases between midnight and 6-10 a.m. (when the body is biologically programmed to sleep) in sleepiness and sleep propensity, lapses of attention and memory, and a wide range of other cognitive performance deficits (Van Dongen and Dinges, 2005). This natural pressure for sleep occurs when someone is awake at night, whether or not the person is working. Morning hours (4-10 a.m.) are a peak time for drowsy driving accidents (Pack et al., 1995) and other industrial catastrophes (Dinges, 1995; Mitler et al., 1988). However, our innate circadian biology begins promoting wakefulness during the day. Performance impairments from a night without sleep actually decline somewhat by 6-10 p.m. (i.e., at 36-40 hours of being awake) relative to the peak for poor performance earlier in the day between 6 and 10 a.m. (i.e., 24-28 hours awake) (Van Dongen and Dinges, 2005). This circadian modulation of sleepiness and alertness was part of the justification for ACGME’s settling on a 30-hour rather than a 24-hour extended duty period (ACGME, 2003, 2004), although some programs scheduled these extended duty periods in a non-circadian fashion (e.g., starting the 30 hours at noon).


Findings on the effects of 30-hour extended duty periods on the performance of physicians (Philibert, 2005) and the reduction in adult intensive care unit (ICU) medical errors when intern duty periods were limited to approximately 16 hours (Landrigan et al., 2004) received much attention in the medical community. These results, however, were not surprising given data that have accumulated over the past 100 years about the effects of sleep deprivation on attention, memory, and a range of cognitive functions (Dinges and Kribbs, 1991; Durmer and Dinges, 2005; Harrison and Horne, 2000; Kleitman, 1963; Patrick and Gilbert, 1896). There is a substantial scientific literature on the cognitive and functional deficits induced when healthy (non-physician) adult volunteers remain awake for 24 to 40 hours (Durmer and Dinges, 2005; Harrison and Horne, 2000; IOM, 2006; Philibert, 2005). In addition, neuroimaging studies have confirmed that a night without sleep results in changes in brain functions that are associated with unstable and inaccurate performance on a wide range of cognitive tasks including attention, working memory, and executive functions such as problem solving and decision making (Bell-McGinty et al., 2004; Chee and Choo, 2004; Chee et al., 2006, 2008; Chuah et al., 2006; Drummond et al., 1999, 2000, 2005; Habeck et al., 2004; Lim et al., 2007; Portas et al., 1998; Thomas et al., 2000; Wu et al., 2006).

Although the majority of healthy adults exposed to wakefulness extended from 16-18 hours to 24-30 hours experience performance-impairing fatigue, there are substantial differences among individuals in the onset and magnitude of cognitive changes induced by a night without sleep (Doran et al., 2001; Leproult et al., 2003; Van Dongen et al., 2004). There is as yet no reliable objective biomarker for differential vulnerability to the effects of sleep loss, although a recent study suggests one possible genetic candidate (Groeger et al., 2008; Viola et al., 2007). Approaches designed to help in dividuals become aware of their own personal vulnerability to the cognitive effects of sleep loss, combined with information on how to prevent these effects, could form the basis of a more personalized fatigue management system (Dinges, 2004; Van Dongen et al., 2007). ACGME is encouraged to look into developing these systems based on the approaches developed in regulated transportation industries for their applicability to residency.


Current ACGME duty hours set an upper limit on duty hours of 24 hours with an additional 6 hours to allow adequate time for patient follow-up, didactic learning, and patient handovers (ACGME, 2003, 2004). ACGME proposed this 30-hour work limit (also referred to as a “long call” schedule or an extended duty period [Knauth, 2007]) in order “to address the effects of acute sleep loss” (ACGME, 2003). However, this limit does not adequately protect against acute sleep loss (Dinges, 2005; Philibert, 2005). An ACGME meta-analysis of 60 studies on the effects of sleep deprivation in 959 physicians found that “Sleep loss of less than 30 hours reduced physicians’ overall performance by nearly 1 standard deviation and clinical performance by more than 1.5 standard deviations” (Philibert, 2005, p. 1392). These very large effect sizes on a large sample of resident physicians leave little doubt that acute total sleep loss of 30 hours diminishes resident performance. Thus, it was concluded that allowing residents to stay awake for 30 hours on duty “may not completely guard against the negative effect of sleep loss on cognitive and clinical performance” (Philibert, 2005, p. 1392).

Both realistic patient simulator studies (Howard et al., 2003) and field studies of residents working extended duty periods (24 or more hours) have often found performance deficits post-call relative to pre-call (Eastridge et al., 2003; Friedman et al., 1971; Kahol et al., 2008; Leonard et al., 1998). However, some studies have not found such deficits (e.g., Jakubowicz et al., 2005; Jensen et al., 2004; Uchal et al., 2005). It is not possible to determine whether the latter studies had inadequate power to detect statistically significant differences as a function of work time. Since all of these field studies vary greatly in the rigor of their study designs and methods, what factors specifically contributed to the different outcomes cannot be determined. As in the more well-designed study by Landrigan and colleagues (2004), factors other than work duration per se (e.g., differences in degree of sleep deprivation) may have contributed to those findings in which extended duty hours reduced resident performance. The prospective controlled study of Landrigan and colleagues suggests that performance deficits associated with extended duty periods (24 + 6 hours) could adversely affect patient safety (Landrigan et al., 2004), and that the increased sleep time resulting from elimination of these long on-call duty periods might have helped to decrease attentional failures stemming from sleep loss during overnight work hours (Lockley et al., 2004). In contrast, retrospective studies, such as one by Ellman (2005), have not shown differences in patient outcomes.

The findings of the meta-analysis conducted by ACGME on the effects of sleep deprivation on physicians (Philibert, 2005) are consistent with previous reviews of the adverse effects of work periods beyond 24 hours on resident performance (Asken and Raham, 1983; Veasey et al., 2002), and with the report on the beneficial effects of limiting ICU interns’ continuous work periods to “approximately 16 hours” (Landrigan et al., 2004, p. 1839). Furthermore, a nationwide web-based survey of 2,737 interns found that extended work duration was associated with an increased risk of percutaneous injuries to interns (Ayas et al., 2006), and the more frequently interns experienced an extended work period of 24-30 hours, the more fatigue-related errors they reported (Barger et al., 2006b; see also Gander et al., 2000). Thus, considerable evidence as reviewed in this chapter now exists to suggest that the 2003 ACGME extended duty hour limit for residents of 24 + 6 continuous hours (ACGME, 2003) is likely to result in increased risks (via performance errors) to both patients and residents than shorter-duration work periods.

The same 2,737 interns queried for information on fatigue-related errors (Barger et al., 2006b) and percutaneous injuries (Ayas et al., 2006) during extended duty periods were also asked about motor vehicle incidents and crashes. Sleep deprivation from extended shifts contributed to significantly elevated risks of motor vehicle crashes, near-miss incidents, and incidents involving involuntary sleep while driving home from the hospital after an extended duty period (post-call) compared to drives home after non-extended work periods. The odds ratio for sleep-deprived drivers was 2.3 for crashes and 5.9 for near-miss incidents (Barger et al., 2005). An earlier smaller survey of residents found comparable results (Marcus and Loughlin, 1996). The willingness of residents to drive when they are drowsy may be, in part, associated with the effects of sleep deprivation on judgment. Recent studies have found that sleep loss can result in greater risk taking (Killgore et al., 2006; McKenna et al., 2007; Roehrs et al., 2004; Venkatraman et al., 2007).

Collectively, the research on resident physicians indicates that sleep loss associated with having to stay awake for an extended period (up to 30 hours or more)—rather than the performance of work per se—is likely the primary reason that neurobehavioral and cognitive performance degrades during residents’ extended duty periods. This conclusion is consistent with findings from laboratory studies of healthy adults showing that the adverse effects on cognitive performance of remaining awake 24 or more hours are found even when people are awake without working (Dinges and Kribbs, 1991; Harrison and Horne, 1999, 2000).


The seminal studies by the Harvard Work Hours, Health and Safety Group on resident duty hours in relation to both patient safety (Barger et al., 2006b; Landrigan et al., 2004) and resident safety (Ayas et al., 2006; Barger et al., 2005) focused on interns (first-year resident physicians) who are the least experienced and work more hours, and therefore are more prone to errors if not properly supervised. Interns are also more sleep deprived than other resident physicians. A national random sample survey that obtained data on sleep from 3,604 first-year (interns) and second-year residents during 1998-1999 found that interns reported obtaining significantly less sleep than second-year residents (Baldwin and Daugherty, 2004). Interns also had significantly more prolonged sleep deprivation and longer single periods without sleep than second-year residents. Additionally, this study found that residents averaging 5 or fewer hours of sleep per night were more likely to report serious accidents or injuries, conflict with other professional staff, use of alcohol, use of medications to stay awake, noticeable weight change, working in an impaired condition, and having made significant medical errors. Residents reporting more instances of inadequate supervision and occasions of working while impaired, as well as those who believed that they should have taken time off for illness but did not, also reported less sleep time and more sleep deprivation (Baldwin and Daugherty, 2004). While these results pre-date the current ACGME resident duty hour limits and consequently may not generalize to residents today, they are consistent with the more recent studies by the Harvard Work Hours, Health and Safety Group indicating that sleep deprivation in interns (the least experienced residents) poses an unsafe condition.


Continuous time spent performing work (referred to as time on task) may also increase the risk of accidents, but this is less clearly understood and documented than the contribution of acute sleep deprivation to the risk of accidents. Reviews across industries of the relative risk of accidents as a function of work hour duration (with attempts to adjust for exposure) generally conclude that the risk of accidents can begin to increase as time working exceeds 8 hours, and especially when it exceeds 12 hours, although the increases in risk after 12 hours of work are not always consistent or large (Caruso et al., 2004; Knauth, 2007; Nachreiner, 2001).

There are very few data, however, to inform work-hour guidelines in health care. A retrospective analysis of 411 recorded medical staff exposures to biological fluid at a university hospital with an emergency medicine residency program found a statistically significant increase (after adjusting for the number of workers per exposure) in exposures during 9-12 hours on duty (Macias et al., 1996). Two studies involving a total of 895 U.S. hospital staff nurses found an increase in self-reported errors and near-errors when work shifts were extended to 12.5 hours or longer (Rogers et al., 2004; Scott et al., 2006), although a smaller (and likely underpowered) study of nurses in Japan who were allowed to nap while working reported that 16-hour night shifts did not result in greater fatigue or difficulties concentrating than 8-hour night shifts (Takahashi et al., 1999).

A report by the National Institute of Occupational Safety and Health on overtime and extended work shifts has concluded that factors other than simply work duration per se contribute to the relationship between work duration and risk (Caruso et al., 2004). These factors are similar to the list at the beginning of this chapter and include shift start time, total hours worked in a week, rotation of work shifts between day and night work, and workload (e.g., Macdonald and Bendak, 2000). Moreover, there is much less information on the effects of work durations beyond 12 hours (Caruso et al., 2004), prompting the National Occupational Research Agenda Long Work Hours Team to propose a framework for future studies of long work hours, “including determinants, outcomes, and moderating factors of long work hours, suggesting that studies need to include more clear and complete descriptions of work schedules, worker characteristics, and the work environment, and need to consider a wider range of possible health, safety, social and economic outcomes for workers, families, employers, and the community. Additional studies are needed on vulnerable employee groups and those critical to public safety. More studies are also needed to develop interventions and test their effectiveness” (Caruso et al., 2006, p. 930).

Maximal Hour Limits per Shift

The evidence reviewed above supports the conclusion that performance is compromised by remaining awake beyond 16 hours (i.e., acute sleep deprivation). Therefore the extended duty shifts (24 + 6 hours) permitted in the current ACGME resident duty hour limits (ACGME, 2003) promote conditions for fatigue-related errors that pose risks to both patients and residents (Ayas et al., 2006; Barger et al., 2006a; Landrigan et al., 2004; Lockley et al., 2007). Limiting continuous work time to 16 hours would reduce these risks. A 16-hour continuous work limit is also reasonable in light of studies that equate the effects on performance from being awake more than 16 hours to the effects of 0.05 to 0.10 percent blood alcohol concentration (Arnedt et al., 2005; Dawson and Reid, 1997; Lamond and Dawson, 1999; Williamson and Feyer, 2000). While 16 hours of continuous work reflects a clear limit relative to safety, there is no compelling evidence that risks to patient safety increase from 8 to 12 hours of work, but some data from nurses suggest that risk may increase after 12 hours of work, although the work-related factors that contribute to this risk are unknown (Bollschweiler et al., 2001; Rogers et al., 2004).

In conclusion, the reviews of safe work hour limits and sleep deprivation indicate that either extended duty periods must be eliminated to improve patient safety relative to resident performance (Landrigan et al., 2004, 2007), or if extended duty periods of 24 + 6 hours are to remain an essential feature of resident training, provision for sleep following 16 hours of work will be needed, before the extended work period continues to 30 hours. The mitigating effects of sleep are discussed below.


This section reviews what is known about sleep obtained by residents during extended duty periods. As a fundamental biological function, sleep both stabilizes waking performance and enhances the ability to learn and remember (Huber et al., 2004; Lim and Dinges, 2008; Stickgold, 2005; Stickgold et al., 2000; Walker and Stickgold, 2006). Continued advances in neurobiology have identified circadian timing and homeostatic mechanisms in the brain (Fuller et al., 2006) that require sleep to be obtained daily in adequate quantity and quality to prevent the physiological and behavioral effects of sleep deprivation. Prevention of sleep deprivation in residents is regarded as among the most essential ways to manage fatigue and its risks (Buysse et al., 2003; Dawson and McCulloch, 2005; Gaba and Howard, 2002; Gabow et al., 2006; Horrocks et al., 2006; Landrigan et al., 2007; Parshuram, 2006; Veasey et al., 2002).

