Clinical Ethics in Anesthesiology. A Case-Based Textbook

5. Practice issues

37. The impaired anesthesiologist - sleep deprivation

Steven K. Howard

The Case

At 3 am, a first-year anesthesia resident was providing an anesthetic to a young adult male undergoing exploratory laparotomy for presumed appendicitis. The resident had started his shift 20 hours earlier at 7 am at this very busy county hospital. There were cases to follow and it was obvious that this resident would be busy until being relieved of duty the following day. As the case was finishing, the resident was drawing up medications to reverse neuromuscular blockade. He drew up 3 mg of neostigmine. He intended to draw up 0.6 mg of glycopyrrolate but instead drew up 30 mg of phenylephrine. The two drugs were in adjacent bins in the drug cart and were packaged in similar appearing vials. Prior to administering the likely lethal concoction, the resident described feeling that “something was not right.” Upon checking the empty vials on the anesthesia cart, the near miss was discovered and the syringe disposed of. The intended reversal agents were then drawn up and administered to the patient without incident. The resident completed two more cases during his call until being relieved the following morning. This incident was not reported to the residents’ supervisor. In three years of anesthesia training, this resident was offered the opportunity to nap on call only once by faculty.

The ASA Guidelines for the Ethical Practice of Anesthesiology outline the ethical responsibilities of anesthesiologists. Section IV.2 states:

The practice of quality anesthesia care requires that anesthesiologists maintain their physical and mental health and special sensory capabilities. If in doubt about their health, then anesthesiologists should seek medical evaluation and care. During this period of evaluation and treatment, anesthesiologists should modify or cease their practice.1

Sleep deprivation has a negative impact on performance, including the “special sensory capabilities” mentioned in the ethical guidelines. Anesthesia care providers are required to care for patients around the clock, presenting special problems for human physiology. Specific strategies have been used in other domains to mitigate the impairing affect of fatigue.

Human error

In 1999, the Institute of Medicine (IOM) published a treatise To Err is Human which uncovered a previously unappreciated level of patient mortality and morbidity associated with human error of health care providers.2 Although the IOM report did not specifically address the issue of care provided by sleep-deprived practitioners, it is likely that a fraction of the preventable errors that occur are secondary to this performance shaping variable. Error investigation in healthcare and in other industries has shown that fatigued workers make errors, although exact rates are difficult to quantify. Physicians are not able to withstand the impairing effect of sleep deprivation. Even though sleep deprivation and fatigue could not be proven to have contributed to the near miss in the case scenario, sophisticated accident investigation performed in other industries would conclude that it played a role.3

Basic sleep physiology

A basic primer in sleep physiology is important to understand the issues of sleep deprivation and its effect on performance. Sleep is a reversible behavioral state of perceptual disengagement from and unresponsiveness to the environment, and it is vital to human survival. It is comprised of two states – non-rapid eye movement (NREM) and rapid eye movement (REM) sleep that cycle throughout the night and can be differentiated electrophysiologically. Both NREM and REM sleep are required for optimal alertness, though the body will preferentially replete lost sleep with slow wave sleep (NREM stages 3 and 4). REM sleep is the stage when dreaming takes place. REM sleep periods get longer as the sleep period extends – the longest REM periods coming before awakening in the morning.

Physiologic sleepiness

Sleep is a basic human need similar to eating and drinking. When deprived of food or drink, we develop hunger or thirst. Similarly, if we obtain less sleep than we require, sleepiness becomes marked and our brain becomes “pressured” to acquire sleep –extreme sleepiness during hazardous activities can become manifest. In these situations the brain may shift uncontrollably between wake and sleep with little or no awareness from the individual. These short sleep episodes (called microsleeps or attentional failures) create safety risks especially when envisioning them occurring in the operating room during patient care or upon driving home after a long call period. This level of sleepiness occurs in all humans if pressed to extremes.

