Study challenges UPF labels - many products may pose low health risk

A new peer‑reviewed study presented at the Nutrition Society Winter Meeting has raised questions about the way ultra‑processed foods (UPFs) are currently classified, arguing that widely used systems may oversimplify the health impacts of processed products.

Researchers from ZOE, a nutrition science company, analysed the diets of 550 adults alongside more than 13,000 branded food items. Their findings suggest that the NOVA system – the dominant framework for identifying UPFs – groups together foods with very different nutritional profiles, potentially leading consumers to avoid products that may not be harmful and, in some cases, could be beneficial.

The NOVA system categorises foods based on the extent and purpose of processing, rather than on measurable health outcomes. As a result, high‑fibre cereals and baked beans are placed in the same “ultra‑processed” category as confectionery and sugary snacks. According to the new analysis, nearly 80% of breakfast cereals are labelled as UPFs under NOVA.

However, when the same cereals were assessed using ZOE’s new Processed Food (ZPF) Risk Scale – which evaluates processing in relation to health markers – 64% were classified as low or no risk. The researchers say this highlights the limitations of treating all UPFs as nutritionally equivalent.

The ZPF scale divides foods into four categories, ranging from “no risk” to “high risk,” based on associations with cardiometabolic health indicators such as inflammation and body weight. In the study, participants who consumed the highest proportion of “high‑risk” processed foods had higher body mass index (BMI) and elevated inflammatory markers. Those who ate more “no‑risk” processed foods showed lower BMI and lower inflammation.

Tim Spector, professor of genetic epidemiology and scientific co‑founder at ZOE, said the findings point to the need for a more nuanced approach. “Instead of demonising all processed foods, the ZPF scale helps identify which ones can be part of a healthy diet and which are more likely to be harmful,” he said.

Dr Federica Amati, ZOE’s head nutritionist, argued that the current debate around UPFs risks confusing consumers. “UPFs account for around 60% of daily energy intake in many countries. It’s unrealistic to expect people to eliminate them entirely, and some processed foods can actually be beneficial,” she said, citing fibre‑rich cereals as an example.

The researchers say the ZPF scale could offer a more practical tool for consumers and policymakers by distinguishing between harmful and benign processing. Professor Sarah Berry, ZOE’s chief scientist and lead author of the study, said the new framework “captures specific processing features driving associations with health” and could support future public health guidance.

The findings add to a growing discussion about how best to classify processed foods in a way that reflects both nutritional quality and real‑world eating patterns. While the NOVA system remains widely used in research and policy, the authors argue that more evidence‑based, outcome‑focused tools may help consumers make clearer choices in an increasingly complex food environment.

Photo by Enkhjin photography on Unsplash