Sensitivity and specificity remain the dominant metrics for evaluating diagnostic tests in clinical epidemiology, and the Spin and Snout mnemonic—“specific findings rule disease in, sensitive findings rule disease out”—continues to be taught as a quick heuristic for interpreting them [1]. Likelihood ratios (LRs), although mathematically derived from sensitivity and specificity, convey diagnostic information in a form better suited to clinical reasoning, particularly within a Bayesian framework [2].
