How to Spot Ultra-Processed Foods: What the NOVA Classification Tells You
It's not just sugar and fat that count. The degree of processing could also shape your metabolism and your gut barrier. Here you'll learn how to read it on the label.
Many people count calories, sugar and fat. The ingredient list often goes unread. Yet the type of processing may be exactly what helps decide how your gut and your metabolism respond to food.
You're standing in front of the refrigerated shelf at the supermarket. The fruit yogurt says “no added sugar” in big letters. Sounds sensible, so it goes into your cart. Turn it over once. The ingredient list has 14 items: flavorings, thickeners, sweeteners, modified starch. None of that is in your kitchen.
Maybe you know this: you eat mindfully and watch your sugar and fat. Even so, you're hungry again two hours later. In the evening, your belly feels bloated, even though nothing “bad” was on your plate. And quietly, a question comes up: you do watch your sugar, so what's going wrong? The short answer: probably nothing at all. You just haven't asked a different question yet.
The usual question is: what's in your food? That means calories, sugar, fat and protein. The second question you rarely hear: what was done to your food? Nutrition researchers have developed a classification system for this. It's called NOVA, and it sorts foods by their degree of processing.
In this article, you'll learn how NOVA works. You'll see what studies on the degree of processing show and where they reach their limits. You'll understand why some emulsifiers might affect the protective layer of your gut. And you'll get three levers that fit into any everyday life.
What to expect here
- The four NOVA groups with examples from the supermarket
- Why the degree of processing could matter more than sugar or fat alone
- What a controlled study with a 508 kcal difference shows
- The emulsifier-gut axis step by step
- IBS, Crohn's and colitis: what studies and practice show
- The 30-second label check
- Where NOVA reaches its limits
- Three levers for the next two weeks
Four groups instead of a nutrition table: how NOVA works
Nutrition tables are useful, but they only tell part of the story. Two products can contain the same amount of sugar. Even so, they may have been made in entirely different ways.
Imagine three things on a table. An apple. A jar of applesauce. And an apple-flavored gummy. All three have something to do with apples. But only the first is still an apple. The second is a cooked apple with sugar. The third often contains hardly any apple, but sugar, glucose syrup, gelling agents and flavoring instead.
This is exactly the gradation NOVA describes. The system divides foods into four groups. What matters is not the nutritional value, but the type and purpose of processing.
Unprocessed or minimally processed
Foods as they come from nature. At most washed, dried, frozen, ground or pasteurized.
- Rolled oats, eggs, milk and plain yogurt
- Fresh meat and poultry
- Fruit, vegetables and plain frozen vegetables
- Legumes and nuts without additives
Processed culinary ingredients
Obtained from group 1, for example by pressing, refining or grinding. They are used for cooking, not for eating on their own.
- Butter and vegetable oils
- Salt
- Sugar and honey
Processed foods
Group 1 plus ingredients from group 2. Usually two or three components you know from your kitchen.
- Bakery bread made from flour, water, salt and yeast
- Cheese
- Pickled vegetables
- Canned tomatoes or legumes
- Salted nuts
Ultra-processed foods
Industrial formulations made from substances extracted from foods or produced in the lab. These are often joined by additives for flavor, color and texture.
- Soft drinks
- Packaged bread with emulsifiers
- Flavored breakfast cereals
- Ready meals and instant soups
- Fruit yogurt with flavoring and thickener
- Protein bars
- Processed meats with additives
A team led by Monteiro summarized the criteria for NOVA 4 in a review. Typical are ingredients that no home kitchen uses: glucose-fructose syrup, protein isolates, hydrogenated fats, flavorings, colorings or emulsifiers. What this means for you: the ingredient list often shows you faster than any nutrition table whether a product belongs to NOVA 4.
Monteiro et al., Public Health Nutr 2019. DOI: 10.1017/S1368980018003762 [Review]Many of these formulations are hyperpalatable. That means sweetness, salt, fat and texture are tuned to be especially rewarding. Your natural satiety signal might hardly be able to keep up.
How big is this issue in Europe and in Germany? Two datasets offer an answer.
A team led by Lauria analyzed the diets of 7,073 children, adolescents and adults from eight countries in the I.Family study. Almost half of the energy came from NOVA 4, with no differences by education or income. What this means for you: ultra-processed food is not a fringe issue for a few households, but everyday life in many families.
Lauria et al., Nutr Metab Cardiovasc Dis 2021. DOI: 10.1016/j.numecd.2021.07.019 [Real-world, cross-sectional analysis]The younger the participants, the higher this share. In children and adolescents, around 70 percent of sugar came from ultra-processed products. The more NOVA 4 on the plate, the higher the energy intake and the lower the diet quality.
A team led by Fedde and Bosy-Westphal followed 182 participants of the Kiel KOPS study for 13.3 years into young adulthood. At around 27 years old, almost half of their energy came from ultra-processed products, even more in those with overweight. What this means for you: in Germany, too, ultra-processed food makes up a large share of young adults' energy.
