162 Walking, Thinking and Daily-Life Activity in Older People with Dementia

Abstract

Abstract Gait has been linked with cognition in cognitively healthy and impaired older people. However, the studies in cognitively impaired people have been small, have not contrasted the strength of cognitive domain associations and have reported inconsistent findings. We examined baseline data of 309 older people with mild-moderate cognitive impairment (age 82±6 years; 47% female) who were participating in a large fall prevention randomised controlled trial. Gait speed was measured at usual pace over 2.4m and cognitive performance was assessed with the Addenbrooke’s Cognitive Examination-III (ACE-III). The ACE-III assesses cognitive domain performance (attention; memory; verbal fluency; language; visuospatial ability). Executive function (EF) was additionally examined using the Frontal Assessment Battery (FAB). Each cognitive domain was associated with gait speed in separate models adjusted for confounders. EF (verbal fluency and the FAB) demonstrated the strongest association which withstood adjustment for attention, memory, language and visuospatial ability. In contrast, visuospatial ability was the only cognitive domain to withstand adjustment for EF (verbal fluency, not the FAB). These findings support higher-order gait regulation. Characterising individuals at risk of negative health outcomes can assist in identifying effective prevention strategies. Forty-five older people with mild-moderate dementia were age-sex matched to two (n=90) healthy controls and all participants (age 81±6 years, 42% female) wore triaxial accelerometers (MoveMonitor, McRoberts) on their lower back for 7-days. Daily-life gait quantity and quality were estimated from the MoveMonitor. Steady-state clinical walking speed was assessed at usual pace over 2.4 to 10.0m. Participants with dementia had reduced gait quantity, slower clinical, habitual daily-life and maximum daily-life walking speeds and multi-domain gait impairment compared to controls. In participants with dementia, clinical walking speed more closely represented habitual daily-life walking speed, whereas in controls, it more closely represented maximum daily-life walking speed. These findings have assessment, functional (e.g. crossing roads) and training/treatment implications.