Ultra-Processed Foods and Your Health: What a Major 2024 Review Found

Reviewed by the HealthyMag Editorial Team. Last updated: July 2026.

What the 2024 BMJ umbrella review actually found

In February 2024, researchers led by Melissa M. Lane published an umbrella review in The BMJ titled “Ultra-processed food exposure and adverse health outcomes: umbrella review of epidemiological meta-analyses.” An umbrella review sits at the top of the evidence pyramid: instead of analysing individual studies, it synthesises existing meta-analyses. This one pulled together 14 meta-analyses comprising 45 distinct pooled analyses, drawing on data from 9,888,373 participants — close to 10 million people.

The headline result was striking for its breadth. Of the 45 pooled analyses, 32 (about 71%) showed direct associations between higher ultra-processed food (UPF) intake and worse health. The outcomes spanned mortality, cancer, and cardiovascular, metabolic, respiratory, gastrointestinal, and mental health. The authors then graded how trustworthy each link was, from “convincing” down to “weak,” based on statistical strength and consistency.

The strongest (“convincing,” Class I) evidence pointed to several specific harms. Higher UPF intake was associated with a roughly 50% greater risk of cardiovascular-disease-related death (relative risk 1.50, 95% CI 1.37–1.63), a higher risk of type 2 diabetes (relative risk 1.12 per increment, 95% CI 1.11–1.13), and notably higher odds of common mental disorders such as anxiety (odds ratio for combined common mental disorders 1.53, 95% CI 1.43–1.63). “Highly suggestive” evidence linked greater UPF intake to about a 21% higher risk of all-cause mortality (relative risk 1.21, 95% CI 1.15–1.27), as well as to depressive outcomes and obesity.

What “ultra-processed” actually means (the NOVA system)

“Ultra-processed” is not just a synonym for “packaged” or “processed.” The review uses the NOVA classification, developed by researchers at the University of São Paulo, which sorts foods by the extent and purpose of processing rather than by nutrient content. NOVA has four groups: unprocessed or minimally processed foods (fruit, eggs, plain milk, dried beans); processed culinary ingredients (oils, butter, salt); processed foods (canned fish, fresh bread, cheese); and ultra-processed foods.

Group four — ultra-processed — describes industrial formulations made largely from substances extracted from foods or synthesised in labs, often containing additives such as emulsifiers, artificial colours, flavourings, and sweeteners you would not find in a home kitchen. Typical examples include soft drinks, packaged snacks and confectionery, mass-produced breads, reconstituted meat products, instant noodles, and many ready-to-heat meals. The distinction matters: canned beans or plain frozen vegetables are processed but not ultra-processed, and the review’s concern is specifically with that fourth group.

Why ultra-processed foods might be harmful

The review reports associations; it does not pin down a single mechanism. But nutrition researchers have proposed several plausible reasons UPFs could genuinely affect health. Many are energy-dense and easy to overeat, delivering a lot of calories in a small volume. They tend to be low in fibre and high in refined starch, added sugar, salt, and certain fats. Some additives and by-products of processing are under investigation for effects on the gut lining and microbiome, though this research is still early.

Two behavioural factors also stand out. UPFs are often engineered to be hyper-palatable and are eaten quickly, which can blunt the body’s fullness signals and encourage larger portions — a pattern supported by a well-known controlled feeding study at the US National Institutes of Health, where people ate more calories and gained weight on an ultra-processed diet than on a minimally processed one matched for nutrients. And simply eating more UPFs tends to displace whole foods like vegetables, legumes, nuts, and fish, so the harm may come as much from what is missing as from what is added.

The honest caveats: association is not proof

This is where careful framing matters. Almost all the evidence in the umbrella review is observational. That means it tracks what people eat and what happens to their health, but it cannot, on its own, prove that UPFs directly cause disease. The signal is strong and consistent across millions of people and many outcomes, which is why it deserves to be taken seriously — but consistency is not the same as causation.

