Multicentric Longitudinal Study Protocol

Total enrolment is approximately 25,000 individuals across 24 clusters. A malaria episode is counted as an incident case only if it occurs at least 28 days after a prior confirmed episode and appropriate antimalarial treatment has been completed. Poisson and negative binomial regression are used to estimate incidence rate ratios, adjusted for cluster-level effects and confounders, with person-time as an offset. Data are collected via REDCap on tablets with real-time synchronisation to a centralised server, using R and STATA for analysis. A multistage sampling approach is used, selecting the highest-API block within each district and then two villages of approximately 1,000 residents each. PCR is applied to a random subsample of RDT-negative dried blood spots to detect submicroscopic infections and confirm parasite species. The study covers 12 districts across 10 Indian states, selected based on API category, eco-epidemiological diversity, and limited prior research coverage.