The association between physical activity and cognition: the Baltimore Experience Corps Trial

Abstract Background Because of population aging, the number of people with AD is expected to triple (11-16 million) by 2050. Over the past decade pharmaceutical interventions focused on AD have had minimal success in targeting the underlying pathology of the disease and modifying disease course. Because of these difficulties, research focus has recently expanded from disease modifying pharmaceutical therapies to lifestyle interventions. Physical activity in particular may have beneficial effects on cognition and brain health, however older adults – especially those with socio-demographic risk factors – are becoming increasingly inactive and have difficulty initiating and adhering to exercise programs. Understanding and evaluating the nature of the benefits of low-intensity activity, the characteristics of dose required to achieve health benefits, and the mechanisms of change particularly within low socio-economic status (SES) populations is essential to designing sustainable lifestyle interventions that can be easily and effectively integrated into communities. The aims of this dissertation are to 1) better characterize and understand the relationship between objectively measured low-intensity physical activity and cognition; 2) inform the feasibility of a civic-engagement volunteer service program to increase physical activity in a socio-demographically at risk population; and 3) explore the potential of objective physical activity devices to be used as biometric signals of disability. Methods The study sample is from the Baltimore Experience Corps Trial (BECT), an intention-to-treat, randomized, controlled effectiveness trial, and the Brian Health Study (BHS), a substudy within the BECT. Physical activity was measured using both objective and self-report measures. Cognitive measures included domain specific psychometric measures and neuroimaging measures obtained from structural brain imaging. Physical function measures included both performance-based and self-report measures. Results Objective measures of physical activity within the low-intensity range were associated in cross-section with physical function and both behavioral and structural measures of brain structure. These objective measures, which are important in order to characterize physical activity and change particularly within the low-intensity range, captured increases in total and low-intensity walking activity associated with the intervention in women within the BECT, and a decrease in moderate- to vigorous-intensity walking activity in men. Self-report measures of physical activity did not sensitively capture low-intensity physical activity or the relationship between physical activity and cognition; however, as expected, sensitively captured moderate- to vigorous physical activity and its association with physical function. Self-report measures additionally indicated declines across both intervention arms in women within the BECT and marginal negative intervention effects particularly in moderate- to vigorous-intensity physical activity, and captured declines in the control group for men, and positive intervention effects at 24 months. Objective measures are extremely useful in order to sensitively capture physical activity, and can be used in order to better measure and understand the relationship between physical function/ activity and cognition