An application of latent class analysis: Determining the sensitivity and specificity of measures of carpal tunnel syndrome.

摘要

The accuracy of a diagnostic measure used to classify a patient as being diseased or disease-free is a valuable piece of information to be used by the physician when making treatment decisions. If a gold standard reference test is available, determining the sensitivity and specificity of a new test is straightforward. However, if a reference test is incorrectly assumed to be perfectly sensitive and specific, the errors of the reference test can result in an under-estimation of the accuracy of the new measure. Latent class analysis, a technique that uses maximum likelihood estimation, can be applied to determine the sensitivity and specificity of a new test when no gold standard exists. In this study, latent class analysis is used to determine the accuracy rates of three commonly used measures of carpal tunnel syndrome: the Tinel's Sign, the Phalen's Maneuver, and the electrical diagnostic test. It is shown that the Tinel's Sign and the Phalen's Maneuver are both highly sensitive and specific. This study suggests that a reduction in the use of the more expensive electrical diagnostic tests might be warranted. Future studies are proposed to confirm these findings.