To assess the association between child ultra-processed food (UPF) consumption and home-school learning environment characteristics during school closures due to the COVID-19 pandemic in schoolchildren with low- and middle income in Chile.

Cross-sectional. UPF consumption was collected using the Nova screener. We apply the structured days hypothesis (SDH) to assess home-school learning environment characteristics with three constructs that summarised school preparedness for online teaching and learning, school closure difficulties for caregivers and child routine. We explored associations between child UPF consumption and home-school environment characteristics using multivariate linear regression analyses after controlling for child demographic and school characteristics.

Low- and middle-income neighbourhoods in southeastern Santiago, Chile.

Children from the Food Environment Chilean Cohort (n 428, 8–10 years old).

Based on the Nova score, child mean consumption of UPF was 4·3 (sd 1·9) groups. We found a statistically significant negative association between child routine for eating, play and study and child UPF consumption when we adjusted for child sociodemographic (model 1: β = –0·19, (95 % CI –0·40, 0·02)) and school characteristics (model 2: β = –0·20, (95 % CI –0·41, 0·00)). Associations between school preparedness for online teaching or school closure difficulties and UPF were not statistically significant.

Variations in child routines during the COVID-19 pandemic were negatively associated with UPF intake in schoolchildren with low- and middle income. Our findings are consistent with the SDH, suggesting the school environment helps regulate eating behaviours. Future research should evaluate what happens when children return to in-person classes at school.