The committee observed that ensuring residents obtain adequate sleep during their scheduled workweeks is a feature currently missing in the ACGME duty hours and resident training culture. Reduced sleep periods are common in many residency programs (Baldwin and Daugherty, 2004). Since current efforts to educate residents about sleep and fatigue management are by themselves not sufficient to increase sleep durations (Arora et al., 2007), requirements for protected sleep periods should be a priority in any new ACGME duty hour limits. Residents should practice good sleep hygiene and learn the importance of avoiding fatigue-related errors by obtaining essential sleep both daily and weekly as a matter of professional responsibility.

Although reduction of resident duty hours alone is one way to achieve more sleep (Lockley et al., 2004, 2006, 2007), it is an indirect and inefficient way to increase sleep given the moderate correlation between resident work hours and sleep time (see Figure 7-1) (Baldwin and Daugherty, 2004; Lockley et al., 2004). Reducing work hours could limit the time available for educational training experiences of residents (Ludmerer and Johns, 2005), without necessarily increasing their sleep time. The most direct way to increase residents’ sleep time—to prevent fatigue risks due to sleep deprivation, and if necessary, allow residents to stay for extended duty periods up to 30 hours for educational purposes—would be to specifically mandate and protect periods of sleep for residents during extended duty. This approach has a long history in fatigue management in operational scenarios from other industries.

Protected Sleep Periods to Counter Fatigue During Extended Duty

The use of limited sleep periods typically between 10 minutes and 3 hours duration (i.e., naps) has been studied in many non-medical work contexts to manage fatigue risks and prevent performance errors. Studies have evaluated the use of planned naps (also called power naps) prior to and during night work, as well as during extended work periods up to 30+ hours (Akerstedt et al., 1989; Bonnefond et al., 2004; Bonnet, 1991; Caldwell and Caldwell, 1998; Dinges, 1989, 1992; Dinges et al., 1987, 1988; Driskell and Mullen, 2005; Kubo et al., 2007; Naitoh, 1992; Naitoh and Angus, 1989; Rosekind et al., 1994, 1995, 1997; Schweitzer et al., 2006; Takeyama et al., 2005). Strategic use of naps and longer sleep periods has been advocated as a fatigue countermeasure for residents during extended duty shifts (Veasey et al., 2002), and planned naps have been studied as fatigue countermeasures for physicians and nurses working 12-hour night shifts (Smith-Coggins et al., 2006). The vast majority of studies have found that naps and longer sleep periods (e.g., 4-5 hours) can help mitigate some of the effects of fatigue during night shifts and extended duty periods, suggesting that naps and longer sleep periods may be a valuable countermeasure to fatigue experienced by residents.

Although some residents take ad hoc naps during 24 + 6 hours extended duty periods when work demands permit, napping during extended duty periods is not addressed by the current ACGME duty hours. Studies indicate that some residents can obtain 1-3 hours of sleep when they nap on call under the current ACGME duty hours. These naps are usually during the nighttime (i.e., between 9 p.m. and 8 a.m.) (Arora et al., 2006, 2007; Lockley et al., 2004; Marcus and Loughlin, 1996). The sleep obtained can vary considerably by the year and subspecialty of the resident (Gabow et al., 2006). Although naps of 1 to 3 hours mitigate some of the effects of sleep deprivation, longer periods of sleep (4-8 hours) afford greater benefits for cognitive performance, even if some subjective fatigue and sleepiness may persist from sleep inertia (Belenky et al., 2003; Driskell and Mullen, 2005; Jewett et al., 1999; Van Dongen et al., 2003).

Benefits of Protected Sleep Periods

Reviews of studies that examined the effects of planned and protected nap sleep ranging in duration from 10 minutes to 4 hours in healthy adults concluded that naps and protected sleep periods can mitigate the effects of sleep loss on cognitive performance and subjective fatigue (Dinges and Broughton, 1989; Stampi, 1992). The longer the sleep, the greater are the benefits for performance, and the longer the benefits last. Thus, a recent meta-analysis revealed that a 15-minute nap could benefit fatigue and performance for a few hours, while a 4-hour sleep period could benefit performance for up to 10 hours (Driskell and Mullen, 2005). Relative to no sleep or a brief nap, 4 hours of protected sleep obtained at night during an extended duty period after about 16 hours of work could reduce fatigue and improve cognitive performance during a terminal portion of up to 10 hours of the 30-hour extended duty period.

Use of Protected Sleep Time by Residents During Extended Duty Periods

Two studies have evaluated the feasibility of deploying a protected pager-free sleep period at night during extended duty periods. One study designed to assess the effects of an on-duty protected nighttime period for sleep on first-year residents’ sleep and fatigue during 24 + 6 hour work periods provided supplementary night float coverage to interns from 12:00 a.m. to 7:00 a.m. The residents on extended duty could finish their work, forward their pagers to a night-float resident physician, and sleep while the other resident cared for their patients. This was compared to a 2-week period when no night-float physician was provided while residents were on call. The nights with protected sleep time and night-float physician coverage resulted in significantly longer sleep (mean sleep = 3.50 hours) and less subjective fatigue during call and post-call than the unprotected nights (mean sleep = 2.37 hours) (p < .001). Protected sleep time, or the minutes that an intern’s pager was “covered” by a night-float resident, was significantly associated with increased on-call sleep duration (r = 0.69). Thus, for each hour that an intern was covered by a night-float physician, the intern received 42 additional minutes of sleep (Arora et al., 2006). Sleep efficiency, which indicates how consolidated (unfragmented) sleep was, also improved for interns with the nap intervention.

Since use of protected night sleep schedule was at the discretion of interns (i.e., not mandated), adherence was relatively low. The authors reported that “interns on the nap schedule forwarded their pagers to the night-float physician on only 22% of available opportunities. In lieu of forwarding their pagers, interns with the nap schedule preferred to forward the care of only their cross-cover patients and to retain care of their own patients” (Arora et al., 2006, p. 795). Interns indicated they preferred the nap schedule when they were on call because it improved sleep, decreased fatigue, and/or gave them a greater ability to focus on care for their own patients without the additional workload associated with cross-covering patients. In other words, interns recognized the advantages of the night float—protected sleep schedule for increased sleep time and reduced fatigue—but they tended not to use it because of concerns about their patients and discontinuity of care (i.e., potential for risks posed by two transitions of care—one at the start and one at the end of their protected sleep period) (Arora et al., 2006). This suggests that teaching residents to work with the night-float physician and nurses as a team, oriented toward patient safety, might help them achieve the protected sleep they need without concerns that they have to remain awake for the sake of their patients.

A study published a decade earlier on the effects of a 4-hour (2 a.m.-6 a.m.) protected sleep time during a 36-hour on-call period found that sleep obtained during the protected period averaged 3.54 hours, but this was not significantly different from the mean sleep of 3.74 hours during the unprotected nights (Richardson et al., 1996). There were also no differences in measures of alertness and performance, despite impressions of attending physicians and supervising residents that the night-float system improved the alertness of the interns (Richardson et al., 1996). However, sleep efficiency (a measure of the consolidation of sleep) and slow wave sleep were markedly increased on the protected sleep nights due to fewer interruptions, indicating that the sleep was likely more restorative. Thus protecting the night sleep period by turning over pager calls to a night-float senior resident resulted in less sleep fragmentation from interruptions—and increased its recuperative value (Bonnet and Arand, 2003; Levine et al., 1987; Wesensten et al., 1999). In fact, there is evidence that merely the perceived risk of sleep interruption (e.g., wearing a pager while attempting to sleep) can fragment sleep, even if no interruption actually occurs (Richardson et al., 1996; Torsvall and Akerstedt, 1988). The study by Richardson and colleagues (1996) also illustrates that the amount of time off for a resident to spend in bed for sleep does not equate to actual sleep acquired. Their interns spent just under an average of 5 hours (295.4 minutes) in bed, but the sleep they obtained averaged 3.67 hours (220.1 minutes)—this is 75 percent sleep efficiency (i.e., total sleep time divided by total time in bed), which is the only estimate of the proportion of time residents could be expected to sleep during a protected nocturnal 5-hour sleep period in the hospital.

Thus, studies of a protected 4- to 7-hour sleep period during an extended duty schedule (Arora et al., 2006; Richardson et al., 1996) that allowed interns to decide whether they wanted to sleep during the protected night nap schedule found that adherence to the protected sleep period schedule was lower than expected. “On only 56 percent of the nights during which coverage was provided did interns make full use of the available time for sleep, i.e. by spending at least 4 hours in bed trying to sleep. The reasons cited in the diaries and in subsequent interviews suggested that with a guarantee of time for sleep at a later point, the covered interns preferred to catch up on incomplete work” (Richardson et al., 1996, p. 724).

The protected sleep schedule investigation of Richardson and colleagues (1996) used a less robust (between-subjects) study design and longer duty hours (36 hours) than the more recent report by Arora and colleagues (2006), which used a within-subjects design and the current ACGME duty hours (30 hours). Nevertheless, the two investigations support the conclusion that providing a protected nighttime period of 4-7 hours for sleep during an extended duty period (on-call schedule) can result in sleep that is less fragmented and somewhat longer than the 1-3 hours typically reported during residents’ extended duty periods (Arora et al., 2006, 2007; Lockley et al., 2004; Marcus and Loughlin, 1996). Since less fragmented, longer duration of sleep is more recuperative (Bonnet and Arand, 2003; Wesensten et al., 1999), protected nocturnal sleep periods during extended work schedules offer a means to prevent the severe fatigue-related performance deficits that can occur when remaining awake for 24-30 hours (Landrigan et al., 2004; Philibert, 2005).


While a protected nighttime sleep of up to 4-5 hours duration appears feasible as a way to prevent acute sleep deprivation in resident physicians during an extended duty period, the limited data available indicate that adherence to such a schedule was relatively poor (22-56 percent). Thus, resident adherence to a protected nocturnal sleep during a 30-hour extended duty period remains a major challenge, especially if they are given the option of deciding when to use the night-float residents or other mechanisms for coverage and obtain at least 4 hours of sleep. Resident unwillingness to obtain sleep when a protected period with pager handoff to others is available appears to be due to their concerns about patients they admitted, continuity of care (i.e., increased hand-offs), and their own workload (Arora et al., 2006; Richardson et al., 1996). Other factors, such as availability of sleep-conducive environments, staff willingness to not interrupt the resident’s sleep, and attending physicians’ encouragement of interns and other residents’ use of the protected nighttime sleep opportunity may also have played a role in adherence in these studies.

The barriers to resident use of protected nocturnal sleep periods during extended work periods could be overcome (1) by requiring residents to use the protected sleep periods for sleep and nothing else (especially if they are being paid and are required to remain in the hospital and work after the rest period), (2) by using more experienced physicians to cover patients for them so residents will have less cause for concern about their patients, (3) by sequestering residents in sleep-conducive environments, (4) by better management of resident workload, and (5) by ensuring that all hospital personnel avoid disrupting the resident’s sleep. A negative aspect of a protected sleep period at night is the possibility of sleep inertia, which refers to a period of grogginess and cognitive performance deficits immediately after awakening from naps and abbreviated sleep periods (Tassi and Muzet, 2000). Ironically, sleep inertia is more severe when sleep is more consolidated (i.e., less fragmented) and deeper, the type of sleep which is associated with better recuperation (Dinges, 1990; Dinges et al., 1985). Sleep inertia usually is overcome within 15-30 minutes by physical activity, environmental and social stimulation, and caffeine (Van Dongen et al., 2001). Once sleep inertia dissipates, alertness and cognitive performance can return to levels that reveal the benefits of the sleep for performance (Dinges, 1990; Dinges et al., 1988; Driskell and Mullen, 2005; Jewett et al., 1999). If 4-5 hours of sleep are obtained in the protected nocturnal period, improved alertness and performance generally will remain for the final 9-10 hours of a 30-hour extended duty period, thus reducing fatigue-related risks, improving learning while providing follow-up care, allowing attendance at didactic sessions, and improving the quality of handovers.


Chronic sleep deprivation occurs when the quantity and quality of sleep being obtained across days is insufficient to prevent daytime sleepiness, elevated sleep propensity, cognitive deficits, and other neurobehavioral problems (e.g., drowsy driving) produced by repeated days of inadequate recovery sleep (Dinges et al., 2005). Residents are most likely to experience chronic sleep deprivation when their daily recovery sleep opportunities are restricted to 7 hours or less in duration (Howard et al., 2002), regardless of the reason for the sleep restriction (e.g., unable to sleep during the daytime following a night shift, inadequate time off from scheduled work, moonlighting, non-work activities that reduce sleep time). The following section contains studies of healthy adults, but not specifically resident populations.

