select search filters
briefings
roundups & rapid reactions
Fiona fox's blog
Scientists comment on a US study investigating colorectal incidence and total ultra-processed food intake, as published in PLOS Medicine.
Dr Rosiered Brownson-Smith, Postdoctoral Researcher, Department of Nutritional Sciences, King’s College London, said:
“This large US cohort study found no association between overall ultra-processed food (UPF) intake and colorectal cancer (CRC). The press release rightly stresses that UPFs are a very varied group of foods. Some contribute whole grains, fibre, dairy and calcium, which are linked to lower CRC risk, while others, such as processed meats, are linked to higher risk. These findings add to the evidence that UPFs should not be considered as a single, uniform category. Different types of UPFs may have different associations with health outcomes, and examining total UPF intake alone may mask these differences.
“The subgroup results fit well with existing evidence. The inverse associations with whole grains, fibre, dairy and calcium, and the positive association with processed meat, match what we already know. What is new is that the whole grain, fibre and dairy associations appear to persist even when the source is ultra-processed. In a statistical model, colorectal cancer incidence was 2% lower among people whose diets contained 10% more minimally processed foods (by weight) in place of ultra-processed foods. That is a very small difference and is unlikely to matter much for individual risk.
“The main caveat is how diet was measured. Intake came from a single food frequency questionnaire (FFQ) at baseline in the mid-1990s. FFQs are poor at distinguishing degrees of processing, and they hold no brand-level or product-level detail. They therefore cannot capture exposure to specific additives such as emulsifiers. Some misclassification is almost inevitable, especially as the food industry reformulates products faster than dietary data or food composition tables are updated. Meat categories can overlap and cooking methods, which strongly influence the carcinogenic potential of meat, are not captured. Fibre is also defined differently across methods, so it is unclear which type of fibre is driving the associations.
“These limitations matter for how the findings are read. A food frequency questionnaire may be picking up general diet quality rather than UPF intake as such. It also cannot properly capture the ‘displacement effect’, where UPFs replace fibre-rich, nutrient-dense foods that protect against CRC. A null result for total UPF here is therefore not evidence that UPFs are harmless.
“The authors also acknowledge that the cohort is older, predominantly white, and was measured once at baseline. The findings should not be generalised to younger people. This is particularly important given the rise in early-onset colorectal cancer, which this study cannot speak to.
“Overall, this is a well-conducted observational study, but it cannot show cause and effect, and it cannot settle whether UPF-derived nutrients are biologically equivalent to those from less processed foods. This study is broadly consistent with existing evidence and current dietary guidance, which recommends eating more fibre-rich foods, whole grains and dairy products, while limiting processed meat intake. However, as an observational study, its findings should not be interpreted as providing definitive evidence about the health effects of UPFs or as a basis for changing dietary recommendations on their own.”
Prof Jules Griffin, Director of the Rowett Institute, University of Aberdeen, said:
“The study by Abar and colleagues is a well conducted analysis of a very large US epidemiology study across 20 years follow-up to examine whether ultra-processed food is associated with the risk of developing colon cancer, with the authors concluding that there is no association with UPF considered as a complete category. The NOVA classification of foods used in the study divides foods into ultra-processed, processed and minimal/no processing categories, and one concern with this classification approach is that foods we consider healthy, for example wholegrain containing breads, can fall into the same category as processed meat products, which high consumption has been shown to be detrimental to health. This study investigates this issue further, finding indeed that foods previously associated with increased risk of colon cancer in the UPF category, such as processed meat are associated with colon cancer, while foods previously highlighted as healthy in the UPF category (whole grain, fibre, dairy, and calcium containing foods) are still associated with reduced risk.
“However, some caution is needed in interpreting these results as the US has one of the highest consumptions of UPF foods, with consumption ranging from 36-50% of total energy consumption in this study, and we know that minimally processed plant-based diets in other countries are associated with much lower colon cancer risks compared with this US population. In the US population most of the wholegrain consumption was associated with UPF foods and it would have been useful to compare with a group of people eating minimally processed wholegrain to see if there were differences. This study also analysed data across twenty years of follow-up, and we know that UPF consumption has increased in both the USA and UK across these time periods, particularly in young adults. Finally, we know from previous studies that diet has an important role in colon cancer risk, and these results shouldn’t be extended to other diseases on the basis of this study. The press release is a very good summary of the research picking out the strengths and limitations of the work.”
