Ultra-processed food: why the debate needs less fear and more clarity
For many people interested in health and wellbeing, the idea of ultra-processed food, or UPF, has become more than a technical term in nutrition research. In public debate, it often serves as shorthand for wider concerns about modern, industrially produced food.
Those concerns are not baseless. A large body of research has found associations between high UPF intake and poorer health outcomes. But the evidence is not always easy to interpret. Many studies rely on self-reported diets and struggle to separate the effects of processing from nutrient quality, eating patterns and wider social factors. The evidence points to the need for more careful use of the term.
In the US, the Food and Drug Administration and the Department of Agriculture began a formal process in 2025 to develop a uniform federal definition of ultra-processed foods, arguing that no single authoritative definition exists for the US food supply. The central question is: what exactly makes a food “ultra-processed”? Is it the ingredients it contains, the way it is made, the extent to which it has been altered from its original structure, or some combination of these?
This helps explain why the topic has become so divisive. Within nutrition research, there is no consensus on how far the UPF category should guide policy or individual dietary advice. Some researchers see it as an important way of identifying harmful patterns in modern diets. Others argue that it is too broad to serve as a sound basis for dietary guidance on its own.
That distinction is important. A category can be useful for tracking population diets while still being too blunt to tell someone whether a particular product belongs in their shopping basket, especially when it tries to capture ingredients, industrial processes, product formulation, marketing, palatability and dietary patterns within one category.
There are also valid concerns about the role of large food companies in shaping diets and public health. Many highly processed products are designed to be cheap, convenient, heavily marketed and easy to overconsume. But the political and commercial problems of the food system are not identical to the scientific problem of classification.
A better approach would distinguish more clearly between products that are ultra-processed and nutritionally poor, products that are ultra-processed but may still have a useful place in the diet, and minimally processed foods that people are encouraged to eat more of. This might include some fortified foods, high-fibre breads or medical nutrition products, depending on their composition and use.
One way to balance warnings about UPFs is to give more attention to positive dietary guidance. In the EAT-UP framework, I propose the term “unrefined plant foods”, or UPs, to describe plant foods whose natural structure remains largely intact. These include whole fruits, vegetables, beans and grains that have not been heavily broken down or reconstituted.
This is not a replacement for the UPF framework. Its main value may be communicative: it balances advice about what to limit with clearer guidance on what to add. Many dietary guidelines already encourage people to eat more fruit, vegetables, legumes and whole grains. Naming these foods more precisely may help make that advice clearer.
Like any food category, unrefined plant foods would need careful definition. The phrase “largely intact” is not self-explanatory, and different researchers, policymakers and consumers may draw the boundary differently. But the value of the concept lies in shifting part of the public health message from avoidance to addition.
Advice based only on avoidance can easily become confusing or punitive. Evidence that higher intakes of whole plant foods are linked with better health also has limitations, including food diaries, self-reporting, cohort studies and the difficulty of separating diet from wider lifestyle factors. Even so, fruit, vegetables, legumes and whole grains are consistently supported across dietary guidelines, public health research and long-standing evidence on diet quality.
These debates also shape how people understand food in everyday life. Dietary advice should avoid creating unnecessary fear around food. When processing is treated as inherently dangerous, the result can be confusion, guilt and anxiety rather than healthier behaviour. In some cases, highly moralised food messaging may even encourage disordered eating patterns, including an unhealthy fixation on foods perceived to be perfectly pure or healthy.
This is also why language needs care. Phrases such as “real food” are often used to mean foods that are minimally processed or close to their original form. But the phrase can also carry assumptions about what counts as proper eating and who is getting it wrong. Public health messages need to take account of differences in income, time, access and daily constraints.
Read more: Why stigmatising ultra-processed food could be doing more harm than good
Improving diets requires more than labelling a broad category of foods as harmful. It requires careful consideration of evidence, behaviour and context. The challenge is to produce advice that is scientifically sound, practical to follow and responsive to the real conditions in which people make food choices.
The UPF debate has rightly placed industrial diets and food quality at the centre of public health discussion. The next step is not to abandon the framework, but to improve it: to define categories more clearly, distinguish between different kinds of processing, and combine warnings about harmful products with practical advice about the foods people can eat more of. In practice, that means combining processing-based classifications with evidence about nutrient profile, fibre content, additives, marketing and the role a food plays in the overall diet.
Beverley O'Hara does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.