Ultra-processed food exposure and adverse health outcomes: umbrella review of epidemiological meta-analyses
BMJ 2024; 384 doi: https://doi.org/10.1136/bmj-2023-077310 (Published 28 February 2024) Cite this as: BMJ 2024;384:e077310Linked Editorial
Reasons to avoid ultra-processed foods
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
Dear Editor,
We have concerns about the recently published umbrella review of 45 pooled analyses of NOVA ultra-processed food (UPF) exposure on health outcomes.[1] Using the GRADE framework, 22 of the meta-analyzed studies were defined of low quality and 19 of very low quality. Despite this, the authors suggest as “public health measures to target and reduce dietary exposure to ultra-processed foods for improved human health.” How can food policies be implemented based on low quality data and on a highly heterogeneous food classification? When hundreds of evaluators were asked to classify with NOVA 231 foods (52% with specified ingredients), only one food was assigned to the same NOVA category by all evaluators [2] and these were nutrition/food science experts!
According to NOVA a food with 5 industrial ingredients would automatically become UPF. Does it really make a difference on health outcomes should the food contain 5 industrial ingredients or 4? Assuming it does, would it depend on type and quantity of those ingredients? Are ascorbic acid, tocopherols, propionic acid, pectin, guar, detrimental to health when added to foods? These are considered safe for human consumption, some have food preserving properties while others may improve palatability, however foods with these ingredients would fall into NOVA-UPF.
Although there are unhealthy UPFs with low nutrient-high energy density or with an unbalanced macronutrient profile, there are also “healthy UPFs”. Foods which fall in NOVA-UPF include many healthy plant-based foods that can substitute animal-based options: legume/soya burgers, soya drinks/yogurts, packaged whole wheat and whole grain breads and breakfast cereals, baked canned beans, peanut butter and even stuffed olives and hummus. These foods can make up to 40% of vegan diets in high income countries and 1/3 of plant-based diets [3] which are considered healthy for us and for our planet.[4] Interestingly, in high income nations, the more vegetarian the diet is, the higher its UPF content.[3] Many of the vegetarian “healthy UPFs” are main contributors to dietary fiber helping to reach the daily recommended intakes of 14g/1000 kcal diet,[5] an amount that the average American adult is still struggling to achieve.[6] In three large US cohorts, whole grains UPFs were inversely associated with risk of type 2 diabetes [7] and in a European cohort, plant-based UPFs were inversely associated with cancer risk.[8] These “healthy UPFs” may contain also carbohydrates of low glycemic index, antiatherogenic unsaturated fatty acids and soy proteins which together contribute to reduce hyperglycemia, hypercholesterolemia, heart disease, type 2 diabetes [9-13] and cancer risk.[14-16] Soy-based meat and dairy alternatives that fall into NOVA-UPFs have similar nutritional values as the unprocessed animal-based counterparts but contain more dietary fiber, less saturated fats and calories,[17] were proved to be cholesterol-lowering in clinical trials [18] and reduced breast cancer recurrence [19] owing to their isoflavone content.[20]
Finally, should NOVA-UPF be country and dataset specific? Pizza may be considered UPFs by NOVA in some countries and in more recent datasets. However, in Italy pizza is generally consumed fresh and contains healthy ingredients, i.e. olive oil, tomatoes, capers, olives, green leafy vegetables, zucchini, eggplants, peppers, onions, garlic, oregano and basil. These are amongst the best contributors to a low dietary inflammatory index which associates with lower disease outcomes.[21 22] How is it possible that a food is at the top of one diet chart and doomed in another? Do we want to reduce the dietary exposure of these proven-healthy foods/ingredients? Would they be replaced with meat? Would we contribute to worse public health at a time when more than ever we need healthy diets that are also planet friendly? Are we creating more confusion and less trust in health professionals?
Livia S.A. Augustin 1, PhD and Carlo La Vecchia 2, MD
1 Epidemiology and Biostatistics Unit, Istituto Nazionale Tumori - IRCCS - Fondazione G. Pascale, Napoli, Italy; 2 Department of Clinical Sciences and Community Health, Universita’ degli Studi di Milano, Milan, Italy.
References
1. Lane MM, Gamage E, Du S, et al. Ultra-processed food exposure and adverse health outcomes: umbrella review of epidemiological meta-analyses. BMJ 2024;384:e077310.
2. Braesco V, Souchon I, Sauvant P, et al. Ultra-processed foods: how functional is the NOVA system? European journal of clinical nutrition 2022;76(9):1245-53.
3. Gehring J, Touvier M, Baudry J, et al. Consumption of Ultra-Processed Foods by Pesco-Vegetarians, Vegetarians, and Vegans: Associations with Duration and Age at Diet Initiation. The Journal of nutrition 2021;151(1):120-31.
4. Willett W, Rockstrom J, Loken B, et al. Food in the Anthropocene: the EAT-Lancet Commission on healthy diets from sustainable food systems. Lancet 2019;393(10170):447-92.
