Compositional associations of time spent in sleep, sedentary behavior and physical activity with obesity measures in children

Background/objectives: The 24 h day is made up of four movement behaviors: sleep, sedentary behavior (SB), light intensity physical activity (LIPA), and moderate-to-vigorous physical activity (MVPA). Previous studies examining associations between movement behaviors and obesity have used flawed statistical methods that did not account for the co-dependent and compositional nature of movement behaviors. Our objective was to use compositional data analysis methods to investigate the relationships between movement behaviors and obesity measures among children. Subjects/methods: Participants were 434 children aged 10-13 years. They wore an accelerometer for 7 days to determine time spent in SB, LIPA, and MVPA. Sleep duration was determined from the accelerometer and a log. The obesity measures were the body mass index, waist circumference, and fat mass index. Compositional data analysis was used to estimate whether the composition of movement behaviors was associated with obesity measures, and the extent to which changing time in movement behaviors within the movement behavior composition was associated with changes in obesity measures. Results: The composition of movement behaviors across the 24 h day was associated with all three obesity measures (p < 0.001). Relative to other movement behaviors, time spent in MVPA was negatively associated with obesity measures (p < 0.01), time spent in LIPA was positively associated with obesity measures (p < 0.05), while time spent in SB and sleep were not associated with obesity measures. The estimates suggested that an 18 min/day increase in MVPA, 21 min/day decrease in LIPA, 87 min/day decrease in SB, or 67 min/day increase in sleep relative to the remaining behaviors would be associated with a 0.1 unit decrease in BMI z-score. Conclusions: The composition of movement behaviors across the day is associated with obesity measures. The findings suggest that changing the movement behavior composition by increasing MVPA and decreasing LIPA would be the most effective approach for improving obesity measures.