A-81 MoCA Cutoffs for English/ Spanish Bilingual Veterans Assessed in English

Abstract

Abstract Objective The Montreal Cognitive Assessment (MoCA) is a well-known screener of global cognitive functioning. Multiple studies have determined optimal cutoff scores for detection of cognitive impairment in various clinical populations. This study aims to determine appropriate cutoff scores in a clinically mixed bilingual (English/Spanish) sample. Methods A sample of n = 57 self-identified bilingual veterans referred for neuropsychological evaluation at a VA hospital completed the MoCA as part of a full battery. All tests were administered in English. The majority were male (96.4%), Hispanic/Latinx, with 14.65 mean years of education. Only MoCA total score without adding one point for ≤12 years of education was included. Descriptive statistics were used for sample characterization. ROC curve analysis assessed diagnostic accuracy of the MoCA for classification of cognitive impairment (CI). The CI group (n = 40) included both major or mild neurocognitive disorder. The nonimpaired group (n = 17) included persons with no CI or psychiatric diagnosis. Results ROC curve analysis was significant (p < .001) with an AUC of .857 (95% confidence interval .746-.969). A cutoff score of ≤24 was yielded an optimal balance of sensitivity (.900) and specificity (.706). Follow-up independent samples t-test and ANOVA were conducted to examine differences between groups. Conclusions Among bilingual individuals, a cutoff of ≤24 on the MoCA maximized sensitivity and specificity of accurately identifying cases of cognitive impairment. Findings have implications for identifying patients requiring further neuropsychological assessment.