Total Daily Sleep Restriction and Cognitive Performance

Chronic sleep restriction can result in cumulative fatigue manifesting as reduced alertness and cognitive deficits in a sleep dose-response manner (Banks and Dinges, 2007; Dinges et al., 2005). Experimental studies of healthy adults chronically restricted to less than 8 hours time in bed at night reveal statistically reliable near-linear cumulative increases in cognitive deficits across days of sleep restriction (see Figure 7-2) (Belenky, 2003; Dinges et al., 1997; Van Dongen et al., 2003). The magnitudes of cognitive deficits found during sleep restriction were sleep dose-dependent between 3 hours and 7 hours per night, both within and between subjects. The less sleep provided each day, the more rapidly performance deficits accumulated over days (Belenky et al., 2003; Van Dongen et al., 2003). By the seventh day of nightly sleep restriction to between 4 and 6 hours time in bed, deficits in attention and unstable alertness (Doran et al., 2001; Dorrian et al., 2005) due to inadequate sleep were substantial. They reached levels equivalent to those found following one night of total sleep deprivation. Between days 9 and 12 of such restriction, performance deficits were equivalent to those found after two consecutive nights without any sleep (Van Dongen et al., 2003). These analyses took into account inter-individual differences in response to sleep loss to ensure the effects were not due solely to the sample populations’ containing more vulnerable individuals.

FIGURE 7-2. Repeated nights of sleep loss result in cumulative cognitive impairment.


Repeated nights of sleep loss result in cumulative cognitive impairment. Higher number of performance lapses indicate poorer performance and more unstable alertness. NOTE: B on the x-axis = baseline day.

A recent large-scale experimental study of chronic sleep restriction using split-sleep schedules (i.e., nighttime sleep plus a daytime nap) also found that reductions in total daily sleep resulted in a near-linear accumulation of impairment regardless of whether sleep was scheduled as a consolidated nocturnal sleep period or split into a nocturnal anchor sleep period and a daytime nap (Mollicone et al., 2008). Thus, the three largest sleep-dose response experiments published to date indicate that cumulative cognitive performance deficits occur in response to chronic restriction of sleep to less than 7-8 hours’ time in bed for sleep. This occurs even if subjects’ ratings and self-reports indicate they believe that are not very sleepy or fatigued (Banks and Dinges, 2007; Belenky et al., 2003; Dinges et al., 2005; Mollicone et al., 2008; Van Dongen et al., 2003). It is important to remember that time in bed does not equate to hours of sleep; hours of actual sleep will be less (Belenky et al., 2003; Richardson et al., 1996; Van Dongen et al., 2003).


Resident duty hours should protect against chronic sleep restriction over consecutive days and weeks of work. A number of work-hour factors are important for the prevention of cumulative sleep deprivation including provision of (1) adequate sleep time to recover after each work day, (2) additional sleep time to recover from extended duty periods (24 + 6 hours), (3) sleep time to recover from repeated days of restricted sleep due to night shift work, (4) sleep time to recover from work hours that result in sleep restriction due to averaging hours over weeks, and (5) off-duty periods with sufficient time for sleep to recover and still have time for personal (quality of life) activities. In the following section the adequacy of the current ACGME duty hour limits is assessed relative to the prevention of chronic sleep deprivation through these means.

Current ACGME Limits Relative to Chronic Sleep Deprivation

Current ACGME duty hours set an upper limit of 80 hours of work per week averaged over 4 weeks (this limit includes all in-house moonlighting) “to safeguard against the negative effects of chronic sleep loss” (ACGME, 2003, 2004). Because the 80 hours is an average, it permits weeks in which work hours could be in excess of 90 hours (Lockley et al., 2004). ACGME has tried to modulate wide swings in weekly hours by saying that the average should be within a rotation and not across rotations (e.g., not allowing a 40-hour ambulatory care rotation to be averaged with a 100-hour surgical one).

The consequences of having people work 80 hours a week on average have not been extensively studied. However, a report by the National Institute for Occupational Safety and Health concluded that “a pattern of deteriorating performance on psychophysiological tests as well as injuries while working long hours was observed across study findings, particularly with very long shifts and when 12-hour shifts combined with more than 40 hours of work a week” (Caruso et al., 2004, p. iv). Not every study, however, found that 84-hour workweeks adversely affect fatigue, performance, and health more than 40-hour workweeks do (e.g., studies of construction workers working 12 hours a day for 7 consecutive days) (Persson et al., 2003, 2006), although more carefully designed studies are needed on long work hours (Caruso et al., 2006). Nevertheless, the reduction of duty hours for ICU interns from a mean of 84.9 hours per week to a mean of 65.4 hours per week was associated with significant reductions in resident medical errors and potentially patient harm (Landrigan et al., 2004). However, as noted previously, it is not known if these benefits were the result of reductions in work duration, increased sleep time, or both, and what role concomitant factors—such as time of day of work—had in the results.

Obtaining Daily Sleep of Sufficient Duration

There are several reasons why the current ACGME limit of 80 hours per week may not protect against chronic sleep loss. Permitting the 80 hours to be an average over 4 weeks provides training flexibility within and between subspecialties, but it also results in the possibility of weeks in which residents can work more than 80 hours. In addition to the potential safety risks posed by overly long work hours (Caruso et al., 2004), the ACGME duty hours stipulate that “adequate time for rest and personal activities must be provided. This should consist of a 10-hour period provided between all daily duty periods after in-house call” (ACGME, 2003). The 10 hours off duty each day for recovery from acute fatigue is consistent with what is federally mandated for most transportation industries.

As noted above, scientific evidence indicates that the recovery potential of sleep in healthy individuals depends heavily on the duration of sleep—whether it is acquired in one continuous period or in a split-sleep period (Banks and Dinges, 2007; Belenky et al., 2003; Dinges et al., 2005; Mollicone et al., 2008; Van Dongen et al., 2003). Residents must have an adequate period for obtaining at least 7-8 hours of sleep per day to avoid chronic sleep restriction. Although studies have not addressed this, it is likely that for residents off duty outside the hospital for 10 hours, sleep will occupy no more than 80 to 95 percent (i.e., 8.0-9.5 hours) of the 10-hour nocturnal period (Ohayon et al., 2004), which leaves 0.5-2.0 hours for residents to transit to and from the hospital, eat, and have time for personal hygiene, domestic activities, and psychosocial requirements.

If the sleep period for residents is during the daytime (following night work), sleep time will be reduced by 2-4 hours due to circadian interface with daytime sleep as well as sleep disturbance from environmental factors such as noise and light (Akerstedt, 2003). If the off-duty period is less than 10 hours (which is possible under the current ACGME rules, which state “should provide” rather than “must provide” 10 hours off), recovery sleep time is reduced and the resident’s cumulative sleep debt grows. As a result, a failure to mandate and enforce the 10-hour “adequate rest” rule poses a challenge to the prevention of chronic sleep restriction and its consequences for cumulative performance impairments in residents, even more so after night shifts and overnight call than day shifts.

There is evidence that the current 80-hour work limit and 10-hour rest advisory are associated with reduced average daily sleep times. The intervention study by Landrigan and colleagues (2004) that restricted the work of interns to approximately 16 hours and eliminated extended duty periods (24 hours or more), reduced weekly hours from an average of 84.9 hours to an average of 65.4 hours (mean decrease in weekly work time of 19.5 hours), and increased sleep time from an average of 45.9 hours to an average of 51.7 hours (mean increase in weekly sleep time of 5.8 hours) (Lockley et al., 2004). Thus, restriction of weekly work hours from 80+ hours to less than 80 hours resulted in a daily average sleep time increase from a mean of 6.6 ± 0.8 hours per day to a mean of 7.4 ± 0.9 hours per day (Lockley et al., 2004). In terms of chronic sleep deprivation, this mean increase of 0.8 hours (48 minutes) more sleep per day is important, because it moves the daily sleep average above the minimum threshold of 7 hours per night that experimental studies have found are needed to prevent cognitive deficits from chronic sleep restriction (see Figure 7-2) (Belenky et al., 2003; Van Dongen et al., 2003). It is also consistent with survey findings of 3,604 residents taken before the ACGME duty hours went into effect, showing that the less sleep residents reported obtaining, the more they reported feeling sleep deprived. When sleep was less than 5 hours per night, they were significantly more likely to report serious accidents or injuries, working in an impaired condition, and having made significant medical errors (Baldwin and Daugherty, 2004).

Although reducing the weekly work hours of ICU interns in the Landrigan et al. study (2004) resulted in more sleep on average being obtained by the interns, there was, as noted previously, only a moderate correlation between work hours and sleep time (e.g., 33 percent of the variance in common) (Lockley et al., 2004). A similar outcome was obtained from the national random sample survey of 3,604 first- and second-year residents in the 1998-1999 training year (Baldwin and Daugherty, 2004). Figure 7-1 shows the relationship of resident sleep time to work hours found in both of these studies.

Although it is clear from these figures that work-hour reductions can result in increased sleep time, there is considerable variability in resident sleep time that remains unexplained, whether comparing a large cohort of residents from multiple subspecialties (Baldwin and Daugherty, 2004) or the same individuals within the same subspecialty and in the same hospital (Lockley et al., 2004). For example, in the study by Lockley and colleagues (2004), about one-third (7/20) of the interns averaged less than 7 hours of reported sleep a night when work hours were reduced during the study (see Figure 7-1B). Thus factors other than just work hours influence resident sleep time. The committee believes that identification of these factors, and ensuring that resident sleep time is protected and optimized, offer important avenues for prevention of fatigue-related resident errors and their risks to patient and resident safety.


Recovery Sleep Following Extended Duty Hours

The current ACGME rules have no minimum off-duty requirement that ensures residents obtain adequate recovery sleep following an extended duty period (24 + 6 hours) in the hospital. Instead, ACGME says only that time for rest and personal activities must be provided and should be 10 hours, which is identical to what is suggested following any work period with a shorter length (ACGME, 2003). However, the limited experimental data that exist regarding the amount of sleep needed to recover from an extended period of wakefulness involving a night of sleep deprivation suggests that 9 hours of time in bed at night is needed for recovery sleep. Six hours of time in bed does not result in full recovery even after five consecutive nights of 6 hours’ time in bed (Jay et al., 2007; Lamond et al., 2007).

In light of these data and the common experience of people needing to sleep longer after a period of total or partial sleep deprivation, resident fatigue could be reduced following extended duty periods (24 + 6 hours) if the mandatory time off duty is greater than the 10 hours suggested by ACGME. It may not be realistic to expect that residents will be able to obtain 9 hours recovery sleep in a 10-hour off-duty period, due to the time required by other non-work-related waking activities as well as the influence of circadian biology on sleep duration (Czeisler et al., 1980; Strogatz et al., 1986). Residents may end a 30-hour extended-duty work period in midday, then due to circadian influences over sleep propensity remain awake until the evening near habitual bedtime (Lavie, 1986; Strogatz et al., 1987). The addition of a 5-hour break for a 4-hour protected sleep period at night in the hospital (with a night float covering for the resident) during the extended (24 + 6 hours) duty period will not change these recovery sleep dynamics. The committee believes that a recovery period of at least 14 hours is necessary following the end of an extended duty period (that includes a 5-hour protected nighttime sleep). This recovery period should include a prohibition on starting work again before 6 a.m. the next morning. The combination of a minimum of 14 hours off duty and a start time of not before 6 a.m. the next day will ensure that residents can acquire at least 8-9 hours recovery sleep during the nocturnal period after an extended duty period in the hospital.

Recovery Sleep Following Night Shifts

As noted at the beginning of this chapter, a number of factors interact to influence the relationship of work and fatigue. The underlying assumption that cumulative work time is the predominant factor is not always substantiated. The time of day at which work takes place—especially work during the night shift when sleep propensity is highest—has proven to be one of the factors that can be more important to the risk of fatigue than the duration of work (Caruso et al., 2006; Feyer and Williamson, 1995).

Circadian biology has a profound effect on the night shift worker by increasing fatigue and decreasing performance at night during the work period and decreasing sleep duration during the daytime (Akerstedt, 2003; Rosa, 2001), and there are substantial inter-individual differences in the severity of these responses (Sack et al., 2007; Van Dongen, 2006). As a result, repeated, consecutive night shifts are associated with a growing sleep debt (due to daytime circadian restriction of sleep by 2-4 hours below sleep obtained at night), and with reduced productivity and increasing night-shift errors (Akerstedt, 2003; Folkard et al., 2005; Rosa, 2001). Studies of residents experiencing night shifts find results consistent with what has been observed in other areas. Daytime sleep duration after night shifts was significantly shorter than nighttime sleep duration following day shifts, and residents had decreased mood and alertness during night shifts compared to during day shifts (Cavallo et al., 2002, 2003).

A review of the risks posed by repeated day shifts versus repeated night shifts concluded that, on average, the risk of an incident on the night shift was more than double that on the day shift, and the risk increased progressively with each consecutive night shift up to four nights (Folkard et al., 2005). Thus, there is evidence that accident risk increases markedly over successive night shifts, but much less so over successive morning or day shifts. The committee notes therefore that using night shifts in place of extended duty periods (24 + 6 hours) will necessitate strategies to prevent the risks posed by fatigue at night and reduced daytime sleep.

Night shifts pose a performance risk to residents, and possibly a risk to patient safety, that could be prevented and mitigated by use of counter-measures (e.g., prophylactic naps prior to night-shift work) (Bonnefond et al., 2004; Bonnet et al., 2005; Horrocks et al., 2006; Knauth and Hornberger, 2003), as well as adequate time for recovery sleep following each night shift. ACGME duty hours currently have no special provisions for the duration of night-shift work—which is typically 12 hours—or for the duration of off-duty recovery time for sleep after night shifts (except for the 10 hours of “adequate rest” requirement). Permitting residents a 12-hour recovery period (rather than 10 hours) between consecutive night shifts would help ensure they either avoid or minimize any chronic sleep debt from cumulative sleep restriction across consecutive night shifts. It would do so by allowing more time for them to obtain both a sleep period in the morning immediately post night shift, and a prophylactic nap before the next nights.

