Additional file 1 of Association of ultra-processed food patterns with overweight and obesity in the German health interview and examination survey for children and adolescents (KiGGS): a longitudinal study

Additional File 1. Spline plots for the associations between ultra-processed food consumption and odds of overweight and obesity. Spline plots for the associations between ultra-processed food consumption and odds of overweight and obesity. Spline plots for the associations between exposure to the ultra-processed food pattern (contribution of ultra-processed foods to total energy intake – % UPF) and odds ratios of overweight and obesity. KiGGS (2003 – 2017). The knots correspond to the 20th, 40th, 60th, and 80th centiles for contributions of 42.38%, 50.16%, 56.92%, and 65.20%, respectively, of ultra-processed foods to total energy intake in the non-overweight sample at baseline (n=4,249) and to 42.36%, 50.07%, 56.84%, and 65.15%, respectively, in the nonobese sample at baseline (n=4,605). Overweight analyses: Wald test for the linear term, P=0.195; Wald test for all nonlinear terms =0.516. Obesity analyses: Wald test for linear term P=0.501; Wald test for all nonlinear terms P=0.066. While plots suggest relatively uniform associations for overweight and a potential non-linear pattern for obesity at higher exposures, Wald tests did not confirm evidence of non-linearity.