with Spencer Greenberg
the podcast about ideas that matter

Episode 212: What we know and don't know about nutrition (with Gil Carvalho)

Enjoying the episode? Want to listen later? Subscribe on any of these apps or stores to be notified when we release new episodes:

May 30, 2024

How do we know what's true in nutrition? Why aren't nutrition studies seemingly as "definitive" as (e.g.) physics experiments often seem to be? What is the "hierarchy of evidence"? Why is there such a disconnect between the kinds of evidence that actually seem to persuade people and the kinds of evidence that scientists view as valid and meaningful? How can we talk about specific foods in ways that avoid labelling them as always good or always bad? Is the Mediterranean diet good for anyone and everyone? Is it better than all other known dietary patterns? Are there healthy ways to do (e.g.) low-carb, high-carb, low-fat, high-fat, and other similar diet types? What do we know about the effects of ketogenic diets? What do we know about the effects of meat-only diets? Are saturated fats always bad? How should we think about mechanistic evidence given for or against a particular food or diet? How much protein should we consume every day? Should we universally reduce our sugar intake? To what extent is excess body fat bad? Since BMI is much criticized, what are the best measures of health for people with excess body fat? Should we avoid blood sugar spikes throughout the day? What percent of people tend to re-gain lost weight after concluding temporary diets? Is caloric intake really the only factor for weight gain or loss? Is the mistrust of nutritionists justified?

Gil Carvalho, MD PhD is a physician, research scientist, science communicator, speaker, and writer. Dr. Carvalho trained as a medical doctor in the University of Lisbon, in his native Portugal, and later obtained a PhD in Biology from the California Institute of Technology. He has published peer-reviewed medical research spanning the fields of genetics, molecular biology, nutrition, behavior, aging, and neuroscience. In parallel with his research career, Dr. Carvalho also has a passion for science communication. He directs and hosts Nutrition Made Simple, which aims to convey fundamental nutrition concepts to a general audience via educational videos. Learn more about him here, and follow him on YouTube, Twitter / X, and Instagram.

SPENCER: Gil, welcome.

GIL: Thanks for having me.

SPENCER: The way you ended up on this podcast is that a colleague of mine said, "You've got to check this guy out on YouTube. He's got the best nutrition stuff." And I was like, "Uh-oh, what is this going to be?" Because you just see so much, so much garbage nutrition stuff. But when I started watching your videos, I was like, "Oh my God, this guy, he'll do a video and be like, 'Let's look at the seven papers ever published on this topic and walk through them and figure out what it actually says.'" So, the level of nuance you bring to this conversation is really mind-blowing. And so, I'm really excited about this conversation so my audience can learn about what's actually known in nutrition and what's just a bunch of BS.

GIL: Well, thanks. I really appreciate the kind words. Yeah, we try to give a little bit of scientific insight without putting people to sleep — that's the mission and the challenge at the same time.

SPENCER: It's actually very heartening that you built up an audience because you might think someone who goes into the depth of all the papers and so on would find it tough to get anyone to listen. So, it's really cool that you're actually able to build an audience with that level of depth. But let's just start with how do we know what's true in nutrition? Because I think, if you go look at a lot of influencers in the space, they're kind of making all these claims. How do you actually get to the bottom of it?

GIL: There are multiple levels. The first level is any evidence provided. With a lot of internet claims, there's no evidence; it's just an opinion or personal preference. And so, sometimes it's as simple as running a scientific search on a scientific database, and you realize the science is overwhelmingly in the opposite direction of a claim. Those are the simplest cases. Sometimes it's a bit more intricate. So, the first kind of realization is that these things are rarely black and white; it's usually kind of levels of gray. The strength of the confidence we have in a certain direction goes up gradually as the evidence mounts in a given direction. That's kind of how science works in medicine and biomedicine in general. It's almost never the case that you run one perfect study and you're done, and you've proved something. It's not like that. It's more like this continuous investigation, almost like you're investigating a crime: you find a little bit of evidence, then there's a witness, and then there's motive, and then there's this and there's that, and your confidence gradually goes up. Science is a bit like that. So, the more evidence and the stronger the evidence in a given direction, the higher your confidence, bottom line.

