Ultra-Processed Food: What It Is, What It's Doing to Us, and the Case for Cooking from Scratch

More than half of the calories the average Briton eats now come from ultra-processed food. Not "processed", ultra-processed, and the difference is the whole point of this piece. Because a homestead kitchen is built on processing: freezing, bottling, fermenting, baking, churning and preserving are all ways of processing food, and they're good, old and useful. What's new, and what the evidence is increasingly worried about, is the industrial version: formulations built from refined substances and additives you would never keep in your own cupboard. Here's the honest picture of what that's doing to us, the one experiment that cuts through the noise, and why the homestead answer holds up even while the scientists are still arguing the details.

The experiment that cut through the noise

Most nutrition debate is a fog of observational studies, so start with the one piece of hard experiment that changed the conversation. In 2019, researchers at the US National Institutes of Health had twenty adults live on a ward where every meal was provided. For two weeks they ate an ultra-processed diet, and for two weeks a minimally-processed one. Crucially, the two diets were matched for calories offered, sugar, fat, salt and fibre, and people could eat as much or as little as they liked.

On the ultra-processed food, they spontaneously ate about 500 calories a day more, and gained around a kilo. On the minimally-processed food, they lost about the same. Same nutrients on paper, very different outcome. It is the clearest evidence we have that ultra-processing itself, the softness, the speed you can eat it, the engineered palatability, drives people to overeat, quite apart from what's on the label. It's a small, short study, twenty people over a month, and it needs replicating at scale. But it removes the usual excuse that "it's just the salt and sugar", and that's why it matters.

What "ultra-processed" actually means, and what it doesn't

This is the distinction everything hangs on, so let's be precise. Nutrition scientists use a system called NOVA that sorts food not by its nutrients but by how much it has been processed, into four groups.

Here's the part the scary headlines usually skip. Groups one to three, unprocessed foods, basic cooking ingredients like butter and oil, and foods made by ordinary kitchen methods, are not the concern. And group three is the homestead larder: tinned tomatoes and pulses, home-style bread, cheese, cured fish, jam, fermented vegetables, most things you'd bottle or preserve yourself. NOVA does not target any of that. Only group four, the industrial formulations, is the problem. So when we talk about cutting ultra-processed food, we are not telling you to stop preserving your glut or baking your own loaf. We're telling you to do more of exactly that, and less of the factory version.

More than half your plate

The scale is genuinely striking, and it's what makes this a national issue rather than a personal quirk. Analysis of the UK's official National Diet and Nutrition Survey found that around 57% of the average adult's calories come from ultra-processed food. For teenagers it's higher, roughly two-thirds. The UK has one of the most ultra-processed diets in Europe.

And it's higher still in lower-income households, which matters for how we talk about it. Ultra-processed food is cheap, shelf-stable, needs no cooking skill and no time, and dominates the shelves where fresh food is hard to reach. That's a food-system outcome, not a failure of willpower, and it's the reason this piece is about skills and the system, never about blame.

The weight of the evidence

Beyond that one trial, the observational picture is large and consistent. A 2024 umbrella review in the BMJ pooled data on nearly ten million people and found higher ultra-processed intake was linked to 32 different adverse health outcomes. The links the reviewers rated strongest included roughly a 50% higher risk of dying from heart disease, meaningfully higher rates of anxiety and common mental disorders, and about a 12% higher risk of type 2 diabetes, with a suggested rise in all-cause mortality too.

We have to be straight about what that is and isn't. It is a very large, consistent association across dozens of studies. It is not proof of cause, because it's observational: people who eat the most ultra-processed food tend to differ in income, smoking and exercise, and untangling the processing from those is hard. Even the UK government's own advisory committee, SACN, put it carefully in 2023, calling the associations "concerning" but the evidence "mostly observational" and urging caution about whether the harm comes from ultra-processing itself or simply from the salt, sugar and fat. That's the honest state of play: strong and growing, not settled.

It's not just the salt and the sugar

The most interesting thread is why ultra-processed food seems to cause harm even when the nutrients match. Part of the answer is physical. Ultra-processed food is usually soft and energy-dense, so you eat it faster and take in more before your body's fullness signals catch up. Grinding, extruding and reconstituting destroys the natural structure of food, which speeds digestion and blunts satiety. And these products are engineered to be moreish, tuned combinations of fat, salt, sugar and texture that encourage you to keep going. The exact recipe of which factor matters most is still being worked out, but the direction is clear: form, not just content.

