Modelling the cost-effectiveness of introducing subsidised malaria rapid diagnostic tests in the private retail sector in sub-Saharan Africa

Abstract

Background: Over the last 10 years, there has been a huge shift in malaria diagnosis in public health facilities, due to widespread deployment of rapid diagnostic tests (RDTs), which are accurate, quick and easy to use and inexpensive. There are calls for RDTs to be made available at-scale in the private retail sector where many people with suspected malaria seek care. Retail sector RDT use in sub-Saharan Africa (SSA) is limited to small-scale studies, and robust evidence on value-for-money is not yet available. We modelled the cost-effectiveness of introducing subsidised RDTs and supporting interventions in the SSA retail sector, in a context of a subsidy programme for first-line antimalarials.

Methods: We developed a decision tree following febrile patients through presentation, diagnosis, treatment, disease progression and further care, to final health outcomes. We modelled results for three 'treatment scenarios', based on parameters from three small-scale studies in Nigeria (TS-N), Tanzania (TS-T) and Uganda (TS-U), under low and medium/high transmission (5% and 50% Plasmodium falciparum (parasite) positivity rates (PfPR), respectively).

Results: Cost-effectiveness varied considerably between treatment scenarios. Cost per disability-adjusted life year averted at 5% PfPR was US$482 (TS-N) and US$115 (TS-T) and at 50% PfPR US$44 (TS-N) and US$45 (TS-T), from a health service perspective. TS-U was dominated in both transmission settings.

Conclusion: The cost-effectiveness of subsidised RDTs is strongly influenced by treatment practices, for which further evidence is required from larger-scale operational settings. However, subsidised RDTs could promote increased use of first-line antimalarials in patients with malaria. RDTs may, therefore, be more cost-effective in higher transmission settings, where a greater proportion of patients have malaria and benefit from increased antimalarial use. This is contrary to previous public sector models, where RDTs were most cost-effective in lower transmission settings as they reduced unnecessary antimalarial use in patients without malaria.

Keywords: diagnostics and tools; health economics; malaria.

Figure 1

Key nodes and branches of decision-analytical model pathway of undifferentiated febrile case, from initial presentation at a retail outlet, through diagnosis and initial treatment, the effectiveness of any treatment, disease progression, further care, to final health outcomes. ACT, artemisinin combination therapy; RDT, rapid diagnostic test.

Figure 2

Comparison of initial treatment received in private retail outlets without RDT availability (control) and with RDT availability (intervention) by test result, across three treatment scenarios: TS-N, TS-T, TS-U. ACT, artemisinin combination therapy; RDT, rapid diagnostic test; TS-N, TS-T, TS-U, three small-scale studies in Nigeria, Tanzania and Uganda.

Figure 3

Cost-effectiveness acceptability curves; at 5% PfPR and 50% PfPR (2017 US$) Vertical dotted lines show country-specific cost-effectiveness thresholds for Madagascar (US$84), Uganda (US$115), Nigeria (US$182), Tanzania (US$283), Ghana (US$521) and Kenya (US$630). DALY, disability-adjusted life year; PfPR, Plasmodium falciparum positivity rate.