There is evidence that night shifts that include backwards rotation in time (i.e., start earlier each consecutive day) are more fatiguing and disruptive of sleep, alertness, and performance than are nights shifts that begin at the same time each day or later each day (Akerstedt, 2003; Driscoll et al., 2007; Rosa, 2001). Since 12-hour night shifts are common in resident work schedules, a 12-hour off-duty period between night shifts would be appropriate to maintain a circadian work-rest schedule and avoid backward rotation between consecutive night shifts. Therefore both the need for sleep and synchronization with circadian rhythms would benefit from allowing residents a 12-hour off-duty recovery period following each night shift. The committee notes that such a schedule could present a challenge to the scheduling of an overlap of shifts to allow for handovers, which are discussed in the next chapter, unless the complementary day shift is longer than 12 hours or other members of the team have schedules that overlapped between shifts.

The scientific literature indicates that night-shift work also requires more days off for recovery due to the repeated challenge it poses to the circadian system and the sleep restriction it engenders, which occurs during the typical four to five nights worked by residents. A review of recovery following different work schedules suggested that as many as 3-4 days may be necessary for recovery after night-shift work due to the disturbances of normal circadian rhythms (Akerstedt et al., 2000). Therefore, recovery from the cumulative sleep loss and circadian perturbation of consecutive night shifts requires at least two nocturnal sleep periods when sleep can naturally occur uninterrupted and extend up to at least 9 hours of time in bed in order for recovery to occur (Jay et al., 2007; Lamond et al., 2007).

Recovery Sleep Following Six Consecutive Days of Work

ACGME duty hours currently require 1 day off in 7, averaged over a 4-week period. This averaging can result in residents working for 14 or even 21 consecutive days to get a consolidated 2-day weekend off. On weeks when residents do not get a day off from work, there is no possibility for ad libitum recovery sleep and little time for non-professional activities other than sleep. Without a day off in 7, as permitted by averaging days off, residents have a greater likelihood of chronic sleep restriction that can progress to levels of severe performance impairment. An experiment in healthy adults revealed that restriction of nightly sleep to between 4 and 6 hours resulted in cumulative performance deficits that reached levels equivalent to 48-64 hours of total sleep deprivation after 9 consecutive days without a day off for extra recovery sleep (Van Dongen et al., 2003). These analyses took into account inter-individual differences in response to sleep loss to ensure that the effects were not due solely to more vulnerable individuals. The findings are shown in Figure 7-2A. They suggest that prolonged work hours without a day off weekly can lead to substantial performance risks.

A review of recovery from work on various schedules concluded that those who work long shifts require more frequent days off for recovery (Akerstedt et al., 2000). Recent studies of recovery from chronic sleep restriction indicate that at least 1 full recovery day (24 hours) for ad libitum sleep time is needed to ensure that performance deficits do not become excessive and pose a risk to performance when work schedules that induce sleep restriction exceed a 6-day period (Belenky et al., 2003; Van Dongen et al., 2003). This can only be accomplished in residents if the 1 day a week is guaranteed in duty hour limits.

To ensure that residents do not continue to accumulate fatigue and performance deficits from chronic sleep restriction due to a single ad libitum recovery day each week, duty hour limits should also ensure that residents have at least one 48-hour period free of duty each month. This will reduce the practice of residents trading work days to consolidate 2 days off in a single weekend.


After reviewing the literature on work duration, sleep, and human performance in this chapter, the committee concludes that there are specific adjustments to the current ACGME resident duty hours that would help prevent and reduce resident fatigue, and thereby enhance resident performance and reduce risks to both patients and residents. The major rationales behind the recommendations are the following: (1) work duration should be limited because human performance degrades after 16 hours whether one is working or not; (2) sufficient time for sleep should be incorporated into daily and weekly work schedules to prevent acute and chronic sleep deprivation, respectively, and allow recovery from accumulated sleep debt; and (3) when extended duty periods are considered an essential aspect of resident training and continuity of care, a protected sleep period should be provided during the extended duty period. These changes will reduce the chances of residents providing patient care when their performance may be less than optimal due to fatigue.

The committee has concluded that the best way to prevent the problem of resident fatigue was to ensure adequate sleep is obtained by residents, and this can be done without changing the overall total of allowable hours of work in a week, retaining its putative training value. Hence the recommendations below, derived from the evidence in this chapter, are focused more on providing predictable and protected time for recovery sleep than on work hours per se. Appendix B provides sample monthly schedules for an individual resident comparing the application of current ACGME duty hour limits and the committee’s recommendations that illustrate enhanced regularity of days off and protected sleep during extended duty periods under the latter.

Focusing on resident off-duty time (for sleep), more so than on-duty time for work, to prevent fatigue-related risks is a novel way to view the necessary balance between the need for long resident duty hours inherent in their intensive training, and the goal of preventing fatigue as a condition of risk. New scientific evidence on the biological causes of fatigue (e.g., sleep need, circadian rhythms) has guided the committee’s recommendations (below) to focus on the need to enhance recovery sleep opportunities for residents, something emphasized in many of the reviews on how to reduce resident fatigue (Baldwin and Daugherty, 2004; Buysse et al., 2003; Cavallo and Mallory, 2004; Gaba and Howard, 2002; Landrigan et al., 2007; Lockley et al., 2006; Parshuram, 2006; Veasey et al., 2002). The committee emphasizes that it will be necessary to educate both residents and faculty on the need for taking advantage of these sleep opportunities as a patient safety and quality-of-care issue.

Recommendation 7-1: ACGME should adopt and enforce requirements for residency training that adhere to the following principles: duty hour limits and schedules should promote the prevention of sleep loss and fatigue; additional measures should mitigate fatigue when it is unavoid able (e.g., during night work and extended duty periods); and schedules should provide for predictable, protected, and sufficient uninterrupted recovery sleep to relieve acute and chronic sleep loss, promote resident well-being, and balance learning requirements. Programs should design resident schedules using the following parameters:

  • Duty hours must not exceed 80 per week, averaged over 4 weeks.
  • Scheduled continuous duty periods must not exceed 16 hours unless a 5-hour uninterrupted continuous sleep period is pro vided between 10 p.m. and 8 a.m. This period must be free from all work and call, and used by the resident for sleep in a safe and sleep-conducive environment. The 5-hour period for sleep must count toward total weekly duty hour limits. Following the protected sleep period, a resident may continue the extended duty period up to a total of 30 hours, including any previous work time and the sleep period.
  • Residents should not admit new patients after 16 hours during an extended duty period.
  • Extended duty periods (e.g., 30 hours that include a protected 5-hour sleep period) must not be more frequent than every third night with no averaging.
  • After completing duty periods, residents must be allowed a con tinuous off-duty interval of
    • A minimum of 10 hours following a daytime duty period that is not part of an extended duty period,
    • A minimum of 12 hours following a night float or night shift work that is not part of an extended duty period, and
    • A minimum of 14 hours following an extended duty period, and residents should not return to service earlier than 6 a.m. the next day.
  • Night-float or night-shift duty must not exceed four consecutive nights and must be followed by a minimum of 48 continuous hours off duty after three or four consecutive nights.
  • At least one 24-hour off-duty period must be provided per 7-day period without averaging; one additional (consecutive) 24-hour period off duty must be provided to ensure at least one continu ous 48-hour period off duty per month.
  • In exceptional circumstances requiring the resident’s physical presence to ensure patient safety or to engage in a critical learn ing opportunity, program faculty may permit, but not require, residents to remain on duty beyond the scheduled time; programs must record for ACGME review the nature of each exception allowed. These exceptions are not to become routine practice. Residency Review Committees should determine at the time of program re-accreditation whether the documented exceptions to scheduled duty hours warrant citation.
  • The ACGME should develop criteria for granting individual programs waivers from one or more of the above scheduling parameters; such criteria should be formulated only to accom modate rare, well-documented circumstances in which patient safety and/or educational requirements of specific programs outweigh the advantages of full compliance with the committee’s recommendations and cannot be addressed by means other than the requested waiver(s); programs that are granted waivers (if any) and the nature of those waivers should be posted on the public access portion of the ACGME website. Included in the application for waiver should be a long-term plan that articu lates how the program will work to avoid a permanent need for the requested waiver. All waivers should be monitored and re viewed on an annual basis to determine suitability for renewal.
  • Programs should provide annual formal education for residents and staff on the adverse effects of sleep loss and fatigue and on the importance of and means for their prevention and mitigation.
  • Sponsoring institutions and programs should ensure that their practices promote and ensure that residents take the required sleep during extended duty periods.


As noted above, the intent of the committee’s recommendations for changes to resident duty hours was to prevent fatigue when possible and to provide measures to relieve both acute and chronic sleep deprivation, recognizing that some fatigue may be inevitable when attempting to provide service in the hospital 24 hours a day, 7 days a week. Because of the diversity of specialty education needs, program sizes, and patient populations, the committee believes some flexibility in duty hour rules is needed for programs to design their own resident training schedules within certain limits supported by the evidence in this chapter.

Maximum Hours of Work per Week

The committee has retained the current 80-hour per week duty hour maximum averaged over 4 weeks (Table 7-1) rather than reduce it or eliminate averaging. This preserves flexibility for each specialty and program site to have what they determine are sufficient hours to achieve their learning goals just as they have now under the 2003 rules. As is currently the case, the committee does not believe all specialties and rotations will require this lengthy workweek. Programs with certain constraints related to number of residents or patient characteristics may require longer hours than other programs. While the 80-hour limit does not have an empirical evidence base demonstrating that it is better or worse for educational outcomes or patient safety than other limits (e.g., 72 or 56 hours), it has been in place for nearly 20 years in some specialties and New York State, and the rule has been in place nationally for 5 years. Board certification results are just beginning to be available nationally across specialties to monitor the achievement of residents (see Chapter 4). Without additional documentation of the actual hours worked by different residency programs and their outcomes, the committee was reluctant to reduce the 80-hour framework. Furthermore reducing hours of work could limit the time for education and training experiences of residents without resulting in increased hours of sleep since, as noted earlier in this chapter, reducing total work hours alone is an inefficient and indirect way to increase sleep time (Baldwin and Dougherty, 2004; Lockley et al., 2004; Ludmerer and Johns, 2005). Studies on the effects of long workweeks are limited in health care and other industries (Landrigan et al., 2004; Persson et al., 2003, 2006), and when performance improves with a shorter workweek, it is not clear if it is due to hours of service or hours of additional sleep obtained.

TABLE 7-1. Comparison of IOM Committee Adjustments to Current ACGME Duty Hour Limits.


Comparison of IOM Committee Adjustments to Current ACGME Duty Hour Limits.

Any of the Residency Review Committees, which set educational standards for each specialty in conjunction with ACGME, may choose to create more restrictive duty hour limits if it considers changes to be necessary for its particular circumstances, such as the severity of patient cases and the constancy of high-intensity work. For example, this has been done in emergency medicine, which limits shift length to 12 hours, totaling 60 hours per week, plus 12 hours for education (ACGME, 2007b); the committee does not recommend any change in the hours for emergency medicine.

Maximum Shift Length

Although the scientific evidence base establishes that human performance begins to deteriorate after 16 hours of wakefulness, the committee did not believe that limiting all shifts to a maximum of 16 hours would leave sufficient flexibility to address the educational needs of all specialties. So extended duty periods of up to 30 hours, the current limit, are allowed with the inclusion of a 5-hour sleep period to address acute sleep deprivation (Table 7-1). The committee anticipates that its new recommendations regarding sleep within extended duty periods and the need for protected and predictable off-duty times for sleep will result in residents being less likely to work in excess of 80 hours per week because the frequency of 30-hour duty period may no longer be averaged (Table 7-1), and when those duty periods occur, a 5-hour period for sleep is incorporated (e.g., see schedules B-1a, B-1b, B-2a, and B-2b in Appendix B as examples).

The committee is not mandating that duty periods be 16 hours, but rather is setting 16 hours as the upper limit for continuous work without a protected sleep period. Only after a 5-hour sleep period can work continue. ACGME currently suggests a 10-hour separation between shifts of any length regardless of whether the work occurs in the day or night time. The committee recognizes that this required 10 hours off between daytime work periods will likely result in routine shift lengths shorter than 16 hours (e.g., 10 to 14 hours per day).

Although under the proposed changes 14-hour days could routinely be scheduled along with the 10-hour interval between shifts to achieve a schedule less than 80 hours a week, the committee is cognizant that this may be a draining schedule whether every day is 14 hours or 14-hour days are in combination with extended duty periods. Bertrand Bell, in developing the rationale behind the 80-hour week for New York State commented, “It is reasonable for residents to work a 10-hour day for 5 days a week. It is humane for people to work every fourth night” (Bell, 2003, p. 40). Samples of schedules that might be drawn up following the committee’s recommendations are compared with schedules following current ACGME duty hour limits (see Appendix B); other variations are possible.

