Ultra-Processed Food at the Supermarket: What Studies Show and How to Choose Without Panic

The new research asks a more precise question

On August 4, 2025, Nature Medicine published a trial of diets with different levels of processing. Researchers randomly assigned 55 adults in England to two sequences of eating: a minimally processed diet and an ultra-processed diet, eight weeks on each, with a break in between. Both diets were designed around Britain's Eatwell Guide.

Participants lost weight on both diets, but the loss was greater on the minimally processed diet. This does not mean that every product in one group causes weight gain. In fact, the results on both diets are a reminder that nutritional composition matters too. The trial involved people with overweight or obesity, rather than testing the population as a whole.

This is where the useful part for shopping begins: understanding what is being classified, what was actually tested, and what can be changed without turning the cart into a list of prohibitions. The term ultra-processed describes a classification system. It is not a medical diagnosis and does not replace examining the particular food.

NOVA's four groups, in plain language

A report published through the UN Food and Agriculture Organization, FAO, in 2019 explains NOVA. It groups foods by the nature and purpose of processing, rather than just their sugar or fat content. Two products with similar nutritional values may therefore fall into different groups.

The first group includes unprocessed or minimally processed foods, such as vegetables, fruit, dried lentils, and frozen vegetables with no added ingredients. Plain yogurt made from milk and cultures can also belong to this group. Refrigeration, freezing, and pasteurization do not automatically make a food ultra-processed.

The second group consists of cooking ingredients, such as oil, salt, and sugar. The third includes processed foods that combine a basic food with such ingredients, for example canned legumes with salt or simple cheeses. The exact classification depends on the composition and production method.

The fourth group includes industrial formulations that may contain ingredients extracted from foods or modified, and additives that shape flavor, texture, and appearance. Soft drinks and many snacks are examples. Packaging alone, the number of ingredients, or a hard-to-pronounce name is not a sufficient test for classification.

What observational studies can tell us

An umbrella review published in BMJ in February 2024 brought together meta-analyses of observational studies. It found associations between greater exposure to ultra-processed food and adverse health outcomes, including type 2 diabetes and heart disease. The strength of the evidence was not the same for every outcome.

Such an association matters, but does not prove that processing itself is the cause. People who eat differently may also differ in income, smoking, physical activity, and access to food. Statistical analysis can account for some of these differences, but not necessarily all of them. Dietary reporting and product classification over the years are also imperfect.

There is therefore no basis for taking a figure from such a study and presenting it as a reader's personal chance of becoming ill because of a particular product. The picture mainly concerns a sustained eating pattern. It does not allow us to calculate the harm from a single snack or attribute a person's illness to one item on an ingredient list.

What the small trial preceding the current debate added

In a trial published in Cell Metabolism in 2019, 20 adults stayed at an NIH research center for a month. Each participant received, in random order, two weeks of an ultra-processed diet and two weeks of an unprocessed diet. Participants could eat as much as they wished of the food served.

During the ultra-processed diet, they consumed more energy and gained weight, and during the other period they lost weight. Random assignment strengthens the possibility that the diets tested caused the short-term difference. It does not show that every ultra-processed product works this way, and did not examine disease developing over years.

The 2025 trial does not resolve every question either. It took place in everyday life, with food supplied to participants, and it was impossible to conceal what they were eating. People with celiac disease or inflammatory bowel disease were not included, so its conclusions should not be applied directly to their therapeutic diets.

The debate also concerns what exactly caused the difference

Scientific responses to the trial published in Nature Medicine raised questions about nutritional differences between the diets and the conclusion that processing should be added to guidelines. This is a crucial distinction: a trial compares actual meals, not an abstract factor called processing that can always be isolated from everything else.

Texture, energy density, composition, and eating speed may change together. If an outcome differs, researchers need to keep investigating which characteristics contributed to it. Uncertainty does not invalidate the finding; it determines how far we can take it and which questions remain open.

At the supermarket, look at both the list and the label

The Ministry of Health's red symbols indicate high amounts of sugar, sodium, or saturated fat under the regulations' criteria. They are not a NOVA label. A product can be ultra-processed without a red symbol, and the absence of a symbol is not a blanket endorsement.

Start with how the food will be used: are you looking for a meal base, a side dish, or a snack? Then compare similar products using their ingredient lists and nutrition panels. The guide to reading labels with a sensitive gut adds questions about individual suitability that the processing classification alone cannot answer.

Frozen vegetables, canned legumes, and plain yogurt can make preparing food at home easier. There is no need to give them up because they have undergone an industrial process. The comparison between frozen and fresh vegetables also starts with what people actually eat, rather than which supermarket department a product is in.

A change you can sustain

Choose one product you eat often and consider an alternative that suits your taste, budget, and preparation time. For example, keep an accessible meal base on hand instead of always relying on snacks. This is a practical suggestion, not a promise of a medical outcome or a requirement to make everything from scratch.

For someone living with bowel disease or a dietary restriction, a suitable packaged product may be an important tool. Do not remove safe, familiar foods simply because of a new headline. Improve your eating pattern with your care team, while leaving room for the realities of your kitchen, budget, and symptoms.

This is general news information and is not a substitute for personal advice from a doctor or dietitian.