Fresh whole vegetables on a white surface, including peppers, broccoli, carrots, sweet potato and Brussels sprouts.

Ultra-Processed Foods: How Strong Is the Evidence?

10 min read

In a 2019 trial, 20 adults lived on a research ward for a month and ate an ultra-processed diet for 2 weeks and a minimally processed diet for 2 weeks. The meals on offer were matched for calories, sugar, fat, salt and fibre, and people could eat as much as they liked. On the ultra-processed diet, they ate about 508 more calories a day than on the minimally processed diet and gained weight.

Large reviews of observational studies, covering nearly 10 million people, also link higher intake of ultra-processed food with a higher risk of death, death from heart disease, type 2 diabetes and poorer mental health. This guide covers what counts as ultra-processed, what the reviews and the trial found, how strong the evidence is, and practical ways to eat less of these foods.

32 of 45
Pooled results linking higher ultra-processed food intake with harm, mostly on low or very low quality evidence
Lane et al. 2024, BMJ, 9.9 million people
21%
Higher relative risk of death from any cause with higher vs lower intake, a fairly strong link on the numbers from low quality studies
Lane et al. 2024, BMJ
15%
Higher relative risk of death from any cause, highest vs lowest intake, in a newer pool of 18 cohort studies
Liang et al. 2025, 1.1 million people
508 kcal
More eaten per day on an ultra-processed diet than a minimally processed one, in a 4-week ward trial
Hall et al. 2019, 20 adults
Sources: Lane et al. 2024, BMJ. Liang et al. 2025, Systematic Reviews. Hall et al. 2019, Cell Metabolism.

What counts as ultra-processed food

The NHS explains that most foods are processed in some way, from pasteurised milk to tinned and frozen foods. Ultra-processed foods have been through more processing and often include ingredients you would not usually have at home, such as preservatives, sweeteners and emulsifiers. The NHS gives sweetened drinks, crisps, packaged cakes and biscuits, ice cream, sweets and chocolate, some ready meals and packaged meat pastries as examples.

Researchers usually sort foods with the NOVA system, which places food in four groups from unprocessed to ultra-processed. The US National Institutes of Health (NIH) describes NOVA as treating foods as ultra-processed if they contain ingredients mainly found in industrial food manufacturing, such as hydrogenated oils, high-fructose corn syrup, flavourings and emulsifiers. The NHS notes that some ultra-processed foods can be part of a healthy diet, such as wholemeal sliced bread, wholegrain or higher fibre breakfast cereals and baked beans.

What the largest review found

A 2024 umbrella review (a review that pools other reviews) in the BMJ by Melissa Lane and colleagues gathered 14 meta-analyses with 45 pooled results and 9,888,373 participants. All the underlying studies were observational, meaning researchers tracked what people ate and their health without assigning diets. Higher ultra-processed food intake was linked with harm in 32 of the 45 pooled results, and 13 found no clear link. The links covered outcomes in death, cancer, and mental, respiratory, heart, gut and metabolic health.

The authors graded the strength of each link. Four were graded convincing: a 50% higher risk of death from cardiovascular disease, a 12% higher risk of type 2 diabetes for each 10% increase in ultra-processed food intake, and 48% and 53% higher odds of anxiety and common mental disorders. These two mental health results come from studies that measured diet and mental health at the same time, so they cannot show which came first. Others were graded highly suggestive, including a 21% higher risk of death from any cause, a 66% higher risk of death from heart disease, and a 22% higher risk of depression. They also rated the quality of the evidence: 4 results were moderate quality, 22 low and 19 very low. The convincing link with cardiovascular death was rated very low quality. These are two different scores. The strength grade looks at the numbers, such as how large and consistent a link is. The quality rating looks at how the studies were done and how likely bias is. A link can look convincing on the numbers and still rest on low quality studies. All these figures are also relative, so they describe a change compared with a starting risk, not your personal chance of getting ill.

Links with higher ultra-processed food intake Higher relative risk or odds, higher vs lower intake, observational studies Heart disease death 66% Common mental disorders 53% Cardiovascular death 50% Anxiety 48% Depression 22% Death from any cause 21% 0% 40% 80%
Source: Lane et al. 2024, BMJ. These are relative figures from observational studies, using different measures (risk ratios, hazard ratios and odds ratios). The anxiety and mental disorder results come from studies that measured diet and health at the same time. Cardiovascular death (highlighted), anxiety and common mental disorders were graded convincing on the numbers, and the rest highly suggestive. The highlighted bar was rated very low quality and the rest low quality. Heart disease death and cardiovascular death come from different meta-analyses.

What a newer review found on death risk

A 2025 review in Systematic Reviews by Shuming Liang and colleagues pooled 18 cohort studies (studies that follow people for years) with 1,148,387 people and 173,107 deaths. People with the highest intake of ultra-processed food had a 15% higher risk of death from any cause than those with the lowest intake. Each 10% rise in the share of ultra-processed food in the diet was linked with a 10% higher risk. The results varied a lot between studies. This review updates earlier work with newer studies, so it is not fully separate from the BMJ review, and its figure is lower than the BMJ review's 21%.

Why these foods might be easy to overeat

The 2019 trial by Kevin Hall and colleagues at the NIH tested this directly. Twenty adults, average age 31, lived at the NIH Clinical Center for a month. In random order, they ate an ultra-processed diet for 2 weeks and a minimally processed diet for 2 weeks. The meals on offer were matched for calories, sugar, fat, carbohydrate, salt and fibre, and for energy density (calories per gram) when drinks were counted. People could eat as much or as little as they liked. An ultra-processed breakfast might be a bagel with cream cheese and turkey bacon, and a minimally processed one oatmeal with bananas, walnuts and skimmed milk.

