Is the Cutoff of the MoCA too High? Longitudinal Data From Highly Educated Older Adults.

BACKGROUND: The Montreal Cognitive Assessment (MoCA) is widely used to evaluate cognitive decline in older individuals. Although, age and education-related norms have been published, the vast majority of diagnostic clinicians use the MoCA cutoff score to identify patients with cognitive impairment. AIM: To identify whether the MoCA cutoff is too stringent for cognitively normal older adults. METHODS: Twenty-seven participants aged 68 to 83 (mean = 75.07, standard deviation [SD] = 4.62), with high education level (mean = 17.14 years, SD = 3.21) underwent cognitive assessment once a year for 5 consecutive years. The cognitive assessment included MoCA; Rey Auditory Verbal Learning Test; Rey Osterrieth Complex Figure test; Wechsler Adult Intelligence Scale Information and Digit Span Subtest; Trail Making Test; Verbal Fluency Test; and Beck Depression Inventory questionnaire. Repeated measures analysis of variance (ANOVA) was used to analyze all standardized scores as well as MoCA standardized and raw scores across all years. RESULTS: Repeated-measures ANOVA for MoCA raw scores yielded significant decline across the years (P < .05). From the second year and forward, the average MoCA total score was below the cutoff of 26/30. However, in substantial contrast, all other neuropsychological scores and the MoCA standardized scores were within the normal range and even above in all years. CONCLUSION: Our study demonstrates that the currently used MoCA cutoff is too high even for highly educated, cognitively normal older adults. Therefore, it is crucial to use the age- and education-related norms for the MoCA in order to avoid misdiagnosis of cognitive decline.

WAIS Information Subtest as an indicator of crystallized cognitive abilities and brain reserve among highly educated older adults: A three-year longitudinal study.

The Wechsler Adult Intelligence Scale-Third Edition (WAIS-III) Information Subtest (IS) is known as a neuropsychological "Hold" test that is relatively resistant to decline with aging. We administered neuropsychological tests among highly educated healthy older adults once a year for three subsequent years. Results showed highly stable performances on the IS across years (Mean Z score: T0 = 1.39, SD = 0.60; T1 = 1.37, SD = 0.77; T2 = 1.50, SD = 0.66; T3 = 1.48, SD = 0.66), that were significantly higher than zero (T0: t = 12.08; T1: t = 9.29; T2: t = 11.71; T3: t = 11.68; for all, p < 0.0001). In contrast, other neuropsychological tests showed differences in performance across time; some performances significantly declined (Rey Osterrieth Complex Figure test-copy, Rey-Auditory Verbal Learning Test, and Montreal Cognitive Assessment test [MoCA]), whereas others were improved, possibly due to practice effects (Rey Osterrieth Complex Figure test- delayed, Rey-Auditory Verbal Learning Test- delayed, and Trail Making Test- part A). Correlation with whole brain volumetric analysis revealed a positive correlation between gray matter volumes and IS scores (r = 0.46, p < 0.05) even when controlling for age and education (partial correlations: r = 0.43; r = 0.45, for both p < 0.05). No significant correlations were found between gray matter and other test scores. Therefore, the WAIS-III Information subtest appears to be an adequate neuropsychological measurement of crystallized ability in highly educated older adults and may be considered as a proxy measure of brain reserve.