This study compared the service utilization, incidents, and hospitalizations of forensic clients and non-forensic clients in a supportive housing program for the mentally ill. 28 forensic consumers were compared to 28 non-forensic consumers matched on age, sex, ethnicity, diagnosis, and hospitalization history across four quarters of services data via ANOVA with repeated measures and substance abuse history as a covariate. Multiple and logistic regression evaluated the contributions of age, sex, length of stay, type of case management, substance abuse history, and incarceration history to the variance of 123 clients for incidents and hospitalizations. Forensic clients modestly utilized more services during their first year in the residence. Incarceration history was the dominant predictor of incidents, but not hospitalizations. These findings may help in determining if a jail diversion program should include a specialized community-housing component.