Subjective sleepiness

Subjective sleepiness (how sleepy we feel) often underestimates the level of physiologic sleepiness. It is conceivable that this subjective – physiologic disconnect makes critical decision making more challenging. This disconnect might help explain why the incidence of single vehicular automobile accidents is higher from 2–6 am, during times of increased physiologic vulnerability. Though we may think we are “OK” to perform, we are often less than optimally alert. The two primary determinants of sleepiness are modulated by sleep homeostasis and circadian rhythms.

The homeostat

Sleep homeostasis refers to the sleep–wake balance of an individual. This balance is a function of sleep need, offset by the quantity and quality of sleep obtained. An individual’s sleep propensity (i.e., sleep pressure) changes when this equilibrium is tipped in either direction. For example, if an individual’s sleep need is 8 hours per night but sleep achieved is 7 hours per night a sleep debt develops and sleep propensity increases. Average adults require greater than 8 hours of sleep per 24-hour period to maintain optimal alertness and function at peak levels. Population studies reveal that sleep need is normally distributed so that for every person who needs 6 hours of sleep for optimal alertness another will require 10 hours. Importantly, humans are notoriously poor at determining their true level of sleep need.

An individual’s sleep need is genetically determined and cannot be trained. As adults age, sleep need remains constant but obtaining adequate quantity and quality of sleep becomes more difficult. Sleep disorders increase with age and other benign issues associated with aging (e.g., urinary frequency in males and menopausal symptoms in females) make sleep consolidation increasingly difficult.

The circadian pacemaker

A circadian pacemaker located in the suprachiasmatic nucleus of the brain governs alertness. Humans are programmed for two nadirs of alertness between 2–6 am and 2–6 pm and these times represent periods of increased vulnerability to performance impairment due to pacemaker-induced sleepiness. The circadian clock is very resistant to alterations and does not adjust rapidly. For healthcare providers, the clock’s resiliency to change is best evidenced by those who work the night shift. It is often difficult to maintain alertness during the night when the body’s clock is “turned off” and similarly it is difficult to gain adequate sleep during the day when clock is “turned on” (e.g., attempting to sleep during the day after an on-call period). Working in opposition to circadian physiology is necessary in the round-the-clock work of health care providers but it comes at the risk of decreased alertness.

Fatigue affects performance

Fatigue adversely affects performance.4 Examples include slowing of cognition, increased performance variability, and decreased motivation. The learning of new information slows, memory is impaired, and non-essential activities are neglected. Picture the fatigued anesthesiologist at 3 am after being awake for a number of hours trying to safely care for patients with these very real constraints of human physiology. These are precisely the special sensory capabilities alluded to in the ASA’s Guidelines for Ethical Practice, and are critical for safe provision of care.

Performance is proven to be altered in sleep-deprived physicians. The Harvard Work Hours, Health and Safety Group showed that reducing shift duration by interns from 30 to less than 17 hours decreased serious medical errors by 35.9%.5 Simulation research of anesthesiologists supports the findings that a clinician’s performance is negatively impacted by sleep deprivation and that increasing sleep (either by prioritizing or napping) can improve performance.6

A robust finding in similar studies is that mood is dramatically affected by sleep loss and fatigue. Negative moods (e.g., anger, depression, fatigue, tension) increase and positive moods decrease as sleep loss accrues.7 The impact of impaired mood on interpersonal interactions between health care providers and between providers and patients has been inadequately studied but is likely to have an overall negative effect.

Comparison of sleep deprivation to ethanol usage

Results of research correlating the affect of sleep loss and alcohol consumption are alarming. Dawson showed that after 17 hours of wakefulness that performance on a tracking task declined to a level of impairment equivalent to a blood alcohol concentration (BAC) of 0.05%.8 After 24 hours of wakefulness, the decline in performance was equivalent to a BAC of 0.1% – over the legal level for driving everywhere in the US. This amount of wakefulness is commonly seen in medical professionals who take “in house” call and creates a source for systemic latent errors thus increasing system vulnerability.9

A study comparing the effects of sleep loss and alcohol consumption on performance of a simulated driving task further supports the correlation.10 With increasing blood alcohol concentration, simulated driving performance became progressively impaired. Speed variability, and off-road events increased, while speed deviation decreased, the result of subjects driving faster. Wakefulness of 18.5 and 21 hours produced changes of the same magnitude as 0.05 and 0.08% BAC, respectively. Finally, wakefulness prolonged by as little as 3 h can produce decrements in the ability to maintain speed and road position as serious as those found at the legal limits of alcohol consumption.

