The Association Between Four Gait Speed Assessments and Incident Stroke in Older Adults: The Health, Aging and Body Composition Study

OBJECTIVE: To examine the association between gait speed and incident stroke and compare the predictive value between four gait speed assessments (6-meter, 20-meter, 2-min, and 400-meter). DESIGN: Prospective cohort study. SETTING: 1,779 older adults from the Health, Aging and Body Composition study. All participants had no history of cardiovascular or cerebrovascular disease at baseline. METHODS: We used Cox proportional hazards regression model to identify the relationship between each of four gait speed assessment and incident stroke. We used the c-statistic, Akaike information criterion (AIC), and Bayesian information criterion (BIC) to compare the predictive validity between four measures. RESULTS: 176 (9.9%) had incident stroke during an average 10.3-year follow-up. After multivariable adjustment, hazard ratio of incident stroke was 0.89 (95% CI: 0.82-0.97), 0.90 (95%CI: 0.82-0.98), 0.88 (95% CI: 0.80-0.97), and 0.86 (95% CI: 0.78-0.95) for 6-meter, 20-meter, 2-min, and 400-meter test, respectively. We found only negligible difference in the c-statistic between four gait speed assessments (range: 0.66-0.67). Similarly, we did not observe huge difference in AIC or BIC between four assessments. CONCLUSIONS: Gait speed was independently associated with stroke among older adults. Different gait speed assessments had similar prognostic value for predicting stroke.