NOVA is a food classification system that sorts everything you eat into four groups based on how much industrial processing it has undergone. Developed in 2009 by Carlos Monteiro, a professor of nutrition and public health at the University of São Paulo in Brazil, the system introduced the term “ultra-processed foods” and created a framework that researchers worldwide now use to study how processing affects health. Unlike traditional nutrition labels that focus on calories, fat, or sugar content, NOVA classifies food by what was done to it before it reached your plate.
The Four NOVA Groups
NOVA divides all food into four categories, each defined by the extent and purpose of processing.
Group 1: Unprocessed or Minimally Processed Foods
These are plant or animal foods that have been altered only enough to make them safe, edible, or storable. Think fresh fruit, dried beans, eggs, plain milk, frozen vegetables, roasted coffee beans, or chilled meat. The processes involved (washing, peeling, freezing, pasteurizing, fermenting without alcohol) don’t add any new substances to the food. A bag of frozen broccoli and a carton of pasteurized milk both fall here.
Group 2: Processed Culinary Ingredients
These are substances extracted from Group 1 foods and used for cooking, not eaten on their own. Olive oil, butter, sugar, salt, flour, and vinegar all belong in this group. They’re made through pressing, refining, grinding, or milling. The idea is that you combine them with Group 1 foods in your kitchen to make meals from scratch.
Group 3: Processed Foods
When you take a Group 1 food and add salt, oil, sugar, or another Group 2 ingredient to it, you get a processed food. Canned fish packed in oil, freshly baked bread from a bakery, cheese, fruits in syrup, and bottled vegetables are typical examples. Most processed foods have two or three ingredients, and you can still clearly recognize the original food. A can of chickpeas in salted water, for instance, is Group 3: you can see the chickpeas, and the only addition is salt and water.
Group 4: Ultra-Processed Foods
This is the category that generates the most attention. Ultra-processed foods aren’t simply modified versions of whole foods. They’re formulations made mostly or entirely from substances derived from foods and additives, with little if any intact Group 1 food remaining. Soft drinks, packaged snacks, reconstituted meat products, instant noodles, and pre-prepared frozen meals are common examples. If you bought commercially made hummus and noticed guar gum listed as a stabilizer on the label, that hummus would land in Group 4, even though homemade hummus from the same chickpeas would not.
How to Spot an Ultra-Processed Food
The ingredient list is your main tool. Two types of ingredients signal that a product is ultra-processed: industrial food substances and cosmetic additives.
Industrial food substances are ingredients you’d never use in a home kitchen. These include high-fructose corn syrup, maltodextrin, dextrose, invert sugar, hydrogenated oils, interesterified oils, hydrolyzed proteins, soy protein isolate, casein, whey protein, and mechanically separated meat. If you see these on a label, the product is almost certainly Group 4.
Cosmetic additives are added to make the final product look, taste, or feel more appealing. Flavors, flavor enhancers, colors, emulsifiers, sweeteners, thickeners, anti-foaming agents, and glazing agents all fall into this category. These additives are used exclusively in ultra-processed food manufacturing, not in home or restaurant cooking.
Why the System Matters for Health
NOVA’s real impact has been in enabling large-scale research on what ultra-processed foods do to the body. A major umbrella review covering the available observational evidence identified 25 health outcomes linked to ultra-processed food consumption. The strongest evidence connected these foods to decline in kidney function (25% higher odds) and wheezing in children and adolescents (42% higher odds). Five additional outcomes had highly suggestive evidence behind them: diabetes, overweight, obesity, depression, and other common mental health disorders.
These are associations from observational studies, meaning researchers tracked what people ate and what happened to their health over time. The patterns are consistent enough across populations that public health organizations have taken notice, but the system was designed as a research and policy tool, not a personal scoring system for every item in your grocery cart.
Where NOVA Falls Short
The system has drawn significant criticism from nutrition scientists, and the objections are worth understanding if you plan to use NOVA as a guide for your own eating.
The biggest issue is that Group 4 is enormous and treats very different foods as equivalent. A fortified whole-grain breakfast cereal and a bag of candy both qualify as ultra-processed, despite having vastly different nutritional profiles. Two versions of popcorn, one made with corn, oil, and salt (Group 3) and another with the same ingredients plus a natural flavor or color (Group 4), receive different classifications even though their nutritional content is nearly identical. Carbonated beverages all get classified as Group 4 regardless of ingredients, so a home-carbonated lemonade made with fresh lemons sits in a lower group than an identical commercially produced version.
Cultural blind spots are another concern. Tofu, a protein-rich staple in Asian cuisines for centuries, has been labeled ultra-processed by some food scientists applying NOVA criteria. This reduces a traditional food with clear nutritional value to the same category as a candy bar. Critics argue the system’s emphasis on industrial formulation creates an arbitrary line: if a process is acceptable at home but harmful in a factory, the distinction seems to be about the setting rather than the food itself.
There’s also a practical problem with the research. The food frequency questionnaires used in most large nutrition studies can’t reliably distinguish between mass-produced commercial bread containing emulsifiers and preservatives (Group 4) and an artisanal loaf from a bakery (Group 3). This blurriness weakens the precision of studies linking ultra-processed food intake to health outcomes.
Finally, some researchers point out that processed foods classified as Group 4 contribute meaningfully to essential nutrient intake in the American diet. For lower-income households, certain ultra-processed foods are cost-effective sources of important nutrients. A blanket recommendation to avoid all Group 4 foods could, in some cases, reduce dietary quality rather than improve it.
How Governments Are Using NOVA
Despite its prominence in research, NOVA has not been widely adopted as the basis for official dietary guidance. Brazil’s dietary guidelines, written with input from Monteiro’s team, were among the first to incorporate processing-based advice, recommending that people prioritize freshly prepared meals and avoid ultra-processed products. A handful of other countries have followed with similar language. But most national dietary guidelines, including those in the United States and Australia, still organize their recommendations around nutrient content and food groups rather than degree of processing.
The system’s influence shows up more in research and public awareness than in formal policy. NOVA gave scientists a common language to study processing, and it gave consumers a new lens for reading ingredient lists. Whether or not you follow NOVA strictly, its core insight is practical: the more an ingredient list reads like a chemistry set rather than a recipe, the further that food has traveled from anything you’d make in your own kitchen.