The (in)validity of sensitivity and specificity

This paper is a legacy of the first author, and after her untimely death re constructed by the second author as a tribute to Irene Guggenmoos's contrib ution to biostatistics. It discusses two different views on diagnostic test ing: the classical view in which sensitivity and specificity of a diagnosti c test are considered universal constants, and the more statistical point o f view that focuses on predictive values. The differences between the two p aradigms are outlined and practical examples are discussed to show that the familiar concepts of sensitivity and specificity must be handled with care and not used indiscriminately. Copyright (C) 2000 John Wiley & Sons, Ltd.