Deinstitutionalization following the introduction of antipsychotic medications in 1954 has received much attention as a major narrative in psychiatry. Little attention has been given, however, to deinstitutionalization before 1954. Using United States census data on discharge and readmission rates of US mental hospitals from 1935 to 1964, this article analyzes deinstitutionalization using an interrupted time-series model, with particular attention to the statistical significance of trends before and after the advent of antipsychotics. Discharge rates significantly increased in the period before antipsychotics, indicating that deinstitutionalization began before 1954, although readmissions during that same period increased at the same rate as discharges. A reasonable inference is that patients discharged in the pre-antipsychotic period were unable to live independently outside the hospital. After 1954, both discharges and readmissions increased significantly, but due to a continuing increase in admissions, no significant decrease in mental hospital populations occurred during the seven-year period after 1954. The decline began in 1961 and coincided with changes in federal policy. The fate of mental patients discharged from hospitals during this second period of deinstitutionalization is examined. The central conclusions are (1) the overall reduction in the population of mental hospitals did not coincide with the 1954 introduction of antipsychotic medications, and (2) deinstitutionalization before and after drugs has been met with inadequate community-based care.