select search filters
briefings
roundups & rapid reactions
Fiona fox's blog
A conference abstract presented at the annual conference of the American Society for Nutrition looks at a new approach for classifying processed foods.
Prof Martin Warren, Chief Scientific Officer and Group Leader, the Quadram Institute, said:
“Refining the definition of processed food is key to improving scientific precision as the current NOVA categories, especially “ultra-processed foods” (UPFs), are too broad and vague, grouping diverse foods together based on processing techniques rather than nutritional composition or health outcomes. Clearly, more precise definitions would allow for more appropriate research on diet and health outcomes.
“This also has implications for policy and regulation, as governments and organizations use NOVA to shape food labelling laws as well as dietary guidelines.
“Currently, there is a mismatch with nutrient profiling with some foods classified as UPFs being nutritionally adequate or even beneficial (e.g., some plant-based alternatives, fortified foods). A refined system could integrate both processing level and nutritional quality, enabling more balanced assessments.
“It’s difficult to tell about the quality of this abstract without more detailed analysis of the paper – but the general description and approach seems logical and robust.
“A step in the right direction but there is a lot of work to do with encouraging people to address the need to adopt the five-a-day recommendation, which has such clear health benefits.”
Prof Eileen Gibney, Professor in the School of Agriculture and Food Science, University College Dublin (UCD), said:
“This is an interesting piece of work.
“It attempts to address some of the criticisms of the current dialogue around the topic of ultra processed foods. As the authors state some of the issues raised in relation to the current definitions used in the UPF discussion is that you can have two distinctly different foods – a sweet or ‘candy’ bar (e.g. chocolates / sweets) in the same category as a fortified sugar-free whole grain breakfast cereal. This makes it complicated to use the concept of UPF in nutritional guidance, and nutritional advice. You can’t ask individuals to simply remove all UPF from a diet, as this leaves little choice for the consumer, and would be incredibly hard for people to follow. What we need to do is to understand which processed foods to minimise, and those that are in fact beneficial in a diet.
“The work presented here looks more closely at the ingredients, determining which are processed and not, as well as their known impact on health, it then considers how much added sugar the food contains, and how the combined ingredients impact on health, penalising foods with ingredients which have evidence for increased risk of disease.
“Essentially this scoring system aims to consider the level of processing (by considering the ingredients within the foods), but also considers evidence that links those ingredients with health outcomes. This more nuanced evidenced based approach appears to then discriminate foods that have been processed for benefit (e.g. sugar free fortified breakfast cereal) versus those that do not give any nutritional or health benefit e.g. a chocolate bar.
“This differentiation is important as it means that we are not simply considering the ‘presence of processing’ in a food, as the existing categorization does, but using an evidence based approach, informed by scientific evidence that demonstrates if a processing step, and/or ingredient actually impacts health. Evidence based approaches to the provision of nutritional advice is really important, and underpins our approach to public health. It will be important that this scoring system is updated as and when new evidence is available.”
Prof Helen Roche, Full Professor of Nutrigenomics (Nutrition and ‘Omics’), Director Of Academic Centre – Conway Institute School of Public Health, Physiotherapy and Sports Science, University College Dublin (UCD), said:
“It is an example of nice research which advances the ways we can enhance and improve classification of healthy versus unhealthy foods, based on sound, systematic science, to better inform the consumer. It is very difficult to distinguish processed from non-processed food and their potential impact on health. Take for example lasagne, if you make it yourself at home versus a highly processed version, which by virtue of inferior ingredients and extensive food processing – the end products are very different in terms of nutritional quality. The new classification system proposed WISEcode UPF™ has the potential to more accurately classify processed versus non-processed foods – which when presented in an app might help support consumers choice towards more healthy food options.”
Prof Alexandra Johnstone, Theme Lead for Nutrition, Obesity and Disease, Rowett Institute, University of Aberdeen, said:
“The press release and abstract are very brief and do not allow for this novel research to be assessed for quality or rigor. The experienced US-based research group present a novel scoring system to classify foods and ingredients according to processing and evidence of impact on health, in comparison to the existing NOVA scale which is commonly used to classify UPF. There is very limited description on the validation of the tool and no perspective on limitations of the dataset. For example, this is being presented at a US nutrition meeting and the trademark terms look to be only relevant for the US food system; it is not clear if this is transferrable in other countries. Prior to a peer-review publication, it is difficult to comment further on the translation of the data.”
