• An effort was made to determine patient characteristics that have differential prognostic significance, depending on treatment assignment to one of three treatment approaches: standard inpatient care (n = 63), brief hospitalization followed by day care (n = 61), and brief hospitalization without day care (n = 51). All were followed by outpatient care. Both demographically and clinically assessed behavioral variables were related to a number of outcome measures, including days in the community, clinical ratings, and family assessment. Generally, the standard treatment was inferior to the two brief treatments. Multiple previous admissions were particularly contraindicative for standard treatment. High overt anger score was especially contraindicative for brief hospitalization without day care and particularly indicative for brief hospitalization with day care.