HIV risk reduction among injection drug users: explaining the lack of anticipated outcomes in a community-level controlled comparison study

A quasi-experimental evaluation design was employed in order to determine whether community outreach efforts would result in reductions in drug- and sexual-related HIV risk behavior among injection drug users, and to gain insight into the potential effect of enhanced outreach on increased client participation in drug abuse treatment and STD/HIV services. Interviews were conducted with 1943 randomly selected drug users in two control neighborhoods and two experimental neighborhoods in the South Bronx. The sample was largely indigent males in their mid 30s, most with less than a high school education, of Hispanic or African American background. Data were gathered before and after enhanced outreach was instituted in the two experimental communities. Needle sharing decreased and bleach cleaning increased in both pairs of neighborhoods; however, sharing of cookers, cotton, and rinse water increased. Unprotected vaginal and anal sex decreased over time, but no between-condition differences in reductions in drug- or sex-related risk behaviors were found. Over time, an increasing proportion of drug users in both study conditions reported having attended substance abuse treatment, but longitudinal outcomes did not favor the enhanced outreach neighborhoods with respect to exposure to outreach messages or acceptance of referrals to chemical dependency treatment or STD/HIV services. In sum, HIV risks declined over time in both conditions, but there were no apparent differences in behavior change or treatment/services entry between the control and enhanced outreach neighborhoods. The study underscores the difficulty of interpreting cross-community comparisons of intervention outcomes in an environment of changing risks and neighborhood dynamics. (C) 2000 Elsevier Science Ltd. All rights reserved.