Comparison of actuarial and conventional MCI criteria in a community‐based prospective cohort study: The Adult Changes in Thought (ACT) study. (31st December 2021)

• Record Type: Journal Article
• Authors: Eppig, Joel S; Madan, Raghav; Arrota, Kayela; McCurry, Susan M; Bowen, James D; McCormick, Wayne; Cholerton, Brenna; Craft, Suzanne; Crane, Paul K.; Larson, Eric B.; Trittschuh, Emily H
• Is Part Of: Alzheimer's & dementia. Volume 17(2021)Supplement 6
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-12-31
• Subjects: Alzheimer's disease -- Periodicals; Alzheimer Disease -- Periodicals; Dementia -- Periodicals; Démence; Maladie d'Alzheimer; Périodique électronique (Descripteur de forme); Ressource Internet (Descripteur de forme); 616.83
• DOI: 10.1002/alz.056303
• Languages: English
• ISSNs: 1552-5260

Abstract

Abstract: Background: Operationalizations of mild cognitive impairment (MCI) are inconsistent. Actuarial criteria have been proposed to improve accuracy. We investigated dementia and Alzheimer's disease (AD) conversion for conventional versus actuarial MCI criteria in a community‐based prospective cohort study. Method: 1413 non‐demented participants were administered measures from memory, language, and attention/executive function domains. We used published norms to generate demographically‐adjusted z‐scores. Conventional MCI criteria were met for scores >1.5sd below the mean on any measure. Actuarial MCI criteria were met for 2 scores >1.0sd below the mean in one domain or a score >1.0sd below the mean in all three domains. Participants were followed biennially to capture dementia incidence. A consensus conference used DSM‐IV dementia criteria and NINDS‐ADRDA AD criteria for diagnosis. We focused on 6‐year outcomes following MCI categorization. Result: At baseline, n=875 (62%) met neither conventional nor actuarial MCI criteria, n=16 (1%) met only actuarial MCI criteria, n=383 (27%) met only conventional MCI criteria, and n=139 (10%) met both. Demographic characteristics are in Table 1. Education differed across criteria (p<0.0001); more people with lower education met only conventional MCI criteria. During follow up (7090 person‐years, mean=5 years), there were 175 incident dementia cases, of whom 110 (63%) met conventional and 53 (30%) met actuarial MCI criteria at their baseline neuropsychological evaluation. In a Cox model adjusting for age, sex, education, and race/ethnicity, the conventional MCI dementia hazard ratio (HR) was 2.93 (95% CI 2.15, 4.01, p<0.001), and the actuarial MCI HR was 3.62 (95% CI 2.60, 5.03; p<0.001). There were 154 incident AD cases; 101 (66%) met conventional and 49 (32%) met actuarial criteria. The adjusted conventional MCI AD HR was 3.25 (95% CI 2.32, 4.56, p<0.001), and the actuarial MCI AD HR was 3.81 (95% CI 2.69, 5.38, p<0.001). Conclusion: Both conventional and actuarial MCI criteria identified individuals at higher risk of developing dementia and AD over 6 years of follow‐up. Conventional criteria flag a higher proportion of those who ultimately developed dementia and AD compared to actuarial criteria. However, hazard ratios for progression to dementia and AD were slightly higher for the actuarial criteria.