On the ultra-processed diet, people ate about 508 more calories a day, mostly from carbohydrate and fat, and gained 0.9 kg on average, of which about 0.4 kg was body fat. They lost a similar amount on the minimally processed diet. People rated the meals as similarly pleasant and reported similar hunger, but they ate faster on the ultra-processed diet. Once drinks were set aside, the solid ultra-processed foods had about 85% more calories per gram, and the authors say this likely added to the extra intake. The NIH notes that small differences in protein between the diets might explain up to half of the extra calories.

What the evidence does not show

The large reviews are built on observational studies, so they show links, not proof that ultra-processed food causes each outcome. People who eat more of these foods might differ in other ways that the studies could not fully measure. The NHS states that it is not known if ultra-processed foods are unhealthy because of the processing itself or because of the calories, fat, salt and sugar they contain.

The Hall trial was small and short, with 20 adults, average age 31, over 4 weeks, and the meals were free and chosen for them. It shows that an ultra-processed diet can lead people to eat more in the short term, not what happens over years. In our view, the evidence supports eating less ultra-processed food, especially foods high in saturated fat, salt or sugar, without treating every ultra-processed food as equally harmful.

How to eat less ultra-processed food

These steps come from NHS advice, with our own notes marked.

1
Have the less healthy ones less often
The NHS says most people would benefit from eating less ultra-processed food that is high in saturated fat, salt or sugar, eaten less often and in smaller amounts.
2
Build meals on whole foods
The NHS describes a healthy diet as plenty of fruit, vegetables and wholegrain or higher fibre starchy foods such as wholewheat pasta, brown rice and whole oats. It also lists wholemeal bread, higher fibre cereals and baked beans as foods that can be part of a healthy diet.
3
Use the label
The NHS suggests checking food labels to compare calories, saturated fat, salt and sugar, and some labels use colour-coding. In our view, a long ingredient list with preservatives, sweeteners or emulsifiers is a sign of ultra-processing.
4
Plan for busy nights
In our view, cooking extra portions and freezing them gives you a quick option on the nights a ready meal would otherwise win. The NIH notes that less-processed food often takes more time and money to prepare.

Frequently asked questions

Are all processed foods bad for me?
No. The NHS explains that most foods are processed in some way and that not all processed foods are unhealthy. It lists wholemeal sliced bread, wholegrain or higher fibre breakfast cereals and baked beans as ultra-processed foods that can be part of a healthy diet.
Does the evidence prove these foods cause disease?
No. The large reviews are built on observational studies, and most results were rated low or very low quality. A small ward trial showed that people ate more and gained weight on an ultra-processed diet over 2 weeks.
Do I have to cut them out completely?
The NHS advice is to eat ultra-processed foods high in saturated fat, salt or sugar less often and in smaller amounts, not to cut out every processed food. In our view, you do not need to label foods as good or bad. If food rules make you anxious, talk to your GP (general practitioner) or a registered dietitian.
Why might these foods lead to overeating?
In the Hall trial, people ate faster on the ultra-processed diet, and its solid foods were more energy dense. Differences in protein might also explain part of the effect. The exact cause is not yet known.
How do I spot ultra-processed food?
The NHS says these foods often include ingredients you would not usually have at home, such as preservatives, sweeteners and emulsifiers. In our view, the ingredient list helps you spot them, and the colour-coded nutrition label for fat, saturated fat, salt and sugar is a quicker guide to which ones to have less often.

The bottom line

A 2024 review of studies in nearly 10 million people linked higher ultra-processed food intake with harm in 32 of 45 pooled results, including death, death from heart disease, type 2 diabetes and poorer mental health (mostly from studies that cannot show which came first), though most of the evidence was rated low or very low quality and all of it was observational. A 2025 review of observational studies found a 15% higher relative risk of death at the highest intake compared with the lowest. A small 2019 trial found that people ate about 508 more calories a day on an ultra-processed diet. In our view, eating less ultra-processed food that is high in saturated fat, salt or sugar, and building meals on fruit, vegetables and wholegrains, is a sensible step in line with NHS advice.

Get a nutrition plan built around your week
DT Fitness London builds nutrition plans around your goals, your schedule and the evidence.
Book a consultation at dtfitnesslondon.com

Related reading

Sources

  • 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. pubmed.ncbi.nlm.nih.gov/38418082
  • Liang S, Zhou Y, Zhang Q, Yu S, Wu S. Ultra-processed foods and risk of all-cause mortality: an updated systematic review and dose-response meta-analysis of prospective cohort studies. Systematic Reviews 2025;14(1):53. pubmed.ncbi.nlm.nih.gov/40033461
  • Hall KD, Ayuketah A, Brychta R, et al. Ultra-processed diets cause excess calorie intake and weight gain: an inpatient randomized controlled trial of ad libitum food intake. Cell Metabolism 2019;30(1):67-77.e3. pubmed.ncbi.nlm.nih.gov/31105044
  • NHS. Processed foods. Last reviewed 12 June 2023. nhs.uk
  • National Institutes of Health. NIH study finds heavily processed foods cause overeating and weight gain. 16 May 2019. nih.gov
Back to blog