Skip to main content
. 2022 Jul 20;14(14):2961. doi: 10.3390/nu14142961

Figure 2.

Figure 2

Marginal effects from a quantile regression of ultra-processed food intake at lunch with an interaction between meal type and income group, stratified by school phase. (A) UPF as % g in primary schoolchildren. Interaction between meal type and income: (Low # school meal—reference; Mid # school meal p = 0.08; High # school meal p = 0.01); (B) UPF as % g in secondary schoolchildren. Interaction between meal type and income: (Low # school meal— reference; Mid # school meal p = 0.68; High # school meal p = 0.77); (C) UPF as % kcal in primary school. Interaction between meal-type and income: (Low # school meal— reference; Mid # school meal p = 0.76; High # school meal p = 0.96); (D) 3D UPF as % kcal in secondary schoolchildren. Interaction between meal-type and income: (Low # school meal— reference; Mid # school meal p = 0.12; High # school meal p = 0.39). Note: Estimates are adjusted for age, sex, survey year, ethnicity, region, and IMD. Coefficients from interaction model given in Supplementary Table S4.