Premium
A concordance index for matched case–control studies with applications in cancer risk
Author(s) -
Brentnall Adam R.,
Cuzick Jack,
Field John,
Duffy Stephen W.
Publication year - 2014
Publication title -
statistics in medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.996
H-Index - 183
eISSN - 1097-0258
pISSN - 0277-6715
DOI - 10.1002/sim.6335
Subject(s) - concordance , estimator , matching (statistics) , statistics , receiver operating characteristic , parametric statistics , measure (data warehouse) , computer science , population , mathematics , econometrics , medicine , data mining , environmental health
In unmatched case–control studies, the area under the receiver operating characteristic (ROC) curve (AUC) may be used to measure how well a variable discriminates between cases and controls. The AUC is sometimes used in matched case–control studies by ignoring matching, but it lacks interpretation because it is not based on an estimate of the ROC for the population of interest. We introduce an alternative measure of discrimination that is the concordance of risk factors conditional on the matching factors. Parametric and non‐parametric estimators are given for different matching scenarios, and applied to real data from breast and lung cancer case–control studies. Diagnostic plots to verify the constancy of discrimination over matching factors are demonstrated. The proposed simple measure is easy to use, interpret, more efficient than unmatched AUC statistics and may be applied to compare the conditional discrimination performance of risk factors. Copyright © 2014 John Wiley & Sons, Ltd.