Why is there no regulation despite evidence that ultra-processed foods are hazardous to long-term health?

Abstract

Non-communicable diseases (NCDs) have exceeded infectious diseases as a global health threat. This shift has occurred concurrently with changes to lifestyle and diet, of which the increased consumption of ultra-processed foods (UPFs) is particularly concerning. UPFs are defined by industrial processing of often cheaply sourced natural food and inclusion of non-nutritional additives such as azo-dye colourants, non-caloric sweeteners, emulsifiers and preservatives. They now account for approximately half of the caloric intake in many countries and notably by children. These products can compromise nutritional quality while promoting hedonistic hunger and disrupting the gut microbiota. Population studies are increasingly demonstrating an association between high UPF consumption and increased all-cause mortality, obesity, cardiovascular disease, cancer risk, mental health and neuro-behavioural disorders.

While association studies cannot account for all potential confounding factors, triangulation through combining large-scale cohort studies with animal/in vitro models and human challenge studies provides more compelling evidence that UPFs increase the risks of NCDs. Despite the increasing evidence, regulatory action on their use has remained limited.

We review the evidence and explore the reasons why effective regulation is lacking, with increasing examples of inappropriate food industry influence on political decisions in several countries. The current focus on reducing consumption of high fat, sugar and salt has distracted attention from the adverse effects of alternative non-nutritional additives. We propose changes to food labelling and call for public health interventions to combat and minimise UPF consumption. These interventions are especially vital for socioeconomically disadvantaged populations and children, who increasingly consume UPFs as attractive and cheaper alternatives to fresh or minimally processed products. Urgent action is needed to promote healthier dietary practices worldwide.

The changing face of lifestyle and world health

Despite occasional pandemics, overall morbidity and mortality from infectious disease has fallen over the last century, while non-communicable diseases (NCDs) have increased. Long-term health and survival are threatened by obesity, metabolic syndrome, cardiovascular disease, allergic and autoimmune diseases, malignancy and mental health disorders.^1,2 The increasing prevalence of NCDs occurred contemporaneously with an increasingly sedentary lifestyle and diets moving towards more heavily processed food, and those high in fats, sugar and salt.

Over the last 10–15 years, research publications have identified associations between ultra-processed food (UPF) consumption and NCDs. There were <10 publications per year before 2015 in Web of Science, but 284 in 2021, simultaneously with progressive increases in UPF sales, which as a percentage of total energy intake constituted 61% in USA, 50.7% in UK and 30% in Brazil. It is notable that Brazil has the most publications highlighting UPF health concerns and a public health campaign to improve dietary practices.³ Ninety-two of 104 studies from a systematic review showed high UPF consumption associated with chronic ill health, and meta-analyses showed significant associations with obesity, type 2 diabetes, cardiovascular disease, depression and all-cause mortality.⁴ As the complexity of the associations has not been fully elaborated, the food industry has successfully countered meaningful legislation.⁵ We have evaluated the strengths and weaknesses of the evidence, to synthesise the justification and means to improve dietary practices.

Infections versus non-communicable diseases (NCDs)

The concept that reduced infections associated with improved hygiene increased the risk of NCDs arose from elaboration of immunological counter-regulatory mechanisms. It became known as the hygiene hypothesis, suggesting that thymus-dependent T-helper 1 cell and T-regulatory cell responses to infection suppress activity associated with NCDs.⁶ However, the immunological balance is more complex and rather than active infection, it is the non-pathogenic microbiome existing in the gut and other epithelial surfaces which has a symbiotic relationship with the human host. A healthy microbiome (eubiosis) benefits digestion, aids vitamin synthesis, stimulates protective immune responses, inhibits pathogen colonisation, and is associated with reduced risks of NCDs. The metagenome (the human microbiome genome), combined with genetic, epigenetic, proteomic, metabolomic and environmental (exposomic) influences outcomes.⁷

Diet and NCDs

A key exposome is diet, with subtle differences altering gene expression through epigenetic changes, the gut microbiome (metagenomics), immune, and metabolic responses. One component of the diet which has altered the balance of energy density and nutritional quality is increasing food processing. Populations exposed to food insecurity cannot maintain a ‘healthy diet’, leading them to resort to cheap UPFs.⁴ UNICEF recently announced that childhood obesity is more common than underweight and linked it to UPF consumption.⁸

Ultra-processed foods

Processing of often cheaply sourced foods alters appearance, taste and nutritional qualities. Manufacturers utilise permitted food additives to restore and sometimes enhance colour, taste and structure. As well as low price and attractive appearance, improved shelf life and ease of cooking have increased sales and consumption worldwide.

