# Table 3 Differences for change in children’s in-care sedentary time and physical activity (follow-up vs. baseline) between groups (accreditation vs. control)

Outcome variables Groups Baseline Follow-up Follow-up vs. baseline Group*time
Model 1
Estimated marginal means (SE) Mean difference (SE) B (95%CI)a P value
Sedentary time (min/h) Accreditation (n = 126) 31.87 (0.43) 30.45 (0.47) −1.41 (0.50) − 0.06(− 1.39,1.27) 0.802
Control (n = 126) 31.96 (0.43) 30.61 (0.42) −1.35 (0.46)
LPA (min/h) Accreditation (n = 126) 21.23 (0.30) 21.78 (0.33) 0.55 (0.36) −0.03(− 0.99,0.93) 0.953
Control (n = 126) 22.00 (0.30) 22.57 (0.30) 0.58 (0.33)
MVPA (min/h) Accreditation (n = 126) 6.90 (0.21) 7.78 (0.28) 0.88 (0.27) 0.01(−0.09,0.11)b 0.802
Control (n = 126) 6.04 (0.21) 6.81 (0.25) 0.77 (0.24)
Model 2
Estimated marginal means (SE) Mean difference (SE) B (95%CI)a P value
Sedentary time (min/h) Accreditation (n = 120) 31.97 (0.48) 30.72 (0.52) −1.25 (0.55) −0.07(−1.43,1.29) 0.918
Control (n = 121) 31.71 (0.47) 30.53 (0.48) −1.18 (0.50)
LPA (min/h) Accreditation (n = 120) 21.41 (0.34) 22.12 (0.37) 0.71 (0.39) 0.08(−0.89,1.05) 0.873
Control (n = 121) 21.75 (0.33) 22.39 (0.34) 0.64 (0.36)
MVPA (min/h) Accreditation (n = 120) 6.65 (0.23) 7.23 (0.29) 0.58 (0.29) 0.01(−0.09,0.11)b 0.891
Control (n = 121) 6.49 (0.23) 7.03 (0.27) 0.54 (0.26)
1. Abbreviations: B unstandardized beta coefficient, SE Standard error, CI Confidence interval, LPA Light-intensity physical activity, MVPA Moderate-to vigorous-intensity physical activity
2. In model 1, the clustering effect of childcare centers and the repeated effect of time were included as random intercepts, and time, group, and time*group were included as fixed effects
3. In model 2, the clustering effect of childcare centers and the repeated effect of time were included as random intercepts, and time, group, time*group, and covariates (child age, child race/ethnicity, parental education, mean precipitation) were included as fixed effects
4. aEstimate for the interaction term between group (accreditation vs. control) and time (follow-up vs. baseline) and can be interpreted as the differences in change between groups for each outcome variable
5. bGiven the non-normal distribution of data for MVPA, the unstandardized beta coefficient was calculated for Log (MVPA)