Validity of the Montreal Cognitive Assessment (MoCA) in older adults treated at an outpatient clinic in a hospital in Lima, Peru

Abstract

Objective: To evaluate the construct validity of the Montreal Cognitive Assessment (MoCA), applied to a Spanish-speaking population. Materials and methods: Transversal, observational and descriptive study of instrument validation. The MoCA, as well as the Mini-Mental Status Examination (MMSE), were applied by three researchers to 110 patients. The internal consistency of the scale was evaluated using Cronbach’s Alpha, and the concordance between researchers, with Cohen’s Kappa coefficient. Sensibility, specificity, positive and negative predictive values, and Spearman’s Rho were calculated to determine the criterion validity, while the construct validity was determined by principal component analysis. Results: 110 patients were studied. Cronbach’s Alpha was 0.89. Weighted Cohen’s Kappa was 0.17 (p = 0.011). The area below the receiver-operator curve was 0.63. Sensibility was 86.2%; specificity, 39.5%; positive predictive value was 33.8%; negative predictive value, 88.9%; quotient of positive probabilities was 1.43, and the quotient of negative probabilities, 0.35. Spearman’s Rho was 0.24. There were 8 factors that explained 65.6% of the variance. Conclusions: The Montreal Cognitive Assessment demonstrates features of validity and reliability to contribute to diagnosis of cognitive decline in the Peruvian population.