There are over 2,500 permitted food additives and while there are variations in approval, the Food and Drug Administration (FDA) in USA and the European Food Safety Authority (EFSA) are particularly influential regulators.⁹ The latter introduced the E-numbering system for approved additives, which is widely employed. The largest combinations of additives are in UPFs, but definitions vary, with most studies using ill-defined sociologically based descriptors. They incorporate the concept that enhancing appearance and taste, and reducing costs, promotes marketing.

A classification of the degree of processing (NOVA) often employed in food consumption surveys was developed by the University of Sao Paulo, Brazil in 2009¹⁰ (Table 1).

Table 1.

The NOVA classification (not an acronym) of degrees of food processing is now used in many population nutrition surveys.¹⁰

NOVA grade	Description	Examples
NOVA-1	Unprocessed or minimally processed natural foods	Drying, crushing, pasteurising
NOVA-2	Pressed, ground or refined processing of natural food	Honey, butter, vegetable oils
NOVA-3	Foods processed to increase shelf-life with no >2–3 additives	Added salt, sugar, oil
NOVA-4	Ultra-processed with >5 additives, some with little or no NOVA 1 food	Confectionery, sweetened drinks, processed meat, fast food, ready to eat

Dietary surveys using the NOVA classification highlighted increasing intakes of UPFs.^3,4,5,10,11 The UK twin birth cohort study is particularly concerning in identifying that UPFs as a proportion of total calorie intake are 46.9% at 21 months and 59.4% at 7 years.¹²

An analysis of over 11,000 adults in the USA tracked for >30 years revealed that those in the 75th centile of UPF intake had a 4% higher risk of mortality compared to those in the 25th centile. The strongest associations with higher mortality were consumption of processed meats, artificially sweetened beverages and breakfast UPFs.¹³

As NOVA-4 foods sometimes contain minimally processed natural foods but have >5 added ingredients, the classification can be challenged. It has led to alternative definitions but all with similar shortcomings. The USA Make Our Children Healthy Again (MAHA) strategy aims to reduce the ‘epidemic’ of chronic disease afflicting young people through a national public health programme. To create a federal approach, the FDA and Department of Agriculture has commenced a public consultation to address the current shortcomings of definition and variability of regulation of UPFs. While the programme emphasises the need for ‘rigorous science’, public trust could be undermined by the lack of scientific evidence underlying the current US political discourse that vaccines cause chronic disease.¹⁴

Hunger and UPFs

The three basic mechanisms triggering appetite are¹⁵:

• 1.Homeostatic hunger; activated by physiological processes which promote hunger to initiate eating and restore energy levels.
• 2.Hedonistic hunger; triggered by the sight, smell or taste creates a desire to eat even when not physiologically hungry.
• 3.Microbiota-driven hunger; activated by the gut microbiome producing metabolites influencing appetite hormones signalling to the brain, altering feelings of hunger or satiety.

While UPFs promote hedonistic hunger, they also alter the gut microbiome.

UPFs and the gut microbiome

Among the synergistic effects of eubiosis, the microbes degrade dietary fibre, generating short-chain fatty acids (SCFAs), which maintain gut barrier through increased tight junction expression, mucin secretion and generation of anti-pathogen compounds. They also moderate appetite and insulin sensitivity, and immune responses by promoting T-regs to reduce inflammation.¹⁶

