‹ Diet

Longevity · Diet deep dive

Ultra-Processed Foods: Is "Processing" Really the Problem?

In one sentence: Diets heavy in ultra-processed food are clearly linked with worse health, and one striking experiment showed people eat about 500 extra calories a day on them, but scientists are still fighting over whether processing itself is the problem or just the junk food that dominates the category.

Twenty people, four weeks, one locked kitchen

In 2018, twenty adults checked into the clinical research center at the US National Institutes of Health in Bethesda, Maryland. For the next four weeks, they wouldn't eat anything the researchers didn't give them.

For two weeks, their meals came mostly from boxes, cans and packages: think breakfast cereals, white bread, packaged snacks, canned foods and processed meats. For the other two weeks, their meals were made from scratch with whole ingredients: oatmeal with fruit and nuts, grilled fish, roasted vegetables, salads, brown rice.

Here's the clever part. The scientists, led by metabolism researcher Kevin Hall, designed both diets to offer the same calories, sugar, fat, fiber, salt and protein at every meal. Participants could eat as much or as little as they liked.

When the results came in, they made headlines worldwide. On the ultra-processed diet, people ate about 500 more calories every day. They gained about 2 pounds in two weeks. On the whole-food diet, they lost about the same amount 2.

It was the first rigorous experiment suggesting that something about ultra-processed food makes us eat more. But what that something is has become one of the biggest debates in nutrition.

Where the term came from

The phrase "ultra-processed food" was coined by a Brazilian nutrition scientist, Carlos Monteiro, and his colleagues at the University of São Paulo. They noticed that rising obesity in Brazil seemed to track not with any single nutrient, but with the spread of industrially made, ready-to-eat products.

So they created a new way to sort foods, called NOVA, based on how much and why foods are processed 1:

Group What it is Examples
1. Unprocessed or minimally processed Whole foods, maybe dried, frozen or cut Fresh produce, eggs, plain milk, dried beans, rice, oats, plain meat
2. Culinary ingredients Extracted from foods for cooking Oil, butter, sugar, salt, flour
3. Processed foods Group 1 plus Group 2 ingredients Canned vegetables, cheese, fresh bakery bread, canned fish
4. Ultra-processed Industrial formulas with ingredients you wouldn't find in a home kitchen Soda, packaged snacks, candy, most breakfast cereals, mass-produced bread, instant noodles, protein bars, flavored yogurt, many plant-based meats

The idea took off. Today, estimates suggest roughly half or more of daily calories in the US and UK come from Group 4. Some countries now advise people to avoid ultra-processed foods in their national guidelines.

What the big studies show

Dozens of large observational studies link higher ultra-processed food intake to obesity, type 2 diabetes, heart disease, depression and early death.

A 2024 "umbrella review" in the BMJ gathered 45 analyses covering nearly 10 million people 3. It rated the evidence as convincing for links with cardiovascular death and type 2 diabetes, among others. For many other outcomes, the evidence was weaker.

That sounds like an open-and-shut case. But many nutrition scientists aren't convinced, and their reasons are worth hearing.

Why the case isn't closed

Who eats the most ultra-processed food?

People who eat lots of ultra-processed food often have lower incomes, less time to cook, more stress, and different smoking, drinking and exercise habits. Researchers adjust for these factors, but statistics can't fully separate the food from the life around it.

The category is enormous

A study finding that "ultra-processed food is linked to diabetes" is lumping together soda, white bread, hot dogs, flavored yogurt and whole-grain breakfast cereal. When studies split the category up, sugary drinks and processed meats often drive the bad associations. Some other ultra-processed foods look neutral, or even good.

The rules can be strange

Critics point to oddities in the NOVA system 4,5:

  • A bag of plain white sugar is Group 2, a "culinary ingredient." A tub of plain whey protein powder is Group 4, "ultra-processed."
  • A supermarket whole-grain loaf with an emulsifier is ultra-processed. A bakery's white bread might not be.
  • Fortified soy milk, which many dietitians recommend, is ultra-processed.

