Plain English with Derek Thompson - Why Are Americans So Unhealthy? Part I: Is Ultra-Processed Food Killing Us?

Episode Date: June 11, 2025

Americans die younger and faster than the residents of almost every other rich country. Why? There's gun violence, drug overdoses, and car crashes. Young people are much more likely to die from these accidents than those in other countries. Just as importantly, Americans are more likely to die from chronic illness, especially heart disease and metabolic diseases. We eat more and worse food. We're arguably exposed to more environmental toxins. We move around less, too. Kevin Klatt, a research scientist at UC Berkeley and a nutritionist, joins us in the first episode of our new miniseries on health. We take on the hottest topic in the diet world today: ultra-processed foods. If you have questions, observations, or ideas for future episodes, email us at [e-mail]. Host: Derek Thompson Guest: Kevin Klatt Producer: Devon Baroldi Learn more about your ad choices. Visit podcastchoices.com/adchoices

Folks, it's Jay Kyle Mann from The Ringer, and as always, basketball is so freaking, freaking good. It's so good, in fact, that the Ringer's NBA draft show is finally back just in time for a ramp up to June. We've got you covered every week as we take an in-depth look at who's got next for the NBA's future. We'll talk the rising and falling stocks of the best and the brightest prospects in the 2025 NBA draft class. From Cooper Flag to Dylan Harper, the VJ Edgecom, and more. tap in with me on the Ringer NBA draft show every Wednesday and make sure that you follow, subscribe, and hit us with those five-star ratings. Today, the first in a new little mini-series. Why are Americans so unhealthy?

Let's begin with the facts. Americans die younger and faster than the residents of just about every rich country. This statistic is indisputable. It's as straightforward as the stat that our population is larger than Portugal. A 30-year-old in America is four times more likely to die in the next 12 months than a similarly aged person in Australia, Austria, Switzerland, Germany, France, the UK, Japan. So why do Americans die young? When you look closely into this question, when you decompose this death gap, a few culprits

become immediately obvious. Gun violence, drug overdoses, car crashes. Young people in America are much more likely. to die from these kind of accidents and tragedies than other similarly rich countries. But these aren't the only factors. Just as importantly, Americans are more likely to die not only from car accidents and gun violence, but from chronic disease, especially heart disease and metabolic diseases like diabetes.

We eat more and worse food. We're arguably exposed to more environmental toxins. We move around less, too. According to one study, Americans take at least 30% percent of the United States. fewer steps a day than people do in Australia or Switzerland, certainly than Japan. On this point, I think, you can blame our car-oriented suburban lifestyle. American cars, you might say, don't just kill us instantly when they slam into us. They sicken us slowly and more subtly when we step into them, replacing walking with sitting,

contributing to this imbalance between the calories we consume, which are often ample, and the calories we expend, which is. is often meager. The result is that the U.S. has the highest prevalence of obesity among OECD countries, about two times higher than the OECD average, and a greater burden of conditions like diabetes. In the next few months, we're going to give you a few episodes that try to get to the bottom of this question. Obviously, the Maha Make America Healthy Again movement is putting health at the center of the news media's attention. And while I disagree with RFK Jr. on vaccines, science spending, I do think he's right about this.

chronic disease in America is serious, and we are sickening and even killing ourselves with our food and our behavior. In future episodes, we'll discuss other topics like the connection between eating, visceral fat, inflammation, GLP1 drugs, but today we dive right in with the hottest topic in diet, ultra-processed foods.

Kevin Klatte is our guest today. He's a research scientist at UC Berkeley and a nutritionist. someone who's not afraid to get nerdy about the basic and deeper questions I have about ultra-processed food. What is it? How is it defined? Why do we think it's so bad for us? How do we know what we think we know about its effect on our bodies? And is ultra-processed food bad because it's tricking us into overeating or because it's poisoning us with industrial chemicals? The category of ultra-processed food is enormous. It includes the

whole wheat bread and tofu that I buy from the grocery store, and the Mountain Dew and ho-ho snacks that I try not to ever buy from the convenience store. And so underneath this discussion of diet and health is a deeper question about science and science communication. If our diet language is so impoverished that we have one phrase for a category that includes my whole wheat bread and a leader of Mountain Dew, don't you think this is a field that would be improved with a bit more nuance? I'm Derek Thompson. This is plain English.

