Very little is known about the effects on sleep and sleepiness of entering or exiting shiftwork. The present study used a longitudinal database (n = 3637). Participants completed a questionnaire on work hours, sleep, and work environment at the start and end of a 5-yr period. Changes in shift/day work status were related to change in a number of subjective sleep variables using logistic regression analysis. The analyses were adjusted for age, sex, and differences in socioeconomic status, work demands, work control, physical workload, marriage status, and number of children. In comparison with constant day work, entering shiftwork (with or without night shifts) from day work increased the risk of difficulties in falling asleep, and leaving shiftwork reduced this risk (odds ratio [OR] = 2.8 [confidence interval, CI = 1.8-4.5]). Also falling asleep at work showed a consistent pattern; an increased risk of falling asleep for those with shiftwork on both occasions, and for those with night work on both occasions. Also entering night work was associated with a strongly increased risk of falling asleep at work (OR = 2.9 [CI = 1.3-6.7]). These results suggest that entering and leaving shiftwork has a considerable impact on sleep and alertness. However, there is a need for large and more extended longitudinal studies to support our findings.