News
Article
Author(s):
Data from a new study suggests the Montreal Cognitive Assessment cut points should be lowered for both English and Spanish screenings for more accurate cognitive impairment screenings.
A commonly used screening tool to detect cognitive impairment is not accurate when used in a diverse population of older adults, according to a new study.1
When investigators screened adults aged ≥ 65 years old from Bronx, New York with cognitive issues, they found Montreal Cognitive Assessment (MoCA) cut points for mild cognitive impairment or dementia were "inappropriately high" when compared to neuropsychological testing, with a a high false-positive rate for detecting cognitive impairment observed with the score.
“These findings underscore the importance of considering cultural factors and social determinants of health when evaluating performances on cognitive screening tools such as the MoCA, particularly in traditionally underserved communities,” said lead investigator Marnina Stimmel, PhD, of the Montefiore Health System and Albert Einstein College of Medicine in Bronx, New York, in a press release.
Stimmel and colleagues wanted to see if currently published English and Spanish cut points for cognitive impairment in the MoCA were appropriate in diverse community-based populations. Efficacy and validity of cognitive screening in diverse populations was previously unclear.2
According to research in 2020 which evaluated the cognitive impairment of 81 African Americans, the recommended MoCA cutoff score for impairment of ≤ 26 resulted in 100% sensitivity with a low 31% specificity for identifying mild cognitive impairment or dementia.3 When having a cutoff score of ≤ 24 points, it demonstrated good sensitivity for detecting impairment (95%) with improved specificity (63%). Furthermore, a cutoff core of ≤ 22 points demonstrated good sensitivity (96%) and specificity (88%).
Although, as stated on MoCA's official website, a Thai sample showed the optimal cut-off score of <24/30, yielding a 81% sensitivity and 86% specificity.4
Investigators of the new study screened 231 participants with 43% Hispanic and 39% Black/African American.2 The cohort included 72% women with a mean age of 73 years old.
About half completed the test in English (49%) and half in Spanish (51%). After screening, a neuropsychologist determined a cognitive status as either normal with subjective cognitive concerns (90 participants, a mean score of 19.9), mild cognitive impairment (133 participants, a mean score of 16.6), and dementia (8 participants, mean score of 10.6). The mean English Montreal Cognitive Assessment score was 18.6, and the mean Spanish score was 16.7.
Published data stated the MoCA cutpoint was a score of ≤ 23, but with this inaccurate cutpoint, the diverse participants in the study received a high false-positive rate (79%). According to receiver operating characteristic curve (ROC) analyses, the English cutpoint should be ≤ 18.5 to classify mild cognitive impairment or dementia, and the Spanish cutpoint should be ≤16.5. The ROC analyses of the English screening had a 65% sensitivity and a 77% specificity, and the analyses of the Spanish screening had a 64% sensitivity and a 73% specificity).
“Lower Spanish and English MoCA cutpoints may improve diagnostic accuracy for identifying cognitive impairment in this group, highlighting the need for the creation and validation of accurate cognitive screeners for ethnoculturally and linguistically diverse older adults,” investigators concluded.
References