Fedde et al., Pediatr Obes 2025. DOI: 10.1111/ijpo.13192 [Real-world, cohort study, long-term data]Old thought: “Processed means unhealthy.”
New view: Processing in itself is not a bad thing. Frozen peas, plain yogurt or canned lentils are valuable companions in everyday life. What matters is whether foods have been turned into an industrial formulation.
Your next step: Do a quick pantry check tonight. Take out five products you eat often. Read the ingredient list of each one and assign it to one of the four groups. You don't have to throw anything away. It's only about taking a look.
And now you know why a look at the ingredient list reveals so much about the degree of processing.
More than sugar and fat: why the degree of processing could matter
Maybe you've often thought you simply lack discipline. You tell yourself you'll snack less, and in the evening you still reach for the bag of chips. That is not a failure of your discipline. Ultra-processed food could be built in a way that makes it hard to stop in time.
The best-known study on this comes from a metabolic ward. There, every bite is weighed and recorded.
A team led by Hall provided 20 adults on an inpatient ward with an ultra-processed diet for two weeks and an unprocessed diet for two weeks. On NOVA 4, they ate around 508 kcal more per day and gained 0.9 kg, while on the unprocessed diet they lost 0.9 kg. What this means for you: the degree of processing could help determine how much you eat, even when the nutrients look the same on paper.
Hall et al., Cell Metab 2019. DOI: 10.1016/j.cmet.2019.05.008 · PMID: 31105044 [RCT, n=20]Both diets offered similar amounts of calories, sugar, fat, fiber and macronutrients. Participants ate as much as they wanted. The extra came from carbohydrates and fat, not from protein. You can read more about this in the article Unprocessed food and satiety.
A newer study brought the experiment closer to everyday life. Both diets followed official dietary guidelines.
A team led by Dicken supplied 55 adults with a minimally processed diet and an ultra-processed diet, one after the other. On the minimally processed diet, weight fell by 2.06 percent, on the ultra-processed diet by 1.05 percent. What this means for you: even a plan that is balanced on paper could turn out differently depending on the degree of processing.
Dicken et al., Nat Med 2025. DOI: 10.1038/s41591-025-03842-0 [RCT, crossover]Supported by large RCTs / meta-analyses: For long-term outcomes, there is now an overview of many meta-analyses. It shows consistent associations. It cannot prove causes, because the underlying data mostly come from observational studies.
A team led by Lane evaluated 45 meta-analyses with almost 10 million participants in total. Of the 45 analyses, 32 linked more NOVA 4 with a higher risk, for example of cardiovascular death, type 2 diabetes or anxiety. What this means for you: the pattern is strikingly consistent, but the quality of the individual evidence is mostly low.
Lane et al., BMJ 2024. DOI: 10.1136/bmj-2023-077310 · PMID: 38418082 [Systematic review, umbrella review]Two teams led by Srour and Fiolet each followed more than 100,000 adults in the French NutriNet-Santé cohort over several years. For every 10 percentage points more NOVA 4, the risk of cardiovascular disease and the risk of cancer were each 12 percent higher. What this means for you: these are associations from observational data, not proof, but they point in the same direction as the experiments.
Srour et al., BMJ 2019. DOI: 10.1136/bmj.l1451 [Real-world, cohort study] · Fiolet et al., BMJ 2018. DOI: 10.1136/bmj.k322 [Real-world, cohort study]In statistical terms, that means a hazard ratio of 1.12 per 10 percentage points. For the heart, the association remained when the researchers accounted for the nutritional quality of the diet. This suggests that it's not only sugar, fat and salt that count.
Four lenses on the cell
Mechanistically plausible, human studies still thin: Why could the degree of processing play a role of its own? Four perspectives make this tangible.
Metabolism. Imagine the food matrix as the framework of a house. In a whole apple or a lentil, sugar, starch and fat are built into cell walls. Your gut first has to break down this framework. That takes time, and so nutrients also reach deeper sections of the gut. That's where L cells sit, which release satiety hormones such as GLP-1. In ultra-processed products, the framework has often already been taken apart. The nutrients could then be absorbed so early that this satiety signal turns out weaker.
Immune system. Your gut wall is covered with a mucus layer. The inner layer is normally almost free of bacteria. If bacteria move too close, the immune system recognizes their components. It recognizes flagellin, the flagellar protein of many bacteria, via the receptor TLR5. It recognizes LPS from the outer envelope of certain bacteria via TLR4. The result could be a quiet, low-grade inflammation. You can read how to recognize silent inflammation in your blood in Silent inflammation and hs-CRP.
Nervous system. Gut and brain are connected via the vagus nerve. Fiber from unprocessed food feeds bacteria that produce short-chain fatty acids such as butyrate. Butyrate is the most important energy source for the cells of the colon lining. It might also help shape signals to the nervous system. Ultra-processed food, by contrast, often contains little fiber. You can read more about this in L-glutamine, butyrate and the gut lining.
Hormones. Rapidly available carbohydrates make blood sugar and insulin rise quickly. Over time, this could strain your metabolism. Fittingly, type 2 diabetes was among the better-supported associations in the big overview. You can read how to flatten blood sugar spikes in Avoiding blood sugar spikes.
