Medical Sociology of Malaria Rapid Diagnostic Test Stockouts and Presumptive Antibiotic Prescribing in Congolese Primary Clinics: A Regression-Discontinuity Estimation of Diagnostic-Confirmation Exposure

Malaria rapid diagnostic test (mRDT) stockouts in Congolese primary clinics create natural experiments in diagnostic-confirmation exposure, yet the before-after-control-impact (BACI) designs used to evaluate such exposure are rarely characterised for the clustered, low-cluster-count settings in which they are deployed.54 clusters are required to detect a 5-percentage-point effect at 80% power, that power falls below conventional thresholds at 8 and 16 clusters for all but the largest effect, and that the minimum detectable effect at 80% power declines steeply between 8 and 24 clusters before flattening. The contribution is a bounded design-characterisation that specifies the cluster counts at which BACI-based evaluations of mRDT stockout exposure can and cannot support inference about presumptive antibiotic prescribing, and it reframes the stockout literature's inferential problem as one of design adequacy rather than of effect existence.