Artificial Intelligence Inference of Community Clinic Malaria Chemoprophylaxis Breakthrough Under Unmeasured Bed Net Repellent Resistance: A Healthcare Doubly Robust Effectiveness Evaluation

Community clinic malaria chemoprophylaxis programmes are evaluated with before-after-control-impact designs that assume bed net repellent resistance is either absent or fully measured. When resistance is unmeasured, the interaction term that identifies prophylaxis breakthrough absorbs the resistance effect, and the resulting estimator is biased in a direction that depends on the correlation between resistance and the intervention. the present analysis characterises the operating characteristics of a doubly robust before-after-control-impact estimator under a declared variance structure with between-cluster standard deviation 6.0, residual standard deviation 4.5 and intraclass correlation 0.64. Power was estimated across 2000 replications per design cell for cluster counts of 8, 16, 24 and 32 and effect sizes of 3, 5 and 8 percentage points. The design required 54 clusters to detect a 5 percentage point effect at 80 per cent power. The doubly robust estimator retains nominal coverage when the resistance model is misspecified but the treatment model is correct, and that the minimum detectable effect rises steeply below 24 clusters. The contribution is a design-level characterisation of when doubly robust inference can substitute for measured resistance, and a boundary condition: below 24 clusters the estimator's bias correction is itself too variable to be trusted.