OBJECTIVE: Growing evidence of an etiologic overlap between schizophrenia and bipolar disorder has become increasingly difficult to disregard. In this study, we examined paternal age, urbanicity of place of birth, being born "small for gestational age," and parental loss as risk factors for primarily schizophrenia and bipolar disorder, but also unipolar depressive disorder and schizo-affective disorder. Furthermore, we examined the incidence of the disorders in a population-based cohort and evaluated our results in the context of the Kraepelinian dichotomization. METHOD: We established a register-based cohort study of more than 2 million persons born in Denmark between January 1, 1955, and July 1, 1987. Overall follow-up began on January 1, 1973 and ended on June 30, 2005. Relative risks for schizophrenia, bipolar disorder, unipolar depressive disorder, and schizoaffective disorder (ICD-8 or ICD-10) were estimated by survival analysis, using Poisson regression. RESULTS: Differences were found in age-specific incidences. Loss of a parent (especially by suicide) was a risk factor for all 4 disorders. High paternal age and urbanization at birth were risk factors for schizophrenia. Children born pre-term had an excess risk of all disorders except schizophrenia if they were born "small for gestational age." CONCLUSIONS: An overlap in the risk factors examined in this study was found, and the differences between the phenotypes were quantitative rather than qualitative, which suggests a genetic and environmental overlap between the disorders. However, large gender differences and differences in the age-specific incidences in the 4 disorders were present, favoring the Kraepelinian dichotomization.