Substance Use, Sexual Networks, and HIV risk among men who have sex with men and transgender women in Peru

BACKGROUND

In Peru, the HIV epidemic is concentrated among men who have sex with men (MSM) and transgender women (TGW) in whom HIV incidence rates are as high as 4.2 per 100 person-years with HIV prevalence reported to be as high as 22% in MSM and up to 30% in TGW. This dissertation seeks to add to the knowledge of the structure of sexual networks, namely the level and predictors of sexual concurrency among MSM and TGW in Peru (Aim 1), and the understanding of risk factors for HIV acquisition in MSM and TGW with high levels of substance use in Lima (Aim 2).

METHODS

Data are from the 2011 Peruvian Biobehavioral Surveillance survey (Aim 1) and the Sabes cohort study conducted in Lima from 2013-2016 (Aim 2). Data were collected with the computer assisted self-interview (CASI) (Aim 1 & 2); HIV testing was performed with Determine 1/2 rapid antibody tests (Aim 1 & 2), pooled nucleic acid amplification test (NAAT), and Western Blot to determine Fiebig Stage at HIV diagnosis. Statistical methods used include Poisson regression and generalized estimating equations (GEE) (Aim 1), and Pearson’s Chi-square, Poisson regression estimated with GEE, and stratified Cox proportional hazards survival analysis with time-varying covariates (Aim 2).

RESULTS

Concurrency is a common practice among MSM and TGW in Peru with a 3-month cumulative prevalence of over 23%. There was evidence of negotiated safety—those with a stable partner were less likely to have condomless anal intercourse (CLAI) with a concurrent non-stable partner. In the Sabes cohort, HIV incidence was 11.7 per 100 person-years of follow-up. Those with alcohol use disorders (AUD) were significantly more likely to attend a venue that served alcohol, binge drink, and use marijuana or amyl nitrites. AUD modified the association between the time-varying behavioral factors and HIV; behavioral risk factors (binge drinking, marijuana use, sex with a casual partner, client, or one-time partner) were most strongly associated with HIV acquisition amongst those with dependent drinking patterns.

CONCLUSION

This study suggests that AUD is linked to HIV risk in two important ways: first, through CLAI with non-stable partners and second, through amplifying the impact of other HIV risk behaviors. These studies suggest that harm reduction approaches, such as negotiated safety with concurrent partners, and treatment of AUDs to decrease alcohol intake could decrease the HIV risk associated with these behaviors.