Are ultra-processed foods bad for you?
The strongest evidence is not about disease, it is about appetite. In a controlled inpatient trial where both menus were matched for calories presented, sugar, fat and fibre, people ate about 500 kcal more per day on the ultra-processed menu and gained roughly 0.9 kg in a fortnight[1]. Long-term cohort data links a high share of these products with worse health outcomes, but cannot prove the food caused it[3]. The useful conclusion is practical: these products are easy to overeat, so build the day around food cooked from recognisable ingredients and keep packaged items at the edge of the plate.
On this page
What "ultra-processed" actually means
The phrase is not a synonym for "junk food", and it is not about how much a food was cooked. It comes from a classification that sorts food by the kind of industrial processing behind it, splitting everything into four groups: unprocessed or minimally processed foods, culinary ingredients such as oil and salt, simple processed foods such as cheese or canned legumes, and ultra-processed formulations[2].
The fourth group is defined by its ingredient list. These are products made largely from substances extracted from foods — protein isolates, modified starches, hydrogenated oils, invert sugars — held together by additives whose job is colour, flavour, texture or shelf life. The practical test in the original definition is simple: look for an ingredient you would never find in a domestic kitchen[2]. By that test, mass-produced packaged bread, flavoured yoghurt, instant noodles, packaged biscuits, processed meat and soft drinks are in the group, while plain bread from a local bakery, rice, lentils, eggs, fresh meat and unsweetened dairy are not.
The one study that tested cause rather than correlation
Almost everything written about this subject rests on surveys. One experiment did better. Twenty adults lived inside a research unit for a month and spent two weeks on an ultra-processed menu and two weeks on a minimally processed one, in random order. The two menus were matched for calories presented, sugar, fat, carbohydrate, fibre and sodium, and people were told to eat as much or as little as they liked.
On the ultra-processed menu they ate about 500 kcal more per day and gained roughly 0.9 kg over the fortnight; on the minimally processed menu they lost about the same[1]. Nobody was dieting and nobody was told to restrain anything. The composition of the food, not willpower, moved intake.
What the long-term evidence can and cannot say
The health claims you see quoted come from cohort studies, which follow large groups for years and ask what they ate. In a French cohort of more than one hundred thousand adults, each 10% rise in the share of ultra-processed food in the diet was associated with about a 12% higher rate of cardiovascular disease[3]. A meta-analysis pooling several cohorts put each 10% rise alongside a 15% higher risk of death from any cause[4].
Those are associations, and people who eat the most of these products also tend to smoke more, move less and earn less, so some of the signal belongs to the life around the food rather than the food itself. The honest summary is that the experimental evidence shows these products make people eat more, and the observational evidence shows that eating a lot of them travels with worse outcomes. That is enough to act on without pretending it is proof of a mechanism.
Why it happens
Three properties do most of the work, and all three are measurable rather than mysterious. The first is energy density: more calories packed into each mouthful, so a normal-feeling portion carries more than it looks like it does. The second is eating rate — participants in the trial ate the ultra-processed meals faster, which gives the signals that end a meal less time to arrive before the plate is empty[1]. The third is what the formulation leaves out. These products tend to be lower in protein and fibre, the two components that do most for fullness per calorie, which is why the same number of calories from them keeps you satisfied for less time. The guide on how much protein you actually need covers that side of it.
What to do about it in a real kitchen
Nothing here calls for a clean-eating rulebook. Three changes carry most of the benefit for most households. Make the base of the day cooked from recognisable ingredients — rice, bread from the bakery, legumes, vegetables, eggs, meat, dairy — and treat packaged products as the edge of the plate rather than its centre. Attack drinks first, because a soft drink is the easiest ultra-processed calorie to remove without missing it. And read the ingredient list rather than the front of the box; our piece on how to read a nutrition label covers the part that is actually regulated.
If you are tracking, keep the arithmetic honest: packaged food is the easy half to log, because the numbers are printed, and home cooking is where most estimates drift. The guide on how your daily calories are calculated explains what you are comparing against, and a calorie counter that works inside WhatsApp is a short enough step that a packet logged at the shop and a pot logged at home end up in the same place. What you will usually find is not a moral story about processed food, but a plain one: the days with more of it are the days you ate more.
Sources
- Hall KD, Ayuketah A, Brychta R, et al. Ultra-processed diets cause excess calorie intake and weight gain: an inpatient randomized controlled trial of ad libitum food intake. Cell Metabolism 2019;30(1):67–77.e3. ↩
- Monteiro CA, Cannon G, Levy RB, et al. Ultra-processed foods: what they are and how to identify them. Public Health Nutrition 2019;22(5):936–941. ↩
- Srour B, Fezeu LK, Kesse-Guyot E, et al. Ultra-processed food intake and risk of cardiovascular disease: prospective cohort study (NutriNet-Santé). BMJ 2019;365:l1451. ↩
- Pagliai G, Dinu M, Madarena MP, et al. Consumption of ultra-processed foods and health status: a systematic review and meta-analysis. British Journal of Nutrition 2021;125(3):308–318. ↩