Arendt and colleagues found that the neurobehavioral performance of residents after a “heavy” (every fourth or fifth night) call was comparable to that of residents after alcohol ingestion to BAC of 0.04 to 0.05% during “light call” conditions.11

Caring for patients while under the influence of alcohol is not ethical and would be viewed as abhorrent behavior. Professionals caring for patients “under the influence” of alcohol would likely lose their license to practice and be shunned by society. Threats to patient safety would be clear. Yet similar levels of impairment are now known to be caused by sleep loss. Are we routinely harming patients because of impaired decision making due to sleep deprivation?

No direct link between fatigue due to sleep loss and patient morbidity and mortality has yet been made, and this is one reason for the maintenance of the status quo. If 4 hours of sleep predict poor patient outcome (or driving off the road after call) healthcare providers would have to prioritize sleep as an important component of professionalism. Instead, in the current culture of health care, staying up all night on call, is still thought of by many as a rite of passage and a badge of honor. If anesthesiologists better recognized the level of impairment engendered by sleep loss, perhaps these behaviors would change.

Ethical responsibilities of anesthesiologist

Responsibilities to patients

Anesthesiologists have an ethical responsibility to be optimally prepared to safely care for patients and this includes avoidance of impairments from any cause. Optimal performance in part depends on level of alertness and this is largely under the control of the individual. There is no BAC equivalent marker for sleepiness and no “sleep police” to deter poor sleep habits, so it is the individual’s responsibility to come to work optimally prepared – not impaired. Excellent sleep habits form the most important strategy for each physician. A variety of strategies have been proven to increase alertness.

Recuperative sleep

Average adults require at least 8 hours per 24-hour day. Sleep blocks less than this can be supplemented by nap periods both at work and after work shifts. Every attempt should be made to mitigate sleep debt prior to call periods by prioritizing sleep. The sleep environment should be conducive to sleep by allowing control over ambient light, noise, and temperature. Pre-sleep routines that give cues for relaxation and sleep should be employed and distractions such as electronic devices, pets, children, etc. should be minimized.

Strategic napping

Napping has been shown to improve alertness and performance.12 As little as 10–15 minutes of sleep will improve performance and longer sleep periods have a more substantial effect. Naps address physiologic sleep need by decreasing sleep debt. Unfortunately, naps are not utilized in our society for mostly cultural reasons – the most common reason given is that napping is a sign of weakness or laziness. Strategic napping has been piloted in a few healthcare facilities but have not reached widespread use at the present time.13

Strategic use of caffeine

Caffeine is the most widely used drug to improve alertness. Chronic use of caffeine limits its strategic use as tolerance develops to routine exposure. Strategic use of caffeine includes: (1) limiting chronic ingestion hence tolerance; (2) ingesting when you have be awake (e.g., during circadian low points); (3) knowing its onset (15–30 minutes) and duration of effect (3–4 hours); and (4) avoiding its use if close to a sleep opportunity. Combining the effect of a nap and caffeine can be used to improve alertness.14

Mindful intake of food and drugs

Use of caffeine, alcohol, and nicotine close to sleep opportunities will impair initiation, consolidation, and maintenance of sleep. Attempting to sleep either hungry or on a full stomach will negatively impact sleep.


Regular exercise improves sleep, but exercise within a few hours to the sleep period can impair sleep. It is best to complete exercise routines 2–3 hours before sleep to avoid this.

Diagnose and treat sleep disorders

There are more than 80 different sleep disorders, the most common of which is obstructive sleep apnea (OSA). Almost all of the disorders have impaired alertness as a common presenting finding. Anesthesiologists with excessive daytime somnolence should consider being evaluated by a sleep medicine professional.

