Assessing the Prevention Impact of HIV Counseling and Testing in the South African Context

Rosenberg, Nora E.

Download PDF

Request Version for Screen Reader

Last Modified

March 21, 2019

Creator

Rosenberg, Nora E.
- Affiliation: Gillings School of Global Public Health, Department of Epidemiology

Abstract

HIV counseling and testing (HCT) is rapidly being brought to scale in South Africa, yet its impact on sexual behavior and HIV incidence is not well understood. The purpose of Aim 1 was to assess whether HIV-infected persons in HIV-discordant couples were less likely to engage in unprotected sex after HCT. Self-reported behavioral data from 500 HIV-infected South Africans enrolled in Partners in Prevention HSV/HIV Transmission Study was used. The sexual behavior of two groups was compared: HIV-infected persons who had just learned their HIV status from HCT ( 30 days before baseline). Among those tested 30 days before baseline the predicted probability of unprotected sex was lower at baseline (0.26), declined futher by month one (0.14) and remained lower at month twelve (0.19). These findings suggest HCT lead to substantial reductions in unprotected sex. In Aim 2, the effect of HCT on future HIV acquisition among HIV-uninfected youth was assessed. A retrospective cohort study of 3959 HIV-uninfected youth 15-24 years-old was conducted using a demographic and health surveillance from KwaZulu-Natal, South Africa (2006-2011). Young persons who reported knowing their HIV status from HCT were compared to those who reported not knowing their HIV status from HCT for time to HIV seroconversion using marginal structural Cox proportional hazards models. In these models, after weighting for confounding and censoring, HCT was protective [HR: 0.59, 95% CI: 0.44, 0.78], underscoring the importance of HCT for youth. In Aim 3 a framework was developed to describe nine awareness patterns within HIV-discordant dyads considering both HCT and HIV disclosure together. It was hypothesized that different types of HCT lead to different awareness patterns and that certain patterns are more strongly associated with uptake of and adherence to behavioral and biomedical HIV prevention strategies. Better understandings of these associations may inform how to optimize HCT delivery for prevention.

Date of publication