Ultra-processed foods: what they are and what the evidence really shows

Ultra-processed foods, or UPFs, are not simply foods that have been cooked, frozen or tinned. The term comes from a specific classification called NOVA and points to industrial formulations built mostly from refined substances and additives you would not keep in a home kitchen. The honest position is that the evidence is genuinely mixed: one tightly controlled trial suggests these foods can make us overeat, a large body of weaker observational research links them to worse health, and the UK's own expert committee says the concept is still hard to pin down. This guide separates what is reasonably well established from what is plausible but unproven, without moralising about anyone's shopping basket.

What "ultra-processed" actually means

The word "processed" is doing a lot of unhelpful work in everyday conversation. Freezing peas, tinning beans, pasteurising milk and fermenting yoghurt are all forms of processing, and none of them are the problem. The useful distinction comes from the NOVA classification, devised in 2009 by the Brazilian epidemiologist Carlos Monteiro and colleagues at the University of Sao Paulo. NOVA sorts foods not by their nutrients but by the nature, extent and purpose of the industrial processing behind them.¹

Group 4, the ultra-processed group, is the one that has attracted the headlines. Monteiro's working definition is that these are industrial formulations made mostly or wholly from substances extracted or refined from foods, such as protein isolates, modified starches, hydrogenated oils and high-fructose corn syrup, combined with cosmetic additives like emulsifiers, colours, flavourings and non-sugar sweeteners.¹ A rough but serviceable test: read the ingredients list, and if it is long and contains substances you would never find in a domestic kitchen, the product is probably ultra-processed.

It is worth naming the obvious weakness early. The category is very broad: a sugary fizzy drink and a fortified wholemeal supermarket loaf can both count as ultra-processed, even though their nutritional profiles are worlds apart. That breadth is exactly why defining and applying the concept has proved contentious, a point the UK's expert committee leans on heavily.⁷

The strongest single study: Hall's metabolic-ward trial

If you read one piece of evidence on UPFs, make it this one, because it is the only randomised controlled trial designed to isolate processing itself. In 2019 Kevin Hall and colleagues at the US National Institutes of Health admitted 20 weight-stable adults to a metabolic ward and fed them, in random order, two weeks of an ultra-processed diet and two weeks of a minimally processed one.² Crucially, the two diets were matched in advance for the calories presented, energy density, total fat, sugar, salt and fibre. Participants could eat as much or as little as they liked at each meal.

On the ultra-processed diet, people ate about 508 kcal more per day and gained roughly 0.9 kg over the fortnight; on the minimally processed diet they ate less and lost about the same amount.²³ Because the menus were matched on paper, the extra eating could not simply be blamed on UPFs being more calorie-dense gram for gram. Something about the foods themselves, or how they were eaten, drove the overeating.

This is a genuine landmark, but it is not the last word, and Hall himself has been careful about that. The trial had only 20 people, ran for four weeks, and took place in an artificial inpatient setting where every meal was supervised. It cannot easily be blinded, and it has not yet been replicated at scale. Treat it as strong proof of concept that UPFs can promote overeating, not as proof of the size of the effect across a whole population eating freely for years.

The observational links, and the confounding problem

Beyond that single trial, almost all the evidence is observational: large groups of people are tracked over years, and those eating more UPFs are compared with those eating fewer. The French NutriNet-Sante cohort is among the most cited. In over 105,000 adults, each 10 percentage-point rise in the share of UPFs in the diet was associated with about a 12 percent higher risk of cardiovascular disease.⁵ In a parallel analysis of nearly 105,000 adults, the same 10-point rise was linked to a 15 percent higher risk of type 2 diabetes.⁶

The broadest summary so far is the 2024 umbrella review led by Melissa Lane, published in the BMJ, which pooled 14 meta-analyses covering 45 separate analyses and close to 10 million participants.⁴ It reported convincing or highly suggestive associations between higher UPF intake and all-cause mortality (about 21 percent higher), deaths from cardiovascular disease, common mental disorders such as anxiety, and being overweight.⁴ That sounds damning, and the consistency across outcomes is striking. But the same review graded most of the evidence as low or very low certainty.

Why might UPFs drive overeating?

