Swedish surveys show ultra-processed foods make up 45% to 52% of energy intake

Researchers at the University of Gothenburg and the Swedish Food Agency analyzed Swedish national dietary surveys covering adults, adolescents, and young children. The data, published in Nutrition Journal, came from three surveys: Riksmaten Adults (2010-2011; n = 1797), Riksmaten Adolescents (2016-2017; n = 3099), and Riksmaten Young Children (2021-2024; n = 1820) doi.orgNutrition Journaldoi.orgOpen source ↗.

Food items were classified using the Nova framework along with recipe disaggregation, independent reviews, and alternative classification scenarios. Under the primary scenario, ultra-processed foods accounted for an average of 45 E% in adults, 52 E% in adolescents, and 46 E% in young children doi.orgNutrition Journaldoi.orgOpen source ↗. When ambiguous food items were classified under stricter criteria, these estimates dropped to between 23 E% and 29 E% doi.orgNutrition Journaldoi.orgOpen source ↗.

Diet quality indicators

A higher share of ultra-processed food energy was associated with less favorable diet quality across most metrics. However, this inverse trend did not apply to salt and saturated fat across all age groups, nor to iron and vitamin E intake in young children doi.orgNutrition Journaldoi.orgOpen source ↗. The researchers noted that fortified foods, such as prepared cereals and porridge, contributed to these specific nutrient profiles in young children.

Body mass index associations

Among adults, each 10 E% increment in ultra-processed food intake corresponded to a 0.61% higher body mass index (95% CI: 0.04-1.18; p = 0.036) doi.orgNutrition Journaldoi.orgOpen source ↗. Adults in the highest tertile of ultra-processed food intake had a higher prevalence of overweight or obesity compared to the lowest tertile after adjustment for energy intake, with a prevalence ratio of 1.16 (95% CI: 1.03-1.30; p = 0.018) doi.orgNutrition Journaldoi.orgOpen source ↗.

No associations between ultra-processed food intake and body mass index or overweight status were observed among adolescents or young children doi.orgNutrition Journaldoi.orgOpen source ↗. The authors concluded that no robust relationships with body mass index were observed across the investigated cohorts doi.orgNutrition Journaldoi.orgOpen source ↗.