All TBR offenders who completed TBR between July 1, 1974, and June 30, 1979, (589) were selected for the study, and a control group consisted of 393 long-term inmates given unconditional prison sentences of at least 2.5 years and released during the same period as the TBR offenders. Recidivism data were obtained for both groups for no less than 3 and no more than 8 years. The study determined how many new offenses and what types of offenses were committed during and after completing TBR, whether there is a relationship between offenses committed after completion of TBR and their criminal records or length of TBR, and how many new offenses and what types of offenses were committed by the long-term offenders. About 60 percent of the TBR offenders were absent without leave at least once, and 25 percent were absent without leave no less than four times. Although 39 percent of them committed an offense during these times, only 11 percent committed a known violent offense. The TBR and long-term inmates did not differ significantly in recidivism, but this does not discount TBR effectiveness, since without it, the recidivism of the mentally ill offenders may have been greater. Only 16 percent of the TBR offenders received a prison sentence exceeding 6 months after discharge from TBR. 4 tables, 7 references.