Artificial Intelligence Distributionally Robust Inference of Community Clinic Dengue Warning-Sign Triage Under Unmeasured Rapid-Test Batch Sensitivity Drift: A Healthcare Safety Evaluation

Rapid diagnostic test (RDT) batch-to-batch sensitivity variation is an unmeasured nuisance that can invert the safety ranking of dengue warning-sign triage rules deployed in community clinics. The framework is evaluated through a replication-based design study of the before-after-control-impact estimator under a declared variance structure (between-cluster SD = 6.0, residual SD = 4.5, intraclass correlation = 0.64), with power estimated across 2000 design cells for cluster counts of 8, 16, 24 and 32 and effects of 3, 5 and 8 percentage points.54 clusters are required for 80% power at a 5-percentage-point effect, and that the intraclass correlation of 0.64 dominates the variance budget, so that doubling cluster counts from 16 to 32 yields a smaller power gain than the nominal formula implies. The contribution is a safety-oriented inferential architecture that separates the estimand (triage harm under batch drift) from the design (cluster-level before-after-control-impact), and that reports a minimum detectable effect rather than a point estimate when batch sensitivity is unidentified. The framework is proposed as a design and monitoring instrument for clinic networks, not as a validated clinical tool.