Dr Amanda Avery, Associate Professor in Nutrition and Dietetics, University of Nottingham, said:
“The NOVA system for classifying foods as ultra-processed or not has served us well since it was first introduced in 2009. But it is time to look for an update given that we know that not all ultra-processed foods are equal and some can contribute to a healthy diet. Also given the ever-increasing number of manufactured food products and increasing level of processing.
“It is unsurprising that AI has been used to create an app with a scoring system using an assessment of ingredients weighted based on current scientific knowledge of the associated health risks, the percentage of calories that come from added sugars, and considerations for ingredients with known health concerns (such as high fructose corn syrup, artificial sweeteners, and salt). Without further information, one assumes that there is also consideration of the positive health benefits of wholegrains providing dietary fibre.
“This scoring system was applied to a large number of foods and many different food ingredients were considered. The USA-based scientists found that the proposed scoring system was better able to differentiate between foods classified as ultra-processed compared to using the NOVA criteria. As one would expect, and hope, there was less differentiation between those foods that are minimally processed.
“It is unlikely that there will ever be a perfect system that accounts for all the nuances that weigh up the risks and benefits of processed food and health. Food manufacturers continue to process food to develop products that are safe and appealing without always considering the wider health impact and of course the health impact is very dependent on how often and how many ultra-processed foods are included in an individual diet. If included occasionally as part of an overall healthy and nutritionally balanced diet, the health risks will be considerably reduced.
“Sadly, whilst such an app may be able to influence healthier food choices, people’s food choices are influenced by a number of factors. Having a greater awareness of the level of processing and ingredients included in a product may not influence choice for everyone. Price for many has a huge influence on the food choices they make, and sadly ultra-processed foods often remain the cheaper option. One exception is that instore brands can often have a better nutritional profile compared to the equivalent branded product and such technology may provide consumers with a greater awareness of this – which is great.
“The abstract being presented is very much describing the development of the app. There does not seem to be any robust evaluation of the use of the app that demonstrates conclusive evidence of the value of the app in improving consumer food choice or the wider health benefits. It would also be good to know if the ability for consumers to be able to compare similar products changes food manufacturing practices to reduce the level of processing and use of artificial ingredients.
“The app has been developed in the USA and whilst a large number of foods and ingredients have been used as part of the development, there are differences in the foods that are available in the UK.”
Abstract title: ‘Ultra-Processed Foods Are Not All Alike: A Novel, Objective Approach to Differentiate Among Processed Foods Including Those Classified As NOVA 4’ by Richard Black et al. It will be presented at the NUTRITION 2025 conference, and is under embargo until 15:00 UK time on Tuesday 3 June 2025.
There is no paper.
Declared interests
Prof Martin Warren: “The Quadram Institute is a UK science national capability strategically supported by the Biotechnology and Biological Sciences Research Council, and also receives funding from other government agencies, national and international charities, and limited funding from industry (six per cent of total funding in 2022/23 came from industry).
Martin’s not got any interests to declare.”
Prof Eileen Gibney: “Eileen R. Gibney is a Professor of Nutrition in University College Dublin, and Director of the UCD Institute of Food and Health. Over the last 5 years she has received research funding through the following; Enterprise Ireland for Technology Centre – Food for Health Ireland (www.fhi.ie) project, co-funded with core partners Carbery, Kerry, Tirlan, Dairygold & Bord Bia; Research Ireland for the Insight Centre for Data Analytics and Co-Centre for Sustainable Food Systems; Horizon Europe most recently in projects such as FNSCloud, PLANEAT and MarieCurie CareerFIT; PhD studentship funding from Société des Produits Nestlé, Switzerland; UCD Foundation and McCarrick Family has provided funding for PhD studentship.
A travel bursary including Registration, Accommodation and Honorarium for attendance and speaking at the Nestle International Nutrition Symposium 2025, was provided by Société des Produits Nestlé, Switzerland.
Eileen R Gibney has completed consultancy work for the following; Société des Produits Nestlé, Switzerland; Irish Advertising Standards Agency, Food Safety Authority of Ireland. No personal payment was received, all payments were made into a research fund through Consult UCD.”
Prof Helen Roche: “I have no conflict of interest with respect to the study I commented on.”
Prof Alexandra Johnstone: “AJ holds voluntary roles within the UK Nutrition Society, Association for the Study of Obesity and British Nutrition Foundation.
FIO Food Grant
https://www.abdn.ac.uk/rowett/research/fio-food/
DIO Food Grant
https://www.abdn.ac.uk/rowett/research/dio-food/.”
Dr Amanda Avery: “Besides my academic position at the University of Nottingham, I also hold a position at Slimming World as Consultant dietitian in the Nutrition, Research & Health Policy team.
I have no other conflicts of interest to declare.”