Nutrition, Diet & Fasting - Diets
Ultra-processed food
The single most actionable finding in modern nutrition, and one of the very few tested in a locked metabolic ward: given the same choice, people eat around 500 calories a day more on ultra-processed food.
"Ultra-processed" is a category from a classification system called NOVA, and it means more than "packaged". It describes foods that are industrial formulations rather than modified foods - typically containing ingredients you would not find in a domestic kitchen, such as protein isolates, modified starches, emulsifiers, and cosmetic additives for colour, flavour and texture. A tinned tomato is processed. A supermarket ready meal, a soft drink and most breakfast cereals are ultra-processed.
The category has real critics, and the criticism is fair: it groups a wholemeal supermarket loaf with a fizzy drink, and it describes how a food was made rather than what is in it. Keep that in mind. It matters because the strongest experiment in this whole classroom used the category anyway, and found something hard to explain away.
The experiment
In 2019 a team at the US National Institutes of Health admitted 20 adults to a metabolic ward for a full month. For two weeks they were offered meals made from ultra-processed food and for two weeks meals made from minimally processed food, in random order, with the two diets matched for calories presented, sugar, fat, fibre and salt. Participants could eat as much or as little as they wanted. Everything eaten was weighed.
On the ultra-processed diet people ate about 500 calories a day more, and gained around 0.9 kg over the fortnight. On the minimally processed diet, the same people lost around 0.9 kg. Nobody was told to eat more or less; they simply did. A randomised crossover trial published in 2025, run to healthy dietary guidelines on both arms, found the same direction of effect.
This is an unusual result for nutrition science, because almost everything else in the field relies on people reporting what they ate - which they do badly. Here intake was measured, each person acted as their own control, and the confounders that plague diet research were locked outside the ward. It is a small study, and the effect size is what matters more than the precision.
Why it happens
Nobody is certain, and the honest candidates are mundane. Ultra-processed food tends to be softer and to require less chewing, so it is eaten faster - and eating rate is one of the better-established predictors of how much someone eats before feeling full. It is more energy-dense, packing more calories into less volume and weight. It tends to be engineered for palatability in ways whole foods are not. And it is lower in protein as a percentage of calories, which matters if appetite is partly driven by hunting for protein.
Additives and emulsifiers get blamed a lot online. That is a live research question with some interesting animal work behind it, not an established human finding, and the eating-rate and energy-density explanations currently do more work with less speculation.
Why this is the most useful page in the classroom
Almost every diet that produces results in practice - keto, carnivore, paleo, whole-food plant-based, Mediterranean - removes most ultra-processed food as a side effect of its own rules. That is a large shared mechanism hiding behind a lot of separate arguments, and it suggests the diets have more in common than their advocates do.
It also sets a floor. Someone unwilling to adopt any named diet still has one change available that has been tested under controlled conditions, and it does not require choosing a side in any of the arguments in this classroom.
Where the evidence stops
Twenty people for four weeks is the strongest evidence, not overwhelming evidence. The long-term data linking ultra-processed intake to disease is observational, with the usual problem that people who eat a lot of it differ in income, time and much else. And the category itself remains contested - a criticism that would matter more if the ward experiment had not deliberately matched the two diets on the obvious nutrients and still found the effect.
Educational content only and not medical advice or a recommendation about any specific food or product.
Related in this classroom
Sources for this classroom
Educational content only. Not medical advice, diagnosis or treatment. Always consult a qualified healthcare professional before changing your health regimen.
