Abstract Body

In settings like Botswana with high HIV prevalence and treatment coverage, it is unknown whether uptake of HIV-prevention and treatment interventions can be further increased.

 

The Botswana Combination Prevention Project, a pair-matched cluster-randomized trial, compared uptake of an intervention package of intensive HIV testing/counseling(HTC) campaigns, linkage to care, expanded antiretroviral treatment(ART), and male circumcision(MC) versus standard-of-care in Botswana from 2013-2018. In mid-2016 universal ART became standard-of-care. We longitudinally followed residents aged 16-64 years of a random ~20% sample of households in 30 communities (15 intervention, 15 standard-of-care) for ~30 months to assess uptake of ART, viral suppression, and MC. HIV testing was conducted annually. To assess change in HTC coverage (documented HIV-negative test within 12 months or knowledge of HIV-positive status) by arm, we recruited an additional sample of residents not enrolled in the longitudinal cohort from six communities/three pairs at study end. For HTC, ART, and viral suppression, we estimated risk ratios(RR) and 95% confidence intervals(CI) (accounting for clustering) using log-linear Poisson regression adjusted for potential baseline coverage imbalances, stratified by time and pair. MC uptake among HIV-uninfected uncircumcised men aged 16-49 years was evaluated using pair-stratified interval-censored Cox proportional hazards.

 

We enrolled 8,974 HIV-negative and 3,596 HIV-positive residents in the longitudinal cohort. An additional 11,767 residents were assessed for HTC uptake at study end. After accounting for baseline differences, HTC coverage was significantly higher in the intervention arm at study end (P<0.0001; Fig.1A). ART coverage and viral suppression increased in both arms, with greater increases in the intervention arm (ART P<0.0001; viral suppression P=0.004; Fig.1B-C). At study end, 98% (95%CI: 93%-100%) of HIV-positive cohort participants in intervention communities were on ART; 96% (95%CI: 92%-100%) were virally suppressed. A small number (348) of 1,873 HIV-negative uncircumcised men reported becoming circumcised, with higher uptake in the intervention arm (P<0.0001).

 

Population levels of HTC, ART, viral suppression, and MC increased in both arms over time, with significantly greater increases in the intervention arm. Remarkably, at study end, nearly all HIV-positive cohort participants in intervention communities were on ART and virally suppressed.