HIV communication programs, condom use at sexual debut, and HIV infections averted in South Africa, 2005

Since the 1990s, communication programs in South Africa have promoted HIV prevention, especially condom use. Although HIV prevalence stabilized after 2000, surveys have not clarified how prevention behavior contributed to that change. We present a secondary statistical analysis of the 2005 national South African Human Sciences Research Council survey that reveals how condom use at sexual debut-which normally occurs before infection from unprotected sex-may have contributed to the reduction in HIV infection.

Threats from measured and unobserved confounding variables are addressed through the use of several complementary statistical methods including structural equation modeling with multivariate probit regression and propensity score matching. The latter is used to create statistically comparable groups of those who used and did not use condoms at sexual debut to estimate the number of HIV infections averted attributed to this behavior.

After controlling for 17 socioeconomic variables, awareness of HIV communication programs had a positive, dose-response relationship with self-reported condom use at sexual debut. Those who used condoms at sexual debut were 1.38 times more likely to have remained uninfected. HIV-negative status was 3.6 percentage points higher among those who used condoms at sexual debut, corresponding to an estimated 139,835 infections averted among sexually active adults by 2005. Never using injectable drugs and having 1 trusted partner also predicted HIV-negative status.

We found a relationship between HIV communication awareness and condom use at sexual debut. Condom use at sexual debut, in turn, was associated with lower subsequent HIV risk.