ABSTRACT

The results of neuropsychological tests may be distorted by patients who exaggerate cognitive deficits. Eighty-three patients with cognitive deficit [Amnestic Mild Cognitive Impairment (aMCI), n = 53; Alzheimer’s disease (AD) dementia, n = 30], 44 healthy older adults (HA), and 30 simulators of AD (s-AD) underwent comprehensive neuropsychological assessment. Receiver Operating Characteristic (ROC) analysis revealed high specificity but low sensitivity of the Delayed Matching to Sample Task (DMS48) in differentiating s-AD from AD dementia (87 and 53%, respectively) and from aMCI (96 and 57%). The sensitivity was considerably increased by using the DMS48/Rey Auditory Verbal Learning Test (RAVLT) ratio (specificity and sensitivity 93% and 93% for AD dementia and 96% and 80% for aMCI). The DMS48 differentiates s-AD from both aMCI and AD dementia with high specificity but low sensitivity. Its predictive value greatly increased when evaluated together with the RAVLT.

摘要

夸大认知缺陷的患者可能会歪曲神经心理学测试的结果。83 名认知缺陷患者 [遗忘型轻度认知障碍 (aMCI)，n = 53；阿尔茨海默病 (AD) 痴呆，n = 30]、44 名健康老年人 (HA) 和 30 名 AD 模拟器 (s-AD) 接受了全面的神经心理学评估。接受者操作特征 (ROC) 分析显示延迟匹配样本任务 (DMS48) 在区分 s-AD 与 AD 痴呆（分别为 87% 和 53%）和 aMCI（分别为 96% 和 57%）方面具有高特异性但敏感性低。通过使用 DMS48/Rey 听觉语言学习测试 (RAVLT) 比值（AD 痴呆的特异性和敏感性分别为 93% 和 93%，aMCI 的特异性和敏感性分别为 96% 和 80%），敏感性显着提高。DMS48 以高特异性但低敏感性将 s-AD 与 aMCI 和 AD 痴呆区分开来。当与 RAVLT 一起评估时，它的预测价值大大增加。