Nutrition: NOVA - why how food is processed matters as much as what's in it

For most of nutrition history, we argued about nutrients — how much fat, how many carbs, which proteins. NOVA changed the question: it asked what we'd done to the food before we ate it.

Read the label on a can of chickpeas: chickpeas, water, salt. Three ingredients, all recognisable. Now read the label on a popular protein bar: glucose syrup, soy protein isolate, palm kernel oil, modified starch, humectant (sorbitol), flavouring, emulsifier (soy lecithin), coating (sugar, cocoa butter, milk powder), salt, vitamins. Seventeen ingredients, half of them industrial.

Both products can deliver 15 grams of protein. By traditional macro tracking, they look identical. They are not.

NOVA is a food classification system that captures what macro tracking misses: the degree to which food has been industrially processed before it reaches your plate. Developed by a research group at the University of São Paulo and adopted by the World Health Organization, NOVA has become one of the most important frameworks in modern nutrition science — and the evidence base supporting it has grown substantially in the decade since its introduction.¹

The four NOVA groups

NOVA 1 — Unprocessed and minimally processed foods

Foods either eaten in their natural state or subjected only to processes that don't fundamentally alter their composition: washing, cutting, freezing, drying, fermenting, or cooking at home. This group includes vegetables, fruit, meat, fish, eggs, milk, plain yogurt, legumes, nuts, seeds, and whole grains.

These are the foods that form the foundation of every traditional diet associated with health and longevity, from the Mediterranean to the Blue Zones.

NOVA 2 — Processed culinary ingredients

Substances extracted from NOVA 1 foods or from nature, used in the preparation and cooking of meals: olive oil, butter, flour, salt, sugar, vinegar. These are not typically eaten alone — they're how we cook. A meal of grilled salmon with roasted vegetables and olive oil is NOVA 1 prepared with NOVA 2 ingredients. The distinction matters: sugar added to a home-baked cake is NOVA 2; sugar added to an industrial product is part of a NOVA 4 formula.

NOVA 3 — Processed foods

Foods made by combining NOVA 1 and NOVA 2 ingredients and applying preservation processes: tinned fish, cured meats, cheese, pickled vegetables, beer and wine, freshly baked bread. These foods are recognisably derived from their base ingredients, are made to last longer than fresh equivalents, and are typically consumed in moderate amounts as part of a broader diet. Their health implications are mixed rather than uniformly negative.

NOVA 4 — Ultra-processed foods

This is the category that has driven most of the research interest in NOVA — and most of the concern. Ultra-processed foods are industrial formulations made mostly or entirely from substances extracted or derived from foods (oils, fats, starches, protein isolates, sugars), combined with additives that have no culinary equivalent: emulsifiers, stabilisers, artificial flavours and colours, flavour enhancers, preservatives designed to extend shelf life for months or years.

The defining characteristics are not individual ingredients but the overall profile: they are designed to be hyper-palatable, to minimise satiety, to last a very long time, and to require minimal preparation. Soft drinks, packaged snacks, fast food, most breakfast cereals, reconstituted meat products, flavoured yogurts, instant noodles, most packaged breads, and ready meals are all NOVA 4.

Why processing matters — the emerging evidence

For most of nutrition epidemiology's history, the central question was: which nutrients cause which diseases? Fat was accused, then rehabilitated. Carbohydrates were accused, then partially rehabilitated. The results were inconsistent partly because nutrients don't travel alone — they come packaged in foods with dozens of other compounds and structural properties that affect how the body processes them.

NOVA reframed the question. Instead of asking "how much saturated fat?", it asks "how much of your diet has been industrially reformulated?" The results have been more consistent.

