EVALUATING MULTIPLE DIAGNOSTIC-TESTS - WITH PARTIAL VERIFICATION

To evaluate diagnostic tests, one would ideally like to verify, for ex ample, with a biopsy, the disease state of all subjects in a study. Of ten, however, not all subjects are verified. Previous methods for eval uation assume that the decision to verify depends only on recorded var iables. Sometimes, particularly if the disease process is not well und erstood, the decision to verify may also depend on unrecorded variable s related to disease. We propose a method to estimate the true- and fa lse-positive rates of multiple tests while adjusting for the effect, o n the decision to verify, of unrecorded variables related to disease. To put the estimates into a more usable form, we develop a simple algo rithm for creating a receiver-operating curve which maximizes the true -positive rate, given the false-positive rate. We apply the methodolog y to data on the early detection of prostate cancer using ultrasonogra phy, digital rectal exam, and prostate specific antigen.