NOVA, the food classification system developed by Brazilian researchers in 2009, sorts everything you eat into four groups based on the nature, extent, and purpose of the processing it has undergone: unprocessed or minimally processed foods, processed culinary ingredients, processed foods, and ultra-processed foods. The distinction that matters most for health is the last one. Studies using NOVA, including the large French NutriNet-Santé cohort and three major U.S. prospective cohorts, have linked high intakes of ultra-processed foods to cardiovascular disease, type 2 diabetes, and weight gain—though these are observed associations, not proven causes.
Understanding the NOVA Food Classification System
NOVA was created to cut through confusion about the loosely used term "processed food," as UC Davis food writers explain. Rather than judging a food by its nutrients alone, the system asks what was done to it and why. A paper by the system's architects, published in a public-health journal, describes NOVA as "the food classification based on the nature, extent and purpose of food processing that places foodstuffs in four groups, one of which is specified as ultra-processed food and drink products."
The framework has been adopted widely in nutrition research. The U.S. National Cancer Institute uses NOVA to classify foods in dietary assessment tools and national survey data, and a World Health Organization report cited by the NOVA authors applied it to compare diets across the USA, Canada, Brazil, and Colombia. Harvard's Nutrition Source notes that the system is recognized by the World Health Organization, the Food and Agriculture Organization, and the Pan American Health Organization—though not currently by the U.S. Food and Drug Administration or USDA.
NOVA Group 1: Unprocessed or Minimally Processed Foods
Group 1 covers the natural edible parts of plants and animals, altered only in ways that do not add ingredients: washing, drying, freezing, grinding, or vacuum-packing. Think fresh vegetables and fruit, whole grains, meat, and even pasta, which UC Davis lists here because it is made from a single basic food without additives. The key test is that nothing meaningful has been added—no salt, sugar, or fat beyond what the food itself contains.
Minimal processing exists mostly to preserve foods or make them ready to eat, not to change their fundamental character. Frozen vegetables, pasteurized milk, and dried legumes all remain Group 1 because the processing serves storage and safety rather than reformulation. This is why NOVA's logic can surprise people: a frozen pea is closer to the garden than a "natural-sounding" packaged snack.
NOVA Group 2: Processed Culinary Ingredients
Group 2 contains substances extracted from Group 1 foods or from nature and pressed, refined, or milled for use in the kitchen: oils, butter, sugar, and salt, along with herbs and spices. These ingredients are rarely eaten on their own; their role is to season and prepare dishes made from Group 1 foods, typically in small quantities.
The category exists to acknowledge a middle ground between whole foods and manufactured products. Olive oil pressed from olives, or cane sugar refined from cane, has certainly been processed—but the purpose is extraction of a single usable substance, not construction of a new food. When these same ingredients appear inside a factory formulation rather than your own cooking, the product they belong to is judged by a different group entirely.
NOVA Group 4: Ultra-Processed Foods
Ultra-processed foods are industrial formulations that go beyond added salt, sugar, and fat to include substances you would not find in a home kitchen: artificial colors and flavors, preservatives, thickeners, emulsifiers, and artificial sweeteners. As Harvard's Nutrition Source describes them, these additives "promote shelf stability, preserve and enhance texture, and increase palatability," and several processing steps using multiple ingredients are typically involved. They are often mass-produced from low-cost ingredients, making them cheap and highly profitable.
Researchers writing in the European Journal of Clinical Nutrition point out that NOVA ties this category to both technological processing and formulation—the specific ingredients used and the sheer number of them in a recipe. That combination is what separates a packaged cookie from a home-baked one made with flour, butter, and sugar. The same researchers caution that the system's descriptive approach "opens the door to ambiguity and differences in interpretation," and that even experts disagree when applying it—so borderline cases are a known feature, not a personal failing.
