Unprocessing the Ultra-Processed Mindset

Why “Ultra-Processed Food” Doesn't Mean “Unhealthy”

“Ultra-Processed Food”: a term that carries tones of dread in the voices of those who utter it; a term that arouses fear when we see it written; a term that, just a few years ago, was barely on anyone’s radar; a term that’s found its way into everyday parlance. Such is the extent of the angst the term instils, you’d be forgiven for contemplating the fate of the human race if we continue ingesting these substances. Maybe we should pause and consider if the anti-ultra-processing hype is truly as existential as some make it out to be. Are we right to fear all ultra-processed foods? What’s caused the explosion in debate?

Back in August 2021, I presented a challenge to this negative perspective when I wrote A Processed Mindset [1]. In it, I commented that there didn’t appear to be a universally accepted definition of ultra-processed food (UPF). This is no longer the case. Now there is a recognised categorisation system: the NOVA classification [2]. (For a summary of the four NOVA groups, including the definition of “ultra-processed food”, please refer to A Processed Mindset.) The term “ultra-processed food” was first used in 2009 by Carlos Monteiro [3], 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. The term has steadily grown in use since and has been further popularised in the UK this year following 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 [4].

When it comes to whether UPF consumption increases the risk of non-communicable disease, the data is compelling: diets habitually high in foods that fall under NOVA Class 4 – i.e. ultra-processed food – are associated with greater risk of metabolic disease and obesity. High UPF consumption has been linked to cardiometabolic disease, stroke, type 2 diabetes, depression, irritable bowel syndrome, functional dyspepsia, cancer, frailty and all-cause mortality in adults, and associated with metabolic syndrome in adolescents and dyslipidaemia in children [5]. The epidemiology can’t be ignored. Van Tulleken covered this extensively in Ultra-Processed People [6].

Despite this, the data has been criticised. It could be that the UPF consumed in the studies was low in fibre and micronutrients and high in saturated fat, salt and sugar. As well as this, it might be that those who eat copious amounts of UPF consume less nutritious fruit, veg, cereals, pulses and seafood. Moreover, as, on the whole, UPF is relatively cheap, it could be that people who consume it in large quantities are more likely to have a lower income, which itself is a risk factor for poor health. Could UPF consumption be a proxy for poverty? As unhealthy behaviours go hand in hand, UPF consumption could signify an unhealthy lifestyle: smoking, lower levels of physical activity and/or high alcohol consumption [7]. Are all these factors merely confounding the data? Most of these issues have been considered, and the majority of the studies demonstrating an association between high UPF intake and disease risk have been adjusted for these variables. This means it’s highly likely that people whose diets include a substantial proportion of mass-produced, hyperpalatable food have a higher risk of adverse health. So, if it’s been shown that people who consume too much UPF have an increased risk of ill health even after adjusting for potential confounders, why, then, am I whinging about NOVA and the term “ultra-processed food”?

Reducing What We Eat

Is nutrition science doing an adequate job of translating the nutritional benefits of food into health policy? Many think not. They feel that food policy is both misguided and misguiding. In his 2007 best-seller, In Defense of Food, food writer Michael Pollan introduced a wide audience to “food reductionism” [8]. This concept refers to the failure to consider food as “food”; rather, we reduce it to being considered in respect of its nutrients. Pollan claimed that the value humans ascribe to food is greater than the sum of its constituent parts. The person initially responsible for highlighting food reductionism was the Australian author and sociologist of food science and politics Gyorgy Scrinis, who felt that the nutrient-led view of food was a significant health issue in itself. After coining the term “nutritionism” in a 2002 essay, “Sorry Marg”* [9], he later, in 2013, devoted an entire book to the issue. In Nutritionism he defined food reductionism as the “dominant paradigm of nutrition science and dietary advice, primarily characterized by a nutritionally reductive approach to food” [10]. Pollan summed up nutritionism by stating that “the key to understanding food is … the nutrient” and argued that foods “are essentially the sum of their nutrient parts” [11]. Scrinis pointed to trans fats, noting that they were once not only permitted but encouraged, and yet they’re now known to be hazardous to health [12]. His worry was that there may be other ingredients that, while currently permitted, may later be revealed to be similarly dangerous. Much of Scrinis’ argument targeted processed food, and in Nutritionism he attempted his own categorisation in respect of the degree of processing, which wasn’t dissimilar to NOVA, but ranked foods and ingredients separately [13].

