A relationship-based care model for jail diversion

This study assessed the effectiveness of a postbooking jail diversion program for a homeless population with mental illness in South Florida, as measured by rate of arrests after admission to the program. The program (termed relationship-based care) is structured to ensure access to psychiatric and primary health care, delivered within a theoretical framework developed for working with this population.

Data were reviewed from the Criminal Justice Information System in Miami-Dade County for 229 adults who were arrested and found to be appropriate for jail diversion. Data for 151 individuals who were consecutively diverted to the relationship-based care program were compared with data for a control group of 78 individuals who had been diverted to other programs in the community. Arrest rates for each participant during the year before diversion were compared with arrest rates for the year after diversion. In addition, for persons in the relationship-based care program, demographic data, type of homelessness (chronic or situational), and number of psychiatric contacts were analyzed to determine the impact of these variables on outcome.

A highly significant reduction in arrest rates for individuals diverted to the relationship-based care program was observed. However, the arrest rate for the control group remained nearly identical before and after diversion. For the relationship-based care group, prediversion arrest rates, duration of participation in the program, and number of psychiatric contacts accounted for a significant portion of the recidivism variance.

The relationship-based care model described here appears to be an effective strategy for reducing criminal recidivism. Length of participation in the program and involvement in psychiatric treatment were correlated with reduced arrest rates. Identifying individuals who are at risk of poor engagement in community services and subsequent premature departure remains a challenge.