Kevin Klatte, welcome with the show. Thanks for having me. You are a diet and metabolism researcher. How did you get into this field and why did you find that subject so interesting? Oh, gosh. Very few people actually ask me this. I actually did lose quite a bit of weight in high school, like 85. pounds, quite substantial. And that sparked my interest in it. But then more as I went into the

sort of academics, there's actually like a single lecture I can remember back to where they taught us the concept of epigenetics. This is sort of the field of what regulates the genes that are expressed. And there's actually little chemical tags that tag directly the DNA and also can tag the proteins that they wrap around. They're called methyl groups, just a carbon and three hydrogen. But those methyl groups, they regulate how tightly coiled the DNA is and whether genes can get turned on or off. And some of those methyl groups can come from the diet. And so it was just really intriguing to me at the time that, you know, diet is a source of this little chemical molecule that is getting added or removed from the DNA and turning on and off genes. And so you don't really think about nutrition and chronic disease risk.

This was just more like a very molecular question. When you change diet, what is happening to our DNA and how the genes are expressed? And so I've gone on and tried to do work in this general field of both how diet influences gene expression and how genetics influence nutrient requirements and how we respond to different diets. And for better or worse, the question of how does diet change who we are, how is diet foundational to human health? this is clearly a question that is having a moment right now. With RFK Jr., the Make America Healthy Again movement, there is surging interest in America's diet

and in ultra-processed foods and seed oils and sugar in particular. And I wanted to have you on the show because I followed your work for a bit, and you are just such a careful and nuanced communicator about the science and nutrition. I wanted to have you walk me through the actual nitty-gritty science of ultra-processed foods, seed oil, sugar,

these sort of cardinal sins of the American diet, as least as they're represented in American media. So let's start with a big-picture thesis statement here. It is a statistical fact that Americans have higher rates of obesity and chronic illness in other countries. And what I would like to know, for starters, is why? Has a scientific community reached anything like a consensus on this answer? Yeah, it sounds like a simple question.

most things in nutrition, simple questions are just have many layers to them. But I think at a really high level, you know, the laws of physics apply to human biology. There has, everyone accepts that there is an energy imbalance occurring where you have what we call positive energy balance where our caloric intake is above our energy expenditure, and that is leading ultimately to rates of body weight gain.

So most basic consensus is that if people consistently eat more, calories, then they burn. The law of physics says, weight goes up. How long has this been happening? How long has the rise in obesity been happening in America? You know, people often pinpoint this to like the 1980s. You see like rates of obesity starting to really increase. But even going back further, we have spotier data before 1960s, but you can look at things like insurance records, as well as military entrance academies. So this is all before we had nationalized surveillance data, but there seems to be a shifting kind of average BMI over time. So we've been in positive energy balance probably at a population level for quite a while.

We just started a much lower BMI, and it's been shifting up and up and up. So you see it's like in the 80s is when we really start the population distribution is shifting towards obesity. But we have had many changes to our food supply with industrialization, just from the availability of the calories, to the things that we're eating, that ultimately have been driving positive energy balance, probably changes in physical activity or at least the amount of energy we expend in the type of work that we do. You can look across the past 150 years, basically,

and come up with everyone has a favorite pet hypothesis of, well, this changed in 1980s, so therefore that probably explains the rates of weight gain afterwards. You have small little experimental studies that are often done saying, like, okay, does changing some aspect of food, whether it be a macronutrient like fat or carbs that people hear about, or whether it's something about the flavor profile, the texture,

these things in controlled laboratory settings, influence energy intake. And so we have no shortage of hypotheses. I would imagine it's probably not one thing, but like many things kind of collected to be, and has their favorite thing. Like, this must explain it all. And in reality, there's probably many, many, many small things that have combined to drive that positive energy

balance that we're seeing. It's interesting because the story that we tell about obesity so clearly shapes the answers that we want to apply. Like if you're the sort of person who says, if you look at the studies, if you look at the stats, obesity started rising in the 1980s, 1990s, we have to focus on something that changed the American food supply in the 1980s. But if you take this longer view and you say, it's possible that American BMI has been rising for decades and decades, maybe even a century, it does rule in other explanations here, like the fact that rich countries tend to live in more calorically rich environments and people use their money to buy energy-dense foods. There's a logical follow-up here, which is, do Americans eat more calories because we're so rich, because we can

buy more calories, period? Or do you also think there's something about the American diet or maybe the Western diet and food supply that is unique to contributing to obesity? I think that's somewhat of a tough question to ask, but if we kind of pick it apart, like I think wealth and the sort of countries increasing their GDP or whatever sort of metrics you want to use for country-level wealth, it tends to come with a westernization of the diet and an increase in chronic disease rates. And so you see this in some low and middle income countries now that are experiencing a westernization. I have a clinical trial more focused on pregnancy nutrition in Tanzania. But over there, you're like dealing with the burden of malnutrition

and micronutrient deficiencies, but then also when we were there, people are like, oh yeah, there's a lot of gestational diabetes that we're catching now and obesity rates are rising. I think most people would argue it's a bit of a both that you need ample calories to support weight gain. But then is there something about the type of calories that we're eating that facilitates overeating? And that's like a hot area of research now. There's been a lot of, I think, sensory science work and particularly smaller, controlled, more laboratory things, defining that the energy density of a food and its texture and the person's eating rate can, like, really dictate how many calories are eaten within a meal.

