That old dog can still hunt—Alternative cutoffs and the recognition trial improve the classification accuracy of the Rey 15-Item Test.

Abstract

This study was designed to test the upper limits of the Rey-15's classification accuracy in cognitively intact university students and clinical patients. A brief battery of neuropsychological tests was administered to all participants. Students were randomly assigned to either the control (n = 80) or experimental malingering (n = 59) condition. Invalid performance in patients (n = 52) was psychometrically defined. The classification accuracy of the Rey-15 (original free recall and the combination score including the recognition trial developed by Boone et al., 2002) was computed across a range of cutoffs. A free recall ≤13 and combination score ≤25 produced a good combination of sensitivity (.41-.70) and specificity (.90-.94) among students. The highest cutoff to reach .90 specificity in clinical patients was free recall ≤11 and combination score ≤23, at .38-.56 sensitivity. Results confirmed the previous reports that more liberal alternative cutoffs maintain adequate specificity. Cutoffs can be raised even further in cognitively intact examinees. The introduction of the recognition trial provided a significant boost to the instrument's classification accuracy. The Rey-15 remains a cost-effective and clinically useful test in both clinical and research settings. Public Significance StatementThe Rey-15 is one of the oldest performance validity test often criticized for its limited clinical utility. This study demonstrated that its classification accuracy can be improved through the use of alternative cutoffs.