Abstract
High-sensitivity cardiac troponin (hs-cTn) is now the recommended biomarker for diagnosis of non-ST-elevation myocardial infarction, but proper interpretation varies based on the assay being used. Nearly uniformly, suggested interpretations of assay-specific hs-cTn results are based on predictive values, which are not applicable to most patients. Through application of a published hs-cTn algorithm to several patient scenarios, we will demonstrate that likelihood ratios are superior to predictive values for patient-centered test interpretation and decision-making. Furthermore, we will provide a blueprint for how to use existing published data presented with predictive values to calculate likelihood ratios. Changing the output of diagnostic accuracy studies and diagnostic algorithms from predictive values to likelihood ratios can improve patient care.