Processing Terminology

Processed Food Confusion Is a Problem of Language

The limits of my language means the limits of my world.
- Ludwig Wittgenstein

Consider for a moment how often you’ve heard the term “ultra-processed food” during the past week. Five years ago, I’m willing to bet if you had heard the term it would have barely registered in your consciousness. Yet now, “ultra-processed food” (UPF) is part of the common vernacular in much of the English-speaking world. But do you really know what it means?

“Yes,” you object. “Of course I do! Ultra-processed foods are those foods that have been heavily processed on an industrial scale and have a long list of ingredients including some that aren’t readily available in a typical home kitchen.”

But is this correct? Does this definition account for any nuance? Does everyone agree with it? Despite its ubiquitousness, not only does the term “ultra-processed food” contribute to much confusion among the public, but most experts argue that a much-needed standardised, formally accepted definition has yet to be agreed. Indeed, by acknowledging that a lack of a definition is a concern, many academics, dietitians, nutritionists and food-industry professionals – including many respected friends of mine – are inferring that they don’t know what the term really means. Despite this, many continue to flippantly use the term when describing those processed foods that we should only be consuming in moderation. Why are people – who ought to know better – continuing to casually use a phrase that they agree lacks a suitable definition? Here we have a problem of language as much as one of nutrition science.

The Processed Mindset Story So Far

I’ve discussed the problems of an overly-maligned perspective of UPF in two previous articles. In A Processed Mindset, first published in August 2021 – where I include a summary of the NOVA categorisation (the most widely accepted classification of food in respect of its degree of processing) – I highlighted some of the problems of the system. I argued that, although many overly processed foods should rightly be consumed with caution, classifications such as NOVA are too literal and should not be relied on, as they wrongly imply that the more a food is processed, the worse it is. I also pointed out that viewing processed foods with a negative mindset is extremely limiting and could lead to people missing out on nutritious foods. Furthermore, I stated that we need to simultaneously consider the environment due to the increasing pressure for the food system to be more sustainable, noting that food-processing technologies are integral to 21st-century eating [1].

Then, in my October 2023 piece, Unprocessing the Ultra-Processed Mindset, I ranted about the unnecessary puristic criticism of food processing. I pointed out that, despite UPF having undoubtedly contributed to many of the food-system-related environmental, ethical and health concerns, food science nevertheless provides opportunities to find solutions to these crises and offers exciting innovations that can support the long-term health of an ever-growing population. I relayed my scepticism of the flippant use of the term “ultra-processed food” and the negativity it carries with it, stating that a “perspective that’s based on a simple categorisation that unnecessarily stigmatises useful foods doesn’t work.” One of the article’s more salient points was, “Humans like to put things in neat little boxes. Here we have a box labelled ‘UPF’ with ‘bad’ rubber stamped on it. This is just not good enough. The world is not black and white; it’s highly nuanced. Nothing is simple, least of all our hugely complex food system.” I did emphasise, however, that the NOVA system is useful for exploring the epidemiology of dietary patterns and disease, but asserted that “NOVA should be confined to academia” and stated my objection to “academics and nutrition communicators pushing the term [UPF] into everyday public use as a means to assess the healthfulness and quality of food, and then criticising anything that falls under the UPF banner” [2].

My focus in this article is on the terminology. What works? What doesn’t? And what are the alternatives? Should there even be a standardised definition of the term “ultra-processed food”? In discussing this, as well as drawing on nutrition science, I’ll equally look to language. Clarity is essential for communicating ideas. When we articulate something, we need to know that our audience has understood our message. Without clarity of comprehension, how can humans collaborate effectively?

Here, not only will I have my nutritionist hat on, but I’ll also be pretending to be a terminology scientist, an etymologist and even a philosopher. I have neither a qualification nor much experience in any of these fields. I don’t care. There are plenty of faux influencers masquerading as nutritionists.

