Hard-to-Reach Populations: Implications for Ending the AIDS Epidemic

乌干达的拉凯地区是东非艾滋病毒流行的最初中心，并继续成为一个地区

The Rakai region in Uganda was the initial epicenter of the HIV epidemic in East Africa and continues to be a high burden area with an HIV prevalence of ~13%. Through the open, population-based Rakai Community Cohort Study (RCCS), we reported that combination HIV prevention (CHP) decreased population-level HIV incidence in Rakai by 42% from 1.17/100 person-years (pys) prior to CHP scale-up to 0.66/100 pys by 2016 (Grabowski et al. NEJM 2017). Implications and limitations from this study raise two issues of global importance. First, mobile persons, typically away for work or school, and, rarely, refusers are a “hard-to-reach” population that is difficult to survey, reducing RCCS participation rates to ~62%. These populations may likewise be hard-to-reach for engagement in HIV services. Ongoing cluster-randomized HIV prevention trials and population-based HIV impact assessments have similar challenges of potential bias due to missing these hard-to-reach populations. Second, despite reaching 59% male circumcision coverage and UNAIDS 90-90-90 goals with 75% viral suppression of all HIV-positive participants in RCCS, HIV incidence reductions were moderate and remained well above the estimated rate needed for HIV elimination (~0.1/100py). To address ongoing HIV transmission in the Rakai region, the PEPFAR program in which RCCS is nested recently began implementing additional CHP interventions: (i) Pre-Exposure Prophylaxis (PrEP); (ii) assisted Partner Notification; and (iii) Same-day antiretroviral therapy (ART). This environment provides a unique opportunity to address the following important questions: (1) To what extent do hard-to-reach populations bias HIV coverage and incidence estimates? (2) Why do some individuals continue to acquire HIV and from whom? (3) Given hard-to-reach populations, can state-of-the-art CHP in a programmatic setting reduce HIV incidence to the levels needed for HIV elimination? Our setting and research infrastructure strongly position us to answer these highly significant questions and inform current and future HIV prevention trials, evaluations, and programs. We thus propose a novel study with the following Aims. Aim 1-We will first determine CHP coverage and HIV incidence among hard-to-reach persons using enhanced surveillance techniques. Aim 2-We will then characterize ongoing sources of incident HIV infection through partner tracing, viral phylogenetics, and sexual network analyses. Aim 3-Finally, we will determine if state-of-the-art CHP can engage hard-to-reach populations and reduce population-level HIV incidence to a level sufficient for HIV elimination by 2030. To our knowledge, no prior HIV population-based studies have empirically determined the potential effects of participation bias on HIV epidemiology and incidence due to non-inclusion of hard-to-reach populations. This study will uniquely address questions on hard-to-reach populations which are critical to understanding the true state of the epidemic, interpreting HIV prevention trials and cross-sectional studies, and informing prospects and pathways to ending the African HIV epidemic.