Compositional Isotemporal Substitution Analysis of 24‐h Movement Behaviours and Speed Capability in Preschoolers

ABSTRACT Background Speed capability is critical for early childhood development, but troubling patterns are emerging in the motor fitness of Chinese preschoolers (3–6 years). This study investigated how compositional 24‐h movement behaviours (sleep, sedentary behaviour [SB], light physical activity [LPA] and moderate‐to‐vigorous physical activity [MVPA]) relate to speed capability. Methods Via compositional data analysis and isotemporal substitution modelling, we assessed relationships between 24‐h movement behaviours (sleep, SB, LPA and MVPA) and speed capability in 275 preschoolers (mean age 4.98 ± 0.76 years). Participants completed 20‐m sprint tests and 7‐day accelerometry. Time‐reallocation effects were quantified through pairwise behavioural substitutions (5‐ to 30‐min durations), with all models adjusted for age, sex and BMI z scores ( z ‐BMI). Results Higher relative MVPA time significantly predicted faster sprint times ( β = −1.302, p < 0.001), while higher LPA predicted slower times ( β = 1.570, p = 0.003). Reallocating 15 min from sleep, SB or LPA to MVPA reduced sprint times by 0.176, 0.201 and 0.385 s, respectively (all p < 0.05). Conversely, reallocating MVPA to other behaviours worsened performance. The effects exhibited asymmetry: displacing time away from MVPA impaired speed capability to a greater extent than equivalent gains in MVPA time improved it. Conclusion MVPA is the strongest positive predictor of speed capability in preschoolers. Optimizing 24‐h movement patterns by reallocating time from LPA or SB to MVPA is associated with enhanced speed performance, supporting targeted interventions for early childhood development.