“It ain’t so much the things we don’t know that get us in trouble,” said the nineteenth-century aphorist Josh Billings. “It’s the things we know that just ain’t so.” These days, for example, many health-conscious adults know that ultraprocessed food (UPF) is a nutritional disaster. Observational research has linked UPF consumption to a wide range of chronic conditions, including obesity, diabetes, cardiovascular disease, cancer, and behavioral disorders. A 2025 childhood health report from the U.S. Department of Health and Human Services’ Make America Healthy Again (MAHA) initiative draws dozens of connections between such food and harmful outcomes.
But as David Ludwig, a professor in the department of nutrition at the Harvard Chan School of Public Health, wrote recently in “Ultraprocessed Food on an Ultrafast Track,” a commentary in the New England Journal of Medicine (NEJM), the appraisal of UPF has often been painted with overly broad brushstrokes. In a second essay, published on Medium, Ludwig compares the alarm over ultraprocessed foods to the mass mystification in “The Emperor’s New Clothes.” “We seem to be rolling out a red carpet, even though the evidence is threadbare!” he writes, adding that the UPF framework “is not based on evidence, it’s based on an ideology.”
The notion of UPF took off in 2009, when, in a commentary in Public Health Nutrition, Brazilian epidemiologist Carlos Monteiro argued that, health-wise, the most salient division of food and drink is by their type, degree, and purpose of processing, rather than their nutritional profiles. The UPF concept has since swept across the nutritional landscape, in part via “brilliant marketing,” says Ludwig. “Ultraprocessed is such an evocative word—ultra signifies an extreme degree.” It does capture, he acknowledges, a legitimate recognition that industrial processing has sullied the relationship between humans and their food supply.
Among the enduring effects of Monteiro and his colleagues’ work is the Nova food classification system, which identifies four categories of foods by their level of processing. (“Nova,” Portuguese for “new,” comes from the title of their article announcing it, which begins with the words, “Uma nova classificação de alimentos…”) Group 1 includes “unprocessed or minimally processed” food such as fresh or frozen fruits, vegetables, grains, or fresh meat. Group 2, “processed culinary ingredients,” embraces seasonings or cooking ingredients derived from natural sources such as salt, sugar, flour, and oils. The “processed foods” of Group 3 can be foods from the first two groups in combination, as with salted nuts or fruits in syrup. Cheeses, breads, and canned vegetables may also qualify, as well as items with added preservatives or flavorings.
Group 4, the category of “ultraprocessed foods,” includes industrially manufactured products made of multiple ingredients (typically including sugars, oils, fats, and salt) and ingredients with little or no culinary use, such as high-fructose corn syrup and hydrogenated oils. That classification incorporates many fast food items and mass-marketed items on supermarket shelves. However, Ludwig says, a food need have only one main ingredient or additive absent from traditional home cooking to earn ultraprocessed status—potentially causing misclassification and confusion. He notes that the category now includes many everyday items such as store-bought whole-grain breads, tofu, cheese, beef patties, and even sparkling water—since carbonation alone qualifies as a chemical additive.
“The nightmarish vision of factories churning out toxic industrial concoctions contrasts with the reality of many foods labeled ‘ultraprocessed,’” Ludwig wrote in an email. “It’s almost a romantic notion that what was done 100 years ago was fine, and what is done today is harmful by default,” he further explains in an interview. “But lots of things people ate a century ago—like sugar, refined flour, and too much salt—are unhealthy.” With Nova, “the food industry could dump all the sugar and salt it wanted into a product without it being called ultraprocessed—but they can’t get away with adding stevia, a safe, sugar-free sweetener.”
In addition, Ludwig says, Nova’s UPF definition ignores the macronutrient content of foods, and “by disregarding nutrients, we throw out the baby with the bathwater.” He adds that processing affects varied nutrients differently. While mechanical processes can alter carbohydrates in unhealthy ways—causing blood sugar to spike, for instance, when people eat processed white bread or drink apple juice—they have little or no effect on the healthfulness of proteins and fats. Squeezing olive oil from olives just produces a healthy oil, he says. Making avocados into guacamole doesn’t diminish their healthful effects. Neither does grinding sesame seeds into tahini or peanuts into peanut butter.
Evidence linking ultraprocessed food to poor health is elusive.
Evidence linking ultraprocessed food to poor health is elusive, Ludwig says, because, as often happens in nutritional research, many confounding variables obscure the links between causes and effects. People who eat large amounts of ultraprocessed food tend to have less education and higher body weight, be more sedentary, smoke more, and drink more. All these factors are established risk factors for disease, independent of UPFs.
Such characteristics also reflect differences in social class, raising another problem. “Telling people to eat beef ribeye at triple the price of hamburger has an elitist streak,” Ludwig explains. “Yet a hamburger has similar nutritional value. It’s kind of like Marie Antoinette saying, ‘Let them eat steak.’”
Since putting forward his provocative argument, Ludwig says he has received mostly supportive feedback from experts in the field, with no serious challenges to the scientific accuracy of his case. Still, one esteemed senior colleague took him to task for “missing the forest for the trees,” worried that the NEJM paper might undermine regulatory efforts to improve the food supply. Ludwig says he understands that criticism yet remains concerned that public health action directed at the wrong target could backfire—as it did during the low-fat diet craze a few decades ago, when an emphasis on consuming carbohydrates over fat made the national obesity crisis worse, not better.
Responding to the present public health crusade, he points out that ultraprocessed foods are less expensive and more convenient than whole foods. “We have eight billion people worldwide who need food,” Ludwig says. “Telling low-income people to cook everything at home from scratch is impractical. To some extent, processed and packaged foods enable us to avoid malnutrition and starvation.”