Murine model and human volunteer studies have shown that altered gut microbiome (dysbiosis) and increased disruption of the gut epithelial barrier are induced by several groups of food additives including emulsifiers, non-caloric sweeteners, preservatives and inorganic nanoparticles. Dysbiosis is associated with obesity, cardio-metabolic disorders, inflammatory bowel disease, allergy and increased cancer risk.^16,17 Oncogenic effects occur through the generation of toxic DNA-damaging metabolites, strongly linked to colorectal cancer.¹⁸ Azo-dye food colourants contribute to gut dysbiosis and are processed by the microbiome to DNA-damaging metabolites.¹⁹

Triangulation of evidence

Longitudinal cohort studies suggest that UPFs are associated with long-term adverse health outcomes, but association cannot be assumed to indicate cause and effect. However, mechanistic and challenge studies of individual additives have indicated the potential that UPFs significantly contribute to chronic ill health, particularly when commenced early in life.²⁰

Methodological triangulation is the use of at least two methods to investigate the same research question.²¹ Three independent approaches increase the credibility of evidence; human longitudinal cohort association studies; in vitro and animal model studies of changes to basic mechanisms associated with NCDs; and human controlled challenges (Fig. 1).

Fig. 1.

The three independent research approaches to the study of the health impact of ultra-processed foods (UPFs) which constitutes triangulation of evidence. While each approach in isolation is not sufficient evidence, if all three point to an adverse effect on the risks of non-communicable diseases (NCDs), the need for action becomes more compelling.

Individual additives

Azo-dye colourants

A systematic review of 13 studies in over 25,000 adults suggested that high UPF consumption increased risks of depression.²² While not indicating which components affect mental health, challenge, animal model and in vitro studies point to likely causes. A systematic review, mostly of controlled challenges, showed that mixtures of food colourants and benzoate preservatives adversely affect behaviour of children with and without pre-existing hyperactive behaviour disorders and supported by murine studies.^23,24 A paediatric study showed an interaction between a polymorphism in the histamine N-methyl transferase gene and an adverse behavioural response to azo-dye and sodium benzoate challenge. The polymorphism reduces histamine degradation and the azo-dye tartrazine causes in vivo and in vitro histamine release, which upsets the balance of histamine three and four receptors in the brain, affecting wakefulness and behaviour.^25,26

Studies have linked high intake of UPFs with increased cancer risks in humans and animal models,²⁷ with a likely mechanism being gut microbial generated azo-reductases producing DNA-damaging, oncogenic metabolites from azo-dye colourants.^19,28

Non-caloric sweeteners

The US study in depression-free women at recruitment who had high UPF intakes related non-caloric sweeteners (NCSs) to increased risks of depression.²⁹ NCSs, and notably sucralose, acesulfame and saccharin, differ from natural sugar in stimulating purinergic, rather than glutaminergic, neurotransmission in murine models.³⁰ Purinergic activity has been associated with depression.³¹

The French prospective study (NutriNet-Santé cohort) of over 100,000 adults showed significant associations between high NCS intake and cardiovascular disease.³² This cohort also revealed an association between various additive mixes including NCSs and Type-2 diabetes.³³ To what extent this is a consequence of obesity remains to be established. However, the World Health Organization in 2023 published a conditional guideline (ISBN: 978-92-4-007361-6) advising that NCSs are not suitable alternatives to sugar for weight control or to reduce the risk of NCDs. Lack of certainty arose from review of short-term randomised controlled trials, showing that NCSs rather than natural sugar intake reduced body weight and BMI, while long-term follow-up cohorts had higher rates of obesity, type 2 diabetes, cardiovascular diseases and all-cause mortality.

Emulsifiers

The NutriNet-Santé cohort showed an association between high emulsifier intake and cardiovascular disease³⁴ and increased risks of all cancers, notably breast and prostate, with high emulsifier intake.³⁵

Advanced glycation end-products

High-temperature processing of meats generates advanced glycation end-products (AGEs). The many different AGE molecules bind to AGE receptors present on many immune active cells. High intakes as they appear in UPFs have been linked to increased risks of cancers and several NCDs.³⁶

