This contradiction troubled Monteiro. If people were buying less fat and sugar, why were they getting bigger? The answer was right there in the data. Brazilians hadn’t really cut down on fat, salt, and sugar—they were just consuming these nutrients in an entirely new form. People were swapping traditional foods—rice, beans, and vegetables—for prepackaged bread, sweets, sausages, and other snacks. The share of biscuits and soft drinks in Brazilians’ shopping baskets had tripled and quintupled, respectively, since the first household survey in 1974. The change was noticeable everywhere. When Monteiro first qualified as a doctor in 1972, he’d worried that Brazilians weren’t getting enough to eat. By the late 2000s, his country was suffering with the exact opposite problem. At a glance, Monteiro’s findings seem obvious. If people eat too much unhealthy food, they put on more weight. But the nutritionist wasn’t satisfied with that explanation. He thought that something fundamental had shifted in our food system, and scientists needed a new way to talk about it. For more than a century, nutrition science has focused on nutrients: Eat less saturated fat, avoid excess sugar, get enough vitamin C, and so on. But Monteiro wanted a new way of categorizing food that emphasized how products were made, not just what was in them. It wasn’t just ingredients that made a food unhealthy, Monteiro thought. It was the whole system: how the food was processed, how quickly we ate it, and the way it was sold and marketed. “We are proposing a new theory to understand the relationship between diet and health,” Monteiro says. Monteiro created a new food classification system—called NOVA—that breaks things down into four categories. Least worrisome are minimally processed foods, such as fruits, vegetables, and unprocessed meats. Then come processed culinary ingredients (oils, butter, and sugar), and after that processed foods (tinned vegetables, smoked meats, freshly baked bread, and simple cheeses)—substances to be used carefully as part of a healthy diet. And then there are ultra-processed foods. There are a bunch of reasons why a product might fall into the ultra-processed category. It might be made using “industrial processes” like extrusion, interesterification, carbonation, hydrogenation, molding, or prefrying. It could contain additives designed to make it hyper-palatable, or preservatives that help it stay stable at room temperature. Or it might contain high levels of fat, sugar, and salt in combinations that aren’t usually found in whole foods. What all the foods share, Monteiro says, is that they are designed to displace freshly prepared dishes and keep you coming back for more, and more, and more. “Every day from breakfast to dinner you are consuming something that was engineered to be overconsumed,” says Monteiro. The concept of ultra-processed food has caught on in a big way since it was first introduced in 2009: Brazil, France, Israel, Ecuador, and Peru have all made NOVA part of their dietary guidelines. Countless health and diet blogs extol the virtues of avoiding ultra-processed foods—shunning them is one thing that both followers of a carnivorous and a raw vegan diet can actually agree on. The label has been used to criticize plant-based meat companies, who in turn have embraced the label. Impossible calls its plant-based burger “unapologetically processed.” Others have pointed out that there’s no way we can feed billions of people without relying on processed food. Open up my kitchen cupboards and you’ll find instant ramen, potato chips, biscuits, canned soup, sweets, and cereal bars—a world of ultra-processed food, all of it ready to eat with either no preparation or just a minimum of effort. It’s not just me that is in thrall to convenient foods. Ultra-processed food makes up almost 57 percent of the average UK diet and more than 60 percent of the US diet. And all of this consumption seems to be doing something to our health. Overconsumption of ultra-processed food has been linked to all kinds of health issues: colorectal and breast cancer, obesity, depression, and all-cause mortality. Figuring out how our diets influence our health is extremely difficult, and any armchair statistician will tell you that correlation does not equal causation, but it does seem clear that consuming too much ultra-processed food isn’t good for us. One reason for this is that ultra-processed foods are often high in salt, sugar, and fat, which almost everyone agrees we should be cutting down on, says Stacey Lockyer, a senior nutrition scientist at the British Nutrition Foundation. But if these foods are unhealthy simply because of their nutrients, then maybe we don’t need the ultra-processed category at all. Could it be that Monteiro’s NOVA categorization is just traditional nutrition science repackaged? Kevin Hall started out as an ultra-processed skeptic. He’s a researcher at the US National Institutes of Health in Bethesda, Maryland, where he studies how diet influences body weight and metabolism. He first heard about the NOVA categorization at a conference in 2015 when a Brazilian researcher mentioned the system to him. Why are you still looking at nutrients when they’re not important anymore, the researcher asked him. “This struck me as a profoundly weird way to think about food,” says Hall. He had spent his entire career studying how nutrients affected the human body. That’s what food was, he thought, just different ways of packaging nutrients together.Still, Hall was intrigued enough by the NOVA categorization that he put together the first randomized control trial comparing ultra-processed and unprocessed diets. In 2019 Hall asked 20 volunteers to stay at a clinical research hospital in Bethesda where they would be fed a diet of only ultra-processed or whole foods for two weeks, then switch to the other diet for the subsequent two weeks. Those on the ultra-processed diet were fed a selection of dishes including tater tots, turkey sausage, Spam, and an ungodly amount of