select search filters
briefings
roundups & rapid reactions
Fiona fox's blog
An unpublished study presented at Nutrition 2026 looks at diet quality and food processing levels.
Dr Beverley O’Hara, Lecturer in Public Health Nutrition at Leeds Beckett University, said:
“This is a large study which addresses a very timely question. This study is different from most studies of ‘ultra processed’ food (UPF) because of the way they have broken down dietary components according to whether or not they are plant derived. This gives much more valuable information about diet than can be ascertained by knowing how processed food is.
“Although the researchers were unable to accurately measure intake of UPF (because the original dietary assessment methods were not designed to do this) the study provides good evidence that it is overall diet quality, not processing, that is important for health.
“Most large studies of this nature have inherent limitations in the way that they measure what people actually eat. Nevertheless, the results clearly show that healthy plant-based diets are strongly protective against type 2 diabetes, whether the constituents of the diet are processed or not. The results also show that both processed and unprocessed healthy plant-based diets are protective against cardiovascular disease.
“These are important findings because they may help people to feel less anxious about eating the processed foods that are affordable, perfectly healthy and part of everyday life. Another problem is that consumers sometimes avoid plant-based meat and dairy alternatives due to worries about ultra processing. These beliefs are impacting the dietary shift towards more sustainable diets.
“The health-protective effects of plant-based foods are probably due to the different types of fibre and the various micronutrients and non-nutrient chemicals that can be found in plants (sometimes called ‘nutritional dark matter’). Consuming a wide range of these foods is important for dietary quality, however achieving this can be challenging, meaning that purchasing them in packaged form may sometimes be a convenient and practical choice.
“Although this is just a conference abstract, the study appears to be an important counterbalance to the large number of poor-quality studies that have led to widespread scaremongering about UPF.”
Dr Aisling Daly, Senior Lecturer in Nutrition at Oxford Brookes University, said:
“There are not enough details presented in the abstract to explain how the six diet indices were developed, so it is unclear what factors were included. To note, the study specifically focussed on plant-based diets, but this is not clear from the press release article title or initial sentences. Additionally, the graphic includes both plant and animal products, which seems unusual since the abstract mentions only plant-based diets. I would be interested to know if any comparison with non-plant based diet indices were conducted, or why not, since this would help add further weight to the discussion around nutrient content or processing levels.
“Another unclear aspect is that the abstract first states 6 diet indices were developed, but the press release specifies only 4 diet indices – this discrepancy is unclear and perhaps should be rectified to make the press article more accurately reflect the research.
“However, it is very interesting to see data being presented which takes into account both elements of food in terms of nutrient quality and processing levels. One of the main criticisms of the NOVA classification system is that it doesn’t distinguish between healthy processing, such as fortification of products with micronutrients, or unhealthy processing, such as the addition of excessive salt or fat for preservation and taste reasons. Having a revised classification system would be a helpful addition to the research literature, so I would be interested in reading the full paper when it is published to see how the healthy and unhealthy diet indices were determined. As the author comments, advising people to blanketly avoid processed foods is not a helpful approach, so having a way to help consumers balance both processing level and nutrient quality would be a beneficial public health strategy, which would likely improve health outcomes, as indicated from this study.
Does this mean it’s not actually the level of processing that matters, but the ingredients? Why?
“Yes, I think this research suggests that we have placed too much emphasis on processing level, without recognising that some processing is beneficial to our health, and other processing is not directly harmful to our health. However there are many processed food products available which are less nutritious and less beneficial to our health, but applying the idea that ‘all processed foods are bad’ is neither accurate, helpful or realistic.
Does it show that a diet of unprocessed-but-unhealthy ingredients (presumably HFSS, low fibre) is worse than one of processed-but-healthier ones? What does ‘processed-but-healthier’ look like?
