Transparent comparisons of Emergency-Department prioritization policies: integrating tail risk, target attainment, and utility analysis

Comparative studies of emergency-department (ED) patient-prioritization rules frequently rely on single, average-based metrics, obscuring clinically important trade-offs. We introduce a three-part evaluation framework that combines tail-sensitive summary statistics, threshold-attainment curves, and stakeholder-informed utility analysis, and apply it in a discrete-event simulation of a 30-bed mixed-acuity ED. Tail statistics expose extremes—for example, the 99th-percentile length-of-stay (LOS) gap between strategies exceeds ten times the corresponding mean gap—and reveal cohort-specific inequities masked by overall averages. Threshold-attainment curves map the fraction of patients meeting every LOS target, showing where strategy rankings switch as service standards tighten or relax. Utility contours translate multi-cohort performance into a single stakeholder score, demonstrating how preferred rules pivot when institutional priorities shift from rapid discharge of low-acuity patients to protection of higher-acuity throughput. Together, the techniques uncover complementary, non-redundant information: each one highlights strengths and weaknesses invisible to the others. The framework therefore offers a transparent, replicable template for selecting prioritization policies that align with local clinical objectives, resource constraints, and risk tolerances.