OptimallyThe short answer
NOVA classifies foods by the nature, extent and purpose of processing. A label can offer clues—such as protein isolates, flavourings, sweeteners or emulsifiers—but no single ingredient list is a perfect shortcut for every product. Ultra-processed foods also vary widely, and NOVA is not a nutrient score or a safety verdict on an additive. A small controlled NIH study found participants ate more and gained weight during its ultra-processed menu phase; it tested two menu patterns in 20 adults, not the isolated effect of processing itself.
What NOVA is trying to describe
NOVA groups foods according to processing. In broad terms, it distinguishes unprocessed or minimally processed foods, culinary ingredients, processed foods made by combining these, and ultra-processed products that are industrial formulations. The original system focuses on the purpose and character of processing, not just whether a food has been cooked, frozen, canned or packaged.
A jar of plain tomatoes and a ready-to-eat snack are both processed in ordinary language, but that alone does not place them in the same NOVA group. Classification can involve judgement at the boundaries.
Read the list as a clue, not a mechanical rule
Protein isolates, modified starches, flavourings, colours, sweeteners and some emulsifiers can be clues to an industrial formulation. They do not share one function or safety profile, and a gum is not automatic proof of ultra-processing in every context.
Compare two packaged breads in the shop. A longer ingredient list can prompt a closer look, but it does not settle the classification. Consider why each ingredient is present and how the product is formulated. For a quick nutrient comparison, the FDA recommends checking serving size and the Nutrition Facts panel too; NOVA and nutrient values answer different questions.
What the NIH feeding study found
In a 2019 NIH randomized crossover trial, 20 adults received an ultra-processed menu and an unprocessed menu for two weeks each. The menus were designed to be similar in presented calories and several nutrients, yet participants ate more and gained weight during the ultra-processed phase. This is valuable experimental evidence about those two menus and participants.
The trial did not isolate processing as the only cause; texture, eating rate and food form could matter. It did not test every food or population.
Use scanning to make the comparison clearer
Optimally can scan a product barcode or photograph an ingredient label, then explain the factors behind its editorial food-quality rating. That can help you notice ingredients and available nutrition information while shopping. It is not a clinical score, and its result is not a formal NOVA determination for every borderline product. Read the explanation and confirm the package details yourself.
Common questions
Does every processed food count as ultra-processed?
No. Processing includes common methods such as pasteurising, freezing and canning. NOVA distinguishes several groups.
Does one additive always make a product ultra-processed?
No single additive is a universal shortcut. NOVA classification considers the formulation and purpose of processing.
Sources & further reading
Author perspectives describe their writers’ ideas. Research and official guidance support the specific points discussed above.
- Monteiro et al., “Ultra-processed foods: what they are and how to identify them”
- NIH Clinical Center: controlled ultra-processed diet trial
Twenty adults, two menu phases of two weeks each.
- FDA: How to Understand and Use the Nutrition Facts Label