Nutrition
Ultra-Processed Foods: What the Evidence Actually Says
Ultra-processed foods are blamed for a lot, but the science is more careful than the headlines. Here is what the NOVA system means, what large studies found, and how to read it without panic.
Ultra-processed foods have become one of the most talked-about topics in nutrition, often with alarming headlines. The term comes from a food-classification system called NOVA, which sorts foods by how much industrial processing they have been through rather than by their nutrients alone. The label covers a wide range of products, from soft drinks and packaged snacks to some breads and breakfast cereals.
This guide explains what the NOVA system actually classifies, what large studies have and have not shown, and why researchers urge caution before treating the link as settled cause and effect.
The essentials at a glance
- NOVA group 4, the ultra-processed category, means industrial formulations made mostly from extracted substances plus additives of no or rare culinary use, not simply any food that is cooked or packaged.
- A 2024 umbrella review pooling studies of about 9.89 million people found that higher intake of ultra-processed foods was associated with around 32 worse health outcomes, though association is not proof of cause (Lane et al., BMJ 2024).
- In a small inpatient trial, people ate about 500 more calories a day on an ultra-processed diet than on a minimally processed one matched for nutrients, suggesting processing itself can drive overeating (Hall et al., 2019). It was a small, short, single trial.
- Most of the evidence is observational, so it can be skewed by confounding: people who eat the most ultra-processed food often have less healthy lifestyles overall.
- The overall pattern of eating matters more than any single food, and the NOVA category is broad and imperfect.
What counts as ultra-processed
The NOVA system, developed by Carlos Monteiro and colleagues at the University of São Paulo in 2009 and used by bodies such as the Pan American Health Organization, Brazil's national dietary guidelines, and the FAO, sorts foods into four groups by degree of processing. Group 1 is unprocessed or minimally processed foods, such as fruit, vegetables, eggs, and plain milk. Group 2 is processed culinary ingredients, such as oils, butter, sugar, and salt. Group 3 is processed foods, such as cheese, canned vegetables, and freshly made bread, where a group-2 ingredient is added to a group-1 food.
Group 4, the ultra-processed category, is different in kind. These are industrial formulations made mostly from substances extracted from foods (such as high-fructose corn syrup, hydrogenated or interesterified oils, modified starches, and protein isolates) combined with additives of no or rare culinary use (such as colors, emulsifiers, and flavorings). They tend to be high in added sugar, fat, or salt and low in fiber. The defining feature is industrial formulation, not simply whether a food was cooked or sold in a package.
Often ultra-processed / usually minimally processed
| Often ultra-processed | Usually minimally processed |
|---|---|
| Soft drinks and energy drinks | Fresh fruit and vegetables |
| Packaged snacks and confectionery | Plain yogurt |
| Instant noodles and packet soups | Eggs |
| Reconstituted meat products | Legumes (beans, lentils) |
| Mass-produced packaged breads | Freshly made bread |
Even with these examples, the category is broad and imperfect. Some wholegrain breakfast cereals and wholegrain breads land in group 4 alongside soft drinks, despite having very different nutritional profiles. That breadth is one reason researchers debate how useful the label is, a point the next section returns to.
What the evidence shows
The largest recent summary is a 2024 umbrella review in the BMJ by Lane and colleagues. It pooled about 14 meta-analyses covering 45 analyses and roughly 9.89 million participants across seven areas of health. Higher exposure to ultra-processed foods was associated with around 32 adverse outcomes. The authors graded the strength of each link. The associations they rated as convincing included cardiovascular-disease mortality (relative risk about 1.50, 95% confidence interval 1.37–1.63), type 2 diabetes (about 1.12, 1.11–1.13), common mental disorders (odds ratio about 1.53, 1.43–1.63), and anxiety (about 1.48, 1.37–1.59). Links they rated as highly suggestive included all-cause mortality (about 1.21, 1.15–1.27), depression (about 1.22, 1.16–1.28), and obesity (about 1.55, 1.36–1.77). Importantly, the certainty of evidence for most outcomes was low or very low, and about 34 of the analyses showed only weak or no evidence.
A separate, smaller study gives the clearest hint that processing itself can matter. In a 2019 inpatient trial run by the US National Institutes of Health, 20 weight-stable adults spent two weeks on an ultra-processed diet and two weeks on a minimally processed one, in random order. The two diets were matched for presented calories, energy density, macronutrients, sugar, sodium, and fiber, and participants could eat as much as they wanted. On the ultra-processed diet they ate about 500 more calories a day and gained roughly 0.9 kg (about 2 lb), while losing a similar amount on the minimally processed diet. Because this was randomized, it points toward processing causing the extra intake, but it was a small trial of just 20 people over two weeks per arm, so it is a strong signal rather than the final word.
These findings come with real caveats. The umbrella-review evidence is observational, which means it can show associations but cannot prove that ultra-processed food causes the outcomes. Residual confounding is a particular concern: people who eat the most ultra-processed food often smoke more, exercise less, and have lower incomes, and those factors affect health independently. NOVA itself is contested. Critics note that group 4 lumps together items as different as soft drinks and wholegrain bread, and it is not clear whether the NOVA label predicts risk any better than simply flagging foods high in fat, salt, and sugar.
Frequently asked questions
- Are ultra-processed foods bad for me?
- Higher intake is associated with worse health outcomes in observational studies, but association is not proof of cause, and the certainty of most of that evidence is low. The overall eating pattern matters more than any single food.
- What makes a food 'ultra-processed'?
- Under the NOVA system it is group 4: an industrial formulation made mostly from extracted substances plus additives of no or rare culinary use, such as emulsifiers, colors, and flavorings. It is not simply any food that has been cooked or packaged.
- Is all processed food unhealthy?
- No. Minimal or basic processing (NOVA groups 1 to 3) is different from the ultra-processed group 4, and even some group-4 items, such as wholegrain bread, behave differently in studies than products like soft drinks.
- Do I need to avoid ultra-processed foods completely?
- General guidance leans toward minimally processed foods most of the time rather than eliminating any single food. Small, sustained swaps within an overall balanced pattern are more practical than a strict ban.
References
- Ultra-processed foods, diet quality, and health using the NOVA classification system · Food and Agriculture Organization of the United Nations (FAO). Accessed 2026-05-27.
- Ultra-processed food exposure and adverse health outcomes: umbrella review of epidemiological meta-analyses (Lane et al.) · BMJ, 2024 (via PMC, PMC10899807). Accessed 2026-06-05.
- Ultra-Processed Diets Cause Excess Calorie Intake and Weight Gain (Hall et al.) · Cell Metabolism, 2019 / NIH (National Institutes of Health). Accessed 2026-06-05.