What is ultra-processed food and why does it matter?

Breakfast. A bowl of flavoured cereal, a carton of orange juice, a yoghurt with fruit at the bottom. Lunch on the go: a packaged sandwich, a bag of crisps, a soft drink. An afternoon snack from the vending machine. Dinner from a ready meal, or pasta with sauce from a jar.

Nothing unusual. Millions of people eat roughly like this every day, in every country, at every income level. And yet almost everything in that list belongs to the same food category, one that, until about fifteen years ago, did not even have a name. A category that a landmark series of studies published in The Lancet in November 2025 linked to damage across nearly every organ system in the human body.

That category is ultra-processed food. Understanding what it actually means, not in the vague, guilt-ridden sense that most nutrition content uses but in the precise, scientific sense, changes the way you read every label, every ingredient list, every supermarket shelf.

The NOVA classification: what was done to your food

To understand ultra-processed food, you need to understand the framework that defines it. In 2009, Carlos Monteiro and his research team at the University of São Paulo developed a classification system called NOVA. Its premise is simple but radical: what matters about food is not just what is in it, but what was done to it.

NOVA divides all food into four groups based on the extent and purpose of processing. It is worth understanding each one, because the differences between them are not intuitive.

Group 1 — Unprocessed or minimally processed foods. Fresh fruit, vegetables, meat, fish, eggs, plain milk, legumes, whole grains. These may have been cleaned, dried, frozen, or pasteurised, but their fundamental nature has not changed. Nothing significant has been added.

Group 2 — Processed culinary ingredients. Butter, oils, flour, sugar, salt, vinegar. Substances extracted from Group 1 foods, used in cooking. You would not eat them on their own, but they are the tools that turn raw ingredients into a meal.

Group 3 — Processed foods. This is where simple industrial transformation begins: adding salt, sugar, or oil to Group 1 foods to extend shelf life or enhance flavour. Canned vegetables, cheese, cured meats, freshly baked bread. The ingredient list is short. You can still identify the original food inside the product.

Group 4 — Ultra-processed foods. This is where the industrial logic changes entirely.

Ultra-processed foods are not simply "more processed" versions of Group 3. They are fundamentally different in purpose and construction. As defined by the NOVA framework, they are

"industrial formulations made with little or no whole food, combining substances derived from foods with additives that have no equivalent in a home kitchen."

The practical test is blunt: look at the ingredient list. If you cannot identify the original food source of most ingredients, or if the list reads like a chemistry inventory, you are almost certainly looking at a Group 4 product.

But there is a subtler point here, one that most discussions miss entirely. The problem with ultra-processed food is not primarily any single ingredient. It is the system they create together. These products are engineered to deliver a specific sensory experience that whole foods cannot replicate: a precise ratio of salt, fat, and sugar; a texture that dissolves effortlessly; a flavour that hits immediately and fades just fast enough to make you reach for more. This is not accident. It is design.

November 2025: when the evidence stopped being deniable

Debates about processed food have been circling for years. What changed in November 2025 was the scale and authority of the evidence assembled in one place.

The Lancet Series on Ultra-Processed Foods and Human Health, published on 18 November 2025, is a three-paper body of work led by Carlos Monteiro and co-authored by researchers from institutions across four continents, including the London School of Hygiene and Tropical Medicine, the University of North Carolina, and the Pan American Health Organization. It was funded by Bloomberg Philanthropies and represents the most comprehensive synthesis of UPF research to date.

The first paper combines systematic reviews, meta-analyses, and original analyses to assess the health consequences of diets rich in ultra-processed food. The second examines how food environments, commercial actors, and global supply chains have driven the rise of UPFs worldwide. The third sets out the policy frameworks needed to address the problem: taxation, marketing restrictions, and subsidies for fresh food.

What makes this Series scientifically significant is the convergence of methods. The authors drew on large observational cohorts including the UK Biobank and the NutriNet-Santé study, as well as controlled feeding trials, a methodologically stronger design that can more directly test causation. In one of those trials, participants eating an ultra-processed diet consumed an average of roughly 500 kilocalories more per day and gained weight compared to those eating minimally processed food, even when both diets were matched for macronutrients and rated equally palatable. The food's physical and chemical structure, not just its calorie content, was driving overconsumption.

The Series identified elevated risk across 12 distinct health outcomes associated with high UPF consumption, including type 2 diabetes, cardiovascular disease, chronic kidney disease, depression, impaired cognitive performance, and all-cause mortality. The lead author, Carlos Monteiro, summarised the findings at the London launch by stating that the evidence indicates ultra-processed foods harm every major organ system in the human body.

It is important to be precise about what this means. The majority of the evidence remains observational, meaning it demonstrates association rather than proven causation. Populations that eat more ultra-processed food also tend to have worse overall diets, lower incomes, and higher exposure to other health risks. Disentangling UPFs from these confounding variables is genuinely difficult. What the Series provides is a convergence: the associations hold across dozens of studies, multiple countries, and different research methodologies, and the controlled trials offer a plausible biological mechanism. At a certain point, the burden of proof shifts.

What these foods actually do inside you

The mechanisms through which ultra-processed food affects health are still being mapped, but several are well-documented. The first paper in the Lancet Series identifies four main pathways.