diet lemonade. The whole-food diet was mostly made up of fruit, vegetables, and unprocessed meat. For both diets, Hall and his researchers provided double the recommended portions of food so participants could eat as much as they liked. The critical part, however, was that the two diets were nutritionally matched, so each contained roughly the same amount of protein, fat, carbohydrates, fiber, and so on. Hall’s study drew a clear link between junk food and excess calorie consumption, but it can’t tell us why people on the ultra-processed diet ate more. After he published the results, Hall was flooded with suggestions from other scientists. Some thought it was because junk food is more calorie-dense. Since processed foods are often deep-fried and high in fat, they pack in more calories per gram than whole foods. Or maybe it was because junk food was eaten more quickly; in the study, people on the ultra-processed diet ate significantly faster than those eating whole foods. Other scientists thought that additives might be playing a role, or that junk food changed the gut microbiome in a way that influenced calorie intake. A big factor might be the effect that ultra-processed foods have on our brain. Alexandra DiFeliceantonio is an assistant professor at the Fralin Biomedical Research Institute at Virginia Tech Carilion who studies how junk food interacts with the brain’s reward systems. “We know a lot more about fat, sugars, and carbohydrates, and how those are signaled in the gut and to the brain. We know a lot less about the role of ultra-processing in altering any of those signals,” says DiFeliceantonio. Her hypothesis is that since ultra-processed foods are rich in easily available calories, they induce a potent reward response in our brains that keeps us coming back for more. DiFeliceantonio’s work draws parallels between junk food and the tobacco industry. In an editorial for the journal Addiction, DiFeliceantonio and her colleague Ashley Geardhardt argue that highly processed foods should be considered addictive substances if we measure them against the standards set for tobacco products. But until we really understand the science behind how ultra-processed food impacts our bodies, policy will always lag behind. “We saw big shifts in things like tobacco policy and policy for opioids when we had really solid, scientific, biological data,” says DiFeliceantonio. So what should health authorities do about it? Government guidelines in Brazil advise people to avoid ultra-processed foods altogether, while French guidelines recommend limiting consumption. But other countries’ guidelines don’t refer to ultra-processed foods at all. In 2021 an independent report commissioned by the UK government proposed a series of reforms aimed squarely at the ultra-processed food industry. The report recommended a tax on sugar and salt used in processed foods, and for large companies to report how much unhealthy food they were selling. The government’s response, published a year later, largely ignored these recommendations. In the UK’s official nutrition guidelines, the only reference to processed food is that people should eat no more than 70 grams of red or processed meat each day. While the role of processing in our diets has come under greater focus, public agencies have been slow to respond. Stanford nutritionist Christopher Gardner sits on the US Dietary Guidelines Committee and is a member of the American Heart Association. “For both of them, processed food is an issue they have to address next, because the public is so interested in this,” he says. “We don’t have a position yet. We need a position on this.” It might also mean that we narrow down our definition of ultra-processed food. Packaged and processed foods are such an important source of nutrition for so many people that we need to be careful before we demonize the entire category, says Hall. They’re convenient, tasty, and cheap. In Hall’s 2019 study, the weekly cost of the ultra-processed meals was $45 cheaper than the whole food diet. “If you design policies to try to eliminate those foods without at the same time providing cheap, inexpensive, easy, convenient alternatives, you’re going to have a lot of people who are going to experience negative consequences of that,” he says. Things get even trickier when you factor in the climate impact of our diets. Most plant-based meats are highly processed, but that doesn’t necessarily make them less healthy than their meat equivalents. Meat substitutes tend to be lower in calories and saturated fat and higher in fiber, but lower in protein. But on an environmental level, plant-based beef is much better than the real thing. “If you’re comparing a highly processed beef burger or pork sausage with its plant-based equivalent, then the plant burger or sausage is generally going to have lower environmental impacts,” says Tara Garnett, a food researcher at the University of Oxford. Monteiro admits that ultra-processed foods are sometimes better than their unprocessed alternatives, but he’s concerned that plant-based burgers might displace other, healthier plant-based foods. Even there the picture is complex. Christopher Gardner ran a trial where people swapped animal meat with plant-based meats for eight weeks. After the plant-based phase of the trial, people lost weight and had low cholesterol concentrations. When it comes to plant-based meats, Gardner says the ultra-processed label might be doing the category a disservice. Monteiro thinks that we can’t afford to wait until we know everything about ultra-processed foods before public health bodies take action. “We are dealing with something very complex. It will take many years to understand all of these mechanisms. But do we need to wait until we know all of this to start to do something to stop this?” he says. For now the science on ultra-processed foods is moving along slowly, but the debate is raging louder than ever. Updated 2-24-2022 11:30 GMT: Alexandra DiFeliceantonio’s academic affiliation was corrected.
