Statistical methodology: III. Receiver operating characteristic (ROC) curves

Abstract

Measures including sensitivity, specificity, and positive and negative predictive values have been traditionally used to assess a diagnostic test's ability to detect the presence or absence of disease. Receiver operating characteristic (ROC) curve analysis allows visual evaluation of the trade- offs between sensitivity and specificity associated with different values of the test result, or different 'cutpoints' for defining a positive result. The purpose of this article is to define, construct, and interpret a ROC curve using a hypothetical example applicable to emergency medicine practice.