Optimize scheduling

The most contentious issue to attempt to change is that of individual work schedules. Monetary and time at work issues are central to most decisions regarding scheduling and often compete with what might be “correct” for both physiological and patient safety reasons. Important factors that need to be taken into account are:

• Length of shift

• Length of rest opportunities while off duty

• Number of consecutive duty periods

• Start and end time of shifts

• Periodic days off of clinical work

Modafinil – an alertness enhancing drug

Modafinil is a non-amphetamine stimulant with low physiologic addictive potential.15 Its current FDA approval is for excessive daytime sleepiness due to narcolepsy, OSA, and shift work sleep disorder. Use of modafinil in these populations is associated with improvements in daytime alertness and small improvements in performance. The development of these types of drugs is creating controversy in the medical profession regarding the ethical use of such stimulants while performing patient care activities. Cephalon, the maker of modafinil issued a statement that the drug is “not intended for use in helping residents work longer hours.”

Physicians’ ethical responsibilities to each other

In an ideal world patients would not suffer consequences of care from impaired providers. Honest dialogue between colleagues regarding coverage if sleep deprivation is present could improve system safety. This type of discussion occurs infrequently, most likely because of ego and a lack of understanding of the impairment caused by sleep loss and the consequent decrease in safety.

Rest breaks

Rest breaks are common in many anesthesia practices. These breaks interrupt the monotony of care, allow for personal time (food and bath room breaks) and allow for brief physical activity. Rest breaks have been shown to transiently improve alertness but they do not alter sleep pressure.16 Longer rest breaks can be used for strategic napping. Strategically placed rest and nap breaks could improve alertness, performance and likely morale and should be built into anesthesia practice when the situation allows.

Work schedules should be created with homeostatic and circadian processes in mind. Anesthesia groups should allow for flexibility to accommodate schedule requests that the providers desire but also have this balanced by the science of sleep medicine.

Hospital responsibilities to groups and physicians

Hospitals have an important role in how providers practice. Production pressure (case throughput) and a culture of safety are important factors especially in anesthesia practice. In hospitals where production pressure is high and a culture of safety is lacking, fatigue related problems are more likely be prevalent. Pressure to perform elective surgery in the middle of the night is but one example of how a hospital can increase the collective sleep pressure of all of its providers – not just the anesthesiologists. Some hospitals have acted proactively to address fatigue-related risk. The Veteran’s Administration has piloted strategic napping in the high acuity areas of its intensive care units and operating rooms.13

Key points

• Professional (and ethical) behavior begins with the practitioner who should come to work prepared and not impaired. Sleep deprivation is an important example of a common and preventable disorder negatively impacting the performance of anesthesiologists.

• Sleep is a basic human need. Sleep deprivation causes increasing physiologic pressure to acquire sleep, and resultant uncontrollable “microsleeps” create the risk of performance lapses and safety risks. In the case of anesthesiology practice, these risks are visited upon patients.

• Fatigue has been shown to affect performance to a similar degree that is associated with legal intoxication with alcohol.

• Anesthesiologists have ethical obligations to patients to obtain appropriate levels of sleep, and to work to optimize work schedules within groups to minimize sleep-disordered performance degradation.

• Anesthesiologists have ethical obligations to each other to promote work breaks, and discourage sleep deprivation among colleagues.

• Anesthesiologists also have obligations to work with hospitals to promote a culture of patient safety and proactively address fatigue-related risks.


1 ASA Guidelines for Ethical Practice of Anesthesiology. HOD October 2008.

2* Kohn, L.T.Corrigan, J.M., and Donaldson, M.S. (1999). To Err is Human: Building a Safer Health System. Washington, D.C.: National Academy Press.

3 Rosekind, M.R.Gregory, K.B.Miller, al. Aircraft Accident Report: Uncontrolled Collision with Terrain, American International Airways Flight 808, Douglas DC-8, N814CK, US Naval Air Station, Guantanamo Bay, Cuba, August 18, 1993. (Report #NTSB/AAR-94/04). Washington, D.C., National Transportation Safety Board, 1994.

