This investigation addresses two issues central to psychiatric hospitalization: (1) Can alternative treatments be effective? and (2) How can patients appropriate for the alternatives be identified? This random design study found that a majority of patients assessed as requiring immediate psychiatric hospitalization could utilize a combination of emergency housing and outpatient treatment to avoid hospitalization. Five key variables examined at intake proved useful in predicting ability to succeed in the alternative treatment setting. Patients who failed in the alternative setting had trouble utilizing leisure time, were grandiose, exhibited eating problems, had difficulty with family, and showed an absence of paranoia.