Improvements in post-release outcomes of former prisoners (FP) are linked to behavior and attitudes influenced by social relationships and social support. Positive social support from family members or friends (i.e., naturally occurring support) plays a beneficial role in decreasing substance misuse and reducing criminal behavior in FPs. Despite evidence for the role positive social support can perform in the lives of FPs, social scientists continue to underutilize social support in interventions with this population.
A randomized controlled trial was conducted to compare the effects of Support Matters, a cognitive-behavioral post-release social support intervention, to routine post-release services offered to a sample of 80 male prisoners released to a large urban county. We hypothesized that Support Matters participants would demonstrate significantly greater improvements than control participants in the proximal outcomes: (a) extent of social support, (b) commitment to social support, and (c) social cognitions/beliefs and the distal outcomes of post-release substance misuse and arrests.
Methods
Support Matters was delivered to 8-total cohorts between October, 2009 and January, 2010. Prisoners (N=80) were recruited prior to release and randomly assigned to treatment (Support Matters) or routine services (control) conditions post-release. Participants completed standardized social support, psychosocial, and substance use instruments pre-release, prior to the intervention period post-release (T1), immediately following the intervention (T2), and 3 months after the intervention (T3; 6 months post-release).
Proximal outcomes were assessed with repeated-measures ANCOVAS examining within and between group effects across T1, T2, and T3. Mauchly's Test of Sphercity was conducted. Distal outcomes were examined using logistic and multiple regression models. All models controlled for covariates that differed significantly between groups prior to intervention enrollment.
Results
Support Matters participants reported increases in (a) tangible social support (p=.043); (b) instrumental support from family members (p=.026); and (c) reciprocity of support to support providers (p=.020). There were neither significant main effects nor intervention-by-time effects for commitment to social support or for changes in thinking patterns or beliefs. Program participation was not statistically significantly related to substance misuse outcomes. However, other things being equal, every one-unit increase in the severity of substance of choice at T3 increased the frequency of substance misuse by 1.38 units. For re-arrest rates, treatment condition effects approached statistical significance (p = .059) and the effects trended in the direction hypothesized. This trend indicates that when holding age, self-assessment of need for treatment, and substance of choice equal, Support Matters participants' odds of arrest are reduced by 95%.
Conclusion
Support Matters is the first trial of a naturally occurring support intervention for FPs with substance use disorders. Given that members of ethnic minority and poor urban communities are grossly overrepresented in the US prison system, improving outcomes of FPs is a pressing issue of social and economic justice. Yet, still preliminary and formative are answers to the question of how best support from naturally occurring relationships can be utilized in formal social support interventions with FPs. This study is one of the first to contribute to that knowledge development endeavor.
