Wu Ya-Huei, Vidal Jean-Sébastien, de Rotrou Jocelyne, Sikkes Sietske A M, Rigaud Anne-Sophie, Plichart Matthieu
Assistance Publique - Hôpitaux de Paris, Hôpital Broca, Paris, France.
Paris Descartes University, Sorbonne Paris Cité, Paris, France.
PLoS One. 2017 Jul 24;12(7):e0181809. doi: 10.1371/journal.pone.0181809. eCollection 2017.
There has been a growing interest in using computerized cognitive assessment to detect age-related cognitive disorders. We have developed a tablet-based cancellation test (e-CT), previously shown as a reliable measure of executive functions and free of effect of familiarity with computer-based devices in healthy older adults. This study aimed to investigate the influence of demographics and current daily use of computer-based devices in older adults with Mild Cognitive Impairment (MCI) and Alzheimer's disease (AD). We further studied the ability of the e-CT to discriminate MCI and AD patients from older adults with normal cognition (NC).
The e-CT was administered to 325 older adults (NC = 112, MCI = 129, AD = 84). Subjects also performed the K-T test, a paper-and-pencil cancellation test from which the e-CT was developed. Multiple linear regression analyses were conducted to assess the contribution of demographics and current daily use of computer-based devices on the e-CT in patient groups. The Receiver Operating Characteristic (ROC) curves and the Area Under the Curve (AUC) were established to compare the efficacy of the e-CT and the K-T test to classify subjects into diagnostic groups.
In the MCI group, age (B = -0.37, p<0.001) and current daily use of computer-based devices (B = 5.85, p<0.001) were associated with the number of correct cancellations of the e-CT. In the AD group, only current daily use of a computer-based device was a significant contributor (B = 6.28, p<0.001). The e-CT (AUC = 0.811; 95% confidence interval [CI]: 0.756-0.867) and the K-T (AUC = 0.837; CI: 0.787-0.887) showed good and comparable diagnostic accuracy to discriminate between MCI and NC subjects. To discriminate between NC and AD, both tests showed high diagnostic accuracy, with the AUC values of 0.923 (CI: 0.876-0.971) and 0.929 (95%CI: 0.886-0.972) for the e-CT and the K-T, respectively.
The e-CT presents satisfying discriminative validity and is a promising tool for detection of early cognitive impairment in older adults.
使用计算机化认知评估来检测与年龄相关的认知障碍的兴趣日益浓厚。我们开发了一种基于平板电脑的划消测验（e-CT），先前已证明它是执行功能的可靠测量方法，并且在健康老年人中不受对基于计算机设备熟悉程度的影响。本研究旨在调查人口统计学特征以及当前老年人日常使用基于计算机设备的情况对轻度认知障碍（MCI）和阿尔茨海默病（AD）患者的影响。我们进一步研究了e-CT区分MCI和AD患者与认知正常（NC）老年人的能力。
对325名老年人进行了e-CT测试（NC = 112，MCI = 129，AD = 84）。受试者还进行了K-T测试，这是一种纸质划消测验，e-CT就是在此基础上开发的。进行了多元线性回归分析，以评估人口统计学特征和当前日常使用基于计算机设备的情况对患者组e-CT的影响。建立了受试者工作特征（ROC）曲线和曲线下面积（AUC），以比较e-CT和K-T测试将受试者分类到诊断组的效能。
在MCI组中，年龄（B = -0.37，p<0.001）和当前日常使用基于计算机设备的情况（B = 5.85，p<0.001）与e-CT正确划消的数量相关。在AD组中，只有当前日常使用基于计算机设备是一个显著的影响因素（B = 6.28，p<0.001）。e-CT（AUC = 0.811；95%置信区间[CI]：0.756 - 0.867）和K-T（AUC = 0.837；CI：0.787 - 0.887）在区分MCI和NC受试者方面显示出良好且相当的诊断准确性。为了区分NC和AD，两种测试都显示出高诊断准确性，e-CT和K-T的AUC值分别为0.923（CI：0.876 - 0.971）和0.929（95%CI：0.886 - 0.972）。
e-CT具有令人满意的区分效度，是检测老年人早期认知障碍的有前景的工具。
