Research explained: cross-sectional study
Ultra-processed food made up about half of Swedish diets and went with lower diet quality, but the weight link was weak
Three national surveys of adults, teenagers and young children found that a bigger share of energy from ultra-processed food tracked with a less nutritious diet on most measures, while links with body weight appeared only in adults and were small.
Ultra-processed foods supplied on average 45% of energy for Swedish adults, 52% for adolescents and 46% for young children, and people who got more of their energy from them tended to have a less nutritious diet on most measures, according to an analysis of three national dietary surveys in Nutrition Journal (published October 13, 2026). Links with body weight were small and showed up only in adults. The study drew wide coverage on health news sites this past week.
What the researchers did
A team from the University of Gothenburg, Chalmers, Karolinska Institutet, Uppsala University and the Swedish Food Agency used Sweden’s three most recent Riksmaten surveys: adults in 2010-2011 (1797 people), adolescents in 2016-2017 (3099) and young children in 2021-2024 (1820). Each food was sorted with the Nova processing system. To deal with foods that are hard to classify, the team broke mixed dishes into their recipes, had items classified independently, and ran alternative scenarios.
The headline share depends heavily on those choices. When unclear items were classified more strictly, the share of energy from ultra-processed food fell to between 23% and 29%.
A higher share went with worse diet quality on most indicators. The exceptions were salt and saturated fat in every age group, plus iron and vitamin E in young children. In adults, each 10 percentage point rise in energy from ultra-processed food was linked to a 0.61% higher BMI, and overweight or obesity was more common in the top third of intake than in the bottom third (prevalence ratio 1.16). No such links appeared in teenagers or young children.
How strong is the evidence?
This is a cross-sectional study: it captures diet and weight at one point in time, so it shows associations, not cause and effect. Intake was self-reported, and the authors themselves conclude that no robust relationships with BMI were observed. The adult survey is now more than a decade old. One author is editor in chief of Nutrition Journal but had no part in the review of this paper; the other authors declare no competing interests. The abstract does not describe funding.
What it means for you
The category “ultra-processed” is broad, and this study shows how much the label itself can shift the numbers. A practical takeaway is to look at what a food brings, not just how it is made. Swapping some packaged snacks for plain oats, whole wheat bread, an apple or raw carrots adds fiber and nutrients without needing to cut out every packaged food. See more options on our grains page.