Accommodating covariates in receiver operating characteristic analysis
the proportion of cases is held constant across centers; scenario 2), the pooled ROC curve is biased; this time it is attenuated with respect to the center-specific ROC curve.This suggests that covariates which impact marker observations among controls should be statistically adjusted in the ROC analysis.
For factors that affect marker observations among controls, we present a method for covariate adjustment. the ROC curve), we describe methods for modelling the ROC curve as a function of covariates.This estimation step is described in more detail in the companion paper (Pepe et al., 2007).Briefly, the CDF can be estimated empirically, or a parametric distribution can be assumed. We bootstrap the data to obtain standard errors for the estimated ROC.Classification accuracy is the ability of a marker or diagnostic test to discriminate between two groups of individuals, cases and controls, and is commonly summarized using the receiver operating characteristic (ROC) curve.In studies of classification accuracy, there are often covariates that should be incorporated into the ROC analysis.