Characteristics of costing study participants.

Tuberculosis preventive treatment (TPT) is an essential intervention recommended for all child contacts in Ethiopia under 15 years who are at risk of tuberculosis (TB) infection. We conducted a cost and cost-effectiveness analysis of home-based versus facility-based TPT provision for child contacts in Ethiopia. As part of the CHIP TB trial, a pragmatic, cluster-randomized trial conducted at eighteen clinics in Ethiopia, clinics were randomized to either a home-based model (intervention arm), administered by community health workers, or a facility-based model (standard of care) for managing child contacts. Cost data were collected from both a health service perspective and a household perspective, capturing all costs relevant to TPT. Costs were evaluated on a per-household basis, with the primary outcome being the difference in median costs per household initiating TPT. A secondary outcome assessed the cost-effectiveness as the incremental cost per additional child contact starting TPT. Probabilistic sensitivity analyses (PSA) were conducted to examine the robustness of findings. At an average cost of US$18.92 per household managed, Home-based contact management, including TPT delivery was cost-saving compared to facility-based TPT delivery (US$27.24 per household managed) assessed based on the partial-societal perspectives, largely due to reductions in household out-of-pocket costs. The home-based strategy was both less costly and had increased TPT initiation in 61.5% of the scenarios assessed in the PSA. Home-based contact management is a cost-saving alternative for households and provides comparable initiation rates to facility-based care, making it a feasible approach to improve TB preventive treatment accessibility. Although it does not entirely replace facility-based care, a hybrid model that respects household preferences and allows flexibility in delivery could enhance TB care access for socio-economically disadvantaged households, potentially reducing health inequities. The trial was registered on clinicaltrials.gov NCT04369326 on April 30, 2020. https://clinicaltrials.gov/study/NCT04369326.