THE DISCREPANCY IN DISCREPANT ANALYSIS

Authors
Citation
A. Hadgu, THE DISCREPANCY IN DISCREPANT ANALYSIS, Lancet, 348(9027), 1996, pp. 592-593
Citations number
7
Categorie Soggetti
Medicine, General & Internal
Journal title
LancetACNP
ISSN journal
01406736
Volume
348
Issue
9027
Year of publication
1996
Pages
592 - 593
Database
ISI
SICI code
0140-6736(1996)348:9027<592:TDIDA>2.0.ZU;2-8
Abstract
Discrepant analysis is a widely used technique that attempts to provid e estimates of sensitivity and specificity in the presence of an imper fect gold-standard. This technique has been applied by many researcher s to estimate the sensitivity and specificity of DNA-amplification tes ts for Chlamydia trachomatis such as the plasmid-based ligase chain-re action and polymerase chain-reaction tests. However, the sensitivity a nd specificity estimates obtained by discrepant analysis are upwardly biased, and this bias remains even when a perfect test is used to reso lve the discrepant results. This technique should not be adopted for e valuating the performance of a diagnostic test.