5. Augustin LSA, Aas AM, Astrup A, et al. Dietary Fibre Consensus from the International Carbohydrate Quality Consortium (ICQC). Nutrients 2020;12(9).
6. Cifuentes L, O'Keefe S. Analysis of 1999-2017 NHANES Data: Minimal Increase and Racial Disparities in U.S. Fiber Consumption Over 18 Years. Nutrition and cancer 2024;76(4):345-51.
7. Chen Z, Khandpur N, Desjardins C, et al. Ultra-Processed Food Consumption and Risk of Type 2 Diabetes: Three Large Prospective U.S. Cohort Studies. Diabetes care 2023;46(7):1335-44.
8. Cordova R, Viallon V, Fontvieille E, et al. Consumption of ultra-processed foods and risk of multimorbidity of cancer and cardiometabolic diseases: a multinational cohort study. The Lancet regional health Europe 2023;35:100771.
9. Glenn AJ, Guasch-Ferre M, Malik VS, et al. Portfolio Diet Score and Risk of Cardiovascular Disease: Findings From 3 Prospective Cohort Studies. Circulation 2023;148(22):1750-63.
10. Jenkins DJA, Willett WC, Yusuf S, et al. Association of glycaemic index and glycaemic load with type 2 diabetes, cardiovascular disease, cancer, and all-cause mortality: a meta-analysis of mega cohorts of more than 100 000 participants. The lancet Diabetes & endocrinology 2024;12(2):107-18.
11. Jenkins DJ, Mirrahimi A, Srichaikul K, et al. Soy protein reduces serum cholesterol by both intrinsic and food displacement mechanisms. The Journal of nutrition 2010;140(12):2302S-11S.
12. Anderson JW, Bush HM. Soy protein effects on serum lipoproteins: a quality assessment and meta-analysis of randomized, controlled studies. Journal of the American College of Nutrition 2011;30(2):79-91.
13. Messina M. Soy and Health Update: Evaluation of the Clinical and Epidemiologic Literature. Nutrients 2016;8(12).
14. Giovannucci E. Insulin and colon cancer. Cancer causes & control : CCC 1995;6(2):164-79.
15. Augustin LS, Dal Maso L, La Vecchia C, et al. Dietary glycemic index and glycemic load, and breast cancer risk: a case-control study. Annals of oncology : official journal of the European Society for Medical Oncology / ESMO 2001;12(11):1533-8.
16. Turati F, Galeone C, Augustin LSA, et al. Glycemic Index, Glycemic Load and Cancer Risk: An Updated Meta-Analysis. Nutrients 2019;11(10).
17. Messina M, Sievenpiper JL, Williamson P, et al. Perspective: Soy-based Meat and Dairy Alternatives, Despite Classification as Ultra-processed Foods, Deliver High-quality Nutrition on Par with Unprocessed or Minimally Processed Animal-based Counterparts. Advances in nutrition 2022;13(3):726-38.
18. Blanco Mejia S, Messina M, Li SS, et al. A Meta-Analysis of 46 Studies Identified by the FDA Demonstrates that Soy Protein Decreases Circulating LDL and Total Cholesterol Concentrations in Adults. The Journal of nutrition 2019;149(6):968-81.
19. Chi F, Wu R, Zeng YC, et al. Post-diagnosis soy food intake and breast cancer survival: a meta-analysis of cohort studies. Asian Pacific journal of cancer prevention : APJCP 2013;14(4):2407-12.
20. Becerra-Tomás N, Balducci K, Abar L, et al. Postdiagnosis dietary factors, supplement use and breast cancer prognosis: Global Cancer Update Programme (CUP Global) systematic literature review and meta-analysis. International Journal of Cancer 2023;152(4):616-34.
21. Shivappa N, Godos J, Hebert JR, et al. Dietary Inflammatory Index and Cardiovascular Risk and Mortality-A Meta-Analysis. Nutrients 2018;10(2).
22. Shivappa N, Bonaccio M, Hebert JR, et al. Association of proinflammatory diet with low-grade inflammation: results from the Moli-sani study. Nutrition 2018;54:182-88.
Competing interests: LSA is a founding member of the International Carbohydrate Quality Consortium (ICQC), has received honoraria from the Nutrition Foundation of Italy (NFI) and research grants from LILT, a non-profit organization for the fight against cancer. CLV serves in the scientific board of the International Sweeteners Association (ISA).
Dear Editor
This study found through an umbrella review that higher intake of ultra-processed foods was associated with higher risk of multiple adverse health outcomes, particularly cardiometabolic, common psychological disorders, and mortality risk. Although the Meta-analysis search for inclusion was not particularly recent, and a portion of the literature and other health conditions maybe not included [1-2], it was overall a fruitful and large study.
