This article reports on the evidence for the validity of psychotic major depression as a distinct subtype based on cross-sectional and 1-year prospective data from the Epidemiologic Catchment Area study. Consistent with findings from previous clinical studies, only about 14% of major depressions were accompanied by psychotic features. Psychotic as compared with nonpsychotic depression had a more severe course, as reflected in increased risk of relapse, persistence over 1 year, suicide attempts, hospitalization, comorbidity, and financial dependency. These differences could not be explained by differences in demographic characteristics or by symptom severity, as assessed by symptom profile or number of symptoms. The boundary problem with schizophrenia and bipolar affective disorder that is seen in clinical studies was also found in this sample. To our knowledge, this is the first study to examine the validity of psychotic depression in a community sample; the findings are consistent with those from clinical samples. They support the clinical significance of psychotic depression and the continuation of its inclusion as a distinct subtype in DSM-IV.