select search filters
briefings
roundups & rapid reactions
before the headlines
Fiona fox's blog
A presentation at the International Congress on Obesity looks at ultra-processed foods pushing aside other food groups.
Prof Gunter Kuhnle, Professor of Nutrition and Food Science, University of Reading, said:
“The NOVA classification was conceived by Professor Monteiro. The definition is ambiguous (and has been interpreted differently by different researchers) which makes it difficult to conduct research and provide advice.
“The body of data we have from scientific evidence does not support some of the claims made (that ultra-processed food causes ill health). In my view there is a high risk that if we focus on level of processing this could distract from the evidenced problems that make it more difficult to find solutions to address diet-related health problems.
“Some ultra-processed foods are high in sugar, fat or salt and known to have an adverse effect on health – while others can be an important source of fibre, wholegrain or fish. There is currently no scientific model that would support the claim that all have an adverse impact on health.
“The study by Kevin Hall and colleagues is often cited as evidence for an adverse impact of UPF on health, however, this it is not that straightforward. It has been claimed that only ultra-processed – but not processed – food have an adverse impact on health. However, the study did not compare ultra-processed food with processed food, but with minimally processed food. It is therefore unable to provide any evidence for a difference between those two categories.
“The ‘umbrella review’ shown in the presentation and referred to in the press release in my view confirms indirectly the lack of evidence: the authors themselves considered most of the evidence suggesting a link between ultra-processed food intake and health to be weak or very weak. This is not surprising as most evidence is derived from observational studies and it is almost impossible to estimate ultra-processed food accurately with the methods currently available (see e.g. the SACN opinion on processing).
“In my view there is no evidence for the claim that ultra-processing creates a worse nutrient profile than processing – mainly because of the wide range of foods. Ultra-processed bread – most bread consumed in the UK – often has higher amounts of wholegrain and fibre than comparable non-ultra-processed breads. Moreover, ultra-processing allows a reduction in salt intake and many supermarket breads meet the recommendations for salt reduction (in contrast to many artisan breads). The claim that ‘ultra-processing’ reduces the amount of phytochemicals when compared to other methods is not supported by evidence in my view.
“The presentation discusses a range of compounds and claims that exposure to them is higher in ultra-processed food – although the evidence is not clear. The formation of contaminants such as acrylamide can be controlled – and in particular the processes during ultra-processing and the analytical methods of producers can offer a much better control. Bisphenols are found in food packaging, and the type of packaging is not part of the ultra-processed food definition – indeed, processed and even minimally processed foods can be packed in plastic.
“The evidence for adverse effects of artificial sweeteners and emulsifiers is very weak – and these compounds are regularly reviewed by the respective authorities/agencies.
“Comparing the food industry with the tobacco industry is not straightforward: for most foods that fall into the ultra-processed food category, there is no strong evidence for any effect on health – be it detrimental or beneficial. Civilisation as we know it would be impossible without a food industry that ensures the production and distribution of foods.”
Dr Hilda Mulrooney, Reader in Nutrition & Health, London Metropolitan University, said:
“It is true that national healthy eating guidelines emphasise the need to focus the diet on foods which are less processed (e.g. the Eatwell Guide states ‘choose wholegrain or higher fibre versions with less added fat, salt and sugar’ in relation to the starchy carbohydrates group. However, that is far less stringent than what this author is suggesting, and the guidance focuses more on nutritional value of foods than their degree of processing.
“Many healthcare professionals including dietitians have advocated for years for taxes on less healthy foods to subsidise the cost of more healthy foods (the Food Foundation calculated recently that per calorie, more healthy foods are twice the cost of less healthy foods). That is not focused specifically on ultra-processed foods as per the NOVA categorisation though, but based on their nutritional content – which is something that NOVA does not take into account – a reason many people have problems with this classification system.
“Taxes on sugar sweetened beverages in the UK have been shown to be successful in driving reformulation and changes in consumer behaviour, far more so than voluntary guidance to reduce sugar content of children’s foods for example. But treating food like tobacco is very simplistic. There is no such thing as a safe cigarette, even second-hand, so banning them is relatively straightforward in that the health case is very clear. However, we need a range of nutrients including fat, sugar and salt, and they have multiple functions in foods – structural, shelf-life – not just taste and flavour and hedonic properties. It is not as easy to reformulate some classes of foods to reduce them and they are not the same as tobacco because we need food – just not in the quantities most of us are consuming.