From individual studies, to systematic evaluations and Meta-analyses, to umbrella review, research on the association between ultra-processed foods and health conditions has received sustained attention in recent years. Is the correlation derived from these studies a result of the unhealthy ingredients such as higher salt, oil and sugar in them, or is it a result of the refinement in the processing and preparation process? In fact, such scientific hypotheses have been reported separately, but why is it that ultra-processed foods have become a category of foods where the mechanisms of health effects become unclear?
In fact, shortly before the publication of this article (12 days), I was concerned about the publication of another umbrella evaluation of "Ultra-processed food consumption and human health:an umbrella review of systematic reviews with meta-analyses" [3]. We often say, "There is no junk food, only irrational diet"; perhaps, ultra-processed foods are not a beast or with no benefit, it may not be appropriate to treat them like pollutants with "exposure" to describe them.
Public health policies and food businesses need to work together to build a safe, nutritious and healthy food environment. At the same time, necessary food processing is essential for human health and sustainable social development, and the rational use of risk-assessed food additives is not likely to cause health harm. Therefore, the "pros and cons" of ultra-processed foods and their "scope" are still being explored. Only through scientific understanding of ultra-processed foods, avoiding "over-fine" processing, reducing the production of high-energy and low-nutrient foods, and reasonably adjusting their proportions in the dietary pattern, is it a wise move in the context of the "food crisis", the "macro-food concept", and the "reduction revolution of salt, oil and sugar".
1. Taneri PE, Wehrli F, Roa-Díaz ZM, Itodo OA, Salvador D, Raeisi-Dehkordi H, Bally L, Minder B, Kiefte-de Jong JC, Laine JE, Bano A, Glisic M, Muka T. Association Between Ultra-Processed Food Intake and All-Cause Mortality: A Systematic Review and Meta-Analysis. Am J Epidemiol. 2022 Jun 27;191(7):1323-1335. doi: 10.1093/aje/kwac039. PMID: 35231930.
2. Cascaes AM, Silva NRJD, Fernandez MDS, Bomfim RA, Vaz JDS. Ultra-processed food consumption and dental caries in children and adolescents: a systematic review and meta-analysis. Br J Nutr. 2022 Jul 27:1-10. doi: 10.1017/S0007114522002409. Epub ahead of print. PMID: 35894293.
3. Barbaresko J, Bröder J, Conrad J, Szczerba E, Lang A, Schlesinger S. Ultra-processed food consumption and human health: an umbrella review of systematic reviews with meta-analyses. Crit Rev Food Sci Nutr. 2024 Feb 16:1-9. doi: 10.1080/10408398.2024.2317877. Epub ahead of print. PMID: 38363072.
Competing interests: No competing interests
Dear Editor
Both the article “Ultra-processed food exposure and adverse health outcomes…” [1] and the Editorial ”Reasons to avoid ultra-processed foods”[2] help a lot to understand the dynamics and harmful effects deriving from the consumption of Ultra-processed foods (UP).
I would like to focus on a further point concerning the possibility of pitfalls related to some methods to produce food, even those commonly considered non-UP.
The Maillard reaction is a nonenzymatic reaction that occurs in organisms and food. Catalyzed by heat, this leads to the generation of dark pigments, flavors, and odors related to the cooking, storage, and nutritional value of foods, especially those with high protein content [3].
Among the great variety of molecules or compounds that are formed in this reaction are the advanced glycation end products (AGEs). Particularly, some AGEs such as N-carboxymethyl-lysine, pyrraline, and pentosidine are highly related to binding to cell receptors, receptor of advanced glycation end products (RAGE), leading to chronic degenerative diseases such as diabetes mellitus, Alzheimer's, high blood pressure, and atherosclerosis, among others. Recently, natural compounds such as polyphenols (flavonoids, nonflavonoids, and phenolic acids) have been associated with inhibitory effects on AGE formation and blocking AGE-RAGE formation [3].
The interaction between RAGE and its ligands mainly results in a pro-inflammatory response and can lead to stress events often favoring mitochondrial dysfunction or cellular senescence [4].
The relationship between harmful Maillard reaction products in dairy processing and many human chronic diseases have gradually come to the fore. Various harmful Maillard reaction products such as lactulosyl-lysine (furosine), furfurals, and advanced glycation end products (AGEs) could be formed during the thermal processing of dairy products, which could lead to various chronic diseases [5].
Food Additives in bread (Emulsifiers, Surfactans, Potassium bromate, and others).
Ingredients such as dough conditioners, crumb softeners, emulsifiers, and surfactants can be added to enhance bread quality [6]. For some of these additives, such as Potassium bromate, there are many reports elucidating its negative impact on human health (potential human carcinogen) [7].
Reflection in conclusion:
It is therefore evident that the altered quality diet, caused by various food production procedures, favors the onset of a series of pathological conditions in the human body that require diagnostic pathways and therapeutic treatments with exorbitant costs.
Therefore, it can be thought that sensitizing patients to adopt both precautions in cooking and in food choices is a valid means of reducing the economic and social costs derived from many degenerative diseases.
