select search filters
briefings
roundups & rapid reactions
Fiona fox's blog
A perspective piece published in Science looks at the evidence around ultra-processed foods (UPFs).
Dr Sumanto Haldar, Lecturer in Nutritional Sciences, Bournemouth University, said:
Does the press release accurately reflect the science?
“While food texture and eating rate manipulation may be important in energy intake and body weight regulation in the shorter term, we don’t know what the longer-term effects (e.g., people exposed to harder texture foods for years) on body weight regulation. Food texture is likely to be one small part of a much large jigsaw, that includes nutritional quality of foods, the dietary context, the food environment etc.
Is this good quality? Are the conclusions backed up by solid data?
“Yes, this is written by established and highly respected researchers in their field and the conclusions that the authors have made are generally supported by the data which they are discussing.
How does this work fit with the existing evidence?
“Currently there is a lot of attention on ultra-processed foods consumption and body weight regulation, cardiometabolic health risk etc, although as the authors rightly acknowledge, most of the evidence is rather observational (mostly cross-sectional associations), rather than being causally linked. Given that overall DIETS high in UPFs are also poorer in nutritional quality, the associations between UPF consumption and poor health outcome had been consistent. Of course this cannot be said for individual foods and therefore does not apply to all foods. The former evidence led to several recent high-profile publications (including the Lancet series on UPFs) and huge media coverage, which in a way is very useful to highlight the importance of diet on health and longevity. However, there are several nuances and this Science article nicely highlights these and supports the fact that it is not the extent of processing per se which may be the real problem.
Have the authors accounted for confounders? Are there important limitations to be aware of?
“This article is a Perspective piece rather than original research – it is a secondary evaluation and commentary on several primary research consisting of randomised controlled trials (RCTs) on UPF and body weight changes and cardiometabolic risk. So there will be different levels of confounder adjustments in different studies, including their inherent limitations. One of the major limitations in all these RCTs are that while body weight regulation, cardiometabolic risk factors are all chronically determined (over months, if not years) the RCTs to date are only up to 8 weeks long (less than 2 months). Therefore, we cannot make conclusive recommendations or form policies based on the current evidence.
What are the implications in the real world? Is there any overspeculation?
“The article supports against the excessive demonisation of processing per se. In fact, it is other nutritional attributes of our diets such as the protein content, the fibre content, the composition of nutrients such as vitamins, minerals and other phytonutrients (such as polyphenols, carotenoids etc). which together contribute to the nutrient density of foods that are likely to be more important. For example, lycopene which is a naturally found carotenoid, with health benefits, that are found in tomatoes have greater bioavailability in tomato paste and sauces than from raw tomatoes.
“Given the need for ensuring food safety and food security, we must continue processing foods to ensure adequate supply of affordable foods to provide basic nutrition for everyone. Of course, there are significant opportunities to improve our food system and ensure that we continue to provide foods that nutritionally dense, affordable and accessible – but this doesn’t necessarily have to be at the cost of stopping food processing, which often help to maximise our limited food resources, ensure food safety and reduce food waste.
“In my view this Science paper and its authors offers an excellent perspective in this era of “ultra-processed foods”.
Prof Jules Griffin, Director of The Rowett Institute, University of Aberdeen, said:
“The perspective article by Magkos and colleagues investigates five randomised controlled trials that study the impact of ultra-processed diets on weight gain and obesity, providing a detailed critique of these studies. They question the evidence provided by these randomised controlled trials, deciding there is overall weak evidence from these trials that ultra-processed food contributes to obesity, but their perspective article has missed some important alternative explanations.
“The authors criticise the current randomised control trials for not taking into consideration the soft texture of ultra-processed foods into their study designs but this was not the primary focus of the previous trials, and soft texture is one reason that has been put forward as an explanation as to why ultra-processed food is consumed in excess. The authors make much of one study where both the ultra-processed food and minimally processed food groups lose weight during the study – it is not unusual for people to lose some weight on nutrition studies, and this is why a cross-over design is used with a control group to compare between the two groups. The important outcome of the study is that the minimally processed group lost more weight under experimental conditions – both groups may have stopped less healthy behaviours like snacking during the study.