Although cultural traditions continue to influence our food choices, in the West we’re less reliant on them. To quote Pollan: “we find ourselves as a species almost back to where we started: anxious omnivores struggling once again to find out what is wise to eat. Instead of relying on the accustomed wisdom of a cuisine, or even the wisdom of our senses, we rely on expert opinion, advertising, government food pyramids, and diet books, and we place our faith in science to sort out for us what culture once did with rather more success” [14]. Numerous techniques of food processing – some of which I discussed, along with their cultural origins, in A Processed Mindset [15] – have been around for millennia. Nutrition science, in its modern guise, however, didn’t really come into full fruition until World War Two [16]. Since then, by way of extensive research, our knowledge of how food and nutrients are linked to disease, performance and cognitive function has expanded exponentially. Despite the numerous phenomenal breakthroughs, we continue to be showered with a mishmash of cherry-picked inconsistencies, leaving the public confused, mistrustful of food policy and with little idea as to what sound dietetic advice looks like.

Prior to NOVA, another attempt to classify processed food came from Canadian biochemist Ross Hume Hall. In his 2000 book The Unofficial Guide to Smart Nutrition, Hall proposed a four-level ranking system for the types of food processing. Foods were ranked based on the degree of processing, the amount of non-nutritional chemicals added to foods and the amount of fibre present in the end food [17]. Hall’s classification has limited real-world validity. For example, Hall rated fresh fruit higher than frozen, which, in turn, he ranked higher than dried and canned, respectively. This fails to account for key factors like nutritional quality and the necessity of particular processes to distribute food. Since frozen produce is typically picked and frozen almost immediately, preserving its nutritional value, levels of key nutrients in it are often greater than in “fresh” veg, which may take several days to land in shoppers’ baskets.

NOVA set out to move away from the nutricentric perspective to instead consider food in respect of how it’s produced. Rather than viewing food as its constituent nutrients, focusing on how it’s prepared offers a new way of looking at what we eat. And this is not without merit. The value humans ascribe to the food we eat certainly is greater than the sum of its constituent parts, as demonstrated by the way food values have been integral to every human culture. Moreover, food contains beneficial non-nutrient factors, such as phytonutrients, which, while not essential for immediate survival, are known to reduce the risk of cardiovascular disease and some cancers.

Hall, Scrinis, Pollan and Monteiro share the desire to be less reliant on the nutricentric way we view food. I’m totally on board with this: the way food policy is devised and the way the majority of 21st-century Westerners consider food has been unhelpfully reductive. My gripe with NOVA, however, is that placing the food we eat into four simple groups based solely on the degree of processing is also reductive: it’s just reductive in a different way. One could argue NOVA is more simplistic than the nutricentric view.

Unprocessing Ultra-Processing

If the use of modern food processing means that deceiving our innate responses is a crime, should food science be found guilty as charged? Is it the case that nutritionists should be ignored? Should we be handing dietitians their P45s? The frustration resulting from the negative messaging surrounding food processing might lead us to ignore the phenomenal benefits granted by nutrition science. The countless advantages afforded to us by nutrition research far, far outweigh the controversies. Food science has enabled crucial discoveries that have improved the health of, literally, billions of people. Many of us are able to choose from a huge variety of foods made possible through technical methods of food preservation and distribution. Crucially, nutrition science supports our future: the learnings provided by ongoing research will be fundamental in helping to tackle the multiple health, ecological and ethical issues.

Despite the numerous advantages, there remains much puristic criticism of food processing. There are valid environmental, ethical and health concerns relating to our food system, and we must acknowledge that UPF has undoubtedly contributed to many of these, but the fact that nutrition science has already helped mitigate a multitude of food-related issues is grounds for continued optimism. Food science presents opportunities that can assist us in finding solutions to the multiple crises we face, from food security to climate change and dwindling biodiversity. With the tools of nutrition science at our fingertips, we have the ability to solve these problems, and exciting innovations will help support the long-term health of an ever-growing population.

