Rest Activity Rhythms and their Association with Cardiorespiratory Fitness and Walking Energetics in Older Adults: Study of Muscle, Mobility and Aging

Purpose It is recognized that disruptions in circadian behavior, such as with shift work or jet lag, are associated with diminished health. This known relationship implies that people with stronger indices of circadian behavior will exhibit improved physiology. To address the association between rhythmic activity behavior and physiology, we proposed that metrics indicative of "more rhythmic"rest-activity patterns would be associated with better cardiorespiratory fitness and walking energetics in a cohort of older adults. Methods Using baseline data from the Study of Muscle, Mobility and Aging (N = 799, age: 76 ± 5 yr, 58% female), we quantified metrics describing rhythmic aspects of rest-activity behavior (amplitude, robustness, time of peak activity, others) from continuous wrist-worn accelerometry. We used linear models to examine cross-sectional associations between rhythmic metrics with VO2peak and walking energetics (cost-capacity ratio at slow and preferred walking speeds) adjusted for age, sex, race, height, health conditions, and other factors. Results Metrics that reflect more rhythmic behavior were associated with VO2peak (higher amplitude: Q1: 18.4 vs Q4: 22.0 mLkg-1min-1; P-trend < 0.001) higher pseudo-F statistic/robustness (Q1: 19.2 vs Q4: 21.3 mLkg-1min-1; P-trend < 0.001), and earlier time of peak activity (Q1 (earliest): 20.9 vs Q4 (latest): 19.2 mLkg-1min-1; P-trend < 0.001). Similar trends were observed with lower cost-capacity ratio at preferred and slow walking speeds (amplitude, pseudo-F statistic, acrophase: P-trend < 0.001 for all). Conclusions More rhythmic activity behavior and earlier time of peak activity were associated with better cardiorespiratory fitness and walking energetics. These findings support the framework that rhythmic activity supports healthy physiology. Further investigations are warranted to determine if declines in rhythmicity of human behavior are predictive of disease.