“The high salt, free sugar and saturated fat content of some ultra-processed food certainly explains some of the associations with ill health effects, but in large observational studies where ultra-processed food intake has been followed, disease risk for cardiovascular disease and all-cause mortality still remain after correcting for individual nutritional components including sugar, saturated fat, salt and fibre intake.
“A better approach than considering these trials in isolation, is to consider the observational epidemiology studies alongside the evidence from randomised controlled trials to determine whether they agree with each other – given a lot of studies do agree, there is a need for more trials to understand what the effects of ultra-processed foods are on health.”
Prof Gunter Kuhnle, Professor of Nutrition and Food Science, University of Reading, said:
“Scientific research relies on disagreement. New ideas need to be questioned, challenged and scrutinised.
“This perspective is therefore very welcome. While it is not a systematic review of the existing literature and presents the opinion of the authors, it is based on data available and raises a number of very important questions about the concept of ‘ultra-processed foods’ and their alleged impact on health. The main criticism now – and in the past – has been that ‘ultra-processed foods’ are a poorly defined and very diverse group of foods that include foods that are already known to have an adverse impact on health (such as soft drinks or processed meats), but also foods where such an impact is unlikely (wholegrain bread, yoghurts or fish fingers). It has also been known that many foods in the ‘ultra-processed’ category are more energy dense and thereby facilitate overconsumption – however, this is not restricted to ultra-processed foods and there are other energy dense foods, a home-baked cake for example.
“The key question is whether the concept of “ultra-processing” is better than the existing nutrient profile – based on food composition – to provide dietary advice to the public. There is currently no evidence that this is the case. A clinical study at Imperial College London [1] showed that as long as a diet largely meets current dietary recommendations, there is no difference between ultra-processed and minimally processed food.
“Processing and indeed ultra-processing have helped to make food more affordable – and the proportion of income spent on food has roughly halved in the last 60 years [2]. The nutrient profiling system, while far from perfect, is known and understood by the public and relies on objective measures that can be linked to a wealth of data.
“The research funding used to promote the concept of “ultra-processed foods” could be used more efficiently to improve the food system and focus on aspects of the diet known to affect health.”
[1] https://www.nature.com/articles/s41591-025-03842-0
[2] https://blog.ons.gov.uk/2018/01/18/celebrating-60-years-of-family-spending/
Dr Seamus Higgins, Associate Professor in Food Process Engineering, University of Nottingham, said:
“The paper is a useful reminder that the science surrounding ultra processed foods or UPFs is far from settled. The authors argue that many of the effects attributed to UPFs can be explained by well-established factors such as calorie density, fibre content, protein levels, food texture and eating rate, rather than processing itself. Importantly, these are the same factors that emerging research into the microbiome, satiety regulation and metabolic health increasingly identifies as key determinants of how various foods interact with human biology.
“But the paper does raise a broader question. Have we become overly focused on food classification systems while overlooking the underlying biological mechanisms that influence health? NOVA was developed as a practical four-category identifier for food processing, not as a definitive indicator for food science. While it has proved valuable in highlighting trends and debate in modern food production, classification is not causation. There is a danger in treating what is meant to be a descriptive framework for consumers as a definitive explanation of how foods interact with human biology.
“At the same time, the debate risks overlooking something the food industry has understood for decades: consumers rarely choose foods according to scientific classifications. Their choices are driven primarily by price, taste and convenience, all of which interact with biological instincts shaped during millions of years of food scarcity. The modern food system has become exceptionally effective at satisfying these ancient drivers, helping explain consumption patterns far more effectively than food classifications alone.
