There are many reasons to criticize Health Secretary Robert F. Kennedy Jr.’s catastrophic health policies, but he might get credit for one thing: He has spawned a real movement away from ultra-processed foods, which are linked to multiple chronic diseases. Driven in part by his advocacy, state legislatures from California to Texas are moving to curb chemical-laden products in schools.
Here’s the problem: There’s never been an agreed-upon definition for what counts as ultra-processed. One group of researchers, however, might just have a solution.
Christina A. Roberto, director of the University of Pennsylvania’s Center for Food and Nutrition Policy, explained that policymakers tend to define ultra-processed food by focusing on the specific chemicals each food contains. That might seem reasonable enough: Identify a list of dyes, preservatives and other additives, and label any product containing them as ultra-processed.
But companies can easily find work-arounds to this approach, Roberto told me. “There’s this very predictable industry response: Industry will constantly innovate and evolve and just make ingredients that aren’t on the list,” Roberto said.
For example, after the Food and Drug Administration banned red dye No. 2 in the 1970s over cancer concerns, manufacturers shifted to red dye No. 3, which remained in use for nearly five decades despite similar safety questions. And when countries barred bisphenol A (also known as BPA) from baby bottles because it could interfere with hormones, companies replaced it with other bisphenols which can have comparable effects on the endocrine system.
In a recent Nature Medicine article, Roberto and two co-authors proposed a different approach: Instead of trying to define what qualifies as ultra-processed, define what does not qualify.
Take yogurt. The traditional method would involve comparing its ingredients against a long list of additives that could pose health risks. Under the alternative framework, researchers would specify what a non-ultra-processed yogurt would look like. In addition to milk and live cultures, permissible ingredients might include fruit, nuts and nutritive sweeteners such as honey. Anything besides that, such as modified food starch, dextrose or dyes, would automatically render the product ultra-processed.
Another example is bread. A non-ultra-processed loaf could contain ingredients such as unbleached flour, grain or seed meals, yeast, oils pressed from seeds or nuts, butter, molasses, baking soda, eggs and milk. Add in components outside that defined set, and the product would be considered ultra-processed.
The same framework could be applied across the food supply. Roberto envisions building a category-based system that mirrors how nutrition standards are implemented. Foods would be grouped into categories based on how they are produced and consumed, and experts would specify, for each type of food, what qualifies as non-ultra-processed. Those definitions, updated periodically as manufacturing practices evolve, could then be used to set procurement rules and school-meal standards.
Over time, policymakers could also distinguish levels of risk within the ultra-processed category. Some of these products may be more nutritious than others. A sweetened yogurt, for example, may provide protein and calcium even if it is ultra-processed. Potato chips, by contrast, tend to be high in sodium and saturated fat, and offer little nutritional value. Those additional characteristics — sodium, added sugar, saturated fat — could be regulated separately, layered on top of the ultra-processed definition.
Roberto’s approach comes with several advantages. First, it would make it far more difficult for companies to sidestep regulation through minor reformulations, and it would shift their incentives. Rather than encouraging companies to invent new ingredients that have not yet been restricted, it would reward the development of non-ultra-processed foods.
Second, it would be more manageable for regulators, who would not need to constantly chase an ever-expanding list of additives. And, crucially, it could also bring clarity for consumers. Health-conscious shoppers who want to avoid ultra-processed foods often struggle to identify them. A category-based system, especially if paired with labeling requirements, would make it easier to tell which foods meet the standard and which do not.
Last year, the Trump administration announced it would issue a uniform definition for ultra-processed foods. If Kennedy truly wants to improve the nation’s diet, he should seriously consider Roberto’s framework, which would help consumers make informed choices and close off industry loopholes. If he fails to deliver on this issue, it will become yet another footnote in his legacy of profound public health failures.
The post What even are ‘ultra-processed’ foods? Here’s a workable definition. appeared first on Washington Post.
