What Ultra-Processed Foods actually do to your body
The science behind UPFs, overconsumption, and why the overall pattern matters
Key takeaways: Ultra-processed foods now account for more than half of energy intake in UK adults. The most comprehensive review to date links high UPF consumption to 32 adverse health outcomes, including cardiovascular disease, type 2 diabetes, depression, and all-cause mortality. A controlled inpatient trial found that people eating ultra-processed diets consumed approximately 500 extra calories per day compared to unprocessed diets — without being instructed to. The mechanisms are not fully resolved, but multiple pathways are plausible: hyper-palatability driving overconsumption, displacement of nutrient-dense foods, gut microbiome disruption, and the physical structure of processed food affecting satiety. The evidence is largely observational and causation is not proven — but the breadth and consistency of findings across millions of participants is substantial. The practical implication is a direction of travel, not a ban: more whole and minimally processed foods, fewer heavily manufactured ones.
Most discussions of ultra-processed foods start with the question of whether they are bad for you. A more useful question is: what do we actually know about the mechanisms? How does a diet heavy in these products affect biological systems — and what does that mean practically for how we eat?
This companion piece explores the evidence in more depth: how researchers classify and study ultra-processed foods, what the trials and large observational studies show, and why the pattern of the whole diet matters more than any single product.
How researchers define and classify ultra-processed foods
The dominant classification framework is NOVA, developed by researchers at the University of São Paulo. NOVA divides foods into four groups based on the extent and purpose of processing:
• Group 1: Unprocessed or minimally processed foods — fresh fruit and vegetables, plain meat and fish, eggs, milk, plain yoghurt, dried pulses, plain nuts.
• Group 2: Processed culinary ingredients — oils, butter, flour, sugar, salt. Used in cooking but rarely eaten alone.
• Group 3: Processed foods — products made by adding salt, sugar, or oil to Group 1 foods. Tinned vegetables, cured meats, cheese, freshly baked bread.
• Group 4: Ultra-processed foods — industrial formulations made mostly or entirely from substances extracted from foods, with little or no whole food. Typically contain additives absent from homemade food: emulsifiers, stabilisers, flavour enhancers, artificial colours, sweeteners.
The NOVA framework has been criticised for its breadth — it groups together products with very different nutritional profiles, from diet soft drinks to mass-produced bread to flavoured yoghurt — and for making the degree of processing the defining variable rather than nutrient content. These are legitimate methodological concerns. Multiple competing classification systems now exist, and researchers continue to debate which best captures health risk.
What the NOVA system does well is draw attention to something that calorie or nutrient-focused frameworks miss: the degree to which a product has been industrially transformed, and what substances have been added in the process, may matter for health independently of its macronutrient content. The controlled trial evidence, discussed below, suggests this is not simply a matter of fat, sugar, and salt content.
The observational evidence: what large-scale studies show
The bulk of the evidence linking UPF consumption to adverse health outcomes comes from large prospective cohort studies — studies that track dietary patterns in tens or hundreds of thousands of people over years and examine how those patterns relate to later health outcomes.
The limitations of this approach are real: dietary self-reporting is imprecise, people who eat more UPFs may differ in other ways from those who eat fewer (lower incomes, higher stress, less sleep, less exercise), and correlation does not establish causation. Researchers attempt to control for these confounders, but residual confounding is always possible.
With those caveats clearly in mind, the findings from large-scale observational research are striking for their consistency and breadth.
KEY RESEARCH
Lane, M. M., Gamage, E., Du, S., et al. (2024). Ultra-processed food exposure and adverse health outcomes: umbrella review of epidemiological meta-analyses. BMJ, 384, e077310. https://doi.org/10.1136/bmj-2023-077310
The 2024 BMJ umbrella review — which synthesised 45 pooled meta-analyses covering 9.9 million participants — found convincing or highly suggestive associations between high UPF consumption and:
• Cardiovascular disease mortality (approximately 50% higher risk in the highest vs. lowest UPF consumption groups)
• Type 2 diabetes (12% higher risk per 10% increase in UPF proportion of diet)
• Depression and anxiety
• Obesity and weight gain
• Colorectal cancer
• All-cause mortality
• Common mental disorders, asthma and wheezing, frailty, and declining kidney function
The review graded evidence quality using established criteria. Many associations reached “convincing” or “highly suggestive” levels — particularly for type 2 diabetes, cardiovascular mortality, and common mental disorders — but overall GRADE certainty was rated low to moderate for most outcomes. This is consistent with the inherent limitations of observational nutrition research rather than weakness specific to UPF studies.
