How should I read a page where the grades disagree?

Start with the strong-evidence entries, because those are the findings most likely to still be true in five years. Then read the contested ones to understand what the honest objections are — not to dismiss them. The order matters. Reading the contested material first tends to produce false confidence in whichever side you already preferred; reading it after the replicated work gives you a baseline to measure it against.

Does this site take a position on ultra-processed food?

Where the literature is clear, yes — and the papers behind that position are on this page. Where it genuinely is not clear, the page says so rather than manufacturing certainty. The commitment is narrower than it sounds: we will not describe a contested question as settled to make a protocol sound more impressive, and we will not bury a trial because it complicates the story.

Why are some of these papers decades old?

This topic spans 2018 to 2025 across 5 journals. Age is not a defect in physiology research — a well-run metabolic-ward study from the 1990s can be more informative than a large observational survey published last year. What matters is the design and whether the finding has held up. Where an older paper has been superseded or failed replication, its grade is lowered rather than the entry being silently deleted.

What should I actually do with this?

Use it to ask better questions. A citation library describes what happened to groups of people in trials; it cannot tell you what is happening inside your body, what interacts with your medication, or what your bloodwork means. The practical route is to read the papers that apply to your situation, then take the specific questions they raise to a clinician who can examine you and look at your labs.

Ultra-Processed Food · 8 cited studies

Research on Ultra-Processed Food

Peer-reviewed research on ultra-processed food and metabolic health — Hall 2019, Dicken 2025, Monteiro NOVA, plus the BMJ umbrella reviews.

See the studies All research

Papers

Strong evidence

18–25

Years covered

01The state of the evidence

Where this literature sits

Ultra-processed food (UPF) is the most consequential dietary variable identified in the last decade of metabolic-health research. The defining study — Hall et al. (Cell Metabolism, 2019) — fed inpatients an ad-libitum UPF diet versus a matched whole-food diet and found a ~500 kcal/day surplus on UPF, with weight gain in two weeks. Dicken et al. (UPDATE trial, Nature Medicine 2025) extended this to outpatient settings: even when UPF diets were aligned with national dietary guidelines, they still produced more weight gain than minimally processed equivalents.

Multi-cohort meta-analyses (Lane 2024 BMJ umbrella review, Touvier 2023) link UPF intake to cardiovascular mortality, type-2 diabetes, depression, and certain cancers. The mechanism isn't a single nutrient — it's the package of low fibre, low protein density, hyperpalatability, fast eating rate, and additive load that drives overconsumption. Below: every major UPF study cited on this site, plus the underlying NOVA classification framework.

Evidence mix on this topic

Strong evidence1
Moderate evidence7

Years covered2018–2025

Journals5

Paywalled summaries0

02The studies

Every paper behind the claim

Filter by grade or flip the order. Each card links straight to PubMed, the DOI or the journal — read the original rather than our summary of it.

Moderate evidence2025
Ultra-processed or minimally processed diets following healthy dietary guidelines on cardiometabolic health (UPDATE trial)
Dicken SJ et al. · Nature Medicine
Crossover RCT: UPF diet matched to dietary guidelines still produced more weight gain than minimally processed diet.
Ultra-Processed FoodSource ↗
Moderate evidence2024
Ultra-processed food exposure and adverse health outcomes: umbrella review of epidemiological meta-analyses
Lane MM et al. · BMJ
Umbrella review across 32 health outcomes: convincing evidence for UPF link to cardiovascular mortality, T2D, common mental disorders.
Ultra-Processed FoodSource ↗
Moderate evidence2023
Ultra-processed foods and cardiometabolic health: public health policies to reduce consumption cannot wait
Touvier M et al. · BMJ
Editorial-review summary of UPF cardiometabolic evidence calling for policy action; useful as orientation reference.
Ultra-Processed FoodSource ↗
Moderate evidence2020
Ultra-processed food consumption and risk of obesity: a prospective cohort study of UK Biobank
Rauber F et al. · European Journal of Nutrition
UK Biobank cohort (~22K adults): higher UPF intake associated with elevated obesity risk over 5-year follow-up.
Ultra-Processed FoodSource ↗
Strong evidence2019
Ultra-Processed Diets Cause Excess Calorie Intake and Weight Gain: An Inpatient Randomized Controlled Trial of Ad Libitum Food Intake
Hall KD et al. · Cell Metabolism
Inpatient RCT: ad-libitum UPF diet caused ~500 kcal/day surplus and 0.9 kg gain in 2 weeks vs matched whole-food diet.
Ultra-Processed FoodSource ↗
Moderate evidence2019
Ultra-processed foods: what they are and how to identify them
Monteiro CA et al. · Public Health Nutrition
Defines the NOVA classification system used to identify ultra-processed foods across the literature.
Ultra-Processed FoodSource ↗
Moderate evidence2019
Ultra-processed food intake and risk of cardiovascular disease: prospective cohort study (NutriNet-Santé)
Srour B et al. · BMJ
French cohort (~105K adults): each 10% UPF increase was associated with a 12% rise in cardiovascular disease risk.
Ultra-Processed FoodSource ↗
Moderate evidence2018
Consumption of ultra-processed foods and cancer risk: results from NutriNet-Santé prospective cohort
Fiolet T et al. · BMJ
10% increase in UPF intake associated with 12% higher cancer risk in 105K-adult French cohort.
Ultra-Processed FoodSource ↗

03Related research

Nothing metabolic sits on its own

Sleep moves insulin. Stress moves visceral fat. Protein moves total intake. The other shelves in the library are usually where the missing half of the answer lives.

The full library Essays on this

04Straight answers

Ultra-Processed Food

Reading this evidence honestly

Of the 8 papers filed under this topic, 1 carry a strong-evidence grade — replicated trials or large meta-analyses — and 0 sit in the contested or mechanism-only tiers where the human outcome has not been settled.

A mixed distribution is normal and honest. Topics where every paper is graded strong are usually topics where the inconvenient papers have been quietly left out.

Strong evidence13%

1 of 8 papers

Moderate evidence88%

7 of 8 papers

Evidence grades describe the strength of the research, not a promise about your outcome.

Papers describe populations. Your labs describe you.

The free assessment takes the research above and asks which parts of it apply to your pattern — then names the protocol the evidence actually supports for it.

Take the free assessment Free calculators

Every claim citedOpen the paper, check us.

Graded honestlyContested is labelled contested.

No guaranteed numbersIndividual results vary.

Educational only, not medical advice. Citing a study is not a recommendation to act on it. Consult your physician before major dietary changes or altering prescribed medication. These statements have not been evaluated by the Food and Drug Administration. Individual results vary; no outcome is guaranteed.

Research on Ultra-Processed Food

Where this literature sits

Every paper behind the claim

Ultra-processed or minimally processed diets following healthy dietary guidelines on cardiometabolic health (UPDATE trial)

Ultra-processed food exposure and adverse health outcomes: umbrella review of epidemiological meta-analyses

Ultra-processed foods and cardiometabolic health: public health policies to reduce consumption cannot wait

Ultra-processed food consumption and risk of obesity: a prospective cohort study of UK Biobank

Ultra-Processed Diets Cause Excess Calorie Intake and Weight Gain: An Inpatient Randomized Controlled Trial of Ad Libitum Food Intake

Ultra-processed foods: what they are and how to identify them

Ultra-processed food intake and risk of cardiovascular disease: prospective cohort study (NutriNet-Santé)

Consumption of ultra-processed foods and cancer risk: results from NutriNet-Santé prospective cohort

Nothing metabolic sits on its own

Reading this evidence honestly

Papers describe populations. Your labs describe you.