Artificial Intelligence Conformal Prediction Sets for Community Clinic Malaria Rapid Diagnostic Test Interpretation Under Variable Lot-Level Sensitivity Drift: A computed Healthcare Overtreatment Cost Evaluation

Rapid diagnostic tests for malaria are deployed in community clinics where lot-to-lot variation in sensitivity shifts the effective operating point of a fixed reading rule, and where the cost of a false positive is borne as unnecessary antimalarial treatment. Because no field data were available, the study is a design and power analysis: a before-after-control-impact estimator was evaluated across a design grid of 8, 16, 24 and 32 clusters, effect sizes of 3, 5 and 8 percentage points, and 2000 replications per cell, under a variance structure with between-cluster standard deviation 6.0, residual standard deviation 4.5 and intraclass correlation 0.64. Fifty-two clusters were required for 80 per cent power at a 5 percentage point effect. The contribution is a formal statement of how lot-conditional conformal coverage degrades when the calibration lot and the deployment lot differ, together with the design conditions under which a cluster-level evaluation of the resulting overtreatment cost is adequately powered.