The most significant shift in nutrition science of the past decade isn’t about a specific nutrient. It’s about a category of food defined not by what it contains but by how it’s made. Ultra-processed food — and the growing body of research linking it to health outcomes — represents a genuinely different way of thinking about food quality, one that goes beyond the ingredient list.

What ultra-processed means — the NOVA classification

The NOVA classification, developed by Brazilian researchers and now widely used in nutrition research, divides foods into four groups based on the extent of industrial processing — not on nutrient content.

Group 4 — ultra-processed food — is defined as industrial formulations made entirely or mostly from substances extracted from foods (oils, fats, sugar, starch, protein isolates) or derived from food constituents (hydrogenated fats, modified starches) or synthesised in laboratories (flavour enhancers, colourings, emulsifiers, texturisers, stabilisers), with little or no whole food.

The examples are familiar: mass-produced bread and pastries, packaged snacks, carbonated soft drinks, instant noodles, chicken nuggets, most breakfast cereals, most flavoured yoghurts, most ready meals. Also: many products marketed as healthy — protein bars, meal replacement shakes, diet foods — that use the same industrial formulation approach.

Why processing matters beyond nutrients

The research on ultra-processed food consistently shows associations with adverse health outcomes — obesity, type 2 diabetes, cardiovascular disease, certain cancers — even after adjusting for nutrient content. This suggests the effect isn’t entirely explained by what the food contains (too much sugar, salt, saturated fat) but by something about the food itself, related to how it was made.

Proposed mechanisms include: ultra-processing disrupts the food matrix that normally modulates digestion rate, leading to faster glucose absorption; additives like emulsifiers may affect the gut microbiome; ultra-processed food is engineered for palatability in ways that undermine the body’s satiety signals; and the form of food — whether nutrients arrive in a whole-food matrix or a fragmented industrial one — may affect metabolism independent of the nutrient composition.

The research is observational and causality is difficult to establish cleanly, but the signal across multiple large cohort studies is consistent enough to take seriously.

What the label shows vs what it doesn’t

The Nutrition Facts panel shows you macronutrients, some micronutrients, and energy content. It doesn’t show you the degree of processing. A product with a “healthy” nutrition profile — low calories, reasonable protein, some vitamins from fortification — can score high on the NOVA ultra-processed scale. The ingredient list is more informative: if it contains emulsifiers (lecithin, carrageenan, polysorbates), stabilisers, flavour enhancers (glutamate compounds), modified starches, or more than a handful of unfamiliar additives, you’re looking at an industrially formulated product regardless of what the nutrition panel claims.

A practical framework

It’s not practical or necessary to eliminate ultra-processed food entirely, and the evidence doesn’t require that conclusion. But the research does support a useful heuristic: foods that look like food — with recognisable ingredients, minimal additives, not reconstructed from extracts — tend to be better for health than foods that are assembled in a factory from separated components. The more the ingredient list contains things you’d find in a kitchen rather than a laboratory, the further from ultra-processed you are.