The arguments against ultra-processed foods reached their apex near the end of last year, when prestigious medical journal The Lancet published a series of papers on the topic.
The series was a broad-based critique of (some would say attack on) UPFs, highlighting the harms they’ve been linked to, the food environment they allegedly create, and the supposedly predatory behaviour of the industry that produces them.
The debate around UPFs goes something like this. On the one hand, critics argue that UPFs are hyper-palatable, are linked to widespread cases of poor health and non-communicable disease, and are designed to be addictive.
The counter-argument posits that the very definition of UPFs, stemming as it does from the Nova classification, is based on processing and has nothing to do with nutrition, meaning that it is not a good judge of a product’s health benefits or drawbacks.
It is easy to boil this debate down into the caricature of it – on the one hand, an evasive and compromised industry using technicalities to defend the indefensible; on the other, a dogmatic band of crusaders using scientifically dubious concepts to hoodwink the public.
Unsurprisingly, this is misrepresentative of the truth on both sides. If we want to understand the debate, we must take both arguments seriously.
Nova is inadequate as the basis of policy
The most widely accepted definition of UPFs is based on the fourth category of the Nova classification, which defines foods developed mainly through industrial processing techniques and containing little-to-no whole food components.
Such individual foods are not, inherently, unhealthy. The Nova classification does not refer to nutritional content, and it is perfectly possible for foods to be both ultra-processed and high in specific beneficial nutrients such as protein. Processing methods are not a direct substitute for health content.
Criticisms focusing on the formal disconnect between nutrition on the one hand and processing on the other are not just nitpicking. They highlight the important fact that a food’s processing method alone cannot capture its full health impact, and thus cannot work as a guide for consumers, regulators, or industry.
There are foods designated UPF under the classification that can provide nutritional benefits – pre-packaged wholegrain bread, for instance, or tinned baked beans. Some UPFs can even have other advantages – think of the sustainability potential of plant-based meat, for instance.
Not all UPFs are the same – even Nova’s creator, Carlos Monteiro, admits this. To provide clarity for regulators and industry alike, a definition which captures such nutritionally diverse foods is not adequate to fully combat the effects of poor health.
Potential harms of the ‘UPF diet’ cannot be dismissed
Ultra-processed foods are, argues the Lancet series, best understood as a dietary pattern. According to the researchers, the growth of UPF consumption in low and middle-income countries has been accompanied by a decline in the consumption of important nutrients like protein, as well as health-promoting phytochemicals.
Addressing the prevalence of UPFs within the food system cannot be a bad thing given these findings, even if some nutrient-rich foods end up being included under the UPF umbrella.
The arguments against UPFs are about looking at the big picture. Sure, using Nova in this context isn’t perfect, but when researchers have linked so many poor health outcomes to these foods, dismissing the concept outright seems counterproductive. UPFs may not all be unhealthy, but a great many are.
Furthermore, to say that there is no relation between health and processing isn’t entirely accurate. Disruptions to the food matrix, caused by processing, can affect the way the body absorbs nutrients, meaning that UPFs have key differences to other foods directly linked to their processing method.
The UPF concept deserves a place in our discourse; it provides a useful shorthand for a dietary pattern which a significant body of research links to harm. Nevertheless, it cannot be the sole basis for regulation.
Nova can inform the solution, but not define it
How can the two sides of this debate, perpetually at loggerheads, inform each other and come to a strong solution?
The relationship between processing and poor health outcomes cannot be dismissed outright. There is such a strong correlation between foods that fit Nova’s definition of UPF, and foods linked to poor health outcomes, that to ignore this link would severely limit the full picture.
Nevertheless, while the concept of UPF can inform policy, Nova alone cannot form its basis.
In any policy or guideline, processing must be taken into account, but it cannot dominate everything. A food’s processing level can affect how it’s regulated, but mitigating circumstances, like nutrient density, should also influence the final verdict.
The UPF concept is crucial to understanding the relationship between food and health, especially in the modern world. But it does not provide the full picture, and relying on it alone certainly cannot provide us with the basis for clear, objective guidelines or regulations.
