Structured-unstructured activity behaviours on preschoolers’ mental and physical health: a compositional and isotemporal reallocation analysis
Participants information and covariate screening
After statistical analysis of participants’ information (Table 1), it was found that sex was a relevant factor for hyperactivity (t = 2.36, P = 0.02), externalizing difficulties (t = 2.34, P = 0.02), pro-sociality (t=-2.11, P = 0.04), height (t = 2.18, P = 0.03), weight (t = 1.16, P = 0.01), BMI (t = 2.15, P = 0.03), grip strength (t = 5.99, P < 0.001), standing long jump (t = 3.35, P = 0.001), and sitting forward bending (t=-3.94, P < 0.001). Therefore, sex will be used as a covariate for the above outcome indicators when establishing a linear regression model in the follow-up.
The area is a relevant factor for emotional symptoms (t=-2.05, P = 0.04), pro-sociality (t = 3.31, P = 0.001), grip strength (t=-5.34, P < 0.001), 15-metre obstacle running (t = 4.05, P < 0.001), walking on the balance beam (t = 2.99, P = 0.003), and physical fitness score (t=-2.21, P = 0.03). Therefore, area will be used as a covariate for the above outcome indicators when establishing a linear regression model in the follow-up.
Age is a relevant factor for height (F = 0.38, P < 0.001), weight (F = 35.44, P < 0.001), grip strength (F = 25.36, P < 0.001), standing long jump (F = 72.29, P < 0.001), two-legged continuous jump (F = 29.20, P < 0.001), 15-metre obstacle running (F = 40.95, P < 0.001), walking on the balance beam (F = 17.77, P < 0.001), body shape score (F = 4.19, P = 0.006), and physical fitness score (F = 4.48, P = 0.004). Therefore, age will be used as a covariate for the above outcome indicators when establishing a linear regression model in the follow-up.
The mental level is a relevant factor for emotional symptoms (F = 134.93, P < 0.001), peer problems (F = 33.37, P < 0.001), internalizing difficulties (F = 156.07, P < 0.001), conduct problems (F = 77.42, P < 0.001), hyperactivity (F = 35.16, P < 0.001), externalizing difficulties (F = 87.96, P < 0.001), and SDQ-Total-Difficulties (F = 208.78, P < 0.001). Therefore, mental level will be used as a covariate for the above outcome indicators when establishing a linear regression model in the follow-up.
Compositional analysis result
Dispersion analysis result
According to the variation matrix (Table 2), In structured PA, the structured MVPA and structured SB have the largest variance (2.52), indicating that these two activities have low correlation; while the structured MVPA and structured LPA have the smallest variance (0.72), indicating that these two activities have high correlation. In unstructured PA, the unstructured MVPA and unstructured SB have the largest variance (2.04), indicating that these two activities have low correlation; while the unstructured MVPA and unstructured LPA have the smallest variance (0.54), indicating that these two activities have high correlation.
Sleep exhibits high variance with other activity behaviours, therefore, sleep behaviour has low correlation with other activities and is not easily influenced.
Compositional analysis
Based on the previous screening of covariates, we used different adjustment variables for different outcome indicators. The final multiple linear regression models established all had significant model fitness (except BMI) (Table 3).
The results showed that in structured activities, structured MVPA was positively correlated with emotional symptoms (β=-0.74, P = 0.01), internalising difficulties (β=-0.87, P = 0.03), height (β = 2.89, P = 0.01), and body shape score (β = 2.17, P = 0.01). Structured LPA was negatively correlated with emotional symptoms (β = 1.09, P = 0.01), internalising difficulties (β = 1.52, P = 0.01), total difficulties (β = 1.74, P = 0.04), height (β=-3.53, P = 0.01), body shape score (β=-2.92, P = 0.008), and 15-meter obstacle running (β = 0.72, P = 0.04). Structured SB was negatively correlated with pro-sociality (β=-1.61, P = 0.04), height (β=-2.60, P = 0.04), grip strength (β=-1.54, P = 0.01), standing long jump (β=-10.53, P = 0.01), seating forward bending (β=-3.73, P = 0.01), and physical fitness score (β=-6.43, P = 0.01).
In unstructured activities, unstructured MVPA was positively correlated with emotional symptoms (β=-0.92, P = 0.01) and internalising difficulties (β=-1.24, P = 0.01). Unstructured LPA was negatively correlated with emotional symptoms (β = 0.90, P = 0.03). Unstructured SB was negatively correlated with emotional symptoms (β = 1.12, P < 0.001), internalising difficulties (β = 1.55, P < 0.001), height (β=-3.60, P = 0.003), body shape score (β=-2.24, P = 0.01), and standing long jump (β=-10.54, P = 0.01).
Isotemporal reallocation analysis result
Changes in preschooliers’ mental and physical as a result of 10-min isotemporal reallocation
Table 4 presents the estimated differences in mental and physical outcomes for reallocations of 10 min between time-use behaviours. Replacing 10 min of structured SB (SMD=-0.55, 95%CI: -1.08 to -0.01; SMD = 0.91, 95%CI: 0.18 to 1.64), structured LPA (SMD=-1.28, 95%CI: -2.30 to -0.27; SMD = 1.76, 95%CI: 0.37 to 3.15), unstructured SB (SMD=-0.63, 95%CI: -1.17 to -0.08; SMD = 0.96, 95%CI: 0.21 to 1.71), unstructured LPA (SMD=-0.66, 95%CI: -1.18 to -0.14; SMD = 1.09, 95%CI: 0.37 to 1.81), and sleep (SMD=-0.60, 95%CI: -1.14 to -0.06; SMD = 0.93, 95%CI: 0.19 to 1.67) with 10 min of structured MVPA can significantly improve internalizing difficulties and body shape in preschoolers.
Replacing 10 min of structured LPA with 10 min of unstructured MVPA can effectively improve internalizing difficulties (SMD=-0.80, 95%CI: -1.31 to -0.29), total difficulties (SMD=-0.94, 95%CI: -1.76 to -0.12), body shape (SMD = 0.99, 95%CI: 0.30 to 1.69), and physical fitness (SMD = 1.52, 95%CI: 0.06 to 2.98) in preschoolers.
The dose-response benefits of reallocating structured MVPA time to other physical activity behaviours
Among the significant results based on 10-minute reallocations, structured moderate-to-vigorous physical activity (MVPA) was found to be more effective in reallocating time compared to other reallocation models. Therefore, models were constructed sequentially with a 5-minute increment up to 30 min of reallocation, to obtain the health dose-response benefits of replacing other activities with structured MVPA time.
As shown in Fig. 3, the time reallocation of structured MVPA is associated with lower internalizing difficulties and better body shape scores. The more time other activities allocate to structured MVPA, the better the improvement effect. Among them, increasing structured MVPA time and reducing structured LPA time have significantly better effects on improving internalizing difficulties and body shape scores of preschoolers than other activity reallocations. Moreover, when the reallocation time reaches 25 min, there is a more significant improvement in benefits.
On the contrary, if structured MVPA is replaced by other activities, there is a downward trend in internalizing difficulties and body shape scores. Moreover, this deterioration occurs faster than the benefits gained from increasing structured MVPA time.
link