I think this sort of bouncing around and everyone having a favorite pet hypothesis, this is a good sort of barometer that may be. be, there's not entirely consensus and there's a lot of guesswork involved. All right, well, let's try one of these hypotheses. I've had a lot of conversations in the last few weeks with scientists about these questions. I really want to do several episodes on the topic of American health and diet. And one topic, one hypothesis, I suppose you could say, that keeps coming up over and over, is ultra processed food. Ultra processed food is a killer. It's the original sin of the American diet. I see this a lot. I hear it all.

lot and I want to stress I'm hearing it not just from scientists that I don't trust, I'm hearing it from scientists that I do trust. But before I ask you, Kevin, the million-dollar question of is ultra-processed food really as bad as people say, I want to ask you, I think, a necessary first question, which is, what is it? What is ultra-processed food? And how do nutrition scientists like you, distinguish between, say, a brownie that I make at home that has a ton of sugar in it and what nutrition scientists would call an ultra-processed brownie? Yeah, so the definition of ultra-processed fruits have changed a little bit over time, but we're now at the point where there's something called the Nova categorizations.

And so there's four categories of processing. You can think of them as individual bins. And so bin number one is relative to... naturally, minimally processed food. Some people would recognize it more as like whole food, so chickpeas, lentils, dairy, et cetera. Category 2 is much more of culinary-type ingredients that we have had throughout history,

so you can think of like oils like honey, or sorry, oils like extra version of olive oil, honey or sugar cane or salt, so things you're using within food preparation. And then category 3 is when you start to mix category 1, in category two components. And what distinguishes, and that's what we just call it processed food in category three.

And then so people have eaten these for thousands of years, ostensibly. In category four is where you get the term ultra-processed, and you really have a focus in the definition on industrialized ingredients. So things that we can only make or add to foods because of modern industrial and like basically modern food chemistry technologies. And so these are everything from preservatives, flavors, colors. There's a lot of things

like individual components that typically add to texture or mouth feel, so you have emulsifiers, things like fractions of food. So the way component that is added and might be used in like a cheese powder for that flavor feel of it. Or, you know, there's triglycerides that are the fats in foods, but then there's monoacylaclysteroids, which are used also for sort of an emulsifier within foods. And so it's like a sub-fraction of the fatty acid, a sub-fraction of the dairy protein, things like multidiction, which are a fraction of starches,

that those are when they're isolated, then used for some functional component. So it's really a mixed bag, but there's many, many, you just need the addition of one of these on an ingredient label for it to become an ultra-processed food. Are all ultra-processed foods equally bad? Is this entire bin, as you said, deserving of equal rebuke across the board? Or in a way, is this a category of food that sometimes discussed as a blanket category

for a set of foods that are very, very different in terms of their effect on the body? Everyone in nutrition thinks like chickpeas are better than Pop-Tarts. That is almost like, let's acknowledge that, leave that to the side. We're getting deep into the weeds of definitions and whether they can be used for policy. And when you look at UPFs, I think there is a tendency for a lot of people who are very big fans of the term to say, like, okay, like, how can you defend eating Pop-Tarts when we should be eating spinach and chickpeas? And it's like, that's not the comparison that I really want to draw out, because within that UPF category, you have an unsweetened almond milk that has an emulsifier in it that is,

now a ultra-processed food. I don't think when people imagine ultra-processed foods and food processing, they're thinking of that. When you look at protein powders that have gone through intense industrial extraction to get from the pea to just the pea protein fraction and then to be powderized and flavored and whatnot, I don't think most people who are, you know, most of, it's kind of funny when you look at Instagram influencers who say ultra-processed foods really bad and then they have like a protein supplement line and you're like, or they have, and like, sometimes they're protein powders or meal powders where they've added all these extra ingredients that would make them a UPF. But there's sort of a cultural connection to the concept of ultra-processed foods

and what makes a processed food that gets intertwined with class and a whole bunch of things that get quite beyond nutrition, I think. But when you look in the UPF category, you have everything from a soda to an unsweetened almond milk. And then when you go into the research, most of the data on ultra-processed foods, apart from a couple trials now, are observations. data, and we can dive into all the nuances of that. But the observational data has already started to show the different classifications of ultra-processed food. So things like yogurts and cold breakfast cereals that are frequently whole grains and fortified

with micronutrients, those are associated with positive health outcomes despite being UPS, and that a lot of the UPS risk signature comes from things that would not surprise you. And so like sugar sweet beverages, grain-based desserts, some frozen dinners, like pizzas, what we often call like a mixed dish, so something you might get at a restaurant food with like noodles and meats and cooked in oils and tons of sodium. And so, like, yeah, these are broad categorizations that we're just looking at. Are they associated with disease outcomes or not? And so sort of the class of the umbrella that UPF, the UPFs fall under,

