Meta-Analysis of Alcohol and Serodiscordant Condomless Sex Among People Living with HIV

Abstract

While observational studies have found that alcohol consumption is associated with serodiscordant condomless sex among people living with HIV (PLHIV), no meta-analysis has yet examined this trend. We conducted a meta-analysis to synthesize empirical evidence on the association between alcohol and condomless sex with partners at risk of HIV acquisition. To meet inclusion criteria, studies: (1) specifically targeted PLHIV or provided stratified data for HIV-infected participants; (2) provided a quantitative measure of alcohol use; (3) provided a quantitative measure of condomless sex with serodiscordant partners; and (4) reported the results of statistical tests examining the relationship between alcohol use and serodiscordant condomless sex. Using random-effects models, weighted effect sizes were calculated. Three separate analyses were conducted to examine serodiscordant condomless sex in association with any alcohol consumption, binge/problematic drinking, and alcohol in a sexual context. A total of 36 independent effect sizes from 27 studies (including 25,065 HIV-infected participants) were pooled in the meta-analysis. Any alcohol consumption, binge/problematic drinking, and alcohol use in a sexual context were each associated with condomless sex with serodiscordant partners [OR 1.64 (95% CI 1.46–1.85); OR 1.65 (95% CI 1.14–2.39); OR 2.88 (95% CI 2.01–4.12), respectively]. Meta-analytic findings demonstrate a consistent positive relationship between alcohol use and serodiscordant condomless sex among PLHIV. Future public health programming for HIV-infected individuals needs to address the role of alcohol consumption in sexual risk-taking behavior.