Semiparametric Covariate-Specific ROC Curve Estimation Under Test-Dependent Sampling

• Activity Number: 611
• Type: Contributed
• Date/Time: Thursday, August 7, 2014 : 8:30 AM to 10:20 AM
• Sponsor: Section on Medical Devices and Diagnostics
• Author(s): Bethany Horton*+ and Haibo Zhou
• Companies: University of North Carolina at Chapel Hill and University of North Carolina at Chapel Hill
• Keywords: ROC Curve ; Test-dependent sampling ; Missing data ; Semiparametric ; Empirical likelihood

Abstract

The receiver operating characteristic (ROC) curve is used to describe the ability of a screening test to discriminate between diseased and non-diseased subjects. As evaluating the true disease status can be costly, it is beneficial for researchers to increase study efficiency by allowing selection probabilities to depend on the screening test. We propose a semi-parametric covariate-specific ROC curve estimator which utilizes a test-dependent sampling (TDS) design where TDS inclusion is dependent on a continuous screening test measure. Incorporating covariates into the ROC curve estimator allows for evaluation of the utility of the screening test for different subsets of a population. Efficiency is gained by incorporating information from both the sampled and un-sampled portions of the population. Disease status is validated only for those who are sampled. Empirical likelihood methods are used to avoid making distributional assumptions for the covariates. Simulation studies show an increase in efficiency for the proposed estimator compared to other existing estimators. This cost-effective design allows for a more powerful study on the same budget.