Effects of Mental Health Court, Specialized Probation, and Standard Probation on Recidivism Outcomes for Persons with Serious Mental Illnesses

Background: Mental health courts and specialized probation units are the two most common alternative to incarceration interventions for persons with serious mental illnesses (SMI).  However, most evaluation research on these programs has been criticized for studying new programs that are still in development, employing short follow up periods, and utilizing standard criminal justice programs as the only comparison condition.  This study addresses these methodological issues by conducting a longitudinal investigation of two longstanding  specialized programs – a mental health court and a specialized probation unit – as well as a standard probation unit, comparing each program’s effects on criminal recidivism.

Methods:  Administrative data were compiled for 864 persons with SMI in three distinct programs in a large Midwestern city: A mental health court, which has been in operation since 2004; a specialized mental health probation unit that was created in 1989, and a standard adult probation program.  For individuals who were discharged from these programs in 2009, criminal history information from first arrest through September, 2014 was received, allowing for five years of post-program analyses.  Logistic and zero-inflated negative binomial regressions were estimated with generalized estimating equations and mixed effects models to test program effects on arrest outcomes, controlling for demographic factors, SMI diagnosis, and successful vs. unsuccessful termination.

Results:  The sample of 864 individuals was 65% male, 35% African American, and 25% white, with a mean age of 38.8.  The two most common diagnoses were schizophrenia (19%) and major depression (21%).  Prior to entering each program, the prior number of arrests averaged 17.8, with an average 30.5 months incarcerated.  Persons exiting mental health court had the highest mean number of post-program arrests (4.4) compared to specialized probation (3.6) and standard probation (2.5).  However, mental health court participants had nearly double the amount of prior arrests and months incarcerated before entering the program.  In fully-adjusted longitudinal regression models, compared with MHC no significant difference in months incarcerated post-probation was found for specialized probationers (β= -.101, p=.78) or standard probationers (β= .504, p=.32).  Successful probation completion was strongly correlated with reduced months of pre-probation incarceration (β= -1.19, p<.0001), as was the number of days on probation (β= -.001, p<.01). Months incarcerated pre-probation (β= .006, p<.05) and a primary diagnosis of schizophrenia or schizoaffective disorder (β= .497, p<.001) were associated with longer incarceration post-probation. 

Conclusions:  This is the first known study to examine effects of two specialized criminal justice interventions on recidivism for people with mental illness.  Results indicate that each program engages a somewhat different client population in terms of criminal history and diagnosis.  Though post-probation recidivism is common, frequency of arrest and length of incarceration post-program is significantly lower for probationers with SMI in all three programs who successfully complete supervision.  However, approximately 50% of the sample did not successfully complete each program, indicating that continued efforts must be made to improve responsiveness of criminal justice interventions to the specific needs of persons with SMI in order to achieve improved outcomes.