Association of total daily physical activity with disability in community-dwelling older persons: a prospective cohort study

Background: Based on findings primarily using self-report measures, physical activity has been recommended to reduce disability in old age. Collecting objective measures of total daily physical activity in community-dwelling older adults is uncommon, but might enhance the understanding of the relationship of physical activity and disability. We examined whether greater total daily physical activity was associated with less report of disability in the elderly. Methods: Data were from the Rush Memory and Aging Project, a longitudinal prospective cohort study of common, age-related, chronic conditions. Total daily physical activity was measured in community-dwelling participants with an average age of 82 using actigraphy for approximately 9 days. Disability was measured via self-reported basic activities of daily living (ADL). The odds ratio and 95% Confidence Interval (CI) were determined for the baseline association of total daily physical activity and ADL disability using a logistic regression model adjusted for age, education level, gender and self-report physical activity. In participants without initial report of ADL disability, the hazard ratio and 95% CI were determined for the relationship of baseline total daily physical activity and the development of ADL disability using a discrete time Cox proportional hazard model adjusted for demographics and self-report physical activity. Results: In 870 participants, the mean total daily physical activity was 2. 9 x 10(5) counts/day (range in 10(5) counts/day = 0.16, 13. 6) and the mean hours/week of self-reported physical activity was 3.2 (SD = 3.6). At baseline, 718 (82.5%) participants reported being independent in all ADLs. At baseline, total daily physical activity was protective against disability (OR per 10(5) counts/day difference = 0.55; 95% CI = 0.47, 0.65). Of the participants without baseline disability, 584 were followed for 3.4 years on average. Each 10(5) counts/day additional total daily physical activity was associated with reduced hazard of developing disability by 25% (HR = 0.75, 95% CI = 0.66, 0.84). The results were unchanged after controlling for important covariates including cognition, depressive symptoms, and chronic health conditions. Conclusions: Greater total daily physical activity is independently associated with less disability even after controlling for self-reported physical activity.