Råvaror och förpackade livsmedel sida vid sida på ett ljust ekbord

Ultra-processed food, what it actually is and why nutrient density matters more

When French nutrition researchers asked trained evaluators to sort ordinary supermarket foods into the four NOVA groups, plain unsweetened yoghurt was placed in all four of them. Agreement between evaluators was poor, and handing them the complete ingredient list did not improve it.

That is an uncomfortable starting point for a term that now turns up in headlines, in political debate and on packaging. The concept is real, with a written definition and a large literature behind it, and it is less precise than the confidence around it suggests. This article goes through the definition, the studies and their weak points, and then turns to what can be measured in a Swedish kitchen: what a diet heavy in ultra-processed products does to the nutrients you actually get, and how we reason about a daily supplement when the week is not perfect.

Health Royals Peak Performance, supplement with vitamin D3, K2 as MenaQ7, folate as Quatrefolic and omega-3 from algal oil

The daily complement at Health Royals

Good food first, and a concentrated complement every day. Peak Performance Formula brings together vitamin D, folate and selenium, which Swedish dietary surveys point to as common gaps, with omega 3 from algae and eight other named components in one daily dose. Vitamin D contributes to the normal function of the immune system.

See Peak Performance, multivitamin for women with omega 3

What does ultra-processed food actually mean?

Ultra-processed food is group 4 in the NOVA classification, and it is defined by how a food is made, not by what nutrients it contains.

NOVA sorts everything we eat into four groups, set out by Carlos Monteiro and colleagues in Public Health Nutrition in 2019. Group 1 is unprocessed or minimally processed food, altered only by drying, grinding, freezing, pasteurisation and similar steps. Group 2 is processed culinary ingredients, such as oil, sugar and salt. Group 3 is processed food, where group 2 ingredients are added to group 1 foods, as in tinned beans, cheese and ordinary bread. Group 4, ultra-processed food, is described as formulations of ingredients mostly of exclusive industrial use, assembled by industrial techniques.

The practical test the authors give is not about nutrition at all. Read the ingredient list. If it contains at least one substance never or rarely used in a kitchen, such as high fructose corn syrup or hydrogenated oil, or at least one cosmetic additive, such as a flavouring, colour, emulsifier, sweetener or thickener, the product is group 4. Nothing in that test refers to energy, salt, sugar or fibre.

Where does the line between processed and ultra-processed go?

The line is drawn at the ingredient list, which means two foods that behave similarly in the body can land on opposite sides of it.

Bread made from flour, water, salt and yeast is group 3. The same bread with an added emulsifier is group 4. So is a wholegrain breakfast cereal fortified with vitamins and minerals. The UK Scientific Advisory Committee on Nutrition made this point in its 2023 position statement: NOVA classifies by processing rather than nutrient content, and therefore captures foods usually regarded as healthier.

How much this matters has been measured. In the European Journal of Clinical Nutrition in 2022, Braesco and colleagues asked evaluators to classify lists of marketed and generic foods. Agreement was low, with Fleiss kappa values of 0.32 and 0.34, a quarter of the foods in each list were assigned inconsistently, and between 26 and 35 percent of the foods classified as group 4 still met the criteria of a healthier nutrient profile. A 2025 systematic review in Food and Nutrition Research identified six competing classification systems, NOVA among them, and concluded that none of them specifies processing techniques clearly.

How much ultra-processed food do people in Sweden eat?

Sweden sits near the top of the European range, with roughly four in ten calories coming from ultra-processed products.

Mertens, Colizzi and Penalvo pooled national dietary surveys from 22 European countries in the European Journal of Nutrition in 2022. Across Europe the share of total energy from ultra-processed food and drink ranged from 14 to 44 percent, with an average of 27.2 percent. In Sweden the figures were 40.6 percent for men and 43.8 percent for women, among the highest measured.

Those numbers deserve a caveat. They come from surveys collected at different times with different methods, with NOVA applied afterwards to food descriptions rather than labels, so the true figure for Sweden is a range, not a decimal. What the surveys show more reliably is where Swedish diets fall short on specific nutrients, the subject of our article on what the Riksmaten survey found.

What do the studies on health outcomes actually show?

