Validity of the C antonese C hinese M ontreal C ognitive A ssessment in S outhern C hinese

Aim The objective of the present study was to investigate the reliability and the validity of the C antonese C hinese M ontreal C ognitive A ssessment ( MoCA ) as a brief screening tool of amnestic mild cognitive impairment ( aMCI ) and A lzheimer's disease ( AD ) in S outhern C hinese older adults. Methods Cognitively normal, aMCI and AD C antonese‐speaking C hinese older adults were recruited from a memory clinic and the community. The E nglish MoCA was translated into C antonese C hinese and then back‐translated. We then evaluated the content validity, reliability, sensitivity and specificity of the C hinese C antonese MoCA . Results We recruited 115 cognitively normal controls, 87 aMCI and 64 AD patients. Only education was positively correlated with the C antonese MoCA score ( r  = 0.46, P  < 0.001). The C hinese C antonese MoCA had a high internal consistency with a C ronbach's alpha of 0.85. In the test–retest reliability assessment, the intraclass correlation coefficient ( ICC ) was 0.95 ( P  < 0.001). The ICC for the interrater reliability was 0.96 ( P  < 0.001). Receiving operating characteristic curve analyses showed an area under the curve of 0.85 and 0.99 for aMCI and AD , respectively (both P  < 0.001). The optimal cut‐off score for detection of aMCI was 22/23, which yielded a sensitivity and specificity of 78% and 73%, respectively. The optimal cut‐off score for detection of AD was 19/20, which gave sensitivity and specificity of 94% and 92%, respectively. Conclusion The C antonese C hinese MoCA is a consistent and reliable instrument. In terms of its validity, the MoCA is better in the detection of AD than aMCI in C antonese‐speaking C hinese persons. It is only fair for the screening of aMCI . Geriatr Gerontol Int 2015; 15: 96–103.