A New Test for Assessing the Covariate Effect in ROC Curves

The ROC curve is a statistical tool that analyzes the accuracy of a diagnostic test in which a variable is used to decide whether an individual is healthy or not. Along with that diagnostic variable, it is usual to have information on some other covariates. In some situations, it is advisable to incorporate that information into the study, as the performance of the ROC curves can be affected by them. Using the covariate-adjusted, the covariate-specific, or the pooled ROC curves, we discuss the implications of excluding or including the covariates in the analysis. Motivated by the above, a new test for comparing the covariate-adjusted and the pooled ROC curve is proposed, and the problem is illustrated by analyzing a real database.