Among the most serious social ills facing urban communities are the overlapping problems of AIDS, drug use and criminal behavior. Designed in response to the observed covariance of AIDS, drug use, and criminal activity, this investigation addresses the critical need for theoretically sound, empirically derived interventions to reduce the risk of HIV transmission during the progression from incarceration to the unprotected environment. Part of a NIMH multisite AIDS prevention trial, the project will test the efficacy of a cognitive-behavioral group skills-building intervention designed to reduce the spread of AIDS among male drug-involved offenders in work-release facilities. The study will be carried at two work release correctional facilities, Lincoln with 528 inmates and Edgecomb with 569 inmates, of whom 79% and 81% are self-reported drug users. The study will include three phases: 1) ethnographic observation, focus groups, and key informant interviews inform the design of the intervention, measurement development, training of interviewers and group leader preparation; 2) next, intervention development is followed by pilot testing of experimental, control, and measurement protocols; and 3) following pretesting of 600 randomly selected work-release participants, 500 are later randomly assigned to skills-building or AIDS information conditions. Subjects in the skills condition will attend 16 two-hour group sessions in the work release facility and 8 booster sessions following release over the next 6 months. Subjects in the AIDS information condition will receive 3 two-hour group sessions in work-release. Study participants will be measured twice within the correctional setting, and measured five times following release for a total of 7 measurement occasions. Outcome variables include sexual risk behavior, drug use and risk-taking, and criminal recidivism. The study will test several hypotheses concerning the longitudinal effects of the intervention on outcome variables and on such mediating variables as coping skills, self-efficacy, social support, and help-seeking, and will delineate a nonrecursive causal model examining the causal linkages between the intervention, mediating and outcome variables over time. This model is based on the proposed assumptions that executing adaptive coping skills will improve social support, summon psychological resources, and enhance psychological well-being-thereby leading to risk reduction. The research will be led by investigators from Columbia University and conducted collaboratively with the New York State Department of Correctional Services and two community agencies serving the needs of offenders who are returning to the community.