They show a consistent statistical association between higher intake and worse outcomes, and almost all of it comes from observational cohorts.

The umbrella review by Lane and colleagues in The BMJ in 2024 gathered 45 unique pooled analyses covering around 9.9 million participants. Four associations reached the highest credibility grade: cardiovascular disease mortality, with a risk ratio of 1.50; a dose response association with type 2 diabetes, risk ratio 1.12 per increment; anxiety outcomes, odds ratio 1.48; and common mental disorders combined, odds ratio 1.53.

European cohorts agree. A multinational analysis of 266,666 participants in seven countries, reported by the International Agency for Research on Cancer in 2023, linked higher intake to multimorbidity of cancer and cardiometabolic disease. The scoping review commissioned for the Nordic Nutrition Recommendations 2023 covered 12 systematic reviews and 44 original studies. Of those 44, 43 were prospective cohort studies and exactly one was a randomised trial.

Is it the processing, or is it what is in the food?

Two controlled feeding trials suggest processing contributes something beyond nutrients, and they disagree about how much.

The first was run by Kevin Hall at the US National Institutes of Health and published in Cell Metabolism in 2019. Twenty adults lived on a metabolic ward for a month and ate two weeks of an ultra-processed diet and two weeks of a minimally processed one, matched for calories, sugar, fibre, fat and carbohydrate, eating as much as they wanted. On the ultra-processed diet they ate about 500 calories more per day and gained 0.9 kg, while losing 0.9 kg on the other diet.

The second, by Dicken and colleagues, appeared in Nature Medicine in 2025. Fifty adults ate for eight weeks on each diet, with a four week gap, and both diets were built to meet national healthy eating guidelines. Weight fell on both, by 2.06 percent on the minimally processed diet and by 1.05 percent on the ultra-processed one, a difference of 1.01 percentage points in favour of the minimally processed diet.

Read together, the two trials say something more modest than the headlines. When an ultra-processed diet is also nutritionally poor, the effect is large. When it is nutritionally decent, the remaining effect of processing is real but small. Most of what group 4 does to a diet runs through what the food contains and what it displaces.

How strong is the evidence, and what does it show?

The association is consistent, and the research is still working out how much of it comes from processing itself and how much from what else tends to go with a diet high in group 4.

The quality grading in the BMJ umbrella review makes the point. Of the pooled analyses assessed with GRADE, 22 were rated low quality and 19 very low quality. Only four reached moderate, and none reached high. The reason is structural: people who eat the most ultra-processed food also tend to smoke more, earn less, exercise less and eat less fruit, vegetables and fibre. When the 2025 Lancet series on ultra-processed food was published, several researchers said so in the expert reaction collected by the Science Media Centre, among them Martin Warren of the Quadram Institute, who argued that socioeconomic confounding is given too little weight.

What the evidence does show is clear. The association is consistent across millions of participants, and both controlled trials found that people ate more, or lost less, on the ultra-processed diet. The mechanism everyone agrees on is the unglamorous one: ultra-processed products tend to be energy dense and nutrient poor, and they take the place on the plate of foods that are not. A diet with four in ten calories from group 4 is, above all, a diet with less room for the nutrients that have to come from somewhere.

What does a diet high in ultra-processed food do to your nutrient intake?

It makes a nutrient dense diet harder to reach, because the foods that carry the nutrients Swedish diets already run low on are the ones being displaced.

This is where the debate stops being about definitions. Nutrients and energy density can be measured in a person, and they are what moves when group 4 takes over the plate. Our article on what Riksmaten found goes through where Swedish intakes fall short, and the nutrients that come out low there are vitamin D, folate and iron, with selenium as a fourth. They come from fatty fish, eggs, leafy greens, pulses and meat and, for vitamin D, from sunlight that is absent for half the year. A diet where four in ten calories arrive ready made leaves less room for those foods, and the gap shows up on no NOVA label.

Good food comes first, and that is the conclusion of the research above. But real life is not always perfect, and a week with more ready meals than planned is ordinary life. That is how we think about a daily supplement at Health Royals: a considered complement to the way you eat, with concentrated nutrients in one daily dose.

