HIV Prevalence in some African Territories: Socio-Economic Drivers

Abstract

In 2020, 35% of all HIV-positive people in the world lived in Eastern and Southern Africa. This work aims at assessing the relationship between socio-economic drivers and HIV prevalence at the sub-national level in these countries. The data used are drawn from the Demographic and Health Survey, in which a subset of respondents is tested for HIV. Using a fractional logistic regression model on clusters of individuals, middle-to-high wealth is positively associated with higher HIV prevalence, while a higher average number of children acts as a protective factor; moreover, higher proportions of people who have never been in sexual relationships lower their cluster’s HIV prevalence but HIV-positive people are more likely to use condoms at last intercourse.