A Challenge to UPF’s Unhealthfulness

There’s no getting away from it: the epidemiology can’t be ignored. Multiple studies have demonstrated that diets habitually high in NOVA Class 4 foods – i.e. UPF – are associated with greater risk of chronic disease [3]. As discussed in Unprocessing the Ultra-Processed Mindset, when it comes to whether regular UPF consumption increases the risk of non-communicable disease (NCD), the data is compelling: there’s a clear link.

However, the results of a prospective cohort study published in The Lancet in November 2023 offer a different take on the if-it’s-UPF-it-causes-disease mindset [4]. The study looked at 266,666 participants from seven European countries. Subjects were free of cancer, cardiovascular disease and type 2 diabetes at recruitment. Foods and drinks consumed over the previous 12 months were assessed and classified according to the NOVA system. Subjects were followed up after a median of 11.2 years. The results? Aligned with previous research, higher UPF consumption was associated with an increased risk of multimorbidity of cancer and cardiometabolic diseases. No surprise there. Among UPF subgroups, associations were most notable for animal-based products and artificially and sugar-sweetened beverages. Again, no real surprise. However, results revealed that some categories of UPF breads and cereals or plant-based alternatives were not associated with risk.

Associations between subgroups of ultra-processed food consumption and risk of cancer-cardiometabolic multimorbidity [5].

These findings present a hefty challenge to the popular belief that all UPF are linked to negative health outcomes*; clearly things are way more complex. The common assumption that all UPF foods are linked to adverse health events is probably wrong. It seems that we can’t simply tar all processed foods with the same UPF brush; other factors are more important than simply how much a food is processed. While the study’s findings need to be replicated, they’re helpful in that they demonstrate that subgroup analyses of UPF are necessary when exploring the association between UPF and NCD. But maybe we shouldn’t be so quick to chastise the NOVA authors. Was their original intention to encourage a blanket categorisation of foods being viewed as “good” and “bad” when they came up with the system?

Processing Origins

Let’s start with the obvious: what does “processing” mean? Well, the English word “process” appears to have originated in the 14th century, from the Latin processus meaning “going forward”, and in the sense of “continuous and regular series of actions meant to accomplish some result” it’s been traced to the 1620s [6]. “Food processing” is the transformation of agricultural products into food, of one form of food into other forms, or “the basic preparation of foods, the alteration of a food product into another form, and preservation and packaging techniques” [7]. Food-processing techniques date back hundreds of thousands of years (for examples, see A Processed Mindset [8]), but the first food processing in the modern sense appears to have been in 1809, when Nicolas Appert invented the hermetic bottling technique to preserve food for French troops. Appert’s technologies contributed to preservation methods such as canning, invented by Peter Durand just a year later [9].

The term “ultra-processed food” is a much more recent invention. It was first used in 2009 by Carlos Monteiro [10], a professor of nutrition and public health at the University of São Paulo, who went on to lead the team who compiled the NOVA system. However, since its conception, the NOVA definition has changed considerably. In his 2019 paper, “Ultra-Processed Foods: Definitions and Policy Issues”, published in Current Developments in Nutrition, Michael Gibney** of the Institute of Food and Health, University College Dublin, explores how the definition of UPF has varied considerably [12]. He points out that there are no firm definitions of UPF and shows that different definitions for this proposed category have evolved over time, starting simply with the addition of food additives and salt [13]. This was followed by considerations of UPF on accessibility, convenience and palatability [14]. Subsequent definitions, he notes, became longer to include more elements, and only later did definitions build on these to include the role of food fortification [15], the importance of ingredients developed in a laboratory [16], a specification for the minimum number of ingredients [17], and then an emphasis on the inclusion of salt, sugars, oils and fats as a starting point for defining UPF [18]. Only later still were specific categories of food additives detailed in the definition to highlight how their intended use is to imitate sensory qualities of minimally processed foods or to specifically disguise undesirable qualities of ultra-processed foods [19].

