Abstract Body

The effect of using HIV prevention services on using HIV treatment services has not been well documented in southern Africa. Using nationally representative data from household surveys conducted in Eswatini, Lesotho, Malawi, Zambia, and Zimbabwe (2015–2017), we examined the correlation of self-reported voluntary medical male circumcision (VMMC) and condom use among HIV-negative adults with use of treatment services by people living with HIV (PLHIV), represented by the UNAIDS 90-90-90 targets, at the community level.

Among HIV-negative adults in the surveys, we estimated the prevalence of self-reported VMMC status and condom use (during last sexual act in the prior 12 months) at the smallest geographic sampling unit (enumeration area [EA]). We used multilevel mixed-effects logistic regression, adjusted for demographic and risk behavior variables at individual level to estimate the correlation between VMMC and condom use at the EA level with the likelihood of PLHIV being aware of their status, currently on ART, or virologically suppressed (VS).

Among 10,861 PLHIV aged 15–64 years (62% women) residing in 1,734 EAs across surveys, 76% had a previous HIV diagnosis, 68% were receiving ART, and 60% were VS. Median EA-level prevalence of HIV infection, VMMC, and condom use was 16% (interquartile range [IQR], 10%–24%), 16% (IQR, 6%–32%), and 72% (IQR, 55%–88%), respectively.

On multilevel analysis, the odds of knowing HIV-positive status, receiving ART, or being VS were significantly higher for PLHIV residing in an EA where ≥75% of the adults reported condom use (adjusted odds ratio [AOR], 1.3 [95% confidence interval (CI), 1.2–1.5]; 1.3 [95% CI, 1.1–1.4]; 1.2 [95% CI, 1.1–1.3], respectively). The odds of knowing HIV-positive status, receiving ART, or being VS were significantly higher for PLHIV residing in an EA where ≥15% of men reported VMMC (AOR, 1.2 [95% CI, 1.1–1.3]; 1.1 [95% CI, 1.0–1.3]; and 1.1 [95% CI, 1.0–1.2], respectively).

In these five countries, community utilization of prevention services was positively correlated with the individual use of treatment services, suggesting that combination prevention services can play a synergistic role in epidemic control.