Optimal Cutoff Scores for Dementia and Mild Cognitive Impairment of the Montreal Cognitive Assessment among Elderly and Oldest-Old Chinese Population

Background: All versions of the Montreal Cognitive Assessment (MoCA) lack population-based data of 80-plus individuals. The norms and cut-off scores for mild cognitive impairment (MCI) and dementia of the MoCA are different among five Chinese versions. Objective: To provide the cut-off scores in detecting MCI and dementia of the Peking Medical Union College Hospital version of the MoCA (MoCA-P). Methods: In a cross-sectional survey, Chinese veterans aged >= 60 years completed the MoCA-P and the Mini-Mental State Examination (MMSE). Results: Among 7,445 elderly veterans, 5,085 (68.30%) were aged >= 80 years old, 2,621 (35.20%) had 6 years of education or less, 6,847 (91.97%) were male, and 2,311 (31.04%) and 984 (13.22%) veterans were diagnosed as having MCI and dementia, respectively. Adding two points and one point to the MoCA scores for the primary and middle school groups, respectively, can fully adjust for the notable impact of education but cannot compensate for the effect of age. In the three age groups (60-79, 80-89, and >= 90 years old), the optimal MoCA-P cut-off scores for detecting MCI were <= 25, <= 24, and <= 23, respectively, and for detecting dementia were <= 24, <= 21, and <= 19, respectively, which demonstrated relatively high sensitivities and specificities. The areas under the curves for the MoCA-P for detecting MCI and dementia (0.937 and 0.908, respectively) were greater than those for the MMSE (0.848 and 0.892, respectively). Conclusion: Compared with the MMSE, the MoCA-P is significantly better for detecting MCI in the elderly, particularly in the oldest old population, and it also displays more effectiveness in detecting dementia.