In a longitudinal epidemiological study of young adults, we estimated the association between sleep disturbance and psychiatric disorders, cross-sectionally and prospectively. A random sample of 1200 was drawn from all 21-30-year-old members of a large health maintenance organization (HMO) in Michigan; 1007 were interviewed in 1989 and 979 were reinterviewed in 1992. Lifetime prevalence of insomnia alone was 16.6%, of hypersomnia alone, 8.2%, and of insomnia plus hypersomnia, 8%. The gender-adjusted relative risk for new onset of major depression during the follow-up period in persons with history of insomnia at baseline was 4.0 (95% confidence interval [CI] 2.2-7.0) and in persons with baseline history of hypersomnia, 2.9 (95% CI 1.5-5.6). When history of other prior depressive symptoms (e.g., psychomotor retardation or agitation, suicidal ideation) was controlled for, prior insomnia remained a significant predictor of subsequent major depression. Complaints of 2 weeks or more of insomnia nearly every night might be a useful marker of subsequent onset of major depression.