Maximum Time Off Between Scheduled Shifts

The committee believes 10 hours is a minimum rest break for residents to receive adequate sleep after daytime work periods and recommends that this become mandatory. Variable amounts of time 10, 12, and 14 hours are recommended as the minimum time off after day, night, and extended duty periods (Table 7-1), respectively, as guided by the evidence for the periods of recovery necessary depending on the time of work during the day and our circadian biology. The committee decided on a requirement of 14 hours off after an extended duty period coupled with a requirement that a resident not return to service before 6 a.m. on the following day. This combination was selected to allow flexibility in the start and ending times of extended duty periods (e.g., 7 a.m. to 1 p.m. on the next day, or 10 a.m. on day 1 until 4 p.m. on day 2) yet ensure that residents are able to have a nocturnal period of sleep before returning to work after an extended duty period. As noted earlier, it is estimated that a person needs 9 hours of time in bed at night to gain sufficient recovery sleep after being up the previous night (Jay et al., 2007; Lamond et al., 2007).

Maximum In-Hospital Frequency of Night Shift

The committee also recommends that 48 hours be given off after three or four consecutive night shifts to provide opportunities for sleep to offset the increasing performance deficits associated with multiple nights of work (Akerstedt, 2003; Folkard et al, 2005; Rosa, 2001). The ACGME does not currently limit the number of consecutive night shifts (Table 7-1).

Mandatory Time Off Duty

The committee recommends 5 rather than 4 days off per month and eliminating averaging of the days off per week to ensure sufficient time weekly for recovery sleep (Table 7-1). Elimination of averaging days off will help prevent schedules where residents may go 2 or 3 weeks without a day off, whether scheduled by the program director or due to residents trading days to have multiple consecutive days off. Offering a fifth day per month that creates a consolidated 48-hour period further enhances recovery from sleep debt and gives residents 2 consecutive days off. The committee recognizes that scheduling a Saturday-Sunday break for every resident, although it may be preferable for resident personal life, may not always be possible, but encourages programs to have at least one of the days always be a weekend day (e.g., Friday-Saturday or Sunday-Monday) so that residents are available when other family and friends are off work to provide some work-life balance.

Protected Sleep Period During Extended Duty Period

The committee debated the best course for continuity of patient care, educational purposes, and addressing fatigue when deliberating on whether to maintain the 30-hour extended duty period. It also debated the appropriate length of a rest period that would be observed by residents yet maximally address acute fatigue so the residents would perform well for the duration of the duty period. A 5-hour period with the expectation of a 4-hour period of sleep best fit the evidence on the amount of sleep required that would improve performance for the balance of the duty period (Driskell and Mullen, 2005). Shorter nap periods, such as 2 hours, provide some relief but are inadequate. Longer periods (e.g., 7-10 hours), while de sirable appeared impractical for two reasons: (1) there would be too many sleep hours counted against duty hour limits, and (2) residents would prefer to and would be inclined to leave the hospital to sleep in their own beds for such a long break, driving home while overtired and not obtaining sufficient sleep before they returned regardless of how the official schedule is drawn. Thus, the committee recommends incorporating a 5-hour sleep period in any duty period over 16 hours and recommends that this sleep period be counted as part of total duty hours.

Given a 5-hour protected sleep period, the committee understands that a resident would only likely obtain up to about 4 hours of sleep (e.g., Arora et al., 2006; Richardson et al., 1996). However a benefit of a mandated rather than optional sleep period is that it is predictable and protected and will improve performance for the balance of the extended duty period (Driskell and Mullen, 2005). The committee recognizes that some residents have in the past preferred to use the protected time to catch up on paperwork (Arora et al., 2006). Consequently it both recommends training for residents in sleep hygiene, so that they will understand the importance of that sleep break, and reductions in resident workloads to make them more manageable within reduced duty hours. Workload limits are discussed in more detail in Chapter 3, but the committee also recommends here confining admissions to the first 16 hours that a resident is on duty during an extended duty period. The extended duty period is one whose limit is frequently violated (see Chapter 2) and limiting admissions will help residents complete their work before the end of their shift.

Educating Residents and Faculty on Benefits of Sleep

The committee does not endorse any specific educational program on the consequences of sleep loss and fatigue management but agrees that residents, their supervisors, and institutional leaders all need to be aware of the patient and resident safety implications of acute and chronic sleep deprivation and act on solutions that will work in their training environment. Both AHRQ and ACGME recommend that residents receive sleep education (ACGME, 2003; Jha et al., 2001), and the education should emphasize obtaining sound sleep and preventing sleep loss as primary tactics for reducing the likelihood of fatigue and its risks (e.g., Horrocks et al., 2006). ACGME might advance the spread of best practices in sleep education by developing or endorsing existing educational material (e.g., high-quality materials developed by individual residency programs), Duke University LIFE Curriculum (Learning to Address Impairment and Fatigue to Enhance Patient Safety), or SAFER—Sleep Alertness, and Fatigue Education in Residency (SAFER Task Force, 2007). A number of articles have emphasized that changing residency culture that perceives long hours as the mark of dedication and professionalism will require more than a single education session, just as it requires more than duty hour regulation (Arora et al., 2008; Owens et al., 2008; Parthasarathy et al., 2007).

Limit on Admitting Patients During Extended Duty Period

Current ACGME rules allow extended duty periods of 30 hours with no provision for sleep, but they do limit when residents may admit patients to the first 24 hours. The balance of the 30 hours under current practice is 6 hours for educational purposes, completion of patient care duties, and handovers of care responsibilities. The 30-hour duty period is frequently violated, with residents often citing insufficient time to complete their workload, patient care responsibilities, and educational requirements. The committee now recommends that residents on an extended duty period only admit patients during the first 16 hours before they take the 5-hour sleep period and then continue on until the end of the 30 hours; the committee believes that allowing residents to admit patients only in the first 16 hours will help them complete their work and permit them to leave the hospital within the time constraints of a 30-hour duty period. Caring for fewer patients would also provide more time for an in-depth evaluation of each patient which has educational advantages as well. Even after the sleep period, residents would now have up to 9 hours to obtain needed test results on these new admissions and hence would have a greater likelihood of receiving the necessary information to make management decisions about a patient they admitted before handing the patient over to another team member. Having a mandatory sleep period will necessitate that either a night float, hospitalist service, or some other mechanism be available to admit new patients and cover the sleeping residents’ patients after residents reach 16 hours on duty.

Adherence to Recommendations and Opportunities for Innovation

In Chapter 2 the committee recommended that ACGME continue to monitor duty hours. The ACGME in employing the substantial compliance model should be firmly intolerant of any systematic duty hour violations. The committee expects that there will not be routine violations, either scheduled or ad hoc, of duty hours. This expectation applies whether it involves a duty period of 10, 16, or 30 hours, or any other variation. The committee expects that there will be rare occasions when the stability of the patient is in question or an exceptional learning opportunity will present itself, and that the application of duty hour rules should be flexible enough to accommodate these instances. However, the committee believes these occasions should be well documented at the local level, reviewed for reasonable cause by the ACGME, and the type and number of exceptions made public by ACGME on its website to determine if these instances can provide guidance for future duty hour adjustments in general and on a specialty specific basis. Having the frequency and the reasons for exceptions available will increase transparency and help address concerns about conflict of interest in the ACGME monitoring process (CIR/SEIU, 2007). Consequently, residency programs must document all such incidents and provide explanations of the “reasonable” violations. Additionally, it is also expected that unusual learning opportunities will mean that supervision is present during those times to mitigate resident fatigue.

ACGME currently also has a process for allowing broader exemptions to duty hour rules for specific programs and for purposes of innovation. ACGME should publish criteria for granting such exemptions, which should be rare, and methods for monitoring them closely for possible increased risks to patient or resident safety. ACGME should list on its website which programs have exemptions and the reasons for the exemptions. The committee encourages the development of innovative projects for improving patient and resident safety including those involving scheduling. These projects must have a rigorous data collection and analysis framework that will advance learning and safety under different duty hour scenarios beyond what is currently known, given that the committee has offered recommendations rooted in a strong evidence base from the human performance and sleep literature. Currently the ACGME is testing scheduling approaches and limits, including different napping strategies during extended duty periods, limiting the separation between shifts to 8 hours, and investigating whether duty hour limits should apply to residents in their last year of training.1 The committee encourages ACGME and the respective Residency Review Committees to document why they need to continue 30-hour duty periods as it was the most contentious part of the duty hour debate before the committee.

Implementation of all of the committee’s duty hour recommendations should include a national evaluation of the following: the changes individual programs make; the extent to which the recommended changes to ACGME duty hour limits actually result in increased sleep for residents and maintenance of alertness and performance during work; the costs of implemented changes; the effect on labor supply and patient coverage; and which specific schedules with protected sleep programs more or less effectively promote sleep and alertness. ACGME and other stakeholders should foster research studies across multiple institutions to examine the effects of duty hour changes and practices (e.g., improved handovers) on (1) serious medical errors and preventable adverse events, (2) resident safety, and (3) resident educational outcomes and well-being. In the interest of both resident and patient safety, it will also be valuable and cost effective to learn from research sponsored by the National Institutes of Health, the National Aeronautics Space Administration, the Department of Defense, and other agencies about ways to determine objectively when residents are impaired by fatigue, since perception of one’s own fatigue is often inaccurate (Van Dongen et al., 2003).


There is no recent national assessment of the degree to which residents and fellows are moonlighting. Anecdotal reports say that duty hour cutbacks created their own demand for moonlighting within teaching hospitals to cover services around the clock;2 however, this would apply primarily to residents beyond the first year. Two studies, one institution-specific and one specialty-specific, conflict on whether moonlighting by residents has increased or remained at the same level since the 2003 limits (Cull et al., 2006; Dola et al., 2006). Currently, ACGME only requires that “in-house” or internal moonlighting for patient care be considered part of the 80-hour weekly limit on duty hours; that moonlighting requires prospective, written permission from the program director; and that resident performance be monitored to ensure no adverse effects that may lead to withdrawal of permission (ACGME, 2003, 2007a). Despite requirements for advance permission, these processes are not always followed; leading to the conclusion that more residents may be moonlighting than their supervisors know (Cull et al., 2006). At this time ACGME Residency Review Committees or individual programs may add additional requirements such as forbidding moonlighting during on-call months or forbidding moonlighting altogether. The Federation of State Medical Boards and legislative proposals have advocated more restrictions on moonlighting (Federation of State Medical Boards, 2007;, 2005).

The committee concludes that all moonlighting for patient care, whether at the training facility (internal moonlighting) or elsewhere (external moonlighting), should come within the 80-hour weekly limit and that all other duty hour parameters should apply. Currently the 80-hour limit applies only to internal moonlighting. Having this provision in the resident contract (if the program chooses to permit moonlighting) gives a clearer expectation that the resident’s primary duty is to the limits set by his or her training program and to patient safety. Residents have a responsibility to be ready for work and not take on too many additional obligations if these activities could interfere with their capacity to learn and to provide safe patient care. Additionally, sponsoring institutions must not require that residents and fellows engage in moonlighting to cover cutbacks in hours on other services. So that moonlighting does not undermine the intent of limiting duty hours, the committee recommends the following.

Recommendation 7-2: The ACGME should immediately amend its current requirements on moonlighting by

  • Requiring that any internal and external moonlighting for pa tient care adhere to the duty hour limits listed above (e.g., 80 hours and all other limits), even if the program has received an exception to schedule longer hours, and
  • Requiring that sponsoring institutions, if they choose to permit moonlighting, include provisions in resident contracts that (1) a resident must request prospective, written permission from the program director for moonlighting; and (2) resident perfor mance will be monitored to ensure that there is no adverse effect of moonlighting on resident performance.


