Related Experiment Video
Updated: May 18, 2026
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
The Beijing version of the Montreal Cognitive Assessment as a brief screening tool for mild cognitive impairment: a
1Center on Ageing Psychology, Key Laboratory of Mental Health, Institute of Psychology, Chinese Academy of Sciences, 4A Datun Road, Chaoyang District, Beijing, 100101, China.
Background:
A cross-sectional validation study was conducted in several urban and rural communities in Beijing, China, to evaluate the effectiveness of the Beijing version of the Montreal Cognitive Assessment (MoCA-BJ) as a screening tool to detect mild cognitive impairment (MCI) among Chinese older adults.
Methods:
The MoCA-BJ and the Mini-Mental State Examination (MMSE) were administered to 1001 Chinese elderly community dwellers recruited from three different regions (i.e., newly developed, old down-town, and rural areas) in Beijing. Twenty-one of these participants were diagnosed by experienced psychiatrists as having dementia, 115 participants were diagnosed as MCI, and 865 participants were considered to be cognitively normal. To analyze the effectiveness of the MoCA-BJ, we examined its psychometric properties, conducted item analyses, evaluated the sensitivity and specificity of the scale, and compared the scale with the MMSE. Demographic and regional differences among our subjects were also taken into consideration.
Results:
Under the recommended cut-off score of 26, the MoCA-BJ demonstrated an excellent sensitivity of 90.4%, and a fair specificity (31.3%). The MoCA-BJ showed optimal sensitivity (68.7%) and specificity (63.9%) when the cut-off score was lowered to 22. Among all the seven cognitive sub-domains, delayed recall was shown to be the best index to differentiate MCI from the normal controls. Regional differences disappeared when the confounding demographic variables (i.e., age and education) were controlled. Item analysis showed that the internal consistency was relatively low in both naming and sentence repetition tasks, and the diagnostic accuracy was similar between the MoCA-BJ and the MMSE.
Conclusions:
In general, the MoCA-BJ is an acceptable tool for MCI screening in both urban and rural regions of Beijing. However, presumably due to the linguistic and cultural differences between the original English version and the Chinese version of the scale, and the lower education level of Chinese older adults, the MoCA-BJ is not much better than the MMSE in detecting MCI, at least for this study sample. Further modifications to several test items of the MoCA-BJ are recommended in order to improve the applicability and effectiveness of the MoCA-BJ in MCI screening among the Chinese population.
