School food environment and healthy eating in Italian schools: evidence from a nationwide teacher survey

School food environments (SFEs) are critical for shaping children’s nutrition and health. In Italy, national guidelines establish standards for school catering, yet decentralized governance generates substantial variability in meal quality, stakeholder engagement, and implementation practices across schools and municipalities. To examine how food quality, governance, and nutrition-related educational activities within school settings are associated with teacher-reported student consumption of fruit, vegetables, and legumes. A survey of 2,431 teachers from all 20 Italian regions was conducted. The questionnaire assessed school food environment characteristics related to governance, infrastructure, food provision, educational activities, and family engagement. Item Response Theory was used to estimate three latent indices: quality of the offer (QO), quality of educational activities (QA), and role of stakeholders in governance (RSG). Ordinal logistic regression models were used to examine the associations between these indices, school-level characteristics, and teacher-reported levels of fruit, vegetable, and legume consumption. 60% of respondents taught in kindergartens, 63% in schools located in urban areas, 54% in state-recognized private schools, and 46% in public schools. Fruit consumption was generally high (69% in the high-consumption category), while vegetable and legume consumption was more evenly distributed across intake levels. QO showed the strongest and most consistent associations with teacher-reported consumption across all food groups. Higher RSG scores were associated with aprobability higher by 2.5 percentage points of high vegetable consumption and a probability higher by 2.0 percentage points of high legume consumption, while no significant associations were observed for QA. Kindergartens showed higher QO scores (+ 0.132) and were associated with higher probabilities of high fruit (+4.0 percentage points) and legume consumption (+6.9 percentage points) compared with primary schools. Overall, structural and governance-related dimensions of the school food environment showed stronger associations with teacher-reported consumption than nutrition-related educational activities. The findings show that meal quality and stakeholder engagement were associated with teacher-reported FVL consumption in Italian schools, while no clear association was observed for nutrition-related educational activities. These results point to the importance of considering different dimensions of the school food environment when examining children’s eating behaviors. They also contribute to ongoing discussions on school meal governance and public food procurement, particularly in decentralized school food systems. Further research is needed to examine these relationships over time and to better understand how school food environments relate to healthier and more sustainable food systems.