Old thought: “Then I'll just have to count calories even more strictly.”
New view: Stricter counting fights against a signal that may have been dampened by processing. It could be easier to choose food that your body recognizes as filling on its own.
Your next step: For three days, note after each meal when you get hungry again. Write down whether the meal came mostly from NOVA 1 to 3 or from NOVA 4. Often a pattern emerges after just a few days.
And now you know why two plates with the same nutrients can still set off different things in your body.
The emulsifier-gut axis: what additives might do to your protective layer
Emulsifiers sound harmless. They appear on countless labels, and hardly anyone thinks about them. Yet they are among the most intriguing points in research on NOVA 4.
An emulsifier binds what doesn't mix on its own: fat and water. Your dish soap basically does the same thing with the grease in the pan. In ice cream, this creates creaminess, in bread a soft crumb, in sauces a stable bond.
Now imagine your gut wall as a castle. The mucus layer is the moat in front of it. Bacteria swim in the outer part, while the inner layer normally stays almost free of them. In the animal model, it looks as though certain emulsifiers could blur the edge of this moat. The moat doesn't disappear. It becomes narrower, and bacteria move closer to the wall.
- The emulsifier reaches the gut. According to EFSA, CMC is not absorbed in the gut. It therefore reaches the colon, where most bacteria live.
- The gut microbiota changes. In animal and lab models, the composition of the bacteria shifts. Some produce more flagellin, meaning more flagellar protein.
- The mucus layer becomes more permeable. Bacteria move closer to the gut wall. The gap that normally protects the wall shrinks.
- The immune system sounds a quiet alarm. Via TLR5 and TLR4, it recognizes flagellin and LPS. A low-grade inflammation could develop.
- Possible consequences. Mice showed weight gain and metabolic disturbances, and susceptible animals developed colitis. In humans, these consequences have not yet been established.
A team led by Chassaing gave mice the emulsifiers CMC and P80 at comparatively low concentrations. The bacteria moved closer to the gut wall, low-grade inflammation and metabolic disturbances followed, and susceptible animals developed colitis. What this means for you: this is biologically plausible, but not yet established in humans.
Chassaing et al., Nature 2015. DOI: 10.1038/nature14232 · PMID: 25731162 [In vivo, mouse]One detail makes this study especially revealing. Germ-free mice without gut microbiota remained unaffected despite the emulsifier. When the team transferred the gut microbiota of treated animals to germ-free mice, the changes appeared there too. So the bacteria seem to be the mediators.
Two studies led by Chassaing and Naimi exposed human gut microbiota in a reactor to various emulsifiers. CMC and P80 altered the microbiota in a lasting way, and some carrageenans and other thickeners clearly did too, while lecithin did not. What this means for you: not every emulsifier is the same, and the bacteria seem to be the mediators.
Chassaing et al., Gut 2017. DOI: 10.1136/gutjnl-2016-313099 [In vitro, ex vivo model] · Naimi et al., Microbiome 2021. DOI: 10.1186/s40168-020-00996-6 [In vitro, ex vivo model]In the 2017 study, the altered microbiota was then transferred to germ-free mice. Signs of inflammation appeared there too. This is biologically plausible, but not yet established in humans.
A team led by Chassaing fed 16 healthy adults a controlled diet for eleven days. Seven received 15 grams of CMC daily, nine received no emulsifier. With CMC, the diversity of the gut microbiota declined, and in two people, bacteria were found in the otherwise almost sterile inner mucus layer. What this means for you: this fits the animal model, but the study is very small and the dose high.
Chassaing et al., Gastroenterology 2022. DOI: 10.1053/j.gastro.2021.11.006 · PMID: 34774538 [RCT, double-blind, n=16]Putting this in context is important. The participants received 15 grams of CMC daily. Whether smaller everyday amounts set off the same thing is a question the study doesn't answer.
A team led by Sellem linked the intake of individual emulsifiers in 95,442 adults with later cardiovascular disease. More celluloses as well as more E 471 and E 472 went along with a slightly higher risk. What this means for you: the risks are small and not proven to be causal, but they fit the lab findings.
Sellem et al., BMJ 2023. DOI: 10.1136/bmj-2023-076058 [Real-world, cohort study]These data also come from the French NutriNet-Santé cohort. You'll find a broader overview of additives and the gut in Emulsifiers, sweeteners and additives in the gut.
Most of the evidence comes from mice and bioreactors. The only controlled human study had 16 participants and a high dose. Observational data show small risks, not causes. That is reason enough for caution, not for alarm.
Old thought: “If it's approved, it can't be a problem.”
New view: Approval and open research questions don't rule each other out. EFSA evaluates additives based on the knowledge available at the time. The question about the gut microbiota is more recent and has not yet been conclusively answered. Both can stand side by side.
Your next step: On your next shopping trip, check three products that often contain emulsifiers: packaged bread, ready-made sauces and ice cream. Compare two versions of each. Often you'll find one with a shorter ingredient list on the same shelf.
