relationships between sleep loss, shift duration, and medical errors among medical residents.
Medical residents work longer hours than virtually all other occupational groups (Steinbrook, 2002). During the first year, medical residents frequently work a 24-hour shift every third night (i.e., 96-hours per week). Two studies found that sleep-deprived surgical residents commit up to twice the number of errors in a simulated laparoscopic surgery (Grantcharov et al., 2001; Eastridge et al., 2003). In a survey of 5,600 medical residents, conducted by the Accreditation Council for Graduate Medical Education, total work time was inversely correlated with reported sleep time. Residents who worked more than 80 hours per week were 50 percent more likely than those working less than 80 hours to report making a significant medical error that led to an adverse patient outcome (Baldwin and Daugherty, 2004). The strongest evidence tying medical errors to sleep-related fatigue from extended work hours comes from an intervention trial designed to limit residents’ work hours (Box 4-1). Earlier attempts to demonstrate patient safety benefits by reducing resident hours were beset by methodological problems (Fletcher et al., 2004).
Residents are not the only health professionals to report medical errors in association with short sleep. Nurses who completed logbooks recording their schedule length, sleep, and errors, reported 3.3 times more medical errors during 12.5 hour shifts than 8.5 hour shifts (Rogers et al., 2004). Nearly 40 percent of the nurses reported having 12-hour shifts; and although their sleep duration was not directly studied, the findings suggest that fatigue is a major factor.
Children with obstructive sleep apnea (OSA) often have problems in development, cognition, behavior, and academic performance, according to detailed reviews of the evidence (Schechter, 2002; Bass et al., 2004). The risk of neurobehavioral abnormalities in children with severe OSA is about three times greater than in children without OSA (Schechter, 2002). The contribution of overnight reduction of oxygen levels in the blood (hypoxemia) in comparison to sleep disruption is unclear. One study shows an association with the lowest level of oxygen during sleep and scores in arithmetic (Urschitz et al., 2005), but other studies show cognitive or behavioral deficits in children who snore without severe sleep apnea (Kennedy et al., 2004; Rosen et al., 2004; Gottlieb et al., 2004; O’Hara et al., 2005). Outcome measures used in numerous studies include intelligence quotient, learning and vocabulary, attention, symptoms of attention deficit hyperactivity disorder (ADHD), and academic performance. For example, two historical