  • ACGME (Accreditation Council for Graduate Medical Education). 2003. Common program requirements for duty hours . Chicago, IL: ACGME.
  • ———. 2004. Report of the workgroup on resident duty hours and the learning environment . Chicago, IL: ACGME.
  • ———. 2007. a. ACGME institutional requirements: Checklist, effective: July 1, 2007 . http://www​​/acWebsite/irc/IRCheckList-07.pdf (accessed February 28, 2008).
  • ———. 2007. b. ACGME program requirements for graduate medical education in emergency medicine . Chicago, IL: ACGME.
  • Akerstedt, T. 2003. Shift work and disturbed sleep/wakefulness. Occupational Medicine 53(2):89-94. [PubMed: 12637592]
  • Akerstedt, T., L. Torsvall, and M. Gillberg. 1989. Shift work and napping. In Sleep and alertness: Chronobiological, behavioral and medical aspects of napping, edited by D. F. Dinges, editor; and R. J. Broughton, editor. . New York: Raven Press. Pp.205-220.
  • Akerstedt, T., G. Kecklund, M. Gillberg, A. Lowden, and J. Axelsson. 2000. Sleepiness and days of recovery. Transportation Research Part F 3:251-261.
  • Arnedt, J. T., J. Owens, M. Crouch, J. Stahl, and M. A. Carskadon. 2005. Neurobehavioral performance of residents after heavy night call vs after alcohol ingestion. JAMA 294(9):1025-1033. [PubMed: 16145022]
  • Arora, V., C. Dunphy, V. Y. Chang, F. Ahmad, H. J. Humphrey, and D. Meltzer. 2006. The effects of on-duty napping on intern sleep time and fatigue. Annals of Internal Medicine 144(11):792-798. [PubMed: 16754921]
  • Arora, V. M., E. Georgitis, J. N. Woodruff, H. J. Humphrey, and D. Meltzer. 2007. Improving sleep hygiene of medical interns—Can the sleep, alertness, and fatigue education in residency program help? Archives of Internal Medicine 167(16):1738-1744. [PubMed: 17846392]
  • ———. 2008. Improving sleep hygiene—Reply. Archives of Internal Medicine 168(11):1230. [PubMed: 18541837]
  • Asken, M. J., and D. C. Raham. 1983. Resident performance and sleep-deprivation—A review. Journal of Medical Education 58(5):382-388. [PubMed: 6343602]
  • Ayas, N. T., L. K. Barger, B. E. Cade, D. M. Hashimoto, B. Rosner, J. W. Cronin, F. E. Speizer, and C. A. Czeisler. 2006. Extended work duration and the risk of self-reported percutaneous injuries in interns. JAMA 296(9):1055-1062. [PubMed: 16954484]
  • Baldwin, D. C., Jr., and S. R. Daugherty. 2004. Sleep deprivation and fatigue in residency training: Results of a national survey of first- and second-year residents. Sleep 27(2):217-223. [PubMed: 15124713]
  • Baldwin, D. C., Jr., S. R. Daugherty, R. Tsai, and M. J. Scotti. 2003. A national survey of residents’ self-reported work hours: Thinking beyond specialty. Academic Medicine 78(11):1154-1163. [PubMed: 14604880]
  • Banks, S., and D. F. Dinges. 2007. Behavioral and physiological consequences of sleep restriction. Journal of Clinical Sleep Medicine 3(5):519-528. [PMC free article: PMC1978335] [PubMed: 17803017]
  • Barger, L. K., B. E. Cade, N. T. Ayas, J. W. Cronin, B. Rosner, F. E. Speizer, and C. A. Czeisler. 2005. Extended work shifts and the risk of motor vehicle crashes among interns. New England Journal of Medicine 352(2):125-134. [PubMed: 15647575]
  • Barger, L. K., N. T. Ayas, B. E. Cade, J. W. Cronin, B. Rosner, F. E. Speizer, and C. A. Czeisler. 2006. a. Impact of extended-duration shifts on medical errors, adverse events, and attentional failures. PLoS Medicine 3(12):e487. [PMC free article: PMC1705824] [PubMed: 17194188]
  • ———. 2006. b. Impact of extended-duration shifts on medical errors, adverse events, and attentional failures. PLoS Medicine 3(12):2440-2448. [PMC free article: PMC1705824] [PubMed: 17194188]
  • Belenky, G., N. J. Wesensten, D. R. Thorne, M. L. Thomas, H. C. Sing, D. P. Redmond, M. B. Russo, and T. J. Balkin. 2003. Patterns of performance degradation and restoration during sleep restriction and subsequent recovery: A sleep dose-response study. Journal of Sleep Research 12:1-12. [PubMed: 12603781]
  • Bell, B. M. 2003. Reconsideration of the New York state laws rationalizing the supervision and the working conditions of residents. Einstein Journal of Biological Medicine 20(1):36-40.
  • Bell-McGinty, S., C. Habeck, H. J. Hilton, B. Rakitin, N. Scarmeas, E. Zarahn, J. Flynn, R. DeLaPaz, R. Basner, and Y. Stern. 2004. Identification and differential vulnerability of a neural network in sleep deprivation. Cerebral Cortex 14(5):496-502. [PubMed: 15054065]
  • Bollschweiler, E., A. Krings, K. H. Fuchs, G. Pistorius, T. Bein, U. Otto, E. Muhl, U. Backes-Gellner, and A. H. Hölscher. 2001. Alternative shift models and the quality of patient care. An empirical study in surgical intensive care units. Langenbeck’s Archives of Sur gery 386:104-109. [PubMed: 11374042]
  • Bonnefond, A., P. Tassi, J. Roge, and A. Muzet. 2004. A critical review of techniques aiming at enhancing and sustaining worker’s alertness during the night shift. Industrial Health 42(1):1-14. [PubMed: 14964612]
  • Bonnet, M. H. 1991. The effect of varying prophylactic naps on performance, alertness and mood throughout a 52-hour continuous operation. Sleep 14(4):307-315. [PubMed: 1947593]
  • Bonnet, M. H., and D. L. Arand. 2003. Clinical effects of sleep fragmentation versus sleep deprivation. Sleep Medicine Reviews 7(4):297-310. [PubMed: 14505597]
  • Bonnet, M. H., T. J. Balkin, D. F. Dinges, T. Roehrs, N. L. Rogers, and N. J. Wesensten. 2005. The use of stimulants to modify performance during sleep loss: A review by the sleep deprivation and stimulant task force of the American Academy of Sleep Medicine. Sleep 28(9):1163-1187. [PubMed: 16268386]
  • Buysse, D. J., B. Barzansky, D. Dinges, E. Hogan, C. E. Hunt, J. Owens, M. Rosekind, R. Rosen, F. Simon, S. Veasey, and F. Wiest. 2003. Sleep, fatigue, and medical training: Setting an agenda for optimal learning and patient care. Sleep 26(2):218-225. [PubMed: 12683483]
  • Caldwell, J. A., Jr., and J. L. Caldwell. 1998. Comparison of the effects of zolpidem-induced prophylactic naps to placebo naps and forced rest periods in prolonged work schedules. Sleep 21(1):79-90. [PubMed: 9485535]
  • Caruso, C. C., E. M. Hitchcock, R. B. Dick, J. M. Russo, and J. M. Schmit. 2004. Overtime and extended work shifts: Recent findings on illnesses, injuries, and health behaviors . Cincinnati, OH: National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention, U.S. Department of Health and Human Services.
  • Caruso, C. C., T. Bushnell, D. Eggerth, A. Heitmann, B. Kojola, K. Newman, R. R. Rosa, S. L. Sauter, and B. Vila. 2006. Long working hours, safety, and health: Toward a national research agenda. American Journal of Industrial Medicine 49(11):930-942. [PubMed: 16948157]
  • Cavallo, A., and M. L. Mallory. 2004. Sleep deprivation, residency training, and ACGME rules: Practical guidelines. Journal of Pediatrics 145(6):717-718. [PubMed: 15580187]
  • Cavallo, A., J. Jaskiewicz, and M. D. Ris. 2002. Impact of night-float rotation on sleep, mood, and alertness: The resident’s perception. Chronobiology International 19(5):893-902. [PubMed: 12405552]
  • Cavallo, A., M. D. Ris, and P. Succop. 2003. The night float paradigm to decrease sleep deprivation: Good solution or a new problem? Ergonomics 46(7):653-663. [PubMed: 12745679]
  • Chee, M. W. L., and W. C. Choo. 2004. Functional imaging of working memory after 24 hr of total sleep deprivation. Journal of Neuroscience 24(19):4560-4567. [PubMed: 15140927]
  • Chee, M. W. L., L. Y. M. Chuah, V. Venkatraman, W. Y. Chan, P. Philip, and D. F. Dinges. 2006. Functional imaging of working memory following normal sleep and after 24 and 35 h of sleep deprivation: Correlations of fronto-parietal activation with performance. Neuroimage 31(1):419-428. [PubMed: 16427321]
  • Chee, M. W. L., C. H. Tan, Z. Hui, S. Parimal, S. D. Weissman, V. Zagorodnov, and D. F. Dinges. 2008. Lapsing during sleep deprivation is associated with distributed changes in brain activation. Journal of Neuroscience 28(21):5519-5528. [PubMed: 18495886]
  • Chuah, L. Y. M., V. Venkatraman, D. F. Dinges, and M. W. L. Chee. 2006. The neural basis of inter-individual variability in inhibitory efficiency following sleep deprivation. Journal of Neuroscience 26(27):7156-7162. [PubMed: 16822972]
  • CIR/SEIU (Committee of Interns and Residents-Service Employees International Union) Healthcare. 2007. Presentation by L. Toni Lewis to the Committee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, December 3, 2007, Washington, DC.
  • Cull, W. L., H. J. Mulvey, E. A. Jewett, E. L. Zalneraitis, C. E. Allen, and R. J. Pan. 2006. Pediatric residency duty hours before and after limitations. Pediatrics 118(6):e1805-e1811. [PubMed: 17142502]
  • Czeisler, C., E. D. Weitzman, M. C. Moore-Ede, J. C. Zimmerman, and R. S. Knauer. 1980. Human sleep—Its duration and organization depend on its circadian phase. Science 210(4475):1264-1267. [PubMed: 7434029]
  • Dawson, D., and K. McCulloch. 2005. Managing fatigue: It’s about sleep. Sleep Medicine Reviews 9(5):365-380. [PubMed: 16099184]
  • Dawson, D., and K. Reid. 1997. Fatigue, alcohol and performance impairment. Nature 388:235. [PubMed: 9230429]
  • Dinges, D. F. 1989. Napping. In Neuroscience year, edited by G. Adelman, editor. . Boston: Birkhäuser. Pp.113-114.
  • ———. 1990. Are you awake? Cognitive performance and reverie during the hypnopompic state. In Sleep and cognition, edited by R. R. Bootzin, editor; , J. F. Kihlstrom, editor; , and D. L. Schacter, editor. . Washington, DC: American Psychological Association. Pp.159-175.
  • ———. 1992. Adult napping and its effects on ability to function. In Why we nap. Evolution, chronobiology, and functions of polyphasic and ultrashort sleep, edited by C. Stampi, editor. . Boston: Birkhäuser. Pp.119-134.
  • ———. 1995. An overview of sleepiness and accidents. Journal of Sleep Research 4(Suppl 2):4-14. [PubMed: 10607205]
  • ———. 2001. Stress, fatigue, and behavioral energy. Nutrition Reviews 59(1):S30-S32. [PubMed: 11255802]
  • ———. 2004. Critical research issues in development of biomathematical models of fatigue and performance. Aviation Space and Environmental Medicine 75(3):A181-A191. [PubMed: 15018283]
  • ———. 2005. Sleep deprivation, impairment of resident performance, and medical error. Journal of Bone and Joint Surgery 87a(11):2581-2582.
  • Dinges, D. F., editor; , and R. J. Broughton, editor. , eds. 1989. Sleep and alertness: Chronobiological, behav ioral and medical aspects of napping. New York: Raven Press.
  • Dinges, D. F., and N. B. Kribbs. 1991. Performing while sleepy: Effects of experimentally-induced sleepiness. In Sleep, sleepiness and performance, edited by T. Monk, editor. . Chichester: John Wiley & Sons. Pp.97-128.
  • Dinges, D. F., M. T. Orne, and E. C. Orne. 1985. Assessing performance upon abrupt awakening from naps during quasi-continuous operations. Behavior Research Methods, Instru ments and Computers 17(1):37-45.
  • Dinges, D. F., M. T. Orne, W. G. Whitehouse, and E. C. Orne. 1987. Temporal placement of a nap for alertness: Contributions of circadian phase and prior wakefulness. Sleep 10(4):313-329. [PubMed: 3659730]
  • Dinges, D. F., W. G. Whitehouse, E. Carota Orne, and M. T. Orne. 1988. The benefits of a nap during prolonged work and wakefulness. Work and Stress 2(2):139-153.
  • Dinges, D. F., F. Pack, K. Williams, K. A. Gillen, J. W. Powell, G. E. Ott, C. Aptowicz, and A. I. Pack. 1997. Cumulative sleepiness, mood disturbance, and psychomotor vigilance performance decrements during a week of sleep restricted to 4-5 hours per night. Sleep 20(4):267-277. [PubMed: 9231952]
  • Dinges, D. F., N. L. Rogers, and M. D. Baynard. 2005. Chronic sleep deprivation. In Prin ciples and practice of sleep medicine, 4th ed., edited by M. H. Kryger, editor; , T. Roth, editor; , and W. C. Dement, editor. . Philadelphia, PA: W. B. Saunders. Pp.67-76.
  • Dola, C., L. Nelson, J. Lauterbach, S. Degefu, and G. Pridjian. 2006. Eighty hour work reform: Faculty and resident perceptions. American Journal of Obstetrics & Gynecology 195(5):1450-1456. [PubMed: 16996453]
  • Doran, S. M., H. P. Van Dongen, and D. F. Dinges. 2001. Sustained attention performance during sleep deprivation: Evidence of state instability. Archives Italiennes de Biologie 139(3):253-267. [PubMed: 11330205]
