17 CROSS-SECTIONAL AND LONGITUDINAL RELATIONSHIPS BETWEEN MOBILITY, NEUROPSYCHOLOGICAL PERFORMANCE AND FALLS IN COMMUNITY-DWELLING OLDER ADULTS: DATA FROM TUDA

Background Impaired mobility is associated with incident cognitive impairment and dementia. However, the complex bi-directional temporal relationships between subtle impairments in neuropsychological performance, mobility trajectories and falls is poorly understood. Methods Using data from the Trinity, Ulster Department of Agriculture (TUDA/TUDA5+) study, we evaluated cross-sectional and longitudinal relationships between impaired mobility, neuropsychological performance and falls using regression models adjusted for important clinical confounders. Older adults with potential cognitive impairment (Mini-Mental State Examination score <25) were excluded. Detailed neuropsychological assessment was performed using the RBANS (Repeatable Battery for Neuropsychological Assessment) and FAB (Frontal Assessment Battery). Impaired mobility was assessed using Irish population-specific age/sex/height-specific Timed-Up-and-Go (TUG) cut-offs. Results Of 4,103 participants (72.9 ± 7.9 years; 67.4% female), just under one-fifth (17.5%) met criteria for impaired mobility. Older adults with impaired mobility had significantly greater likelihood of impaired neuropsychological performance, in particular for language (OR 1.77; 1.35-2.31; p<0.001) and attention (OR 1.69; 1.37-2.08; p<0.001) domains. In 953 participants followed for a median 5.2 (IQR: 4.83-7.26) years, impaired mobility at baseline significantly predicted incident impairment in immediate memory (OR 2.56; 1.33-4.95; p<0.001). Stronger relationships were seen for impaired neuropsychological performance predicting mobility decline rather than impaired mobility predicting cognitive decline (all p<0.001). Both impaired mobility and neuropsychological performance were associated with incident falls, particularly for impairments in executive function and attention (all p<0.001). Impaired mobility in isolation had poor performance as a sole test to predict incident cognitive impairment (AUC: 0.55-0.65). Conclusion In both cross-sectional and longitudinal analyses, impaired mobility is associated with subtle impairments in neuropsychological performance. Whilst impaired neuropsychological performance was a greater predictor of impaired mobility rather than vice versa, our findings highlight the complex relationship between mobility and cognitive trajectories in older adults, emphasising the need for comprehensive cognitive and falls assessment in those presenting with new-onset subtle impairments in mobility and cognition.