Differential gender effects of depression on use of HIV medications among HIV-positive Puerto Rican drug users

Abstract

Many barriers to the use of HIV medications have been identified. Research findings have also shown a gender disparity in HIV care behaviors. However, interaction effects of gender with the potential barriers to use of HIV medications among HIV-positive minority drug users remain under-studied. This study examined interaction effects of gender with potential moderating factors (i.e., individual and network characteristics) on the use of HIV medications. Analyses were based on 260 HIV-positive Puerto Rican heroin and cocaine users, recruited in New York (N=178) and Puerto Rico (N=82) in 1998–2003. HIV status was assessed using OraSure, and heroin or cocaine use was verified by urinalysis. All participants were tested and interviewed at baseline and six-month follow-up (183 males; 77 females). In predicting use of HIV medications at follow-up (HIVMEDF), use of HIV medications at baseline (HIVMED), individual characteristics (e.g., depression), network characteristics (e.g., having an intravenous drug user [IDU] sex partner), recruitment site, and interaction effects of these variables with gender, were examined in multiple logistic regression analysis. Use of HIV medications was low (29% at baseline; 40% at follow-up). HIVMED, recruitment site, gender, and depression had significant main effects on HIVMEDF. Depression also had a significant interaction effect with gender on HIVMEDF. Unlike men, women with depression were less likely than women without depression to use the medications. The findings indicate that gender-specific issues should be addressed by treatment programs for HIV-positive drug users, with particular efforts needed to enhance use of medications for depressed women.

Keywords:

• gender
• depression
• use of HIV medications
• Puerto Rican drug users

Acknowledgements

This data analysis was supported by the National Institute on Drug Abuse, Grant No. R03DA022918. Data used were obtained from a project (ARIBBA) funded by the same Institute, Grant No. R01DA10425. The authors also appreciate Dr. Rafaela R. Robles and her staff for their collaboration in conducting the ARIBBA project in PR.