The underlying determinants of South Africa's generalized HIV epidemic
Thesis summary
This thesis argues that the most parsimonious explanation for why only a handful of countries have generalized HIV epidemics is due to the increased connectivity of their sexual networks. Sexual partner concurrency, or having two partners at a time, is argued to be a critical factor in promoting this increased network connectivity.
Looking at the global patterning of HIV and HSV-2, this thesis finds that there is a close correlation between concurrency rates and prevalence of these infections. Furthermore we find that this ecological correlation extends to racial and ethnic groups within South Africa. Detailed analyses of sexual behaviours from high and low HIV prevalence groups within South Africa suggest that concurrency is the behaviour that covaries most closely with HIV prevalence. At an individual level we find partner concurrency to be associated with prevalent HSV-2 infection in women and in a different population-based survey with symptoms of an STI in men and women.
Understanding the reasons for the high concurrency rates in certain communities is clearly of crucial importance. In a number of studies looking into this issue we seek to understand if concurrency is driven predominantly by cultural or socioeconomic factors. A global ecological study finds no evidence of a relationship between concurrency prevalence and various markers of economic deprivation/inequality and gender-inequality. Our analysis of the correlates of concurrency in both Cape Town and Carltonville found no evidence of a relationship between concurrency and economic indicators. Concurrency rates were however consistently higher in blacks, compared to other racial groups, from their first sex partner and this pattern continued through following relationships. This fits with a range of qualitative data which indicates that it is culturally acceptable, particularly for men, to have main and side partners throughout Southern and Eastern Africa. If South Africa is to significantly reduce its HIV incidence, it will need to decrease the interconnectedness of its sexual networks and to do this will require norm change.