“Yes, the study creates a four way split rather than a two way split of “ultra processed vs unprocessed” commonly reported in the media, showing that the overall combination of foods in our diet is more important than solely the level of processing. Processed foods are needed in the diet, and for many people they are essential for convenience, cost-efficiency, and food preservation reasons. The current narrative that ‘processed foods are bad’ is not accurate and is damaging and worrying for people who need to rely on these foods. The concept of processed foods is not the issue, rather the quality of the processed foods being purchased should be considered – for example we encourage people to include tinned beans in their diet for protein and fibre benefits, while also being more sustainable environmentally, affordable, having a long shelf life and being simple to cook. If we only apply the processed food lens, these tins of beans would be ‘bad’, but when we combine processing level with nutrient content, and other practical concerns, these tins of beans are now ‘good’. It is also important to note the NOVA classification has four levels of processing and the ultra processed foods are those of most concern throughout the research literature. However, oftentimes fortifying foods with micronutrients for health benefits means these foods become ‘UPFs’. This study helps to break down this siloed mindset around UPFs, which I believe is a very welcome addition to both the research arena and the public health promotion arena. I would be very interested to read the full article to understand how their diet indices were developed. Additionally, I would like to see this research extended to include animal products in the same way, or to understand why animal products were not included in the present study.”
Seamus Higgins, Associate Professor in Food Process Engineering at the University of Nottingham, said:
“This is an interesting study because it attempts to separate diet quality from food processing—two factors that are often closely linked and remain the subject of considerable scientific debate. It reinforces the view that a healthy dietary pattern remains one of the strongest predictors of long-term health. However, these are preliminary findings from a conference abstract rather than a peer-reviewed paper and should therefore be interpreted with appropriate caution.
“The study suggests that processing alone is not the whole story, but neither does it demonstrate that processing can be disregarded. Food structure, formulation, nutrient quality, the sensory characteristics of food—including taste, texture and aroma—and the biological and psychological drivers of food choice are all closely intertwined, making it difficult to attribute health outcomes to any single characteristic.
“A further limitation is that the analysis focused on the plant-derived components of participants’ diets rather than the whole diet. As animal-derived foods make a substantial contribution to typical US dietary patterns, caution is needed before generalising these findings to overall diets.
“Likewise, reducing food to a single measure—whether calories, fat, sugar or processing—risks oversimplifying the far more complex biological responses to different foods and eating behaviours.”
Prof Alberto Fiore, Professor in Food Chemistry and Technology at Abertay University, said:
“This is exactly the kind of evidence the field has needed, and it is considerably stronger than most UPF research to date because the researchers deliberately designed the study to separate diet quality from processing level, rather than conflating the two as most prior cohort studies have done. Following over 200,000 adults across three long-running US cohorts for decades gives this study real statistical power, and the four-quadrant design — cross-referencing healthy versus unhealthy with ultra-processed versus minimally-processed — is a genuinely elegant way to isolate the variable that actually matters.
“The result is unambiguous and important: within each diet-quality group, health outcomes were ‘almost identical regardless of whether the diet was ultra-processed or minimally-processed’. That single finding does more to clarify this debate than years of headline-grabbing observational studies that simply reported a raw association between NOVA category 4 intake and disease risk without ever checking whether nutrient quality was doing all the work. It confirms directly what has long been suspected: a fortified wholegrain cereal and a can of cola should never have been treated as equivalent health risks simply because both happen to sit in the same processing category, and this study now has the data to demonstrate that error empirically rather than just argue it conceptually.
“The one caveat worth flagging is that this is a conference abstract presented at NUTRITION 2026, not yet a peer-reviewed publication, and the analysis focused specifically on plant-derived dietary components rather than the full diet — so some caution is warranted until the full paper is published and independently reviewed. But even with that caveat, this is a well-designed, adequately powered study that directly tests the hypothesis rather than assuming it, and its conclusion is the right one: healthy diets lower risk and unhealthy diets raise it, regardless of processing level. Dietary guidance should follow the data and refocus on nutrient quality — through frameworks like the Mediterranean diet or plant-based diet indices — rather than continuing to police processing level as a proxy for health.”
Declared interests
Beverley O’Hara: “Conflict of interests: none. Food industry funding: none. Advisory roles: none. Committee membership: Yorkshire network lead for the Association for the Study of Obesity.”
Alberto Fiore: “I confirm that I do not have any conflict of interest, and I do not know the authors and not getting paid by any industry.”
Seamus Higgins “previously held technical and leadership roles within the food industry, including with Nestlé and Premier Foods. Since joining the University of Nottingham approximately ten years ago, he has had no employment, consultancy, advisory, financial, or other commercial relationships with these companies and reports no competing interests in relation to this article.”
Aisling Daly: “I have no conflicts of interest to declare. I am a Senior Lecturer in Nutrition at Oxford Brookes University and a Registered Nutritionist with the Association for Nutrition.”
For all other experts, no reply to our request for DOIs was received.