Missing Data in Diagnostic Device Studies: Methods and Case Studies
*Xiao-Hua Zhou, University of Washington 

Keywords: Missing disses status; Verification bias; inverse probability weighting; imputation

The accuracy of a diagnostic test can be measured by its sensitivity, specificity, positive predictive and negative predictive values. More generally, a receiver operating characteristic (ROC) curve may be used to represent the accuracy of a diagnostic test. To calculate these measures, we need to determine the disease status of each patient in the sample. The procedure that establishes the patient's disease status is referred to as a gold standard. However, for some studies, only a subset of the patients with diagnostic test results are chosen to receive the gold standard assessment. If the study population consists of only verified cases, the estimated accuracy of the diagnostic test may be biased. This type of bias is called verification bias. In this talk, I will discuss how to correct for verification bias in estimation of the ROC curves and covariate specific ROC curves and their areas. I will illustrate verification bias correction methods with a data set from a large clinical study on Alzheimer's disease.