Recently the city of San Francisco sued several food giants (Kraft Heinz, Mondelez and Coca-Cola) for deceptively marketing “ultraprocessed foods” and then externalizing the resulting health care costs onto the public. They cite a recent Lancet review which concludes:
“The totality of the evidence supports the thesis that displacement of long-established dietary patterns by ultra-processed foods is a key driver of the escalating global burden of multiple diet-related chronic diseases.”
What are ultraprocessed foods, and what is the evidence base for their health effects? I don’t think this story is as clear as the Lancet authors claim. There is likely some real signal here in the data, but the details are fuzzy.
The first problem is that there is no operational definition of “ultraprocessed”. The basic idea is these are foods that are highly processed and ready-to-eat, as opposed to “whole” foods that are minimally processed, such as raw vegetables. But the processing of good into a ready-to-eat form is not itself a health risk. Processed is being used as a marker for specific features that are common but not universal in processed foods. As the Lancet authors summarize:
“…overeating driven by high energy density, hyper-palatability, soft texture, and disrupted food matrices; reduced intake of health-protective phytochemicals; and increased intake of toxic compounds, endocrine disruptors, and potentially harmful classes and mixtures of food additives.”
But of course not all processed foods have all of these features. It is also not clear if each of these features represents a genuine health risk by themselves. For example, a lot of attention has been focused on food dyes, with RFK and the FDA planning to phase out 8 “artificial” food dyes by 2026. The evidence, however, does not support a specific health risk from any of the dyes, they are just being targeted for being “artificial”. Prior to the MAHA takeover, the FDA approved food dyes based on adequate safety data. Any small potential for adverse effects would only come from consuming unrealistically large amounts of such dyes. But they are easy to demonize.
Given that there are multiple regulatory agencies reviewing and updating the evidence, existing food additives are unlikely to present any significant health risk. If there is evidence for a specific risk from a specific additive, the regulations can easily be updated. Essentially the Lancet authors are criticizing processed foods for containing ingredients that have already been reviewed and approved.
Also, demonizing specific ingredients is also not science-based or rational. It results in measures that are unlikely to have any health benefit, and in fact may be counterproductive. For example, Coke has promised to replace corn syrup with cane sugar in their American products. This is essentially complete nonsense – sugar is sugar, but making people believe that cane sugar is somehow more healthful because it’s natural is deceptive.
Where there is some legitimacy, I think, is in the fact that processed foods can often be calorie dense. People can end up consuming far more calories than they think, and this can contribute to obesity. The evidence for this is mostly observational, and this make it difficult to form firm conclusions. Also, may foods in the processed category are obviously high calorie – no one should be shocked that Twinkies are calorie-dense. But even if we assume this connection is real, and not just due to confounding factors, there are solutions other than banning processed food. Processed food is cheap and convenient, that is why they are so popular, and banning them so that people are forced to buy fresh ingredients and cook is unlikely to go over well.
There are other less radical solutions. The first, which is already partly in place, is transparency – food labels should clearly indicate how many calories are in the food. Often this can be a bit deceptive, with counterintuitive serving sizes, but easily remedied with regulation. Labeling of other ingredients, such as salt, fat, and sugar content, is also helpful.
Perhaps we need to take this transparency issue further, however. Calorie dense foods could be easily indicated with a prominent warning label. This would motivate companies to stay below the threshold for being considered calorie dense. Further, lower calorie foods could also be indicated on the label, which food companies could chase by reformulating their products.
There may be other measures as well, to help people make healthier decisions easier and to avoid competition leading to the most calorie-dense foods possible.
What I fear from the current war on ultraprocessed foods is that they are focused on the wrong things and/or are making unrealistic policy recommendations. Banning food dyes and promoting the use of cane sugar are perfect examples of how these efforts can go wrong. Food companies will also happily makes these essentially cosmetic changes without altering the essence of their products. Telling people to buy more fresh produce and spend more time cooking (while these are things I personally endorse) is unlikely to achieve the hoped-for results. Cheap and easy are hard to give up.
