Who Gets to Decide If a Food Is ‘Ultra-Processed?’

Whole grain breakfast cereal, Greek yogurt, and 100{7b6cc35713332e03d34197859d8d439e4802eb556451407ffda280a51e3c41ac} whole wheat bread have more in common than being part of a “balanced breakfast”—they can all be classified as “ultra-processed” foods.

Ultra-processed foods (UPFs) are everywhere. Some estimates have found that up to 73{7b6cc35713332e03d34197859d8d439e4802eb556451407ffda280a51e3c41ac} of the American food supply is ultra-processed foods.

However, experts don’t agree on what to do about ultra-processed foods—or even how to classify them.

How Processed Foods Were Originally Classified

Ultra-processed foods were first defined by Carlos A. Monteiro, MD, PhD, and his team in 2009. The researchers wanted to prioritize ultra-processed foods in public health nutrition and developed the NOVA food classification system as a way to categorize levels of food processing.

NOVA (which is a name and not an acronym) classifies all foods into four groups:

• Group 1: Unprocessed or Minimally Processed Foods  
• Group 2: Processed Culinary Ingredients
• Group 3: Processed Foods
• Group 4: Ultra-Processed

Monteiro officially defined ultra-processed foods as “industrial formulations made mostly or entirely with substances extracted from foods, often chemically modified, and from additives, with little if any whole food added. Sequences of processes are and must be used to obtain, alter, and combine the ingredients and to formulate the final products (hence ‘ultra-processed’).”

In other words, ultra-processed foods are designed to be cheap, tasty replacements for the foods included in the three other NOVA categories. Soda, chips, and ice cream are common examples of ultra-processed foods.

During the recent American Society for Nutrition (ASN) conference, Monteiro said that “the end product of food ultra-processing are products that perhaps we shouldn’t call foods because they are very far from the original foods.”

How Can You Tell If a Food Is Ultra-Processed?

Classifying canned soups, pastries, pre-prepared burgers, and sports drinks as ultra-processed might seem obvious, but it’s not always easy to tell ultra-processed foods from processed foods.

In a 2019 paper, Monteiro explained that one way to identify ultra-processed foods is to look at the nutrition facts label. If the product is made with at least one NOVA Group 4 ingredient or an additive, it can be considered ultra-processed.

However, when French researchers put NOVA to the test, they found inconsistencies. For the study, a cohort of food and nutrition professionals were asked to classify an assortment of foods using NOVA.

The researchers observed that the health experts did not consistently classify the same foods into the same NOVA groups. Based on the study’s findings, the researchers concluded that the NOVA system needed to be reevaluated.

Véronique Braesco, PhD, a researcher with VAB-nutrition and the study’s first author, told Verywell that even though NOVA is widely used, this research was actually one of the first attempts to test the robustness of the system.

“We were puzzled that the classification itself is a tool on which many papers are based [and] many decisions are based,” said Braesco. “[And] which to our knowledge, and after having read papers, didn’t seem to be validated.”

Are Ultra-Processed Foods Unhealthy?

Going by the NOVA definitions, almost everything we eat is processed to some extent. Refrigerating, boiling, and chopping a whole food item are forms of processing.

However, Monteiro argues processing isn’t the problem—ultra-processing is. In fact, plenty of experts agree that ultra-processing should be a focus for public health nutrition.

In 2019, a landmark randomized controlled trial led by Kevin D. Hall, PhD, found an association between ultra-processed foods and weight gain compared to an unprocessed diet.

Are Ultra-Processed Foods Part of a Larger Problem?

Some experts argue that we need to better understand the mechanisms behind ultra-processed foods before making broad recommendations.

Arne Astrup, MD, PhD, a prominent nutrition researcher and a professor at the University of Copenhagen, told Verywell that “it’s not surprising that in all these observational studies, you will see that a high intake of ultra-processed foods is linked to obesity and diabetes.”

However, Astrup pointed out that ultra-processed foods could be a marker of an overall “unhealthy” diet. He also suggested that the findings from Hall’s randomized control trial could be explained by the high amounts of sugar and low amounts of fiber in an ultra-processed foods diet.

According to Astrup, it’s also worth noting that Hall’s study only lasted two weeks—therefore, longer trials would need to be done to verify the results.

Monteiro stands by his argument and told Verywell that NOVA does not classify foods as healthy or unhealthy.

While he recognizes that it would be unrealistic for people to avoid all UPFs, Monteiro said that “the lower the consumption the better” as “what the evidence shows is that a dietary pattern dominated by UPFs is unhealthy.”

What Do Experts Say About Ultra-Processed Foods?

Even though researchers don’t agree on using NOVA to define ultra-processed foods, policies and consumer tools are still being developed using the classification system.

For example, Brazil, Peru, Belgium, Ecuador, Israel, Maldives, and Uruguay are just some of the countries that include recommendations to limit ultra-processed foods in their population’s dietary guidelines. France recently set a goal of reducing ultra-processed food consumption by 20{7b6cc35713332e03d34197859d8d439e4802eb556451407ffda280a51e3c41ac}.

It has even been proposed that the U.S. Department of Agriculture (USDA) address ultra-processed foods in the 2025-2030 Dietary Guidelines for Americans.

During the ASN conference held virtually last month, Monteiro and Astrup debated the use of ultra-processed foods for informing dietary guidelines.

As the lead researcher behind NOVA, Monteiro argued that even though more research is needed to understand the exact mechanisms, there is enough evidence to support recommendations for ultra-processed food consumption in dietary guidelines.

‘TrueFoods’ vs. Ultra-Processed Foods

Americans will have to wait to see if the next Dietary Guidelines will reference ultra-processed foods. In the meantime, research groups like the team behind the TrueFoods Database are trying to make it easier for consumers to figure out if their foods are ultra-processed.

Babak Ravandi, PhD, a postdoctoral researcher at Northeastern University who helped develop the tool, told Verywell that TrueFoods was built on the back of NOVA.

While Ravandi calls NOVA’s creation a “landmark” in public health, his concern is that the system only offers four classification categories, which means that “it classifies a huge chunk of foods as equally ultra-processed.”

Ravandi argues that this makes it hard for people and policymakers to find alternative choices. That’s where TrueFoods tries to help.

With TrueFoods, foods are given a score between 0-100 that compares their level of processing with other similar products on the market. The AI that powers it relies on data from the Food and Drug Administration (FDA) and USDA.

However, Ravandi said that the data still needs to be perfected to eliminate the margins of error that are seen in the database. For example, TrueFood scores some seemingly ultra-processed foods too low—like the score of 11 given to Cheetos Flamin’ Hot Popcorn.

Only time will tell if the scientific community will be able to reach an agreement about the best way to categorize the levels of processed foods.

Ravandi said that most researchers can at least agree on the two extremes—that naturally occurring whole foods are at one end of the spectrum and convenience foods are at the other. It’s everything in between that’s still being debated.

However, Ravandi is hopeful that “we are moving one step closer toward coming to consensus.”