Peak Performance is our example of that thinking. It is a multivitamin for women with omega 3 from algae, built around twelve named constituents. OMEGATEX® algal oil at 500 mg gives 375 mg of omega 3, of which 300 mg DHA and 50 mg EPA. Vitamin D3 sits at 25 µg and vitamin K2 at 90 µg as MenaQ7®. Folate is 200 µg as Quatrefolic® 5-MTHF, vitamin B12 is 10 µg as MecobalActive® methylcobalamin, then thiamine 1.1 mg, niacin 16 mg, chromium 20 µg, zinc 5 mg as L-OptiZinc® and selenium 27.5 µg. Two plant extracts complete the list: Hytolive® olive fruit extract at 150 mg, giving 15 mg hydroxytyrosol, the same daily amount used in its randomised, double-blind, placebo-controlled human study, with 450 mg extra virgin olive oil, and Extramel® melon extract at 10 mg, a source of SOD at the dose used in two placebo-controlled human studies in which perceived stress, fatigue and sleep-related problems were measured, which we have written about separately. The dose logic follows the ingredient: the plant extracts sit at their studied dose, the vitamins and minerals at a nutritional level chosen for daily use over years. For us, the ingredient cannot be separated from its source, manufacturer and studied dose. It is the concentrated part of a routine whose larger part is what you cook. That is also why the amber glass jar with its scent core is made to stand on the kitchen counter, with refill pouches after the first jar, two jars a year instead of twelve.

The nutrients that recur in the Swedish survey data also have a settled place in European nutrition, which matters when a company decides what belongs on a label.

For these nutrients the EU has authorised health claims. Vitamin D contributes to the normal function of the immune system and to the maintenance of normal bones. DHA contributes to the maintenance of normal brain function, at an intake of 250 mg DHA a day. Niacin and vitamin B12 contribute to normal energy-yielding metabolism, and niacin, folate and vitamin B12 contribute to the reduction of tiredness and fatigue. Nothing in them says anything about processing, which is the point: the measurable question in a diet high in ultra-processed food is whether the nutrients arrive.

Why do Swedish and Nordic authorities not give advice based on processing?

Because both the Nordic Nutrition Recommendations and the Swedish Food Agency concluded that the category adds nothing to advice already framed around foods and nutrients.

The Nordic Nutrition Recommendations 2023 devote a chapter to the question. The committee decided not to formulate any specific recommendation on ultra-processed food, and concluded that categorising foods as ultra-processed does not add to the classifications and recommendations already in NNR2023, while acknowledging strong evidence for an association with weight gain and obesity. The background scoping review for NNR2023 judged the evidence strong enough to support advice limiting intake, and the committee declined to write it. Both documents are public, and the disagreement between them is the honest state of the field.

Livsmedelsverket followed the same logic. Its 2025 report on general dietary advice for adults includes a chapter on ultra-processed foods, but the advice itself is unchanged and is framed around energy dense and nutrient poor foods. On additives, the European Food Safety Authority assesses each approved additive against an acceptable daily intake, and EFSA has re-evaluated more than 70 percent of the 315 additives approved before 2009.

What can you do with this in practice?

Use the ingredient list, but read it for the things that are actually measurable.

  • Read past the group number. Salt, sugar, fibre, energy density and how much of your week is home cooked say more than whether a product falls in group 4. A fortified wholegrain cereal and a soft drink are both ultra-processed, and treating them as equivalent is a common misreading of this research.
  • Move some meals home. Both controlled trials point the same way, and cooking more of your own food is a reasonable response.
  • Keep an eye on the nutrients that go missing. Fatty fish, eggs, pulses and leafy greens are the foods group 4 displaces, and they are where vitamin D, folate and omega 3 come from.
  • If you add a supplement, add one with a reason for every ingredient. A named raw material, a named manufacturer, a dose with a documented rationale, and a company that shows what has been studied.

Our article on the evidence behind anti-inflammatory eating covers a whole dietary pattern rather than a single category, and our wellbeing page collects the rest.

We have done the selection so that you do not have to. Peak Performance carries the omega 3, vitamin D, folate and the other nutrients above at the amounts listed, with the doses and their reasons published. If the check holds, you can stop looking.

See Peak Performance Formula

Further reading

Sources