Determining sample size to evaluate and compare the accuracy of binary diagnostic tests in the presence of partial disease verification

Calculating sample size to evaluate the accuracy of a binary diagnostic test and to compare the accuracy of two binary diagnostic tests is an important question in the study of diagnostic statistical methods. In the presence of partial disease verification, the disease status of some patients in the sample is unknown, so that the calculation of sample size can be complicated. A method to calculate sample size when evaluating the sensitivity and the specificity of a binary diagnostic test andwhen comparing the sensitivity and specificity of two binary tests in the presence of partial disease verification is proposed. The results obtained were applied to the diagnosis of coronary stenosis. © 2008 Elsevier B.V. All rights reserved.