This post was written in response to a comment on this blog in response to a comment on this blog.
Thanks for commenting. I am aware that you submitted a comment using a real name and email address that let me easily identify you. I have anonymised your comment in case a troll is impersonating you.
Whether really from a professor of nutrition, or not, your comment was from someone with some experience in this area, I think, and so I felt it worth responding to the issues raised.
Let me begin by noting that you are an academic in a field where funding from industry or industry-front organisations is commonplace. I note no mention of this on your website, either for you or your department. Might you please disclose all such sources of funding? I think it would be appropriate for you to do so. For the record, I have none relating to this issue.
I then note your comments, and will respond to them after reproducing your commends to contextualise my responses. First, you said:
The main problem is that there is no consensus that UPF as a category has a meaningful impact on health. All data comparing UPF with processed foods are from observational studies which rely on data that does usually not allow to identify UPF reliably. This is one reason why neither the UK nor the US nor the Nordic countries advise against UPF consumption.
I think your claim is misleading. Foods have been characterised using a NOVA scale, as follows:
- Group 1 - Unprocessed or minimally processed foods
- Group 2 - Processed culinary ingredients
- Group 3 - Processed foods
- Group 4 - Ultra-processed food and drink products
You claim that:
The difficulty in estimating intake means that even for supermarkets, it would be virtually impossible to distinguish between NOVA 3 and NOVA 4. Would – for example – chicory root extract make a good UPF? Would high speed mixing make bread UPF?
This, I suggest, is disingenuous. You seek spurious accuracy as an excuse for inaction, and yet you must know three things.
The first is that all data has a selection bias. The process of selecting what data to collect is based on ill-informed judgement in all cases. The data actually collected usually confirms the selection bias. To pretend that the data we have is insufficient because it lacks sufficient granularity is, in that case, absurd. All data is incomplete. All of it has a measurement problem at boundaries (which is where we usually measure it). Your first claim is then scientifically spurious. All supposed scientific data is subject to the classification problems you note. Are you suggesting we stop using data as a consequence?
Second, as I have already noted, decision making will either a) always be judgemental, or b) be judgemental, but be potentially informed by data. And let me be clear, even when data is supposedly used, judgement is always paramount. If you are in doubt, look at the decisions of the Bank of England Monetary Policy Committee. They collect vast amounts of data. They apply contested theory to it. They use a criteria for public policy determination which is simple in the extreme. They change lives by doing so. And how are the decisions made? On gut feelings, at best. This is how public policy decision making works. Prevailing narratives support the decisions made, whether right or wrong.
Third, you show no awareness of this when making this comment:
Regarding the intake of children: due to the difficulties of estimating UPF intake, UK bread is usually assumed to be always UPF – so a homemade sandwich as packed lunch will always be classed as UPF. Likewise fish fingers, even if they are generally considered to be a healthy and sustainable source of fish.
Again, I suggest this is disingenuous. You apparently extrapolate from single micro-situations (a sandwich, a fish finger) to entire population data, suggesting that data on the latter is invalid because of a single micro-data exception. Are you being serious? I have no idea, because you go on to say:
There is a wealth of data for nutrient composition and its effect on health. Why should we change this approach to something popular but without much evidence?
This comment directly contradicts the paragraph that preceded it, which I have noted, and the one before that, also noted. The data you now claim we should use is apparently, also according to you, invalidated by classification problems, and by micro-data outliers meaning we cannot use it, and yet you want to. Your argument quite literally makes no sense because it is riddled with contradictions.
So, let's talk reality. Easy to follow ways of determining which foods are NOVA 3 and 4 are readily available (and you must be aware of them)and the average person with none of your training would find them easy to follow. This is exactly the type of heuristic almost invariably used for public policy making purposes. Why do you have a problem with this?
And as to why should we act, please explain why it is that we should not given that since those foods that contain high quantities of sugar, and most especially fructose, plus salt, fats and extensive chemically created artificial ingredients intended to both extend life and induce consumption via addiction, became commonplace forty years ago:
- Obesity has risen?
- Type 2 diabetes has risen?
- Metabolic diseases have risen?
- All conditions related to glucose induced inflammation have risen?
Are you suggesting:
- This has not happened?
- There is no connection with UPF consumption?
- There is another cause, in which case please suggest what it is?
Might you please also explain why , if UPF is not an issue, then:
- The food industry is the biggest employer of scientists in the UK, because processed food does not require them?
- The sugar lobby is so powerful?
- Scientific papers funded by the sugar lobby are ten times more likely to find no problem with sugar in diets than do those not funded by the sugar lobby?
I look forward to your responses.
Thanks for reading this post.
You can share this post on social media of your choice by clicking these icons:
There are links to this blog's glossary in the above post that explain technical terms used in it. Follow them for more explanations.
You can subscribe to this blog's daily email here.
And if you would like to support this blog you can, here:
Buy me a coffee!
Thank you for taking the time to reply. Please do feel free to publish my name – all information you need to know about me are easily available on my website, including funding.
However, I believe there are some misunderstandings:
I do not dispute that NOVA is being used, I do dispute that there are reliable data on health outcomes. Even the umbrella review published in BMJ does not find very strong evidence for most outcomes. But more importantly – almost none of the observational studies really do measure UPF intake, because FFQs, the instruments most commonly used, are unable to do so (which is usually even acknowledged by the authors).
Your comment that this is “spurious” suggests that you have not worked a lot with nutritional data: if we do not know what people consume, we don’t know the impact on health (we have published on this – albeit in a different context, but the problem remains the same – https://elifesciences.org/articles/92941).
I have to admit I do not understand your comment about sandwiches – bread is the main source of UPF in the UK – it is one of the main reason UPF intake appears to be so high on a population level.
Why does referring to food composition data contradict my statement? Food composition data can be measured (with known problems) and it can – and is – used to make recommendations. It is inherently different from NOVA as the source e.g. of sugar is irrelevant. Under NOVA, a food with the self-same composition can be NOVA 3 or NOVA 4, depending on whether it has been made for profit or not.
You state: “So, let’s talk reality. Easy to follow ways of determining which foods are NOVA 3 and 4 are readily available (and you must be aware of them)and the average person with none of your training would find them easy to follow.” And this is simply wrong: it is impossible to distinguish between NOVA 3 and 4 exclusively on ingredients, as purpose of production is also part of the criteria. There are also many aspects of food production where it is not clear whether they affect the classification (e.g. high speed mixing).
