Graham, Susan MMartinez, Miguel Angel2023-08-142023-08-142023-08-142023Martinez_washington_0250O_25860.pdfhttp://hdl.handle.net/1773/50165Thesis (Master's)--University of Washington, 2023Background: Gay, bisexual, and other men who have sex with men (GBMSM) in sub-SaharanAfrica experience disproportionate rates of stigma and its social repercussions, including high rates of alcohol use. The burden of HIV in sub-Saharan Africa also disproportionally affects GBMSM populations. This work aimed to assess the variability of drinking patterns and determine associations between unhealthy alcohol use and HIV risk behaviors and adherence to HIV prevention medication among GBMSM in Kenya. Methods: A secondary analysis of the Anza Mapema Mbili study, which followed HIV negativemen for one year, was conducted to assess the prevalence of unhealthy drinking, as measured by phosphatidyl ethanol (PEth), and determine its association with (1) several HIV risk factors (condomless anal intercourse, multiple sex partners, and forced sexual violence) and (2) adherence to pre-exposure prophylaxis (PrEP). Descriptive statistics were used to identify differences between participants with and without unhealthy drinking at baseline. McNemar’s test was used to assess whether the prevalence of unhealthy drinking changed over study visits. Generalized estimating equations with logit links and exchangeable correlation structures were used to determine associations between unhealthy drinking and study outcomes after adjustment for potential confounders. Results: The prevalence of unhealthy drinking as determined by the first collected PEthmeasure was 56% (95% CI: 47% – 64%). The prevalence of unhealthy drinking did not change significantly over the next study visit where PEth was collected. Participants who had unhealthy drinking were, on average, 3.4 years older as compared to those without unhealthy drinking. Further, a larger proportion of those in the unhealthy drinking category were classified as having “Hazardous/Harmful Drinking” via the AUDIT test as compared to those without unhealthy drinking (62.0% vs. 27.4%, respectively). A borderline significant association was found between unhealthy alcohol use (<= 200 ng/ml of PEth) and lower odds of having multiple sex partners, but there were no statistically significant associations with condomless anal intercourse, forced sex, or PrEP adherence. Conclusion: Unhealthy alcohol use, as measured by PEth, was found in the majority of GBMSMstudy participants and did not change over follow-up. Unhealthy alcohol use was associated with older age and AUDIT score > 8 but was not strongly associated with HIV risk behaviors or PrEP adherence. Further work is needed to understand better the role of alcohol use, drinking patterns, and social harms in HIV risk perception and engagement in preventative measures among GBMSM.application/pdfen-USnoneAlcoholGBMSMHIVKenyaPrEPEpidemiologyHealth servicesThesis