Xu Yangfan, Lin Yangyang, Yi Lingrong, Li Zhao, Li Xian, Yu Yuying, Guo Yuxiao, Wang Yuling, Jiang Haoying, Chen Zhuoming, Svendrovski Anton, Gao Yang, Molloy D William, O'Caoimh Rónán
Department of Rehabilitation Medicine, The Sixth Affiliated Hospital, Sun Yat-sen University, Guangzhou, China.
Department of Rehabilitation Medicine, The Second Affiliated Hospital, Chongqing Medical University, Chongqing, China.
Front Psychol. 2020 Apr 3;11:558. doi: 10.3389/fpsyg.2020.00558. eCollection 2020.
Cognitive frailty describes cognitive impairment associated with physical decline. Few studies have explored whether short cognitive screens identify frailty. We examined the diagnostic accuracy of the Chinese versions of the Quick Mild Cognitive Impairment (Q-CN) screen and Montreal Cognitive Assessment (MoCA-CN) in identifying cognitive frailty.
Ninety-five participants with cognitive symptoms [47 with mild cognitive impairment (MCI), 34 with subjective cognitive disorder, and 14 with dementia] were included from two outpatient rehabilitation clinics. Energy (work intensity) and physical activity levels were recorded. Cognitive frailty was diagnosed by an interdisciplinary team using the IANA/IAGG consensus criteria, stratified on the Clinical Frailty Scale (CFS). Instruments were administered sequentially and randomly by trained assessors, blind to the diagnosis.
The mean age of the sample was 62.6 ± 10.2 years; median CFS score was 4 ± 1 and 36 (38%) were cognitively frail. The Q-CN had similar accuracy in differentiating the non-frail from cognitively frail compared to the MoCA-CN, AUC 0.82 versus 0.74, respectively ( = 0.19). At its optimal cut-off (≤55/100), the Q-CN provided a sensitivity of 83% and specificity of 67% versus 91% and 51%, respectively, for the MoCA-CN (≤23/30). Neither was accurate in separating MCI from cognitive frailty but both accurately separated cognitive frailty from dementia.
Established short cognitive screens may be useful in identifying cognitive frailty in Chinese adults with cognitive complaints but not in separating MCI from cognitive frailty. The QCN had similar accuracy to the MoCA-CN and a shorter administration time in this small and under-powered study, necessitating the need for adequately powered studies in different healthcare settings.
认知衰弱描述了与身体机能下降相关的认知障碍。很少有研究探讨简短的认知筛查能否识别衰弱。我们检验了中文版快速轻度认知障碍筛查量表（Q-CN）和蒙特利尔认知评估量表（MoCA-CN）在识别认知衰弱方面的诊断准确性。
从两家门诊康复诊所纳入了95名有认知症状的参与者[47名轻度认知障碍（MCI）患者、34名主观认知障碍患者和14名痴呆患者]。记录了能量（工作强度）和身体活动水平。由一个跨学科团队使用国际老年营养学会/国际老年学协会（IANA/IAGG）共识标准诊断认知衰弱，并根据临床衰弱量表（CFS）进行分层。由经过培训的评估人员依次随机发放量表，评估人员对诊断结果不知情。
样本的平均年龄为62.6±10.2岁；CFS评分中位数为4±1，36名（38%）为认知衰弱。与MoCA-CN相比，Q-CN在区分非衰弱与认知衰弱方面具有相似的准确性，曲线下面积（AUC）分别为0.82和0.74（P=0.19）。在最佳截断值（≤55/100）时，Q-CN的灵敏度为83%，特异度为67%，而MoCA-CN（≤23/30）的灵敏度和特异度分别为91%和51%。两者在区分MCI与认知衰弱方面均不准确，但都能准确区分认知衰弱与痴呆。
既定的简短认知筛查可能有助于识别有认知主诉的中国成年人中的认知衰弱，但无法区分MCI与认知衰弱。在这项样本量小且效能不足的研究中，Q-CN与MoCA-CN具有相似的准确性且施测时间更短，因此需要在不同医疗环境中开展效能充足的研究。