You dismiss my claim about chicory root – but would chicory root extract in a yoghurt (often used as prebiotic) render a yoghurt NOVA 4 or not?
Obesity and Type 2 diabetes has risen, but I do not think intake of UPF has anything to do with it except for food composition and texture: UPF is cheap and affordable way to consume HFSS – but it is the composition, not the processing.
“The food industry is the biggest employer of scientists in the UK, because processed food does not require them?”
Do you have data on this? But food processing is fairly important and research intensive – why should they not hire scientists.
Thank you for your comment, and thank you too for letting me identify you.
For the sake of those who have not read other comments in this thread, you suggest you are Professor Gunter Kuhnle of Reading University. https://www.reading.ac.uk/food/our-staff/gunter-kuhnle You work at the High Sinclair Nutrition unit. Despite your claim, I can see no evidence that this fully discloses its sources of funding, and provide no detail of its commercial funding at all. Might please now provide that? https://www.reading.ac.uk/food/our-facilities/hugh-sinclair-unit
I have already posted some replies to your later comments on this blog, simply because they were easier to deal with than this one and I have had other things to do today. However, I will now address these comments. I do so whilst noting that it is incredibly difficult to work out what your arguments are. It would really help if you could actually explain yourself better than you do, because opacity appears to be your stock in trade.
I also repeat that I would like you to answer the questions I posted in my original piece, which I do not think you have done .
What you have instead, I think, suggested is the processing is not a problem with regard to food, and at a very simple level (and that is the level that you appear to be working at) that is obviously true. We do, of course, have to process most of our food stuffs to eat them. I think we can agree on that. But, to suggest that the manufacturing processes of industry, which substantially change the material input into those food stuffs, and the likely nature of the food produced, has no impact on health is as absurd as the claims that were once made by the tobacco industry that said that smoking had no consequence for health. You simply cannot input substantially more sugar, and especially fructose, into the food production process and not have a consequence. Nor, I suggest, can you be indifferent to the nature of the fats in products, although you provide no indication that you have any such concern although as a matter of fact, it is now very obvious that some fats are decidedly beneficial to health, despite all the claims made by the sugar lobby over many years, and the number that are harmful is much lower than was previously thought. Your reference to a generic HFSS food (high in fat, sugar and salt) as if they are the problem makes no sense in that context, although it might have a significantly higher observational quality than your comment that sugar content is of no consequence.
I reiterate my points. You have not explained your funding. You have not explained your methodology. You have not explained what you think the problem that we face in public health cover I g obesity, etc, are caused by. You have not explained why you are indifferent to the excess consumption of sugars and other likely to be addictive components of ultra processed foods. You have not shown any clear understanding of the management of data sets, and have a relied on pedantry and the use of outliers to claim the data cannot be proved as a basis for justifying changes in public policy, which is a well-known technique used in the past by both the tobacco and carbon industries. You have not, in fact, presented an argument.
You obviously came here with a purpose. Might you now address these issues?
If I post on a public blog, I would expect to be identified.
My website includes all details about my funding. It is easy to find and to read. I’m not sure what else is needed.
It is quite simple: I do believe UPF is a con to create money for some influencers – and it makes it much more difficult to address the actual problems in nutrition. I’ve outlined the reasons: the data are incredibly weak, publishers use very questionable methods to promote studies (the BMJ ‘umbrella’ review was not properly peer reviewed for example) and the narrative is quite appealing, but that doesn’t make it right.
I think public health should focus on facts – even if they are difficult and counter-intuitive.
Of course, processes can impact health – in both directions (e.g. by making nutrients more available). But one should look at individual processes and then modify, change or ban them. Combining all processes does not help anyone.
Then why not focus on sugar? Why not reduce soft-drink intake? Why include foods with effect on health whatsoever?
Elsewhere, you have complained about me being pedantic regarding the assessment of intake – yet it is exactly this type of being pedantic that provided much better data on the association between fat and health (regarding the alleged conspiracy of the sugar industry, I suggest a number of papers that question this – it appears to be much more difficult). Some saturated fatty acids – e.g. C:16 – are clearly associated with health risks, others, e.g. odd-chain fatty acids as found in dairy, are not. But it is really only the technology of the last decade, that has allowed us to understand this much better, because we can measure the actual fat intake.
I cannot recall having said this.
Hang on -after all that sugar funding you don’t know why it is sugar is causig a problem in our diets? Is that what you want us to believe? and you have the nerve to suggest those who disagree with you are doing so to profit when your university does all it can to prevent disclosure of the extent to which your work is funded by big sugar?
I think we need to close this. I don’t engage with those who peddle hypocsisy and straightforward misinformation on this scale. Answer my questions on funding and its influence on your work – very precisely – or we will know exactly what you are about – which is finding ways to maximise corprpate profits, whetever the cost for real people.
I am not responsible for the FOI policy of my employer – neither am I responsible for the contents of the website; this is why I direct you to my own website for which I am responsible.
But you might have missed that I was one of the main authors of studies showing the adverse effects of sugar on health.
Your personal website does not refer to Mars, or PepsiCo.
After a v quick search on -“professor of nutrition” NOVA UPF – I found this link, which has the merit of including declarations of interest.
https://www.sciencemediacentre.org/expert-reaction-to-presentation-on-ultra-processed-foods-global-diets-and-policy-implications-press-released-from-and-being-presented-at-the-international-congress-on-obesity-ico-2024/
The arguments against a focus on UPF as a category for informing public policy, do sound to me, v much like special pleading – a version of “until we know everything, perfectly, we musn’t challenge the status quo, even when the status quo is intolerable and unsustainable (but making a lot of money for a few, at the expense of the many)”.
The references to tobacco as a cautionary tale are ingenuous, as for decades the tobacco industry said the same as “the professor of nutrition” about a lack of causality with regard to negative health outcomes (death) from smoking, and their VAPE successors are doing the same now.
The cost of waiting for “certainty” will be astronomically high for very many millions of ordinary people, and the public services they depend on. The tsunami of obesity, heart disease, arthritis, & type 2 diabetes is already overwheming our health defences.