Multiple additives/UPFs

Some cohort studies delineated which components of UPFs were linked to disease outcomes, but it is equally possible that there is a compounding effect of multiple ingredients. The strongest evidence is an obesogenic effect of high UPF intake by increasing appetite, calorie intake and weight gain.37, 38 A 2-year prospective study in Chilean pre-school children associated high UPF intake with adiposity markers.³⁸ A Canadian study revealed a positive linear regression of associations between every 10% increase in UPF-derived energy intake in children at 3 years of age and adiposity at 5 years, which was statistically significant in boys but not in girls.³⁹ UK and European studies linked high intakes of energy-dense, low-nutrient foods (characterised by UPFs) with overweight and obesity in children.^40,41 A cross-over challenge study in 20 inpatient adult volunteers fed either a high UPF or healthy fresh food diet (matched for calorie content) each for 2 weeks in randomised order. They were permitted additional food as desired and during the UPF period calorie intake was significantly higher, associated with weight gain, which did not occur during the fresh food period.⁴² Combining data from eight longitudinal cohort studies with over 500,000 participants associated high UPF intake with obesity, cardiovascular disease, type 2 diabetes and some cancers.³ Comparing the highest with lowest quartiles of UPF exposure of over 197,000 adults in the UK Biobank study, all-cause cancer and cancer deaths were significantly increased, particularly ovarian cancer.²⁷ A systematic review and meta-analysis showed high UPF exposure associated with colorectal and breast cancer risk, but not prostate cancer risk.⁴³ Could the cumulative effect of UPFs in young people explain the increase in malignancies in young adults (<50 years) which includes colorectal, breast and uterine cancers?⁴⁴

A US online and telephone survey of over 2,000 nationally representative adults (>50 years) found addiction to a high UPF diet in 12.5%, linked with obesity, poor mental and physical health, and social isolation, more in women than men.⁴⁵ Early life exposure to highly sweetened weaning foods and drinks creates lifelong taste preferences.⁴⁶

There are many other components of UPFs which require investigation, including inorganic nano-particles and food flavours. Plastic food- wraps are a concern because in vitro human cell culture studies highlight that the chemicals which migrate from the plastics into the food have potential for adverse health effects.⁴⁷

Strengths and weaknesses of current evidence

Population long-term cohort association studies linking UPF consumption with NCDs employ variable definitions, ascertainment of intake and outcomes. Inference of cause and effect is compromised by confounding because an unhealthy lifestyle commonly co-segregates with high intakes of UPFs.

In vitro and animal challenge studies have sometimes employed exposures which are not relevant to humans. High concentrations of sodium benzoate and potassium sorbate impaired apoptosis gene transcription (which normally clears pre-cancerous cells) in a cell culture line, matched by animal challenge studies, were conducted in conditions which cannot be directly translated to humans.⁴⁸ Nevertheless, there is solid evidence from triangulation and the health benefits of fresh foods as in the Mediterranean diet.⁴⁹ UPFs or a fresh food diet are mutually exclusive, therefore regulatory action is required without further delay.

In summary, UPFs are inexpensive, shelf-stable products that exacerbate poor nutrition and risks of NCDs by displacing nutrient-rich foods.⁵⁰ This is exemplified by a review of systematic reviews showing convincing or highly suggestive evidence of associations with 32 specific adverse health parameters and all-cause mortality.⁵¹

Why no regulatory action?

There is excessive influence of UPF manufacturers in committees advising politicians.^52,53 They engage in high-profile health promotion while simultaneously lobbying against regulatory measures.⁵⁴ It is suspected that the reluctance to act extends beyond regulators to the medical profession and health systems.⁵⁵ An exemplar of attempts to affect policy is from a Unilever South Africa leaflet released in 2025:

‘Should we turn our backs on ultra-processed foods?’

‘Elimination of all UPFs from the diet, irrespective of their nutritional quality, would be akin to throwing the baby out with the bath water. This will not help fix the broken food system but may serve to worsen the existing disparities in food security.’

