Short-term outcomes were evaluated for 65 children who were followed for 2, 4, or 6 months after psychiatric hospitalization. Child (e.g., aggression), parent (e.g., depression), family dysfunction (caretaker inconsistencies), and the modalities of treatment (e.g., point system) are described. Analyses of variance revealed no effects of follow-up interval or length of stay. A regression analysis revealed that low improvement was predicted by child attention deficit disorder with hyperactivity and depressive symptoms, older age, neurological dysfunction, and history of physical abuse. High improvement compared with low improvement children had a more successful adjustment in several critical roles and exhibited fewer individual problem behaviors.