As policymakers work to define what constitutes an “ultra-processed food” (UPF) for future dietary guidance, one important question remains unresolved: should differences in how cooking oils are produced be a factor?
The Dietary Guidelines for Americans (2025–2030) urge consumers to minimize UPF consumption. Yet there is no federal definition of what constitutes a UPF. Last October, FDA and USDA solicited public comments on that very question.
Last week, a panel of health experts convened by Healthy Eating Research (HER) issued a report recommending that the federal government use the NOVA classification system, developed in Brazil, as the basis for identifying UPFs.
While NOVA considers multiple factors, including ingredients and processing techniques, the HER panel recommends a policy-oriented implementation of NOVA that would classify a food as ultra-processed if it contains at least one “marker ingredient,” defined as a cosmetic additive or ingredient of non-culinary use.
When it comes to cooking oils, however, this approach raises concerns.
First, NOVA itself has a significant limitation. As NAOOA noted in comments submitted to FDA and USDA, cooking oils vary widely in how they are produced. At one end of the spectrum is mechanically extracted, cold-processed and raw extra virgin olive oil. At the other are seed oils produced using petroleum-based chemical solvents and multiple high-heat processing steps.
Yet NOVA makes no distinction among cooking oils, classifying all of them regardless of how they are produced as “culinary ingredients,” implying there is no material difference in processing.
If federal policy follows this approach, it risks creating a backlash from the many health-conscious consumers who have become increasingly concerned about the industrial processes used to produce certain seed oils–not unlike the response we saw to the policy decision made concerning the herbicide glysophate.
Indeed, the cover story of the latest edition of INFORM, the magazine of the American Oil Chemists’ Society, asks whether more effective methods should be adopted to detect residual hexane in light of the focus on seed oil safety in the U.S.
Exempting cooking oils from UPF classification as NOVA has done could thus leave those U.S. consumers who are seeking to avoid using UPFs in the foods they prepare in the dark when it comes to cooking oils, and undermine public confidence in the government’s efforts to define and reduce UPF consumption.
Second, the HER panel’s proposed reliance on ingredient statements to find marker ingredients that would indicate the food is an UPF is problematic. To begin with, processing aids may not appear in ingredient statements when they are subsequently removed during production. When the processing aid that was used is intrinsically hazardous and could pose a safety concern if not properly handled and removed, the fact that it was used in the first place should still indicate that the product is an UPF.
For example, hexane that is commonly used as a processing aid in the extraction of some cooking oils would be hazardous to health if not properly removed. Thus, should cooking oils be subject to UPF classification as I have argued, oils that are extracted with the use of hexane should arguably be classified as UPF regardless of whether the hexane is required to be included on the label or not.
Another problem with focusing solely on ingredient declarations should cooking oils be subject to UPF classification is that it overlooks the health implications of processing techniques themselves.
Solvent-extracted oils are typically exposed to high temperatures multiple times during production: during extraction, during solvent removal, and again during refining. Other oils may be heated only during refining. The extent of these thermal processes should be taken into account given that they may alter the chemical composition and stability of oils and contribute to the formation of undesirable compounds like trans fats. They may also impact the stability of the oil, increasing the likelihood that harmful compounds are produced when the oils are subsequently used in high-heat cooking applications.
The HER panel does offer a recommendation worth considering. It suggests that exemptions from policy measures aimed at reducing UPF consumption, such as restrictions on procurement by schools, might be provided to foods that meet the FDA definition for “healthy,” for instance. Under such an approach, a cooking oil that might be considered a UPF could be exempt if it is low in saturated fat. That may be a reasonable policy outcome.
But exemption from policy actions should not change a product’s UPF classification. Consumers deserve transparency about how foods are processed—including distinctions among cooking oils—regardless of whether the government classifies it as healthy.
A meaningful definition of ultra-processed food should include all foods, including cooking oils that NOVA designates as culinary ingredients, and should account for the use of marker ingredients and/or the degree of processing involved that could materially impact the healthiness of the food.
Only then can consumers truly make informed choices about the foods they bring into their homes.