4 Philibert, I. (2005). Sleep loss and performance in residents and nonphysicians: a meta-analytic examination. Sleep28, 1392–1402.

5 Landrigan, C.P.Rothschild, J.M.Cronin, al. (2004). Effect of reducing interns’ work hours on serious medical errors in intensive care units. N Engl J Med351, 1838–48.

6* Howard, S.K.Gaba, D.M.Smith, al. (2003). Simulation study of rested versus sleep-deprived anesthesiologists. Anesthesiology98, 1345–55.

7 Howard, S.K.Gaba, D.M.Rosekind, M.R., and Zarcone, V.P. (2002). The risks and implications of excessive daytime sleepiness in resident physicians. Acad Med77(10), 1019–25.

8* Dawson, D. and Reid, K. (1997). Fatigue, alcohol and performance impairment [Scientific Correspondence]. Nature388: 235.

9 Reason, J. (1990). Human Error. Camridge, Great Britain, Cambridge University Press.

10 Arnedt, J.T.Wilde, G.Munt, P., and MacLean, A. (2001). How do prolonged wakefulness and alcohol compare in the decrements they produce on a simulated driving task? Accid Anal Prev33, 337–44.

11* Arnedt, J.T.Owens, J.Crouch, al. (2005). Neurobehavioral performance of residents after heavy night call vs after alcohol ingestion. JAMA294, 1025–33.

12 Sallinen, M.Harma, M.Akerstedt, al. (1998). Promoting alertness with a short nap during a night shift. J Sleep Res7, 240–7.

13 Joint Commission Resources: Strategies for Addressing Health Care Worker Fatigue. (2008). Oakbrook Terrace, IL, The Joint Commission on Accreditation of Healthcare Organizations.

14 Bonnet, M.H. and Arand, D.L. (1994). Impact of naps and caffeine on extended nocturnal performance. Physiol Behav56, 103–9.

15 Czeisler, C.A.Walsh, J.K.Roth, T. et al. (2005). Modafinil for excessive sleepiness associated with shift-work sleep disorder. N Engl J Med353, 476–86.

16 Neri, D.F.Oyung, R.L.Colletti, al. (2002). Controlled breaks as a fatigue countermeasure on the flight deck. Aviat Space Environ Med73, 654–64.

Further reading

Borbely, A.A. and Achermann, P. (2005) Sleep homeostasis and models of sleep regulation, in Principles and Practice of Sleep Medicine, 4th Edition. Edited by Kryger MHRoth TDement WC.Philadelphia, PA, Elsevier Saunders. pp. 405–17.

Carskadon, M.A. and Dement, W.C. (2005). Normal Human Sleep: An Overview, in Principles and Practice of Sleep Medicine, 4th Edition. Edited by Kryger MHRoth TDement WC. Philadelphia, PA,Elsevier Saunders, pp. 13–23.

Lockley, S.W.Cronin, J.W.Evans, E.E., et al. (2004). Harvard Work Hours H, and Safety Group: Effect of reducing interns’ weekly work hours on sleep and attentional failures. N Engl J Med351, 1829–37.

National Sleep Foundation: 2008 Sleep in America Poll, 2008 website: Accessed June 1, 2009

Rosekind, M.R.Graeber, R.C.Dinges, D.F., et al. (1994). Crew Factors in Flight Operations IX: Effects of Planned Cockpit Rest on Crew Performance and Alertness in Long-Haul Operations. NASA Technical Memorandum #108839. Moffett Field, CA, NASA Ames Research Center

Van Dongen, H.P. and Dinges, D.F. (2005). Circadian rhythms in sleepiness, alertness, and performance, in Principles and Practice of Sleep Medicine, 4th Edition. Edited by Kryger MHRoth TDementWC. Philadelphia, PA, Elsevier Saunders, pp. 435–43.