A Prospective Study of Gait Performance and Subsequent Cognitive Decline in a General Population of Older Japanese

Gait speed is a good predictor of cognitive decline in later life. However, it is not known whether step length or step frequency is better for predicting such decline. This study aimed to compare the predictive value for cognitive decline of gait performance measures in a population of older Japanese. Among 853 cognitively intact adults aged 70 years or older who participated in a baseline survey, 666 (mean age, 75.5 [SD 4.4] years; women, 59.3%) were reevaluated at least once during the subsequent 4-year period. Cognition was assessed by Mini-Mental State Examination (MMSE), and cognitive decline was defined as a decrease of three points or more on the Mini-Mental State Examination during follow-up. During a median follow-up of 2.7 years, 110 adults (16.5%) had cognitive decline. Among the measures of gait performance (speed, step length, and frequency), step length was the most predictive of cognitive decline. After controlling for important confounders, older men in the lowest and middle tertiles of step length at maximum speed and older women in the lowest and middle tertiles of step length at usual speed were 4.42 (95% confidence interval: 1.65-11.8), 2.17 (0.82-5.71), 5.76 (2.15-15.4), and 2.44 (0.94-6.35) times as likely to develop cognitive decline, respectively, as those of the same sex and walking speed who were in the highest tertile. Step length was an independent predictor of cognitive decline in a general population of older adults and may be a better predictor than overall gait speed of such decline.