Ultra-what?

by Janice McAllister

The term “ultra-processed food” has become increasingly popular in recent years, and its pervasive use has triggered a frightening effect on consumers. The meaning of this term, however, is not clear at all, and you only have to ask around what an ultra-processed food is to realize that everyone will give a different answer. 

Unfortunately, as the American writer H.P. Lovecraft put it, “the oldest and strongest emotion of mankind is fear, and the oldest and strongest kind of fear is fear of the unknown.” Thus, ultra-processed foods (UPFs) have become the monster under the bed: nobody knows what it is, but it is still scary.

The term “ultra-processed” was coined in 2009 by the Brazilian epidemiologist Carlos A. Monteiro through the NOVA classification, which divides foods into four groups, according to their degree of processing. The last of these groups (NOVA 4) is reserved for so-called ultra-processed foods. The NOVA classification implies that the more a product has been processed and the more is unhealthy. However, no evidence confirms this view, as decades of research have proven that what really matters in a healthy diet are nutritional balance, variety, appropriate energy intake, portion size, and frequency of consumption.

Nevertheless, the NOVA classification has increasingly been used in scientific research and in policies, adding a lot of confusion in the already complex debate on healthy nutrition.

Even the definition of ultra-processed foods offered by Monteiro himself in a recent article published in The Lancet shows a surprising degree of vagueness, with constant recourse to terms such as “typically”, “often”, “usually”, “almost” and “especially”.

This vagueness is exactly what makes the definition unmeasurable and, as a result, incompatible with the scientific method, which by its very nature requires the kind of mathematical precision that allows experiments to be replicated.

The direct consequence is that the NOVA 4 group brings together foods that are completely different from one another in nutritional composition, reference portion size, calorie content, and role in the diet. To give just a few examples, NOVA 4 includes sugary drinks, yoghurts, tuna in brine, stock cubes, and wholemeal bread.

It is therefore no coincidence that, in the two studies carried out to test how well the NOVA system is understood, both nutrition experts and consumers were rarely able to assign a food to the correct NOVA group.

The term itself, moreover, does not meet scientific criteria when applied to food. The prefix “ultra” (as in ultrasound, ultraviolet, ultramundane, ultracentenarian) comes from the Latin for “beyond” and denotes the crossing of a boundary. The term “ultra-processed”, meaning “beyond processing”, makes no sense.

Yet, using a term with no scientific or legal standing can easily become a pretext for the outright demonization of industrial products. Monteiro's definition in The Lancet is in fact categorical in describing an ultra-processed food as necessarily “branded” and “industrially produced”. Other recurring expressions include “mass produced”, containing additives “of exclusive industrial use”, “industrial formulations” and so on. In practice, according to Monteiro, the same product raises the risk of disease or obesity if it is branded, while that risk disappears if it is unbranded.

In food technology, processes can take many forms (chemical, mechanical, thermal) and are often beneficial and necessary. Yet the NOVA classification treats them as if they were all the same and had the same impact on the finished product. This is hardly surprising since, as we have seen, NOVA is obviously not renowned for either accuracy or reliability.

This lack of a clear rationale, consistency, and objectivity should make it evident to everyone that NOVA cannot be trusted as a valid public health strategy. And that the use of the term “ultra-processed” is simply ultra-unreasonable, ultra-unfounded, ultra-muddled and ultra-illogical: in short, the stuff of ultra-incompetents.