Ultra-processed food and what to make of it

Ultra-processed food has become the dominant nutritional debate of recent years, with genuinely important evidence and a definition that does not always do what people want it to.

The classification

The NOVA system groups foods by degree and purpose of processing rather than by nutrient content.

Group one covers unprocessed and minimally processed foods.

Group two covers culinary ingredients such as oil, salt and sugar.

Group three covers processed foods — bread, cheese, tinned vegetables, cured meat — made by combining groups one and two.

Group four covers ultra-processed foods: industrial formulations typically containing ingredients not used in home cooking, such as protein isolates, modified starches, emulsifiers, sweeteners, colourings and flavourings.

Which places mass-produced bread, breakfast cereals, flavoured yoghurts and plant-based meat substitutes alongside confectionery and soft drinks — a grouping that is the main criticism of the system.

What the evidence shows

Substantial and worth taking seriously.

Large prospective cohort studies across multiple countries have associated higher ultra-processed food consumption with cardiovascular disease, type 2 diabetes, obesity, depression and all-cause mortality.

Umbrella reviews of this literature have found consistent associations across many outcomes, with the usual caveats of observational research.

The single most informative study is a controlled feeding trial in which participants ate ultra-processed or minimally processed diets matched for calories, sugar, fat, fibre and macronutrients — and consumed several hundred more calories a day on the ultra-processed diet, gaining weight.

That trial was small and short and it is the strongest evidence that something about the food itself, rather than only its nutrient content, affects intake.

The proposed mechanisms

None fully established.

Energy density and softness, which allow faster eating and less satiety per calorie.

Hyperpalatability from specific combinations of fat, sugar and salt.

Displacement of whole foods and fibre.

Effects of emulsifiers and sweeteners on the gut microbiome, which is an active research area.

Compounds from packaging and processing.

And the food environment — availability, price and marketing — which promotes exactly these products.

The criticisms of the framework

Which are legitimate.

The category is heterogeneous: wholemeal supermarket bread and a fizzy drink are classified together, which is nutritionally implausible.

Classification is not always reproducible between researchers.

Confounding is substantial, since people eating more ultra-processed food differ in income, education, smoking and activity.

Some fortified processed foods make meaningful nutritional contributions at population level.

And advice to avoid ultra-processed food entirely is not achievable on a low income or with limited time, which risks becoming another source of guilt without changing anything.

The practical position

Which most nutrition bodies have converged on.

Reduce consumption of the clearly problematic categories: sugary drinks, confectionery, crisps, processed meat and most convenience snacks.

Increase whole and minimally processed foods: vegetables, fruit, pulses, wholegrains, nuts, eggs, fish.

Do not agonise over supermarket bread, tinned beans, frozen vegetables and plain yoghurt, which are convenient, cheap and nutritionally useful whatever their classification.

Cook where you can, and use convenience where you must.

And treat the classification as a rough signal rather than a rule, since the underlying advice — more plants, more fibre, fewer sugary and highly palatable snacks — is unchanged from before the framework existed.

The cost dimension

Which the debate frequently ignores.

Ultra-processed foods are cheaper per calorie than fresh foods in most markets, which is why consumption correlates with income.

They also require no cooking facilities, no skills and no time.

Which means that framing this purely as individual choice ignores the constraints many households face, and that population-level solutions — reformulation, marketing restrictions, pricing and food environment policy — are where the leverage is.

For an individual on a limited budget, pulses, oats, frozen vegetables, eggs and tinned fish are the cheap minimally processed foods, and they compare well on cost with the processed alternatives.