Teun Remmers

236 | Appendix 3 Supplementary Digital Content 5: Relationship between weather elements and physical activity, and interactions between weather variables and day type ( n =4599 days, n =307 children) Vigorous PA std. B (95% CI) -2 restricted log likelihood = Rainfall (0.01–7.8 mm per day versus no rainfall) Rainfall (7.9–46.0 mm per day versus no rainfall) 0.055 (-0.001 to 0.112) -0.010 (-0.096 to 0.076) Humidity -0.044 (-0.082 to -0.006) Day length -0.147 (-0.274 to -0.019) Temperature -0.625 (-1.017 to -0.233) Temperature squared 0.473 (0.086 to 0.859) Solar Radiation exposure 0.157 (0.109 to 0.206) Day type (weekday versus weekend day) 0.343 (0.286 to 0.400) Gender (boys versus girls) 0.532 (0.396 to 0.667) Age in years -0.046 (-0.117 to 0.025) Body Mass Index -0.102 (-0.164 to -0.041) Solar * Temperature -0.513 (-0.725 to -0.302) Solar * Temperature squared 0.398 (0.188 to 0.608) Temperature * Day type 0.563 (0.181 to 0.946) Temperature squared * Day type -0.626 (-0.999 to -0.254) Table shows fixed effects (as standardized regression coefficients (std. B) with 95% confidence intervals (CI) from linear mixed models, with a random intercept for each child, and a repeated term for each child across each repeated observation, with autoregressive covariance structure. Additional testing for 3-way interaction did not significantly improve the model. Bolded text indicates significant at p<0.05.