Home-based testing and referral to clinics for biomedical prevention reduced HIV incidence in two districts in rural Kenya and Uganda by 70% after two years of the intervention, the Conference on Retroviruses and Opportunistic Infections (CROI 2026) in Denver heard last week.
The SEARCH Community Precision Health study randomised 16 communities in Kenya and Uganda to continue quarterly home visits by a community health worker, plus clinic-based HIV testing, PrEP, and HIV treatment, or to an enhanced intervention.
The enhanced ‘Community Precision Health’ intervention consisted of quarterly home visits and home-based HIV testing plus clinic referrals for HIV prevention and HIV treatment. All participants received HIV education to help them assess their HIV prevention needs. If they anticipated the potential for HIV exposure in the near future, they received a clinic referral. HIV-negative participants referred to a clinic were offered dynamic choice prevention, which allows users to switch between different options according to their needs.
Community health workers in the intervention arm used an app, the Community Health Toolkit, to record the number of visits and household members, HIV test results and make referrals to a clinic. Clinic providers used the app to flag up when participants did not attend a clinic appointment, so that a community health worker could follow up.
Dynamic choice prevention offered a range of options to participants, including oral PrEP, injectable cabotegravir PrEP, the dapivirine ring, or PEP in Pocket (PEP supplied in case of future need), all of which were available through the public health system. Participants could move between these options depending on their circumstances during the study.
The trial followed 42,366 participants in the control arm and 42,234 participants in the intervention arm and lasted two years. More than half of the participants were female (57% in the control arm and 55% in the intervention arm), the median age was 31, just over half were married, and 31% in the control arm and 35% in the intervention arm were never married.
The primary study outcome was the reduction in HIV incidence. HIV incidence was assessed using a LAg avidity recency assay, which detects HIV infection within the past four to five months.
HIV status was ascertained in 95% of participants in the control arm and 96% of participants in the intervention arm. Testing for recent HIV acquisition was carried out for 99.9% of participants.
At the end of year 2, HIV incidence was 70% lower in the intervention arm than the control arm (0.06% vs 0.19% had acquired HIV). There were seven recent HIV infections in the intervention arm and 22 in the control arm.
Subgroup analysis by country, age and sex showed that incidence was significantly lower in the intervention arm for all variables except male sex. HIV incidence among men was substantially lower than among women in the control and intervention groups, resulting in no statistically significant difference between arms in HIV incidence among men.
A secondary analysis showed that four times as many participants without HIV in the intervention arm used a biomedical prevention product compared with the control arm (1.67% vs 0.41%). A similar difference was seen in all subgroups, although uptake was lower in Uganda compared to Kenya and in the 15-24 age group compared to older participants.
Among people with HIV, there were no significant differences between control and intervention groups in knowledge of HIV status, the proportion of HIV-diagnosed people on antiretroviral treatment, or in viral suppression among those on treatment. A similar proportion of the populations living with HIV in control and intervention communities had viral loads suppressed below 400 copies (81% vs 82%).
Presenting the study results, Professor Gabriel Chamie of the University of California, San Francisco, said that increased HIV prevention coverage was a primary driver of the reduced HIV incidence. The study design overcame multiple barriers to the provision of PrEP and PEP in rural areas by using community assessment to develop awareness and community health workers to deliver oral PrEP and PEP supplies.
More than 95% of community health workers involved in the study reported that the app was easy to use, and no providers reported difficulty in communicating with community health workers through the platform or co-ordinating home deliveries of PrEP or PEP. Ninety per cent of participants were satisfied with confidentiality and community health workers’ support for prevention product delivery during the study.
Asked whether the intervention was scalable or transferable to other settings, Chamie explained that the Community Precision Health intervention had been designed in consultation with national ministries of health in Kenya and Uganda to leverage existing resources, including clinics, prevention products and community health workers. The study took place in rural areas where the rate of viral suppression was high. It’s uncertain what impact the intervention would have in urban areas, he said.
In this study, participants had access to dynamic choice prevention, which enabled them to switch between options according to their needs. “What’s critical to dynamic choice prevention is having access [in the public health system] to a range of prevention options – choice of options is key,” Chamie said.
References
Chamie G et al. The impact of SEARCH Community Precision Health on HIV incidence in rural Kenya and Uganda. Conference on Retroviruses and Opportunistic Infections, Denver, abstract 163, 2026.