A 2026 briefing from the UK Parliamentary Office of Science and Technology (POST) reviewed the accumulated evidence and concluded that regular high UPF consumption is associated with poor dietary patterns and increased risk of negative health outcomes across multiple systems — a conclusion that informed the UK government’s increased policy attention to this area.
The trial evidence: what happens when you change the diet
Observational data shows association; trials test causation more directly. A landmark inpatient randomised controlled trial published in Cell Metabolism in 2019 remains the most direct experimental evidence available.
KEY RESEARCH
Hall, K. D., Ayuketah, A., Brychta, R., et al. (2019). Ultra-processed diets cause excess calorie intake and weight gain: an inpatient randomized controlled trial of ad libitum food intake. Cell Metabolism, 30(1), 67–77. https://doi.org/10.1016/j.cmet.2019.05.008
In this trial, 20 adults were admitted to a clinical research facility and randomly assigned to either an ultra-processed or an unprocessed diet for two weeks, then crossed over to the other diet for two weeks. Both diets were carefully matched for total calories offered, energy density, macronutrients (fat, carbohydrate, protein), sugar, sodium, and fibre. Participants were told to eat as much or as little as they wished.
The result was striking: on the ultra-processed diet, participants consumed approximately 508 more calories per day than on the unprocessed diet, ate faster, and gained weight (approximately 0.9 kg). On the unprocessed diet, they lost weight. This occurred even though the two diets were matched for all the variables traditionally used to explain overeating.
This finding is significant because it suggests that something about the degree of processing itself — beyond fat, sugar, salt, and fibre content — influences energy intake. The mechanisms are not yet fully understood, but the trial result is difficult to explain through nutritional content alone.
Proposed mechanisms: how UPFs may drive harm
Several biological pathways have been proposed to explain the associations between UPF consumption and adverse health outcomes. None is definitively established, and they are not mutually exclusive — the effects of a diet heavy in ultra-processed foods are probably multi-pathway.
Hyper-palatability and overconsumption
UPFs are specifically engineered to be highly appealing through precise combinations of fat, sugar, salt, and texture. This engineering can override normal satiety signalling, making it easy to consume more than the body needs without registering fullness. The Hall trial supports this mechanism directly: participants ate faster on the UPF diet, with a shorter inter-meal interval — consistent with reduced satiety signalling.
Displacement of nutrient-dense foods
Diets high in UPFs are consistently associated in nationally representative surveys with lower intake of fruits, vegetables, legumes, nuts, and whole grains — the foods with the strongest evidence for health benefit. When UPFs make up a large proportion of caloric intake, there is mechanically less room for the foods associated with reduced chronic disease risk. Meta-analyses of dietary survey data across multiple countries find that higher UPF intake is inversely associated with dietary fibre, micronutrient density, and anti-inflammatory food components.
Gut microbiome disruption
Several additives commonly found in UPFs — particularly emulsifiers and certain preservatives — have been shown in animal and early human studies to alter the composition and function of the gut microbiome. The gut microbiome is increasingly understood to play a central role in metabolic health, immune function, and the gut-brain axis. Low dietary fibre — characteristic of most UPF-heavy diets — also reduces the diversity and abundance of beneficial bacterial species. This area of research is emerging and causally complex; human evidence remains limited compared to animal models.
Food matrix and physical structure
The physical structure of food — its texture, viscosity, and the degree to which its cellular structure has been disrupted or reformed — influences how quickly it is digested and how it signals satiety. Highly processed foods are often consumed in a form where the food matrix has been substantially altered: cells ruptured, starches pre-gelatinised, proteins denatured. This may accelerate digestion and absorption, affect glycaemic response, and reduce satiety independent of nutrient content. Research in this area is active but not yet conclusive in humans.
High sugar, salt, and saturated fat content
The nutrient profile of most UPF-heavy diets is independently problematic: high in added sugars (contributing to glycaemic load and dental health impacts), high in sodium (contributing to hypertension and cardiovascular risk), high in saturated and trans fats (contributing to dyslipidaemia and cardiovascular disease), and low in fibre (contributing to bowel cancer risk and microbiome impoverishment). While these nutrient effects do not explain all the associations observed, they are well-established contributors to chronic disease risk and cannot be separated from the UPF effect in practice.
