The term "ultra-processed foods" (UPFs) has become a dominant concept in American food discourse over the past decade, influencing media coverage, consumer guides, and even policy debates led by figures like Robert F. Kennedy Jr. However, a growing body of evidence suggests that the scientific foundation for demonizing UPFs is weaker than commonly assumed.
Definitional Ambiguity and Observational Weaknesses
The trouble begins with the definition. UPFs generally refer to packaged foods containing ingredients not typically used in home kitchens, such as high-fructose corn syrup and maltodextrin. Yet nutrition scientists struggle to clearly distinguish normal processing from ultra-processing. A study once classified tofu as ultra-processed, highlighting the inconsistency.
Much of the evidence linking UPFs to poor health outcomes like heart disease and cancer comes from observational studies that cannot separate correlation from causation. These studies often rely on diet surveys that lack the detail to determine whether reported foods like white bread or yogurt were actually ultra-processed.
Randomized Controlled Trials Under Scrutiny
To address these shortcomings, researchers conducted randomized controlled trials (RCTs), the gold standard for establishing cause and effect. Several landmark RCTs found that UPFs led to overeating and weight gain. However, a recent analysis in Science examined five such trials and discovered that the ultra-processed and non-ultra-processed diets differed in important ways beyond processing. The UPF diets were more calorie-dense and lower in fiber, which can independently cause overeating. Thus, the effects attributed to ultra-processing may have other causes.
Faidon Magkos, a professor at the University of Copenhagen and co-author of the Science analysis, stated, "While UPFs are often characterized by properties such as soft texture, higher energy density, and lower fiber content, these features are neither unique to UPFs nor consistently present across them."
The NOVA Classification System and Its Criticisms
UPF research relies on the NOVA classification system, which sorts foods into four groups: unprocessed or minimally processed (group 1), processed culinary ingredients (group 2), processed foods (group 3), and ultra-processed foods (group 4). Critics argue that the categories are overly broad and inconsistently applied. For example, wheat gluten, often added to breads, is classified as ultra-processed despite its long history in Asian cuisines and common home use.
Kevin Klatt, a nutrition scientist at the University of Toronto, noted that the NOVA definition "has absolutely nothing to do with processing." He pointed out that edamame pasta undergoes intense industrial processing yet is categorized as minimally processed because its only ingredient is edamame.
Implications for Policy and Consumer Choices
The debate matters beyond academia. Health officials like Robert F. Kennedy Jr. have used the UPF concept to push dietary guidelines that stigmatize plant-based alternatives while promoting red meat and full-fat dairy. This has contributed to declining popularity of plant-based meats and milks, which are often healthier and more sustainable.
Federal and state officials are considering regulations based on the UPF concept for programs like school meals, potentially leading to reformulations that do not improve health. Meanwhile, certifications for "non-ultra-processed" foods may lend a health halo to conventional junk foods like cookies.
Kevin Hall, a prominent nutrition scientist, acknowledged the critiques but argued that they "present plausible hypotheses" and that further research is needed. Magkos and colleagues call for more refined trials to isolate the specific attributes of UPFs.
Ultimately, the focus on ultra-processing may distract from well-established nutritional factors like calorie density, fiber content, and added sugars. As Klatt suggested, the UPF framework lacks a reasoned organizing principle beyond casting aspersions on an arbitrary set of ingredients.
