Risk of arrest among public mental health services recipients and the general public.

Objective: This study compared arrest rates on a broad range of offenses in a cohort of public mental health service recipients and in the general population. Methods: Administrative data from a state mental health agency were merged with data capturing arrests over a 9.5-year period in a cohort of persons with severe and persistent mental illness who used public mental health services and were aged 18-54 (N=10,742). The cohort's arrest rates for eight offense categories were compared with those of the general population for persons in the same age group over the same period (N=3,318,269). The data for the cohort that received mental health services were weighted by age and gender to align the cohort's demographic characteristics with those of the general population. Results: The service use cohort members' odds of experiencing at least one arrest in any charge category were significantly higher than those of the general population (odds ratio [OR]=1.62, 95% confidence interval=1.52-1.72); odds were higher across all charge categories, with ORs ranging from 1.84 for drug-related offenses to 5.96 for assault and battery on a police officer. Aside from the crime of assault and battery on a police officer, the largest ORs were associated with misdemeanor crimes against persons and property and with crimes against public decency. ORs associated with felony charges, while significant, tended to be slightly smaller in magnitude. Conclusions: The offenses for which persons with serious mental illness are at greatest risk of arrest are many of those targeted by current diversion programs. These findings suggest the need for additional research addressing the ways in which individual psychopathology and socioenvironmental factors affect risk of offending in this population. (Psychiatric Services 62: 67-72, 2011)