| | Model 1 | Model 2 | Model 3 |
|---|
(SB/LPA + SB/MVPA) difference | (LPA/MVPA) difference | (LPA/SB + LPA/MVPA) difference | (SB/MVPA) difference | (MVPA/SB + MVPA/LPA) difference | (SB/LPA) difference |
|---|
β (95% CI) | p | β (95% CI) | p | β (95% CI) | p | β (95% CI) | p | β (95% CI) | p | β (95% CI) | p |
|---|
Baseline BMI (kg/m2) | 0.002 (− 0.014, 0.018) | 0.848 | 0.009 (− 0.013, 0.031) | 0.434 | 0.010 (− 0.013, 0.033) | 0.40 | 0.009 (− 0.011, 0.029) | 0.382 | −0.011 (− 0.033, 0.012) | 0.342 | − 0.004 (− 0.017, 0.010) | 0.604 |
Change in BMI (kg/m2) | 0.042 (0.011, 0.073) | 0.009 | 0.045 (−0.008, 0.098) | 0.095 | 0.02 (−0.036, 0.075) | 0.487 | 0.057 (0.009, 0.105) | 0.021 | −0.059 (− 0.115, − 0.004) | 0.037 | 0.026 (− 0.006, 0.058) | 0.115 |
Baseline FM% | 0.007 (− 0.003, 0.016) | 0.150 | 0.013 (− 0.001, 0.027) | 0.072 | 0.009 (− 0.002, 0.019) | 0.126 | 0.013 (− 0.001, 0.026) | 0.072 | − 0.015 (− 0.031, 0.000) | 0.056 | 0.001 (− 0.005, 0.008) | 0.679 |
Change in FM% | 0.019 (0.008, 0.036) | 0.031 | 0.020 (−0.004, 0.045) | 0.108 | 0.009 (−0.010, 0.029) | 0.351 | 0.025 (−0.001, 0.052) | 0.057 | −0.027 (− 0.055, 0.001) | 0.062 | 0.011 (− 0.001, 0.022) | 0.082 |
- Note: All models were adjusted for age, country, education, follow-up duration, baseline and follow-up wear time, baseline movement-behavior composition, residence change, baseline employment status, change in employment status, baseline health status and change in health status, baseline marital status and follow-up marital status. Baseline adiposity models were additionally adjusted for change in adiposity, while change in adiposity models were additionally adjusted for baseline adiposity. The outcome variable in each model represents the difference between the respective pivot coordinates at follow-up and baseline. For ease of interpretation, the first pivot coordinate has been expressed as the sum of individual log-ratios
- A positive beta suggests that an increase in the respective explanatory variable is associated with an expected increase in the time spent in the movement behavior that is in the numerator of the pivot coordinate at the expense of the movement behavior(s) that is (are) in the denominator of the pivot coordinate. A negative beta suggests that an increase in the respective explanatory variable is associated with an expected decrease in the time spent in the movement behavior that is in the numerator of the pivot coordinate in favor of the movement behavior(s) that is (are) in the denominator of the pivot coordinate
- β = unstandardised regression coefficient, expected change in the pivot coordinate associated with a unit increase in the explanatory variable, CI confidence interval, SB sedentary behavior, LPA light physical activity, MVPA moderate-to-vigorous physical activity, BMI body mass index, FM% fat mass percentage, Change change baseline to follow-up
