Why Ultra-Processed Foods Make You Overeat Without Trying | Dr. Kevin Hall

Added: 2026-05-08

[00:00:00]How do we actually know that when you look at population studies and I think you you said earlier about 50% of the average Americans calories are coming from this category of of ultra-processed foods or maybe even more depending on age.

[00:00:15]How do we know that it's the foods themselves that are increasing risk of a cardiometabolic disease or some other outcome and it's not the fact that they're cheap and the the people that are consuming them are often of lower socioeconomic status and and other other variables that could be at play. I think that that's where you get into the limitations of different kinds of studies, right? So in these observational cohort studies for example, you can start to make statistical adjustments for for some of these things and try to see if the effect goes away or get stronger as a result of making those statistical adjustments. But at the end of the day, there are certain confounds that you can't get around, right? I mean, these are not experiments, right? You can analyze the data in different ways to try to get around it, but it's an important factor. I've argued that we need those kinds of studies as well as a sort of more controlled studies where we take certain factors out of the equation. So for example, you ask the question about, you know, these foods tend to be less expensive and that's certainly a factor. They tend to also be more convenient. They tend to also be a lot easier to prepare and you don't need as much equipment or skill to prepare them, right? All of those factors are important and so when I first started getting into this this topic, I was interested in, well, what if you remove all those things? And what if you just provide people with the foods themselves? Is there something about the foods themselves and you create meals that are highly you know, have a lot of calories from ultra-processed foods versus a lot of calories from minimally processed foods and no ultra-processed foods, but matching for the nutrients in the meals that you're giving people.

[00:01:58]Is there something over and above those other factors that that we mentioned that are important, you know, the price, the convenience and those kinds of things? Do people behave differently with these ultra-processed meals? And that was the first study that we decided to do in this topic. It ended up being the first randomized controlled trial on this question and what we found was that despite presenting people with the same number of calories, the meals were matched for a variety of nutrients of concern and giving them very simple instructions, just eat as much or as little as you'd like. You shouldn't be trying to change your weight and in fact, they didn't know what the primary outcome of the study was. We were giving them twice the number of calories that they would need to maintain their weight and what they didn't know is we were measuring their leftovers. Now, when the same people were exposed to the highly ultra-processed food environment, they were eating about 500 calories per day more on average than when the same people were exposed to the minimally processed food environment and gaining weight in the ultra-processed food environment and losing it in the minimally processed food environment. So that says that when at least in that one study and you know, there's there's very few of these kinds of randomized trials now that the food environment that people were in when it was composed of ultra-processed foods, 80% of the calories, they spontaneously chose to overeat calories and gain weight without knowing that they were doing so or intending to do so.

[00:03:19]And the same people were losing weight in the other case. And that's taking away, they didn't have to prepare the foods, buy the foods, they didn't have to have any skill and they rated them equally pleasant, which was also interesting, right? So it tells us something even beyond this, which I think is important, which is that the food environment that we find ourselves in is somehow integrating with our innate biology to shift where we're regulating our body weight and shift our appetite in very profound ways, which might give us some insights about, you know, why is it that we've had this increase in obesity prevalence over the past many decades? Is it something about the food environment and what is it about the food environment? And I think that just saying it's ultra-processed food in general is a bit of a cop out. I think we want to know what the mechanisms are.

[00:04:04]What is it about ultra-processed foods over and above the cheapness and the convenience and the fact that they're heavily marketed and all that sort of stuff. Our studies are suggesting there's something about the foods themselves that are causing people to overeat calories. And just just to kind of underscore a point that you you made there, those diets were matched for macronutrients, so carbs and fat and protein. So what was different >> off in the They were a little off in the protein. So let's talk about protein and then talk to us about what else was was different, I guess, in terms of the characteristics. Absolutely, sure. So one of the things that was pointed out afterwards is that the ultra-processed diet was a little bit lower in the amount of protein that we gave them. I think it was, you know, within a couple of percentage points difference. So practically matched for protein, but when you talk to folks like Stephen Simpson and David Rabinheimer, who are big proponents of this protein leverage hypothesis, which I think is, you know, a fascinating idea, you know, they were making claims that, oh, maybe you could explain the entirety of the effect based on the fact that we gave them a slightly lower protein content in the ultra-processed food environment context. I don't think that that's true.