REFERENCES :
[1] Ultra-processed food exposure and adverse health outcomes: umbrella review of epidemiological meta-analyses BMJ 2024; 384 doi: https://doi.org/10.1136/bmj-2023-077310 (Published 28 February 2024)
[2] Reasons to avoid ultra-processed foods BMJ 2024; 384 doi: https://doi.org/10.1136/bmj.q439 (Published 28 February 2024)
[3] Sergio Luis Valle-SánchezRoberto, Rodríguez-Ramírez , et al. : Natural inhibitory compounds of advanced glycation end products (AGEs) from the Maillard reaction Chapter 9 in: Studies in Natural Products Chemistry, 7 October 2023. https://www.sciencedirect.com/science/article/abs/pii/B9780443189616000184
[4] Teissier T, Boulanger É. The receptor for advanced glycation end-products (RAGE) is an important pattern recognition receptor (PRR) for inflammaging. Biogerontology. 2019 Jun;20(3):279-301. doi: 10.1007/s10522-019-09808-3. Epub 2019 Apr 9. PMID: 30968282.
[5] Li M, Shen M, Lu J, Yang J, Huang Y, Liu L, Fan H, Xie J, Xie M. Maillard reaction harmful products in dairy products: Formation, occurrence, analysis, and mitigation strategies. Food Res Int. 2022 Jan;151:110839. doi: 10.1016/j.foodres.2021.110839. Epub 2021 Dec 2. PMID: 34980378.
https://doi.org/10.1016/j.foodres.2021.110839
[6] Vargas MCA, Simsek S. Clean Label in Bread. Foods. 2021 Aug 31;10(9):2054. doi: 10.3390/foods10092054. PMID: 34574163; PMCID: PMC8466822.
[7] Shanmugavel V, Komala Santhi K, Kurup AH, Kalakandan S, Anandharaj A, Rawson A. Potassium bromate: Effects on bread components, health, environment and method of analysis: A review. Food Chem. 2020 May 1;311:125964. doi: 10.1016/j.foodchem.2019.125964. Epub 2019 Dec 9. PMID: 31865111.
Competing interests: No competing interests
Dear Editor
A frequent characteristic of ultra-processed foods is their poor mechanical consistency, which does not require vigorous and prolonged chewing activity, as is necessary for natural foods.
It is therefore easy to deduce that instinctively the individual is not actually aware of having reached the right nutritional requirement (risk of overnutrition, often of sub-optimal quality).
Transglutaminases are widely used in several industrial processes, including the food and pharmaceutical industries.
In commercial food processing of meat, fish, dairy, and baking products [1], Transglutaminase modifies the functional properties of food proteins by incorporation of amines, crosslinking, deamidation, and bonding surfaces of foods.
Production sources of commercial transglutaminase are transglutaminases of mammals, non-mammals, Microbial transglutaminase (mTG) and recombinant transglutaminase [2].
Multiple mTG linked proteins are immunogenic in patients with celiac disease. In the study conducted by Lerner and Matthias [3], the authors indicate that the use of this enzyme can further increase antigenic load presented to the immune system and increase the risk for gluten-sensitive populations. In a recent study, Matthias et al. [4] have suggested that mTG increases immunogenicity in children with celiac disease because mTG antibodies correlate to intestinal damage in the same degree as transglutaminase human tissue antibodies. The authors suggested that further investigation is necessary to elucidate the role of anti-mTG antibodies in this disease [4].
Other articles further confirm this hypothesis. mTG, being a protein's glue, by cross-linking it creates neoepitope complexes that are immunogenic and potentially pathogenic in celiac disease. Despite low sequence identity, it imitates functionally its family member, the endogenous tissue transglutaminase, which is the autoantigen of celiac disease [5,6].
Could we therefore think that using Transglutaminase in the production of ultra-processed foods (also including Gluten Free) might induce cases of celiac disease?
REFERENCES:
[1] Strop P. Versatility of microbial transglutaminase. Bioconjug Chem. 2014 May 21;25(5):855-62. doi: 10.1021/bc500099v. Epub 2014 Apr 17. PMID: 24694238.
[2] Duarte, L., Matte, C.R., Bizarro, C.V. et al. Review transglutaminases: part II—industrial applications in food, biotechnology, textiles and leather products. World J Microbiol Biotechnol 36, 11 (2020). https://doi.org/10.1007/s11274-019-2792-9
[3] Lerner A, Matthias T (2015) Possible association between celiac disease and bacterial transglutaminase in food processing: a hypothesis. Nutr Rev 73:544–552. https://doi.org/10.1093/nutrit/nuv011
[4] Matthias T, Jeremias P, Neidhöfer S, Lerner A (2016) The industrial food additive, microbial transglutaminase, mimics tissue transglutaminase and is immunogenic in celiac disease patients. Autoimmun Rev 15:1111–1119. https://doi.org/10.1016/j.autrev.2016.09.011
[5] Aaron L, Torsten M. Microbial transglutaminase: A new potential player in celiac disease. Clin Immunol. 2019 Feb;199:37-43. doi: 10.1016/j.clim.2018.12.008. Epub 2018 Dec 10. PMID: 30543926.