And now you know why an inconspicuous word on the label could affect the protective layer of your gut.
IBS, Crohn's and colitis: what studies and practice show
I know this pattern very well. The following story is a composite and has been anonymized. It stands for many similar cases.
The starting situationA family with two school-age children, both parents working. Everyday life is tightly scheduled, and they rarely cook. The mother has been living with a diagnosis of irritable bowel syndrome for years.
The sentence in the first conversation“Somehow all of us have tummy trouble,” she says. Bloating, changing bowel habits, stomachaches before school. First, we check medically to make sure nothing serious is behind it.
A look at the foodThe family keeps a food diary for a week. The result surprises everyone: most of it comes from NOVA 4. Sliced white bread, breakfast cereals, fruit yogurt, ready-made sauces and frozen pizza.
The turning pointNo bans, no diet. First, the family changes breakfast, with rolled oats, plain yogurt and fruit. A few weeks later, a home-cooked evening meal is added.
After about two monthsAll four of them have abdominal complaints noticeably less often. I can't claim causality, but I document the temporal connection.
Sometimes change doesn't begin with giving things up, but with a new base.
Whether it was the food, more calm at the table or something else can't be separated. Maybe you still recognize something from your own everyday life. A gut that constantly grumbles costs energy and quietly shapes how you plan your day.
What do studies say about observations like these? The data are intriguing, but not consistent. That is exactly why a closer look is worthwhile.
A team led by Schnabel recorded diet and gut symptoms at the same point in time in 33,343 adults from the French NutriNet-Santé cohort. In the top quarter for the share of ultra-processed foods, the odds of irritable bowel syndrome were around 25 percent higher than in the bottom quarter. What this means for you: it is a snapshot that cannot establish a cause.
Schnabel et al., Am J Gastroenterol 2018. DOI: 10.1038/s41395-018-0137-1 · PMID: 29904158 [Real-world, cross-sectional analysis]In this group, ultra-processed foods provided around a third of the energy. In 10.5 percent of participants, the criteria for irritable bowel syndrome were met. People who eat a lot of convenience food may also live differently in other ways, for example with more stress or less sleep. And conversely, a sensitive gut could change food choices. Confounding factors like these can never be fully accounted for.
A team led by Narula followed 116,087 adults from 21 countries in the PURE study for a median of 9.7 years and analyzed their consumption of ultra-processed foods. Five or more servings a day were associated with a 1.82-fold risk of Crohn's disease or colitis compared with less than one serving. What this means for you: the association is clear, but an observational study does not prove a cause.
Narula et al., BMJ 2021. DOI: 10.1136/bmj.n1554 [Real-world, cohort study]During the study, 467 people were newly diagnosed with Crohn's disease or ulcerative colitis. Soft drinks, sweetened convenience products, salty snacks and processed meat each showed higher risks. White meat, unprocessed red meat, dairy products, starches, fruit, vegetables and legumes, by contrast, showed no association. This fits the idea that the degree of processing could play a role of its own.
A team led by Meyer followed 413,590 adults from eight European countries in the EPIC cohort for an average of 13.2 years. There, ultra-processed foods showed no clear association with Crohn's disease or colitis, but eating a lot of unprocessed food went along with less Crohn's disease. What this means for you: the studies partly contradict each other, which is why an honest assessment has to mention it.
Meyer et al., Clin Gastroenterol Hepatol 2023. DOI: 10.1016/j.cgh.2022.09.031 [Real-world, cohort study]The comparison was between the top and bottom quarters for the share of unprocessed foods. The hazard ratio for Crohn's disease was 0.57, with a confidence interval of 0.35 to 0.93. That corresponds to a risk just over 40 percent lower. Different populations, questionnaires and case numbers could explain why PURE and EPIC differ. One shared message remains: a base of minimally processed foods doesn't come off badly in either study.
A team led by Bhattacharyya had twelve people with ulcerative colitis in a quiet phase of the disease eat a carrageenan-free diet. In the group that also received carrageenan capsules, there were three relapses, while the placebo group had none. What this means for you: this is a signal, but twelve participants can't support a firm conclusion.
Bhattacharyya et al., Nutr Healthy Aging 2017. DOI: 10.3233/NHA-170023 [RCT, small pilot study]In my clinical practice, I observe that many people with irritable bowel syndrome first think of individual foods, such as lactose, gluten or onions. The degree of processing as a whole comes into view less often. Yet a simple, minimally processed foundation can be a good first step, even before long lists of forbidden foods.
If you want to dig deeper: I put possible causes of irritable bowel syndrome into context in IBS: finding the causes. You'll find the integrative view of inflammatory bowel disease in Crohn's disease and colitis from an integrative view. You can read what's behind a permeable gut wall in Leaky gut and zonulin.
Blood in your stool, unintended weight loss, symptoms that wake you at night, fever or persistent diarrhea should be checked by a doctor. The same applies to new symptoms from around the age of 50. A change in diet does not replace a medical workup.
