Ultra-Processed Food in the UK: The Numbers

Roughly half of what UK adults eat is ultra-processed. Analyses of the UK National Diet and Nutrition Survey put ultra-processed food at about 50 to 57 per cent of adult daily energy intake — among the highest shares in Europe — and UK adolescents are higher still, at around 65 to 66 per cent. Those are the honest headline figures. This page collects the verified UK statistics on ultra-processed food in one place, with every number sourced or left out.

That last part matters. UPF coverage is full of recycled, approximately-remembered numbers. Below, each figure comes from a named study or an official document, and the limits of each are stated alongside it.

What percentage of the UK diet is ultra-processed?

The most-cited answer comes from Rauber and colleagues, who applied the NOVA classificationto the UK National Diet and Nutrition Survey (NDNS) covering 2008 to 2014. They found that ultra-processed food supplied about 56.8 per cent of total energy in the UK diet — and about 64.7 per cent of free sugars.

Other analyses of NDNS data land slightly lower, which is why the honest range is roughly 50 to 57 per cent rather than a single fixed number. The variation comes from different survey years and from judgement calls in coding foods to NOVA groups — more on that below.

For adolescents the picture is starker. UK teenagers get around 65 to 66 per cent of their energy from ultra-processed food. The staples of a typical teenage diet — mass-produced sliced bread, most breakfast cereals, soft drinks, packaged snacks — are typically NOVA 4, so the number is less surprising than it first looks.

UK UPF consumption: the verified numbers

What SACN actually says

SACN — the Scientific Advisory Committee on Nutrition, which advises the UK government — published its position statement on processed foods and health in July 2023. Of eight food-processing classification systems it reviewed, NOVA was the only one it judged potentially suitable for UK use, but it flagged concerns about how consistently NOVA can be applied in practice, and said the mostly observational evidence should be “treated with caution”.

Crucially, SACN did not conclude that ultra-processed food is harmful independent of its nutrient content. That is often misreported. Its April 2025 rapid evidence update reviewed the newer research and held the same position: the associations remain of concern, the limitations remain significant, and “on balance, most people are likely to benefit from reducing consumption of processed foods high in energy, saturated fat, salt and free sugars and low in fibre.”

In other words, official UK advice still runs through the Eatwell Guide and the familiar levers of energy, saturated fat, salt, sugar and fibre — not through a UPF label. There is no NHS or government UPF-specific dietary guidance.

What the health evidence shows — and what it does not

The largest summary of the evidence is the 2024 umbrella review by Lane and colleagues in the BMJ: 14 reviews, 45 pooled meta-analyses, around 9.9 million participants. It found higher ultra-processed food exposure consistently associated with 32 adverse health outcomes, with the strongest evidence grades for cardiovascular-disease mortality, type 2 diabetes, common mental disorders and all-cause mortality.

Those findings are consistent and large-scale, but they are associations from observational cohorts. People who eat more UPF tend to differ in many other ways, and no cohort study can fully untangle that.

The strongest causal signal comes from a single randomised controlled trial: Hall et al., 2019. Twenty inpatients spent two weeks each on an ultra-processed diet and an unprocessed diet, matched for presented calories, sugar, fat, sodium, fibre and macronutrients. On the ultra-processed diet, participants ate about 500 kcal a day more and gained weight. It is a genuinely important result — and it is one small, short-term study. Both halves of that sentence deserve equal weight.

How reliable are these numbers?

Every UPF statistic depends on coders assigning foods to NOVA groups, and that step is less precise than the headline percentages suggest. In a 2022 study by Braesco and colleagues, over 150 experts classified the same foods and reached only “fair” agreement — a Fleiss kappa of about 0.32 for marketed foods and 0.34 for generic foods. The same product can land in different groups depending on the coder and the ingredient detail available.

That does not make the UK figures worthless — a diet that is half ultra-processed by one coder's reading is not suddenly whole-food by another's. But it is a real limit, and it is why serious estimates come as ranges. It is also worth repeating that ultra-processed and unhealthy are correlated, not synonymous: wholemeal sliced bread and fortified plant milks can be NOVA 4 and still nutritionally reasonable, while plain block cheese is NOVA 3 and still high in saturated fat and salt.

What to check on the label

Statistics describe the national diet. Your own basket is decided one label at a time. Two things are worth knowing when you read one:

• Preservatives, antioxidants and stabilisers used for durability do not by themselves make a food ultra-processed — NOVA permits them in group 3. The markers that typically signal NOVA 4 are cosmetic additives (flavourings, colours, non-sugar sweeteners, emulsifiers) and industrial ingredients rarely used at home, such as protein isolates, maltodextrin and invert sugar.
• The group depends on the specific product, not the category. Most supermarket versions of a food may be NOVA 4 while a simpler version on the same shelf is not — so check the label rather than assuming.

If you would rather not memorise marker lists, a UPF scannerdoes the reading for you: scan a barcode and see the product's NOVA group, its additives and its overall score in seconds.

Sources

• UK National Diet and Nutrition Survey UPF analysis — Public Health Nutrition (Cambridge)
• SACN position statement: Processed Foods and Health (July 2023)
• SACN rapid evidence update on processed foods and health (April 2025)
• Lane et al., BMJ 2024 — umbrella review of UPF exposure and health outcomes
• Hall et al., Cell Metabolism 2019 — randomised controlled trial of ultra-processed diets
• Braesco et al., 2022 — reliability of NOVA classification
• British Nutrition Foundation on the SACN position statement