# Ultraprocessed Food Consumption and Behavioral Outcomes in Canadian Children

**Authors:** Meaghan E. Kavanagh, Zheng Hao Chen, Sukhpreet K. Tamana, Theo J. Moraes, Elinor Simons, Stuart E. Turvey, Padmaja Subbarao, Piushkumar J. Mandhane, Kozeta Miliku

PMC · DOI: 10.1001/jamanetworkopen.2026.0434 · JAMA Network Open · 2026-03-03

## TL;DR

Higher consumption of ultraprocessed foods in preschool children is linked to worse behavioral and emotional outcomes by age 5, suggesting dietary changes could support healthier development.

## Contribution

This study is the first to show a link between ultraprocessed food intake in early childhood and adverse behavioral outcomes later in life.

## Key findings

- Each 10% increase in ultraprocessed food energy intake was associated with higher behavioral and emotional symptom scores.
- Replacing 10% of ultraprocessed food energy with minimally processed foods was linked to lower behavioral symptom scores.
- Ultraprocessed foods made up nearly half of preschool children's energy intake in Canada.

## Abstract

Are ultraprocessed foods (UPF) associated with behavioral and emotional functioning among preschool children?

In this cohort study of 2077 Canadian children, higher UPF intake at age 3 years was associated with adverse behavioral and emotional symptoms at age 5 years. Modeling the substitution of a portion of UPF with minimally processed foods was associated with better behavioral and emotional symptoms.

These findings suggest that UPF consumption in early childhood may adversely influence behavioral and emotional development, and that ongoing public strategies promoting minimally processed foods in place of UPF could help support children’s development.

This cohort study examines the associations between intake of ultraprocessed foods at age 3 years and behavioral outcomes at age 5 years among children in Canada.

Ultraprocessed foods (UPF) contribute to nearly half of energy intake among preschool-aged children in Canada, yet their impact on behavioral and emotional functioning remains underexplored.

To examine the associations between UPF intake at age 3 years and behavioral outcomes at age 5 years.

This cohort study included children with dietary and behavioral data from September 2011 to April 2018 in the CHILD Cohort Study, a prospective Canadian pregnancy cohort. Data analysis was done between February to July 2025.

UPF intake at age 3 years was assessed by a 112-item food frequency questionnaire and categorized using the NOVA system.

Behavior was measured using the validated Child Behavior Checklist (CBCL; range 0-100; higher scores indicating more adverse symptoms). UPF intake measured as a continuous percentage of energy was examined using multivariable-adjusted linear regression models accounting for maternal diet, birth factors, infant feeding, and sociodemographic and early-childhood characteristics. A multivariable-adjusted substitution model estimated the association of statistically replacing 10% energy from UPF with minimally processed foods (MPF) among all children.

Among 2077 participants, 1092 (52.6%) were male; 1376 children (66.2%) were White, 480 children (23.1%) were multiracial, and 221 children (10.7%) were identified as another ethnic group. At age 3 years, UPF contributed a mean (SD) of 45.5% (11.6%) of total energy intake. At age 5 years, the mean (SD) CBCL scores were 44.6 (9.1) for internalizing, 39.6 (9.4) for externalizing, and 41.2 (9.0) for total behavior. Each 10% increase in energy from UPF was associated with higher CBCL internalizing (β = 0.81 [95% CI, 0.43 to 1.19]), externalizing (β = 0.47 [95% CI, 0.08 to 0.87]), and total (β = 0.64 [95% CI, 0.27 to 1.01]) scores. Substitution of 10% energy from UPF with MPF was associated with lower internalizing (β = –0.91 [95% CI, –1.33 to –0.49]), externalizing (β = –0.49 [95% CI, –0.93 to –0.06]), and total (β = –0.70 [95% CI, –1.12 to –0.29]) scores.

In this cohort study of preschoolers in Canada, higher UPF intake was associated with adverse behavioral and emotional symptoms by age 5 years. These findings suggest that replacing UPF with MPF during the preschool years may support healthier behavioral development, with potential benefits for long-term mental health. These findings also support ongoing policy actions that promote MPF and underscore the need for early-life dietary interventions.

## Full-text entities

- **Genes:** MSLN (mesothelin) [NCBI Gene 10232] {aka MPF, SMRP}, CRP (C-reactive protein) [NCBI Gene 1401] {aka PTX1}
- **Diseases:** obesity (MESH:D009765), anxiety disorders (MESH:D001008), major depression (MESH:D003865), hyperactivity (MESH:D006948), UPF (MESH:D005517), sleep problems (MESH:D012893), inflammatory (MESH:D007249), cardiometabolic diseases (MESH:D024821), congenital abnormalities (MESH:D000013), disruptive behavior (MESH:D019958), neuroinflammation (MESH:D000090862), anxiety (MESH:D001007), behavior problems (MESH:D001523), Trisomy 21 (MESH:D004314), fearfulness (MESH:C000719212), aggression (MESH:D010554), depression (MESH:D003866), disability (MESH:D009069), internalizing symptoms (MESH:D000082122), CBCL (MESH:D002653), CHILD (MESH:D002658), Atherosclerosis (MESH:D050197), externalizing symptoms (MESH:D012816), adverse (MESH:D064420), mental health disorders (OMIM:603663), CHILD (MESH:C562515), attention-deficit/hyperactivity disorder-related behaviors (MESH:D001289)
- **Chemicals:** iron (MESH:D007501), sugar (MESH:D000073893), zinc (MESH:D015032), vitamin D (MESH:D014807), omega-3 fatty acids (MESH:D015525), vitamin B12 (MESH:D014805), folate (MESH:D005492), sodium (MESH:D012964), phthalates (MESH:C032279), artificially and sugar-sweetened beverages (-), bisphenol (MESH:C543008), fatty acids (MESH:D005227)
- **Species:** gut metagenome (species) [taxon 749906], Homo sapiens (human, species) [taxon 9606]

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC12958080/full.md

## Figures

3 figures with captions in the complete paper: https://tomesphere.com/paper/PMC12958080/full.md

## References

85 references — full list in the complete paper: https://tomesphere.com/paper/PMC12958080/full.md

---
Source: https://tomesphere.com/paper/PMC12958080