A landmark study in Cell Metabolism in 2019 used a randomised controlled design — the gold standard in nutrition research, which is rare — to test ultra-processed versus minimally processed diets directly. Participants eating ad libitum (as much as they wanted) on a NOVA 4-heavy diet consumed an average of 508 kcal more per day and gained weight. The same participants on a minimally processed diet lost weight. The diets were matched for presented calories, sugar, fat, carbohydrates, and fibre. The difference was the degree of processing.²

A 2022 umbrella review in The BMJ synthesising evidence from 45 pooled analyses found that higher consumption of ultra-processed foods was consistently associated with elevated risk of cardiovascular disease, type 2 diabetes, obesity, depression, and all-cause mortality.³ The associations held even after controlling for total calorie intake and nutrient composition — suggesting that the processing itself, not just what's in the food, is driving outcomes.

The mechanisms under investigation include: disruption of gut microbiome diversity from emulsifiers and additives; breakdown of the food matrix that normally slows digestion and supports satiety; hyper-palatability engineering that overrides normal appetite regulation; and the displacement of nutrient-dense foods from the diet.

What this means practically

The research does not suggest that eating a biscuit will cause disease, or that any NOVA 4 food consumed is a health threat. What it shows is that a dietary pattern dominated by NOVA 4 foods — which in many Western countries now accounts for more than 50% of calorie intake — is associated with worse health outcomes across nearly every measured dimension.⁴

The practical insight is directional rather than prescriptive. A diet weighted toward NOVA 1 and 2 — whole foods, home cooking, minimally processed ingredients — is consistently associated with better outcomes regardless of the specific macronutrient profile. A diet weighted toward NOVA 4 is consistently associated with worse outcomes, for reasons that appear to go beyond the nutrients contained.

This matters for how you read your nutrition data. Hitting a protein target with a protein isolate shake and processed meat is not the same as hitting it with chicken, eggs, and lentils, even if the numbers match. The NOVA score attached to each meal is an attempt to surface that difference in a way that total macro tracking cannot.

How to improve it

• Cook from whole ingredients as your default. A meal of chicken, vegetables, and olive oil is NOVA 1 regardless of how simple or elaborate the recipe. The act of cooking from scratch almost guarantees a low NOVA score.
• Read the ingredient list, not the nutrition panel. A short ingredient list of recognisable components is a reliable proxy for NOVA 1–2. A long list with industrial additives is NOVA 4, regardless of what the front-of-pack claims.
• Replace the most frequent NOVA 4 items first, not the occasional ones. If you eat a packaged breakfast cereal every morning, switching to oats or eggs has more impact than avoiding the occasional biscuit. Frequency is what determines dietary pattern.
• Fermented foods (yogurt, kefir, kimchi, sourdough) are NOVA 3 at most. They carry additional gut health benefits and are meaningfully different from ultra-processed options, even when they contain some processing.
• Packaged foods are not all NOVA 4. Tinned tomatoes, frozen vegetables, plain yogurt, tinned fish, and many cheeses are NOVA 1–3. Convenience and processing are not the same thing.
• Be sceptical of health claims on packaging. "High protein", "low fat", "added fibre", "no artificial colours" are front-of-pack marketing. NOVA 4 products routinely carry these claims. The ingredient list tells you more.
• Batch cooking reduces the friction that drives NOVA 4 eating. When whole-food options are immediately available, the appeal of ultra-processed convenience decreases significantly.
• Aim for 80%+ of calories from NOVA 1–2 as a practical target. This leaves room for processed foods (cheese, tinned fish, bread) and the occasional ultra-processed item without requiring perfection.

The 100 Great Years perspective

NOVA matters to 100 Great Years because longevity isn't just a matter of hitting nutrient targets. It's about building a dietary pattern that supports your body across decades — one that works with your gut, your metabolism, and your appetite regulation, rather than against them.

Ultra-processed foods have been engineered to be consumed in large quantities. That engineering works. The research on their long-term health consequences has become hard to dismiss. Tracking NOVA scores alongside macros is an attempt to give you a fuller picture of what you're actually eating — not to create anxiety about individual choices, but to surface patterns that macro numbers alone cannot show.

A diet built mostly on whole foods, prepared at home from recognisable ingredients, is as close to a consensus longevity recommendation as nutritional science has produced. NOVA gives that recommendation a framework.

Find out how your nutrition scores

Get your free Health and Wealth scores in 5 minutes.

This article is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional before making decisions about your health.