EM and SEM algorithms to compare the weighted kappa coefficients of two diagnostic tests in the presence of partial verification and discrete covariates

• Published in: Journal of statistical computation and simulation Vol. 90; no. 18; pp. 3454 - 3476
• Main Authors: Roldán-Nofuentes, J. A., Sidaty-Regad, S. B.
• Format: Journal Article
• Language: English
• Published: Abingdon Taylor & Francis 11.12.2020; Taylor & Francis Ltd
• Subjects: 6207; Algorithms; Coefficients; Diagnostic systems; Diagnostic tests; Discrete covariate; EM and SEM algorithms; Hypotheses; Kappa coefficient; partial verification; Verification; weighted kappa coefficient
• ISSN: 0094-9655; 1563-5163
• DOI: 10.1080/00949655.2020.1804903

The weighted kappa coefficient of a binary diagnostic test is a measure of the beyond chance agreement between the diagnostic test and the gold standard, and depends on the sensitivity and the specificity of the diagnostic test, on the disease prevalence and on the relative importance between the false positives and the false negatives. This manuscript studies a hypothesis test to compare the weighted kappa coefficients of two binary diagnostic tests when, in the presence of partial disease verification, a discrete covariate is observed in all individuals. The EM algorithm is applied to estimate the weighted kappa coefficients and the SEM algorithm is applied to estimate their variances-covariances. Simulation experiments were carried out to study the size and the power of the proposed hypothesis test. The results were applied to a real example on the diagnosis of the Alzheimer's disease.