“There is an argument in my view for taxing high fat, salt and sugar foods to subsidise the cost of healthier options – but in my view this should be based on their nutritional profile, not categorisation as ultra-processed. That is so wide a category that it includes foods that make valuable contributions to dietary intakes of several population groups.
“One slide states that UPF cause several diseases; associations have certainly been shown but this is not the same as demonstrating causality, and the chronic diseases in question are multi-factorial in origin.”
Dr Duane Mellor, Dietitian and Spokesperson for British Dietetic Association; and Honorary Academic Fellow at Aston University, said:
“This is an interesting discussion of previously published analyses of studies which explored the risk associated with ultra-processed foods and diseases ranging from type 2 diabetes and cardiovascular disease, to poor mental health and some cancers. It does not provide any new data, but looks to combine thoughts from a range of sources and theories.
“The point about dietary intakes being unhealthy in a number of countries is not really disputed, with diets in many countries reported to be in too high in added fat, salt and sugar and low in fibre and minerals including potassium. It also highlights how the food industry markets foods that are highly palatable and tend to be less healthy than foods which are minimally processed. However, it is not straightforward to draw parallels between the food industry and tobacco industry, as food is essential to life, tobacco is not. Also to have a safe food supply in cities our modern society needs some processing to prevent food from becoming contaminated and spoiling which might result in illness which include diseases like pathogenic strains E.coli which there has been a recent outbreak of in the UK.
“It also doesn’t acknowledge the limitations of the NOVA classification, which does not consider beer or wine as ultra-processed foods but includes traditional foods like tofu and tempeh as ultra-processed! This doesn’t take into account other research which highlights the potential of these soya based foods to reduce cholesterol for example and may understate the risks of alcohol consumption. This classification which is potentially open to subjective interpretation, which was developed in one country, does not appear to take into account the traditional foods of other cultures and countries. If a classification of ultra-processed food is to be used, it would need to be objective and possibly have a locally appropriate approach to suit the food systems and supply in other countries. Also because of its descriptive approach to classifying foods it would also potentially be open to complaints and litigation with food producers arguing that their food is not ultra-processed by comparing its production and ingredients list to a non-ultra processed equivalent.
“Additionally, when looking at some of the research that examined the effects of ultra-processed foods, for example the recent study from Harvard, it excluded ultra-processed wholegrain cereals from the ultra-processed foods in the diet. So, not all of the research uses the same definition of ultra-processed foods. This highlights that if ultra-processed food is to be used in regulation of foods, there needs to be clear and agreed definitions. Currently, we have the NOVA classification developed by Professor Monterio’s group in Brazil, which due to its simplicity has been used in thousands of analyses of previously collected data to produce research papers. There is a clear need to get a globally agreed objective definition that can be applied without needing to make judgments about a food so that if it is possible foods can be reliably classified according to their degree of processing and how healthy they might be.
“One potential aspect which is interesting though, is the point made that while we know foods high in added fat, salt and sugar make up too much of our diets, we need to be very aware that reformulating these foods to replace fat, salt and sugar with other ingredients does not automatically make them healthier. This is perhaps best shown with soft drinks, where research does not suggest that either sugar sweetened drinks or non-sugar or artificially sweetened drinks are linked to positive health outcomes. So, it perhaps does suggest we might need to be mindful as we seek to support and promote communities to eat a healthy diet, that if reformulation of products is undertaken, we need to avoid unintended consequences or risks from the new food product, which might be lower in fat, sugar and/or salt but may not be as healthy an option as we might have been thinking up until now.
“Countries do need to work harder to support healthier diets in their populations, but we need suitable and objective ways of doing this – currently the NOVA ultra-processed foods classifications is open to subjective interpretation, and a simple way to quantify the degree of ultra-processing a food might making it harder to effectively regulate.”
Presentation title: ‘Ultra-processed foods and the pandemic of obesity: the thesis and the evidence’ by Carlos A. Monteiro was presented at the International Congress on Obesity (ICO 2024) and was under embargo until 04:01 UK time on Thursday 27 June 2024.
There is no paper.
Declared interests
“Prof Gunter Kuhnle: “https://kuhnle.co.uk/coi.html
Employment
Research funding
Committee membership
Scientific organisations
Others
Dr Hilda Mulrooney: “I have no conflicts of interest to declare.”
Dr Duane Mellor: “I have discussed and been a consultant about the concept of UPF for EUFIC, APPGs (both unpaid as part of my previous academic role) and members of the food industry (including WSRO, Mars and Danone as a paid independent consultant).”