Predictors of arrest during forensic assertive community treatment.

Objective: An emerging adaptation of the assertive community treatment model is forensic assertive community treatment (FACT), which aims to prevent criminal recidivism. This study examined predictors of arrest among patients in a prototype program and considered the implications of study findings for future development of the FACT model.

Methods: Demographic and clinical data from all 130 patients treated in Project Link from 1997 through 2003 were merged with a statewide criminal record database to identify variables associated with arrest.

Results: Poisson regression revealed a history of arrests for violent offenses before treatment and evictions from residential treatment, and antisocial traits were associated with arrest during treatment. Substance abuse was not associated with arrest, but the sample lacked heterogeneity for this factor.

Conclusions: Variables associated with arrest were similar to those seen in the general population. In the absence of a standardized model of intervention, FACT programs should incorporate interventions that target modifiable risk factors in order to prevent criminal recidivism among high-risk patients.