“Perhaps the greatest challenge is that much of nutrition science still relies on calories – a measurement system developed more than 125 years ago to quantify the energy value of food rather than its biological effects. Calories remain useful, but they were never designed to measure satiety, metabolism, microbiome interactions, hormonal responses or long-term health outcomes. Indeed, food labelling regulations for calorific value continue to accept a degree of inaccuracy of up to ±20%, reflecting their origins as a measure of fuel value rather than biological response.
“As our understanding of the microbiome and human metabolism continues to evolve, the question may not simply be whether food is processed, but whether the tools we use to measure and classify food have kept pace with the science. The future of nutrition research may lie less in debating categories of food and more in understanding how food, biology and behaviour interact within the complex systems that shape human health.
“It is also important to recognise what this article is and what it is not. This is a Perspective piece rather than a new experimental study or systematic review. As such, it does not present new data but offers an expert interpretation of the existing evidence base.
“The article presents a well-argued and scientifically credible perspective, particularly in highlighting the challenges of separating the effects of food processing from factors such as calorie density, fibre content, texture and eating rate. However, it should not be interpreted as a dismissal of the wider evidence base. Large observational studies, including recent analyses published in The Lancet, continue to report strong associations between high UPF consumption, obesity and adverse health outcomes. The existence of this relationship is increasingly difficult to dispute. The more important question is whether these outcomes are driven by processing itself – including the various techniques, ingredients and additives used in modern food manufacture – or by the nutritional, behavioural and biological factors that often accompany highly processed foods.”
‘Ultraprocessed foods and obesity: Interpreting the evidence’ by Faidon Magkoet al. was published in Scienceat 19:00 UK time on Thursday 4 June 2026.
DOI: 10.1126/science.aef3495
Declared interests
Dr Sumanto Haldar: “Declarations:
– for any of your own research
Yes, I have previously worked on industry funded research, although none of this industry had any say on the design, study conduct, laboratory analyses, data collection, management and interpretation or the writing, reviewing and approval of the manuscripts published from these works. The industry funded projects I worked with were Wilmar International, Nestle, Pinduoduo Incorporated (HongKong Walnut Street Limited).Most of these fundings did include conference travel to international conferences that were either partially/fully funding the trips.
– to attend scientific meetings
Yes, have attended scientific meetings that were part of industry funded projects.
– current or previous advisory roles or committee membership
No formal advisory roles, but I am currently in discussion with The Watercress Company (TWC) to develop minimally processed nutritionally dense food products.
– previous employment in companies
None and neither do I have any investment, shares or company roles within any company.
– if you have received industry funding for some of your other research, even if it’s not directly linked to this particular story
Received funding from Wilmar International, Nestle, Pinduoduo Incorporated (HongKong Walnut Street Limited).”
Prof Jules Griffin: “I am a consultant for Sitryx, a company specialising in designing drugs to target immunometabolism.
I receive funding from the Scottish Government, European Union, Medical Research Council, Biotechnology and Biological Sciences Research Council and the British Heart Foundation.
I hold shares in GlaxoSmithKline and Haleon plc.”
Prof Gunter Kuhnle: “I am a former member of the UK Committee on Toxicity of Chemicals in Food, Consumer Products, and the Environment; a current member of the Advisory Committee on Novel Foods and Processes; the Director of the Chemical Analysis Facility at the University of Reading, which provides analytical services to academic and commercial clients; have received research funding (2010–20) from Mars for work on flavanols; and have received consultancy payments from RSM UK and EQT, paid to the University of Reading. As a member of the EFSA ANS panel (2018-2019) and the UK’s COT (2019-2025) I was involved in the evaluation of food additives.”
Dr Seamus Higgins: “Other than being the Author of Food and Us, The Incredible Story of How Food Shapes Humanity, I have copied a recent disclosure declaration from my The Conversation articles.
Disclosure statement: Seamus Higgins does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.”