Is oat milk ultra-processed? Usually, yes. Most commercial oat drinks add gums, emulsifiers or flavourings, and under the NOVA classificationany one of those puts a product in group 4 — ultra-processed. The added oil in most cartons is not a marker by itself, but it almost never travels alone. And group 4 is not inevitable. A minimal oat drink made from oats, water and an enzyme can sit in NOVA group 3, and a few sold in the UK do. If you are asking whether oat milk is a UPF, the honest answer is that the recipe decides, and recipes vary more than the cartons suggest.
The same applies across the chiller aisle. The plant milk UPF question — oat, soya, almond, the lot — has no single answer, because NOVA classifies products, not categories. Two cartons of oat drink sitting next to each other on the shelf can land in different groups. Here is how to tell them apart.
Is oat milk ultra-processed? What the ingredient list says
Pick up a typical carton of oat drink in a UK supermarket and the ingredient list runs well past oats and water. Most commercial versions add some combination of: a plant oil (usually rapeseed or sunflower), stabilising gums such as gellan or guar, emulsifiers such as lecithin, and sometimes phosphates used as acidity regulators. Some add flavourings or sugar on top.
Under NOVA, emulsifiers, gums, flavourings and non-sugar sweeteners are classic markers of ultra-processing — ingredients whose job is to build texture, stability and taste in an industrial formulation. One marker is enough for group 4. So most supermarket oat milks are, by the book, NOVA 4. That is the straightforward part of the answer.
What makes a plant milk a UPF — and what does not
The line matters, so it is worth drawing precisely. NOVA does not classify foods by how many additives they contain, and it does not treat every E-number as a red flag. Preservatives, antioxidants and stabilisers used purely to keep a food safe and durable do not, by themselves, make it ultra-processed — Monteiro and the FAO explicitly leave foods with only those additives in group 3. What tips a product into group 4 is the presence of ingredients rarely used in home kitchens, or additives whose purpose is cosmetic: to imitate or enhance flavour, colour and texture.
For oat drinks, that means the gellan gum and the lecithin are the deciding ingredients. The added oil is not — oil is a culinary ingredient under NOVA, in a drink as in a pan — but it is the giveaway that you are holding an industrial formulation, and it almost never travels without the gums that do decide. Added vitamins and minerals are not deciding either. Fortification does not move a product between NOVA groups either way — a point that matters for plant milks more than almost any other category, as we will get to.
One more honesty note: NOVA is not a precision instrument. When over 150 experts classified the same foods in a 2022 study (Braesco et al.), they reached only fair agreement — a Fleiss kappa of about 0.32 to 0.34. Borderline products like plant milks are exactly where coders disagree. Treat the group as a useful signal, not a verdict carved in stone.
The exceptions: minimal oat drinks
A short-list oat drink — oats, water, a pinch of salt and an enzyme used to break down the oats — carries no ultra-processing markers. No oil, no gums, no emulsifiers, no flavourings. A drink like that sits in NOVA group 3, and some UK brands now market exactly this kind of recipe. We will not promise any specific product qualifies, because manufacturers reformulate without ceremony: the carton you bought last year is not evidence about the carton on the shelf today. Check the label, every time.
The trade-off is real, though. Minimal recipes usually skip fortification too, which is precisely what many people want from a milk alternative. More on that below.
Shelf, chilled and barista: the UK complication
UK supermarkets sell oat drink in two places — the long-life cartons on the ambient shelf and the fresh-looking cartons in the chiller — and the same brand can use different recipes in each. The chilled carton is not automatically the less processed one; the position in the shop tells you nothing about the ingredient list.
Barista editions deserve their own mention. They are engineered to foam, stretch and resist splitting in hot coffee, and they usually carry more additives than the standard version — typically extra oil, acidity regulators and stabilisers. If you are choosing an oat milk on processing grounds, the barista carton is usually the furthest from minimal. That does not make it dangerous. It makes it a more engineered formulation, which is exactly what NOVA is designed to flag.
A NOVA 4 label is not the whole story
Here is the part most UPF commentary skips. Plant milks are one of the clearest cases where ultra-processed and unhealthy come apart. Researchers have argued that fortified plant milks can be nutritionally useful despite their NOVA 4 label, because the added calcium, iodine and B12 do a real job — particularly for anyone using oat drink in place of cows' milk, which is a significant source of those nutrients in the UK diet. A minimal NOVA 3 oat drink with no fortification wins on processing and loses on nutrients. Neither carton is simply “better”; they are different trade-offs.
Keep the wider evidence in proportion too. The large studies linking higher UPF intake to worse health outcomes are overwhelmingly observational — they show association, not proof of cause — and they say nothing specific about oat drinks. NOVA tells you how a product was made. It was never meant to be the only number on the carton that matters, which is why Cull scores nutrition, additives and processing separately rather than collapsing them into one verdict.
What to check on the label
Thirty seconds with the back of the carton settles it:
- Oils— rapeseed or sunflower oil is a culinary ingredient under NOVA and does not decide the group by itself, but in a drink it is a tell of industrial formulation, and it rarely appears without the gums and emulsifiers that do.
- Gums and emulsifiers— gellan gum, guar gum, lecithin. Any of these typically means group 4.
- Phosphates, flavourings, non-sugar sweeteners — the same. One marker is enough; a longer list is not “worse”, it is the same group.
- Added vitamins and minerals— these do not make it ultra-processed, and if the drink is replacing cows' milk, calcium and iodine fortification is worth actively looking for.
- Oats, water, salt, enzyme — and nothing else — likely NOVA 3.
Or skip the memorising. Cull's UPF scannerreads the barcode and shows the NOVA group, the additives and the score in one place — useful when you are comparing four cartons in a cold aisle.
The label-first habit carries over to the rest of the trolley. The same logic decides whether your bread is ultra-processed, whether your breakfast cereal makes the cut and which yoghurts stay minimally processed. Categories do not have NOVA groups. Products do.
Sources
- FAO (2019) — Ultra-processed foods, diet quality, and health using the NOVA classification system
- Monteiro et al. (2016) — NOVA. The star shines bright, World Nutrition
- Nature Food — on the validity and utility of NOVA and the concept of ultra-processed foods for science and policy
- Advances in Nutrition (ScienceDirect) — plant-based milk alternatives and processing
- Patient.info — are dairy-free milks like oat or soya good for us?
- Braesco et al. (2022) — reliability of NOVA classification, European Journal of Clinical Nutrition
- Open Food Facts — NOVA groups for food processing