There is no single agreed mechanism, but several plausible and overlapping hypotheses, ranked here roughly by how much experimental support each has.

Energy density and hyper-palatability

UPFs tend to pack more calories into each mouthful and are engineered to hit combinations of fat, sugar and salt that are easy to keep eating past the point of fullness. This is the most established lever, and it overlaps closely with how UPFs are formulated and marketed.²

Soft textures and fast eating

Soft, easy-to-chew foods are eaten faster, and faster eating means more is swallowed before the brain registers fullness. In controlled work led by Ciaran Forde's group, ultra-processed meals were eaten at a markedly higher rate of calories per minute, and slowing the eating rate cut energy intake by around a quarter.⁹ Eating rate now looks like one of the better-supported explanations for Hall's 500-calorie gap.

Displacement of whole foods

The more of the plate UPFs occupy, the less room there is for vegetables, legumes, whole grains and the fibre and micronutrients they carry. Some of the apparent harm of UPFs may simply be the harm of a diet low in minimally processed food, a point our guide to gut microbiome diversity picks up, since fibre is the main fuel for a healthy gut community.

Additives and the gut lining

This is the most speculative strand. Some emulsifiers used to improve texture and shelf life may disturb the gut. In a small randomised controlled-feeding study, the emulsifier carboxymethylcellulose altered the gut microbiota and reduced its diversity in some participants over just 11 days, with a couple of people showing more marked changes.¹⁰ This is mechanistically interesting and biologically plausible, but it is preliminary, involved tiny numbers, and is a long way from proving that everyday additive exposure causes disease.

The UK view: what SACN concluded in 2023

In July 2023 the Scientific Advisory Committee on Nutrition, the body that advises the UK government, published a position statement after reviewing the evidence.⁷ Its conclusions are worth reading in full, because they are more measured than most headlines. SACN agreed that the observed associations between UPFs and poor health are concerning. But it raised three serious reservations.

• The evidence is almost entirely observational, with the limitations described above, and should therefore be treated with caution.⁷
• The NOVA concept is difficult to define and apply consistently, grouping very different products together, which makes firm conclusions hard.⁷
• It is unclear whether any harm comes from processing itself, or simply from the fact that many UPFs are energy-dense and high in saturated fat, salt and free sugars and low in fibre, or from UPF intake being a marker of an otherwise less healthy lifestyle.⁷

For that reason SACN did not introduce new UPF-specific limits. Its practical point is reassuring in its simplicity: existing UK advice, the Eatwell Guide, already steers people toward the same foods, more vegetables, fruit, wholegrains and pulses, and less of the sugary, salty, fatty products that dominate the ultra-processed category. The committee called for better research and a UK-applicable classification rather than alarm.

A practical, non-perfectionist bottom line

The least useful response to all this is fear, or a hunt for dietary purity. Not all UPFs are equal: a wholemeal supermarket loaf, baked beans and a fortified plain yoghurt sit very differently from a daily habit of sugary drinks, processed meats and confectionery. Chasing a zero-UPF diet is impractical for most people on a budget and a schedule, and the evidence does not justify treating every packaged food as poison.

A more honest and achievable goal is to shift the ratio of your diet toward minimally processed foods most of the time. Build meals around vegetables, fruit, pulses, whole grains, eggs, fish and plain dairy; keep the most calorie-dense, fast-to-eat products as the smaller part of the week rather than the backbone of it; and slow down at the table. These same moves underpin our guides to metabolic health and losing fat while keeping muscle, because the mechanisms overlap. If you would rather map your own diet and goals to specific actions, the stack builder can help you organise the picture, and the start here guide shows how to sequence changes without overhauling everything at once.

What to ask your GP

What to do next

This article is educational and does not constitute medical advice, diagnosis, or a treatment recommendation. Medication uses described as “off-label” are not licensed for that purpose in the UK and should only be considered under qualified clinical supervision. Always speak to your GP, pharmacist, or a registered specialist before starting, stopping, or changing any treatment. If you have severe or alarm symptoms - unintentional weight loss, blood in your stool, difficulty swallowing, persistent vomiting, a fever, or severe pain - seek urgent medical care.