Cost-effectiveness analysis of G6PD diagnostic test for Plasmodium vivax radical cure in Lao PDR: An economic modelling study

BACKGROUND: Plasmodium vivax (Pv) infections were 68% of the total malaria burden in Laos in 2019. The parasite causes frequent relapses, which can be prevented by primaquine (PMQ). Testing for glucose-6-phosphate-dehydrogenase (G6PD) deficiency is recommended before giving PMQ to avoid haemolysis. Because of the risk of haemolysis in G6PD intermediate deficiencies among females, Laos uses the PMQ 14-days regimen only in G6PD normal females. Among G6PD point-of-care tests, qualitative tests cannot differentiate between G6PD normal and intermediate females. Quantitative tests are required to differentiate between G6PD normal and intermediate deficiencies. However, the quantitative test lacks the cost-effectiveness evidence necessary for decision-making for large-scale adoption. This study examined the cost-effectiveness of quantitative G6PD test, with either supervised PMQ treatment or unsupervised PMQ treatment, against the usual unsupervised PMQ 8-weeks strategy. Supervised PMQ 8-weeks strategy without G6PD testing was also compared against the unsupervised PMQ 8-weeks strategy since the former had recently been adopted in malaria high burden villages that had village malaria volunteers. A budget impact analysis was conducted to understand the incremental cost and effect needed for a nationwide scale-up of the chosen strategy. METHODS: A decision tree model compared the cost-effectiveness of implementing four strategies at one health facility with an average of 14 Pv cases in one year. The strategies were unsupervised PMQ strategy, supervised PMQ strategy, G6PD test with unsupervised PMQ strategy, and G6PD test with supervised PMQ strategy. Disability Adjusted Life Years (DALYs) was the effect measure. Costs were calculated from a payer perspective, and sensitivity analyses were conducted. One Gross Domestic Product (GDP) per capita of Laos was set as the cost-effectiveness threshold. Budget impact analysis was conducted using the health facility wise Pv data in Laos in 2020. FINDINGS: Supervised PMQ strategy was extendedly dominated by G6PD test strategies. When compared against the unsupervised PMQ strategy, both G6PD test strategies were more costly but more effective. Their Incremental Cost-Effectiveness Ratios (ICER) were 96.72US$ for the G6PD test with unsupervised PMQ strategy and 184.86US$ for the G6PD test with supervised PMQ strategy. Both ICERs were lower than one GDP per capita in Laos. Following the sensitivity analysis, low adherence for PMQ 14 days made both G6PD test strategies less cost-effective. The lower the Pv case number reported in a health facility, the higher the ICER was. In the budget impact analysis, the expected budget need was only half a million US$ when the G6PD test rollout was discriminately done depending on the Pv case number reported at the health facilities. Indiscriminate roll out of G6PD test to all health facilities was most expensive with least effect impact.