Associations Between Aspects of Diet and Internalizing and Externalizing Behaviors in Children Aged 4 Years

Background/objectives: Mental health challenges are increasing worldwide. Identifying preventable factors for such challenges is important and will have the greatest impact if identified in young children. In this study we aimed to explore the association between aspects of diet and child mental health at the age of 4 years. Methods: This is a secondary analysis of data from the Early Food for Future health study in Norway, a randomized controlled trial aiming to improve diet at ages 6–12 months. Cross-sectional data from 363 children aged 4 years and their mothers are used. Diet was assessed using food frequency questions. Five food scores were developed: vegetables, fruits, fruit and vegetables, sweet/salty snack score, and soft drink score. The Child Behavioral Checklist (CBCL) was used to assess internalizing and externalizing behaviors. Crude and multivariable linear regression models are presented. Results: We found that frequency of consumption of vegetables and total frequency of consumption of fruit and vegetables in 4-year-old children were inversely associated with internalizing behavior (β −0.207, 95% CI: 0.351, −0.063), while frequency of consumption of sweet and salty snacks was positively associated (β 1.807, 95% CI: 0.276, 3.337) with externalizing behavior and frequency of consumption of fruit and vegetables was inversely associated (β −0.188, 95% −0.336, −0.041). All were independent of maternal education, measures of financial difficulties and maternal mental health. Conclusion: An inverse association was observed between child frequency of consumption of fruit and vegetables and internalizing and externalizing behaviors, whereas frequency of consumption of sweet/salty snacks was positively associated with externalizing behavior. A varied and healthy diet early in life may promote child mental health, with potential large returns for society. Given the observational nature of the data, causal inference is limited and intervention studies are needed.