Walking Speed, Processing Speed, and Dementia: A Population-Based Longitudinal Study

Abstract

Slow walking speed has been shown to predict dementia. We investigated the relation of walking speed, processing speed, and their changes over time to dementia among older adults. This study included 2,938 participants (age 60+ years) in the population-based Swedish National study on Aging and Care in Kungsholmen, Sweden, who were free from dementia and severe walking impairment at baseline. Walking speed was assessed with participants walking at their usual pace and processing speed was defined by a composite measure of standard tests (digit cancellation, trail making test-A, pattern comparison). Dementia at 3- and 6-year follow-ups was diagnosed according to Diagnostic and Statistical Manual of Mental Disorders-IV criteria. Of the 2,232 participants who were reassessed at least once, 226 developed dementia. Logistic regression models showed that each standard deviation slower baseline walking speed or decline in walking speed over time increased the likelihood of incident dementia (odds ratios 1.61, 95% confidence interval [CI] 1.31-1.98; and 2.58, 95% CI 2.12-3.14, respectively). Adjustment for processing speed attenuated these associations (odds ratios 1.26, 95% CI 1.01-1.58 and 1.76, 95% CI 1.33-2.34). Mixed-effects models revealed statistical interactions of time with dementia on change in walking and processing speed, such that those who developed dementia showed accelerated decline. At baseline, poorer performance in processing speed, but not in walking speed, was observed for persons who developed dementia during the study period. Processing speed may play an important role for the association between walking speed and dementia. The slowing of walking speed appears to occur secondary to slowing of processing speed in the path leading to dementia.