Predictors of Cognitive Aging Profiles Over 15 Years: A Longitudinal Population-Based Study

The present study aimed to characterize profiles fi les of cognitive aging and how these can be predicted from interindividual differences in demographic, lifestyle, health, and genetic factors. The participants were 1,966 older adults (mean baseline age = 71.6 years; 62.9% female), free from dementia at baseline and with at least two cognitive assessments over the 15-year follow-up, from the population-based Swedish National Study on Aging and Care in Kungsholmen. The cognitive assessment comprised tests of semantic and episodic memory, letter and category fl uency, perceptual speed, and executive function. First, we estimated the level and change within each of the cognitive domains with linear mixed effect models, based on which we grouped our sample into participants with " maintained high cognition," " " moderate cognitive decline," " or " accelerated cognitive decline." " Second, we analyzed determinants of group membership within each cognitive domain with multinomial logistic regression. Third, group memberships within each cognitive domain were used to derive general cognitive aging profiles fi les with latent class analysis. Fourth, the determinants of these profile fi le memberships were analyzed with multinomial logistic regression. Follow-up analyses targeted profiles fi les and predictors specifically fi cally related to the rate of cognitive change. We identified fi ed three latent profiles fi les of overall cognitive performance during the follow-up period with 31.6% of the sample having maintained high cognition, 50.6% having moderate cognitive decline, and 17.8% having accelerated cognitive decline. In multiadjusted analyses, maintained high cognition was predicted by female sex, higher education, and faster walking speed. Smoking, loneliness, and being an epsilon 4 carrier were associated with a lower likelihood of maintained high cognition. Higher age, diagnosis of diabetes, depression, and carrying the apolipoprotein E epsilon 4 allele increased the likelihood of accelerated cognitive decline. Factors at baseline that could significantly fi cantly predict profile fi le membership within the specific fi c cognitive domains included age, sex, years of education, walking speed, diabetes, and the epsilon 4 allele. Of note, these factors differed across cognitive domains. In sum, we identified fi ed demographic, lifestyle, health, and genetic factors of interindividual differences in domain-specific fi c and general cognitive aging profiles, fi les, some of which are modifiable. fi able.