Replacing as little as 1 min of MVPA, LPA, or SED with sleep was associated with significant reductions in emotional and behavioural problems in Chinese preschoolers, while increasing sleep by reducing any other movement behaviour was consistently associated with improvements across all SDQ subscales.
Key Findings
Results
Replacing 1 minute of MVPA, LPA, or SED with sleep was associated with significant reductions in total difficulties scores in Chinese preschoolers.
Sample consisted of 1316 children aged 3-4 years recruited through multistage stratified cluster sampling across seven major administrative regions in China.
Beta coefficients for total difficulties when replacing MVPA, LPA, or SED with sleep were all β = -0.01.
95% CIs: βMVPA = -0.01 (95%CI -0.02 to -0.01); βLPA = -0.01 (95%CI -0.01 to -0.001); βSED = -0.01 (95%CI -0.02 to -0.01).
Compositional isotemporal substitution analysis was used to predict changes in EBPs due to reallocating time among 24-h movement behaviours.
EBPs were evaluated using the parent-rated Strengths and Difficulties Questionnaire (SDQ).
Results
Replacing 1 minute of MVPA, LPA, or SED with sleep was associated with significant reductions in externalising problems.
βMVPA = -0.01 (95%CI -0.01 to -0.002); βLPA = -0.003 (95%CI -0.01 to -0.0003); βSED = -0.005 (95%CI -0.01 to -0.002).
All three substitutions (from MVPA, LPA, and SED to sleep) yielded statistically significant results for externalising problems.
Externalising problems were assessed as a subscale of the SDQ.
Results
Reallocating time from MVPA or SED to sleep was associated with significantly reduced internalising problems, but reallocating from LPA to sleep was not significant.
βMVPA = -0.01 (95%CI -0.01 to -0.002); βSED = -0.01 (95%CI -0.01 to -0.004) for internalising problems.
The substitution of LPA with sleep did not yield a statistically significant result for internalising problems.
Internalising problems were assessed as a subscale of the SDQ.
Results
A significant predicted improvement in prosocial behaviour was observed only when time was reallocated from SED to sleep.
β = 0.002 (95%CI 0.0002 to 0.004) for prosocial behaviour when reallocating from SED to sleep.
Substitutions from MVPA or LPA to sleep were not significantly associated with prosocial behaviour.
Prosocial behaviour was assessed as a subscale of the SDQ.
Results
Transferring time from MVPA or SED to LPA was associated with decreases in internalising problems and conduct problems.
For internalising problems: βMVPA = -0.004 (95%CI -0.01 to -0.0002); βSED = -0.004 (95%CI -0.01 to -0.002).
For conduct problems: βMVPA = -0.002 (95%CI -0.004 to -0.0003); βSED = -0.01 (95%CI -0.002 to -0.0002).
LPA was measured using 24-h accelerometry over five consecutive days.
Results
Increasing LPA at the expense of sleep was associated with worsening total difficulties and externalising problems.
Reallocating time from sleep to LPA was significantly associated with increases in total difficulties and externalising problems.
This finding indicates that the benefit of LPA must be balanced against maintaining sufficient sleep duration.
Sleep duration was parent-reported rather than accelerometer-measured.
Methods
The study used compositional multiple linear regression and compositional isotemporal substitution to analyse 24-hour movement behaviour data in a nationally representative sample of Chinese preschoolers.
1316 children aged 3-4 years were recruited as part of the Chinese cohort of the SUNRISE International Study of Movement Behaviors in the Early Years.
Recruitment used multistage stratified cluster sampling in urban and rural areas across seven major administrative regions in China.
MVPA, LPA, and SED were measured using 24-h accelerometry over five consecutive days; sleep was parent-reported.
Compositional data analysis treats the four 24-h movement behaviours (MVPA, LPA, SED, sleep) as parts of a whole that sum to 24 hours.
What This Means
This research suggests that how preschool-aged children (3-4 years old) spend their time across a full 24-hour day — including physical activity, sedentary time, and sleep — is meaningfully linked to their emotional and behavioural wellbeing. Using data from over 1,300 Chinese children as part of a large international study, researchers found that even small increases in sleep time (as little as 1 minute), taken from any other activity including moderate-to-vigorous physical activity, light physical activity, or sitting/screen time, were consistently associated with fewer behavioural and emotional problems such as conduct issues, emotional difficulties, and overall problem scores.
The study also found that replacing time spent sitting or doing vigorous activity with lighter physical activity (like walking or playing gently) was linked to fewer internalising problems (such as anxiety or emotional issues) and fewer conduct problems. However, there was a notable trade-off: increasing light physical activity by cutting into sleep time was associated with more behavioural difficulties, not fewer. This suggests that the benefits of active play depend on children also getting enough sleep, and that sacrificing sleep for more activity may be counterproductive.
This research suggests that for preschoolers, sleep may be the most consistently beneficial component of the daily routine for emotional and behavioural health. It highlights the importance of considering the whole day — not just encouraging physical activity in isolation — when thinking about children's development. Promoting a balance between diverse light physical activity opportunities and sufficient sleep, rather than maximising any single behaviour, appears to be key for supporting young children's emotional and behavioural wellbeing.
Cheng T, Zhang Y, Zhang M, Lu Z, Zhou J, Xu M, et al.. (2026). Effects of 24-h Movement Behaviours on Emotional and Behavioural Problems in Chinese Preschoolers Using Compositional Data Analysis.. Child: care, health and development. https://doi.org/10.1111/cch.70239