there's many things that I think you could design a reasonably healthy diet with. And there are many things that we would probably argue to minimize or eliminate and everything in between. And that is not only true of the UPF category, but also things in like category three. And so when you start to, you can look at the variability of products within category four, but then when you really focus in on what makes category four UPF, it's that industrialized ingredients. And so if you compare like a group three, you're like homemade white bread that's just refined flour and a yeast and salt and maybe a tiny bit of olive oil or whatever, that is quote unquote not UPF. And so by the cultural thinking around it should be healthier than the UPF whole grain bread that you get at a store that has a preserve in it that makes it a UPF.

So there is, when you really start to whittle down into like, can we actualize this UPF term to into policy, communicate to people, it gets, they need to be able to read food labels and look at all the ingredients and know what's industrial or not and to figure out what's a UPF and then be able to make these comparisons and they're not really black and wide. You know, the more I report on science, the more I find that like the single most underrated question in science is how do we know that, right? I was scandalized, I think, when I was reading all these headlines about the dangers of moderate alcohol consumption. And I'm someone who loves

the occasional glass of wine, sip of whiskey. And so obviously, I was very motivated to want to investigate this question. You know, is red wine great for my heart? Is whiskey killing me? Is beer giving me cancer? And you go into the observational studies on moderate, alcohol use. And they're just a mess. We did a whole podcast on this. The vast majority of these studies comparing people who don't drink alcohol to people who drink moderate alcohol have this bucket of people who don't drink alcohol that's often totally confounded. It's people who are sick so they don't drink alcohol. It's people who maybe used to be alcoholic so they no longer drink alcohol. It's people that are religious and don't drink alcohol. It's people who therefore

are both so Mormon that they don't drink alcohol and also so sick that they don't drink alcohol. It doesn't make any sense to think of this as a coherent group, and yet it's being compared to this also extremely diverse group of moderate drinkers. And so I came out of that study being like, or that research thinking, Jesus Christ, every single time I do an episode on science, I have to ask questions about methodology now, despite the fact that it might bore some of my listeners. So we're a little bit into this episode. Maybe I still have some of them here with me. How do we know what we know about ultra-processed foods being bad for us? Is there a canonical study in this space that proves to careful folks like you that,

yeah, this is probably something to look out for? Do we have big randomized trials in this space? What do we got going for us on the evidence front? Because that is the crux of science. It's the, I think people focus on the conclusions of science, and it's really the methods that get you to the conclusions is the scientific process, because most of the data is observational. So these are a large cohort studies where you ask people about what they eat through different types of dietary recall methods.

And in the best versions of these cohort studies, you ask them multiple times because people's diet changes over time. And particularly as people are increasingly getting sick or developing risk factors, their diets can change. So I guess you follow these cohorts up for long periods of time. They have diabetes incidence, increased diabetes rates. They have increased cardiovascular disease or cancer. You can look at a whole wealth of outcomes. tens, if not hundreds of thousands of people. And then you relate how what they reported eating

at different time points to those disease events. There's two major tools that are used in these cohort studies to assess diet, but the major one is a food frequency questionnaire, which if anyone hasn't looked at one of these, I really encourage you to go look at one. Their whole goal is to do what the title says, like food frequency. So they're really trying to rank how frequently you can consume different food groups. And these are typically already laid out for you on the paper. So like how many times a

week do you eat red meat? It could be daily three times a week, one to three times a week, three to five times a week, like once a month. And so there's different categories you check this off. And their whole goal is just to get a feel for like really to be able to classify high consumers versus low consumers and distinguish them. And that's mostly how they're validated. And so the food frequency questionnaires are just kind of food groupings. It's very vague, minimal information. So you don't get a label. So when you call things a UPF in that case, there's a bit of a leap of faith involved and not a great way to cross-reference it. So the tool that we're using is not really designed to distinguish the crux of what makes something a group four instead of group three, which is the

addition of the industrial ingredients because the FFQs do not have that level of detail. So you have a tool that's not designed to assess UPS where people are trying to assess UPS. To sum up where I am on this, I think the observational data suggests that Western diets are probably not great for people. They seem to certainly correlate with a lot of obesity, cardiovascular disease, metabolic disease. They seem not great. That said, we've got this category of ultra-processed foods that's getting a lot of attention. And you just told me, that a breakfast of mountain dew and twinkies is an ultra-processed breakfast, but also a breakfast of almond milk and whole-grained bread that has one preservative

that I bought from Trader Joe's is also an ultra-processed breakfast. And it seems like if you have no nuance here, if the policy remedy is to slap the label of ultra-processed food might kill you on all of the almond milk and half the Greek yogurt and half the tofu and half or all of the protein powder. And it's the same label as what you've got on the Pop-Tarts. People are going to become numb to this stuff. It doesn't make any sense intuitively. And people will just ignore it. So I think the next place to go here is what is it about the worst ultra-processed foods, Kevin, that makes them the most bad.

