EXPERTS’ OPINION   Free access

Minerva Anestesiologica 2008 September;74(9):481-8

Copyright © 2008 EDIZIONI MINERVA MEDICA

lingua: Inglese

Statistical methods for evidence-based medicine: the diagnostic test. Part II

Antonelli P. ¹, Chiumello D. ², Cesana B. M. ¹

¹ Medical Statistics and Biometry Unit, Department of Biomedical Sciences and Biotechnologies, University of Brescia, Brescia, Italy; ² Department of Anesthesia and Intensive Care “Maggiore, Policlinico, Mangiagalli, Regina Elena” Hospital IRCCS Foundation University of Milan, Milan, Italy

Part II of this work addresses the use of likelihood ratios for the calculation of a positive predictive value (PPV) and a negative predictive value (NPV) for patient diagnosis. These two values are the most useful indices in the application of a diagnostic test in order to make a diagnosis for a patient. Furthermore, the relevance of sample size is also addressed by providing background theory and pertinent tables. These basic topics for assessing the validity of a diagnostic test and for performing appropriate and valid case vs control studies are explained in a simple way for a clinical audience with a working example.