Mention the term ‘ultra-processed foods’ (UPFs) to a group of food industry professionals, and a heated debate is almost guaranteed to follow. UPFs are gaining ground as a primary cause of the dietary ills of the world, among media and researchers alike, but it is a concept of much controversy.

While scientists, industry actors and authorities in many countries have accepted that there is a clear definition of what constitutes UPFs, that UPFs are potentially harmful and therefore should be limited, many others are fighting the use of this term.  

In a recently published 70-page report commissioned by the Swedish Nutrition Foundation (SNF), a group of researchers calls for industry and media to stop using the term ‘ultra-processed food’.  

The report starts by explaining that while food processing can have negative effects, it also adds a lot of value by improving digestibility, sustainability, bioavailability, accessibility, and more. The authors then move on to review eight different classification systems of processed food:

• NIPH (National Institute of Public Health, Mexico)
• ARC-EPIC (International Agency for Research on Cancer, Europe)
• NOVA
• IFPRI (International Food Policy Research Institute, Guatemala)
• IFIC (International Food Information Council Foundation, US)
• FSANZ (Food Standard Australia New Zealand)
• UNC (the University of North Carolina, US)
• Siga (France)

While the report acknowledges that the use of classification systems can be helpful by allowing us to study the effects of dietary patterns, it is largely critical towards all of the above classifications and definitions. The authors claim that they are inconsistent, non-systematic and vague, using undefined terms like “natural” and “healthy”. Meanwhile, they don’t measure the degree of processing, sometimes using a mix of processing degree and ingredients to define something as (ultra)processed.

Another weakness, according to the report, is that there is no consensus of what constitutes processing. The way the term is used in classification systems differs from how it is used in research and food technology. For example, milk and yoghurt count as processed from a food technology perspective – but according to NOVA, these are unprocessed foods. Carbonated soft drinks count as ultra-processed in NOVA but not in food technology, and so on.

Indeed, research has confirmed that the amount of foods that are considered processed vary greatly between the different systems. For example, NOVA classifies 8-10% of foods as highly or ultra-processed; IARC-EPIC classifies around 46-47% as highly or ultra-processed, according to studies published in 2021 and 2022.

Finally, SNF’s report highlights that from a consumer perspective, ‘ultra-processed food’ is difficult to understand. It references a study from Brazil which showed that consumer understanding of the term was low – despite it being used in the country’s dietary guidelines for eight years.

The authors conclude that “Currently, there are many ambiguities and confusions as to what is included in the concept ‘ultra-processed’. The methodology for classifying ultra-processed foods is not sufficiently developed and quality-assured for it to be used. The term ‘ultra-processed foods’ should therefore not be used, but instead we should describe specifically which foods are referred to.”

A copy of the report, published in Swedish with a summary in English, can be accessed here: https://nutritionsfakta.se/2023/05/25/idag-den-25-maj-lanserar-nutritionsfakta-rapporten-processade-livsmedel/