One of the issues with ultra-pressed foods is that they're associated with almost every negative outcome. And the mechanism for why something drives obesity is probably not the same mechanism for why it might drive different types of cancer and may or may not be the mechanism it drives cardiometabolic disease. These things can technically be all linked.

I think that's why a lot of the literature has focused on what about ultra-processed foods might drive obesity, which is itself a risk factor for cardiometabolic disease and cancer. but you see associations between UPFs and like everything. So there are many UPF advocates who say it's, well, sometimes it's the food additive component, and sometimes it's the obisogenic component of them, and sometimes it's the fat composition,

and sometimes it's the sodium, and so you get this total gish galop. When you have such a broad umbrella, you can, that things fall under, you can point to any number of theoretical things that might be causing it. But I think, again, the research is focused most on what about them might drive excessive calorie intake and obesity.

And so there's a whole field of sensory science that has been studying what about food might influence everything from our pleasure to how much the calories that we eat from it that has sort of begun to define. And again, these are largely in like controlled laboratory settings and not as much the real world. But factors like the energy densities, the amount of calories per actual weight of the food, is a major determinant of our energy intake. And just to pause you there, energy density is calories per gram, which I kind of think of as like calories per bite.

So if you compare a raw lettuce leaf to a single chito, they might weigh roughly the same, but the chito has many more calories. It has more energy density. Other than energy density, what else about ultra-processed foods do you think contributes to weight gain or disease? There is also things like our eating rate is a major determinant of calorie intake, and so processed foods tend to have a different texture. They don't spend as much time needing to be chewed, and so you can eat them quicker, and so that could be a factor that plays into it.

People have also started to study, like, sort of the, what we call, like, the hedonic response to food or palliability. Some people are actually measuring, like, how much dopamine gets released upon eating it with these very fancy sort of neuroscience techniques, And so there's various ways you can kind of capture, like, how much pleasure the food gives you. But some folks have purported that they are sort of like hyper dopaminergic and we're getting, quote, unquote, addicted to them. And that that's why we're overeating and they're overriding our satiety cues. Sorry, and just squeezing in here to do some vocab.

Satiety means feeling full. Pallitability is close to what I think most people would just call taste. So you're saying that ultra-processed food is associated with, energy density, calories per bite, palatability, taste, and a dopamine response that's overriding our satiety cues. So, like, we feel so good about the food that we eat even when we're full. But I guess this goes right back to my biggest surprise here, which is, how do we think about this enormous category of food that seems to include everything from unsweetened almond milk to Twinkies? The latter has a completely different everything than the former.

different calorie density, sweetness, texture, dopamine response. Again, this comes back to, like, I don't personally think of unsweetened almond milk because, like, overriding my satiety cues are really, like, driving excessive intake. I think taking a mechanistic viewpoint is really important for identifying the attributes of the specific UPSs that might be a problem. But when you apply it to the entire food category, it's like, well, they're not all hyper-palatable, they're not all high-energy density, they're not all soft and easy to consume in large quantities all at once. So you get to the point, I think, where the UPF term, you go, like, why are we even caring about this? We should be studying the attributes of the specific things that people are

calling UPFs that drive intake, for example. And I think that's where a lot of the research is going, although we are producing research on this topic at a truly snail's pace. So one thing that's surprising about that answer is that you didn't mention a lot of macronutrients. Like, you didn't mention fat, you didn't mention sugar. And certainly, there's been periods of American diet science or dietary recommendations where the enemy was clear. The enemy was fat. Or I think more recently in the last decade, certainly this is a story that I invested in quite a bit. The enemy is sugar.

Where are we, Kevin, on sugar? Like how? Let me ask the question this way. If I'm your patient and I tell you that I've been struggling with my weight for the last few years, but my plan for the foreseeable future is to consider sugar an enemy, and I'm basically going to do my best to no matter what, make sure no sugary beverages, don't have too much dessert, stay away from bread that is high in sugars, and only eat like, you know, those super grainy breads that are like one gram of sugar per serving.

