A Q&A with the co-author of the new book ‘Food Intelligence’
“[STAT’s Sarah Todd] Are there food policies being floated by MAHA right now that you think are particularly right or particularly wrong?
[Former NIH researcher on ultra-processed foods Kevin Hall] I hope that there really is something horrible about these food dyes that they’re going to be able to address by replacing them. But someone from the FDA I had a conversation with said that they were worried that if we replace synthetic food dyes with natural food dyes, the natural food dyes are coming from plants. And plants accumulate heavy metals depending on the soil that they’re grown in. Because we have to concentrate these food coloring dyes so highly, are they going to come with an increased heavy metal exposure? I don’t know the answer to that question. But in nutrition science, we always have to worry about the replacement.
[..] With the study that’s wrapping up now that I was leading at the NIH, we were able to design a diet that was very, very high in ultra-processed food yet didn’t cause people to overeat. And it had nothing to do with food dyes, it had nothing to do with food additives and things like that. It was targeting other properties of these foods.
One was calorie density, which tends to be high in ultra-processed foods because you extract a lot of water from them. And the other was hyper-palatable foods, which are defined as foods that cross certain pairs of nutrient thresholds.
[Todd] Like high in both sugar and fat?
[Hall] Exactly. So by designing a diet that was still high in ultra-processed foods but low in both of those properties, calorie density and hyper-palatable foods, that suggested people don’t overeat on those kinds of diets, and yet they report the meals being equally pleasant. So that says people don’t have to give up on the taste and how much they like the meals. If you design them in an intelligent way and you target the mechanisms by which these foods cause overeating, then we can give policymakers and regulators the correct targets to go after when it comes to decreasing obesity rates, for example. [..]
[Todd] [..] What kind of reformulation do you think would be most impactful, and what do you think about what MAHA’s been up to so far?
[Hall] [..] What we should be focusing on are the things in our diet that constitute a large chunk of the calories.
There have been some modeling studies that have suggested if you actually did improve the nutritional quality of pizza, on average, at the population level, that could have a significant impact on public health. Mike Lean has done this in the U.K. with a nutritionally good-for-you pizza that’s still tasty — it costs a little more, but we could subsidize that if we were smart about our policies. [..]
[Todd] What do you see as some of the most harmful narratives that we have still around nutrition, weight, and weight loss?
[Hall] [..] There are lots of open questions that still need really good science to unpack. How much of the variability in glucose that we see in people’s blood is harmful if you don’t have diabetes? That’s another thing that the MAHA folks are promoting, that everybody should have a continuous glucose monitor. All this wearable stuff, the science on that is not really all that great in people without diabetes, and in fact, it could be potentially damaging for some people.
Again, it’s this idea that if only you knew enough about your own body, then we don’t have to worry about the food environment. We don’t have to worry about these systemic problems that are driving population health downhill. If you happen to be one of the fortunate and privileged few that could just learn enough to biohack your way to health, then that’s the solution for everybody.
This has been my criticism of some of the investments in human nutrition science at the NIH. They spent more than $170 million on this precision nutrition health program, which is coming up with algorithms that will make the perfect diet for you. It’s like, OK, that’s just going to make things more complicated and expensive for people. And it’s not going to be a public health intervention.”
Full article, STAT+, 2025.7.30