OBJECTIVE: The purpose of the study was to examine the course of involuntarily administered medication in a state hospital population. METHOD: The authors retrospectively examined the records of all 51 involuntarily medicated patients in six state hospitals in New York City in a single calendar year. Clinical course was recorded for the period of involuntary medication and for 12 months thereafter. These patients were compared to 51 patients on the same wards who accepted medication. RESULTS: Clinicians assessed involuntarily medicated patients as more dangerous to themselves or others and less delusional after treatment than the comparison patients. Long-acting intramuscular antipsychotics were prescribed more frequently for involuntarily medicated patients. No differences were observed in rates of discharge, outpatient cooperation, or rehospitalization. Half of the patients in both groups remained continuously institutionalized, and of those who left the hospital, only 30% of the involuntarily medicated group and 40% of the comparison group took medication as outpatients. CONCLUSIONS: For these chronically severely ill patients, involuntary medication did not appear to enhance insight or cooperation or result in rapid return to the community. Involuntary medication is often a necessary short-term, in-hospital management strategy, but it does not replace the need to develop comprehensive, long-term inpatient and community-based approaches to the management of treatment refusal.