This study assessed clinical and demographic differences between 74 geriatric psychiatry outpatients with early-onset vs late-onset depression. The following data were considered: age, gender, marital status, years of education, number of prescription medications and active medical diagnoses (including presence of various categories of medical disorder), presence of any comorbid dementia or other psychiatric disorder, age of depression onset, number of depressive episodes and MMSE score. Fifteen patients (20.3 %) had an early onset of depression (before age 60 years) and 59 (79.7%) had a late onset of depression. Early-onset patients had significantly more episodes of depression than late-onset patients (4.2 vs 1.9, t = 4.74, p < 0.001). Patients with early-onset depression also had a higher mean number of prescribed medications (5.3 vs 3.5, t = 2.29, p = 0.025) and active medical disorders (4.6 vs 3.1, t = 2.89, p = 0.005). Specifically, early onset of depression was associated with an elevated prevalence of cardiac disease (53.3% vs 23.7%, chi(2) = 5.0, df = 1, P = 0.025), diabetes (46.7% vs 16.9%, chi(2) = 6.0, df = 1, P = 0.015), gastrointestinal disorder (40.0% vs 12.0%, chi(2) = 6.5, df = 1, p = 0.011) and arthritis (26.7% vs 6.8%, chi(2) = 4.9, df = 1, p = 0.027). These findings support previous reports that people with a history of depression experience greater medical morbidity than those without a history of depression. The study groups did not differ with respect to MMSE score or presence of a concurrent dementia disorder. These results were unexpected given previous studies that indicate greater cognitive impairment in late-vs early-onset depression. The potential contribution of increased vascular risk factors among the early-onset depression group may have partly contributed to the finding of no difference in cognition between groups in the present study. (C) 1997 John Wiley & Sons, Ltd.