The cost of action can be borne by the broad shoulders of a few food industry shareholders.
I agree with you.
A little more background here about research organisations on the relationship between food and health, and their sponsorship links
For example, the “Hugh Sinclair Unit of Human Nutrition (HSUHN)”
https://www.reading.ac.uk/hsuhn/
“We collaborate with centres of excellence throughout the world and have strong links with the food industry.”
but the details (of the commercial sponsors at any rate) are hard to come by…
https://www.reading.ac.uk/hsuhn/research
“Research sponsors include research councils, including Biotechnology and Biological Sciences Research Council (BBSRC), Medical Research Council (MRC), European Union; charities including Wellcome Trust, British Heart Foundation, Diabetes UK, and a wide range of industrial partners. Our collaborators include the University of Reading’s centres of excellence: Institute for Cardiovascular and Metabolic Research (ICMR) and Centre for Integrated Neurosciences and Neurodynamics (CINN).”
The industrial sponsors seem very shy about divulging their identities.
Maybe it’s hidden deep in the university accounts?
A little more here from the declarations in the paper I originally referred to in my first post:
https://duckduckgo.com/?t=lm&q=BBSRC+TUKFS&ia=web
which also mentions Mars Inc.
https://en.wikipedia.org/wiki/Mars_Inc.
(that well known health food purveyor…)
and UKRI:
https://www.ukri.org/opportunity/transforming-uk-food-systems-for-health-and-environment/
Plenty there to dig around in.
RobertJ
Given I tried to hide my commentator’s identity I astonished that your ability to identify the organisation at which he works. I accept that might just be chance.
The commentator is https://www.reading.ac.uk/food/our-staff/gunter-kuhnle who is on the staff at the unit that you identify. There is a conflict as to his title: the High Sinclair Unit describes him as an associate professor, but the university says he is a full professor, and that I will accept.
I think your comments entirely appropriate.
I note that the sources of funding are entirely opaque when it comes to the industrial partners and that significantly worries me. I think that unacceptable.
I will be raising that issue with Professor Kuhnle
Richard
Wouldn’t it be better to focus on content? In contrast to many of my critics, I am completely transparent. I do not receive findings from NGOs (which often don’t have to be declared) or media corporations.
The comparison with tobacco is appealing but very misleading for a very simple reason: we don’t need tobacco, but we need food – and without the food industry, we would not be able to enjoy the food safety and security we have (and food would be much more expensive). The original and main purpose of food processing has been to make food safe and prevent spoilage – and this also reduces cost.
The main question however is: why should we introduce a concept for which there are very limited data and that has been developed in a completely different context (changes in the food system in Brazil) – when we have a system that we know works (food composition). People who reject HFSS in favour of NOVA probably cause much more damage as they stop known and established methods in favour of completely untested ones.
Your funding is opaque in the extreme, as far as I can see.
If I am the critic you refer to, you can find data on it all quite easily.
Your second paragraph is absurd: when addiction is created (and it us to UPF) need is nit an issue. Why pretend otherwise? This sounds like an excuse for the food industry.
And let’s be clear, you are very good at criticising, but so far you have yet to make a single constructive observation about what you think the cause of the problems that we face, and what you would do about them. If this is to progress further, please do so now. Otherwise, I will quite reasonably conclude that you are here for one reason only, which is to justify the continued production of addictive foods that are causing immense problems for health in our society.
Where is it opaque? What information is missing? I’m happy to provide any clarification.
You are not.
The idea of “UPF” addiction is promoted by a small group of people – most of which appear to have a very clear financial interest in doing so (books, consultations and even ‘retreats’). Do you also take this into consideration?
The discussion on UPF diverts essential resources from research that will actually help to reduce obesity and diet-related NCDs. UKRI TUKFS (here is an interest – I’m part of one of those projects) does a lot of good work to find ways to improve the food system, without having to resort to UPF. There is no evidence that UPF is in any way superior to anything we have at the moment.
Please answer the oiunhts I raised.
Why won’t ypou do so. I asked:
And as to why should we act, please explain why it is that we should not given that since those foods that contain high quantities of sugar, and most especially fructose, plus salt, fats and extensive chemically created artificial ingredients intended to both extend life and induce consumption via addiction, became commonplace forty years ago:
– Obesity has risen?
– Type 2 diabetes has risen?
– Metabolic diseases have risen?
– All conditions related to glucose induced inflammation have risen?
Are you suggesting:
– This has not happened?
– There is no connection with UPF consumption?
– There is another cause, in which case please suggest what it is?
Might you please also explain why , if UPF is not an issue, then:
– The food industry is the biggest employer of scientists in the UK, because processed food does not require them?
– The sugar lobby is so powerful?
– Scientific papers funded by the sugar lobby are ten times more likely to find no problem with sugar in diets than do those not funded by the sugar lobby?
Please answer those questions. Is that really so hard?
@ Gunter Kuhnle
About opacity & transparency, could you name the commercial /industrial organisations that are referred to, but NOT named in
https://www.reading.ac.uk/hsuhn/
&
https://www.reading.ac.uk/hsuhn/research ?
If you decline to do so, please could you explain the lack of transparency and how the general public benefit from such secrecy?
Where a 3rd party requires a declaration of interest, as in
https://www.sciencemediacentre.org/expert-reaction-to-presentation-on-ultra-processed-foods-global-diets-and-policy-implications-press-released-from-and-being-presented-at-the-international-congress-on-obesity-ico-2024/
the UPF giant Mars Inc. was named. Again, perhaps you could be transparent about their specific involvement with the research? (Declaration – I used to eat a lot of Mars bars as a schoolboy.)
See my pther comments Robert
Reading has form in refusing to disclose data on this issue requested under FoI legisaltion. I wonder why? Apparently it would prejudice their commercial interests. I am sure it would.
In order to answer your question probably, could you please explain what extensive chemically created artificial ingredients are?
Many factors have changed in the past decades – why should it be ultra-processed foods and not e.g. a more sedentary lifestyle or simply easier availability of foods?
No
Independent of food composition? Yes. UPF is simply a marker of HFSS.
Can you please explain a) where you got the numbers from and b) why food processing does not need scientists?
How should I know?
Let’s answer your last question first.
