Artificial Intelligence Benchmarking of District Hospital Chest Radiograph Tuberculosis Triage Under Unmeasured Portable Generator Voltage Sag Artefact: A Healthcare Fairness-Robustness Evaluation

Portable generator voltage sag introduces unmeasured artefact into district hospital chest radiographs used for tuberculosis triage, yet the operating characteristics of before-after-control-impact designs intended to detect triage degradation under such artefact remain uncharacterised. Power was estimated across 2000 design cells defined by cluster counts of 8, 16, 24 and 32 and effect sizes of 3, 5 and 8 percentage points. The analysis identified 56 clusters as the minimum required for 80% power at a 5 percentage point effect. Minimum detectable effect curves declined monotonically with cluster count, and coefficient estimates were centred on their nominal values with coverage consistent with the declared variance structure. The results establish that district-level triage evaluations under unmeasured voltage sag artefact require cluster counts substantially larger than those typical of single-site audits, and that the intraclass correlation of 0.64 dominates the design's sensitivity. The contribution is a formal characterisation of the fairness-robustness evaluation problem for artificial intelligence tuberculosis triage under power-quality artefact, expressed as a design theorem with explicit boundary conditions.