The Montreal Cognitive Assessment has been recommended by the Movement Disorder Society (MDS) for screening cognitive disorders in Parkinson's disease (PD), although validations of language variants are required. This study aimed to determine the reliability and validity of the Vietnamese Montreal Cognitive Assessment (MoCA-V) in Vietnamese PD patients.

PD patients from a movement disorder clinic at a tertiary hospital in Vietnam were recruited. Participants underwent the MoCA-V and Mini-Mental State Examination (MMSE) assessments. Patients were classified into three cognitive groups: dementia, mild cognitive impairment (MCI), and normal cognition (NC), based on the MDS diagnostic criteria. The reliability of the MoCA-V was determined through internal consistency and test-retest reliability. Validity was assessed through concurrent validity, convergent validity, and criterion validity. Exploratory analysis of diagnostic accuracy was performed as a secondary objective.

Among 40 PD patients (dementia: 37.5%, MCI: 52.5%, NC: 10.0%), the MoCA-V demonstrated good internal consistency (Cronbach's alpha = 0.83; 95% confidence interval [0.74–0.90]) and high test-retest reliability (total score's intraclass correlation coefficient = 0.84 [0.69–0.92]; P < 0.001). The total score showed strong correlations with the patient's cognitive status (Kendall's tau-b = 0.63; P < 0.001) and the MMSE total score (Spearman's rho = 0.86; P < 0.001), indicating high concurrent and convergent validity. The scale's criterion validity in discriminating cognitive impairment (including dementia and MCI) from NC and dementia from non-dementia was high (area under the curve [AUC] = 0.98) and moderate (AUC = 0.88), respectively. Notably, the MoCA-V demonstrated superior sensitivity compared to the MMSE in differentiating cognitive impairment from NC.

The MoCA-V is a reliable and valid instrument for cognitive assessment in Vietnamese PD patients across all PD stages, with superior sensitivity to the MMSE for detection of cognitive impairment.