Background: Zambia has scaled up differentiated service delivery (DSD) models for antiretroviral treatment (ART) to provide more client-centric care and increase service delivery efficiency. The current DSD landscape includes multiple models of care based on guidelines, resources, partner inputs, and other factors. We used local data to identify cost-effective combinations of DSD models that will maximize benefits and/or minimize costs to guide future DSD expansion. Methods: We developed a mathematical Excel-based model using retrospective retention and viral suppression data from a national cohort of ART clients (≥15 years) between January 2018-March 2022 stratified by age, sex, setting (urban/rural), and model of ART delivery. Outcomes (viral suppression and retention in care), provider costs, and costs to clients for each model were estimated from the cohort and previously-published data. For different combinations of the nine DSD models in use, we evaluated the incremental cost to the health system per additional ART client virally suppressed on treatment compared to the 2022 base case. Results: Of the 125 combinations of DSD models evaluated, six were on the cost-effectiveness frontier (CEF): 1) six-month dispensing (6MMD)-only; 2) 6MMD and adherence groups (AGs); 3) AGs-only; 4) fast track refills (FTRs) and AGs; 5) FTRs-only; and 6) AGs and home ART delivery. 6MMD-only was cost-saving compared to the base case, increased the proportion of clients virally suppressed by 1.6%, and decreased costs to clients by 16.6%. The next two scenarios on the CEF, 6MMD+AGs and AGs-only, each cost an additional $245 per person virally suppressed, increased the total number of individuals suppressed on treatment by 2.8% and 4.8%, respectively, and increased costs to clients by 63% and 143%, respectively. Conclusions: Mathematical modelling using existing data can identify cost-effective mixes of DSD models and allocations of clients to these models, while ensuring that all client sub-populations are explicitly considered. In Zambia, providing 6MMD to all eligible clients is likely to be cost-saving, while health outcomes can be improved by allocating clients to selected models based on sub-population.

The authors have declared no competing interest.

Funding for the study was provided by the Bill & Melinda Gates Foundation through OPP1192640 to Boston University and INV-037138 to the Wits Health Consortium. The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript.

I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.

Yes

The details of the IRB/oversight body that provided approval or exemption for the research described are given below:

This study protocol was approved by ERES Converge IRB (Zambia), protocol number 2019-Sep-030; the Human Research Ethics Committee (Medical) of the University of Witwatersrand (South Africa), protocol number M190453; and the Boston University IRB (United States), protocol number H-38823.

I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals.

Yes

I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).

Yes

I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable.

Yes

Non-human subjects data are contained in the manuscript. Human subjects data are owned by the Zambian Ministry of Health and their use was approved by the ERES Converge IRB (Zambia). Full data are available upon approval from the Zambian Ministry of Health and appropriate ethics committees.