The association between incarceration and psychiatric disorders has been noted. Yet, existing studies are cross-sectional or examine the risk of recidivism, which has limited the predictive validity of psychiatric disorders as a risk factor for incarceration. To overcome this limitation, this study used a prospective cohort to examine whether psychiatric diagnoses in early adulthood predicted incarceration throughout a 30-year follow-up. It tested the association between psychiatric diagnoses with future incarcerations, their number and durations, controlling for education and ethnic status. This study merged data from three sources in Israel: a prospective 10-year birth cohort study of young adults aged 25-34, conducted in the 1980s (N = 4914) that included a psychiatric interview; data from the Prison Service, including the cause, number and duration of incarcerations; and from the Vital Statistics Registry on death records. Multivariate analysis showed that substance-use disorders, antisocial personality and lower levels of education predicted future incarceration, their number and maximum duration. The remainder diagnoses were not significantly associated with future incarceration. Results limited the prediction of future incarcerations to persons diagnosed with substance use and antisocial personality, and do not support an independent predictive association between additional psychiatric diagnoses and future incarceration.