Background and Purpose:
We determined the diagnostic accuracy of the Montreal Cognitive Assessment (MoCA) for poststroke neurocognitive disorder defined according to the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, criteria in a prospective observational study.
Methods:
Consecutive participants able to complete a cognitive test battery and MoCA 3 months poststroke were included. The reference standard of neurocognitive disorder was defined as a score of ≥1.5 SD below the normative mean in ≥1 cognitive domain on the cognitive test battery.
Results:
Among 521 participants (43.6% women; mean age/SD, 71.5/12.0 years; mean education/SD, 12.4/3.8 years), the area under the receiver operating characteristic curve of MoCA for neurocognitive disorder was 0.80 (95% CI, 0.76–0.84). Using the standard MoCA cutoff <26, sensitivity was 0.71 (0.69–0.79) with specificity of 0.73 (0.66–0.76). MoCA cutoff of <27 gave higher sensitivity (0.82 [0.77–0.85]) at the expense of specificity (0.60 [0.53–0.66]).
Discussion:
MoCA has reasonable accuracy for poststroke neurocognitive disorder diagnosed using the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, criteria.
Registration:
URL:
https://www.clinicaltrials.gov
. Unique identifier: NCT02650531.