Accuracy of the Montreal Cognitive Assessment tool for detecting mild cognitive impairment: A systematic review and meta‐analysis

Nayaar Islam,Rola Hashem,Maryse Gad,Aime Brown,Brooke Levis,Christel Renoux,Brett D. Thombs,Matthew DF McInnes
DOI: https://doi.org/10.1002/alz.13040
2023-03-21
Abstract:INTRODUCTION This systematic review evaluates the accuracy of the Montreal Cognitive Assessment (MoCA) for detecting mild cognitive impairment (MCI). METHODS We searched MEDLINE, PSYCInfo, EMBASE, and Cochrane CENTRAL (1995–2021) for studies comparing the MoCA with validated diagnostic criteria to identify MCI in general practice. Screening, data extraction, and risk of bias assessment were performed independently, in duplicate. Pooled sensitivity and specificity for MoCA cutoffs were estimated using bivariate meta‐analysis. RESULTS Thirteen studies [2158 participants, 948(44%) with MCI] were included; 10 used Petersen criteria as the reference standard. Risk of bias of studies were high or unclear for all domains except reference standard. Sensitivity and specificity were 73.5%(95% confidence interval: 56.7–85.5) and 91.3%(84.6–95.3) at cutoff <23; 79.5%(67.1–88.0) and 83.7%(75.4–89.6) at cutoff <24; and 83.8%(75.6–89.6) and 70.8(62.1–78.3) at cutoff <25. DISCUSSION MoCA cutoffs <23 to <25 maximized the sum of sensitivity and specificity for detecting MCI. The risk of bias of included studies limits confidence in these findings.
clinical neurology
What problem does this paper attempt to address?