A calm gut is not a luxury. It is freedom. Freedom to leave the house without a toilet plan. Freedom to simply be part of a meal with friends.
Old thought: “My gut is just sensitive, there's nothing I can do about it.”
New view: A sensitive gut can react to many things, including the degree of processing of your food. This is not a question of blame, but a possible starting point. You can try out what changes when your base becomes simpler.
Your next step: Keep a simple log for one week. Note every meal and roughly assign it to a NOVA group. Next to it, write down how your gut feels two to three hours later. After seven days, you'll often spot patterns you hadn't noticed before.
And now you know why a restless gut might have more to do with the degree of processing of your food than any nutrition table reveals.
The 30-second label check
At the supermarket, you don't have time for biochemistry. You're standing in front of the shelf, and maybe someone with a shopping cart is pushing behind you. You need a method that works in half a minute.
The good news: the ingredient list tells you a lot. In the EU, ingredients must be listed in descending order by weight. So whatever comes first makes up the largest part of the product.
Your 30-second check
- Count the ingredients. The longer the list, the more likely a product belongs to NOVA 4. There is no set cutoff, but very long lists are a clue.
- Look for things that don't belong in a kitchen. Is there anything on the label you won't find in any kitchen? Maltodextrin, glucose syrup, modified starch, flavorings or protein isolates are typical examples.
- Check the E numbers. Emulsifiers, thickeners and sweeteners are typical markers of NOVA 4. Pay special attention to the numbers in the table below.
- Read the first three ingredients. If sugar, refined oils or white flour come right at the front, the product consists largely of them.
- Ask the home cooking question. Could you make the product yourself with ingredients from your kitchen? If not, it probably belongs to NOVA 4.
A word on seals and marketing claims: labels like “organic,” “vegan” or “protein” say little about the degree of processing. Even a vegan protein bar made from organic ingredients can belong to NOVA 4. Seals describe origin or composition, not processing.
| E number | Name | Typically found in | What studies show |
|---|---|---|---|
| E 466 | Carboxymethylcellulose (CMC) | Ice cream, sauces, desserts, baked goods | Mouse: bacteria closer to the gut wall, low-grade inflammation. Human, small study with 16 people over 11 days: lower gut microbiota diversity in the group given 15 g of CMC daily. |
| E 433 | Polysorbate 80 (P80) | Ice cream, desserts, sauces, fine baked goods | Mouse and lab model: changes similar to those with CMC. Not yet established in humans. |
| E 407 | Carrageenan | Pudding, cream products, plant-based drinks, meat products | Lab model: especially clear changes with various carrageenans. Pilot study in ulcerative colitis with twelve people: three relapses on carrageenan, none on placebo. |
| E 471, E 472 | Mono- and diglycerides of fatty acids and their esters | Bread, baked goods, margarine, ice cream | Large cohort: higher intake associated with slightly higher cardiovascular risk. |
| E 460 to E 468 | Cellulose and cellulose derivatives | Grated cheese, sauces, baked goods | Large cohort: higher total intake associated with slightly higher cardiovascular risk. |
| E 322 | Lecithins | Chocolate, margarine, baked goods | Lab model: no clear changes in the gut microbiota. |
The table is not a list of forbidden foods. It shows where research is taking a closer look. Lecithin hardly stood out in the lab model, while CMC clearly did.
Old thought: “Now I have to memorize all the E numbers.”
New view: You don't need an encyclopedia in your head. The length of the list and the home cooking question are enough for most decisions. The table is just your cheat sheet for the cases where you want to take a closer look.
Your next step: Do the 30-second check today on three products from your refrigerator. Take photos of the ingredient lists if you like, and compare them with an alternative on your next shopping trip.
And now you know why half a minute in front of the shelf can often tell you more about your food than the advertising on the front.
Where NOVA reaches its limits
Maybe you're asking yourself at this point: isn't this all too simple? The question is justified. NOVA is also critically discussed in research.
A classification system is only as good as its criteria. A French team therefore tested how consistently experts work with NOVA.
A team led by Braesco had French food and nutrition experts assign products to the NOVA groups. Agreement was low: kappa 0.32 for 120 branded products with an ingredient list, 0.34 for 111 generic foods without an ingredient list. What this means for you: even experts don't always classify the same product the same way.
Braesco et al., Eur J Clin Nutr 2022. DOI: 10.1038/s41430-022-01099-1 · PMID: 35314769 [Real-world, methods study]Kappa measures how much several people agree. A value of 1 means full agreement, a value of 0 means no more than chance. The researchers themselves rate their result as low agreement, even when the ingredient list was available.
But there were also clear cases. For one group of 90 branded products, 91 percent of the judgments went to NOVA 4. For 28 generic foods, by contrast, the judgments were spread fairly evenly across all four groups. For your everyday life, this means: for many ready-made products, the classification is fairly clear. In the gray zone in between, NOVA is not an exact measurement.
In 2018, the EFSA expert panel on food additives re-evaluated cellulose and its derivatives, including CMC. At the reported uses and levels, it saw no safety concerns. What this means for you: by regulatory standards, there is no cause for concern, yet open research questions remain.
