Correlates of population attributable fractions.

BackgroundHigh incidence rates of HIV, sexually transmitted infections (STIs), and teenage pregnancy are major challenges facing South Africa. The role of socio-economic factors in driving these incidence rates is complex, with high socio-economic status protecting against some risk behaviours (condomless sex, early sexual debut, and casual/transactional sex in females) but increasing other risk behaviours (e.g., male engagement in casual and commercial sex). We aimed to model the effect of socio-economic status, and associated economic strengthening interventions, in South Africa.Methods and findingsWe extended a previously-developed agent-based model of HIV, STIs, and fertility in South Africa to assess effects of education, employment, and per capita household income on sexual behaviours. We estimated these effects from literature and from calibration of the model to African randomized controlled trials of economic strengthening interventions. Population attributable fractions (PAFs) were calculated. We considered three intervention types, all targeting households with log per capita income below the national average: school support to reduce school dropout; vocational training for unemployed adults; and unconditional cash transfers. We estimate that low socio-economic status accounted for 13% of new HIV infections, 7% of incident STIs (gonorrhoea, chlamydia, and trichomoniasis) and 31% of teenage births in South Africa, over 2000−2020. However, because of uncertainties regarding effect sizes, confidence intervals around these PAFs are wide (1,50% for HIV; −1,19% for STIs; and 10,76% for teenage births). Over 2025−2040, none of the interventions are estimated to reduce HIV, STIs, or teenage births significantly, due to limited impact on secondary economic outcomes. The greatest impact would be that of school support on teenage births (a 5% reduction, 95% CI: −1,12%). Key limitations include the assumption of uniform STI treatment access across socio-economic strata, and the exclusion of possible socio-economic effects at a community level.ConclusionsAlthough poverty is likely to be a significant driver of HIV, STIs, and teenage pregnancy in South Africa, precise quantification is challenging. Recently trialled economic strengthening interventions have insufficient impact on socio-economic status to reduce HIV and STIs significantly at a population level.