Population-level coverage and correlates of enrollment in community-based antiretroviral therapy (ART) refill models following differentiated service delivery scale-up in Africa

A multi-country study

Bibliographic Data

ID19444173
AuthorsLipin Lukose (0000-0001-8409-3491, Johns Hopkins University), Robert Kairania (0000-0002-9015-7209, Rakai Health Sciences Program), Fred Nalugoda (0000-0001-6119-9293, Rakai Health Sciences Program), Gertrude Nakigozi (0000-0001-6193-2790, Rakai Health Sciences Program), Godfrey Kigozi (0009-0008-9142-6870, Rakai Health Sciences Program), Robert Ssekubugu (0000-0003-1245-4774, Rakai Health Sciences Program), Steven J Reynolds (0000-0002-5403-2759, Rakai Health Sciences Program), Larry W Chang (0000-0001-7985-4567, Johns Hopkins University), Caitlin E Kennedy (0000-0001-6820-063X, Johns Hopkins University), M Kate Grabowski (0000-0002-4451-3436, Johns Hopkins University), M Grabowski (0000-0002-8072-4919, Johns Hopkins University), Joseph G Rosen (0000-0003-4991-4033, Johns Hopkins University, corresponding author)
Year2025
Pages1-16
Publication date2025-12-16
Peer ReviewedYes
Open AccessNo
TypeARTICLE
VenueAIDS Care (JOURNAL)
Journal identifiersISSN: 0954-0121 • E-ISSN: 1360-0451
PublisherInforma UK Limited (PUBLISHER • GB)
DOI10.1080/09540121.2025.2604109
PMID41399124
OpenAlexW4417359636
LanguageEN
References cited48

Community-based antiretroviral therapy (ART) distribution is a promising strategy to increase HIV treatment accessibility in Africa, yet population-level coverage and correlates of enrollment are understudied. We pooled data from seven Population-based HIV Impact Assessment surveys (2020–2023) and used survey-weighted multivariable logistic regression with jackknife variance estimation to identify socio-demographic and clinical correlates of enrollment in community, relative to facility-based, ART distribution models. Across countries, 11,629 virally suppressed ( 12 months (mean age: 42.1 years, 67.0% female) self-reported their assigned ART refill model, 2.9% of whom received ART from a community-based pickup point (range: 0.5% in Mozambique to 7.6% in Zimbabwe). Correlates of enrollment in a community-based refill model included rural residence (adjusted odds ratio [adjOR] = 2.24, 95% confidence interval [CI]:1.20–4.21), past-year mobility (adjOR=1.56, 95%CI: 1.05–2.31), ART dispensing intervals of >6 months (adjOR=2.36, 95%CI: 1.11–5.02), and transportation-related barriers to facility-based HIV care (adjOR=1.43, 95%CI: 1.01–2.03). Among virally suppressed males, superior virologic control at the <200 copies/ml (adjOR=7.12, 95%CI: 1.59–31.97) and <400 copies/ml (adjOR=6.08, 95%CI: 1.34–27.57) levels was observed among those enrolled in a community-based, relative to facility-based, ART distribution model. Findings affirm the potential of community-based refill models to reach populations susceptible to HIV care attrition, but coverage estimates remain low, indicating implementation barriers

Antiretroviral therapy · Confidence interval · Health care · Logistic regression · Odds ratio · Relative risk · Residence · Service delivery framework · Adolescent Sexual and Reproductive Health · HIV/AIDS drug development and treatment · HIV/AIDS Research and Interventions