An epidemiologic sample of first admission psychotic patients is used to examine naturalistic treatment of psychotic mania. Specifically, we examined if 15-20 year olds receive different medications, or respond differently from patients over age 30 who are also early in their course of illness. The major difference in the two groups, besides their age, is the presence of comorbid externalizing disorder and substance abuse in the younger group. This report further examines the impact of this comorbidity on the presence of further episodes and overall outcome. The findings indicate that acute treatment is very similar in the two groups, and that over a 4 year follow-up, about 40% of patients discontinue medication. In spite of that, 32% of the YOUTH subjects and 48% of the ADULT group did not have a second episode. Of those externalizing YOUTH with a single episode, 64% took no medication during follow-up. However, the major difference between admission, 24 month follow-up and 48 month follow-up was the discontinuation of substance/alcohol use and the improvement in functioning. Medication did not invariably make that difference. This community sample reveals that at 48 months, a significant minority of young and adult subjects hospitalized with psychotic mania will have a single episode, and that short term course in younger patients is considerably worse than longer term course.