Why exactly are ultraprocessed foods so hard to resist? This study is trying to find out

It was 9am on a Friday in March, and Ernest Jones III was hungry. From a hospital bed at a research facility at the National Institutes of Health in Maryland, he surveyed his meal tray: Honey Nut Cheerios with fibre-enriched whole milk, a plastic-wrapped blueberry muffin and margarine.
“Simple, old school,” one of those “Saturday morning breakfasts from back in the day,” said Jones, 38, who is studying to become a pastor.
He was about halfway through his 28-day stay at the NIH, and Jones was one of 36 people participating in a nutrition trial that is expected to be completed in late 2025. For one month each, researchers will draw participants’ blood, track their body fat and weight, measure the calories they burn, and feed them three meticulously designed meals per day.
The subjects don’t know it, but their job is to help answer some of the most pressing questions in nutrition: Are ultraprocessed foods harmful to health? Are they a major driver of weight gain and obesity? And why is it so easy to eat so many of them?
If researchers can answer these questions, they say, perhaps there are ways to make ultraprocessed foods healthier.
THE TROUBLE WITH PROCESSING
Ultraprocessed foods encompass a large range of foods and drinks that are made using methods and ingredients you wouldn’t typically use or find in a home kitchen. Sodas, processed meats and flavoured yoghurts are part of this category, as are most breakfast cereals, packaged breads and plant milks.
Ultraprocessed foods are a major source of calories in the United States – making up about 58 per cent of those consumed by children and adults, according to one recent estimate.
Kevin Hall, a senior investigator at the NIH who is leading the trial, said that there is a “mountain of epidemiological data” linking ultraprocessed foods to poor health – including 32 health concerns like heart disease, Type 2 diabetes, obesity, depression and certain gastrointestinal conditions and types of cancer.
But many questions remain, including whether it is the ultraprocessed foods themselves that cause those conditions, or whether it is something about the lives of the people who eat them. Scientists also still don’t know why ultraprocessed foods might cause poor health.
There’s a “gaping hole in our knowledge,” Dr Hall said – one he hopes to begin to fill with this research.
THE ‘MOST IMPORTANT STUDY DONE IN NUTRITION IN YEARS’
In 2019, Dr Hall and his colleagues published the results of a trial that has been hailed as one of the most influential studies in nutrition.
In it, 20 adults lived at the NIH and spent two weeks following a diet made of ultraprocessed foods and two weeks following one prepared from unprocessed foods. Both diets had similar nutrient levels, and the participants were instructed to eat as much or as little as they liked.
The results were surprising: During the ultraprocessed weeks, the participants consumed about 500 more calories per day than they did during the unprocessed weeks, and gained an average of two pounds (0.9kg). By the end of the unprocessed weeks, they had lost about two pounds.
It was “the most important study done in nutrition in years,” said Marion Nestle, an emeritus professor of nutrition, food studies and public health at New York University.
Researchers hadn’t directly tested how ultraprocessed foods might influence calorie consumption and weight gain before, she said, and the answer was relevant to a large portion of Americans. It also rigorously controlled the participants’ diets for a month – something most nutrition studies don’t do.
But the trial was small and hadn’t been replicated, Dr Hall said. It also didn’t explain why people tend to overeat ultraprocessed foods.
So Dr Hall is using this new study to replicate those findings, and to test two theories about why the foods may lead to weight gain.
One idea is that they often contain certain combinations of enticing nutrients – like fats, sugars, sodium and carbohydrates – that might trigger the brain’s reward system in a way that makes people want to eat a lot of them.
After munching on salty potato chips, “your brain is like, ‘Oh my god, we need another bite of that,’” even if “your stomach is like, ‘Oh, please don’t do that, we’re so full,’” said Tera Fazzino, an assistant professor of psychology at the University of Kansas. Dr Fazzino defined the term used to describe this phenomenon, called hyperpalatability.
A second hypothesis, Dr Hall said, is that ultraprocessed foods often contain many calories per bite. And because they can be less filling than unprocessed foods, you might consume more of them without thinking in order to feel satisfied.
Dr Hall thinks that if food companies could make ultraprocessed foods less calorie-packed and irresistible, we might be less likely to consume extra calories and gain weight.
TRACKED, SCANNED AND CHAPERONED
Every day at 6.30am, a nurse knocked on the door of Mr Jones’s room to wake him, check his blood pressure and take his weight.
At 9am, it was time for his first meal. Every tray was carefully prepared in a basement kitchen at the NIH, each ingredient weighed to the nearest tenth of a gram. Participants were instructed to eat as much or as little as they liked. Some meals contained as many as 2,000 calories – the amount some people have in an entire day.
Once he was finished, Jones’ tray was whisked back to the basement, where scientists weighed any leftovers to calculate exactly how much he had eaten.
