Covariate-specific ROC curve analysis can accommodate differences between covariate subgroups in the evaluation of diagnostic accuracy
We present an illustrative application of methods that account for covariates in receiver operating characteristic (ROC) curve analysis, using individual patient data on D-dimer testing for excluding pulmonary embolism.
Source: Journal of Clinical Epidemiology - Category: Epidemiology Authors: Jenny Lee, Nick van Es, Toshihiko Takada, Frederikus A. Klok, Geert-Jan Geersing, Jeffrey Blume, Patrick M. Bossuyt, the IPD study team Tags: Original Article Source Type: research