[00:05:10]We did some analysis after the fact and I can tell you about a new study that we ran where we actually matched the protein a lot better and we see similar effects. So I don't think that that's the main reason that that happened, but it's important to point out that was a limitation of the study. We didn't quite match for protein. Even though we say like we matched for, you know, sugar content of the diets, which is true. But of course, there's a lot more added sugar in the ultra-processed environment and no added sugar in the or free sugars in the in the minimally processed diet condition, right? The sugars were contained in fruits and you know, those kinds of whole foods where it also comes along with a lot of fiber. And even though we matched for fiber in the two conditions, the sort of whole food condition had a lot of insoluble fiber and what we were doing in the ultra-processed diet condition was adding a fiber supplement, right? Which is a lot of soluble fiber. So even though, you know, it's one of those things in nutrition, right? You name something, but it actually has a lot more subcategories that might make a big difference and I think that's an important caveat when it comes to doing these kinds of nutritional studies. And what was some of the other main differences between the the foods or when you went away and kind of reflected back on the fact that participants were eating about 500 more calories per day on the ultra-processed diet despite doing your best to match those nutrients, what were some of the other things that that maybe kind of suggest why people were eating more calories? Yeah, so so first maybe I'll take a step back and say, you know, there's lots of different ways you can design these kinds of trials and and the way that we ended up doing that trial was not the way that I wanted to do it initially. Like I wanted to do it in such a way that we would take a ultra-processed product and then we would have our chefs prepare a minimally processed version of that same product.

[00:07:01]So we would have a food-by-food comparison, right? The minimally processed version with the ultra-processed version trying to match for the macronutrients on a food-by-food basis. When I proposed that idea, basically the kitchen staff revolted.

[00:07:17]They were like, there's no way we can do this. This is like a we would have to, you know, increase our staff by a factor of 10 or something like that and and it's just not a doable process. And so we actually iterated back and forth on, okay, well, if we can't do it that way, what if we took processed version of the of the food and an ultra-processed version and did it that way. So same food still and then the issue was, well, not all of those versions are in our our branded food database and so we would have to then chemically analyze the the ones that weren't in the food database so that we would be sure that we matched the nutrients. And that chemical analysis of food is not cheap, right? So then that becomes reasonable scale of of expense that we couldn't afford. So we ended up settling on this idea of, okay, well, look, we're going to choose very different foods in these different diets, right? So there's no equivalent of a chicken nugget on the minimally processed diet, right? That's that's clearly belongs in the ultra-processed diet. There's like a grilled chicken, obviously, in the minimally processed version, but there's no like chicken nugget equivalent in the minimally processed diet. And so there's very different foods in these two sets of meals. And so we created these 7-day rotating menus that on average over the course of the day were matched at this very high level for the amount of fat, carb, sugar, sodium, fiber and then also the overall energy density of the meals that we gave people, the number of calories per gram of stuff that we presented to them were matched. Now, later on people started to say, oh, well, wait a second. Yeah, you matched for overall energy density, but in the one case, you've got all these beverages that you're dissolving this fiber supplements in. That's the ultra-processed diet condition. But if I look on the plates, right? I'm not in the beverage glasses, on the plates, those those foods on the plates are higher in energy density.

