Young women in South Africa (SA) are experiencing the highest HIV prevalence (16-30%) and incidence (3 to 5%) rates in the world, with very high male to female HIV transmission rates (2 to 6%). Although sexual behaviors have been the focus of HIV prevention research, it is clear that these high rates cannot be explained by risk behaviors alone. Recent observational studies have identified that hormonal contraception (HC), and in particular injectable contraceptive (ICs) such as DMPA may increase the risk of HIV transmission in women and this risk is significantly higher in younger women (<20 years). Herpes Simplex Virus-2 (HSV-2), has been shown to be associated with a significant increase in HIV infection in women. Given that ICs (DMPA and NET-EN) are the most commonly used contraception among young women in SA, it is critical we have more empirical data to understand the mechanism by which ICs modify the risk of HIV and HSV-2. We propose to conduct a prospective cohort study among 600 sexually active women aged 16 to 24 years based on current use of ICs or non-HCs (condoms, IUD) so as to assess:
Aim 1 : the impact of IC on incidence of HIV and HSV-2 and point prevalence of HPV and common bacterial STIs. Women will be tested for HIV every 3 months and samples for HSV-2, HPV and bacterial STIs will be collected quarterly, tested annually and all positive samples will be back-tested to ascertain the approximate time of infection.
Aim 2 : Through the use of computer assisted self-interviews, we will collect sensitive sexual behavior information from all women (N=600) during quarterly visits and assess impact of IC and associated risk behaviors.
Aim 3 : A subset of women (N=30), will be seen at weekly intervals for one menstrual cycle to assess changes in genital tract micro flora, and mucosal immunity associated with IC use as well as exposure to infections, controlling for changes in endogenous hormones. Discrete Kaplan-Meier curves and log-rank tests will be used to estimate the incidence rates of HIV and HSV-2, stratified by contraceptive use. Cox regression models will be used to identify effects for fixed covariates at baseline (IC versus non-HC), and time-independent (demographics) or time-dependent (contraceptive exposure, changes in genital structure, immune cells, sexual behavior) covariates. Survival analyses will be used to assess the impact of IC on HIV progression (time to death, CD4 decline, etc.). Marginal structural modeling will be used for all recurrent and multiple events to account for the time-dependent confounding in our study. In South Africa, young sexually active women are at high risk for HIV and HSV-2 infection, as well as other common STIs, and a majority use ICs. These data will provide guidance on the safety and efficacy of ICs for women at risk for STIs and HIV, and assist in the development of new multi-purpose reproductive health technologies that combine reproductive health, HIV and STI prevention.

Public Health Relevance

Sexually active young women in South Africa are in need of interventions that prevent unwanted pregnancy, STI and HIV due to high rates of HIV and unwanted pregnancies in this population. This study will be among the first to examine the correlation between IC (DMPA or NET-EN) and both HIV and HSV-2 incidence, as well as the impact on other common STIs and sexual risk behaviors in this vulnerable, young population.

National Institute of Health (NIH)
Eunice Kennedy Shriver National Institute of Child Health & Human Development (NICHD)
Research Project (R01)
Project #
Application #
Study Section
Special Emphasis Panel (ZHD1-DSR-A (55))
Program Officer
Kaufman, Steven
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Johns Hopkins University
Schools of Public Health
United States
Zip Code