UPFs We Do Not Need a Definition, We Need an Algorithm
At the WSRO Annual Meeting in Verona, I joined a lively panel on ultra-processed foods with Marcia Terra and Dr. Ros Miller, moderated by Annie Denny. My central message was simple: the future of prevention should not depend on a rigid definition of UPFs but on smarter tools that help citizens make better choices.
The UPF debate is becoming a policy machine: a category becomes a label, a warning, a tax, a target for reformulation, and eventually a new definition of “healthy”. This risks turning complex questions of nutrition, lifestyle, and public health into regulatory shortcuts.
Processing is not the enemy. The real question is how food is consumed: how much, how often, in which food matrix, within which diet, with how much movement, and with which actual health outcomes. Public health should move beyond fear-based categories and toward food literacy, positive nutrition, AI, personalized prevention, and consumer empowerment.
Key Takeaways
- We do not need a definition. We need an algorithm.
- Citizens need tools that help them choose, not rigid classifications that replace judgment.
- Processing is not the enemy.
- Processing has always been part of human food culture: preservation, safety, accessibility, innovation, and pleasure.
- Context matters more than categories.
- Portion, frequency, food matrix, diet, movement, lifestyle, and individual needs are essential.
- Market effects are not health outcomes.
- Changing labels, sales, or reformulation is not the same as reducing obesity, diabetes, CVDs, or NCD burden.
- The alternative is not inaction. It is smarter prevention: food literacy, positive nutrition, movement, AI, wearables, personalized prevention, and empowerment.
READ MY NEWSLETTER PAGANINI NON RIPETE: WE DO NOT NEED AN UPF DEFINITION
A very lively and insightful first panel at the WSRO Annual Meeting in Verona allowed me to discuss one of the most controversial issues in today’s food and public-health debate: ultra-processed foods.
It was a pleasure to exchange views with the legendary Marcia Terra and Dr. Ros Miller, moderated by Annie Denny, in a conversation that was both serious and genuinely enjoyable.
My key message was simple:
We do not need a definition. We need an algorithm.
This does not mean replacing citizens with technology. It means the opposite. We need smarter tools that help people understand food, portion, frequency, food matrix, movement, personal needs, and lifestyle context.
The UPF debate is no longer only about science. It is becoming a policy machine.
A category becomes a label.
A label becomes a warning.
A warning becomes a tax.
A tax becomes reformulation pressure.
Reformulation becomes compliance.
Compliance becomes the new definition of “healthy”.
This is the problem.
FOOD IS STRATEGIC INFRASTRUCTURE: READ MY TALK AT THE ASSICA GENERAL ASSEMBLY IN ROME
A classification created for research can quickly become a regulatory shortcut. It can shape labels, taxes, restrictions, advertising rules, procurement policies and reputational pressure. Once this happens, the debate shifts from understanding food to managing behaviour.
Processing is not the enemy. Human beings have always processed food: cooking, drying, fermenting, grinding, baking, preserving, freezing, and pasteurizing. Processing has improved safety, extended shelf life, increased accessibility, supported innovation and created pleasure.
The real question is not whether a food is processed. The real question is:
How much?
How often?
In which food matrix?
Within which diet?
With how much movement?
With what lifestyle?
With which personal needs?
And with which actual health outcomes?
A rigid UPF definition cannot answer these questions. A well-designed algorithm could.
Definitions can be useful for research and monitoring. But citizens do not live inside definitions. They live inside habits, routines, constraints, budgets, cultures, pleasures, and bodies. A static classification cannot fully capture portion size, frequency, total diet, physical activity, sleep, stress, metabolic variability, or personal goals.
ARE WE LIVING IN A SIMULATION?
This is why the future of prevention should not be based only on classifying food. It should be based on helping people make better decisions in real life.
The UPF debate also risks repeating a familiar mistake in nutrition policy: reducing a complex health challenge to one visible target. This is politically attractive because it is easy to communicate. But simplification is not the same as understanding.
Obesity, diabetes, cardiovascular diseases and NCDs are complex and multifactorial. They cannot be explained by one nutrient, one ingredient, one product category or one processing classification.
This does not mean that all products are equal. It means that policy should be proportionate, evidence-based and capable of distinguishing between symbolic action and actual health outcomes.
- Changing a label is not the same as reducing obesity.
- Changing a recipe is not the same as improving diet quality.
- Changing sales is not the same as reducing diabetes.
- Changing perception is not the same as building food literacy.
READ MY EDITORIAL: CHILE CAN LEAD AGAIN AND OVERCOME WARNING LABELS
Public policy should not be judged only by what it intends to discourage. It should be judged by what it actually changes in health, society, markets, consumers and value chains.
This is why unintended consequences matter. Policies based on broad categories may affect consumer understanding, food anxiety, SMEs, traditional products, innovation, trade, supply chains, affordability and trust in institutions.
A product may be reformulated to comply with a score or avoid a warning without becoming meaningfully healthier in a broader dietary sense. A company may innovate for compliance rather than quality. A consumer may learn to fear a category instead of understanding a diet.
The alternative is not inaction. It is smarter prevention.
A better agenda should focus on food literacy, positive nutrition, movement, portion and frequency, responsible innovation, AI, wearables, personalized prevention and consumer empowerment.
Food is biology.
Food is culture.
Food is technology.
Food is a pleasure.
Food is a responsibility.
Food is a choice.
WHAT POLICIES FOR AFFORDABLE DIETS: MY TALK AT SMEs CONNECT
UPF should not become the new tobacco. Food is not tobacco. Food is essential to life, culture, identity, and social relations.
The question is not whether some products should be improved, consumed less frequently, or better understood. Of course they should.
The question is whether we want public health to be built on fear-based classifications or on knowledge, responsibility and better tools.
We do not need a definition. We need an algorithm.
