Poverty or inequality as an underlying cause of HIV in Africa? The HIV-poverty thesis re-examined

Background: Explanations for Africa's disproportionate burden of HIV/AIDS have been linked to the continent's endemic poverty. However, contrary to theories of poverty as the underlying cause of HIV in Africa, an increasing body of evidence at the national and individual levels indicates that wealthier countries in Africa, and wealthier individuals within countries, are at heightened risk for HIV. This study will test the hypothesis that rather than HIV being a disease associated with poverty, HIV infection increases under conditions of rising socio-economic inequality.

Methods: Examining demographic and health survey data from six African countries (Zimbabwe, Lesotho, Tanzania, Kenya, Ivory Coast and Ghana), this study utilizes a multi-level model of the relationship between HIV infection and economic inequality. Bivariate relationships between the outcome measure (HIV serostatus) and economic inequality at the regional level were assessed within each of the six countries. All multivariate models were run as a two-level, hierarchical random coefficient models in Stata adjusted for clustering at the regional level.

Results: Within each country, regional HIV prevalence was positively correlated with regional economic inequality reaching statistical significance in Ghana and Lesotho (p<0.1). Results from the two-level random intercept model demonstrated that individual wealth quintile and regional gini coefficient are positive and significant (p<0.01). As hypothesized, taking advantage of the pooling of data reveals that wealthier individuals, rather than poorer individuals as commonly thought, are at higher risk for HIV infection and that the probability of infection increases with rising regional (within-country) inequality.