  • Dorrian, J., N. L. Rogers, and D. F. Dinges. 2005. Psychomotor vigilance performance: A neurocognitive assay sensitive to sleep loss. In Sleep deprivation: Clinical issues, pharmacol ogy and sleep loss effects, edited by C. Kushida, editor. . New York: Marcel Dekker. Pp.39-70.
  • Drake, C. L., T. Roehrs, G. Richardson, J. K. Walsh, and T. Roth. 2004. Shift work sleep disorder: Prevalence and consequences beyond that of symptomatic day workers. Sleep 27(8):1453-1462. [PubMed: 15683134]
  • Driscoll, T. R., R. R. Grunstein, and N. L. Rogers. 2007. A systematic review of the neurobehavioural and physiological effects of shiftwork systems. Sleep Medicine Reviews 11(3):179-194. [PubMed: 17418596]
  • Driskell, J. E., and B. Mullen. 2005. The efficacy of naps as a fatigue countermeasure: A meta-analytic integration. Human Factors 47(2):360-377. [PubMed: 16170944]
  • Drummond, S. P. A., G. G. Brown, J. L. Stricker, R. B. Buxton, E. C. Wong, and J. C. Gillin. 1999. Sleep deprivation-induced reduction in cortical functional response to serial subtraction. NeuroReport 10:3745-3748. [PubMed: 10716202]
  • Drummond, S. P., G. G. Brown, J. C. Gillin, J. L. Stricker, E. C. Wong, and R. B. Buxton. 2000. Altered brain response to verbal learning following sleep deprivation. Nature 403(6770):655-657. [PubMed: 10688201]
  • Drummond, S. P. A., A. Bischoff-Grethe, D. F. Dinges, L. Ayalon, S. C. Mednick, and M. J. Meloy. 2005. The neural basis of the psychomotor vigilance task. Sleep 28(9):1059-1068. [PubMed: 16268374]
  • Durmer, J. S., and D. F. Dinges. 2005. Neurocognitive consequences of sleep deprivation. Seminars in Neurology 25(1):117-129. [PubMed: 15798944]
  • Eastridge, B. J., E. C. Hamilton, G. E. O’Keefe, R. V. Rege, R. J. Valentine, D. J. Jones, S. Tesfay, and E. R. Thal. 2003. Effect of sleep deprivation on the performance of simulated laparoscopic surgical skill. American Journal of Surgery 186(2):169-174. [PubMed: 12885613]
  • Ellman, P. I., I. L. Kron, J. S. Alvis, C. Tache-Leon, T. S. Maxey, T. B. Reece, B. B. Peeler, J. A. Kern, and C. G. Tribble. 2005. Acute sleep deprivation in the thoracic surgical resident does not affect operative outcomes. Annals of Thoracic Surgery 80(1):60-65. [PubMed: 15975341]
  • Federation of State Medical Boards. 2007. State-specific requirements for initial medical licen sure . http://fsmb​.org/usmle_eliinitial.html (accessed March 17, 2008).
  • Feyer, A. M., and A. M. Williamson. 1995. Work and rest in the long-distance road transport industry in Australia. Work and Stress 9(2-3):198-205.
  • Folkard, S., D. A. Lombardi, and P. T. Tucker. 2005. Shiftwork: Safety, sleepiness and sleep. Industrial Health 43(1):20-23. [PubMed: 15732299]
  • Friedman, R. C., J. T. Bigger, and D. S. Kornfield. 1971. The intern and sleep loss. New En gland Journal of Medicine 285:201-203. [PubMed: 5087723]
  • Fuller, P. M., J. J. Gooley, and C. B. Saper. 2006. Neurobiology of the sleep-wake cycle: Sleep architecture, circadian regulation, and regulatory feedback. Journal of Biological Rhythms 21(6):482-493. [PubMed: 17107938]
  • Gaba, D. M., and S. K. Howard. 2002. Patient safety: Fatigue among clinicians and the safety of patients. New England Journal of Medicine 347(16):1249-1255. [PubMed: 12393823]
  • Gabow, P. A., A. Karkhanis, A. Knight, P. Dixon, S. Eisert, and R. K. Albert. 2006. Observations of residents’ work activities for 24 consecutive hours: Implications for workflow redesign. Academic Medicine 81(8):766-775. [PubMed: 16868436]
  • Gander, P. H., A. Merry, M. M. Millar, and J. Weller. 2000. Hours of work and fatigue-related error: A survey of New Zealand anaesthetists. Anaesthesia Intensive Care 28(2):178-183. [PubMed: 10788970]
  • 2005. S. 1297—109th congress: Patient and Physician Safety and Protection Act of 2005 . http://www​​.xpd?bill​=h109-1297&tab=summary (accessed March 29, 2008).
  • Groeger, J. A., A. U. Viola, J. C. Y. Lo, M. Von Schantz, S. N. Archer, and D. J. Dijk. 2008. Early morning executive functioning during sleep deprivation is compromised by a PERIOD3 polymorphism. Sleep 31(8):1159-1167. [PMC free article: PMC2542962] [PubMed: 18714788]
  • Habeck, C., B. C. Rakitin, J. Moeller, N. Scarmeas, E. Zarahn, T. Brown, and Y. Stern. 2004. An event-related fMRI study of the neurobehavioral impact of sleep deprivation on performance of a delayed-match-to-sample task. Cognitive Brain Research 18(3):306-321. [PubMed: 14741317]
  • Harrison, Y., and J. A. Horne. 1999. One night of sleep loss impairs innovative thinking and flexible decision making. Organic Behavioral Human Decision Processes 78(2):128-145. [PubMed: 10329298]
  • ———. 2000. The impact of sleep deprivation on decision making: A review. Journal of Ex perimental Psychology: Applied 6(3):236-249. [PubMed: 11014055]
  • Horrocks, N., R. Pounder, and R. C. P. Working Group. 2006. Working the night shift: Preparation, survival and recovery—A guide for junior doctors. Clinical Medicine 6(1):61-67. [PMC free article: PMC4954435] [PubMed: 16521358]
  • Howard, S. K., D. M. Gaba, M. R. Rosekind, and V. P. Zarcone. 2002. The risks and implications of excessive daytime sleepiness in resident physicians. Academic Medicine 77(10):1019-1025. [PubMed: 12377678]
  • Howard, S. K., D. M. Gaba, B. E. Smith, M. B. Weinger, C. Herndon, S. Keshavacharya, and M. R. Rosekind. 2003. Simulation study of rested versus sleep-deprived anesthesiologists. Anesthesiology 98(6):1345-1355. [PubMed: 12766642]
  • Huber, R., M. F. Ghilardi, M. Massimini, and G. Tononi. 2004. Local sleep and learning. Nature 430(6995):78-81. [PubMed: 15184907]
  • IOM (Institute of Medicine). 2006. Sleep disorders and sleep deprivation: An unmet public health problem . Washington, DC: The National Academies Press.
  • Jakubowicz, D. M., E. M. Price, H. J. Glassman, A. J. G. Gallagher, N. Mandava, W. P. Ralph, and M. P. Fried. 2005. Effects of a twenty-four hour call period on resident performance during simulated endoscopic sinus surgery in an Accreditation Council for Graduate Medical Education-compliant training program. Laryngoscope 115(1):143-146. [PubMed: 15630383]
  • Jay, S. M., N. Lamond, S. A. Ferguson, J. Dorrian, C. B. Jones, and D. Dawson. 2007. The characteristics of recovery sleep when recovery opportunity is restricted. Sleep 30(3):353-360. [PubMed: 17425232]
  • Jean-Louis, G., D. F. Kripke, S. Ancoli-Israel, M. R. Klauber, and R. S. Sepulveda. 2000. Sleep duration, illumination, and activity patterns in a population sample: Effects of gender and ethnicity. Biological Psychiatry 47(10):921-927. [PubMed: 10807965]
  • Jensen, A., R. Milner, C. Fisher, J. Gaughan, R. Rolandelli, and H. Grewal. 2004. Short-term sleep deficits do not adversely affect acquisition of laparoscopic skills in a laboratory setting. Surgical Endoscopy and Other Interventional Techniques 18(6):948-953. [PubMed: 15095080]
  • Jewett, M. E., D. J. Dijk, R. E. Kronauer, and D. F. Dinges. 1999. Dose-response relationship between sleep duration and human psychomotor vigilance and subjective alertness. Sleep 22(2):171-179. [PubMed: 10201061]
  • Jha, A. K., B. W. Duncan, and D. W. Bates. 2001. Fatigue—Sleepiness and medical errors. Chapter 46. In Making health care safer: A critical analysis of patient safety practices, edited by K. G. Shojania, editor; , B. W. Duncan, editor; , K. M. McDonald, editor; , and R. M. Wachter, editor. . Evidence Report/Technology Assessment No. 43. Pp.519-533 . Rockville, MD: Agency for Healthcare Research and Quality.
  • Kahol, K., M. J. Leyba, M. Deka, V. Deka, S. Mayes, M. Smith, J. J. Ferrara, and S. Panchanathan. 2008. Effect of fatigue on psychomotor and cognitive skills. American Journal of Surgery 195(2):195-204. [PubMed: 18194679]
  • Killgore, W. D. S., T. J. Balkin, and N. J. Wesensten. 2006. Impaired decision making following 49 h of sleep deprivation. Journal of Sleep Research 15(1):7-13. [PubMed: 16489997]
  • Kleitman, N. 1963. Sleep and wakefulness. Chicago, IL: University of Chicago Press.
  • Knauth, P. 2007. Extended work periods. Industrial Health 45(1):125-136. [PubMed: 17284884]
  • Knauth, P., and S. Hornberger. 2003. Preventive and compensatory measures for shift workers. Occupational Medicine 53(2):109-116. [PubMed: 12637595]
  • Kubo, T., H. Takeyama, S. Matsumoto, T. Ebara, K. Murata, N. Tachi, and T. Itani. 2007. Impact of nap length, nap timing and sleep quality on sustaining early morning performance. Industrial Health 45(4):552-563. [PubMed: 17878627]
  • Lamond, N., and D. Dawson. 1999. Quantifying the performance impairment associated with fatigue. Journal of Sleep Research 8:255-262. [PubMed: 10646165]
  • Lamond, N., S. M. Jay, J. Dorrian, S. A. Ferguson, C. Jones, and D. Dawson. 2007. The dynamics of neurobehavioural recovery following sleep loss. Journal of Sleep Research 16(1):33-41. [PubMed: 17309761]
  • Landrigan, C. P., J. M. Rothschild, J. W. Cronin, R. Kaushal, E. Burdick, J. T. Katz, C. M. Lilly, P. H. Stone, S. W. Lockley, D. W. Bates, and C. A. Czeisler. 2004. Effect of reducing interns’ work hours on serious medical errors in intensive care units. New England Journal of Medicine 351(18):1838-1848. [PubMed: 15509817]
  • Landrigan, C. P., C. A. Czeisler, L. K. Barger, N. T. Ayas, J. M. Rothschild, and S. W. Lockley. 2007. Effective implementation of work-hour limits and systemic improvements. Joint Commission Journal on Quality and Patient Safety 33(11 Suppl):19-29. [PubMed: 18173163]
  • Lavie, P. 1986. Ultrashort sleep-waking schedule. 3: Gates and forbidden zones for sleep. Electroencephalography and Clinical Neurophysiology 63(5):414-425. [PubMed: 2420557]
  • Leonard, C., N. Fanning, J. Attwood, and M. Buckley. 1998. The effect of fatigue, sleep deprivation and onerous working hours on the physical and mental wellbeing of pre-registration house officers. Irish Journal of Medical Sciences 167(1):22-25. [PubMed: 9540294]
  • Leproult, R., E. F. Colecchia, A. M. Berardi, R. Stickgold, S. M. Kosslyn, and E. Van Cauter. 2003. Individual differences in subjective and objective alertness during sleep deprivation are stable and unrelated. American Journal of Physiology-Regulatory Integrative and Comparative Physiology 284(2):R280-R290. [PubMed: 12529281]
  • Levine, B., T. Roehrs, E. Stepanski, F. Zorick, and T. Roth. 1987. Fragmenting sleep diminishes its recuperative value. Sleep 10(6):590-599. [PubMed: 3432859]
  • Lim, J., and D. F. Dinges. 2008. Sleep deprivation and vigilant attention. Annals of the New York Academy of Sciences 1129:305-322. [PubMed: 18591490]
  • Lim, J., W. C. Choo, and M. W. L. Chee. 2007. Reproducibility of changes in behaviour and fMRI activation associated with sleep deprivation in a working memory task. Sleep 30(1):61-70. [PubMed: 17310866]
  • Lockley, S. W., J. W. Cronin, E. E. Evans, B. E. Cade, C. J. Lee, C. P. Landrigan, J. M. Rothschild, J. T. Katz, C. M. Lilly, P. H. Stone, D. Aeschbach, and C. A. Czeisler. 2004. Effect of reducing interns’ weekly work hours on sleep and attentional failures. New England Journal of Medicine 351(18):1829-1837. [PubMed: 15509816]
  • Lockley, S. W., C. P. Landrigan, L. K. Barger, and C. A. Czeisler. 2006. When policy meets physiology—The challenge of reducing resident work hours. Clinical Orthopaedics and Related Research (449):116-127. [PubMed: 16770285]
  • Lockley, S. W., L. K. Barger, N. T. Ayas, J. M. Rothschild, C. A. Czeisler, and C. P. Landrigan. 2007. Effects of health care provider work hours and sleep deprivation on safety and performance. Joint Commission Journal on Quality and Patient Safety 33(11 Suppl):7-18. [PubMed: 18173162]
  • Ludmerer, K. M., and M. M. E. Johns. 2005. Reforming graduate medical education. JAMA 294(9):1083-1087. [PubMed: 16145029]
  • Macdonald, W., and S. Bendak. 2000. Effects of workload level and 8- versus 12-h workday duration on test battery performance. International Journal of Industrial Ergonomics 26(3):399-416.
  • Macias, D. J., J. Hafner, J. C. Brillman, and D. Tandberg. 1996. Effect of time of day and duration into shift on hazardous exposures to biological fluids. Academic Emergency Medicine 3(6):605-610. [PubMed: 8727632]
  • Marcus, C. L., and G. M. Loughlin. 1996. Effect of sleep deprivation on driving safety in housestaff. Sleep 19(10):763-766. [PubMed: 9085483]
