Ultra-processed foods linked to childhood behavioral issues

There is a link between consumption of ultra-processed foods (UPFs) in early childhood and behavioral and emotional development, according to a recent study published in JAMA Network Open.¹

Researchers from the University of Toronto found increased odds of fearfulness, anxiety, hyperactivity, and aggression in children consuming UPFs in early childhood. This was the first study to use detailed, prospective data to determine the link between UPFs and standardized behavioral assessment in kids.¹

“The preschool years are critical for child development, and it’s also when children begin to establish dietary habits,” said Kozeta Miliku, MD, PhD, an assistant professor at the University of Toronto and principal study investigator.¹

Classifying food intake

Participants of the cohort study included pregnant women recruited from 4 sites across western Canada between 2008 and 2012 with singleton births at a gestational age of at least 34 weeks and 4 days, alongside no congenital abnormalities.² Exclusion criteria included trisomy 21, withdrawing before the 3-year visit, and missing data at 3 years.

The NOVA classification system was used to assess UPFs at the 3-year visit, with relevant data obtained by a caregiver-reported 112-item semiquantitative food frequency questionnaire (FFQ). There were 4 NOVA groups that FFQ items were mapped to, including the following²:

• Unprocessed and minimally processed foods (MPFs)
• Processed culinary ingredients
• Processed foods
• UPFs

Energy intake in each NOVA group was divided by the total daily energy intake and multiplied by 100 to determine the percentage energy contributed. Subgroups for UPFs included breads and cereals, sweets and desserts, ready-to-eat or ready-to-heat mixed dishes, animal-based products, savory snacks, artificially and sugar-sweetened beverages, plant-based alternatives, and sauces, spreads, and condiments.²

Behavior outcomes and participant demographics

Investigators used the validated preschool version of the caregiver-reported Child Behavior Checklist (CBCL) to evaluate behavioral and emotional outcomes at the 5-year visit. The CBCL included 99 items discussing symptoms, with responses ranging from 0 meaning not true to the child to 2 being very true or often true.²

A greater CBCL indicated more caregiver-reported behavioral problems. Covariates included prenatal stress, maternal UPFs during pregnancy, maternal education, maternal marital status, child sex, caregiver-reported ethnicity, gestational age at birth, older siblings, household income, study site, energy intake at 3 years, and body mass index.²

There were 2077 children included in the final analysis, 52.6% of whom were male, 66.2% White, 23.1% multiracial, 10.7% another ethnicity, and 57.1% from households with an income of at least CAD$100,000. A median energy intake of 1488.6 kcals/d was reported at the 3-year visit, with UPFs accounting for a mean 45.5% of energy intake and MPFs, a mean 37.9%.²

Energy was most often obtained from sweets and deserts, breads and cereals, animal-based products, and ready-to-eat or ready-to-heat mixed dishes, with rates of 12.5%, 11.9%, 7.9%, and 6.1%, respectively. Mean internalization, externalizing, and total CBLC scores at the 5-year visit were 44.6, 39.6, and 41.2, respectively.²

Impact of UPFs on behavioral scores

Increased CBLC scores were reported across all domains at the 5-year visit for every 10% energy contribution from UPF, at β = 1.02 for internalizing, β = 0.99 for externalizing, and β = 1.05 for total score. While slightly attenuated in the multivariable-adjusted regression analyses, these associations were still significant at 0.81, 0.47, and 0.64, respectively.²

When substituting UPFs with MPFs, lower CBLC scores were reported across all domains for every 10% of substitution, at β = −0.91 for internalizing, β = −0.49 for externalizing, and β = −0.70 for total score. Rises in internalizing and total behavior scores at 1.76 and 1.65, respectively, were reported for increased artificially and sugar-sweetened beverage intake.²

Additionally, scores of β = 0.79 and β = 1.11 were reported for increased intake of bread and cereal and ready-to-eat or ready-to-heat mixed dishes, respectively. However, no other associations were reported for UPF subgroups. Overall, the data indicated a significant link between UPF intake in early childhood with behavioral and emotional problems.²

“Our findings underscore the need for early-life interventions such as professional advice for parents and caregivers, as well as public health campaigns, nutrition standards for child-care providers, and reformulation of some packaged foods,” said Miliku.¹

References

1. Ultra-processed foods in preschool years associated with behavioural difficulties in childhood. News release. University of Toronto. March 3, 2026. Accessed March 11, 2026. https://www.eurekalert.org/news-releases/1118583
2. Kavanagh ME, Chen ZH, Tamana SK, et al. Ultraprocessed food consumption and behavioral outcomes in Canadian children. JAMA Netw Open. 2026;9(3):e260434. doi:10.1001/jamanetworkopen.2026.0434