[00:09:10]And the only reason you matched the overall energy density is cuz you were adding, you know, these these high-fiber lemonades in the ultra-processed diet condition. And there's this idea that energy density of beverages doesn't really contribute much to satiation. And so we should really look at the non-beverage energy density of the of the meals. And then yeah, it turns out that the ultra-processed diet had almost double the non-beverage energy density as compared to the minimally processed diet. And people have been studying this for years, Barbara Rolls's group probably most famously for studying how energy density contributes to calorie intake in a very powerful way. And then we had this other interesting discussion. So we all already told you about the protein discussion, right? So that was another factor that we were were potentially playing a major role.

[00:09:58]Kieran Ford was a co-author on that first paper and he's very well known for his work on texture of food and oral sensory processing. And the fact that ultra-processed foods tend to be softer, easier to chew and swallow, and you tend to eat them more quickly. And indeed, people were eating the meals in the ultra-processed diet condition on a gram per minute basis more quickly than the minimally processed diet condition. So, that was another factor that could play a role, especially in combination with energy density of the of the foods on the plate. And then finally, Tera Fazzino from University of Kansas, who I think has been on your program, She has, yes.

[00:10:37]>> up Yeah, so she said to me, "You know, I know you matched the overall diets for, you know, salt and sugar and fat and carbs, but what about the number of individual foods on the plate that cross these pairs of thresholds that she's defined as being hyperpalatable combinations, right? So, even though the overall meals were matched for these nutrients, what about the individual foods? Did you present people with more individual foods that were either high in both fat and sugar, fat and salt, or carbs and salt?" And it turned out that, yep, we certainly did. We were presenting people with more of these pairs of combinations. And since then, we've worked with folks like Jeffrey Brunstrom in the UK who have looked at this the ratio of carbs and fat in in the individual foods that we were preparing and how that might play a role as well as the micronutrient properties of the of the overall diet. So, there's a variety of differences between these two things and lots of theories about what the main contributor was behind why people chose to over-consume calories on ultra-processed diets. Okay, let me throw a few things back to you because there was there was a lot of in there, a lot of good stuff. So, just to tie this back to the beginning of the conversation, the reason this is important is that if we can understand what are the actual traits or characteristics that are driving when I what I found really interesting, and I want to double-click on this in a moment, driving a a kind of subconscious wanting for urge for for more calories despite equal pleasantness. If we can work out what that is, then we can perhaps provide advice to the food industry or regulation or whatever it is, incentives, to create this category of food in a way where people do not have that same urge to consume extra calories. Yeah, that's right. I mean, if we understood mechanisms by which ultra-processed foods in general are are thought to be causing harm, and that's still an open question, is there is there a causal relationship there? And we think that at least down the pathway of excess calorie intake, you know, that seems like there probably is a causal pathway. If we've understood that piece, then we could give both the food industry, consumers, and regulators and policy makers the information that they would need to say, "How do we minimize consumption of the problematic foods?" One thing I've been thinking about more lately as I get older is brain health, staying sharp, maintaining focus, and protecting cognitive function long-term. My days involve constantly shifting between screens, emails, social media, phone calls, recording podcasts.

[00:13:16]All of this requires a lot of cognitive processing power and cognitive reserve.

[00:13:21]And there are stretches where I feel absolutely dialed in, and there are other times where the fog creeps in. And what I've come to appreciate is that there are biomarkers that you can measure that act as a window into your brain health. Things like inflammation, blood sugar regulation, nutrient levels, and hormonal health all influence how your brain performs today and over time.

[00:13:42]If we're not measuring those systems, we're making assumptions that what we're doing is working. That's one of the reasons I use Function. It gives you over 160 lab tests each year, so you can see what's happening beneath the surface. For me, tracking things like inflammatory markers and markers of metabolic health like triglycerides and HbA1c has directly informed how I train, how I eat, and how I recover. Rather than just doing things blindly, I have data to help guide the lifestyle choices that I'm making and where I may or may not need to pay more attention. Own your health for $365 a year, that's a dollar a day. Learn more and join using my link functionhealth.com/simonhill and use the gift code Simon25 for a $25 credit towards your membership.