[6] Lerner A, Matthias T. Processed Food Additive Microbial Transglutaminase and Its Cross-Linked Gliadin Complexes Are Potential Public Health Concerns in Celiac Disease. Int J Mol Sci. 2020 Feb 8;21(3):1127. doi: 10.3390/ijms21031127. PMID: 32046248; PMCID: PMC7037116.
Competing interests: No competing interests
Dear Editor,
The abundance and cultural acceptance of ultra-processed food has been a worrying phenomenon which doesn’t affect society equally. Literature has shown that those living in socioeconomic deprivation are more likely to consume ultra processed foods. (1) The social, environmental, and political factors influencing ultra processed food consumption are all too often outside the individual's control. Focusing on individual interventions such as food labelling is unlikely to have significant impact.(2)
The population already struggle to meet current dietary recommendations with rates of ‘5-a-day’ uptake ranging from 10.6%-52% of adults in England, unsurprisingly deprivation has a role to play with uptake lower in the unemployed. (3) Expecting individuals to take on board yet another dietary recommendation whilst also manging competing priorities such as time and resources to prepare meals, and access and affordability to healthier foods is misguided.
A focus on these upstream factors is essential rather than once again blaming the individual for the choices they are never truly making themselves. Considering the traditional social determinants of disease is not sufficient to overcome this health risk. We need to focus on the commercial determinants, pushing companies to prevent production of these products and ensure that these products are not targeted at particular population groups. The Global Burden of Disease estimated that 19 million global deaths a year are from tobacco, alcohol, ultra-processed food, and fossil fuels; all of which have are heavily influenced by commercial sector practices.(4)
The pathways through which commercial sector practices can influence health are complex.(4) Corporations hold great power in shaping social norms with access to resources that far outweigh those of traditional public health bodies. Use of political frameworks is needed to effectively overcome these barriers. We’ve seen the power of legislation to regulate the tobacco industry.(5) The public health community need to come together across systems to allow a unified message to corporations that current practices are unacceptable and our populations deserve better.
1- Marchese L, Livingstone KM, Woods JL, Wingrove K, Machado P. Ultra-processed food consumption, socio-demographics and diet quality in Australian adults. Public Health Nutrition. 2022;25(1):94–104. doi:10.1017/S1368980021003967
2- Thomson K, Hillier-Brown F, Todd A, McNamara C, Huijts T, Bambra C. The effects of public health policies on health inequalities in high-income countries: an umbrella review. BMC Public Health. 2018;18(1):869
3- OHID. Public health profiles: Public Health Outcomes Framework. Updated May 2023. Accessed: 20/03/24
4- Gilmore AB, Fabbri A, Baum F, Bertscher A, Bondy K, Chang HJ, Demaio S, Erzse A, Freudenberg N, Friel S, Hofman KJ, Johns P, Abdool Karim S, Lacy-Nichols J, de Carvalho CMP, Marten R, McKee M, Petticrew M, Robertson L, Tangcharoensathien V, Thow AM. Defining and conceptualising the commercial determinants of health. Lancet. 2023 Apr 8;401(10383):1194-1213. doi: 10.1016/S0140-6736(23)00013-2. Epub 2023 Mar 23. PMID: 36966782.
5- Hill SE, Johns P, Nakkash RT, Collin J. From silos to policy coherence: tobacco control, unhealthy commodity industries and the commercial determinants of health. Tob Control. 2022 Mar;31(2):322-327. doi: 10.1136/tobaccocontrol-2021-057136. PMID: 35241606.
Competing interests: No competing interests
Dear Editor,
Good Day. Vanakkam.
Thank you very much for that interesting article.
Ultra processed Food is now a Scope creep.
There is clear need of Ultra processed food under certain trying physical and medical conditions. Current widespread use is a Scope creep. However this rampant widespread use of Ultra processed Food by people of all ages all over the world is a major health concern all over the world. Numerous publications have attested to the harmful effects of these Foods.
Ultra processed foods provide the customer with comfort, taste & convenience at the cost of nutrition and health. The preservatives which prevent the growth of microbes & extend the shelf life of these products are unlikely to nourish the trillions of microbes in our guts nor our 37 trillion cells. The earlier we move away from them, the healthier we will become.
Professor Dr Pandiyan Natarajan,
Professor Emeritus,
The Tamil Nadu Dr MGE Medical University,
Chennai, India,
Chief Consultant in Andrology and Reproductive Medicine,
Chettinad Super Speciality Hospital (Retired)
Nova IVF Fertility. Apollo 24/7
Competing interests: No competing interests
Dear Editor,
This paper rightly pointed out the dangers that can be observed from excessive consumption of ultra-processed food and the difficulties defining this for future public health policy.
