It is critical to discover why some people's cognitive abilities age better than others'. We applied multivariate growth curve models to data from a narrow-age cohort measured on a multi-domain IQ measure at age 11 years and a comprehensive battery of thirteen measures of visuospatial, memory, crystallized, and processing speed abilities at ages 70, 73, and 76 years (n = 1091 at age 70). We found that 48% of the variance in change in performance on the thirteen cognitive measures was shared across all measures, an additional 26% was specific to the four ability domains, and 26% was test-specific. We tested the association of a wide variety of sociodemographic, fitness, health, and genetic variables with each of these cognitive change factors. Models that simultaneously included all covariates accounted for appreciable proportions of variance in the cognitive change factors (e.g. approximately one third of the variance in general cognitive change). However, beyond physical fitness and possession of the APOE e4 allele, very few predictors were incrementally associated with cognitive change at statistically significant levels. The results highlight a small number of factors that predict differences in cognitive ageing, and underscore that correlates of cognitive level are not necessarily predictors of decline. Even larger samples will likely be required to identify additional variables with more modest associations with normal-range heterogeneity in aging-related cognitive declines.

Later‐life changes in brain tissue volumes—decreases in the volume of healthy grey and white matter and increases in the volume of white matter hyperintensities (WMH)—are strong candidates to explain some of the variation in ageing‐related cognitive decline. We assessed fluid intelligence, memory, processing speed, and brain volumes (from structural MRI) at mean age 73 years, and at mean age 76 in a narrow‐age sample of older individuals (n = 657 with brain volumetric data at the initial wave, n = 465 at follow‐up). We used latent variable modeling to extract error‐free cognitive levels and slopes. Initial levels of cognitive ability were predictive of subsequent brain tissue volume changes. Initial brain volumes were not predictive of subsequent cognitive changes. Brain volume changes, especially increases in WMH, were associated with declines in each of the cognitive abilities. All statistically significant results were modest in size (absolute r‐values ranged from 0.114 to 0.334). These results build a comprehensive picture of macrostructural brain volume changes and declines in important cognitive faculties during the eighth decade of life. Hum Brain Mapp 36:4910–4925, 2015. © 2015 The Authors. Human Brain Mapping Published by Wiley Periodicals, Inc

BackgroundStudies investigating the risk factors for or causation of dementia must consider subjects prior to disease onset. To overcome the limitations of prospective studies and self-reported recall of information, the use of existing data is key. This review provides a narrative account of dementia ascertainment methods using sources of existing data.MethodsThe literature search was performed using: MEDLINE, EMBASE, PsychInfo and Web of Science. Included articles reported a UK-based study of dementia in which cases were ascertained using existing data. Existing data included that which was routinely collected and that which was collected for previous research. After removing duplicates, abstracts were screened and the remaining articles were included for full-text review. A quality tool was used to evaluate the description of the ascertainment methodology.ResultsOf the 3545 abstracts screened, 360 articles were selected for full-text review. 47 articles were included for final consideration. Data sources for ascertainment included: death records, national datasets, research databases and hospital records among others. 36 articles used existing data alone for ascertainment, of which 27 used only a single data source. The most frequently used source was a research database. Quality scores ranged from 7/16 to 16/16. Quality scores were better for articles with dementia ascertainment as an outcome. Some papers performed validation studies of dementia ascertainment and most indicated that observed rates of dementia were lower than expected.ConclusionsWe identified a lack of consistency in dementia ascertainment methodology using existing data. With no data source identified as a “gold-standard”, we suggest the use of multiple sources. Where possible, studies should access records with evidence to confirm the diagnosis. Studies should also calculate the dementia ascertainment rate for the population being studied to enable a comparison with an expected rate.Electronic supplementary materialThe online version of this article (doi:10.1186/s12888-017-1401-4) contains supplementary material, which is available to authorized users.