Information Borrowing Methods for Covariate-Adjusted ROC Curve

• Activity Number: 340
• Type: Contributed
• Date/Time: Tuesday, July 31, 2012 : 10:30 AM to 12:20 PM
• Sponsor: Biometrics Section
• Author(s): Zhong Guan*+ and Jing Qin and Biao Zhang
• Companies: and National Institute of Allergy and Infectious Diseases and University of Toledo
• Keywords: Bootstrap ; Covariate-adjusted ROC curve ; Density ratio model ; Semiparametric likelihood ; Sensitivity ; Specificity

Abstract

In medical diagnostic testing problems, the covariate adjusted receiver operating characteristic (ROC) curves have been discussed recently for achieving the best separation between disease and control. Due to various restrictions such as cost, the availability of patients, and ethic issue quite frequently that only limited information is available. As a result, it is less likely to have large enough overall sample size to support reliable direct estimations of ROCs for all the underlying covariates of interest. Therefore, it is desirable to use indirect estimates that borrow strength by employing values of the variables of interest from neighboring covariates. In this paper we discuss two semiparametric exponential tilting models, where the density functions from different covariate levels share a common baseline density and the parameters in the exponential tilting component reflect the difference among covarities. With the proposed models, the estimated covariate adjusted ROC is much smoother and more efficient than the nonparametric counterpart without borrowing information from neighboring covariates. A simulation study and a real data application are reported.