Freedom of Information requests make clear you are primarily funded by Mars, Pepsico and Roquette Freres SA. You know all avout sugar funded papers then. Why is this not made clear on any of your websites? What have you to hide? How do they ninfluyence your research. Prciose answers are required please. Why not share the conditions of the funding?
Your other answers are trite, mere speculation and not worth commenting on. Is that really the best you can do? Are you not woprried abouit your reputation?
They say no such thing for two reasons. 1) they refer to the research group, not to me as a person and 2) they exclude any non-commercial funding, e.g. by UKRI.
Why would I? I still don’t understand why I should.
Have you looked at my website? It provides a detailed summary of my funding.
I struggle to understand: I think I have achieved one thing though – you have considered whether FFQs (or 24h recalls) are really reliably instruments for dietary research.
Your website makes no mention of funding as far as I can see – and I opened every page.
Why do none of your claims stack?
@RobertJ
I am not responsible for the content of the website or for the research of my colleagues – likewise, I will not have all information you request. You can – very easily – find all information about me on my website and I am more than happy to answer questions based on those.
Of course – I have a long-standing research collaboration to investigate the effects of flavanols on health.
I have named three companies not named there. Why aren’t they named? You claimed total transparency. It is glaringly obvious that your university fights that. Why? You are on its Senate, I note. You should know.
“a version of “until we know everything, perfectly, we musn’t challenge the status quo”
Exactly. Delay is the new denial. The fossil fuel industry delayed meaningful action for over 30 years after their scientists informed them of the CO2 problem and its effects on the climate. They hushed up their own research and then began a highly-successful rearguard action by claiming there was no certainty that CO2 emissions were destabilising the climate. This approach exactly copied the playbook of the tobacco manufacturers in the 1960s and 1970s and we see it now being used by the food manufacturers. And it’s always with the same sole motive – to protect their profits.
Agreed
Apologies for a late post on a very interesting post, but here are a couple of links showing how children’s health in the UK has been declining over a number of years:
https://foodfoundation.org.uk/sites/default/files/2024-06/TFF_Children%27s%20Health%20Report.pdf
https://www.ft.com/content/b5035c40-a6ed-4a88-8b8e-1e245f1473e9
Big Food companies employ scientists with the sole objective ensuring that they remain competitive by reformulating and changing the production technologies to reduce raw material costs and to improve the palatability.
A very brief check using ChatGTP shows that Nestle is now the world’s largest consumer products food company, with Pepsico 2nd and Unilever 3rd. Between then they declare employing at least 11,700 scientist and technologists working in this highly competitive industry. Everyone of these employees is excited by their science and technology innovations but are hobbled by their contracts of employment and their needs to provide income for their families. The rarely publish their findings and so the unintended consequences of their work often becomes common knowledge too late for regulators to take affective action and their employers are happy to continue selling the fruits of their work for as long as possible in order to maximise profitability.
Every mouthful of food from cradle to grave is an opportunity for these companies to persuade us that the food tastes so good that we will consume more than is necessary. These products are called Hyper Palatable Foods – HPF – and are designed to be addictive.
Whether we call them UPF or HPF, there is so much evidence to suggest that, as these types of food have become cheap and common place, dietary intake has changed and/or increased and the health of the population has deteriorated.
Recent findings suggest that as Big Tabacco bought into the food industry, it used it’s research clout to innovate HPF’s – suggesting a link between knowing about addictive behaviour and the rise of the number of these food products.
Here are some more links:
https://lsa.umich.edu/psych/news-events/all-news/faculty-news/many-of-today-s-unhealthy-foods-were-brought-to-you-by-big-tobac.html
https://foodfoundation.org.uk/sites/default/files/2024-06/TFF_Children%27s%20Health%20Report.pdf
https://youtu.be/E0UWKV6jJ3k
And – a declaration of interest – a long time ago, I worked in research in the food industry and then in increasingly senior positions for 20 years.
Finally, Richard and other commentators are right to be sceptical of dissenting comments from academics and researchers. They always find it difficult to move from an entrenched position in the face of a vocal opposition….irrespective of the source of their salaries.
Thank you
‘Thoroughly agree with RobertJ on this.
All I see here is the same sort of ‘delaying’ behaviour we saw in the European automotive industry about emissions – contest, look for any flaw in the argument, plead for fairness, delay, delay, delay and hope that it just goes away. I think that the tobacco industry led the way on this sort of spoiling behaviour.
In terms of the behaviour of the food producers, having spent time refining their production processes, changing all this will be costly so of course they will resist. But they should never have been allowed to do this in the first place, but short term profit has come before long term health.
It is another example of how the private sector maximises its opportunities for profit whilst leaving society to pick up the hidden costs of those decisions. It is simply not on.
And it is yet another example of the transfer of risk onto others. This transfer of risk ‘ethos’ is a sickness in higher management through many sectors of the economy – even the public sector. It is unethical for start because even though the Neo-libs like to tell you what a marvellous knowledge processor the market is, we also know about asymmetries of information between producers, middlemen and buyers are legion and it is ‘caveat emptor’ apparently with all the blame heaped on the end user for not knowing enough (which is also wrong when one looks are the money and psycho-babble expended on marketing).
Further, the fact that food science exists at all, is used as badge of ‘credibility’, the message being ‘we know what we are doing, leave us alone’ yet most of the science is going into extending shelf and delivery lives and getting people hooked on the product – again, the shadow of the tobacco industry looms large.
Much to agree with
Declaration of (no) interest:
I have no expertise or links to any of this except some university training in animal, particularly ruminant (not human) nutrition many decades ago. Ruminants have a completely different carbohydrate metabolism, based around the “Krebs cycle” so I’m not to be relied on for advice in that area unless you are a cow!
I had no knowledge of the professor of nutrition before reading today”s blog.
As for identifying anyone, DuckDuckGo (never Google) was my friend, and the skill is choosing the search parameters from the original blog post, and the luck was the relevant links chosen by the algorithm being near the top. I also decided that a 2024 date was a big clue.
At least it wasn’t Gillian McKeith!!
Thanks
That would of course be true if there wasn’t an alternative. We know very well that overconsumption of energy results in weight gain – and we know the adverse effects of excessive intakes of sugar and fat. There is no need to introduce a classification system that is highly ambiguous and has never been shown to have a benefit.
