Associations of 24-H Movement Behavior Composition with Estimated Cardiorespiratory Fitness in School-Aged Children: A Compositional Data Analysis

Background/Objectives: This study aimed to examine the association between 24-h movement behavior composition and estimate cardiorespiratory fitness in school-aged children using compositional data analysis, and to model the theoretical differences in estimated cardiorespiratory fitness associated with isotemporal reallocations of time between movement behaviors. Methods: A cross-sectional study was conducted in 222 schoolchildren aged 8 to 12 years (mean age 9.94 ± 0.69 years), with most participants aged 10 years. Twenty-four-hour movement behaviors were assessed objectively using wrist-worn accelerometers, and cardiorespiratory fitness was estimated from the 20 m shuttle run test using the Léger equation. Daily time-use composition was analyzed using isometric log-ratio coordinates and adjusted linear regression models were fitted. Estimated differences in cardiorespiratory fitness associated with 30-min isotemporal reallocations between behaviors were then modeled. Results: The 24-h movement behavior composition was significantly associated with estimated cardiorespiratory fitness. In isotemporal models, reallocating 30 min from sedentary behavior to sleep was associated with the largest modeled difference in estimated cardiorespiratory fitness, whereas other reallocations showed smaller estimated differences depending on the behavior displaced. Age was positively associated with estimated cardiorespiratory fitness, while sex showed a limited association. Bivariate analyses revealed weak or inconsistent associations, supporting the value of the compositional approach for capturing the interdependent nature of daily time use. Conclusions: Twenty-four-hour movement behavior composition was associated with estimated cardiorespiratory fitness in school-aged children. These findings support the use of compositional approaches to examine sleep, sedentary behavior, and physical activity jointly. However, given the cross-sectional design and the modeled nature of the reallocations, the estimated differences should be interpreted cautiously and not as direct causal or physiological effects.