Development of a new wealth index for Tanzania: the moderated effect of the implementation of 1,7- malaria reactive community-based testing and response (1,7-mRCTR) by socioeconomic position (SEP) with malaria prevention.

INTRODUCTION: The China-UK-Tanzania pilot project of 1,7-malaria reactive community-based testing and response (1,7-mRCTR) approach was implemented in Tanzania between 2015 and 2018. This project targeted villages with the highest malaria incidence to conduct screening and treatment. While socioeconomic factors are known to be strongly associated with malaria burden, their specific impacts on malaria prevention behaviours during the 1,7-mRCTR implementation period remained unclear. This study aimed to construct a household wealth index and investigate its association with malaria prevention outcomes within the context of 1,7-mRCTR. METHODS: We used data from two cross-sectional household surveys conducted in 2015 (baseline) and 2018 (endline), covering 19 686 households. A 12-item wealth index was constructed using Mokken scale analysis, with weighted wealth scores calculated via multiple correspondence analysis to categorise households into wealth tertiles. Using logistic regression within a Difference-in-Differences (DID) framework, we assessed the association between household wealth and the household ownership of useful long-lasting insecticidal nets (LLINs), use of LLINs and use of antimalarial drugs. RESULTS: Analysis of the pooled data showed that households in the first (poorest) tertile had significantly lower odds of owing LLINs (OR=0.62, 95% CI 0.54 to 0.70, p<0.001) and using LLINs (OR=0.53, 95% CI 0.45 to 0.62, p<0.001) compared to the third (wealthiest) tertile. The DID analysis, accounting for the interaction between the intervention period (2018 vs 2015) and wealth tertile, showed a significantly greater increase in the odds of owing LLINs (OR=1.26, 95% CI 1.03 to 1.56) and using LLINs (OR=1.88, 95% CI 1.25 to 2.82) among households in the first tertile compared with the third tertile. CONCLUSION: The wealth index effectively differentiated household socioeconomic status, revealing significant wealth-based disparities in malaria prevention behaviours. Importantly, the implementation of the 1,7-mRCTR approach appears to have had a disproportionately positive effect on poorer households, leading to a reduction in wealth-based inequalities related to key malaria prevention measures.