文摘
Sexual offender literature has focused on the importance of recidivism reduction in an effort to increase public safety. While cognitive-behavioral therapy (CBT) programs are currently considered a mainstream treatment method, it is essential to study recidivism as an indicator of treatment effectiveness. This meta-analysis examined research published since 1970 to determine the overall effectiveness of treatment in reducing recidivism among adult male sexual offenders. Treatment time implementation and CBT treatment features were also assessed as subgroup variables. Each study was reviewed to determine whether it evaluated treatment effectiveness, as defined by recidivism. Recidivism must have been defined as rearrest, or reconviction for a general, sexual, or violent/combination offense after treatment concluded. Studies that did not include recidivism as outcome data were excluded. Qualifying studies included those utilizing CBT treatments for adult male sexual offenders in a community or institutionalized setting. The reported recidivism rates of a treated sexual offender sample must have been compared with a control group using similar criteria for both groups. All articles were coded by trained researchers using a subject specific manual. To ensure interrater reliability, all articles were blind coded and discrepancies were discussed until resolved. Results were computed using Comprehensive Meta-Analysis software from Biostat. The results from each of the 25 studies identified were converted into weighted effect sizes utilizing a random effects model (REM), which resulted in 42 effect sizes. Significant overall effect sizes were found for sexual and violent/combination recidivism； however, multiple indices indicated heterogeneity in the effect sizes. Assessment of treatment time implementation as a subgroup variable indicated significant differences in the overall effectiveness of treatment by decade with treatment delivered during the 1990s related to lower levels of sexual and violent/combination recidivism. Finally, while no significant differences in the effect sizes of studies were found among CBT treatment features for general recidivism, there was considerable variability among the studies employing just relapse prevention than studies with a combination of relapse prevention and Risk-Needs-Responsivity. The results are discussed as to the implications for future treatment and research in the area of sexual offender treatment.
