In a large cohort of community-based older persons, greater total daily physical activity was associated with less report of disability at baseline and longitudinally, including in persons age 80 and older. The results were unchanged after adjusting for a wide variety of potential confounding variables including global cognitive function, vascular risk and disease burden, joint pain, body mass index, and depressive symptoms. In addition, the exclusion of persons with baseline dementia did not alter the results. Total daily physical activity remained associated with functional independence even after controlling for self-reported physical activities. These analyses support the link between total daily physical activity and disability and suggest an active lifestyle including high levels of not only exercise but also non-exercise physical activity may contribute to the maintenance of independence even in very old adults.
Our study of Rush Memory and Aging Project participants enhances the prior understanding of late-life physical activity and disability in three important ways. First, our study shows actigraphy measures can be obtained in community-dwelling older persons, including persons over age 80. As prior studies mainly reported the feasibility of actigraphy in children and middle-aged adults
], extending the range to persons over the age of 80 points to actigraphy being a useful performance-based measure of activity throughout the lifespan.
Second, we also were able to show that total daily physical activity is a predictor of developing ADL disability, including persons aged 80 years and older. While prevention of ADL disability is a public health target for maintaining healthy populations, achieving a 10% reduction in disability in a decade
] requires focused efforts on populations at high risk for developing disability such as persons over the age of 80. Public health efforts to encourage higher levels of total daily physical activity in this group may be useful especially since total daily physical activity remained a predictor even after controlling for other common age-related conditions associated with disability such as lower level of cognitive function. Prior work in the Rush Memory and Aging Project has highlighted the relationship between total daily physical activity and cognitive function
], and cognitive decline is predictor of disability in older persons. Our study highlights that the association between total daily physical activity and disability is not completely explained by level of cognitive function and identifies the need for further studies to examine the mechanism by which greater total daily activity is associated with less disability.
Third, we were able to show that total daily physical activity and self-report physical activity provided separate and complementary predictive information for developing disability by measuring non-overlapping facets of physical activity in older adults. Why total daily physical activity and self-report physical activity provide complementary information requires further inquiry. Self-report of physical activity over an extended period may be highly influenced by level of physical activity in the last 24 hours and may capture intermittent intensity of physical activity. Total daily physical activity assessed continuously over multiple days incorporates all physical activity and may capture the mean levels for physical activity over the entire measurement period. Accumulating evidence in young and middle age individuals suggests that any movement (goal-directed or non-goal directed) may have important health outcomes
] by increasing total daily energy expenditure
]. The results of this study extend the prior findings in younger persons to the oldest of the old. Older persons, for whom participation in formal exercise may be constrained because of underlying health problems, may nonetheless benefit from increases in their levels of non-exercise physical activities.
Given that actigraphy is inexpensive and portable, total daily physical activity may be a useful method for clinically titrating physical activity interventions in older adults at risk for developing ADL disability. In our analysis, about 1 in 6 older persons without initial disability had no self-reported physical activity and their mean total daily physical activity level was 2.5
counts/day. This information can be used to define a clinically useful two-step screening and intervention algorithm. In the first step, self-report physical activity may be utilized to quickly screen older adults without disability but at higher risk for developing disability. As self-report physical activity has limitations including recall bias, older persons with no self-report physical activity then can undergo measurement of total daily physical activity. If total daily physical activity is less than 2.5
counts/day, more efforts can be made to encourage increasing total daily physical activity. Based on the current study findings, an increase in level of total daily physical activity from 2.5
counts/day to 3.0
percentile) could potentially result in a 14% reduction in hazard of developing ADL disability. Whether an increased in total daily physical activity is achieved can be monitored using actigraphy. While possible, such an evaluation and intervention algorithm in clinical settings requires further validation.
This study has several strengths. First, concomitant measures of total daily physical activity by actigraphy and a self-report measure of physical activity were available in a large cohort of community-based persons with a high rate of follow-up participation. Second, detailed cognitive function data enabled us to examine the effect of global cognition and dementia on the relation between physical activity and disability. Third, we were able to examine the association of total daily physical activity and disability in an older community-dwelling cohort. This study also has limitations including the use of a volunteer cohort. Second, total daily physical activity as measured by actigraphy does not classify the types of activities performed. Finally, the high education level of participants may limit the generalizability of the results to persons with less than 12 years of formal schooling.