A new study finds that nine in ten research papers disputing the definition of ultra-processed foods have ties to the food industry. The result will not tell you what to have for lunch, but it does illuminate a scientific argument with real consequences, because the definition being fought over increasingly shapes food labels and public health advice.
The definition comes from the NOVA classification, which groups foods by how they are made rather than only by what nutrients they contain. Fresh foods and simple staples come first. At the far end sit ultra-processed products: industrial formulations such as fizzy drinks, packaged snacks, instant meals and processed meats, typically built from refined ingredients, additives and long shelf lives.
Counting the Papers and Their Backers
The researchers gathered published papers that dispute the ultra-processed definition and checked the funding and affiliations behind each one. Around ninety percent had links to food companies or industry-connected bodies. That is the study’s central, verifiable claim, and it is about the structure of the debate: the loudest scepticism is rarely independent of the interests affected.
Those interests are easy to understand. the US Food and Drug Administration oversees much of American food labelling, and the World Health Organization advises governments worldwide on diet and health. If definitions used by such bodies classify a product as ultra-processed, that product may face warning labels, advertising limits or taxes, so companies rationally fund research that questions the classification.
Arguments Worth Hearing Anyway
Industry backing does not make every criticism wrong. The NOVA system genuinely does group together products with very different nutritional profiles, and some researchers with no industry funding share the concern that processing is a rough stand-in for healthfulness. Nutrient-based approaches have their own blind spots, but the methodological debate is a real one, conducted by serious people on both sides.
Where readers should be alert is in the leap from caveat to conclusion. Showing that a definition is imperfect is not the same as showing the foods it flags are harmless. Our earlier coverage at this recent coverage made a similar distinction in a different research area: questioning a measure is healthy science, while quietly discarding everything the measure reveals is advocacy.
What the Health Evidence Can and Cannot Say
Studies of whole diets repeatedly find that people eating the most ultra-processed food tend to have worse health outcomes. Those studies are observational, so they cannot prove the processing itself is the cause; income, habits and total calories are tangled up in the pattern. Presenting the association as settled proof would overclaim the evidence, and this article will not do that.
Equally, dismissing the pattern because it is observational would be its own kind of error, as we noted when covering early research in this related report. Public health often has to act on the best available pattern while better evidence is gathered, and the pattern here is consistent enough that regulators in several countries already use it.
Practical Takeaways for Shoppers and Policymakers
For policymakers, the study argues for transparency rules with teeth: research submitted in labelling consultations should carry clear funding declarations, and advisory panels should balance funded voices with independent ones. Sunlight will not end the argument, but it lets everyone see the argument’s sponsors.
For shoppers, the message is steadier than the headlines. You do not need to memorise NOVA groups to eat sensibly. Build most meals from recognisable ingredients, treat heavily marketed convenience products as occasional rather than default, and when a study about food makes a dramatic claim in either direction, scroll down and read who paid for it. That habit, more than any label, is the study’s real gift to consumers.
Regular readers may wonder how much of this changes next week’s shopping list. Honestly, very little, and that steadiness is reassuring rather than disappointing. The study is best filed as media literacy rather than dietary instruction: it teaches you to ask a better question about food research, not to fear a new category of product. Households that cook often, share meals and keep heavily processed convenience foods in a supporting role are already aligned with the broad direction of the evidence, whatever refinements the great definition debate eventually produces.