Jones didn’t know about the food scales in the basement, or that the number of calories he consumed was the key part of the study. He also wasn’t allowed to see his weight, over fears it might influence how much he ate.
Every week, the characteristics of the meals would change, depending on what the researchers were testing. One week was all unprocessed foods, like unsweetened yoghurts, nuts, baked cod, stir-fried beef, rice and lots of vegetables.
For the other three weeks, at least 80 per cent of the meals’ calories were made up of ultraprocessed foods – breakfast cereals, sausages, deli meats, sweetened yoghurts, baked goods – with slight tweaks between weeks to test how their calorie concentrations and palatability might affect how much the participants ate.
For the entire study, Jones wore a continuous glucose monitor on his upper arm to track blood sugar fluctuations. He also wore activity trackers on one of his wrists, an ankle and his waist to account for any variations in his activity. One day per week, his blood was drawn before breakfast and then six times more over the next three hours to measure his insulin, glucose, lipid and hunger and satiety hormone levels, as well as his inflammation markers.
Several times per day, an iPad would ping with survey questions about his mood, appetite and satisfaction with the meals.
Once a week, Jones would also get a full body scan to measure his body fat and a test to see how many calories he burned while resting in bed. And he’d be sealed into a dormlike room called a metabolic chamber for a 24-hour period every week, where researchers would measure how many calories he burned while eating, watching TV, riding a stationary bike, sleeping and doing other activities.
To see how the diets affected his gut microbiome, Mr. Jones also had to give a stool sample once a week – his least favourite part of the study.
“That was the one thing that I was like, ‘I don’t know if I want to do this,’” Jones said.
In his downtime, he watched a lot of documentaries and sports – including all of the March Madness games – and he read, wrote in his journal, and attended divinity courses and church services online.
Living at the NIH took some getting used to. Jones was the only trial participant at the NIH during his stay; the researchers don’t have space or staffing to accommodate more than one or two subjects at a time. And he was not allowed to snack or have caffeine, which can affect metabolism, Dr Hall said, and people’s preferences for coffee creamer and sweeteners would complicate things. Alcohol was also forbidden.
Jones missed sipping hot tea and eating homemade popcorn when he did his homework in the evenings, and he would have liked an occasional Jolly Rancher or honey bun. He sometimes craved a soda, which, despite being a major ultraprocessed product in the United States, was not part of the study.
He also missed his daily walks, which often stretched for more than 10 miles in his mother’s Richmond, Virginia, neighborhood, where he’d lived for the year before coming to the NIH.
During the trial, Jones agreed to ride a stationary bicycle for an hour each day so he could get a standard amount of exercise. He was allowed to step outside or take short walks around campus, but they had to be chaperoned – to prevent “stopping by the vending machine or the cafeteria”, Dr Hall said.
Jones asked to go outside only a handful of times during his month at the facility, but he was thrilled when a nurse handed him a pair of eclipse glasses on Apr 8 and told him it was time to go watch the celestial event.
WHAT A TRIAL LIKE THIS MIGHT TELL US
If the study can give researchers some clarity about why ultraprocessed foods may cause unintentional weight gain, the results could help to guide nutrition policies, said Josiemer Mattei, an associate professor of nutrition at the Harvard TH Chan School of Public Health. For instance, she said, policymakers could develop labels for certain foods to warn about their potential health risks.
The ultraprocessed category includes so many foods and drinks that it’s impractical – and maybe not necessary – for most people to avoid them all, Dr Hall said.
But if the trial suggests that some of these foods cause weight gain because they are packed with calories or engineered to be extremely tasty, those findings can help distinguish which of the foods may be okay to eat, and which are most important to avoid, Dr Hall said. Food manufacturers could potentially use that information to make processed foods that are less likely to cause weight gain, he said, such as by reducing their sodium or sugar, or by adding fiber, which adds bulk without adding calories.
Carlos Monteiro, a nutritional epidemiologist at the University of São Paulo in Brazil, who defined the term ultraprocessed foods with his colleagues in 2009, is sceptical that the companies would willingly implement these changes, though. Making a product less irresistible, for example, could cut into their profits, he said.
Still, Dr Hall thinks it’s worth a try to push for less harmful ultraprocessed foods, in part because it’s unlikely that Americans will have the desire or time to go back to making everything from scratch, he said.
Dr Hall’s family eats plenty of unprocessed foods, but chicken nuggets and frozen pizzas still make occasional appearances in his home. They’re convenient and his young children enjoy them. “I don’t want to not have chicken nuggets as an option,” he said.
He acknowledged that it’ll take more research to understand how these foods affect health before they can be re-engineered so that they don’t cause weight gain.
“It sounds pie in the sky,” he acknowledged. But, he added, “I think there’s a real shot at this.”
By Alice Callahan © The New York Times Company
The article originally appeared in The New York Times.