ABSTRACT
This chapter describes the methodology for making inferences with respect to the basic measures of test accuracy and will begin with a section on the design of such studies. The elements of good design will be explained in the context of a protocol submission of a trial to assess the accuracy of a diagnostic test in a clinical situation. The submission procedure at the MD Anderson Cancer Center (MDACC) is very formal and all submissions are required to contain evidence of a well-designed experiment. The first step in protocol submission is the review by the Department of Biostatistics and Applied Mathematics.
