Just how bad are ultra-processed foods?

In the mid-1990s, Carlos Monteiro, a nutritional epidemiologist in Brazil, noticed something alarming: the obesity rates among the children in his country rose rapidly.

To understand why, he and his colleagues at the University of São Paulo examined the data on this the purchasing pattern of food of Brazilian households to see if they have changed in recent years. The researchers found that people bought less sugar, salt, cooking oil and staples such as rice and beans, and more processed foods such as soft drinks, sausages, instant noodles, packaged bread and biscuits.

To describe that second category of food, Dr. Monteiro, the team introduced and defined a new term in the scientific literature – ultra-processed foods, or UPFs. They would later link UPFs for weight gain in children and adults in Brazil.

Since then, scientists have found associations between UPFs and diabetes a range of health problemsincluding heart disease, type 2 diabetes, obesity, gastrointestinal disease and depression, as well as previous deaths.

That's worrying experts say, as ultra-processed foods have become an important part of people's diets worldwide. They make an allowance 67 percent of calories consumed by children and teenagers in the United States, for example.

But many questions remain. What exactly are ultra-processed foods? And how strong is the evidence that they are harmful? We asked experts these and other questions.

To study foods based on the way they were processed, Dr. Monteiro and his colleagues created a food classification system called Nova, named after the Portuguese and Latin words for “new.” It has since been adopted by researchers around the world.

The Nova system sorts food four categories:

  • Unprocessed or minimally processed foodssuch as fresh or frozen fruits and vegetables, beans, lentils, meat, poultry, fish, eggs, milk, yogurt, rice, pasta, cornmeal, flour, coffee, tea and herbs and spices.

  • Processed culinary ingredientssuch as cooking oil, butter, sugar, honey, vinegar and salt.

  • Processed foods made by combining foods from Category 1 with the ingredients from Category 2 and preserving or modifying them using relatively simple methods such as canning, bottling, fermenting and baking. This group includes freshly baked bread, most cheeses and canned vegetables, beans and fish. These foods may contain preservatives that extend their shelf life.

  • Ultra-processed foods made using industrial methods and ingredients you wouldn't normally find in grocery stores, like high fructose corn syrup, hydrogenated oils, and concentrated proteins like soy isolate. They often contain additives such as flavorings, colorings or emulsifiers to make them look more attractive and tasty. This includes soft drinks and energy drinks, chips, candies, flavored yogurt, margarine, chicken nuggets, hot dogs, sausages, lunch meats, boxed macaroni and cheese, infant formula and most packaged breads, plant-based milks, meat substitutes and breakfast cereals.

“If you look at the ingredient list and see things you wouldn't use in home cooking, those are probably ultra-processed foods,” says Brenda Davy, a nutrition professor at Virginia Tech.

Notably, the Nova system does not classify foods based on nutrients such as fat, fiber, vitamins or minerals. It's “independent of diet,” says Maya Vadiveloo, an associate professor of nutrition at the University of Rhode Island.

That led to it debate among nutritionists on whether it is useful to describe a food's healthfulness, in part because many UPFs — such as whole-wheat bread, flavored yogurt and infant formula — can provide valuable nutrients, said Dr. Vadiveloo.

Most research linking UPFs to poor health is based on observational studies, where researchers ask people about their diets and then track their health over many years. In a large overview of studies published in 2024, scientists reported that consuming UPFs was linked to 32 health problems, with the most compelling evidence for deaths from heart disease, type 2 diabetes and common diseases. mental health problems such as anxiety and depression.

Such studies are valuable because they can look at large groups of people – the 2024 review included results from almost 10 million people – over the many years it can take for chronic health problems to develop, says Josiemer Mattei, associate professor of nutrition at the University of California. Harvard TH Chan School of Public Health. She added that the consistency of the link between UPFs and health problems increased her confidence that there was a real problem with the foods.

But the observational studies also have limitations, says Lauren O'Connor, a food scientist and epidemiologist who previously worked at the Department of Agriculture and the National Institutes of Health. It's true that there is a link between these foods and chronic diseases, she said, but that doesn't mean UPFs directly cause poor health.

Dr. O'Connor wondered whether it would be useful to group such “very different” foods – like Twinkies and breakfast cereals – into one category. Certain types of ultra-processed foods, such as soft drinks and processed meat, are clearly more harmful than others. UPFs love flavored yogurt and whole wheat breadin contrast, have been associated with a reduced risk of developing type 2 diabetes.

Clinical trials are needed to test whether UPFs directly cause health problems, said Dr. O'Connor. Only one such study has been done, which was small and had some limitations, she said.

In that study, published in 2019, twenty adults of varying body sizes lived for four weeks at a National Institutes of Health research hospital. For two weeks they ate mostly unprocessed or minimally processed foods, and for another two weeks they ate mostly UPFs. The diets contained similar amounts of calories and nutrients, and participants could eat as much as they wanted at each meal.

During the two weeks they followed the ultra-processed diet, participants gained an average of two pounds and consumed about 500 more calories per day than they did on the unprocessed diet. During the period they were on the unprocessed diet, they lost about two pounds.

That finding could help explain the link between UPFs, obesity and other metabolic disorders, said Kevin Hall, a nutrition and metabolism researcher at the National Institutes of Health, who led the trial. But the study needs to be repeated, where Dr. Hall is working on now.

There are many “strong opinions” about why ultra-processed foods are unhealthy, said Dr. Hall. “But there's actually not a lot of rigorous science” about what those mechanisms are, he added.

Since UPFs are often cheap, convenient and accessible, they probably are crowd out healthier foods from our diet, said Dr. Hall.

But he and other scientists think the food could have more direct effects on health. It's easy to eat too many of them, perhaps because they contain combinations that are hard to resist carbohydrates, sugars, fats and salt, are high in calories and easy to chew. It's also possible that the resulting blood sugar spikes could damage arteries or blood vessels increase inflammationor that certain thing food supplements or chemicals can disrupt hormones, a “leak” intestine or disrupt the intestinal microbiome.

Researchers, including Dr. Hall and Dr. Davy, are beginning to conduct small clinical trials that will test some of these theories. Such studies can help identify the most harmful UPFs and even suggest how to make them healthier, said Dr. Hall.

But most researchers think there are several ways the food causes harm. “Rarely is there a single dietary factor that fully explains the relationship between food and a particular health outcome,” said Dr. Vadiveloo.

In 2014, Dr. helped Monteiro with writing new dietary guidelines for Brazil who advised people to avoid ultra-processed foods.

Other countries like Mexico, Israel And Canada have also explicitly recommended avoiding or limiting UPFs, or “highly processed foods.” The American dietary guidelines do not contain such advice, but an advisory committee currently does examine the evidence on the impact of UPFs weight gain, which could impact 2025 guidelines.

It's difficult to know what to do with UPFs in the United States, where so much food is already ultra-processed and lower-income people in particular may depend on them, said Dr. Hall.

“Ultimately, they are an important food source, and food is food,” added Dr. Mattei added. “We really can't vilify them,” she said.

While research continues, expert opinions differ on how people should approach UPFs. Dr. Monteiro said it's safest to avoid them altogether — for example, swapping flavored yogurt for plain yogurt with fruit, or buying a loaf of fresh bread from a local bakery instead of packaged bread, if you can afford it .

Dr. Vadiveloo proposed a more moderate strategy, focused on limiting UPFs that don't provide valuable nutrients, such as soda and cookies. She also recommended eating more fruits, vegetables, whole grains (ultra-processed or otherwise), legumes, nuts and seeds.

Cook at home as much as possible, with minimally processed foods, said Dr. David. “We can't say much about that at the moment.”