TRANSACTIONAL SEX AND HIV AMONG MEN IN RAKAI, UGANDA: A MIXED METHODS STUDY

Background: Addressing HIV risk associated with transactional sex (relationships motivated by implicit material exchange for sex) is important in curbing sub-Saharan Africa’s generalized HIV epidemic. Research on the relationship between transactional sex and HIV among men is limited. This mixed methods study aimed to understand characteristics of men who engage in transactional sex, their risk for HIV acquisition and onward transmission of HIV, and how to best measure transactional sex. Methods: We used data from the Rakai Community Cohort Study (RCCS), an open, population-based cohort study in Rakai District, Uganda to examine prevalence of HIV and viremia among 7,678 men by self-reported transactional sex and HIV incidence among 7,762 men who were HIV-negative at baseline by self-reported transactional sex over four person-intervals. In-depth interviews (IDI) were conducted with men (N=26) who did (n=21) and did not (n=5) report transactional sex in the RCCS. Cognitive interviewing IDIs were used to examine men’s understanding of three transactional sex measures and explored changes in transactional sex relationships during the COVID-19 pandemic. Data were analyzed thematically. Results: Prevalence of transactional sex among men was 9%. Overall, transactional sex was not associated with prevalence of HIV (adjusted prevalence risk ratios (adjPRR):1.04;95%CI:0.87,1.18) or viremia (adjPRR:1.05; 95%CI:0.81,1.37). However, prevalence of viremia was higher among boda boda drivers who engaged versus did not engage in transactional sex (adjPRR:2.46;95%CI:1.11,5.47). We observed a positive, but non-statistically significant effect of transactional sex on HIV incidence (adjIRR:1.50;95% CI:0.86,2.61). Cognitive interviews revealed that the measure used in quantitative analyses may have misclassified some transactional relationships and underestimated transactional sex prevalence. An improved measure developed by Wamoyi and colleagues performed well in identifying transactional sex. Economic insecurity and restricted mobility during Uganda’s COVID-19 lockdowns led to the dissolution of many transactional sex relationships, illustrating a strong link between material provision and the formation and continuation of transactional sex relationships. Conclusion: This mixed methods study provides insight into multiple facets of the relationship between transactional sex and HIV outcomes. Our findings highlight the importance of using improved transactional sex measures and the need for a more nuanced conceptualization of men’s participation in transactional sex