Alzheimer's Disease Research Unit, Yale University School of Medicine, New Haven, CT, USA; Department of Psychiatry, Yale University School of Medicine, New Haven, CT, USA.
Alzheimer's Disease Research Unit, Yale University School of Medicine, New Haven, CT, USA; Department of Psychiatry, Yale University School of Medicine, New Haven, CT, USA.
Am J Geriatr Psychiatry. 2018 Dec;26(12):1258-1267. doi: 10.1016/j.jagp.2018.09.009. Epub 2018 Sep 21.
To investigate optimal cutoff scores and the effects of normative adjustments on the performance of the Montreal Cognitive Assessment (MoCA) as a screening instrument for Mild Cognitive Impairment (MCI) and dementia due to Alzheimer's disease (AD-dementia).
499 adults 48 to 91 years-old enrolled in the Alzheimer's Disease Neuroimaging Initiative (ADNI) and were administered the MoCA during baseline. Participants were classified as either cognitively normal (CN), MCI, or AD-dementia by clinical assessment. Receiver operating characteristic (ROC) analyses were performed using raw MoCA scores, education-adjusted MoCA scores, and a regression-based adjustment derived from the National Alzheimer's Coordinating Center data (NACC). Test performance characteristics were calculated for various cutoffs after each normative correction method.
Areas under the curve (AUC) were similar for raw, education-adjusted, and NACC-adjusted MoCA scores, and demonstrated minimal improvement when adjustments of increasing complexity were applied. Our results suggest that the optimal cutoff score for distinguising MCI is 24 and for distinguising AD-dementia is 22.
This study adds to the understanding of how normative adjustments affect the sensitivity and specificity of the MoCA. Suggested corrections based on education alone do not yield improved test characteristics, but small improvements are attained when a regression-based correction that accounts for age, sex, and education is applied. Furthermore, optimal cutoffs for distinguishing CN from MCI or CN from AD-dementia were lower than previously reported. Optimal cutoffs to detect MCI and AD-dementia may vary in different populations, and further study is needed to determine appropriate use of the MoCA as a screening tool.
研究蒙特利尔认知评估（MoCA）作为轻度认知障碍（MCI）和阿尔茨海默病所致痴呆（AD 痴呆）筛查工具的最佳截断分数和规范调整的效果。
499 名 48 至 91 岁的成年人参加了阿尔茨海默病神经影像学倡议（ADNI），并在基线时接受了 MoCA 评估。参与者根据临床评估分为认知正常（CN）、MCI 或 AD 痴呆。使用原始 MoCA 评分、教育调整的 MoCA 评分和基于国家阿尔茨海默病协调中心数据（NACC）的回归调整进行了接收器工作特征（ROC）分析。在每种规范校正方法之后，计算了各种截断值的测试性能特征。
原始、教育调整和 NACC 调整的 MoCA 评分的曲线下面积（AUC）相似，并且在应用越来越复杂的调整时，表现出最小的改善。我们的结果表明，区分 MCI 的最佳截断分数为 24，区分 AD 痴呆的最佳截断分数为 22。
这项研究增加了对规范调整如何影响 MoCA 的敏感性和特异性的理解。仅基于教育的建议校正并不能提高测试特征，但应用考虑年龄、性别和教育的基于回归的校正可获得微小的改善。此外，区分 CN 与 MCI 或 CN 与 AD 痴呆的最佳截断值低于先前报道的水平。区分 MCI 和 AD 痴呆的最佳截断值可能在不同人群中有所不同，需要进一步研究以确定 MoCA 作为筛查工具的适当用途。