Despite there being no firm definition from the outset, the term has steadily grown in use and has been further popularised over the last couple of years following a number of high-profile academics appearing on popular podcasts and, in the UK, by the publication of the book Ultra-Processed People: Why Do We All Eat Stuff That Isn’t Food … and Why Can’t We Stop? by Chris van Tulleken [20].

Ultra-Processed Caution

In a December 2022 paper published in the BJN, Gibney flags NOVA’s subjectivity bias in that it merely reflects the personal opinions of the authors [21]. As an example, Gibney points out that the “NOVA definition of UPF refers to… ‘additives whose function is to make the final product palatable or more appealing’”. However, he continues, palatability “is not just a function of a given food but is primarily determined by genetic, phenotypic and environmental factors”. He notes that the subjective nature of the definition also makes reference to “the ability of food additives to make a food ‘more appealing’”, and queries whether a preservative used in a bread makes it “more appealing” than a similar additive-free bread. Gibney cautions against the NOVA recommendation that all UPF foods be avoided. His key concern centres on NOVA and public health messaging: “If the degree and nature of processing of foods are to be considered as an important driver of public health nutrition, then some level of objectivity in the definition of highly processed foods is needed”. He points out that some types of bread and breakfast cereals are classed as UPF, yet they positively impact nutrient intake, and that further gains can be achieved through reformulation. In opposing the “reformulation of foods on the grounds that one cannot make an unhealthy food (subjectively defined) healthy”, NOVA ignores efforts to encourage reformulation of foods to be lower in salt, added sugars and fat. He asserts that if “the degree and nature of processing are to be considered within the strategies of public health nutrition policies, a robust, objective, evidence-based definition must be devised and the criteria for considering a food as highly processed must first take account of that food’s impact on population nutrient intake”.

A Meaningless Term

So, we have a recently developed term that’s clearly unfit for purpose. It’s commonly used to describe heavily processed foods that, if consumed regularly in large quantities, may contribute to NCD risk. It’s found its way into daily usage, not only by the lay public, but by qualified and experienced nutrition professionals, many of whom, while accepting that the term needs a standardised definition, continue to use it freely despite the fact that they should know better.

Does this mean that there needs to be a standardised definition of UPF? Some feel that we already have one: NOVA 4. However, because the classification continues to change, even NOVA can’t provide a firm definition of UPF. On the issue of if there should be a legal definition of UPF at all, I’m torn. On the one hand, it seems clear that there shouldn’t be, as the health attributes of a food depend on way more than its degree of processing. Yet, like it or not, “ultra-processed food” has found its way into common parlance, accompanied by much confusion. So, as it’s likely the term is here to stay, there probably should be an accepted definition.

In Ultra-Processed People, Chris van Tulleken feels that there should be a firm definition and praises the Monteiro team because “they drew a line” in coming up with one. He states that “their genius was to decide that a line could be drawn at all – that there is harmful food and that it can be defined” (emphasis in the original) [22]. This is despite the fact that, as highlighted by Michael Gibney, this line has been moved several times since NOVA’s conception [23]. I don’t, however, disagree with van Tulleken: robust epidemiology mandates firm definitions when exploring how food groups relate to disease risk. Crucially, however, this only works in academic research. In casual conversation, these sorts of rigid definitions lead to people being confused. Van Tulleken feels that if a product has been developed in a way that promotes its overconsumption, then it’s a UPF: “if I’m struggling with whether to call a food UPF, then it probably is UPF” [24]. On this point, in Unprocessing the Ultra-Processed Mindset I stated, “On the surface, this makes sense. However, the fact that there is a debate about whether or not some foods fall under the UPF definition makes me question the concept of pigeonholing when it comes to branding a food as unhealthy” [25]. As I stated in the article, in order to meaningfully categorise foods, such a system would have to incorporate the presence of key nutrients and the degree of processing, along with cultural, social and environmental considerations. Good luck in coming up with a user-friendly methodology that deals with all that!

If a workable system is likely not possible, should we be looking to language instead?

