Cognitive and physical activities are associated with cognitive resilience in a memory clinic cohort

Lifestyle factors such as participation in cognitively stimulating activities and physical activity are hypothesized to foster neural connections and enhance resilience, thereby attenuating cognitive loss in the context of Alzheimer's disease (AD) and other neurodegenerative diseases. Nonetheless, the relationship between these factors and important clinical outcomes of cognition and brain atrophy is not well understood. We assessed cognitive and physical activity levels in a large, tertiary memory clinic cohort with various clinical and aetiological diagnoses. Furthermore, we investigated whether cognitive and physical activities relate to resilience against brain atrophy across the AD continuum. In the memory clinic-based Amsterdam Dementia Cohort (ADC), 4033 individuals completed the self-reported questionnaires to quantify their engagement in cognitive (lifetime, past and current) and physical (current) activities. Firstly, we examined differences in activity scores across diagnostic groups [i.e. Alzheimer's and non-Alzheimer's types of dementia, mild cognitive impairment (MCI) and subjective cognitive decline (SCD)] using linear models adjusted for age and sex. Secondly, in a subset on the AD continuum (i.e. amyloid-β-positive with SCD, MCI or Alzheimer's dementia; n = 904), we used linear mixed-effects models adjusted for age and sex to assess whether cognitive and physical activities had an interactive or additive effect on concurrent cognition and rate of decline (global cognition, memory and executive functioning) at a given level of magnetic resonance imaging-based temporoparietal brain atrophy. We also tested associations with clinical progression and with mortality using Cox survival models. Lifetime and current cognitive activity, and current physical activity were generally lower in more cognitively unimpaired groups (all P < 0.001), while differences in past cognitive activity between diagnostic groups were not significant (P = 0.08). Within the AD continuum, at similar levels of temporoparietal atrophy, higher cognitive and physical activities were associated with better cognition at baseline (past cognitive activity: β_Std = 0.15-18, P _FDR < 0.001; and physical activity: β_Std = 0.9-0.11, P _FDR < 0.05). In longitudinal analyses, neither factor was related to cognitive decline nor clinical progression. Current cognitive activity [hazard ratio (HR) = 0.82 (0.73-0.92), P _FDR < 0.001] and physical activity [HR = 0.88 (0.79-0.99), P _FDR < 0.05] were associated with reduced mortality risk in the total sample, while past cognitive activity was linked to reduced mortality only in MCI [HR = 0.54 (0.36-0.8), P _FDR < 0.01]. While associations between current cognitive and physical activities with better concurrent cognitive performance might be (partially) explained by reverse causality, the observed effects of past cognitive activity suggest that early and mid-life participation in cognitively stimulating activities may provide a cognitive benefit once AD manifests.