Clinical utility of the Rey 15-Item Test, recognition trial, and error scores for detecting noncredible neuropsychological performance in a mixed clinical sample of veterans
Clinical utility of the Rey 15-Item Test, recognition trial, and error scores for detecting noncredible neuropsychological performance in a mixed clinical sample of veterans
复制标题
DOI：
10.1080/13854046.2017.1333151
复制
发表时间：
2018-01-01
影响因子：
3.9
通讯作者：
O'Rourke, Justin J. F.
中科院分区：
文献类型：
--
作者：
Bailey, K. Chase;Soble, Jason R.;O'Rourke, Justin J. F.
目的：本横断面研究比较了Rey 15-Item Test （RFIT）、Recognition Trial和Error Scores在混合临床退伍军人样本中识别不可信表现的效果，并将其与另一种广泛使用的效度测量方法Test of Memory Malingering （TOMM）进行了比较。方法：选取62名在临床评估时完成RFIT （Recall/Recognition Trials）、TOMM和Word Memory Test （WMT）的退伍军人。以WMT为标准，71% （N = 44）为有效，29% （N = 18）为无效。结果：在有效参与者中，25%的RFIT召回失败，而78%的无效参与者通过（敏感性：22%；特异性：75%；诊断优势比[DOR]: 0.86）。识别试验将识别无效表现的敏感性提高到39%，但25%的有效参与者得分仍低于临界值（特异性：75%；DOR: 1.91）。RFIT Recall and Recognition Trial logistic回归和受试者工作特征（ROC）分析均无统计学意义，分类准确率分别为71和72.6%，曲线下面积（auc）为。52和。55。RFIT错误评分也无法区分有效性组。相比之下，TOMM具有更强的心理测量特性（灵敏度：50%；特异性：97.7%;DOR: 43；分类准确率：82.3%;AUC: 0.91）。此外，认知障碍患者的RFIT回忆和识别失败率分别高出14%和22%，而认知障碍患者和非认知障碍患者的RFIT回忆和识别失败率分别为95%和100%。结论：尽管RFIT在退伍军人神经心理学家中经常使用，但它在检测不可信表现方面的能力有限，并且在混合临床退伍军人样本中错误分类了很大比例的有效参与者，这表明它在这一人群中的效用有限。
Objective: This cross-sectional study examined the Rey 15-Item Test (RFIT), Recognition Trial, and Error Scores for identifying noncredible performance in a mixed clinical veteran sample compared to another widely used validity measure, the Test of Memory Malingering (TOMM). Method: Sixty-two veterans who completed the RFIT (Recall/Recognition Trials), TOMM, and Word Memory Test (WMT) during clinical evaluation were included. Using the WMT as the criterion, 71% (N = 44) were classified as valid and 29% (N = 18) as invalid. Results: Among valid participants, 25% failed the RFIT Recall, whereas 78% of invalid participants passed (sensitivity: 22%; specificity: 75%; diagnostic odds ratio [DOR]: .86). The Recognition Trial increased sensitivity to 39% for identifying invalid performance, but 25% of valid participants still scored below cut-off (specificity: 75%; DOR: 1.91). RFIT Recall and Recognition Trial logistic regression and receiver operating characteristic (ROC) analyses were nonsignificant, with respective classification accuracies of 71 and 72.6% and areas under the curve (AUCs) of .52 and .55. RFIT Error Scores also failed to differentiate validity groups. In contrast, TOMM had stronger psychometric properties (sensitivity: 50%; specificity: 97.7%; DOR: 43; classification accuracy: 82.3%; AUC: .91). Moreover, RFIT Recall and Recognition failure rates were 14 and 22% greater, respectively, among those with cognitive impairment, whereas 95% of those with impairment and 100% without passed the TOMM. Conclusion: Despite frequent use among VA neuropsychologists, the RFIT displayed limited ability to detect noncredible performance and misclassified a large percentage of valid participants in this mixed clinical veteran sample, suggesting limited utility with this population.