Practice Profile: Adult Reentry Programs

Ndrecka (2014) examined the effectiveness of adult reentry programs through a meta-analysis of 53 experimental and quasi-experimental evaluations, that resulted in the calculation of 58 effect sizes. Studies were included in the meta-analysis if they 1) evaluated a reentry program defined as a correctional program that focuses on the transition of individuals from prison into the community, including treatment or services that have been initiated while the individual is in custody and a follow-up component after the individual is released; 2) had at least one measure of criminal behavior as an outcome measure; 3) used an experimental or quasi-experimental design; 4) provided enough information to calculate an effect size; 5) were published after 1980; and 6) included only adults in its sample.

A thorough search of the literature was conducted using eight different methods, which led to the inclusion of the 53 eligible studies.  First, a keyword search was conducted using a variety of databases. Second, the bibliographies of four articles that describe reentry program research were examined to find additional studies. Third, websites for government agencies, such as the National Institute of Justice, were searched for publications and reports on reentry programs. Fourth, the National Reentry Resource Center was searched, which includes reports of reentry program evaluations. Fifth, the websites for each state’s department of correction and office of juvenile corrections were visited and examined to discover evaluation reports. Sixth, the websites for certain research agencies were examined to find unpublished studies. Seventh, the Ancestry method was used to examine the bibliographies of each study that was collected for inclusion in the analysis. Lastly, a keyword search was used with Google and Google Scholar to find other published and unpublished studies. The final 53 eligible studies used in the meta-analysis that resulted in 58 effect sizes and had a combined sample size of 289,125 participants and publication dates that spanned from 1980 to 2013.

Of the included studies, 27.6 percent provided mixed/combination services in their programming, 20 percent employed a therapeutic community model, and 17.2 percent offered transition employment programming. About 35 percent of the programs involved the combination of a prison and a community-based correctional facility as the location of treatment. Additionally, the majority (53.4 percent) of programs included in the meta-analysis lasted 3 months or longer, and about 47 percent had voluntary participation. Regarding treatment targets, about a third of the programs targeted employment and vocational training, and about 29 percent targeted substance abuse. Lastly, about 43 percent of programs applied a mixed treatment format, and about 40 percent applied a case-management treatment format. Of the samples in the included studies, 72.4 percent employed a mixed-race population, 12.1 percent employed a mostly Black population, 3.4 percent employed a mostly white population, and 12.1 percent lacked this data.  

The source of recidivism data was collected via self-report, official records, or other sources (not named). The form of effect size selected was the correlation coefficient (r) and while both a random effects and fixed effects model was reported for the overall effect sizes, the more conservative random effects model was reported for all of the findings.

Berghuis (2018) conducted a meta-analysis to aggregate the results from randomized controlled design studies that examined the effects of reentry programs to reduce recidivism for adult males. To be included, studies had to 1) include a target population of formerly incarcerated adult males (18 years or older at the time of release); 2) include interventions to aid in reintegration back into the community by providing assistance with employment, housing, social support, and/or substance use; 3) use a randomized controlled design; and 3) measure recidivism as a primary outcome. Recidivism measures included rearrest, reconviction, and reincarceration.

A comprehensive search of relevant databases, journals, and other sources yielded a total of 8,179 titles. From those results, a total of nine randomized controlled studies were included in the review. Six of the studies were published in academic journals, two were government reports, and one was a dissertation. The sample size of the included studies ranged from 40 to 1,813 (across all nine studies, the total sample size was 4,585). All participants in the studies were men. The average age ranged from 27 to 37 years. Participants were predominately non-white, and their most common offenses were property and drug crimes. Eight studies took place in prisons, and one study took place in a community-based setting following release from prison. Studies that provided specific information about location indicated that they were conducted in Florida; Minnesota; North Carolina; Philadelphia, Pa.; and Washington.

Of the nine studies, five focused on unimodal programs, (one aspect of reentry), and four focused on multimodal programs (several aspects of the reentry). The unimodal programs addressed work and employment-related issues, including work release interventions, temporary paid jobs, support services, and job placement. The multimodal programs tended to have several phases and focused on continuity of care. These programs started prior to release, and continued with post-release services that included housing, subsidized employment, mentoring, and drug treatment.

Random effects analysis was conducted on all included studies. Effect sizes in the form of odds ratios were calculated for each study.