How would you recommend or object to my impossible simple, of the entire field of nutrition science by just saying, hey, man, I'm just going to stay away from sugar. How am I doing? So to original contention, I didn't say sugar or fat, it's because they are technically sources of calories that get wrapped up in the energy density. And like the food form matters a whole bunch there. So like a pure oil, which we would think of the fat and almonds, which we would think of as fatty, differ in their energy density. So I always tell patients like you imagine a plain totally unsweetened no oil added muffin. It tastes great. But

that muffin is going to be like the same volume and take up the same space as your stomach, even when you add in a bunch of oil and a bunch of sugar. And so that's like the energy per volume there is much higher. It's not stretching your stomach and giving you as filling as much. So macronutrients are kind of factored in in the energy density. They're also factored in in palatibility. And so, you know, highlighting any single one of them starts to get at a lot of these attributes around taste and energy density, but isn't telling the complete story, I think. But inevitably, when somebody says, like, I'm just going to avoid all sugar, well, nobody's, like, I think with sugar, you know, people talk about being addictive. Nobody's going and, like,

in the middle of the night, wake up at 2 a.m. and, like, eating white sugar. Like, it's always kind packaged in a way that is mixed often with starches, with sugar, with sodium, with added oils in a palatable setting. And that sugar, the sweetness is like a reinforcing factor. So when you avoid all sugar, you're just like cutting out all these products that have all these attributes that people are studying and are likely factors for why we overeat. But so you can set up kind of nutritional guardrails for yourself by saying, I'm going to eat a super low-fat diet.

And if you actually follow that and you eat like beans and whole grains, like it doesn't taste that good if you actually are strict about it. It's kind of boring. It becomes bland. You've limited the variety of food. There are other things that drive our intake that are not intrinsic to the food, like just the variety, advertising, constant stimuli, things that are, there's a lot about obesity that we've been talking about

that's like assuming that intrinsic factors of the food are the reason people overeat, but there's so much in our environment from like, again, food advertising variety that might be stimuli as well. And so the guardrails that people have set up historically, and there's many anecdotes about I went low carb or I went low fat and I went no sugar, I went no gluten. And it was amazing for me. You'll find that they've drastically cut down the number of foods you can eat within your daily environment. They often affect palatibility. They often cut out these super high energy density foods. There's a common mechanisms by which they all could. converge. They actually have nothing to do with the intrinsic badness of fat or sugar themselves,

but the way those rules work in your environment will cut out just many, many, many sources of calories that have some of the attributes that we think drive intake. One last follow-up on ultra-processed food before we get to a bit of a lightning round on diet nutrition. I think I still just want to understand what the dominant thesis is for why ultra-processed foods or the Western diets generally is so bad for us. And it seems a little bit like a door number one, door number two situation, although maybe the answer here is that it's both doors. Door number one is these foods are a really, really efficient delivery mechanism for excess sugars and excess calories that put us into caloric surplus so that we're eating more calories

than we can expend, and then being in caloric surplus for a long period of time makes us obese. That's explanation number one. Another explanation is that ultra-processed foods are bad, in part because they introduce non-food chemicals that essentially poison the body and cause inflammation. So what interpretation or blend of interpretations makes most sense to you? It's about being an efficient mechanism for delivering calories to the body or ultra-processed foods are bad because in a way they're not quote-unquote real foods. They introduce chemical poisons to the body that cause inflammation.

I think the latter is more of a philosophy than it is science. Just saying like there's chemicals. There are many chemicals, like, say that in kind of a scary way. It can be vitamins. It can be preservatives. It can be emulsifiers. And we should, I think you have to look at them kind of one by one. And there's certainly arguments around how those food additives get approved

because it's more of a toxicological framework. of like find the dose that has no effect in mice and then divide it by 10 to 100 and then set that as the upper limit for humans and then kind of like the food industry can use it as long as it's below that dose. The latter option is like I think it captures a lot of media attention because it's like, ah, they're adding toxic chemicals to our food, the colorance and people, the administration's talking about this. It's like we have blockbuster data that says that this is bad.

And almost always it's massive doses in a rodent that humans could never consume. in a rodent species-specific mechanism that it, like, is this true of some of the red dyes, that is the reason that increases cancer. It's still at a risk that, like, if you were eating that much red dye, you'd be eating so many skittles, I'd be much more word for you about the nanoscutters you're eating

than, like, than the food additive component. So that's not to say that there's not specific classes that we're, like, studying a little bit more. There's been more studies of, like, emulsifiers, for example, and folks who are prone to inflammatory bowel disease, And so there are some additives where it's like, yeah, we should investigate that a little bit more. But the fact that we're looking into emulsifiers a little bit more in specific use cases is not an indictment of, I think, like, all food additives need to be studied necessarily. But there is a few that, like, we should take a look at, you know, argue, like assess for their risk in relationship to chronic diseases.