To use the climate change analogy: NOVA is a bit like someone claiming that not CO2 itself, but how it was generated is the issue, i.e. a domestic oil heating is fine, but an industrial one is not.
Politely, this does not seem like a comment made by a professor. I am still wondering whether you are a troll.
Using the generic terms sugar and fat in the way that you do is, as you should know, of little help. There are extremely beneficial fats as well as harmful ones. Likewise, not all sugar is harmful, but fructose definitely is in the quantities now consumed, which livers cannot handle.
Your last comment is an absurd comparison. I will make no further comment on it.
What I will say is that you do need to make clear who you are responding to when posting comments here or your comments will appear without context, rendering them largely meaningless. If you want to take part in public discourse debate then that is one of the rules of the game.
You can read all about my depravity in van Tulleken’s book – he dedicates several pages to me.
It depends on what one is interested in: there are obviously “healthy” fats (healthy sugars is more difficult, unless of course you include the more complex ones). However, the fats (and sugars) generally used as macronutrients all have the same impact on hsealth.
In that case, would you be able to point me towards a formatting guide? I’m more than happy to follow the rules of the room!
Please answer the questions:
And as to why should we act, please explain why it is that we should not given that since those foods that contain high quantities of sugar, and most especially fructose, plus salt, fats and extensive chemically created artificial ingredients intended to both extend life and induce consumption via addiction, became commonplace forty years ago:
– Obesity has risen?
– Type 2 diabetes has risen?
– Metabolic diseases have risen?
– All conditions related to glucose induced inflammation have risen?
Are you suggesting:
– This has not happened?
– There is no connection with UPF consumption?
– There is another cause, in which case please suggest what it is?
Might you please also explain why , if UPF is not an issue, then:
– The food industry is the biggest employer of scientists in the UK, because processed food does not require them?
– The sugar lobby is so powerful?
– Scientific papers funded by the sugar lobby are ten times more likely to find no problem with sugar in diets than do those not funded by the sugar lobby?
You did well in answering the apologist’s comment Richard. Whatever the “professor” of nutrition’s background is, 2 big things jumped out to me:
S/he’s might be a trained scientist who would immediately recognise that obesity data needs not only to be correlated with ‘what’ the affected people eat, but also how much (case-by-case) they eat, and whether/how-well the affected parties digestive systems work (and how they are affected by physical impairment. digestive function, microbiome variety,etc.. etc. But – the point is – the data exists, is broadly correlated already and just needs refinement, not denial.
S/he does not seem to acknowledge that there is a wider, epidemiological dimension that needs to be considered. I’d expect that to be a priority to a prof of nutrition. Isn’t that the whole point of nutrition as a discipline? As above, s/he must already know that plenty of pertinent data already correlates metabolic illnesses like type2 diabetes, cardiovascular disease, hypertension, muscle and joint problems, some cancers, etc., etc, with obesity.
The professor must know (better than many/most of us) of the existence of studies such as the 2022 Lancet study, which, if I remember it properly, also correlated wealth to changing diet and becoming overweight. Lancet concluded that globally 1 in 8 people are obese and that obesity is a bigger problem on every continent (except SE Asia) than conventional ‘malnutrition’ where stunting, wasting away and being underweight demonstrate the condition. This Lancet study might have been the first to categorise obesity as a ‘malnutrition’ (not sure). Anyway, it drew out a probable (or as the lawyers say “plausible”) correlation between being overweight and consuming calorifically-rich, nutritionally-poor food.
For heaven’s sake, Professor, even the WEF and WHO are addressing the issue, the former probably to protect the (food industry) culprits, and the latter to show they can be more than authoritarians that promote/devise Public Health Emergencies of International Concern (PHEICs) on epidemiologically dubious data.
I read the professor’s post as an apologia for the food industry, Richard. If his/her age is sufficient, s/he probably first developed the techniques many years ago for the tobacco companies.
* https://www.weforum.org/stories/2024/03/global-obesity-healthy-lives/
https://www.who.int/news/item/01-03-2024-one-in-eight-people-are-now-living-with-obesity
There are more responses to come on this from the professor in question. I have been delayed in getting to them by other work.
I appreciate your comments and share your sentiments.
It is easy to find criticisms of scientific studies in public health, but the criticism isn’t constructive unless it points out a major flaw (what you quote doesn’t) or discusses the study in relation to evidence for alternative hypotheses. By the nature of public health almost all evidence will be observational, but that isn’t a block to good science (other rigorous disciplines are based on observational data, e.g. astronomy or palaentology). It would be totally impractical and ethically highly controversial to mount a randomised controlled trial where large numbers of individuals keep to defined diets over periods of years to decades.
The absence of “perfect” data isn’t a reason not to use the evidence that is available.
I would take more seriously an argument that the risk to health really comes from high glycaemic index foods (those from which simple sugars are rapidly absorbed) rather than the level of processing involved in their production. Since many UPFs are also high GI there would have to be careful scrutiny of the evidence to distinguish the alternative hypotheses, and you could have constructive discussions about how to determine where the biggest risk is. Arguably the difference in practical terms could be negligible: dietary advice about food choice would probably come up with very similar recommendations whether using ultraprocessing or high GI as the characteristic to avoid, simply because of the huge overlap.
One area where your critic does have a small point is in the difficulty of classifying food. In Richard’s original blog he said supermarkets “knew” which of the foods they sell are UPFs. In fact there is a significant grey area (discussed in van Tulleken’s book) and any analysis using their data would need to involve relevant experts in assigning classifications at an item by item level. That doesn’t mean it couldn’t be done though. Or that it wouldn’t be a source of interesting data.
(Though I am not sure about supermarkets allowing researchers to link purchases to individual identifiable customers. As far as I know supermarkets do allow access to their databases – ten years ago a colleague of mine was using them for his research – but only on an anonymised basis. It would quite likely be illegal without the explicit permission of each loyalty card holder).
Thanks, Jonathan.
Other comments will be coming soon on this thread.
I need your point about high GI foods and UPF foods. The rate of overlap is likely to be very high, meaning as you say that whichever indicator is used, the likely outcome will be the same. Sucrose is an obvious common element. I can live with what works.