  • McKenna, B. S., D. L. Dickinson, H. J. Orff, and S. P. A. Drummond. 2007. The effects of one night of sleep deprivation on known-risk and ambiguous-risk decisions. Journal of Sleep Research 16(3):245-252. [PubMed: 17716272]
  • Mitler, M. M., M. A. Carskadon, C. A. Czeisler, W. C. Dement, D. F. Dinges, and R. C. Graeber. 1988. Catastrophes, sleep, and public policy: Consensus report. Sleep 11(1):100-109. [PMC free article: PMC2517096] [PubMed: 3283909]
  • Mollicone, D. J., H. P. A. Van Dongen, N. L. Rogers, and D. F. Dinges. 2008. Response surface mapping of neurobehavioral performance: Testing the feasibility of split sleep schedules for space operations. Acta Astronautica 63(7-10):833-840. [PMC free article: PMC2633932] [PubMed: 19194521]
  • Nachreiner, F. 2001. Time on task effects on safety . Journal of Human Ergology 30(1-2): 97-102. [PubMed: 14564865]
  • Naitoh, P. 1992. Minimal sleep to maintain performance: The search for sleep quantum in sustained operations. In Why we nap: Evolution, chronobiology, and functions of polyphasic and ultrashort sleep, edited by C. Stampi, editor. . Boston: Birkhäuser. Pp.199-216.
  • Naitoh, P., and R. G. Angus. 1989. Napping and human functioning during prolonged work. In Sleep and alertness: Chronobiological, behavioral, and medical aspects of napping, edited by D. F. Dinges, editor; and R. J. Brougthon, editor. . New York: Raven Press. Pp.221-246.
  • Ohayon, M. M., M. A. Carskadon, C. Guilleminault, and M. V. Vitiello. 2004. Meta-analysis of quantitative sleep parameters from childhood to old age in healthy individuals: Developing normative sleep values across the human lifespan. Sleep 27(7):1255-1273. [PubMed: 15586779]
  • Owens, J. A., A. Avidan, D. Baldwin, and C. Landrigan. 2008. Improving sleep hygiene. Archives of Internal Medicine 168(11):1229-a-1230. [PubMed: 18541835]
  • Pack, A., A. M. Pack, E. Rodgman, A. Cucchiara, D. F. Dinges, and C. W. Schwab. 1995. Characteristics of crashes attributed to the driver having fallen asleep. Accident Analysis and Prevention 27(6):769-775. [PubMed: 8749280]
  • Parshuram, C. S. 2006. The impact of fatigue on patient safety. Pediatric Clinics of North America 53(6):1135-1153. [PubMed: 17126687]
  • Parthasarathy, S., K. Hettiger, R. Budhiraja, and B. Sullivan. 2007. Sleep and well-being of ICU housestaff. CHEST 131(6):1685-1693. [PubMed: 17400688]
  • Patrick, G. T. W., and J. A. Gilbert. 1896. On the effects of loss of sleep. Psychological Review 3(5):469-483.
  • Persson, R., P. Orbaek, H. Ursin, G. Kecklund, and K. Osterberg. 2003. Effects of the implementation of an 84-hour workweek on neurobehavioral test performance and cortisol responsiveness during testing. Scandinavian Journal of Work Environment & Health 29(4):261-269. [PubMed: 12934719]
  • Persson, R., P. Orbaek, G. Kecklund, and T. Akerstedt. 2006. Impact of an 84-hour workweek on biomarkers for stress, metabolic processes and diurnal rhythm. Scandinavian Journal of Work Environment & Health 32(5):349-358. [PubMed: 17091202]
  • Philibert, I. 2005. Sleep loss and performance in residents and nonphysicians: A meta-analytic examination. Sleep 28(11):1392-1402. [PubMed: 16335329]
  • Portas, C. M., G. Rees, A. M. Howseman, O. Josephs, R. Turner, and C. D. Frith. 1998. A specific role for the thalamus in mediating the interaction of attention and arousal in humans. Journal of Neuroscience 18(21):8979-8989. [PubMed: 9787003]
  • Richardson, G. S., J. K. Wyatt, J. P. Sullivan, E. J. Orav, A. E. Ward, M. A. Wolf, and C. A. Czeisler. 1996. Objective assessment of sleep and alertness in medical house staff and the impact of protected time for sleep. Sleep 19(9):718-726. [PubMed: 9122559]
  • Roehrs, T., M. Greenwald, and T. Roth. 2004. Risk-taking behavior: Effects of ethanol, caffeine, and basal sleepiness. Sleep 27(5):887-893. [PubMed: 15453546]
  • Rogers, A. E., W. T. Hwang, L. D. Scott, L. H. Aiken, and D. F. Dinges. 2004. The working hours of hospital staff nurses and patient safety. Health Affairs 23(4):202-212. [PubMed: 15318582]
  • Rosa, R. R. 2001. Examining work schedules for fatigue: It’s not just hours of work. In Stress, workload, and fatigue, edited by P.A. Hancock, editor; and P. A. Desmond, editor. . Mahwah, NJ: Lawrence Erlbaum Associates. Pp.513-528.
  • Rosekind, M. R., R. C. Graeber, D. F. Dinges, L. J. Connell, M. S. Rountree, C. L. Spinweber, and K. A. Gillen. 1994. Crew factors in flight operations IX: Effects of planned cockpit rest on crew performance and alertness in long-haul operations . Moffett Field, CA: NASA Ames Research Center.
  • Rosekind, M. R., R. M. Smith, D. L. Miller, E. L. Co, K. B. Gregory, L. L. Webbon, P. H. Gander, and J. V. Lebacqz. 1995. Alertness management: Strategic naps in operational settings. Journal of Sleep Research 4(Suppl 2):62-66. [PubMed: 10607214]
  • Rosekind, M. R., D. F. Neri, and D. F. Dinges. 1997. From laboratory to flightdeck: Promoting operational alertness. Paper read at Proceedings: The Royal Aeronautical Society’s Symposium on Fatigue and Duty Time Limitations—An International Review, London, UK.
  • Sack, R. L., D. Auckley, R. R. Auger, M. A. Carskadon, K. P. Wright, M. V. Vitiello, and I. Zhdanova. 2007. Circadian rhythm sleep disorders: Part 1, Basic principles, shift work and jet lag disorders. Sleep 30(11):1460-1483. [PMC free article: PMC2082105] [PubMed: 18041480]
  • SAFER Task Force. 2007. Sleep, alertness, and fatigue in residency . http://www​​/store/product/details.aspx?pid=116 (accessed June 1, 2007).
  • Schweitzer, P. K., A. C. Randazzo, K. Stone, M. Erman, and J. K. Walsh. 2006. Laboratory and field studies of naps and caffeine as practical countermeasures for sleep-wake problems associated with night work. Sleep 29(1):39-50. [PubMed: 16453980]
  • Scott, L. D., A. E. Rogers, W. T. Hwang, and Y. Zhang. 2006. Effects of critical care nurses’ work hours on vigilance and patients’ safety. American Journal of Critical Care 15(1):30-37. [PubMed: 16391312]
  • Smith-Coggins, R., S. K. Howard, D. T. Mac, C. Wang, S. Kwan, M. R. Rosekind, Y. Sowb, R. Ballise, J. Levis, and D. M. Gaba. 2006. Improving alertness and performance in emergency department physicians and nurses: The use of planned naps. Annals of Emergency Medicine 48(5):596-604. [PubMed: 17052562]
  • Stampi, C., editor. , ed. 1992. Why we nap. Evolution, chronobiology, and functions of polyphasic and ultrashort sleep. Boston: Birkhäuser.
  • Stickgold, R. 2005. Sleep-dependent memory consolidation. Nature 437(7063):1272-1278. [PubMed: 16251952]
  • Stickgold, R., L. James, and J. A. Hobson. 2000. Visual discrimination learning requires sleep after training. Nature Neuroscience 3(12):1237-1238. [PubMed: 11100141]
  • Strogatz, S. H., R. E. Kronauer, and C. A. Czeisler. 1986. Circadian regulation dominates homeostatic control of sleep length and prior wake length in humans. Sleep 9(2): 353-364. [PubMed: 3505735]
  • ———. 1987. Circadian pacemaker interferes with sleep onset at specific times each day—Role in insomnia. American Journal of Physiology 253(1):R172-R178. [PubMed: 3605382]
  • Takahashi, M., H. Fukuda, K. Miki, T. Haratani, L. Kurabayashi, N. Hisanaga, H. Arito, H. Takahashi, M. Egoshi, and M. Sakurai. 1999. Shift work-related problems in 16-h night shift nurses. (2): Effects on subjective symptoms, physical activity, heart rate, and sleep. Industrial Health 37(2):228-236. [PubMed: 10319571]
  • Takeyama, H., T. Kubo, and T. Itani. 2005. The nighttime nap strategies for improving night shift work in workplace. Industrial Health 43(1):24-29. [PubMed: 15732300]
  • Tassi, P., and A. Muzet. 2000. Sleep inertia. Sleep Medicine Reviews 4(4):341-353. [PubMed: 12531174]
  • Thomas, M., H. Sing, G. Belenky, H. Holcomb, H. Mayberg, R. Dannals, H. Wagner, D. Thorne, K. Popp, L. Rowland, A. Welsh, S. Balwinski, and D. Redmond. 2000. Neural basis of alertness and cognitive performance impairments during sleepiness. I. Effects of 24 h of sleep deprivation on waking human regional brain activity. Journal of Sleep Research 9(4):335-352. [PubMed: 11123521]
  • Torsvall, L., and T. Akerstedt. 1988. Disturbed sleep while being on-call—An EEG study of ships engineers. Sleep 11(1):35-38. [PubMed: 3363267]
  • Uchal, M., J. Tjugum, E. Martinsen, X. G. Qiu, and R. Bergamaschi. 2005. The impact of sleep deprivation on product quality and procedure effectiveness in a laparoscopic physical simulator: A randomized controlled trial. American Journal of Surgery 189(6):753-757. [PubMed: 15910732]
  • Van Dongen, H. P. A. 2006. Shift work and inter-individual differences in sleep and sleepiness. Chronobiology International 23(6):1139-1147. [PubMed: 17190701]
  • Van Dongen, H. P. A., and D. F. Dinges. 2005. Circadian rhythm in sleepiness, alertness and performance. In Principles and practice of sleep medicine. 4th ed., edited by M. H. Kryger, editor; , T. Roth, editor; , and W. C. Dement, editor. . Philadelphia, PA: W. B. Saunders. Pp.435-443.
  • Van Dongen, H. P. A., N. J. Price, J. M. Mullington, M. P. Szuba, S. C. Kapoor, and D. F. Dinges. 2001. Caffeine eliminates psychomotor vigilance deficits from sleep inertia. Sleep 24(7):813-819. [PubMed: 11683484]
  • Van Dongen, H. P., G. Maislin, J. M. Mullington, and D. F. Dinges. 2003. The cumulative cost of additional wakefulness: Dose-response effects on neurobehavioral functions and sleep physiology from chronic sleep restriction and total sleep deprivation. Sleep 26(2):117-126. [PubMed: 12683469]
  • Van Dongen, H. P., M. D. Baynard, G. Maislin, and D. F. Dinges. 2004. Systematic interindividual differences in neurobehavioral impairment from sleep loss: Evidence of trait-like differential vulnerability. Sleep 27(3):423-433. [PubMed: 15164894]
  • Van Dongen, H. P. A., C. G. Mott, J. K. Huang, D. J. Mollicone, F. D. McKenzie, and D. F. Dinges. 2007. Optimization of biomathematical model predictions for cognitive performance impairment in individuals: Accounting for unknown traits and uncertain states in homeostatic and circadian processes. Sleep 30(9):1129-1143. [PMC free article: PMC1978411] [PubMed: 17910385]
  • Veasey, S., R. Rosen, B. Barzansky, I. Rosen, and J. Owens. 2002. Sleep loss and fatigue in residency training: A reappraisal. JAMA 288(9):1116-1124. [PubMed: 12204082]
  • Venkatraman, V., Y. M. L. Chuah, S. A. Huettel, and M. W. L. Chee. 2007. Sleep deprivation elevates expectation of gains and attenuates response to losses following risky decisions. Sleep 30(5):603-609. [PubMed: 17552375]
  • Viola, A. U., S. N. Archer, L. M. James, J. A. Groeger, J. C. Y. Lo, D. Skene, M. von Schantz, and D. J. Dijk. 2007. PER3 polymorphism predicts sleep structure and waking performance. Current Biology 17:1-6. [PubMed: 17346965]
  • Walker, M. P., and R. Stickgold. 2006. Annual review of psychology. Sleep, Memory, and Plasticity 57:139-166. [PubMed: 16318592]
  • Weinger, M. B., and S. Ancoli-Israel. 2002. Sleep deprivation and clinical performance. JAMA 287(8):955-957. [PubMed: 11866625]
  • Wesensten, N., T. J. Balkin, and G. Belenky. 1999. Does sleep fragmentation impact recuperation? A review and reanalysis. Journal of Sleep Research 8(4):237-245. [PubMed: 10646163]
  • Williamson, A. M., and A.-M. Feyer. 2000. Moderate sleep deprivation produces impairments in cognitive and motor performance equivalent to legally prescribed levels of alcohol intoxication. Occupational and Environmental Medicine 57:649-655. [PMC free article: PMC1739867] [PubMed: 10984335]
  • Wu, J. C., J. C. Gillin, M. S. Buchsbaum, P. Chen, D. B. Keator, N. K. Wu, L. A. Darnall, J. H. Fallon, and W. E. Bunney. 2006. Frontal lobe metabolic decreases with sleep deprivation not totally reversed by recovery sleep. Neuropsychopharmacology 31(12):2783-2792. [PubMed: 16880772]



Personal communication, I. Philibert, Accreditation Council for Graduate Medical Education, May 2, 2008.


Personal communication, David Meltzer, University of Chicago, August 12, 2008.

Copyright 2009 by the National Academy of Sciences. All rights reserved.
Bookshelf ID: NBK214949


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (3.5M)

Related information

  • PMC
    PubMed Central citations
  • PubMed
    Links to PubMed

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...