But we really lack good biomarkers for things that people care about that you can measure. So if you can't assess these things in observational studies, like say you think a dye or some additive is causing cancer, we're not assessing that die in your self-reported dietary intake data, so you can't really link it to cancer that well. And then when you come to a trial, you might be able to change the amounts of the die and study things, but you're not going to wait in a trial 20 years

to see somebody develops cancer. We don't have like a short-term biomarker of cancer risk for most cancers. So it does get really hard to study, and it's why we have this toxicological approach where you're finding the dose that does nothing to a mouse and then dividing that by 10 to 100 as a safety factor. And so if folks don't trust that toxicological approach, the other types of data we have are really inadequate

to answer the questions that people care about are like, are these additives toxic for these long-term disease risk? It really feels to me like this industry has lurched from one boogeyman to another. Fat is the boogeyman of the 1980, early 1990s, and then we go to a low-fat diet. And then sugars, the big boogeyman, the 2000s, and now we're onto ultra-processed foods,

which are, I think, the hottest boogeyman in town. It's not that the boogeyman aren't scary, I hear you saying, right? Eating too much fat is probably bad, eating too much sugar, bad, eating too many ultra-processed foods, almost certainly bad. But maybe one lesson that I'm taking from this conversation is that the boogie manification, if we can call it that, the boogie manification of these categories is distracting us from one big thing that we know for certain,

which is that people in Western food cultures are just in chronic energy surplus. And it might not even matter so much if the calories come from fat or sugar, ultra-process, pop-tarts, whatever. The most clear thing is we're just eating so much more food than we used to. And because it's very difficult,

maybe to sort of sexy up surplus caloric balance, surplus energy balance or something, we have to sexy up the boogeyman. And so we're sort of lurching from one generalization to another. Is my description there, my sort of brief history

of diet controversies, do you agree with it, or am I overgeneralizing myself? You know, we've said don't eat fat, we've said don't eat carbs, we've said try to eat Mediterranean diets. Like the basics are not sexy, as you say, and people are constantly looking for a new way to say junk food kind of seems,

like something that has like a bit of a Puritan or a moralized take on it. But there's just this intrinsic desire for you say, these are bad, these are good, eat this, don't eat that, regulate this, don't regulate that. And the food system has just gotten so massive, so globalized, so many products. Like, as we said, you've got processing techniques that are probably good and you've got processing techniques that are probably bad and, like, or detract from the, nutritional qualities of food and how you get into that nuance and regulate around it and comment

on that I think I will always have something to like kind of shit post about on Twitter because of this constant lead. And I don't work like directly in public health. So I have a lot of like empathy for people in public health trying to communicate about these things because saying like, you know, eat your colorful fruits and vegetables and and you know, make food fun for people and get people back to cooking in the home and all the sorts of things. There's been eight million things tried that it doesn't change the reality that people interact with the food environment. It is overwhelmingly the inverse of what dietary guidelines recommend.

How you change, I really think needs to be like serious policy that we just have to like experiment with what might move the needle or not. But I don't know that there's actual political will because that requires picking economic winners and losers. It requires getting the agricultural community on board. It requires like a lot. lot of interventions that most, I think, economists would tell you not to do, might end up with a lot of food waste if you start producing things people don't really want to eat. So there are some learned

lessons from programs like, I think the school lunch program where there's been incremental changes over time to get more whole grains, get sodium down, get saturated fat down that we can maybe learn from in different sectors. And maybe this really does need to be like so local, localize that we do sector-by-sector policy changes. But there are many hypotheses and not a lot of political will to really do anything besides change the messaging within media. Now the media ecosystem has gotten so silo. Like, you've got the low-fat plant-based vegan people on Instagram. You've got the low carbors on Twitter. You've got the carnivores on Reddit. You've got seed oil people mixed into all this. And so it's gotten harder and harder to communicate about these

things and trying to address misinformation, target the right information to those who need it. Can we close, Kevin, with a lightning or lightning-ish round of some other food fads. I'd love to get your brain on. We can do, like, you know, food fads in 60 seconds. Like, each answer is like about a minute long. So not entirely lightning, but lightning-ish round. An impossible task for me, as you've learned from listening to this podcast, and I will try. Seed oils. Are they as bad as the RFK folks say? There is a hypothesis that has been around for a while that we eat too many omega-6 fats found in seed oils and two little omega-3s that has elements of nuance in it, but people say that the omega-6s are uniquely pro-inflammatory. There's not really any data to support that.

There's a whole bunch of biochemical theory I cannot cover in 60 seconds, but I'll probably have a post on this at some point breaking it down. But I think there's general consensus in the nutrition community that people don't eat enough omega-3s. We're not eating enough faddy fish. We're not enough flax and chia. There's different types of megathes in those. But the messaging, I think, needs to be like there are healthy omega-3 rich foods, but there's also independent health effects of omega-6s that they lower blood clestral on things. And so what was sort of like, we're eating too much omega-6 and not enough omega-3, there's now pretty strong consensus we're eating enough omega-3. There's still some uncertainty on the omega-6 side.

but I would put getting seed oils out of your diet pretty far down my list, unless you're eating seed oils exclusively in the form of Doritos, in which case I would argue. The seed oils are intended to be used to cook vegetables and dress a salad and things and not necessarily added to a muffin to increase the energy density. So when people cut out seed oils and say, it's amazing, I'm not too surprised because most people are getting seed oils from very easily overconsumed processed snack foods.