Dr Aisling Daly, Lecturer in Nutrition, Oxford Brookes University, said:
“This study is a nice addition to the literature, by providing analysis of subtypes of UPFs and the connections with colorectal cancer, rather than treating all NOVA UPF foods as one homogenous group. By identifying that specific types of UPF are linked with a higher risk of CRC and others are linked with lower risk, this helps ease concerns about UPFs as one homogenous group, and the findings reflect the dietary guidance already provided in relation to foods and nutrients known to be related to colorectal cancer. The authors also rightly note that the study population is limited to older, predominantly white people, which is a limitation of the study’s wider application. However, the finding of an association between red and processed meat and increased colorectal cancer incidence, and of an association between fibre and calcium intake and decreased colorectal cancer incidence, fits in with what we know from previous research – and this study seems to suggest the association is with the nutritional content of the food and the types of food sources rather than its level of processing alone.”
Dr Beverley O’Hara, Lecturer in Public Health Nutrition, Leeds Beckett University, said:
“This study is a great example of how use of the NOVA classification in nutrition research adds nothing to our ability to unpick the health effects of foods. The authors report that the only UPF foods that were associated with a reduced risk of developing colorectal cancer included whole grain and fibre-containing foods, and that meat intake from UPF sources were associated with a higher incidence of colorectal cancer. These associations are already firmly established and so applying the NOVA lens adds nothing new.
“It must also be noted that the dietary assessment method used for this study was not designed to capture ultra-processed food (UPF) intake and is therefore at high risk of food misclassification. Furthermore, food composition databases were not developed to classify foods according to their level of processing. These limitations in both the classification and quantification of UPFs substantially reduce the confidence that can be placed in the findings.
“Applying a food classification system such as NOVA to food intake data retrospectively could be said, in my opinion, to constitute ‘research waste’. In other words, research funds would be much better spent on different kinds of studies that would be better able to understand the relationship between dietary components and risk of developing various cancers, as well as other noncommunicable diseases.”
Prof Gunter Kuhnle, Professor of Nutrition and Food Science, University of Reading, said:
“This is a very comprehensive study investigating the association between the consumption of ultra-processed food (UPF) and the risk of colorectal cancer. Like most studies investigating associations between ultra-processed food intake and health, it has a major limitations that make the results difficult to interpret: the dietary assessment instruments used have not been designed to estimate UPF intake, and this will affect the results. The authors took great care to address this limitation, but it is still a serious limitation and whether the associations found are due to ultra-processed food intake or just a general dietary pattern is difficult to say.
“There is one very important finding that is not affected by the difficult of estimating UPF intake: the study has shown that in contrast to processing and ultra-processing, diet quality is strongly associated with colorectal cancer risk. This is not surprising, as it confirms what has been known before: a diet low in fibre and high in processed meat is associated with an increased cancer risk.”
‘Associations of food processing and dietary risk factors with colorectal cancer incidence in the NIH-AARP Diet and Health Study: A cohort study’ by Leila Abar et al. was published in PLOS Medicine at 19:00 UK time on Thursday 15 October 2026.
DOI: 10.1371/journal.pmed.1005251
Declared interests
Dr Rosiered Brownson-Smith: “None to declare.”
Prof Jules Griffin: “I am a consultant for Sitryx, a company specialising in designing drugs to target immunometabolism.
I have received funding from the European Union to investigate endocrine disrupting chemicals and hold a grant from UK Research and Innovation examining the health benefits of a fish diet.
I have received funding from Shimadzu Ltd to attend the Metabolomics Society to present results from a collaborative project.
I hold shares in GlaxoSmithKline and Haleon plc.”
Dr Aisling Daly: “I have no conflicts of interest to declare.”
Dr Beverley O’Hara: “I have no conflicts of interest to declare.”
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.”
For all other experts, no reply to our request for DOIs was received.