Factor structure of the Montreal Cognitive Assessment in Parkinson disease

Objectives: The Montreal Cognitive Assessment (MoCA) is a common tool for screening mild cognitive impairment (MCI) and dementia. Studies in multiple clinical groups provide evidence for various factor structures mapping to different cognitive domains. We tested the factor structure of the MoCA in a large cohort of early Parkinson disease (PD).

Materials and methods: Complete MoCA data were available from an observational cohort study for 1738 patients with recent-onset PD (64.6% male, mean age 67.6, SD 9.2). Confirmatory factor analysis (CFA) was applied to test previously defined two-factor, six-factor, and three-factor models in the full sample and in a subgroup with possible cognitive impairment (MoCA < 26). Secondary analysis used exploratory factor analysis (EFA; principal factors with oblique rotation).

Results: The mean MoCA score was 25.3 (SD 3.4, range 10-30). Fit statistics in the six-factor model (χ² /df 17.77, root mean square error of approximation [RMSEA] 0.10, comparative fit index [CFI] 0.74, Tucker-Lewis index [TLI] 0.69, standardised root mean square residual [SRMR] 0.07) indicated poorer fit than did previous studies. Findings were similar in the two-factor and three-factor models. EFA suggested an alternative six-factor solution (short-term recall, visuospatial-executive, attention/working memory, verbal-executive, orientation, and expressive language), although CFA did not support the validity of the new model.

Conclusions: The factor structure of the MoCA in early PD was not consistent with that of previous research. This may reflect higher cognitive performance and differing demographics in our sample. The results do not support a clear, clinically relevant factor structure in an early PD group, suggesting that the MoCA should be followed with detailed assessment to obtain domain-specific cognitive profiles.