Incidence of HIV infection in rural KwaZulu-Natal in the context of the epidemiology and impact of HIV/AIDS in South Africa.

Thesis (M.Med.)-University of KwaZulu-Natal, Durban, 2007.South Africa has had one of the fastest growing HIV epidemics in the world and almost 30% of women attending public antenatal clinics (ANC) are currently infected with the virus. But as the epidemic is starting to level off and antiretroviral therapy (ART) is becoming increasingly available, few methods exist to determine the impact of ART or other interventions on the epidemic in South Africa. This thesis explores the epidemiology and dynamics of HIV infection and investigates the potential impact of ART. Methods Total and age-specific prevalence data are analysed in time and space and are used to investigate patterns of infection in men and women, urban and rural, and low and high risk populations. Dynamical models are developed to estimate incidence from age-specific prevalence and trends over time and are compared to laboratory-based estimates of recent HIV sero-conversion. Incidence is estimated in different populations in South Africa. A dynamical model is developed to estimate the impact of ART on the future course of the HIV epidemic. Results HIV prevalence varies geographically and by age, sex and race. The average female-tomale HIV prevalence ratio is 1.7 and prevalence peaks at an older age among men than women. The age at which prevalence peaks among women has increased from 23.0 to 26.5 years between 1995 and 2002. Four patterns of infection are identified: among pregnant women attending ANCs, among men and women in the general population, and among migrant workers. HIV incidence among ANC attendees peaked in the mid to late 1990s (at 6.6% per year nationally) with variation between provinces. Current estimates of HIV prevalence and incidence among the general population in South Africa (aged 15-49 year) are 18.8% and 2.4% per year, respectively. Age-specific incidence estimates from dynamical models and laboratory methods are in good agreement provided the window period for the laboratory method is increased. Over the next ten years the provision of ART could avert 1 to 1.5 million deaths depending on whether it is provided when the CD4 cell count falls to 200 or 350 cells/ul. By 2015 about 1.1 million people will be receiving ART but this will have little impact on the incidence of HIV and scaling up of prevention efforts remains urgent. Conclusions The thesis explores some of the determinants and patterns of HIV prevalence and incidence in South Africa in order to find better ways to manage the epidemic of HIV, monitor changes and evaluate progress in control efforts. In order to fight the epidemic we need to mobilize the best possible science in support of those people and communities affected by the epidemic