More Junk Food Please

Maybe we could be using a widely accepted term that already exists. Most people understand “junk food” to mean “a pre-prepared or packaged hyperpalatable convenience food that’s high in fat, sugar and/or calories, and typically low in fibre and micronutrients”. The original term is said to have been coined by Michael F. Jacobson of the Center for Science in the Public Interest in 1972 [26], although its first use has been traced back to 1952, when a headline in the Lima, Ohio, News read: “‘Junk Foods’ Cause Serious Malnutrition”. The piece covered a reprint of a 1948 article from the Ogden, Utah, Standard-Examiner, titled, “Dr. Brady’s Health Column: More Junk Than Food”. In the article, Dr Brady writes, “What Mrs. H calls ‘junk’ I call cheat food. That is anything made principally of (1) white flour and or (2) refined white sugar or syrup. For example, white bread, crackers, cake, candy, ice cream soda, chocolate malted, sundaes, sweetened carbonated beverages” [27].

So, we have a term that’s been in common use for at least 50 years. Granted, “junk food” has no universally accepted definition (although, as we’ve seen, neither does “ultra-processed food”), but this lack of a rigid definition is the very reason why “junk food” works. The term also accounts for those highly palatable foods that are high in fat, sugar and/or calories and low in fibre that don’t fall under the UPF banner, such as cakes, certain varieties of crisps (potato chips), fast-food kebabs, and fish and chips. In casual conversation, most people intuitively understand that something described as “junk” is, by and large, likely not to carry desirable healthful attributes and should only be consumed in moderation. Moreover, it excludes those nutritious, fibre-rich gems that NOVA brands negatively as UPF, such as granary bread, wholewheat breakfast cereals and baked beans.

There is much confusion over UPF and how the category relates to the healthfulness of food. “Junk food” doesn’t cause confusion in the same way that “ultra-processed food” does, and in a complex world, it can be the very imprecision of language that leads to successful communication. In considering this, I’m reminded of the quote from 20th-century Austrian philosopher, Ludwig Wittgenstein: “The limits of my language means the limits of my world” [28].

* Like all research, this study has its limitations, such as: the NOVA classification was implemented on dietary data captured more than 20 years previously; potential changes in modifiable behaviours during follow-up, especially after the diagnosis of NCD, were not possible to account for; unmeasured confounders, such as family history of NCD, could have affected the results.
** Michael Gibney sadly died in February this year [11].

