A New Methodologic Look at Describing the Performance of Diagnostic Classification Procedures in Medicine
A simple method is delineated for correcting the fundamental measures of validity of binary diagnostic tests, sensitivity and specificity, for the presence of chance agreement between classification result and true status. While recent proposals of such chance—correction techniques result in an undesirable dependence of the measures on the disease prevalence, our approach overcomes this deficit. Point and interval estimates of the chance—corrected measures are derived.
Unable to display preview. Download preview PDF.
- BENNETT, B.M. (1983). Further results on indices of diagnostic screening II. Biometrical Journal, 25, 453–457.Google Scholar
- GEFELLER, O., and BRENNER, H. (1994). How to correct for chance agreement in the estimation of sensitivity and specificity of diagnostic tests.Methods of Information in Medicine, 33, 180–186.Google Scholar
- WHITE, A., and LANDIS, J.R. (1982). A general categorical data methodology for evaluating medical diagnostic tests. Communications in Statistics — Theory and Methods, All, 567–605.Google Scholar