The University of North Carolina Global food research programme published an evidence-based list of rejoinders to industry claims about UPFs.⁵⁶

Industry lobbying blocked UK local authorities from banning outdoor junk food advertising through threats of loss of advertising income.⁵⁷ A UK briefing report (2024) aimed to inform members of parliament highlighted the evidence from association studies and explained the limits of assuming causal relationships and problems of UPF classification. While it classified some associations as ‘convincing evidence’, it underplayed data from challenge and mechanistic studies and concluded that more research was required before public health policies can be formulated.⁵⁸ Industry has sometimes influenced scientific research and public narrative. The Sugar Research Foundation funded and directed a review which favoured the sugar industry, and downplayed sugar’s role in heart disease and failed to disclose industry sponsorship,⁵⁹ which shaped subsequent public health policy. A 2020 BMJ analysis noted that industry-funded nutrition studies were more likely to produce outcomes favourable to sponsors.⁶⁰ The industry-led influence obscuring the dangers of sugar raises concern that similar dynamics are delaying action on UPFs.

Other tactics include marketing that portrays products as healthy despite having poor nutritional content⁶¹ such as plant-based ultra-processed meat alternatives. They are marketed as health promoting, but are typically highly processed with additives to mimic the taste and texture of real meat.⁶²

Confused and blocked public health strategies

Allura red, an azo food colourant, has potential carcinogenic effects, particularly increasing colon carcinomas, through microbial azo-reductases generating DNA-damaging metabolites.^19,27,63 This resulted in withdrawn approval in some European countries, but it is still approved by FDA and EFSA. FDA in January 2025 withdrew approval for FD&C Red No 3 (known as erythrosine in other countries), based on potential carcinogenicity in murine models through mechanisms not relevant to humans and with no evidence from human studies. Erythrosine is still approved for use by European and UK regulators. The current focus on azo-food colourants as hazardous to health distracts from a comprehensive approach to optimising nutrition and lifestyle.⁶⁴

A review of evidence suggests that high sugar-sweetened beverage (SSB) intake is associated with increased metabolic risks associated with obesity,⁶⁵ prompting manufacturers to use NCSs but without evidence that this change is either beneficial, or substituting one hazard for another? Some Californian cities have introduced SSB taxes, but it is unclear whether this applies to NCSs. Furthermore, industry lobbying at state level has limited the ability of local governments to make independent decisions in the interests of public health.⁶⁶

A recent UK government capitulation to food industry pressure is highlighted by the Department of Health and Social Care (DHSC) changing its guidance to food outlets ahead of new legislation. The Food and Drinks federation representing large corporations such as Nestle, Unilever and Mars disputed the guidance and demanded a change. The advice will now only limit promotion of foods and drinks high in fat, salt or sugar. While encouraging ‘healthier eating’, there will be no statement about processed foods.

What must change?

E-numbers

Hitherto, the only significant regulatory change initiated by EFSA has been food labelling of azo-dyes and sodium benzoate as potentially affecting children’s behaviour.²⁰ Manufacturers have moved to using colourants from natural sources, some of which are potentially health-promoting. However, as all approved colourants have E-numbers, consumers are unlikely to discriminate. Even vitamin C, ascorbic acid, has an E-notation when used as an anti-oxidant. Thus, the E system is not fit for purpose and we suggest that it should be replaced by one which highlights ‘healthy’ additives (maybe H-notation), leaving the rest as E.

Food labelling

Permitted food additives are listed in the contents box in small lettering on the back of product containers. Only common food allergens are indicated in bold-type. Much of the information in the nutritional value facts box, as required by law, is misleading.⁶⁷ Studies of the effects on purchasing behaviour have produced conflicting results. A 2018 meta-analysis of published studies suggested small improvements in selecting healthy eating options, but some authors had funded associations with the food industry.⁶⁸ A consumer survey sponsored by the UK Food Standards Agency (FSA) concluded that labels alone are unlikely to lead to changes in purchasing behaviour.⁶⁹ A survey from India indicated that awareness of food labels and their influence on purchasing decisions was significantly influenced by the demographics (age, occupation, family income and eating habits) of the subjects interviewed.⁷⁰