In short, critics say the system sorts foods by how they're made, not by what they do to your body.

We already have a good tool

Many researchers argue we can already spot problem foods: they're high in fat, sugar or salt and low in fiber and protein. In this view, "ultra-processed" mostly overlaps with "junk food," and the new label adds confusion, guilt and anxiety, especially for families who rely on affordable, convenient staples 6.

The defenders respond

Supporters of the concept have answers 1,7:

  • Processing may matter beyond nutrients. Changing a food's texture (making it soft and quick to eat), adding certain emulsifiers, and engineering flavor combinations might affect how fast we eat, how full we feel, and possibly our gut bacteria.
  • Remember the NIH experiment. The diets were matched for sugar, fat, salt and fiber, and people still overate. Something beyond the nutrition label was going on.
  • It's simple advice that works. "Eat mostly foods your grandmother would recognize" is easy to follow, and people who follow it usually eat better.

Back to the locked kitchen

So what did make people eat 500 extra calories in Kevin Hall's study? Even Hall has said his experiment couldn't answer that. Looking closely, there were some differences between the diets that the matching didn't cover:

  • People ate the ultra-processed meals faster. Faster eating gives fullness signals less time to catch up.
  • The solid ultra-processed foods were more energy-dense, packing more calories into each bite.

Hall's team and others are now testing these ideas one by one:

  • Energy density: Does lowering calories per bite reduce overeating?
  • Eating speed: Do soft, easy-to-chew textures drive overconsumption?
  • "Hyper-palatable" combinations: Do certain mixes of fat with sugar, or fat with salt, override our fullness signals?
  • Additives and the gut: Some emulsifiers changed gut bacteria and promoted inflammation in mice. Human studies are just starting.
  • A healthier ultra-processed diet: If people are fed ultra-processed foods chosen to be lower in energy density and less hyper-palatable, do they still overeat?

That last experiment may be the most important. If the answer is no, the problem isn't processing. It's the kind of food that processing usually produces.

What we don't know: Which feature of ultra-processed food drives overeating, whether specific additives harm people at normal intakes, and whether healthier ultra-processed foods carry any risk at all.

Two approaches, two countries

The ultra-processed debate isn't just academic. Countries have taken different paths.

In 2014, Brazil became the first country to build its national dietary guidelines around processing. Its advice is simple: make natural or minimally processed foods the basis of your diet, and avoid ultra-processed foods. The guidelines also encourage cooking, eating at regular times and eating with others. Several other countries have since borrowed the idea.

Chile took the other road. In 2016, it began requiring black warning labels on packaged foods high in sugar, salt, saturated fat or calories, and limited how those foods could be marketed to children. That approach targets nutrients, not processing.

Both strategies are still being studied. Early research on Chile's labels found people bought fewer sugary drinks. Whether Brazil's processing-based advice changes health over the long run is harder to measure.

What the two diets might look like

To get a sense of the difference, here's an illustrative day of each (not the exact NIH menus):

  • Mostly ultra-processed: sweetened cereal with milk, a packaged muffin, a deli-meat sandwich on mass-produced white bread with chips, a flavored yogurt, and a frozen pizza for dinner.
  • Mostly minimally processed: oatmeal with fruit and nuts, an apple, leftover chicken with rice and vegetables, plain yogurt with berries, and a homemade stir-fry or bean chili.

Notice that the second day needs more planning, more time and often more dishes. That's part of why ultra-processed foods won the modern kitchen.

Where everyone agrees

Underneath the fight, there's a lot of common ground:

  • Diets built around soda, sweets, salty snacks, fast food and processed meat are linked with poor health.
  • Most people would benefit from eating more vegetables, fruit, legumes, whole grains, nuts, fish and plain dairy.
  • "Ultra-processed" and "unhealthy" overlap a lot, but they aren't the same thing.

