Sex ratios of HIV prevalence: Uniting demographic, epidemiological and sociological perspectives to understand gender imbalances in generalized HIV/AIDS epidemics

Analysis of national prevalence surveys from sub-Saharan Africa reveals that while women tend to have higher HIV prevalence than men, prevalence sex ratios vary widely across countries, reflecting highly disparate burdens of disease by gender in some populations. However, small sample sizes and rates of survey non-response ranging between 3 and 30 percent restrict the capacities of these surveys to detect significant differences in gender disparities measured between countries.;To identify potential sources of gender disparities in the distribution of HIV, a new conceptual framework is proposed. The framework incorporates the proximate determinants terminology to categorize risk factors for HIV infection broadly and a multi-level approach to recognize those risk factors that operate on the level of the individual, sexual partnership, or population.;The conceptual framework guides an analysis of determinants of the gender distribution of HIV risk within two populations: Tanzania, where women were about 40 per cent more likely than men to test positive for HIV, and Kenya, where women were more than three times more likely than their male peers to test HIV-positive. Results reveal that very different epidemic processes are responsible for producing the observed gender imbalances in HIV prevalence in the two countries. In Tanzania, patterns of sexual mixing placed women at greater risk of encountering an HIV-positive partner than men, and this risk factor explained the entire gender-gap in the likelihood of testing positive for HIV. In Kenya, men's lesser biological susceptibility to HIV infection, enhanced by high prevalence of male circumcision, additionally contributes to the highly imbalanced risk of HIV infection by gender.;That no one risk factor or set of risk factors explains the observed variation in the gender distribution of HIV prevalence across populations underscores the need for continued high-quality national prevalence surveys to produce precise, unbiased estimates of HIV prevalence in both women and men. Only with reliable information on the gender distribution of HIV risk may adequate gender-sensitive responses to the HIV/AIDS epidemic be prioritized effectively and targeted efficiently.