Half of your kid’s food might be ultra-processed and that’s a problem

By Tarun Sai LomteReviewed by Susha Cheriyedath, M.Sc.Jun 19 2025

New Australian research reveals that ultra-processed foods make up nearly half of kids' diets, posing obesity risks for older children aged 10–12.

Study: Ultra-processed food intake and risk of obesity among schoolchildren aged 8–12 years living in Victoria, Australia. Image Credit: K-FK / Shutterstock

In a recent study published in the journal Pediatric Obesity, researchers examined associations between ultra-processed food (UPF) intake and obesity indicators in schoolchildren.

Global rates of childhood obesity and overweight have markedly increased in the past decades. Changing eating patterns are a critical aspect of this increased prevalence. The global food supply has been increasingly characterized by the mass promotion and production of highly processed, palatable, energy-dense foods, known as UPFs.

The dietary prevalence of UPFs has rapidly increased, with Australians purchasing 134 kg of UPFs per capita in 2019. Reducing UPF intake has been a focal point of nutritional recommendations in many countries. Evidence on associations between UPF intake and obesity or overweight measures is robust in adults but less consistent in children.

About the study

In the present study, researchers examined associations between UPF intake and obesity indicators in children. They used data from the Salt and Other Nutrients in Children study, which involved schoolchildren. Between 2010 and 2013, a convenience sample of primary schools in Victoria, Australia, was selected (cohort 1). In 2018-19, schools comparable to the previous sample were recruited (cohort 2). Only individuals aged 8 years or older were included.

Subjects completed a 24-hour dietary recall; cohort 1 completed a face-to-face dietary recall, while cohort 2 completed a web-based dietary recall using an online dietary assessment tool. Reported food intake was coded to foods within the AUSNUT-2011-2013 food composition database. Subsequently, each food item was categorized into NOVA groups using a previously established database.

The proportion of energy and weight consumed from each NOVA group in the total diet was estimated. Further, anthropometric data, including height, weight, and waist circumference, were collected at school. Body mass index (BMI) z-scores, waist-to-height ratio (WHtR), and the International Obesity Task Force definitions of underweight, overweight, healthy, and obese were used as three distinct indicators of obesity.

Children’s parents completed questionnaires on sociodemographic characteristics, including age, sex, area-level and parental education measures of socioeconomic status, geographic location, and parents’ education. UPF intake quartiles were determined, and chi-squared and t-tests examined sociodemographic characteristics across quartiles. Linear regression modeling was used to assess the differences in UPF intake across sociodemographic characteristics.

Multiple linear regression was used to examine the association between a 10% difference in UPF intake (% of total energy or grams) and BMI z-score, while logistic regression was employed to investigate the association between UPF intake and WHtR and weight status. Model 1 was unadjusted, while model 2 was adjusted for sex, age, residential location, socioeconomic status, and whether dietary recall captured food intake on a school or non-school day. The final model was additionally adjusted for energy intake.

Findings

In total, 682 children aged 10.2 years, on average, were included. Participants were slightly more likely to be male (53.4%), had a healthy weight (73%), resided in major cities (76.5%), and had a high socioeconomic status across both area-level (54%) and parental education (64%) measures. Around 4% of children were obese, 16% were overweight, 6.6% were underweight, and 24% had a high WHtR. Further, most dietary recalls (75%) captured food intake on a school day. All children consumed some UPF on the day of dietary recall, with UPFs contributing 47.2% of the total energy intake.

Minimally processed or unprocessed foods contributed 32% of energy intake, processed culinary ingredients contributed 6.8%, and processed foods contributed 13.7% of energy intake, respectively. The top sources of energy from UPFs were pastries, cakes, and buns (4.8%), confectionery (4.5%), biscuits (4.7%), fast food dishes (3.9%), and breakfast cereals (4.1%). Further, boys consumed 425 kJ/day more absolute energy from UPF than girls, though UPF as a proportion of total intake didn't differ by sociodemographics.

When considering UPF as a proportion of total diet, intake showed no differences across age, sex, location, or socioeconomic groups. UPF intake was not associated overall with any obesity indicator. However, results varied by age and UPF measurement metric.

Age-stratified analysis revealed that among children aged 10–12 years, a 10% increase in UPF, measured as the proportion of total food weight (g/day), was associated with a 0.07-unit higher BMI z-score and a 19% higher odds of abdominal obesity. No association was found for overweight/obesity classification or for UPF measured by energy proportion. This was significant after adjusting for energy intake.

Conclusions

In sum, UPF intake accounted for nearly half of schoolchildren's total energy intake, with no sociodemographic differences in proportional consumption. The associations between UPF intake and obesity indicators varied by age, with no associations in children aged 8–9 and a small positive association specifically for UPF measured by weight proportion with abdominal obesity and BMI z-score in those aged 10–12. The sample's lower obesity prevalence (4% vs 25% national rate) and high-SES skew should be considered when interpreting results. Further research is needed to understand the potential impact of UPF consumption on adiposity outcomes.