If the knowledge gleaned from food science adds nutritional and culinary benefits and allows greater accessibility, then the advantages are vast. For sure, the majority of the foods that fall under NOVA’s UPF classification should be consumed in moderation. But the UPF definition is a sweeping generalisation, and this calls its effectiveness into question. Numerous nutritious foods are unduly vilified solely because they fall under this “ultra-processed” banner, a definition which is, after all, just the opinion of a group of researchers and their disciples. An overly simplistic system of categorisation that demonises foods is profoundly unscientific; one might even go as far as to say it’s ideological. A 2022 report in Nutrition Research Reviews was similarly critical, accusing NOVA of bias, claiming that it “classifies foods according to the assumed ‘purpose’ for which they have been designed and produced. This approach introduces a subjective (perhaps ideological) bias in the food classification process that should be, on the contrary, as independently objective as possible.” The authors contended that the classification “suffers from a lack of biological plausibility so the assertion that ultra-processed foods are intrinsically unhealthful is largely unproven” [18].

In Ultra-Processed People, Chris van Tulleken conceded that some foods fall at the margins of the UPF definition. He used canned baked beans in tomato sauce – a British staple – as a useful example, pointing out that they’re a healthy, affordable food, and that some varieties may fall into NOVA class 3 and others into NOVA 4 due to the ingredients. He admitted that, in this instance, “we meet the limitations of NOVA, a system designed to look at dietary patterns rather than to evaluate individual foods”, and he continued: “There is almost certainly a spectrum of UPF, yet exactly how or whether any one product will be harmful is impossible to tell because we don’t just eat one food – we eat a range of foods” [19]. Advocates of NOVA feel that, as well as the list of ingredients, other factors are integral to the UPF definition, including how a food is labelled and marketed and its hyperpalatability. Van Tulleken asserted: “if I’m struggling with whether to call a food UPF, then it probably is UPF”. He felt that if a product has been developed in a way that promotes its overconsumption, then it’s a UPF [20]. 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. What about foods that fall under the UPF banner whose flavour profile doesn’t encourage overconsumption? What about products containing one or two so-called UPF ingredients that are high in fibre and protein, rich in key micronutrients, and low in sugar, salt and calories? What about foods that are helping to combat the increasing environmental and ethical pressures on the global food system?

Confine NOVA to Academia

I’m highly sceptical of the flippant use of the term “ultra-processed food” and the emotion it carries with it. Humanity faces immense challenges: our future is dependent on the advances of nutrition science. We have more than 8 billion (and rising) mouths to feed, food-system-derived greenhouse gas emissions to reduce, food waste to mitigate, and inhumane and unsustainable agricultural practices to address, while simultaneously allowing our fast-moving world to continue at its ever-rapid pace. We desire access to an array of foods grown far from where we live, provided safely, in good condition and with a long shelf life. A perspective that’s based on a simple categorisation that unnecessarily stigmatises useful foods doesn’t work. 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.

To be clear, the concerns regarding a large number of commonly eaten UPFs must not be dismissed. Quite the contrary: many UPFs should be consumed with caution. Nor am I wholly dismissive of NOVA and the term “ultra-processed”. NOVA has a clear usefulness for exploring the epidemiology of dietary patterns and disease, in the same way that we look at, say, the Mediterranean diet. But what I object to is academics and nutrition communicators pushing the term 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. NOVA should be confined to academia.

The three principal concerns I raised in A Processed Mindset are:

1. Classifications such as NOVA are too literal and should not be relied upon.

2. Such classifications wrongly imply that the more a food is processed, the “worse” it is.

3. Viewing processed foods with a negative mindset is extremely limiting and will, likely, lead you to unnecessarily miss out on some very nutritious foods. [21]

With the anti-ultra-processing dogma becoming more widespread, my concerns have intensified. The heightened pressure for a more sustainable food system tells us that food processing technologies are integral to 21st-century eating. Innovative and cutting-edge practices enable us to produce delicious and nutritious food and drink at scale and at affordable prices. Processing techniques allow us to buy raw ingredients from farmers and make safe food via distribution channels to remote regions, and to reduce food waste. Processing supports specific dietary requirements, such as plant-based proteins for vegans or safe products for those with specific food intolerances. Importantly, processing has a role in helping people make healthier food choices. Manufacturers have tweaked recipes to add more vegetables to sauces, reduce sugar content and fortify foods with specific micronutrients or fibre. The ability to process food is hugely valuable. This extends to the highly questionable intensive livestock agriculture techniques. If meat can be kept fresher for longer, less is wasted. Exploiting an animal only to throw away its nutrition-rich carcass seems both practically and morally irrational.