Diet and mental health: the UPF dimension
The association between UPF consumption and mental health outcomes — particularly depression and anxiety — has been replicated across multiple large cohort studies and is included in the BMJ umbrella review’s “convincing” evidence category. The mechanisms here are even less well understood than for metabolic and cardiovascular outcomes.
Several pathways are plausible. The gut-brain axis — the bidirectional communication between gut microbiota and the central nervous system, mediated through neural, hormonal, and immune signals — may link dietary patterns to mood and cognitive function. Specific nutrients with established roles in neurotransmitter synthesis (omega-3 fatty acids, B vitamins, zinc, magnesium) are characteristically depleted in UPF-heavy diets. Chronic low-grade inflammation, promoted by poor diet quality and gut dysbiosis, has been independently associated with depressive symptoms.
The confounding challenge is particularly acute in this domain: people who are depressed or anxious may eat more UPFs as a consequence rather than a cause, and social and economic factors that predict poor mental health also predict higher UPF consumption. The observational associations are consistent and replicated, but establishing the direction of causality is methodologically difficult.
What UK diets actually look like — and the scale of the challenge
Ultra-processed foods account for approximately 57% of total energy intake in UK adults, rising to higher proportions in children and adolescents. This makes the UK one of the highest UPF-consuming populations in Europe. The National Diet and Nutrition Survey consistently shows that this high UPF intake is accompanied by:
• Fibre intake averaging approximately 19g per day, against a recommended 30g
• Free sugar intake approximately double the recommended maximum of 5% of total energy
• Only around 1 in 3 adults meeting the 5-a-day fruit and vegetable recommendation
• Fewer than 1% of the population meeting all current dietary guidelines simultaneously
These figures reflect the food environment, economics, time constraints, and cultural norms that shape eating behaviour. They are not primarily a story of individual failure. They do, however, illustrate why even modest population-level shifts away from UPFs toward whole and minimally processed foods could have substantial public health effects.
Important caveats: what the evidence does not show
Responsible interpretation of UPF research requires acknowledging several limitations that are sometimes obscured in popular coverage.
First, the evidence does not establish that UPFs cause harm independent of their nutrient content. The Hall trial shows that degree of processing affects intake independent of matched macronutrients — but this does not rule out that the effects observed in observational studies are primarily driven by the poor nutritional profiles of most UPF-heavy diets rather than processing itself.
Second, the NOVA classification groups together products with very different risk profiles. A pot of low-fat plain yoghurt and a bag of crisps are both classified as ultra-processed, but the evidence does not support treating them as equivalent health risks. The associations are strongest for processed meats, sugar-sweetened beverages, and industrially produced baked goods and snack foods.
Third, individual variation is real. The same dietary pattern affects different people differently, depending on genetics, gut microbiome composition, physical activity, sleep, stress, and many other factors. Population-level associations tell us about risk at the group level, not individual destiny.
Fourth, most of the evidence is observational, and GRADE certainty for most outcomes is rated low to moderate. The findings are consistent and replicated — but they are not the same quality of evidence as well-powered randomised trials of many years’ duration, which are practically impossible in nutrition research.
Putting it together: UPFs and health-span
From a health-span perspective, the evidence on ultra-processed foods converges on several practical conclusions.
The strongest risk signal is for diets in which UPFs constitute a very large proportion of total intake. The associations with cardiovascular disease, type 2 diabetes, and mortality are most consistently observed at high levels of UPF consumption, not at moderate or occasional intake.
The most effective practical response is not to eliminate all processed foods but to increase the proportion of whole and minimally processed foods in the diet. Frozen vegetables, tinned pulses, plain dairy, eggs, fresh or frozen fish, and whole grains are all minimally processed, nutritionally valuable, and compatible with busy lives and constrained budgets.
The displacement effect is worth particular attention: a diet in which UPFs are reduced tends to create space for the foods most consistently associated with health benefit — more vegetables, more legumes, more whole grains, more variety. The positive effects of eating more whole foods and the negative effects of eating fewer UPFs are, in practice, largely the same intervention.
As part of Eat Well — Pillar Two of The New 5-a-Day — reducing ultra-processed food intake connects directly to all the other evidence-based dietary principles: more dietary variety, better gut microbiome health, improved metabolic markers, and the broader pattern of consistent, sustainable eating that supports long-term wellbeing. The science does not demand perfection. It supports a clear and achievable direction of travel.
A note on medical advice: The content in this post is intended to inform and inspire, not to replace professional medical guidance. If anything you’ve read raises questions or concerns about your own health, please speak to your GP or another qualified health professional.