But do I ever tell my patients, not to avoid canola oil and cooking their vegetables? No. The keto diet, what's the cost-benefit ratio there? Well, it's an amazing benefit for its intended medical use in epilepsy and other rare disorders. The definition of the keto diet is its own podcasts, I think, and the history of how it's evolved. I think in general, keto just means low-carb to people. Like, the original medical ketogenic diet is low in protein, too.

But almost everyone describes eating a ketogenic diet as this high-protein. low-carb thing. It is not magic by, there's no magic metabolic mechanism that anybody's found. It might increase energy expenditure just slightly, but high-protein diets do that as well. It's probably because you're restricting almost everything you can eat your food environment and whittling it down to just a few food items.

I'm not surprised people lose weight on it. But in general, it doesn't long-term, particularly once you get to the point if you stop losing weight, you just maintain your weight, it doesn't look so great for your blood cholesterol levels, but there's a lot of variability in that for sure, and it depends on the type of fat you eat and whatnot. But there's no dietary guideline that recommends it for the population abroad

because you to design a ketogenic diet that hits kind of your estimates of all the micronutrients, adequate fiber and things, and all the things that we think are related to the lower risk of chronic disease. Like a designing ketogenic diet that looks like that is pretty tough, and it's not the heavy meat cheese butter version that most people are eating. So any diet by any name can be actualized in 20 different ways. And so if you're doing keto that's like high avocado, dark leafy greens, low in starches, like that version of a keto and nuts and seeds, like that version of a keto diet is much less

than the like red meat, butter, and carnivore diet version. The Mediterranean diet, go to Sargentia, eat what they eat, fish, leafy greens, legumes, lots of olive oil, glass of wine in night. Maybe you can drop the glass of one in night. not maybe unnecessary. You just talked about how bad alcohol epidemiology is. Maybe not a necessary component of the Mediterranean diet. But how solid is the evidence

that the basics of the Mediterranean diet are really useful basics for someone who just wants a one-stop shop too? Here's a diet that will do me good. If you go buy a Mediterranean diet cookbook and follow that, you're probably good and moving in a healthy direction. I don't think the Mediterranean diet is magic necessarily.

It's also representing like 12 plus countries that all have really diverse cuisines. And so I think it's sort of gotten dulled down into this like, okay, eat lots of fruits and vegetables and whole grains and maybe poultry and fish, a little bit of red meat and eggs and then use lots of olive oil, which is like great. Like sofritos and things that are like, they're using like leaders of olive oil all the time. And so I don't think an American can really adapt that. there's a blockbuster trial called the predomid trial that randomized people already in the Mediterranean to essentially eating a slightly more Mediterranean diet plus getting a extra virgin olive oil or nuts and seeds

or eating a controlled diet, which is just the background Mediterranean diet they're already eating that lowered risk of largely of stroke. And so there is, you know, there's strong evidence. Like we don't get many big blockbuster randomized controlled trials. But eating that kind of diet, If we just gave most Americans like a leader every couple weeks of olive oil, I'm not sure that they would go through it quite as quickly. So adapting that sort of, the Mediterranean diet is one of many traditional, regional diets. I think you can point to that are high in plant foods, fiber, healthy fats, fish, and there's all sorts of variation across these different regions,

but that are all typically associated with health benefits because they're skewing a lot of these, modern hyper-processed foods that are not so great that we talked about. And then focusing on nutrient-dense things. So I don't want to, the Mediterranean diet, I think gets critiqued a lot for being like, okay, Westerners decided to study this Westernized diet. And is it really that uniquely better than the traditional Tanzanian diet or the traditional Japanese diet? No, but we haven't invested in the research infrastructure to study those sorts of things.

So I don't want to, like, if you tell people of different cultures to eat a Mediterranean diet, they might look at you like. Like, how do I even do that? So the principles of it need to be extracted out and adapted to whatever a person's culture and history is. And it's not an easy one-to-one transfer of the Mediterranean diet of what they actually eat, like the foods they eat to other folks. Kevin Klat, thank you very much. I really learned a lot from this. And I appreciate the detail that you went into to explain what we know and don't know about diet, which is a lot.

I appreciate it. Thank you, man. Yes. We need to fund a whole bunch more nutrition research to learn more about it. So hopefully that revolution is coming right after the food colors.