A key component of this future policy must be a fairness within socioeconomic classes for access of a balanced, healthy diet.
We know that consumption of ultra-processed food is higher in lower socio-economic classes (1), and that this is due to a combination of factors including, but not limited to, advertising aims and accessibility to healthier alternatives (1, 2).
Previous public health policy has limited the options of these groups, which partly allows for expansion of alternative food-stuffs, such as the beginning of ultra-processed food (3). This article suggests some of these policies as inspiration to curb ultra-processed food intake throughout the country, such as improved labeling. These strategies are known to benefit lower socio-economic classes less, in comparison to other possible strategies (3).
Alternatives to ultra-processed foods, and the resources to create healthier meals, must become more accessible, not simply restricting a cheaper food-option.
References:
(1) Coyle, D.H., Huang, L., Shahid, M. et al. Socio-economic difference in purchases of ultra-processed foods in Australia: an analysis of a nationally representative household grocery purchasing panel. Int J Behav Nutr Phys Act 19, 148 (2022). https://doi.org/10.1186/s12966-022-01389-8
(2) Baraldi LG, Martinez Steele E, Canella DS, et alConsumption of ultra-processed foods and associated sociodemographic factors in the USA between 2007 and 2012: evidence from a nationally representative cross-sectional studyBMJ Open 2018;8:e020574. doi: 10.1136/bmjopen-2017-020574
(3) Thomson K, Hillier-Brown F, Todd A, McNamara C, Huijts T, Bambra C. The effects of public health policies on health inequalities in high-income countries: an umbrella review. BMC Public Health. 2018;18(1):869
Competing interests: No competing interests
Dear Editor
A pervasive issue in health research is overlooking uncertainty in interpreting findings. Lane et al employ the GRADE methodology, which is designed to evaluate the certainty of evidence. Using this they judged the majority of evidence in this area to be of "very low" or "low" certainty, suggesting a significant degree of uncertainty regarding the true impact of UPF (as defined by the authors) on the health outcomes assessed. This critical aspect of scientific inquiry—acknowledging the limits of our confidence in the data—is often glossed over, leading to potentially misleading conclusions about the strength of the evidence. Despite assessing it, Lane et al. fail to adequately communicate the highly uncertain findings in their conclusions. This raises the question: why assess uncertainty if it is not then communicated effectively?
Secondly, introducing a "credibility" method alongside GRADE to evaluate the evidence poses additional challenges. This method aims to assess the reliability of findings from meta-analyses but lacks thorough validation for its effectiveness. Without a robust framework to establish its credibility, its application may introduce inconsistencies in how evidence is interpreted and presented. The reliance on such an unvalidated method undermines the rigour of the review process, casting doubt on the conclusions drawn about the relationship between UPF consumption and adverse health outcomes.
The composition of the review panel for the BMJ paper has (rightly) raised concerns. [1] One reviewer's limited research background as a naturopath, especially regarding UPF and related methodologies, combined with seemingly non-impartial views on UPF, [2] coupled with the inclusion of the creator of the NOVA classification system who also wrote the accompanying editorial for this paper—compromises the objectivity and thoroughness of the peer review process.
Reflecting on the BMJ's historic role in championing evidence-based medicine, I'm compelled to express my concerns about a potentially systematic issue of ignoring uncertainty. Another systematic review published in the BMJ in February 2024, utilising the GRADE methodology, identified evidence of very low to low certainty and again overlooked [3] this uncertainty in its conclusions. This trend towards overconfidence in findings necessitates a return to a more transparent and cautious approach in research communication, ensuring that discussions and conclusions faithfully represent the uncertainty and limitations of the evidence. Addressing this is vital for upholding the BMJ's legacy in leading evidence-based medicine and ensuring health decisions and policies are informed by fair and reliable assessments of the best available evidence.
1. McCartney, M. [@mgtmccartney]. (2023, March 7). ‘do you have links to the pieces?oh my.’ [Post]. X.
2. Work.Life. (2016, September 7). Member Spotlight: Katie Ruane from Harley Street Naturopath.
3. Nunan, D. [@dnunan79]. (2024, February 16). For example - in the news study ALL the evidence was low or very low Certainty (quality) of Evidence (CoE). [Post]. X.
Competing interests: No competing interests
Dear Editor,
The formation of Action on Smoking and Health (ASH) by the Royal College of Physicians in 1971 to lobby and campaign for tobacco control was a major step forward in public health. Since poor diet has overtaken smoking as the leading killer in many western countries, is it now time for a coalition against UPFs along the lines of ASH?