I also note you say that there is controversy over the boundaries of what is, and is not, UPF, but that you recognise that it would be entirely possible to resolve these issues with a little extra research. Given the vast numbers of scientist engaged in the food industry, I think that that is not a problem.
Finally, I really do not think that it would be necessary to link through purchases with individuals at supermarkets. I see absolutely no value in doing so, because nobody would be suggesting that this data would be a diagnostic tool for those at risk. This data should be used for public health purposes, in which case high-level information is all that is required. I very strongly suspect that this gives rise to no data shoots of any consequence.
Yes Gunter Kuhnle, ultra-processed food as a category needs delineation and better (probably narrower) definition. I can see the point your examples are drawing, but I’d have difficulty categorising all (supermarket) bread simply as ‘UPF’ – the stuff that’s baked with a hairdryer and boasts “added” b vitamins on the wrapper and the ‘taste the difference’ multi seed loaf might be equally ‘processed’ but the fats, preservatives, flours, etc. are relevant to the product’s nutritional profile, no? A working definition of UPF would have to include more than the amount of processing the factory, lab or artisan had to provide for its production, no? It should, I’d guess, need to quantify nutritional benefit against nutritional harm as a minimum.
But yes, we need a tighter definition of ‘U-PF’.
This must be a misunderstanding – I do not endorse FFQs, which should be quite obvious from my research. They are useful for some aspects of nutrition research, but not this area.
What do you make of it? All the claims you make here about UPF are large based on UPF.
How would you do this on an individual level? I actually do such research, and we are unable to estimate waste without getting detailed information from study participants.
It depends on what it is you would like to know: if it is health effect based on individuals, we need data on individual consumption.
So please do tell me: why do you believe the studies claiming an adverse effect of UPF on health when you reject all methods that support such a link?
Public health is a macro issue.
You clearly are unable to discuss it as you do not comprehend that. As I said, if you did not disclose funding I would go no further. You won’t.
You clearly seek to obstruct constructive debate. I now know why you are so criticised. Thread closed.
Thanks Richard. I wrote the above before your correspondent identified himself as Professor Kuhnle and you engaged in dialogue. It turns out that he believes the best indicator of risk is HFSS determined from food composition, which will also have a high overlap with UPF; however in criticising your comments on UPFs he didn’t really make the case for processing not being part of the problem (and from what I read, among other things ease of sugar absorption and gut benefits of fibre do change with processing). He does indicate a pragmatic benefit of using HFSS, foods can be classified objectively in the lab from chemical analysis, but that doesn’t relate to where the ultimate risk factors lie.
It is strange he seems not to understand why perception of conflict of evidence is going to be increased if someone obscures a source of research or personal funding. Most food scientists ending up having some funding from companies – it is not a field with many other sources of funding – but the honorable approach is to be completely transparent about that and let the reader judge whether the academic argument might have been influenced.
I perhaps misunderstood your previous blog where you referred to the supermarket data including that on shoppers as well as sales. You prompted me to see if I could find any information on my colleague’s findings but I was unsuccessful. Tragically he became ill and died from cancer over a short period and it is very possible it never got published; he was looking at family diets based on supermarket “food baskets” (till receipts associated with cards), and I think the data made available to him allowed comparisons of locations with different socioeconomic and public health challenges.
Thanks, Jonathan
That sounds like useful research
All observational work requires good, validated methods. That might seem to be pedantic, but neither astronomy nor palaeontology would be where they are had they ignored that existing methods have flaws. FFQs – the method most commonly used to estimate food intake – have never been designed to measure UPF. If one looks e.g. at the Harvard FFQ (which is very often used), it combines a range of breakfast cereals with porridge – clearly different NOVA categories. Likewise, the UK Biobank uses a very limited dietary assessment instrument.
We do work on better methods to estimate intake, but this is very difficult for a group that is so diverse and where there are many ambiguities (e.g. sausages – made completely from meat and spices) are UPF, bacon made with nitrite is not).
Another possibility is texture – as the study by Kevin Hall has show, UPF are easier (and quicker) to eat when compared to minimally processed foods. And this is something most people will have noticed (and is incidentally the problem with fruit juices: it is very easy to consume much larger amounts of fruits than if one had to chew).
There are ways to do this – but as you say, there are many difficulties, mainly regarding data protection rules. I believe there is work on this going on.
Please do not discuss FFQs on a publiuc plaform without explaining them. If you wanty to ngage here say what you mean.
Let me add this:
I note you refer to the use of food frequency questionnaires to undertake the study of nutrition. I have to admit that I find this quite astonishing. I also am a little surprised that you think it’s appropriate to comment on a blog using an abbreviation without explaining what it means. Are you aware of what is necessary when engaging in public discourse?
A simple web search provided me with the following information:
A food frequency questionnaire (FFQ) is a survey that asks about how often and how much food and drinks a person consumes over a period of time. FFQs are a common tool for assessing diet in large studies and epidemiologic research.
Here are some things to know about FFQs:
### How they work
FFQs typically include a list of foods and beverages, and ask participants to indicate how often they eat each item. The frequency categories are usually presented in a tick box format.
### What they’re used for
FFQs are used to estimate nutrient intake and to relate food consumption to disease outcomes.
### How they’re created
When creating an FFQ, researchers consider the foods and food groups to include, and ensure that the list captures habitual intake.
### Strengths and weaknesses
FFQs can be prone to measurement error, especially for recall periods longer than a week. They can also be time consuming to administer if they’re too long.
Politely, these might have been at forefront of research when the unit at which you work was created in the 1940s, but that anybody might now so use such a method of research seems to be quite astonishing. There are a number of good reasons for saying. Firstly, those willing to participate in such research will be few and far between, and create an enormous problem with selection bias. Secondly, as any GP knows when asking a patient about their alcohol consumption, people are incredibly inclined to lie when answering questions with regard to any food or drink intake. Thirdly, people simply forget to do these things correctly. Fourthly, as the above commentary notes, these questionnaires have to be very simple to be used, meaning that the degree of complexity required to identify any difference between processed and ultra processed food consumption is very unlikely to be picked up by such a survey. I am astonished that any university ethics committee might now consider their use, so inherently unreliable must they.
