Feasibility and acceptability of task sharing collection of HIV viral load dried blood spot samples with community lay cadres: A cross-sectional diagnostic validation study in Zimbabwe

Viral load (VL) testing is a critical tool for clinical management of HIV, yet healthcare worker (HCW) shortages remain a key barrier to VL sample collection. This study assessed the feasibility and acceptability of VL dried blood spot (DBS) sample collection by community lay cadres (CLCs) compared to collection by trained HCWs. We implemented a cross-sectional diagnostic validation study across 10 purposefully selected public health facilities in Zimbabwe over six weeks in March-April 2024. Two DBS samples were collected from 374 participants: a reference sample collected by a HCW and a validation sample collected by a CLC. A subset of 173 CLC collections were observed using a checklist, and surveys were conducted with participating clients and CLCs. Diagnostic comparability was assessed using the proportion of matched pairs with agreement on viral load suppression status and Gwet's AC1 statistic, while survey and observation data were analyzed using descriptive statistics. No samples were rejected by the laboratory, but two samples (one collected by a HCW and one by a CLC) were classified as invalid. Of the 372 paired tests analyzed, 96.0% (95%CI 93.2-97.7%) had concordant results, and the Gwet's AC1 was 0.9564, indicating almost perfect agreement. All critical checklist items were done properly in 90.2% (95%CI 85.7-94.7%) of observed CLC collections. All CLCs reported confidence in performing DBS sample collection (89% very confident; 11% somewhat confident), and 94% of clients indicated willingness to have samples collected by a CLC in the future. These findings suggest that task-sharing VL DBS sample collection with CLCs is a feasible strategy, supported by strong diagnostic comparability, high CLC competency, and client acceptability. Policymakers should consider formalizing task shifting of VL sample collection within facilities. Further evidence on the feasibility of community-based DBS collection is needed.