Optimizing HIV case identification among children and understanding remaining gaps in pediatric HIV testing in Kinshasa, DRC.

BACKGROUND: It is critical to identify children living with HIV and initiate lifesaving antiretroviral treatment (ART) early. The Pediatric Accelerated Case Finding Effort focused on contact elicitation and HIV testing of ART clients' biological children. We describe HIV testing and seropositivity rates following the initiative and gaps along the index testing cascade to inform pediatric HIV case finding optimization. METHODS: This mixed-methods study involved collecting monthly data on index testing outcomes, including elicitation (identifying biological children < 15 years), HIV testing and linkage to treatment from March 2020 to July 2021 in 35 facilities in Kinshasa. Data were summarized and presented for the first month (as a baseline proxy) and the entire study period. Qualitative data were collected from 14 healthcare workers participating in in-depth interviews and 33 community health workers in four focus group discussions. Audio recordings were transcribed and translated from Lingala or French into English and coded using MAXQDA software. Data were thematically analyzed according pediatric case finding barriers and strategies. RESULTS: At baseline (March 2020), among 3337 eligible female index clients, 1634 (49.0%) underwent elicitation to identify children with unknown HIV status. By July 2021, all eligible clients (n = 11,734) had contacts identified. Of the contacts, 9871/11,848 (83.3%) were HIV-tested. Of contacts tested, 662 (6.7%) were diagnosed as HIV-positive, with 535 (80.8%) age 5-14 years; 99.5% initiated treatment. Providers attributed gaps in HIV testing primarily to testing refusals for children due to non-disclosure among parents and logistical or financial obstacles to transportation for tracing. COVID-19 movement restrictions and exposure fears also limited provider interactions for testing. Provider-implemented strategies included transport reimbursement, extensive counseling and alternative approaches to child testing for parents in sero-discordant relationships. CONCLUSIONS: Following intensified efforts around pediatric case finding, we found a high HIV positivity yield of 6.7% among previously undiagnosed children, with 81% of them aged ≥5 years. While elicitation improved over time, contact tracing for HIV testing remained the largest gap, missing opportunities to reach 17% of undiagnosed children. Ensuring adequate resources for tracing and HIV testing and supporting disclosure among couples, while emphasizing elicitation of ART clients' biological children can help to optimize pediatric case finding.

Virological and care outcomes of community ART distribution: Experience with the PODI+ model in Kinshasa, Democratic Republic of the Congo.

INTRODUCTION: Differentiated service delivery models for HIV treatment can minimize unnecessary burdens on health systems and promote efficient delivery of antiretroviral therapy (ART). Under the PODI+ (poste de distribution communautaire) model, ART multi-month dispensation (MMD) was provided by lay workers (peers) in communities. We compared outcomes among clinically stable adults living with HIV receiving MMD via PODI+ or health facility (HF). METHODS: Clients receiving MMD at nine HFs and two PODI+ sites in Kinshasa were followed prospectively for one year (2018-2020). Medication possession ratio (MPR) was measured as proportion of total days with medication during the study through record abstraction at 3-month intervals. Viral load was assessed at enrollment and 12 months. We compared MPR and viral load suppression by arm and examined associations and potential confounders using unadjusted and adjusted odds ratios (AOR). Likert-style client satisfaction was collected during 12-month interviews and described by arm. RESULTS: Odds of maintaining viral load suppression at 12 months for PODI+ participants were two times that for HF participants. In adjusted models, PODI+ participants had 1.89 times the odds of being suppressed at 12 months compared to HF participants (95% CI: 1.10, 3.27). No significant differences in MPR were found between groups (OR: 0.86, 0.38-1.99). Older participants had significantly higher odds of MPR (AOR: 1.02, 95% CI: 1.01, 1.03) and viral suppression (AOR: 1.03, 95% CI: 1.00, 1.07). Satisfaction with services was ≥87% overall, but PODI+ participants rated time spent at site, provider attributes and other care aspects more favorably. CONCLUSIONS: Participants receiving MMD via peer-run community distribution points had similar MPR, but better virological outcomes and greater satisfaction with care than clinically similar participants receiving MMD through facilities. PODI+ could be a useful model for expansion to serve larger clinic populations from overburdened health facilities, particularly as policy shifts towards more inclusive MMD eligibility requirements.