There are two alternative potential methods that I can imagine, without undertaking any further research into what is actually done to appraise the basis of peoples nutrition and its long-term impact upon health.
One would be to undertake a controlled study, keeping two groups in complete isolation so that their precise food intake could be observed over a period of time, with its impact on their health over a period of at least weeks, if not months, being measured as a consequence. However, there would again be a massive problem with selection bias in the sample, and I really doubt that any university ethics committee would permit such a study on that basis because it would, effectively, require imprisonment of those engaged in it, and you can see all the problems with that. As a result, this normal basis for biological research is not available to you as a nutritionist.
Alternatively, you could do work on very large samples of the sort that I suggested, analysing the actual food sold by all our leading supermarkets, comparing that profile with known data on the likely processed/ultra post food status of the items in question. Given that we know that the vast majority of food is not purchased to be thrown away, this would provide a quite extraordinarily accurate view of the actual consumption of food in this country. Nothing else could do that. Why wouldn’t you want to use such a method?
Why should I not mention FFQs? After all, they are the basis for almost all the claims you make here.
Almost the entire UPF literature is based on FFQs, so I would expect anyone who questions my point about unreliable exposure data to understand what FFQs are.
Please do have a look at the many papers you quote about UPF – almost all of them use FFQs; they are the most common used instrument used in nutrition research (they have been developed in the 80s in Harvard).
I completely agree with you – but then why do you disagree with my observation that these data are incredibly unreliable?
One could supplement food or make subtle changes to the diet – e.g. substituting UPF bread for non-UPF bread (and other foods of course – bread is just an easy example). But I agree, it would be difficult.
The problem with purchase data is that they are usually per household and not per person, and they don’t account for food waste. The UK’s national diet and nutrition survey (NDNS) provides an incredibly detailed account of food intake using a 24h recall (i.e. asking someone in an interview about their food intake within the past day) and collects urine sample to measure objective biomarkers. This is probably the best we have at the moment, and one could obviously expand this to ask about actual foods consumed (i.e. brand).
You endorse FFQs in other comments – hence my suggestion that you were a believer in them.
And given that this is a public site – you should not believe anyone knows what an FFQ is. It is arrogant of you to do so. I had to make good your omission. Do ypu know waht public engagement is?
Your arguments about food purchase data are utterly spurious – and waste is eaasily calculated – as you must well know. But also – in aggregate all of this falls out – we are looking at public health here. Do have any idea what that means? I continually get the feeling that you do not, or choose not to do so.
On the other hand, 24hour recall is ridiculously unreliable and yet again, like urine smapling, subject to massive selaction bias.
I don’t think anyone will disagree with this statement – and I have done so as well. But this has nothing to do with the degree of processing, use of additives or the intent of processing. A home-made cake or honey meets these criteria – but they are not ultra-processed. Likewise, supermarket whole-grain bread is low in salt (compared to artisan bread), high in fibre and not especially energy-dense, yet it is UPF.
The “UPF” concept is used by a number of people to generate income – by selling books and media appearances. But even a report that is reasonably biased against UPF – such as the one by the House of Lords committee – does not see UPF as the main issue.
I am preparing answers to your other submitted comments at this point in time, but I will answer this one first.
I do so in several ways. First, I know that you have not responded to any of the questions that I posed to you in my original piece. You have entirely ducked the real issues here. Why is that? Would you please now answer them?
Secondly, if you think the House of Lords exonerated ultra process foods, you must be referring to a different report from the one that I read, because they expressed their considerable concern with regard to those foods, and demanded action about them. Why misrepresent what they said?
Thirdly, why are you so pedantic? If you have any understanding of large data sets (and your comments suggest that you do not) then you would of course know that in every such larger data said there will be imperfections and problems in data allocation at boundary points, which I made clear in my original comments. To suggest that a home-made cake confounds the quality of a large data set is, to be polite, quite ridiculous. The number of home-made cakes in this country is quite small, and in fact by any definition of ultra processed food would very clearly not be UPF, but be processed, although I accept that there might be an issue around flour.
Fourthly, there is literally no one who I know who looks at these issues who does not think that the supermarket bread is anything but UPF. Why do you disagree? Might you please explain how it can have such an extraordinary shelf life? If it is not ultra processed, including significant numbers of ingredients that are very clearly not needed to make bread unless the aim is to increase its sweetness, extend it shelf life, and increase its toxicity.
That, brings me back to my main point now, which is please answer the questions I posed. You are saying others are promoting UPF For their own financial reasons. What are your financial reasons for denying it exists, and what is your reason or explanation for the phenomena which are observed, and how do you explain the link between the rising ultra processed foods and obesity, type two diabetes, and other factors, or are you suggesting that there has been some change in the human genome that has created these ourcimes? Please provide reasoned comments.
A very interesting discussion – increasingly it is not just food but everything else in modern life that is ultra processed – including economics and politics.
Everything is becoming more and more uneccessarily complex and distant from its core purpose.
The working definition of UPF is that it contains ingredients you could not find in your kitchen at home and if we concentrate on just two things -first, emulsifiers which are widespread in today’s manufactured food. Brand owners don’t want lumps or hard bits upsetting the mouthfeel! But they could well be destroying the gut biome by emulsifying it – emulsifiers certainly will not improve it. So when we do eat unprocessed food such damaged guts may not be able to release its proper nutritional value. Second are the seed oils which are so disgusting in both smell and taste that without chemical processing and purification they would never, ever be eaten.
That, as these unusual things have become more and more widespread and manufactured foods more widely consumed and at the same time populations, including children, have suddenly and consistently been getting more obese is an odd coincidence. The argument that these substitutes make food cheap is all very well – but it is not actually food as we used to know it.
And if food price inflation results in human length of life and good health inflation then it will be worth it.
Thanks, Peter
I am not posting further comments from Gunter Kuhle as they say ntohing new – and that breaches the moderation polocy here.
This subject has fascinated /interested me for more than two decades.
This is just one example of the anomalies we face when trying to eat well for our health.
NHS Dietary advice re carbohydrate choice
Top of the list
Baked potato with skin
Atkins Glycemic index
Third on the list and therefore to be eaten sparingly if at all
Baked potato with skin
@RoberJ
I have provided a detailed list here, well beyond what is commonly expected (5 years are usually sufficient). You can look of course at all my publications and my ORCID record where all relevant information are provided.
