Determining appropriate screening tools and cut-points for cognitive impairment in an elderly Chinese sample.

The establishment of normative data and screening cut-points for cognitive tasks is important to ensure the effective and timely detection of mild cognitive impairment (MCI) and Alzheimer's disease (AD). These need to be culturally relevant and account for known factors that impact on cognition such as age, education, and gender. In this study, 1,068 elderly Chinese residents of Shanghai completed a comprehensive series of cognitive tasks as part of a community screening study with 1027 meeting criteria for analysis, age M(SD) = 72.54 (8.40). MCI was detected in 267 individuals, AD in 50, and 710 had normal cognition. Receiver Operator Characteristic curve analysis indicated that the Mini-Mental State Examination (MMSE) and the Montreal Cognitive Assessment (MoCA) best differentiated normal cognition from MCI and AD. We present suggested cut-points to differentiate between normal cognition and MCI and AD for the total sample, and when split according to education levels, age, and gender. Trends suggest that the MoCA was better suited to detecting MCI, and the MMSE was better for detecting AD. For younger and more educated participants, only a slight impairment was necessary to meet screening criteria, while a larger impairment was necessary for older and less educated participants. Both tasks had a high negative predictive values for MCI and AD, and variable positive predictive values. The cut-points presented can be used to inform future work using the MMSE and MoCA to screen for MCI and AD in older Chinese people. (PsycINFO Database Record (c) 2016 APA, all rights reserved).