Evidence suggests that high SSB intake is associated with increased metabolic risks associated with obesity, but high SSB intake is likely confounded by other differences in diet including higher consumption of UPFs,⁷¹ and beverage manufacturers are switching to NCSs, substituting one hazard for another? Some cities in California introduced SSB taxes but it is unclear whether this applies to NCSs. Furthermore, industry lobbying at state level to limit the ability of local governments to make independent legally binding decisions is concerning.⁷² Rather than the current small back of package confusing information, prominent front of package (FoP) warning labels for products high in ingredients with evidence of adverse health effects could improve informed choices. The benefits of FoPs have been established in Chile, with large reductions in purchasing of foods high in sugar, sodium and calories.⁷³ Hitherto proposals on FoPs notably by FDA have focused on foods high in sodium, sugar and saturated fats.⁷⁴ While the FDA binary FoP (healthy yes/no) label improved purchasing choices, a more comprehensive scoring system (‘1 - least healthful to 100 - healthful’) was significantly better.⁷⁵

Conclusion

As the global health landscape moves towards NCDs instead of infectious disease, there is evidence which links UFPs to multiple health concerns. Urgent public health action is needed to promote healthy dietary practice while reversing rising UPF consumption.

Public health regulatory measures

To achieve improvements in nutrition, the following actions are recommended:

• •Immediate bans on additives with evidence of harm and no nutritional benefit such as azo-dye colourants.
• •More rigid restrictions on marketing of UPFs to children, and banned provision of UPFs by schools and public institutions.
• •Local and national government intervention to legislate against advertising and marketing of unhealthy UPFs.
• •Governments must address food insecurity by reducing costs through subsidising farming, wholesalers and outlets that provide fresh food.⁷⁶ This sentiment is echoed by many of the studies quoted in this review. The best example that this strategy can be effective comes from the USA Supplement Nutrition Assistance Program (SNAP) for populations exposed to food insecurity which provided benefit cards for food purchasing, and those participating had lower risks of obesity and cardiovascular disease risk scores compared with non-participants.⁷⁷
• •To achieve the above, government should establish a task-force combining expertise in food and medical sciences to evaluate all evidence independent of industry influence; where necessary, commission research to address unresolved issues; allow all agencies including industry to present evidence; and finally, to make recommendations exclusively in the interests of public health on legislation, approval and notation of additives, advertising, FoP food labelling, and subsidies to improve availability of fresh foods.

Quality improvement to achieve change

We appreciate that implementation will be slow due to barriers, such as existing policy structures, regulatory fragmentation, industry opposition and economic considerations. There has been timely publication in 2025 of several policy documents by the UK government which are concordant with our aims; the National Health Service ‘Fit for the future 10 year health plan’ includes a focus on moving from sickness to prevention; the Health Research Authority strategy 2025–8 aims to boost research that improves health; the Medicines and Healthcare products Regulatory Agency 2025 roadmap includes a new code for post-marketing surveillance which could be applied to UPFs; and the Life Sciences sector plan from the Departments of Business and Trade, Science and Innovation, and DHSC aims for better health outcomes. While the FSA policy is focused on ‘food you can trust’, it is notably not concordant or included in any of the above policies and appears prone to industry influence. While the aims are all laudable, they are weak on detail of the means to achieve the desired outcomes. This does not diminish the vision and we believe that by employing the science of change management and quality improvement, the aims will be achieved. Our recommendations not only provide the vision but by identifying the barriers, involving all the relevant stake-holders and using an action/effect programme with plan-do-study-act cycles, the mission can be achieved.^78,79 We hope this review addresses Lord James Nicholas Bethell’s call that doctors must step up and campaign against the damaging promotion of ‘junk food’ by the food industry.⁸⁰

CRediT authorship contribution statement

John O. Warner: Writing – review & editing, Writing – original draft, Project administration, Methodology, Data curation, Conceptualization. Alisha Khan: Writing – review & editing, Writing – original draft, Formal analysis, Data curation.

Declaration of competing interest

John Oliver Warner reports a relationship with the Food Standard’s Agency’s (FSA) Advisory Committee for Novel Foods and Processes and the Food Allergy and Intolerance Group that includes: board membership completed 12 years ago. John Oliver Warner received bursaries for giving invited lectures over the last 5 years from Friesland Campina; Danone/Nutricia; Kingslac; and Airsonett. He and his team received research grants from FSA to conduct studies of the effects of artificial food additives on child behaviour. He recently contacted FSA to express his concern about the lack of regulation of UPFs. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.