The case for such a body is further advanced by the new umbrella review of observational studies by Lane and colleagues demonstrating that UPF consumption is associated with an increased risk of a wide variety of chronic diseases [1]. Although the studies do not prove causality, the likelihood for this is greatly strengthened by the consistency of the data (93% of the pooled analyses suggest harm from UPFs) and by the many biologically plausible mechanisms outlined by the authors. These are some of the criteria first delineated by Austin Bradford Hill for helping establish causality in observational studies, and who, along with Richard Doll, pioneered the links between smoking and lung cancer.
The rationale for action against UPFs is strong. As the main source of sales for many multi-national food companies in the UK, HFSS (foods high in fat, sugar, salt) - most of which are UPFs - are heavily promoted and vigorously defended [2]. Links between the food industry and members of the Scientific Advisory Committee on Nutrition advising on UPFs is just one of the many examples where the food industry has undesirable links with academics, nutrition organisations and policy makers [3, 4].
Consumers and health professionals alike are confused [5]. There is a clear need for a body free of conflicts of interest that can raise awareness, advise on alternative dietary strategies and lobby both the food industry and government to promote healthier foods.
Of course we need more randomised control trials to investigate causality between UPFs and chronic diseases. But as Hill said, "All scientific work is incomplete... That does not confer upon us a freedom to ignore the knowledge we already have, or to postpone the action it appears to demand at a given time" [6].
1 Lane MM, Gamage E, Du S, et al. Ultra-processed food exposure and adverse health outcomes: umbrella review of epidemiological meta-analyses. Bmj 2024;384:e077310.10.1136/bmj-2023-077310
2 Bite Back. Fuel us, don’t fool us: Are food giants rigging the system against children’s health? (Manufacturers). Bite Back, 2024.
3 Coombes R. Row over ultra-processed foods panel highlights conflicts of interest issue at heart of UK science reporting. Bmj 2023;383:2514.10.1136/bmj.p2514
4 Scientific Advisory Committee on Nutrition. SACN statement on processed foods and health, 2023.
5 EIT Food Consumer Observatory. Consumer perceptions unwrapped: ultra-processed foods (UPF) A pan-European study from the EIT Food Consumer Observatory on consumer perceptions of ultra-processed foods, 2024.
6 Hill AB. The Environment and Disease: Association or Causation? Proc R Soc Med 1965;58:295-300
Competing interests: No competing interests
Re: Nunan’s Rapid Response to ‘Ultra-processed food exposure and adverse health outcomes: umbrella review of epidemiological meta-analyses’
Dear Editor,
Nunan highlights a pervasive issue in health research is overlooking uncertainty in interpreting findings, potentially leading to misleading conclusions about the strength of the evidence (1). He asks: why assess uncertainty if it is not then communicated effectively?
Conclusion word limits, especially in abstracts, often restrict detailed discussion. Given the extensive scope and numerous pooled analyses in our umbrella review, we primarily emphasise meta-evidence meeting our credibility and quality criteria for the strongest evidence available. These criteria were determined using two assessment tools: objective evidence classification criteria and subjective GRADE methodology, respectively. Both tools, pre-specified and consistently reported throughout our study, are commonly used and recommended within the umbrella review literature.
In our conclusion, we prioritise pooled analyses with convincing ("class I") or highly suggestive ("class II") credibility, coupled with either "moderate" or "low" quality, collectively filtering out those with high uncertainty. A more detailed analysis of potential sources of uncertainty and directions for future research can be found in our discussion section.
Nunan further highlights the challenge of introducing the unvalidated evidence classification criteria alongside GRADE to evaluate the evidence.
In nutrition science, no single rigorously validated tool exists to evaluate evidence. While the GRADE system is commonly employed for assessing evidence quality, its uncertain reliability in evaluating complex diet-related evidence, especially within prevalent observational studies, is questioned (2). On the other hand, although recent umbrella review guidelines advocate for the use of the evidence classification criteria, they also acknowledge the limitations of relying on strict cut-off points (3).
Therefore, to ensure a balanced evaluation of evidence and to avoid overreliance on single tools, we employed both the evidence classification criteria and the GRADE methodology.
Nonetheless, aligning with Tobias et al. (2021), we concur that the nutrition science community must collaborate to establish a consensus on developing fit-for-purpose tools. These tools, tailored for effectively synthesising and grading diet-related evidence, are essential given our field's complexities and the strengths and weaknesses of existing evidence grading systems (2).
