Evaluation of sensitivity, specificity predictive values and their comparison for two diagnostic procedures

ABSTRACT

Methods for evaluating sensitivity [zeta], specificity [eta] and predictive values [p's] for a laboratory technique are proposed for detection of a particular disease D. The methods for comparing sensitivities, specificities and their predictive values for two diagnostic procedures say t[1], and t[2] for a specific disease D are discussed. It is assumed that a definite result can be obtained on the presence or absence of disease D in each of a series of n patients together with their diagnoses by the two procedures. In this paper, Chi-square [x[2]] tests are discussed for homogeneity of zeta, eta and p's. In the end, two examples are illustrated to give practical concept, for using these methods: