B1457
Title: Comparison of the average kappa coefficients of two binary diagnostic tests subject to a paired design
Authors: Jose Antonio Roldan Nofuentes - University of Granada (Spain) [presenting]
Maria del Carmen Olvera Porcel - University of Granada (Spain)
Abstract: The fundamental parameters to assess and compare the performance of binary diagnostic tests are sensitivity and specificity. When considering the losses associated with an erroneous classification with the binary diagnostic test, the parameter that is used to assess the binary diagnostic test is the weighted kappa coefficient. The weighted kappa coefficient depends on the sensitivity and the specificity of the binary diagnostic test, on the disease prevalence of the population studied and on the relative loss between the false positives and the false negatives (weighting index). Another parameter to assess and compare binary diagnostic tests are the average kappa coefficients. The average kappa coefficient of a binary diagnostic test is a chance corrected index between the binary diagnostic test and the gold standard, and it depends on the sensitivity and the specificity of the diagnostic test and on the disease prevalence. Several hypothesis tests are studied to compare the average kappa coefficients of two binary diagnostic tests subject to a paired design. Simulation experiments were carried out to study the type I errors and the powers of the hypothesis tests studied. The results were applied to a real example.