EFSA ANS Panel, EFSA J 2018. DOI: 10.2903/j.efsa.2018.5047 [Regulatory document]A second point concerns the quality of the evidence. Researchers often rate it with a system called GRADE. Imagine it as a school grade for how reliable studies are. In the large review led by Lane, none of the 45 analyses reached the highest level. Four were rated moderate, 22 low and 19 very low.
This is mainly because most of the data come from observation. People who eat a lot of ultra-processed food often also differ in exercise, sleep, smoking or income. Such differences can be adjusted for statistically, but never fully.
The Kiel KOPS data also show how complex the picture is. Among the children and adolescents, who were around 13 years old, there was even an inverse association. There, an eating pattern with a lot of ultra-processed food went along with a lower BMI and less fat mass. Only among the young adults did the expected association with overweight appear. In women, consumption in childhood and adolescence was linked to consumption in adulthood, but not in men.
Another detail comes from the study led by Dicken. Both diets there followed the UK dietary guidelines, and participants lost weight on both. Mild gastrointestinal complaints were common on both diets. So the degree of processing is one factor among several, not the only one.
NOVA assesses processing, not nutrients. A pure fruit juice without additives can end up in NOVA 1, even though it supplies a lot of sugar. A packaged whole-grain sandwich bread with an emulsifier, on the other hand, ends up in NOVA 4, even though it contains fiber. So combine NOVA with a look at the nutrition facts and with common sense.
Old thought: “If NOVA is so controversial, the whole concept is worthless.”
New view: NOVA is a tool, not a law. It doesn't replace looking at nutrients, amounts and your everyday life. But it asks a question that is often missing. How much was your food changed before it landed on your plate?
Your next step: Use NOVA as a first filter, not as a verdict. If a product lands in NOVA 4, also look at sugar, fiber and how often you eat it. A product for special occasions counts for less than your daily breakfast.
And now you know why NOVA is a good map, but not the terrain itself.
Three levers for the next two weeks
Maybe you're thinking now: that sounds like a lot of effort. The good news: you don't have to change your whole life. Three small levers are enough to start with.
Lever 1: The kitchen question at the shelf
For every packaged product, ask yourself a single question: could you make this yourself from ingredients in your kitchen? If yes, it probably belongs to NOVA 1 to 3. If not, a second look at the ingredient list is worthwhile. The question takes three seconds and sharpens your eye with every shopping trip.
Lever 2: One meal a day from NOVA 1 to 3
Choose one meal a day that you'll put together from groups 1 to 3 from now on. Breakfast is often the easiest. Rolled oats with plain yogurt, fruit and nuts take five minutes to prepare. Once that's a habit, add the next meal after one to two weeks.
Lever 3: Swap instead of giving up
Bans rarely last long. So swap individual products for less processed versions from the same category.
| Instead of | Try | The difference |
|---|---|---|
| Packaged sandwich bread with emulsifiers | Bakery bread made from flour, water, salt and yeast | Short ingredient list, NOVA 3 |
| Fruit yogurt with flavoring | Plain yogurt with fresh fruit | No flavoring, no thickener |
| Flavored breakfast cereals | Rolled oats with nuts and fruit | NOVA 1, often more fiber |
| Ready-made sauce from a jar | Canned tomatoes with olive oil and herbs | Ingredients you know |
| Soft drink | Water with lemon or unsweetened tea | No added sugar |
In my clinical practice, I observe that small swaps often last longer than big bans. This is experience from practice, not a study. Try out what fits into your everyday life.
Your self-check after two weeks
- How many of your meals came mostly from NOVA 1 to 3?
- Has your feeling of hunger between meals changed?
- How did your gut feel, especially in the evening?
Eating well doesn't begin with bans. It begins with the question of how much kitchen and how much factory is in your food.
Not only sugar and fat matter: the degree of processing could also help shape how your body responds to food.
Old thought: “Eating healthily means giving things up.”
New view: Eating healthily can also mean adding something new. Every meal made from real ingredients is a step, not a sacrifice.
Your next step: Today, choose the one meal you'll start with. Write it down and stick with it for two weeks. Then do the self-check.
And now you know why small, repeated steps can often change more than a radical fresh start.
If you don't just want to read but want to get started right away: below this article, you'll find the option to book an appointment.
Frequently asked questions about ultra-processed foods
What is the NOVA classification?
NOVA sorts foods into four groups. What counts is the type and purpose of processing, not the nutrients. Group 1 covers unprocessed or minimally processed foods, group 2 culinary ingredients such as oil, butter or salt. Group 3 covers processed foods such as bakery bread or simple cheese. Group 4 covers ultra-processed products made from industrial formulations.
How can I spot ultra-processed foods in the supermarket?
A good clue is a long ingredient list with substances you won't find in any kitchen. These include, for example, maltodextrin, glucose syrup, flavorings, emulsifiers, thickeners or sweeteners. Also ask yourself the home cooking question: could you make the product yourself with ingredients from your kitchen? If not, it probably belongs to NOVA 4.
