Predicting neuropsychological testing outcomes and research clinic diagnosis of MCI and dementia in Parkinson’s disease using the MoCA

Abstract Background and Objectives The Montreal Cognitive Assessment (MoCA) is frequently used in cohort studies in Parkinson’s disease (PD) as a simple and quick tool for assessing global cognitive abilities of patients. However, cut-off values for distinguishing between normal cognition, mild cognitive impairment (MCI), and dementia differ across the literature. We comprehensively evaluate the accuracy of the MoCA for patient stratification and whether it can improved by including additional routinely collected information. Methods We use longitudinal data from PD and healthy control (HC) participants of the PPMI cohort which – in addition to the MoCA – conducts detailed neuropsychological testing and records diagnoses of MCI and dementia made at the research clinics. Multilevel logistic regression was used to predict (1) impairment in detailed neuropsychological testing and (2) clinician diagnoses from the MoCA in conjunction with other routinely collected information on basic demographics or functional impairment as recorded in MDS-UPDRS 1.1. Model performance was assessed using the area under the ROC curve (AUC). Optimal cut-offs for patient stratification were derived according to Youden’s J, common screening and diagnosis criteria, and an equal proportions criterion, that is, the cutoff at which the proportion of observations with the outcome equals the proportion of observations below cutoff. Results We analysed data from 1,094 PD patients and of 267 HC. Education-adjusted MoCA scores predicted impairment in 2 or more domains with an AUC of 0.86 (95% CI 0.84, 0.88). Youden’s J was maximized at cutoff ≤ 24 with sensitivity 74.7 (70.5, 79.3) and specificity 83.1 (82.0, 84.2); cutoff ≤ 21 equated proportions. The MDS-UPDRS 1.1 was a better predictor of research clinic diagnosis of PD-MCI or PDD than the MoCA. A combination of MDS-UPDRS 1.1 and education adjusted MoCA discriminated diagnosis of any impairment from no impairment with an AUC of 0.87 and dementia from no dementia with an AUC of 0.96. Discussion Optimal MoCA cutoffs for PD-MCI or PDD depend on their purpose. For post-hoc stratification for research purposes, we recommend considering a cutoff that equates proportions. Identifying suitable cutoffs from the literature - for research or in clinic use - needs to take into account the respective PD population.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.