Carlos Monteiro, the University of São Paulo nutrition scientist who first coined the term “ultra-processed food”, wants these products sold with warnings like those on cigarettes. Ahead of the International Congress on Obesity in São Paulo in June 2024, Monteiro told the Guardian that front-of-pack warnings should be introduced “similar to those used for cigarette packs”.
Labels are only one part of the plan. Monteiro also called for public health campaigns modeled on those against tobacco, a ban or heavy limits on advertising, and an end to sales in schools and health facilities. The plan also includes heavy taxes on ultra-processed foods, with the money used to make fresh foods cheaper, according to the same Guardian report.
Who Carlos Monteiro is and what “ultra-processed” means
Monteiro and colleagues first used the term about 15 years before the 2024 congress, when they designed the Nova food classification, the Guardian reported. Nova sorts food by how much it is processed and why, instead of looking only at nutrients such as fat or sugar. It places every food and drink in one of four groups: minimally processed foods, processed culinary ingredients, processed foods, and ultra-processed foods, as the same report explains.
Ultra-processed foods are industrial products built from substances taken out of whole foods, often with additives such as flavours, colours, emulsifiers, and sweeteners that make them more appealing, according to a guide to Nova written by Monteiro’s team. Common examples are breakfast cereals, protein bars, fizzy drinks, ready meals, and fast food.
Why Monteiro compares ultra-processed food to tobacco
Monteiro set out three points of likeness. Both tobacco and ultra-processed foods cause many serious illnesses and early deaths. Both are made by large international companies that spend their profits on aggressive marketing and on lobbying against regulation. Both are harmful by design, so changing the recipe cannot fix them, in Monteiro’s view.
Monteiro also argued that food companies know their products must be cheaper, more convenient, and tastier than home cooking to compete. To maximise profits, the products must also be cheap to make and eaten in excess, Monteiro said. For that reason, Monteiro judges that studies and reviews alone no longer warn the public strongly enough.
What the evidence behind the warning shows
The largest recent summary of the evidence is an umbrella review of 45 pooled analyses in The BMJ, published in February 2024 by Lane, Marx, and colleagues at Deakin University in Australia. An umbrella review collects existing meta-analyses (studies that pool the results of many earlier studies). Together, the pooled analyses covered 9,888,373 participants, the same paper reports.
Higher intake of ultra-processed food was linked to worse results for 32 of the 45 health measures (71%), the review found. These covered death, cancer, and mental, breathing, heart, gut, and metabolic health. The review graded the evidence as convincing for a 50% higher risk of death from heart and blood vessel disease (risk ratio 1.50). According to the same review, the evidence was also convincing for type 2 diabetes, with a 12% higher risk per step up in intake (dose-response risk ratio 1.12).
Convincing evidence also linked ultra-processed food to anxiety (odds ratio 1.48) and to common mental disorders (odds ratio 1.53), the paper reports. A second tier of “highly suggestive” evidence covered death from any cause (risk ratio 1.21), death from heart disease (hazard ratio 1.66), and obesity (odds ratio 1.55), according to the review. The same tier included sleep problems (odds ratio 1.41), wheezing (risk ratio 1.40), and depression (hazard ratio 1.22), the authors report.
All of these studies were observational: they show that people who eat more ultra-processed food have more of these problems, and they cannot prove that the food is the cause. Using the GRADE system (a standard scale of how much trust a result deserves), the review rated 4 pooled analyses as moderate quality, 22 as low, and 19 as very low. The authors conclude that the findings give a reason to develop and test public health measures that cut ultra-processed food intake, the paper says.
The case against the tobacco comparison
Hilda Mulrooney, a reader in nutrition and health at London Metropolitan University, called the comparison “very simplistic” in the same Guardian report. No cigarette is safe, so the health case against tobacco is clear, Mulrooney argued. People need fat, sugar, and salt, which also give foods their structure and shelf life, so some foods are harder to reformulate. In Mulrooney’s view, people need food, just in smaller amounts than most of them now eat.
Why warning labels on ultra-processed foods make sense
Monteiro’s call rests on solid ground where it matters most. The links with heart deaths, type 2 diabetes, anxiety, and common mental disorders reached the top evidence class in the same BMJ review, the largest on the subject. A warning on the front of a pack is a cheap way to put that information where people make their choice. What the evidence does not yet show is how much a label would change what people buy, which is exactly the kind of test the BMJ authors ask for.
The tobacco comparison is not new on Gromeus: a UCL doctor has argued that ultra-processed food makers act like Big Tobacco. Update (2024): in September, Monteiro widened the call and argued that governments should tax ultra-processed foods like instant noodles and ice cream.
Update (2025): in November 2025, Monteiro led the first paper of a three-part Lancet series on ultra-processed foods and human health. Drawing on more than 100 prospective studies, meta-analyses, randomised trials, and lab studies, the paper concludes that diets built on ultra-processed foods are a key driver of the global rise in diet-related chronic diseases. According to the same paper, its two companion papers set out the policy actions needed to protect diets based on fresh and minimally processed foods.
Sources and related information
The Guardian – Ultra-processed foods need tobacco-style warnings, says scientist – 2024
Andrew Gregory’s report of 27 June 2024 carries Carlos Monteiro’s call for tobacco-style warnings, the other policy proposals, and the reply from Hilda Mulrooney of London Metropolitan University.
The BMJ – Ultra-processed food exposure and adverse health outcomes: umbrella review of epidemiological meta-analyses – 2024
Lane, Marx, and colleagues graded 45 pooled analyses covering 9,888,373 participants and report the risk ratios, odds ratios, and evidence classes cited above (BMJ 2024;384:e077310, DOI 10.1136/bmj-2023-077310). The open-access full text is on PubMed Central.
The Lancet – Ultra-processed foods and human health: the main thesis and the evidence – 2025
Monteiro and colleagues opened the Lancet series of November 2025 with their review of the evidence on ultra-processed diets, which concludes that these diets are a key driver of diet-related chronic disease.
Public Health Nutrition – Nutrition and health. The issue is not food, nor nutrients, so much as processing – 2009
This commentary by Carlos Monteiro is the early paper arguing that processing matters for health, the idea behind the Nova classification.
Public Health Nutrition – Ultra-processed foods: what they are and how to identify them – 2019
Monteiro’s team wrote a practical guide to spotting ultra-processed foods under Nova, including the typical ingredients and additive classes used to define them.



