OBJECTIVE: Major depressive disorder is associated with elevated mortality rates that increase with the severity of depression. The authors hypothesized that patients with psychotic depression would have higher mortality rates than patients with nonpsychotic depression. METHOD: Survival analytic techniques were used to compare the vital status of 61 patients with psychotic major depression with that of 59 patients with nonpsychotic major depression up to 15 years after hospital admission. Medical status was assessed with the Cumulative Illness Rating Scale. Dexamethasone suppression test (DST) data were available for 101 patients. RESULTS: The mortality rate for subjects with psychotic depression was significantly greater than that for those with nonpsychotic depression, with 41% versus 20%, respectively, dying within 15 years after hospital admission. A proportional hazards model with age and medical status entered as covariates confirmed a significantly higher mortality rate in patients with psychotic depression (hazards ratio=2.31). A positive DST result was associated with psychotic depression but was not related to vital status. CONCLUSIONS: Patients with psychotic depression have a two-fold greater risk of death than do patients with severe, nonpsychotic major depression.