Abstract
BACKGROUND: To address knowledge gaps, this study examined social determinants, such as education attainment and HIV prevention, among sexually active men (SAM), with a focus on voluntary medical male circumcision (VMMC). METHODS: Two nationally representative surveys, the Eswatini Demographic and Health Survey 2006 and the Eswatini HIV Incidence Measurement Survey 2016, were used to estimate whether or not VMMC at the individual and community levels contributes to HIV disparities to any meaningful extent. Multilevel logistic regression models further explored the educational gradient in HIV infection for 2006-2007 and 2016 with regard to VMMC among SAM, while adjusting for household poverty, sexual practices and individual characteristics. RESULTS: Among SAM with tertiary education, HIV prevalence declined from 25.0% in 2006-2007 to 10.5% in 2016. A 51% decrease in HIV prevalence was found to be associated with an increase in VMMC (adjusted odds ratio 0.49; 95% CI 0.40 to 0.60). Compared with SAM with tertiary education, those who had a lower level of education were more likely to have HIV infection and this education gradient effect had become particularly profound in 2016. CONCLUSIONS: VMMC began to be promoted in 2008 in Eswatini and results suggest its effect, along with the education attainment effect, significantly resulted in a meaningful reduction in HIV prevalence among SAM by 2016.