The first is passive overconsumption. Ultra-processed foods are engineered for high energy density and rapid eating. Their texture is soft, uniform, and requires minimal chewing, which means you consume them faster than whole foods. Your satiety signals, which take roughly twenty minutes to reach your brain after eating begins, have no time to catch up. You eat more than you needed before your body registers that you have eaten enough.

The second is displacement of protective foods. A diet dominated by UPFs is, by definition, a diet with less room for vegetables, legumes, whole grains, and other foods associated with reduced chronic disease risk. The harm is partly what these foods do, and partly what they replace.

The third pathway involves exposure to potentially harmful additives. Some emulsifiers, for example, have been linked in experimental studies to disruption of the gut microbiome and low-grade intestinal inflammation. This research is still developing, and it would be premature to draw firm conclusions about specific additives. But the cumulative, chronic exposure to complex mixtures of food additives, at concentrations that have never existed in human diets before the industrial food era, is a legitimate area of scientific concern.

The fourth involves contaminants introduced by industrial processing itself: compounds formed during high-heat manufacturing, migration from packaging materials, and endocrine-disrupting substances found in certain plastic food containers. Again, the science here is evolving, but the direction of evidence is not reassuring.

The industry angle: this is not an accident

Here is what most nutrition content quietly omits.

Ultra-processed foods are not the incidental result of companies trying to feed people conveniently. They are the deliberate output of an industrial logic optimised for profit, shelf life, scalability, and above all repeat purchase. Food technologists work explicitly to engineer what the industry internally calls "bliss points": combinations of salt, sugar, fat, and texture calibrated to maximise palatability while minimising satiation. The goal is a product you continue eating past the point of hunger, and return to buy again tomorrow.

The Lancet Series documents this in detail. Ultra-processed food companies spend billions annually on marketing, disproportionately targeting children and lower-income consumers. They fund research designed to cast doubt on unfavourable findings. They lobby against regulation. When mandatory limits on sugar or fat force reformulation, products are adjusted to stay below regulatory thresholds while remaining structurally ultra-processed. A reduced-sugar breakfast biscuit with seventeen ingredients is still a Group 4 product. The NOVA framework cuts through this precisely because it focuses on the nature of the product rather than its nutrient profile, making it much harder to game.

This is not a conspiracy theory. It is documented commercial strategy, described in industry communications, academic analyses of corporate political activity, and the Lancet Series itself.

How to recognise ultra-processed food in practice

You do not need a food science degree to apply NOVA logic in a supermarket. A few reliable signals.

The ingredient list is your primary tool. Ingredients are listed by descending weight. If the first several ingredients are substances you would not find in a home kitchen (maltodextrin, soy protein isolate, modified starch, carrageenan, xanthan gum), you are almost certainly looking at a Group 4 product. The length of the list alone is a rough indicator: a real food has a short one.

Look for clusters of additives. Emulsifiers (lecithin, mono- and diglycerides, carrageenan), stabilisers, "natural flavourings" or "nature-identical flavourings", artificial colourings, and non-sugar sweeteners are characteristic of ultra-processed formulations. One or two additives might indicate Group 3. A dense cluster almost always signals Group 4.

Health claims on the front of the packaging have no bearing on NOVA classification. "Organic", "natural", "no added sugar", "high protein", "source of fibre": none of these tell you anything about how the product was made or what industrial substances it contains. Some of the most aggressively health-marketed products in any supermarket are unambiguously ultra-processed.

Pay attention to texture. Ultra-processed foods are engineered for uniformity: soft, dissolving, consistent from one unit to the next. Real bread stales and has an irregular crumb. Real cheese has an uneven texture. Real yoghurt separates. Industrial uniformity is a signal worth noticing.

This connects naturally to the next question most people ask once they start paying attention: what are all these ingredients actually doing, and how do I read a label properly? That is the subject of our next piece.

The part nobody wants to say out loud

There is a conversation about ultra-processed food that the mainstream health discourse consistently avoids: this is partly a class issue.

Fresh, minimally processed food costs more, perishes faster, and takes longer to prepare. For a single parent working irregular hours, or a student managing a tight budget, the practical calculus around food is entirely different from that of someone with time, money, and a well-stocked kitchen. The Lancet Series acknowledges this explicitly. Its policy recommendations are structural rather than individual: taxation of UPFs, marketing restrictions, subsidies for fresh food, and redesign of food environments.

The framework implicitly recognises something that individual-focused nutrition advice rarely does: telling people to eat better without addressing the economic conditions that make ultra-processed food dominant is not public health. It is the displacement of a systemic problem onto the people least equipped to solve it alone.

Understanding ultra-processed food is not about dietary guilt or lifestyle optimisation. It is about understanding a system: how it was built, who benefits from it, and what it was designed to do to you. That knowledge does not automatically change what ends up on your plate. But it changes what you are looking at when you stand in a supermarket aisle.

A note from Daitymon

We live in an era in which the most sophisticated technology in human history is routinely deployed to engineer food that is cheaper to produce, longer-lasting on a shelf, and precisely calibrated to override the signals your own body sends you about when to stop eating. This is presented, more or less without irony, as progress. The least we can do is understand what we are dealing with.

Our next piece tackles food labels directly: what the numbers mean, what the claims do not say, and what to actually look for.