A Meta-Analysis of HIV Prevention Interventions Targeting Women With Criminal Justice Involvement
Women are one of the fastest growing segments of the U.S. criminal justice system. Female offenders have HIV rates that are three to fifteen times that of the general population of women. In correctional settings, HIV prevalence is also significantly higher for women than men, even after controlling for race and ethnicity. Only in recent years has HIV prevention science begun to build an evidence base in targeting criminal justice-involved women. What is currently lacking is evidence of the relative efficacy of HIV prevention interventions for women with criminal justice involvement. We also do not thoroughly understand the effect of an intervention’s setting within the criminal justice landscape or its theoretical orientation on reducing HIV risk. Toward this end, we conducted a systematic review and meta-analysis of HIV prevention interventions for women with criminal justice involvement.
Methods
We conducted a systematic review of the literature published from 1980 until April 2013 on HIV prevention interventions for women with criminal justice involvement, using 26 search terms in 13 online article databases. Inclusion criteria for studies consisted of: 1) criminal justice involvement of participants, 2) an exclusively female sample or gender-specific outcomes, 3) presence of a control/comparison group, and 4) at least one quantitative behavioral HIV risk outcome. Articles were coded on multiple study factors, including setting, theoretical orientation, race/ethnicity, and juvenile/adult sample composition.
We calculated the standardized mean difference (SMD) or Cohen’s d for each intervention by dividing the mean difference by the pooled standardized deviation of the two groups. Effect sizes for the studies, based on the SMD were estimated using a random effects model, using the statistical software, Comprehensive Meta-analysis. Particular attention was paid to intervention setting and specific theoretical basis, if any.
Results
A total of eighteen journal articles were found that addressed HIV interventions targeting women with criminal justice involvement. Six articles were eliminated either because the same intervention was reviewed in multiple articles or because the study did not have a control/comparison group, leaving a final sample of twelve and a total nof 3,087. Nine studies were randomized control trials, and three had quasi-experimental designs. While ten interventions were delivered in prison or jail settings, only two were delivered in the community. The effect size for reduction in sexual risk behavior outcomes (i.e. number of unprotected sexual occasions) was -0.239 [-0.386, -0.091; SE=0.087, p=0.002]. Forest plots were created to illustrate the range of effect size across interventions.
Conclusions & Implications
Extant HIV prevention interventions are moderately efficacious at addressing the reduction of HIV sexual risk for women offenders, although there is need for further intervention development and implementation across the criminal justice spectrum. Given the results of this systematic review, this development should emphasize the use of cognitive-behavioral and social cognitive approaches to maximize the potential impact of HIV prevention interventions for this vulnerable population.