Is every processed food unhealthy?
No. Processing in itself is not a bad thing. Frozen peas, plain yogurt or canned lentils are valuable companions in everyday life. What matters is whether foods have been turned into an industrial formulation with many additives.
Isn't it simply about too much sugar and fat?
Not only. In the study led by Hall, both diets offered similar amounts of calories, sugar, fat and fiber. Even so, 20 adults ate around 508 kcal more per day on the ultra-processed diet and gained 0.9 kg. On the unprocessed diet, they lost 0.9 kg. In the NutriNet-Santé cohort, the association with cardiovascular disease remained even after accounting for nutritional quality.
Which emulsifiers are best studied?
The most frequently studied are carboxymethylcellulose (E 466) and polysorbate 80 (E 433). In mice, both altered the gut microbiota and led to low-grade inflammation. In a small human study with 16 people, seven received 15 grams of CMC daily for eleven days. In this group, the diversity of the gut microbiota declined. In the lab model, various carrageenans and other thickeners stood out clearly, lecithin hardly at all.
Are approved emulsifiers a cause for concern?
In 2018, EFSA re-evaluated cellulose and its derivatives, including CMC. At the reported uses and levels, it saw no safety concerns. At the same time, research questions about the gut microbiota remain open. Animal and lab data are biologically plausible but not yet established in humans. That is reason enough for caution, not for alarm.
Can ultra-processed foods promote irritable bowel syndrome?
That has not been proven. A team led by Schnabel recorded diet and gut symptoms at the same point in time in 33,343 adults. In the quarter with the highest NOVA 4 share, the odds of irritable bowel syndrome were around 25 percent higher than in the lowest. As a snapshot, the study cannot establish a cause. Conversely, a sensitive gut could also change food choices.
Is there a link with Crohn's disease and ulcerative colitis?
The data are mixed. In the PURE study, 116,087 adults from 21 countries were followed for a median of 9.7 years. There, five or more servings a day were associated with a 1.82-fold risk compared with less than one serving. In the European EPIC cohort with 413,590 adults, by contrast, no clear association emerged. There, however, eating a lot of unprocessed food went along with less Crohn's disease. Both are observational data, not proof of a cause.
How much ultra-processed food is eaten in Europe and Germany?
The European I.Family study examined 7,073 children, adolescents and adults from eight countries. Almost half of their energy came from NOVA 4, even more among younger participants. In the Kiel KOPS study, young adults around 27 years old also got almost half of their energy from NOVA 4.
Is the NOVA classification scientifically undisputed?
No. In a study led by Braesco, French experts often classified the same products differently. Agreement was kappa 0.32 for branded products and 0.34 for generic foods. In addition, in the review led by Lane, none of the 45 analyses reached the highest GRADE level. Four were rated moderate, 22 low and 19 very low.
Do I have to give up ultra-processed foods entirely?
No. Shifting your base may make more sense. Start with one meal a day from NOVA 1 to 3 and swap individual products for simpler versions. In the study led by Dicken, participants lost weight on both balanced diets. On the minimally processed diet, weight fell by 2.06 percent, on the ultra-processed diet by 1.05 percent.
How quickly might I notice changes?
That is very individual, and there are no promises. In the study led by Hall, differences in energy intake appeared within each two-week diet period. In the small CMC study, the gut microbiota changed within 11 days. The study led by Dicken ran for eight weeks per diet. A self-check after two weeks can show you whether your sense of hunger or your gut has changed.
Connections to other topics
The degree of processing touches on many topics around the gut, metabolism and inflammation. These articles explore individual aspects in more depth.