Although, as discussed, the data linking high UPF consumption and poor health is compelling, there nevertheless remains a risk of bias, even when controlling for confounding factors. Those testing the link between high UPF consumption and metabolic disease will likely have their hypotheses validated because the majority of the UPF tested will be low in fibre, protein, essential fats and micronutrients, and high in sugar, salt and saturates. Despite attempting to adjust for these variables, there’s no getting away from the fact that the UPF consumed by subjects in these trials will likely have a poorer nutritional profile and will, therefore, be correlated with disease risk, irrespective of how the foods are processed. Of course, the hyperpalatability and energy density of high-UPF diets may mean more is consumed, but calorie for calorie, this research bias is hard to fully mitigate.

The fact that the way in which a food is marketed is part of the UPF classification seems silly. While we should be questioning the rationale behind some nutrition and health claims, especially as they’re used to mask unfavourable attributes of some products, I can’t get my head around why the mere presence of a marketing claim should be an indication that a food is a UPF and, consequently, unhealthy. Weirdly, there seems to be a blanket demonisation of breakfast cereals by NOVA proponents, even though many cereals have nutritional qualities. Take Nestlé Shredded Wheat – a 100 percent wholewheat product that’s a nutritious inclusion to any breakfast, particularly because it’s consumed with milk – which wouldn’t be categorised as NOVA 4 by even the most ardent NOVA aficionado. Should Shredded Wheat not be marketed using a health claim? Surely, doing so helps to promote these nutritious gems over sugar-laden alternatives.

It’s easy to forget that many of us are fortunate enough to be in the position to criticise UPF. The ready access to a vast choice of food that the West has been enjoying for decades is what many in lower-income nations are now justifiably desiring. A fast-paced world where people don’t have the time – or the will – to prepare meals is one in which they rely on convenience. A dynamic society means we can enjoy many comforts. Access to convenient sustenance allows us to work and spend quality time with our friends and families, and allows society to develop innovations that benefit others and help to improve our future. Even within higher-GDP countries, like the UK and US, higher UPF intake has been associated with those on a lower income [22]. The anti-UPF fearmongering serves to worsen how those who can afford little in the way of minimally processed produce feel about their situation.

Now, more than ever, we must rely on food technology to provide sustainable nutrition. In the same way as our early agrarian forebears granted the benefits of selectively breeding for sustenance, we look to food technologies, including ultra-processing. But rather than a blanket slating of UPF, let’s ensure that the way in which we process a food enhances its nutritional quality, makes it more accessible and sustainable, or aids convenience. Whole ingredients may undergo processing to prolong their shelf life, boost their organoleptic qualities and aid their healthfulness with the addition of certain nutrients. At the same time, we could apply caution to processing techniques that appeal to consumers’ hedonistic urges and contribute to overconsumption. While we’re dependent on food technology in order to address the multiple challenges, we can ensure the right food processing methods are used in the right way for the right reasons.

The Nutritionally Complete Elephant in the Room

I should address the elephant in the room: I’m co-founder of Huel and the nutritionist behind the products. Huel develop and produce nutritionally complete, affordable food with minimal impact on animals and the environment. Under NOVA, Huel products fall under Group 4. It is undeniably in my own interest to challenge a perspective that could be seen to vilify products that I’m involved with. Aside from the fact that I’ve a financial interest, I’m emotionally attached, too. If I’m not directly behind a nutrition decision, I’ve been involved in some way because I personally approve the nutrition of all Huel products. Thus, even if I were to no longer work for Huel, it’s impossible for me to avoid this bias. All I can do is be transparent and strive to keep this front of mind as I communicate my ideas. However, while I must remain sincere about this and other conflicts of interest, any reasonable objection to my ideas should challenge the nature of my argument and the evidence I present, rather than focusing on my biases. We all have biases, and, crucially, they need not be financially motivated. Proponents of NOVA, for example, are equally biased, as they’re committed to it as an idea. All we can do is be aware of our biases, try to mitigate them (knowing that frequently we’ll fail) and, most importantly, be transparent about them.