The additives worth knowing about, graded honestly

Additives are where credibility is won or lost, so here's each on its own merits rather than lumped together into a scare.

Additive / group	What the evidence actually shows	How worried to be
Nitrites in processed meat (bacon, ham, sausages)	The WHO's cancer agency classes processed meat as a Group 1 carcinogen; about 50g a day (two rashers) is linked to roughly 18% higher bowel-cancer risk	Genuine and well-established, though the absolute risk rise is modest. A real reason to make processed meat a treat, not a staple
The 'Southampton Six' food colours (E110, E104, E122, E129, E102, E124)	A UK government-funded trial linked them, with a preservative, to increased hyperactivity in children; the law now requires a warning label	Worth avoiding for children especially. Effects were modest and varied, and many UK makers reformulated them out
Emulsifiers (e.g. carboxymethylcellulose, polysorbate-80)	Animal studies and one small human trial suggest they can disturb the gut lining and microbiome; early days	Emerging, not proven in humans. One to watch, and a nice contrast with fermenting, which feeds the gut
Aspartame (artificial sweetener)	Classed a 'possible' carcinogen by IARC in 2023 on limited evidence; the WHO's intake committee left the safe daily limit unchanged, and you'd need something like 9 to 14 cans of diet drink a day to exceed it	Low. A hazard flag, not a red alert at normal intakes. Don't let anyone scare you with this one

The point of grading them honestly is that a couple are genuinely worth acting on, and a couple are routinely overstated. A brand that cries wolf on aspartame loses the right to be believed about the emulsifiers.

The honest counter-argument

A credible case makes the other side's argument well, so here it is. The NOVA system is arguably too blunt: it sorts by processing, not nutrition, so a fibre-rich wholemeal supermarket loaf and a tin of baked beans land in the same "ultra-processed" group as cola and crisps, which is obviously silly on nutrition alone. Most of the health evidence is observational and tangled up with income and lifestyle. And "processed is bad" is a genuine oversimplification, since processing has fed populations, cut waste and reduced food poisoning.

We take all of that seriously, and we don't need to overclaim, because three things survive the scepticism. The regulator itself agrees the associations are real and consistent; it's arguing about cause, not existence. The controlled trial removes the main objection, because when you match the nutrients exactly, people still overeat the ultra-processed food. And, most usefully, the homestead answer is robust to the whole debate: whether the harm turns out to be the ultra-processing, the additives, the nutrients or the confounders, cooking more of your own food from simple ingredients improves every one of those at once. You don't have to win the mechanism argument to be right about the kitchen.

The homestead answer

None of this needs to be daunting, and it certainly isn't about perfection. It's about quietly taking back a few of the skills the food industry took over.

Doing more of your own processing, less of theirs

1. 1

   Cook a bit more from scratch

   The single highest-leverage move. Even swapping a few regular ready-made buys for home-cooked versions shifts a chunk of your diet out of group four. Start with our from-scratch basics.

2. 2

   Batch-cook and freeze

   Freezing is good processing. A Sunday cook-up and a full freezer is the real answer to 'no time midweek', and it's cheaper per portion than ready meals.

3. 3

   Preserve the glut

   Bottling, fermenting, jam and chutney turn a seasonal surplus into a stocked larder with no industrial additives. Fermenting even adds good gut microbes, the opposite of what the emulsifiers may disturb.

4. 4

   Bake your own bread

   Home bread sidesteps the emulsifiers and preservatives in most packaged loaves, and it's one of the most satisfying skills to have.

5. 5

   Read the label, and don't obsess

   Use the home-kitchen test: a short list of words a cook would recognise is the green flag. You don't have to be perfect, just tilt your regular buys the right way.

To do more of this, our from-scratch basics show you how to make everyday staples like butter, yoghurt, soft cheese and stock, our guide to batch cooking and freezer meals covers the low-effort midweek strategy, and the preserving and fermenting sections turn a glut into a full larder.

But let's be honest about who can

One last thing, and it's not a footnote. Cooking from scratch is not equally easy or cheap for everyone. It takes time, skill, kitchen kit, and energy, both the gas and electric to cook and the personal energy left after a long shift, and for people on tight budgets or in areas with poor access to fresh food, the ultra-processed option is genuinely the practical one. That is a problem with the food system, not with the people caught in it. So nothing here is about judging anyone for what's in their trolley. It's about offering the skills, and the cheapest, lowest-effort versions of them, to anyone who wants to reclaim a bit more of their own kitchen. Eat what you can manage, cook what you can, and don't let anyone, including us, make you feel bad about a ready meal on a hard week.