Contribution of modifiable dementia risk factors to cognitive performance and subjective cognition in middle‐aged adults

Abstract

AbstractBackgroundCharacterization of the contribution of modifiable risk factors (MRF) to dementia generally consider MRFs individually, despite strong evidence that MRFs co‐occur. In a large group of middle‐aged adults, the prevalence and co‐occurrence of MRFs for dementia was determined, spanning five broad domains (mood, lifestyle behaviours (e.g., physical inactivity), cardiovascular health, cognitive/social engagement, and sleep). We then investigated relationships between MRFs, both individually and combined, to cognitive performance and subjective cognition.MethodMiddle‐aged adults (n = 1610), most (70%) with a family history of dementia, enrolled in the Healthy Brain Project, completed an extensive set of questionnaires about their physical and psychological health and lifestyle. Participants also completed the Cogstate Brief Battery (CBB), and the Cognitive Function Instrument (CFI), a measure of subjective cognition. Participants were classified according to the number of domains (mood, lifestyle behaviours, cardiovascular health, cognitive/social engagement, and sleep, ranging from 0‐5) in which they reported at least one MRF. Age, sex, education and ethnicity were adjusted for in analyses.ResultMost participants (65%) reported at least one MRF in two or more domains (Fig 1A). Compared to participants reporting no MRFs, participants who reported at least one MRF in 3‐5 domains showed worse memory performance and reported greater subjective cognitive concerns, with magnitudes of differences moderate to large (d = 0.30‐0.93; Fig 1B). Participants who reported at least one MRF in five domains also showed worse attention than those reporting no MRFs (d = 0.58). When individual MRFs were considered simultaneously, MRFs in the mood (e.g., anxiety symptomatology) and cognitive/social engagement domains (e.g., leisure activities) were associated with worse attention and memory performance. Individual MRFs reflecting mood and sleep symptomatology were associated with greater subjective cognitive concerns.ConclusionIn middle‐aged community dwelling adults, multidomain MRFs for dementia are highly prevalent and co‐occur, and are associated with poorer cognitive outcomes. This suggests that the presence of multiple MRFs as early as midlife may have negative neurological outcomes, however, this will need to be explored in future neuroimaging studies. These findings indicate that multidomain lifestyle prevention trials in middle‐aged adults may be useful to delay or prevent future cognitive impairment or dementia.