Alcohol and HIV Seroconversion in Men Who Have Sex with Men

Sander, Petra

Download PDF

Request Version for Screen Reader

Last Modified

March 21, 2019

Creator

Sander, Petra
- Affiliation: Gillings School of Global Public Health, Department of Epidemiology

Abstract

Previous findings linking alcohol consumption and HIV seroconversion among men who have sex with men (MSM) have been inconsistent. This study argues those findings may be limited by inadequate control for confounding due to time-updated factors affected by prior alcohol consumption. We examined prospective data from 3,725 HIV-seronegative MSM enrolled in the Multicenter AIDS Cohort Study (MACS). Participants made semiannual study visits from 1984-2007, self-reported alcohol consumption, and underwent HIV testing. Potential predictors of alcohol consumption were explored. Demographic and time-updated factors including smoking, self-reported depression symptoms, illicit drug use and sexual practices were considered. Associations were modeled using logistic and lognormal multivariable regression. We next examined the joint effects of alcohol consumption and unprotected receptive anal intercourse (URAI) on HIV seroconversion. At baseline, median alcohol consumption by seronegative MACS men was 5 drinks/week (interquartile range (IQR): 1-12) and 18% were nondrinkers. Over 47,261 follow-up visits, median alcohol consumption was 2 drinks/week (IQR: 0-8) and 26% of participants were nondrinkers. Comparing users of marijuana, poppers and cocaine to non-users, the odds ratio (OR) for any current drinking among drinkers at the prior visit was 2.90 (95% CL: 2.21, 3.80). For this group, the median number of drinks consumed per week was 1.14 (95% CL: 1.11, 1.18) times the median consumed by non-users. Over follow-up, 529 HIV seroconversions occurred. After accounting for several measured confounders using a joint marginal structural Cox proportional model, the hazard ratio (HR) of HIV seroconversion for moderate drinking (1-14 drinks/week) compared to nondrinking was 1.10 (95% CL: 0.78, 1.54) and for heavy drinking (>14 drinks/week) was 1.61 (95% CL: 1.12, 2.29) compared to nondrinking. The HRs for heavy drinking compared to nondrinking for participants with 0-1 or >1 URAI partner were 1.37 (95% CL: 0.88, 2.16) and 1.96 (95% CL: 1.03, 3.72), respectively (robust joint Wald P for interaction = 0.42). These findings suggest first, that illicit drug use may identify MSM whose alcohol puts them at risk for adverse alcohol-related health outcomes, and second, that alcohol interventions to reduce heavy drinking among MSM should be integrated into existing HIV prevention activities.

Date of publication