Broader Perspectives

The crux of my concern is not to challenge the data linking a high consumption of foods that fall under the NOVA UPF definition to disease risk, but to challenge the definition itself. I’m uneasy with the idea that any food falling under NOVA Group 4 carries with it implicit negativity. Under the NOVA categorisation, a food that contains a single so-called UPF ingredient but is otherwise nutrient-rich and packed with quality ingredients – such as fibrous cereals, fruit, nuts, seeds or pulses – becomes tarred with the same brush as foods that are high in sugar and lacking in key nutrients, and that doesn’t sit right with me.

Society is intent on categorising things, and this includes the food we eat. Clearly, neither the nutricentric nor the NOVA perspective works. Maybe, instead, we need a system that encompasses the multiple attributes of food: a strategy that incorporates key nutrients – the sugar content, the amount of protein, the presence of essential fats, its fibre – soluble and insoluble – content, the quantity of salt, and the levels of vitamins, minerals and phytonutrients – along with how and to what degree it’s been processed. A system should also account for cultural, social and environmental considerations. Granted, that’s a lot to examine, and a public-friendly methodology may not even be possible. It might be that the complexity of the food we eat, how we eat it and the value we ascribe to it makes any attempt to categorise our sustenance an impossible task. Nevertheless, something like my proposal seems to be the only way we could move away from both reductive and inadequate perspectives. In any case, food technology – including ultra-processing – is integral to a 21st-century food system. In his book Seeds of Science, the historian Mark Lynas pointed out that “if we resisted anything but natural farming technology (circa 1960), we would need a plot of land the size of two South Americas to feed the planet” [23].

* Scrinis was referring to margarine, which he views as the ultimate nutritionism product.

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: 19 October 2023).
2. Monteiro, C. A. et al. (2016) ‘NOVA. The Star Shines Bright’, World Nutrition, 7(1-3), 28-38.
3. 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.
4. 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.
5. 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, 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, 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. 
6. ibid (4), pp52-67.
7. ibid (4), pp59-60.
8. Pollan, M. (2008) In Defence of Food: The Myth of Nutrition and the Pleasures of Eating, London: Penguin, p28.
9. Scrinis, G. (2002) ‘Sorry, Marge’, Meanjin, 61(4), 108-16.
10. Scrinis, G. (2013) Nutritionism: The Science and Politics of Dietary Advice, New York: Columbia University Press, p258.
11. ibid (8), p28.
12. ibid (10), pp144-9.
13. ibid (10 ), pp218-9.
14. Pollan, M. (2006) The Omnivore’s Dilemma: A Natural History of Four Meals, London: Penguin, p303.
15. ibid (1).
16. Hwalla, N. and Koleilat, M. (2004) ‘Dietetic Practice: The Past, Present and Future’, Eastern Mediterranean Health Journal, 10(6), 716-30.
17. Hall, R. H. (2000) The Unofficial Guide to Smart Nutrition, New York: IDG Books.
18. Visioli, F. et al. (2022) ‘The Ultra-Processed Foods Hypothesis: A Product Processed Well Beyond the Basic Ingredients in the Package’, Nutrition Research Reviews, 1-11. Available at: https://doi.org/10.1017/s0954422422000117 (Accessed: 19 October 2023).
19. ibid (4), pp157-9.
20. ibid (4), p160.
21. ibid (1).
22. (a) Baraldi, L. G. et al. (2018) ‘Consumption of Ultra-Processed Foods and Associated Sociodemographic Factors in the USA Between 2007 and 2012: Evidence from a Nationally Representative Cross-Sectional Study’, BMJ Open, 8, e020574; (b) Juul, F. et al. (2021) ‘Current Intake of Ultra-Processed Foods in the U.S. Adult Population According to Education-Level and Income’, Current Developments in Nutrition, 5(S2), 418; (c) Leung, C. W. et al. (2022) ‘Food Insecurity and Ultra-Processed Food Consumption: The Modifying Role of Participation in the Supplemental Nutrition Assistance Program (SNAP)’, American Journal of Clinical Nutrition, 116(1), 197-205.
23. Lynas (2018), cited in McIntyre, L. (2021) How to Talk to a Science Denier: Conversations with Flat Earthers, Climate Deniers, and Others Who Defy Reason, Cambridge, MA: MIT Press, p132.