Abstract
SUMMARY This review of experimental studies focuses on the disturbances of sleep that may occur in connection with irregular work hours and their relation to ensuing alertness. Three aspects of sleep are covered: the duration of sleep, the stages of sleep, and the continuity of sleep. A sleep curtailment as small as 2 h has clear negative effects on alertness. Sleep disturbed as frequently as every minute clearly affects alertness, but such procedures also affect the stages of sleep and decrease total sleep time. However, fragmentation rates of one per 10 minutes also induce sleepiness without affecting sleep content or duration. There is no clear evidence for slow-wave sleep (SWS) being more important than other stages, but designs may not have been sensitive enough. In connection with fragmentation studies it is suggested, however, that the effects are stronger if more SWS is lost. In summary, the review suggests that the relatively mild disturbances of sleep in shift work may contribute to reduced alertness although there are probably other, more potent, factors.