NOVA classification
Spotting ultra-processed foods
this articleAdditives
Emulsifiers and sweeteners in the gut
IBS
Finding the causes
Leaky gut
Gut barrier and zonulin
Crohn's and colitis
The integrative view
Satiety
Unprocessed food
Inflammation
Fries or banana
Fiber
30 grams a day
Scientific sources
- Monteiro CA, Cannon G, Levy RB, et al. Ultra-processed foods: what they are and how to identify them. Public Health Nutr. 2019;22(5):936-941. DOI: 10.1017/S1368980018003762 · PMID: 30744710 [Review]
- Hall KD, et al. Ultra-Processed Diets Cause Excess Calorie Intake and Weight Gain: An Inpatient Randomized Controlled Trial of Ad Libitum Food Intake. Cell Metab. 2019;30(1):67-77.e3. DOI: 10.1016/j.cmet.2019.05.008 · PMID: 31105044 [RCT, n=20, inpatient]
- Lane MM, Gamage E, Du S, et al. Ultra-processed food exposure and adverse health outcomes: umbrella review of epidemiological meta-analyses. BMJ. 2024;384:e077310. DOI: 10.1136/bmj-2023-077310 · PMID: 38418082 [Systematic review, umbrella review]
- Chassaing B, Koren O, Goodrich JK, et al. Dietary emulsifiers impact the mouse gut microbiota promoting colitis and metabolic syndrome. Nature. 2015;519(7541):92-96. DOI: 10.1038/nature14232 · PMID: 25731162 [In vivo, mouse]
- Chassaing B, Compher C, Bonhomme B, et al. Randomized Controlled-Feeding Study of Dietary Emulsifier Carboxymethylcellulose Reveals Detrimental Impacts on the Gut Microbiota and Metabolome. Gastroenterology. 2022;162(3):743-756. DOI: 10.1053/j.gastro.2021.11.006 · PMID: 34774538 [RCT, double-blind, n=16]
- 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. DOI: 10.1136/bmj.l1451 · PMID: 31142457 [Real-world, cohort study]
- Fiolet T, Srour B, Sellem L, et al. Consumption of ultra-processed foods and cancer risk: results from NutriNet-Santé prospective cohort. BMJ. 2018;360:k322. DOI: 10.1136/bmj.k322 · PMID: 29444771 [Real-world, cohort study]
- Narula N, Wong ECL, Dehghan M, et al. Association of ultra-processed food intake with risk of inflammatory bowel disease: prospective cohort study. BMJ. 2021;374:n1554. DOI: 10.1136/bmj.n1554 · PMID: 34261638 [Real-world, cohort study]
- Schnabel L, Buscail C, Sabate JM, et al. Association Between Ultra-Processed Food Consumption and Functional Gastrointestinal Disorders: Results From the French NutriNet-Santé Cohort. Am J Gastroenterol. 2018;113(8):1217-1228. DOI: 10.1038/s41395-018-0137-1 · PMID: 29904158 [Real-world, cross-sectional analysis]
- Naimi S, Viennois E, Gewirtz AT, Chassaing B. Direct impact of commonly used dietary emulsifiers on human gut microbiota. Microbiome. 2021;9(1):66. DOI: 10.1186/s40168-020-00996-6 · PMID: 33752754 [In vitro, ex vivo model]
- Bhattacharyya S, et al. A randomized trial of the effects of the no-carrageenan diet on ulcerative colitis disease activity. Nutr Healthy Aging. 2017;4(2):181-192. DOI: 10.3233/NHA-170023 · PMID: 28447072 [RCT, small pilot study]
- Chassaing B, Van de Wiele T, De Bodt J, et al. Dietary emulsifiers directly alter human microbiota composition and gene expression ex vivo potentiating intestinal inflammation. Gut. 2017;66(8):1414-1427. DOI: 10.1136/gutjnl-2016-313099 · PMID: 28325746 [In vitro, ex vivo model]
- Dicken SJ, Jassil FC, Brown A, et al. Ultraprocessed or minimally processed diets following healthy dietary guidelines on weight and cardiometabolic health: a randomized, crossover trial. Nat Med. 2025;31(10):3297-3308. DOI: 10.1038/s41591-025-03842-0 · PMID: 40760353 [RCT, crossover]
- Sellem L, Srour B, Javaux G, et al. Food additive emulsifiers and risk of cardiovascular disease in the NutriNet-Santé cohort: prospective cohort study. BMJ. 2023;382:e076058. DOI: 10.1136/bmj-2023-076058 · PMID: 37673430 [Real-world, cohort study]
- Braesco V, Souchon I, Sauvant P, et al. Ultra-processed foods: how functional is the NOVA system? Eur J Clin Nutr. 2022;76(9):1245-1253. DOI: 10.1038/s41430-022-01099-1 · PMID: 35314769 [Real-world, methods study]
- EFSA Panel on Food Additives and Nutrient Sources added to Food (ANS), Younes M, Aggett P, et al. Re-evaluation of celluloses E 460(i), E 460(ii), E 461, E 462, E 463, E 464, E 465, E 466, E 468 and E 469 as food additives. EFSA J. 2018;16(1):e05047. DOI: 10.2903/j.efsa.2018.5047 · PMID: 32625652 [Regulatory document]
- Fedde S, Stolte A, Plachta-Danielzik S, Müller MJ, Bosy-Westphal A. Ultra-processed food consumption and overweight in children, adolescents and young adults: Long-term data from the Kiel Obesity Prevention Study (KOPS). Pediatr Obes. 2025;20(4):e13192. DOI: 10.1111/ijpo.13192 · PMID: 39622513 [Real-world, cohort study, long-term data]
- Meyer A, Dong C, Casagrande C, et al. Food Processing and Risk of Crohn's Disease and Ulcerative Colitis: A European Prospective Cohort Study. Clin Gastroenterol Hepatol. 2023;21(6):1607-1616.e6. DOI: 10.1016/j.cgh.2022.09.031 · PMID: 36243353 [Real-world, cohort study]
- Lauria F, Dello Russo M, Formisano A, et al. Ultra-processed foods consumption and diet quality of European children, adolescents and adults: Results from the I.Family study. Nutr Metab Cardiovasc Dis. 2021;31(11):3031-3043. DOI: 10.1016/j.numecd.2021.07.019 · PMID: 34518085 [Real-world, cross-sectional analysis]