Background: We report preliminary findings on the effectiveness of the Asibonisane Community Responses (CR) program, which aims to prevent HIV transmission, reduce intimate partner violence, and increase HIV service utilization through community outreach efforts in informal settlements, where barriers to reaching the 90-90-90 goals may differ from the rest of the country.
Methods: A stepped-wedge randomized evaluation was conducted in 18 informal settlements in KwaZulu-Natal where the CR program was being implemented. Program rollout was randomized across evaluation sites allowing for comparisons between control and intervention areas. Using a two-stage random sampling approach, we interviewed 768 women ages 18-24 and 758 men ages 18-35 in early 2017. We conducted two cohort follow-up rounds at seven-month intervals; a final round is underway. We report results from round 3 of data collection; results will integrate data from the final round, and account for baseline levels of testing and treatment.
Results: Respondents reported high levels of economic insecurity and mobility: at baseline, 40% of women and 27% of men had moved within the last two years. In control communities, women (66%) were more likely than men (66% vs. 45%) to report being tested for HIV, and being on ARV treatment (97% vs. 73%) in the last six months, among those who reported being HIV positive. At follow-up, men in intervention communities were significantly more likely than men in control communities to report having been tested for HIV (55% vs. 45%, p=0.04) and to report being on ARV treatment (95% vs. 73%, p=0.02) in the last six months. We found no effects on testing or treatment for women.
Conclusions: Levels of HIV testing indicate that both men and women fall considerably short of the 90% target of knowing their status; treatment levels for men also fall short of the 90% target. Preliminary findings from the evaluation indicate promising results in both testing and treatment for men, an often difficult to reach population. We find no significant program effects for women. This study adds to the growing evidence base on approaches to achieving the 90:90:90 targets, with a focus on a highly vulnerable population.