Health Implications of Ultra-Processed Food Consumption
The strongest research case against ultra-processed foods comes from cohort studies that explicitly use NOVA. A prospective study of the NutriNet-Santé cohort examined ultra-processed food intake and cardiovascular disease risk; the same French cohort was analyzed for BMI change and risk of overweight and obesity in a 2020 PLOS Medicine paper. Three large U.S. prospective cohorts were used to study ultra-processed food consumption and type 2 diabetes risk, reported in Diabetes Care in 2023.
These studies observe associations rather than prove causation, and the Academy of Nutrition and Dietetics has reviewed the evidence base with that framing. One mechanism the NOVA authors themselves emphasize is displacement: their summary of NOVA-based studies concludes that ultra-processed products "displace unprocessed or minimally processed foods and freshly prepared dishes and meals." A diet heavy in Group 4 crowds out the foods and meals built from Groups 1 and 2. The European Journal of Clinical Nutrition authors add an open question—whether the health links stem from food structure or from specific ingredients and additives—which remains unresolved.
NOVA Food Classification Decision Framework
Use the sequence of questions below to classify any food. Answer them in order; the first "yes" places the food in a group. The framework applies NOVA's exclusionary criteria—the purpose and extent of processing—to resolve ambiguous cases.
| Step | Question | If yes | Worked example |
|---|---|---|---|
| 1 | Is it an edible part of a plant or animal, altered only by washing, drying, freezing, grinding, or vacuum-packing, with no added ingredients? | Group 1: Unprocessed or minimally processed | Frozen peas, fresh fruit, plain milk |
| 2 | Is it a substance extracted and refined for cooking—oil, butter, sugar, salt, herbs, spices—used to prepare dishes rather than eaten alone? | Group 2: Processed culinary ingredient | Olive oil, table salt |
| 3 | Is it a simple combination of Group 1 and Group 2 items, such as food preserved with salt or made at home from few ingredients? | Group 3: Processed food | Canned vegetables with salt |
| 4 | Does it contain industrial ingredients not used in home cooking—emulsifiers, artificial flavors, colors, preservatives, thickeners, or artificial sweeteners—across multiple processing steps? | Group 4: Ultra-processed | Soft drinks, packaged snacks |
The classic ambiguous case is yogurt. Harvard's Nutrition Source explains that plain yogurt is minimally processed, but fruited yogurt with added sweeteners could be labeled either processed or ultra-processed depending on how much sweetener and which chemical additives it contains. Run it through the matrix: plain yogurt stops at Step 1; a sweetened fruit version fails Step 1, and its final group depends on whether the ingredient list crosses the line into industrial additives at Step 4.
Two habits make the framework reliable in practice. First, read the ingredient list rather than the front of the package—NOVA classifies by ingredients and process, not marketing language. Second, expect genuine borderline cases; expert reviewers have documented disagreement even among food and nutrition specialists, so a defensible either-way call on a borderline product is consistent with the system, not a misuse of it.
Making Informed Food Choices
The evidence-based takeaway is to prioritize foods from NOVA Groups 1 and 2, and to be mindful of intake of Group 4 ultra-processed foods. This article explains a classification system and reports research associations; it is not medical or dietary advice, and it teaches the method of classification rather than judging specific branded products.
A measurable indicator of progress is a shift in your grocery purchases toward fewer items with long ingredient lists containing industrial additives, and an increase in whole or minimally processed foods. You can check this directly: compare labels on your next shopping trip and count how many products list emulsifiers, artificial flavors, or sweeteners among their ingredients.
Sources
- The UN Decade of Nutrition, the NOVA food classification and the trouble with ultra-processing - PMC
- Using Nova with ASA24 & NHANES — EPI
- What to Know About Processed and Ultra-Processed Food — UC Davis
- Processed Foods and Health • The Nutrition Source — The Nutrition Source - Harvard Chan School
- Ultra-processed foods: how functional is the NOVA system? - European Journal of Clinical Nutrition — Nature
