Research points to best practices to reduce recidivism

The Ohio residential correctional programs -- halfway houses and community-based correctional facilities -- that are most successful at reducing recidivism among offenders enjoy an impressive track record.

An offender participating in the state's most successful programs is 50 percent less likely to engage in criminal activity in the two-year period following release vs. offenders who receive no guidance or services. In contrast, the least successful of these programs actually increased the chances of criminal recidivism. The least successful programs increased recidivism rates among offenders by 32 percent in comparison to recidivism rates among offenders receiving no follow-up guidance or services.

That's according to University of Cincinnati criminal justice research to be presented July 17-23 at the 32nd International Congress on Law and Mental Health at Humboldt University in Berlin, Germany. That research was carried out by Paula Smith, UC assistant professor of criminal justice; Edward Latessa, professor and director of UC's School of Criminal Justice; Lori Brusman-Lovins, UC research associate of criminal justice; and Matt Makarios, assistant professor of criminal justice at the University of Wisconsin-Park Side.

Their research, "The Importance of Correctional Program Characteristics and Their Relationship to Offender Outcomes," focused on 64 residential correctional programs in Ohio -- 44 halfway houses and 20 community-based correctional facilities -- involving 20,000 adult offenders. The three-year research project was sponsored by the Ohio Department of Rehabilitation and Corrections.

The purpose of the research is to assist in designing more effective correctional programs.

Program Characteristics That Lead to Success

While some variation in success rates can be attributed to the individuals enrolled in those programs, the three-year UC study found five program elements that were key to successful outcomes.

These were:

• Leadership and implementation. This category refers to a strong network of support for the residential correctional program both within the facility and in the wider community. Explained UC's Latessa, "This is referring to whether the program's leadership is on the shop floor and involved in day-to-day operations. Do the related systems -- the probation, court and state structures -- support the correctional facility? Is there volunteer support from the neighborhood? If the facility is hosting community picnics, that's a good sign. If there are pickets when I show up, I know that's a bad sign."

• Staff. This category refers to appropriate ongoing training in effective techniques and approaches in providing treatment. It also refers to quality of supervision, provision and communication of ethical guidelines, and regular meeting times.

• Offender assessment. This category refers to appropriate and effective assessment for actuarial risks and needs of participating offenders, followed up by good case plans for individual offenders. In other words, certain offender characteristics -- like marital status, language ability and reading level -- contribute to the likelihood that an offender will return (or not) to criminal activity. The treatment program personnel need to be able to accurately judge these risks and take the risks into consideration when devising a case plan for an individual.

• Treatment components and core correctional practices. The best treatment programs follow a model that is based on data-driven evidence of what works and what doesn't. Said Latessa, "It's also about follow through, having a treatment model that incorporates appropriate rewards and appropriate consequences and then following up with those rewards or consequences."

• Quality assurance. Finally, good programs monitor themselves and their participants. They track their own recidivism rates. They debrief frequently to encourage learning. They conduct exit interviews of participants. The continually reassess to make sure they're doing what they're supposed to do.

UC's Latessa and Research Associate Brian Lovins will also make a panel presentation.

.