Alcohol use and viral suppression in HIV-positive Kenyan female sex workers on antiretroviral therapy

Background Excessive alcohol intake has been associated with poor adherence to antiretroviral therapy (ART). The impact of alcohol on viral suppression is particularly important among groups at high risk of HIV transmission, such as female sex workers (FSWs). Few studies have directly evaluated the association between alcohol use and HIV viral load. We hypothesized that hazardous or harmful alcohol use is associated with detectable plasma viral load among HIV-positive FSWs. Methods A prospective cohort study was conducted among HIV-positive FSWs in Mombasa, Kenya. Hazardous or harmful alcohol use was assessed yearly and defined as an Alcohol Use Disorders Identification Test (AUDIT) score >= 7. Detectable viral load was assessed every six months and defined as >= 180 c/mL. Adherence measures were collected monthly and included late ART refill (>48 hours) and self-reported adherence, using both a validated self-rating scale of ability to take medication and visual analog scale (VAS) of ART use in the last month. Generalized estimating equations were used to estimate adjusted relative risks (aRR) and 95% confidence intervals (CI). Results This analysis included 366 participants followed monthly between October 2012 and March 2018. At baseline, AUDIT scores indicated hazardous alcohol use (AUDIT 7-15) in 14.3%, harmful alcohol use (AUDIT 16-19) in 1.4%, and alcohol dependency (AUDIT >= 20) in 1.4% of participants. After adjusting for potential confounders, a combined exposure including hazardous, harmful, and dependent alcohol use was not associated with detectable viral load (aRR 1.10, 95%CI 0.63-1.92) or late ART refill (aRR 1.13, 95%CI 0.82-1.56), but was associated with lower self-rated ability to take medication (aRR 2.38, 95%CI 1.42-3.99) and a lower rate of self-reported perfect ART adherence by VAS (aRR 2.62, 95%CI 1.84-3.71). Conclusions In this FSW cohort, while participants reporting hazardous, harmful, or dependent alcohol use were not more likely to have a detectable viral load, they were more likely to report lower ART adherence. These results suggest that interventions targeting alcohol use among this population of FSWs may not have a large impact on viral suppression.