The scorecard

Question What the evidence suggests Confidence
Do people overeat on typical ultra-processed diets? Yes, in a well-controlled short trial Moderate
Are high ultra-processed diets linked with disease? Yes, consistently Moderate
Is processing itself the cause? Unknown Low
Are all ultra-processed foods harmful? Unlikely Moderate

Common questions

How can I tell if a food is ultra-processed? Check the ingredient list for things you wouldn't use at home: emulsifiers, flavor enhancers, colors, modified starches, hydrolyzed proteins, high-fructose corn syrup. But remember that tells you about processing, not necessarily about health.

Which ultra-processed foods are reasonable choices? Depending on the product: whole-grain cereals low in sugar, plain yogurt and kefir, fortified soy milk, whole-grain sliced bread, canned fish, plain protein powders. Fiber, protein, added sugar and salt are a better guide than the label.

Are additives safe? They're reviewed by regulators, and for most there's no good human evidence of harm at normal intakes. Some, like certain emulsifiers, are being restudied.

Isn't this really about cost? Partly. Ultra-processed foods are often cheaper and faster. Frozen vegetables, canned beans, oats, eggs, rice and peanut butter are affordable alternatives that can replace some of them.

Are plant-based meats a good swap? They're ultra-processed by definition, and their nutrition varies a lot by brand. One small Stanford trial found that swapping red meat for plant-based meat products for eight weeks lowered LDL cholesterol. They may be a reasonable bridge, but beans and lentils are the better-proven option.

Is all bread ultra-processed? Not all. Most mass-produced packaged bread is. Bread made with just flour, water, yeast and salt counts as processed but not ultra-processed.

Are frozen fruit and vegetables ultra-processed? Plain frozen produce counts as minimally processed, the same group as fresh. Frozen meals and breaded or sauced products are a different story.

Are protein bars bad? They're ultra-processed, and many are closer to candy bars with added protein. Some are reasonable as an occasional convenience. Look at the sugar and fiber, and know that bars with lots of sugar alcohols can upset some stomachs.

Do I have to cook everything from scratch? No. Even small shifts count: a home-cooked dinner a few nights a week, oats instead of a sugary cereal, nuts or fruit instead of packaged snacks. Researchers studying this don't expect perfection, and neither should you. Swapping the worst offenders matters far more than eliminating every product with an unfamiliar ingredient.

The takeaway

Twenty people in a locked kitchen proved that a typical ultra-processed diet makes us eat more. What they didn't prove is why, and that "why" will shape food advice for the next decade.

  • Start with the clear offenders: sugary drinks, sweets, salty packaged snacks, fast food and processed meats.
  • Don't fear useful processed foods like canned beans, frozen vegetables, plain yogurt and whole-grain bread.
  • Read what's in the food, not just how it's made.
  • Slow down. Eating more slowly and choosing foods that need chewing may help with fullness.
  • Cook when you can, and skip the guilt when you can't.

The science of ultra-processed food is still being written. For now, the best advice is the oldest: mostly real food, and not too much of the rest.

References

  1. Monteiro CA, et al. A new classification of foods based on the extent and purpose of their processing. Cad Saude Publica. 2010;26(11):2039-49. PubMed 21180977

  2. 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. PubMed 31105044

  3. Lane MM, et al. Ultra-processed food exposure and adverse health outcomes: umbrella review of epidemiological meta-analyses. BMJ. 2024;384:e077310.

  4. Gibney MJ, et al. Ultra-processed foods in human health: a critical appraisal. Am J Clin Nutr. 2017;106(3):717-724. PubMed 28793996

  5. Poti JM, Braga B, Qin B. Ultra-processed food intake and obesity: what really matters for health, processing or nutrient content? Curr Obes Rep. 2017;6(4):420-431. PubMed 29071481

  6. Weaver CM, et al. Processed foods: contributions to nutrition. Am J Clin Nutr. 2014;99(6):1525-42. PubMed 24760975

  7. Monteiro CA, et al. Ultra-processed foods: what they are and how to identify them. Public Health Nutr. 2019;22(5):936-941. PubMed 30744710

This article is for general education and isn't a substitute for personalized medical advice.