Notes & References:
1. Collier, J. (2023) ‘A Processed Mindset. Does “Ultra-Processed” Mean “Unhealthy”?’, Thought For Food, 31 July. Available at: https://jamescollier.substack.com/p/a-processed-mindset (Accessed: 30 April 2024).
2. Collier, J. (2023) ‘Unprocessing the Ultra-Processed Mindset. Why “Ultra-Processed Food” Doesn’t Mean “Unhealthy”’, Thought For Food, 23 October. Available at: https://jamescollier.substack.com/p/unprocessing-the-ultra-processed (Accessed: 30 April 2024).
3. Numerous meta-analyses, for example: (a) Pagliai, G. et al. (2021) ‘Consumption of Ultra-Processed Foods and Health Status: A Systematic Review and Meta-Analysis’, British Journal of Nutrition, 125(3), 308-18; (b) Askari, M. et al. (2020) ‘Ultra-Processed Food and the Risk of Overweight and Obesity: A Systematic Review and Meta-Analysis of Observational Studies’, International Journal of Obesity, 44(10), 2080-91; (c) Lane, M. M. et al. (2021) ‘Ultraprocessed Food and Chronic Noncommunicable Diseases: A Systematic Review and Meta-Analysis of 43 Observational Studies’, Obesity Reviews, 22(3), e13146; (d) Moradi, S. et al. (2021) ‘Ultra-Processed Food Consumption and Adult Diabetes Risk: A Systematic Review and Dose–Response Meta-Analysis’, Nutrients, 13(12), 4410; (e) Suksatan, W. et al. (2021) ‘Ultra-Processed Food Consumption and Adult Mortality Risk: A Systematic Review and Dose–Response Meta-Analysis of 207,291 Participants’, Nutrients, 14(1), 174.
4. Cordova, R. et al. (2023) ‘Consumption of Ultra-Processed Foods and Risk of Multimorbidity of Cancer and Cardiometabolic Diseases: A Multinational Cohort Study’, The Lancet, 35, 100771.
5. ibid (4).
6. Online Etymology Dictionary (n.d.) Processing. Available at: https://www.etymonline.com/search?q=processing (Accessed: 30 April 2024).
7. Britannica (2024) Food Processing. Available at: https://www.britannica.com/technology/food-processing (Accessed: 30 April 2024).
8. ibid (1).
9. Cablevey (n.d.) ‘The Evolution of the Processed Foods Industry’, Cablevey Blog. Available at: https://cablevey.com/the-evolution-of-the-processed-foods-industry/ (Accessed: 30 April 2024).
10. Monteiro, C. A. (2009) ‘Nutrition and Health. The Issue Is Not Food, nor Nutrients, So Much as Processing’, Public Health Nutrition, 12(5), 729-31.
11. The Irish Times (2024) ‘Michael Gibney Obituary: Prominent Food Scientist Was a “Colossus in Nutrition” and an “Original Thinker”’, 9 March. Available at: https://www.irishtimes.com/obituaries/2024/03/09/michael-gibney-obituary-prominent-food-scientist-was-a-colossus-in-nutrition-and-an-original-thinker/ (Accessed: 30 April 2024).
12. Gibney, M. J. (2018) ‘Ultra-Processed Foods: Definitions and Policy Issues’, Current Developments in Nutrition, 3(2), nzy077.
13. ibid (10).
14. Monteiro, C. A. et al. (2010) ‘A New Classification of Foods Based on the Extent and Purpose of Their Processing’, Cadernos de Saude Publica, 26(11), 2039-49.
15. Moubarac, J. C. et al. (2014) ‘Food Classification Systems Based on Food Processing: Significance and Implications for Policies and Actions: A Systematic Literature Review and Assessment’, Current Obesity Reports, 3(2), 256-72.
16. Costa Louzada, M. L. et al. (2015) ‘Ultra-Processed Foods and the Nutritional Dietary Profile in Brazil’, Revista de Saude Publica, 49, 38.
17. Monteiro, C. A. et al. (2016) ‘NOVA. The Star Shines Bright’, World Nutrition, 7(1-3), 28-38.
18. Martínez Steele, E. et al. (2016) ‘Ultra-Processed Foods and Added Sugars in the US Diet: Evidence from a Nationally Representative Cross-Sectional Study’, BMJ Open, 6(3), e009892.
19. Moubarac, J. C. et al. (2017) ‘Consumption of Ultra-Processed Foods Predicts Diet Quality in Canada’, Appetite, 108, 512-20.
20. Van Tulleken, C. (2023) Ultra-Processed People. Why Do We All Eat Stuff That Isn’t Food… and Why Can’t We Stop? London: Cornerstone Press.
21. Gibney, M. J. (2023) ‘Ultra-Processed Foods in Public Health Nutrition: The Unanswered Questions’, British Journal of Nutrition, 129(12), 2191-4.
22. ibid (20), p51.
23. ibid (12).
24. ibid (20), pp59-60.
25. ibid (2).
26. O'Neill, B. (2006) ‘Is This What You Call Junk Food?’, BBC News, 30 November. Archived from the original on 10 October 2019. Available at: https://web.archive.org/web/20191010164339/http://news.bbc.co.uk/2/hi/uk_news/magazine/6187234.stm (Accessed: 30 April 2024).
27. Popik, B. (2008) ‘Junk Food’, The Big Apple, 26 December. Available at: https://www.barrypopik.com/new_york_city/entry/junk_food/ (Accessed: 30 April 2024).
28. Wittgenstein, L. (1921) Tractatus Logico-Philosophicus. Reprint, Milton Park: Routledge, 2001.