The biggest caveat is confounding. People who eat the most ultra-processed food also tend, on average, to smoke more, exercise less, sleep worse, earn less, and eat fewer vegetables. Good studies try to statistically adjust for these factors, but they can never fully separate the food from the life around it. The review’s authors themselves flagged that much of the underlying evidence was of low or very low methodological quality and called for better research.

It is also worth remembering that not all processing is equal, and NOVA has genuine critics. The category is broad: it lumps a sugary energy drink together with a wholegrain packaged bread or a fortified plant milk, even though their health profiles differ. Some scientists argue that nutrient content, not processing category alone, should drive dietary advice. The truth is probably that both matter.

The practical, non-alarmist takeaway

None of this justifies panic or guilt over every wrapper in your cupboard. Ultra-processed foods are woven into modern life, they are often cheaper and more convenient, and an occasional packaged snack is not going to undo an otherwise sound diet. The evidence points to a dose-response pattern — more UPF, more risk — which means the useful lever is proportion, not perfection.

Practically, that looks like building meals around foods that need little explanation: vegetables, fruit, beans and lentils, eggs, plain dairy, nuts, whole grains, fish, and unprocessed meats. Cooking a little more from scratch, swapping sugary drinks for water or unsweetened options, and reading ingredient lists (long lists of unfamiliar additives are a reasonable rough signal) all nudge the balance in a helpful direction. Small, sustainable shifts beat dramatic overhauls that do not last.

The realistic goal is not zero packaged food, but shifting the balance toward whole and minimally processed foods most of the time.

Frequently Asked Questions

What counts as ultra-processed food?

Under the NOVA classification used in the review, ultra-processed foods are industrial formulations made mostly from extracted or synthesised ingredients, typically with additives like emulsifiers, sweeteners, colours, and flavourings. Common examples are soft drinks, packaged snacks, confectionery, mass-produced breads, instant noodles, reconstituted meat products, and many ready meals. Canned beans, plain yoghurt, or frozen vegetables are processed but not ultra-processed.

Are ultra-processed foods really that bad?

The evidence is serious enough to act on: a 2024 BMJ umbrella review of nearly 10 million people found consistent links between higher UPF intake and about 32 adverse outcomes. But these are associations, not proof of direct harm, and the risk appears to rise with how much you eat. An occasional treat is not the problem; a diet dominated by ultra-processed foods is the concern.

Is all processed food unhealthy?

No. Processing is a spectrum. Freezing vegetables, canning beans, pasteurising milk, and baking plain bread are all forms of processing that can be perfectly healthy and even improve safety or nutrition. The review’s concern is specifically the “ultra-processed” fourth category, not processing in general.

Does this prove ultra-processed foods cause disease?

No. The umbrella review is built almost entirely on observational studies, which can show strong, repeated associations but cannot establish causation by themselves. People who eat the most UPFs also tend to share other risk factors like smoking and inactivity, and the authors noted much of the underlying evidence was of low quality. It is a compelling signal, not a verdict.

Which ultra-processed foods are worst?

The review graded outcomes rather than ranking individual products, so there is no official “worst” list. That said, sugar-sweetened beverages and processed meats consistently show up as higher-risk across the wider nutrition literature. In general, ultra-processed items that are energy-dense and high in added sugar, salt, or refined starch are the most reasonable to cut back on first.

How do I eat fewer ultra-processed foods?

Aim for gradual, sustainable swaps rather than an overnight overhaul. Build meals around whole and minimally processed foods, cook a bit more from scratch, replace sugary drinks with water, and glance at ingredient lists — a long list of unfamiliar additives is a rough clue you are holding an ultra-processed product. The goal is shifting the balance most of the time, not achieving zero.

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. doi:10.1136/bmj-2023-077310.
• Monteiro CA, Cannon G, Levy RB, et al. Ultra-processed foods: what they are and how to identify them (the NOVA classification). Public Health Nutrition. 2019;22(5):936–941.
• 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 (US National Institutes of Health).