Balancing time use for children's fitness and adiposity: Evidence to inform 24-hour guidelines for sleep, sedentary time and physical activity

Abstract

Purpose: Daily time spent on one activity cannot change without compensatory changes in others, which themselves may impact on health outcomes. Optimal daily activity combinations may differ across outcomes. We estimated optimal daily activity durations for the highest fitness and lowest adiposity.

Methods: Cross-sectional Child Health CheckPoint data (1182 11-12-year-olds; 51% boys) from the population-based Longitudinal Study of Australian Children were used. Daily activity composition (sleep, sedentary time, light physical activity [LPA], moderate-to-vigorous physical activity [MVPA]) was from 8-day, 24-hour accelerometry. We created composite outcomes for fitness (VO2max; standing long jump) and adiposity (waist-to-height ratio; body mass index; fat-to-fat-free log-ratio). Adjusted compositional models regressed activity log-ratios against each outcome. Best activity compositions (optimal time-use zones) were plotted in quaternary tetrahedrons; the overall optimal time-use composition was the center of the overlapping area.

Results: Time-use composition was associated with fitness and adiposity (all measures p<0.001). Optimal time use differed for fitness and adiposity. While both maximized MVPA and minimized sedentary time, optimal fitness days had higher LPA (3.4 h) and shorter sleep (8.25 h), but optimal adiposity days had lower LPA (1.0 h) and longer sleep (10.9 h). Balancing both outcomes, the overall optimal time-use composition was (mean [range]): 10.2 [9.5; 10.5] h sleep, 9.9 [8.8; 11.2] h sedentary time, 2.4 [1.8; 3.2] h LPA and 1.5 [1.5; 1.5] h MVPA.

Conclusion: Optimal time use for children's fitness and adiposity involves trade-offs. To best balance both outcomes, estimated activity durations for sleep and LPA align with, but for MVPA exceed, 24-h guidelines.

Fig 1. Participant flow.

LSAC = Longitudinal Study of Australian Children; zBMI = body mass index z-score; ilr = isometric log ratio; VO_2max = predicted maximal aerobic power; SEP = socioeconomic position.

Fig 2. Relationship between incrementally increasing durations of individual activity behaviors and fitness/adiposity composite summary z-score.

Higher z-scores represent better outcomes. Adjusted for age, sex, puberty and socioeconomic position. Fitness additionally adjusted for body mass index z-score and included quadratic term for the activity composition. Note: Jitter was applied to data points to enable visualization of overlapping points. Each data point represents one of the possible permutations of activity compositions (in 10-min increments) within the study sample’s empirical activity footprint (i.e., the ranges of activity durations observed in the sample truncated at ±3SD of univariate distributions of behaviors). Care must be taken when interpreting the relationship between individual activity behaviors and outcomes. Although we can describe the shape of relationships in terms of individual behaviors (e.g., MVPA is beneficially associated with outcomes), this description pertains to the average situation only, as shown by the loess line. There is substantial variation around this line because the observed relationship with the activity in question (e.g., MVPA) depends on the values of the remaining activity behaviors (e.g., sleep, sedentary time and LPA). LPA = light physical activity; MVPA = moderate-to-vigorous physical activity.

Fig 3. Estimated best zones for fitness (orange) and adiposity (blue) corresponding to time-use compositions associated with the best 15% of each outcome.

Red denotes the best overall zone, where fitness and adiposity best zones overlap. Models adjusted for sex, age, puberty and household socioeconomic status. Fitness models additionally adjusted for body mass index z-score and included a quadratic term for the activity composition. The four plots show different rotations/perspectives of the same 3-D tetrahedron. Activities are at 100% (24 hours) at the corresponding apices of the tetrahedron, and 0% at the opposite base. A datapoint in the exact center of the tetrahedron would have equal shares of each activity (25%, or 6 hours). The activity in brackets is in the foreground. For an interactive 3-D plot, please see S3 File. The compositional mean of the overlap zone between the polygons (shown in red) is indicated by the black dot (h/day): Sleep = 10.2; Sedentary = 9.9; LPA = 2.4; MVPA = 1.5. LPA = light physical activity, MVPA = moderate-to-vigorous physical activity.