The Relationship Between Accelerometer-measured Sedentary Behaviors And Physical Activity Energy Expenditure Measured By Doubly-labeled Water

Abstract

PURPOSE: The aim was to study the association between accelerometer determined total sedentary time (ST) and breaks in ST with physical activity energy expenditure (PAEE) from doubly-labeled water (DLW). Further, results based on accelerometer counts from only the vertical axis (VT) and from all three axes (VM3) were compared. METHODS: Data from 748 women and 659 men (43-83 years) were analyzed. Participants wore the Actigraph GT3X accelerometer on the hip during waking hours for seven days twice during the study about 6 months apart. Using the average of the two measurements, total ST was estimated using cut-points of 100 (VT) and 200 (VM3) counts/min. A break in ST was defined as at least one minute of a higher activity count within a sedentary bout. PAEE (kcal/day) was estimated by subtracting resting metabolic rate and the thermic effect of food from total energy expenditure derived from one DLW measurement, with a small subsample having a repeat. Partial Spearman correlations and linear regression models were used to determine associations between total ST and breaks per sedentary hour with PAEE. Correlations and models were adjusted for age, weight, fat free mass, accelerometer wear time and moderate-to-vigorous physical activity (MVPA). RESULTS: Mean total ST using VT and VM3 were 9.0 and 7.9 h/day of wear time, respectively, in women and 9.3 and 8.5 h/day in men. Corresponding mean breaks per sedentary hour were 10.4 and 11.0 in women and 9.2 and 9.4 in men. Partial Spearman correlation coefficients between ST and PAEE ranged from -0.34 to -0.44 and correlations between breaks per sedentary hour and PAEE from 0.27 to 0.34. After adjustment for within-person variation in PAEE, correlations for ST ranged from -0.38 to -0.68 and correlations for breaks per sedentary hour from 0.35 to 0.51. In adjusted models, each hour of ST was associated with lower PAEE (β= -66 to -103 kcal, p < .0001) and each break per sedentary hour was associated with higher PAEE (β= 21 to 39 kcal, p < .0001). Estimates were slightly higher for VT vs. VM3. CONCLUSIONS: This study found that total ST was inversely associated and breaks in ST was positively associated with PAEE in both men and women independent of body composition and MVPA. Replacing ST with light physical activity and more frequent movement throughout the day may have an impact on an individual’s PAEE.