You referred us to your website for transparent declarations about funding.
You dissembled when we asked in more detail about funding.
There isn’t any information about funding on your website.
Eventually you made declarations here when pressed but didn’t provide links to where that information was published on your website, the place you claimed published that information.
I feel that I have been misled and I do not regard you as a reliable source of information. I do not trust you and will not be engaging further. Have a nice day.
Since some people believe my declaration of interest is so important, I put this here.
Currently, I am Co-I on a BBSRC funded research project (https://gtr.ukri.org/projects?ref=BB%2FV004905%2F1) and have no other funding. In the past, I have received research funding from the EU (Flaviola – https://www.flaviola.org/ and PHYTOME – https://cordis.europa.eu/project/id/315683/reporting). I have also received research funding from Mars to develop and apply a biomarker of flavanol intake and investigate associations between intake and health (https://pubmed.ncbi.nlm.nih.gov/33087825/), but this has ended more than 5 years ago.
I am obviously also employed by the University of Reading and have family links to a vineyard.
And since there are claims I would reject adverse effects of sugar: one of my more interesting work (https://pubmed.ncbi.nlm.nih.gov/28723954/) showed for the first time in a larger dataset that sugar intake is associated with obesity.
I’m not sure how this affects my main point: the data linking UPF intake with health are very weak because we have no reliable information on actual intake. The focus on UPF is likely detrimental to public health, as it shifts the focus away from an established approach, i.e. food composition, to a new and – in the opinion of many experts – very questionable approach.
See the comment by Jonathan
Your reticence has already undermined your credibility
But none of those links are from your own site, are they?
You did refer us to your own site, for the full transparency you claimed was there.
I’m even more sceptical now.
“blood out of stone” comes to mind.
Correct
“Credibility shot” was my thought.
In Tulleken’s ‘Ultra Processed People’s introduction says if food has any ingredient not in a ‘standard home kitchen’ it is UPF.
But surely, as Richard suggests, we don’t really want to be nit picking about very precise definitions , or even about ‘ingredients’ vs ‘processing’ – the massive ‘industry’ devoted to making food containing the wrong kind of sugar fat and salt , addictive , together with the dramatic rise in obesity and diabetes etc etc is enough.
If we indeed need more research – at least we should be adopting the precautionary principle – banning or taxing rogue ingredients, processing banning advertising etc etc.
Also a drive to encourage people to cook and even grow their own food – make it fun for kids etc.
If ten percent of the TV advertising breaks did this – as a public service it could help.
Much to agree with
https://www.taxresearch.org.uk/Blog/2024/12/31/why-has-obesity-increased/comment-page-1/#comment-1000029
@Gunther Kuhnle
I note that your reply to my questions gives me no more information than I already had.
I’m amazed that you claim, as a professor, to know so little about your colleagues’ funding.
Frankly, I don’t buy it.
But let’s stick to YOUR research and YOUR site which I am assuming is
https:kuhnle.co.uk
I did some site searches:
Site: https://kuhnle.co.uk
“sponsor” – no results
“funding” – no results
“Mars” – no results
“industry” – no results
“Ltd” – no results
“Limited” – no results
“Inc” – no results
“Foundation” – no results
“Trust” – no results
“FFQ” – no results
“Flavanol” – gives one result,
https://kuhnle.co.uk/icph_2024/lessons_learned_kuhnle.pdf
but sadly that throws up an “Oops that page can’t be found” result
Taking the pdf title on its own doesn’tt turn anything up either⁷ and the site’s own search function for flavanol takes me to the title of a paper by someone else.
Maybe I’m searching for the wrong terms in the wrong site?
Could you please be transparent and actually provide some clickable links to where you DO disclose your funding? Because so far, you HAVEN’T and I’m forced to the conclusion that you simply don’t want to – but you can prove me wrong and I will go to bed a wiser happier man having grown in knowledge.
I’m sure you don’t want me to draw the wrong conclusions.
it used to be MSG – now it is UPF – I have tried to avoid both at different stages of my life and especially recently, and I do get questioned by the medics about sugars, salt & additives etc – fortunately I can and do cook most of my meals from scratch so I know what is in them and can maximise nutrition. I cannot tolerate over flavoured foods, especially re salt(s) – I have found this thread very enlightening – thank you Richard and your supporters.
Thanks, Susan. It has been a bit distracting, but I learned a lot, even if I did not wish to know it all.
While I am very cynical of the food industry, including food research, I also believe that many in the industries believe that their objective is to produce cheap nutritious food, and that is the box in which the work.
Likewise there are many politicians who believe they understand politics and economics, and many in the pharmaceutical industry who believe that they produce safe beneficial drugs.
There was also a time that I also believed the BBC and some newspapers disseminated impartial neutral news.
I have turned that grumpy cynical old man I was once cynical about, and now realise they knew more than I gave them credit for. That reevaluation began only a few years ago.
It is hard not to be very cynical. Spar says: “Obesity & insulin resistance is the risk factor for Type 2 Diabetes and not the consumption of
excess sugar.” https://blackdotresearch.sg/sugar-risk-type-2-diabetes/
The article goes on to explain that “Excess consumption of sugar is not listed as a risk factor for type 2 diabetes as eating a lot of sugar does not appear to cause diabetes directly.”
It would be a bit like saying that drivers do not cause road traffic deaths, that is caused by the impact of cars with people.
Agreed
That is an absurd claim
I feel the same as you but would describe it in a different way. We are not old and cynical, we’re oldER and wiser. We have years of experience and we’ve seen “it all before”. (Alright, a lot of it before). We no longer accept things at face value, we use healthy degree of scepticism before reaching conclusions on important topics.
I think the world needs more people who operate this way on a daily basis.
I’m a theologian and a retired educator, rather than a scientist, but find the intelligible academic articles I’ve read, such as this (open access) one below, support Richard’s point and gives me sufficient cause to be concerned both personally and for public health.
https://www.tandfonline.com/doi/full/10.1080/23311932.2024.2438405#d1e285
The evident tensions and suspicions in the debate are understandable to me though, having tread Marion Nestle and Michael Pollen’s “Food Politics :How the Food Industry Influences Nutrition and Health”, which I recommend.
Thanks