Accuracy of the Short-Form Montreal Cognitive Assessment Chinese Versions

Abstract

Background: There is a strong need for short and effective methods to screen for cognitive impairment. Recent studies have created short forms of the Montreal Cognitive Assessment (s-MoCA) in English-speaking populations. It is also important to develop a validated Chinese short version to detect cognitive impairment. Methods: Item response theory and computerized adaptive testing analytics were used to construct abbreviated MoCAs across a large neurological sample comprising 6,981 community-dwelling Chinese veterans. Results: Six MoCA items with high discrimination and appropriate difficulty were included in the s-MoCA. The Chinese short versions (sensitivity 0.89/0.90, specificity 0.72/0.77) are similar in performance to the full MoCA in identifying cognitive impairment (sensitivity 0.91, specificity 0.82). Conclusions: These short variants of the MoCA may serve as quick and effective instruments when the original MoCA cannot be feasibly administered in clinical services with a high patient burden and limited cognitive testing resources.