References
1. Nunan D. Re: Ultra-processed food exposure and adverse health outcomes: umbrella review of epidemiological meta-analyses. [Rapid Response] BMJ 2024. Available from: https://www.bmj.com/content/384/bmj-2023-077310/rr-1
2. Tobias DK, Wittenbecher C, Hu FB. Grading nutrition evidence: where to go from here? Am J Clin Nutr 2021;113(6):1385-87. doi: 10.1093/ajcn/nqab124
3. Fusar-Poli P, Radua J. Ten simple rules for conducting umbrella reviews. Evid Based Ment Health 2018;21(3):95-100. doi: 10.1136/ebmental-2018-300014 [published Online First: 2018/07/15]
Competing interests: All authors have completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf and declare: MML, EG, DNA, AJM, SG, FNJ, AO, and WM are affiliated with the Food & Mood Centre, Deakin University, which has received research funding support from Be Fit Food, Bega Dairy and Drinks, and the a2 Milk Company and philanthropic research funding support from the Waterloo Foundation, Wilson Foundation, the JTM Foundation, the Serp Hills Foundation, the Roberts Family Foundation, and the Fernwood Foundation; MML is secretary for the Melbourne Branch Committee of the Nutrition Society of Australia (unpaid) and has received travel funding support from the International Society for Nutritional Psychiatry Research, the Nutrition Society of Australia, the Australasian Society of Lifestyle Medicine, and the Gut Brain Congress and is an associate investigator for the MicroFit Study, an investigator-led randomised controlled trial exploring the effect of diets with varying levels of industrial processing on gut microbiome composition and partially funded by Be Fit Food (payment received by the Food and Mood Centre, Deakin University); AMJ is secretary for the International Society for Nutritional Psychiatry Research (unpaid) and an associate investigator for the MicroFit Study; SG is affiliated with Deakin University, which has received grant funding support from a National Health and Medical Research Council Synergy Grant (#GNT1182301) and Medical Research Future Fund Cardiovascular Health Mission (#MRF2022907), is affiliated with Monash University, which has received grant funding support from Medical Research Future Fund Consumer-led research (#MRF2022907), is secretary for the Australian Cardiovascular Health and Rehabilitation Association—Victoria and Tasmania—(unpaid), and has received travel funding support from the Institute for Mental and Physical Health and Clinical Translation and SOLVE-CHD (solving the long-standing evidence-practice gap associated with cardiac rehabilitation and secondary prevention of coronary heart disease); PB has received funding from an Australian Research Council Future Fellowship award (project number #FT220100690) and from Bloomberg Philanthropies; ML is affiliated with the Institute for Physical Activity and Nutrition, School of Exercise and Nutrition Sciences, Deakin University, which has received grant funding support from the Australian Research Council (#DP190101323), and has received royalties or license funding from Allen and Unwin (Public Health Nutrition: from Principles to Practice) and Routledge, Taylor and Francis Group (Healthy and Sustainable Food Systems) and consultation and remuneration funding support from WHO and Food Standards Australia New Zealand (in his role as a board member); CMR is affiliated with Johns Hopkins Bloomberg School of Public Health and Johns Hopkins University, which have received grant funding support from the National Heart, Lung, and Blood Institute, Bloomberg American Health Initiative, and the National Institute of Diabetes and Digestive and Kidney Diseases, was chair of the Data and Safety Monitoring Boards for the SUPER Trial: Effect of Dietary Sodium Reduction in Kidney Disease Patients with Albuminuria and the ADEPT Trial: A Clinical Trial of Low-Carbohydrate Dietary Pattern on Glycemic Outcomes, was on the Editorial Board of Diabetes Care (unpaid), was the immediate past chair for the Early Career Committee of the Council on Lifestyle and Cardiometabolic Health, American Heart Association (unpaid), and the Nutritional Epidemiology Research Interest Section of the American Society for Nutrition (unpaid), and has received funding support as an associate editor of Diabetes Care and editorial fellow of the Journal of the American Society of Nephrology; FNJ has received fellowship funding support from the National Health and Medical Research Council (#1194982) and payment or honorariums for lectures, presentations, speakers bureaus, manuscript writing, or educational events from the Malaysian Society of Gastroenterology and Hepatology, JNPN Congress, American Nutrition Association, Personalised Nutrition Summit, and American Academy of Craniofacial Pain, is a Scientific Advisory Board member of Dauten Family Centre for Bipolar Treatment Innovation (unpaid) and Zoe Nutrition (unpaid), has written two books for commercial publication on the topic of nutritional psychiatry and gut health, and is the principal investigator for the MicroFit Study; AO has received fellowship funding support from the National Health and Medical Research Council (#2009295) and is affiliated with Deakin University, which has received grant funding support from the Medical Research Future Fund, Dasman Diabetes Institute, MTP Connect—Targeted Translation Research Accelerator Program, the National Health and Medical Research Council, Barwon Health, and the Waterloo Foundation, and has received funding support for academic editing and as a grant reviewer from SLACK Incorporated (Psychiatric Annals) and the National Health and Medical Research Council, respectively, and travel funding support from the International Society for Nutritional Psychiatry Research; WM is president of the International Society for Nutritional Psychiatry Research (unpaid), has received fellowship funding support from the National Health and Medical Research Council (#2008971) and Multiple Sclerosis Research Australia, consultation and remuneration funding support from Nutrition Research Australia and ParachuteBH, and travel funding support from the Nutrition Society of Australia, Mind Body Interface Symposium, and VitaFoods, and was the acting principal investigator and is an associate investigator for the MicroFit Study.