SPENCER: It seems that, in physics, you should have to have definitive experiments. They build this apparatus, everyone who understands it and who's an expert on how it was done does the experiment. Everyone's like, "Wow, okay, we changed our minds." But this doesn't seem to really happen that much in nutrition. I assume part of it is that even if you had the perfect experiment, it is like, "Well, you've only done it on one population. You only followed that population for a certain amount of time. That population might have a lot of other things going on. Maybe it wouldn't generalize to another situation." And I'm curious to hear more about why you think we need many studies. Why can't there just be one perfect study?

GIL: You touched on several points there. Reproducibility is really key. It's really the currency of biomedicine because of the multiple variables, as you intelligently pointed out. There are multiple populations, there are multiple conditions at baseline. Sometimes we talk about, "Okay, is this food healthy or not?" Well, at what amount? Instead of what? What else is the person eating? These things all matter, and a lot of times they do affect the results. So, you do want to see a variety of studies. Also, each type of study gives you slightly different answers. If you run a randomized controlled trial, it's one experimental design; it addresses certain questions. If you look at cohort studies, it's a slightly different angle. You can do experiments in model organisms like lab animals or test tubes. You can answer different questions there. So, it's really by putting all the pieces of the puzzle together that you get to see the big picture.

SPENCER: A lot of influencers in nutrition don't even bother to cite evidence. But when they do, it often seems very selective. I know you like to think about a hierarchy of evidence. Can you just briefly walk us through that hierarchy?

GIL: It's basically the idea that evidence is not all created equal. So, somebody says, "Here's a study." That's a good start. But what kind of study is it? Is it in humans? Is it a case report? Is it just a description of a patient? Was it a randomized trial? How large is it? Is it what we call a meta-analysis, like when you put together a bunch of studies? The hierarchy of evidence is essentially a very simple idea to organize that. Briefly, you have, let's say, anecdotes at the very bottom. So, "My cousin says that he started eating XYZ and he feels this." It's not useless, it's not zero, but it's a very low level of confidence. Then you'd go to something like experiments in lab animals or test tube or petri dish data. Then you'd get to human data, but human data with a lot of variables. Things like what we call ecological data. So, for example, "There's a tribe somewhere in the Amazon that eats a lot of this, and they have this disease very often." It's interesting, but there's a lot of moving parts. So, that's still relatively low. Then you get to cohort studies, which are closer to the top. Higher still would be randomized control trials. Obviously, there's nuance to all these steps, but I'm just giving you kind of the short answer. And then above randomized trials would be meta-analyses, basically pulling together a bunch of trials or a bunch of cohort studies, things that look at the data systematically.

SPENCER: It seems to me that one of the big problems that occurs is that how convincing people find evidence is very different from how good the evidence is. The anecdote from your cousin might be more influential, or pointing to, "Oh, let's go look at the longest-lived people in the world, and let's make a little documentary about their lives." That can be really persuasive, but you actually don't know which of the 50 things those people are doing is the reason they live so long, or is it just genetics that you couldn't possibly mimic?

GIL: Yep, that's exactly right. I think it's almost a mirror image of the types of evidence that are most gripping for the general public versus the types of evidence that are the most conclusive scientifically. The reason is that I think our minds are wired to listen to stories and to individual stories and to see somebody's face, and we perceive that as real. Whereas, things like numbers and bar graphs and p-values are not something that we're wired to really pay attention to. It's something that feels abstract and not real. The reason that these higher levels of the evidence hierarchy are more compelling is precisely that they make an effort to eliminate other variables. So, in general, with an anecdote, if I tell you, "Hey, a year ago I started eating more peanuts, and I think I lost some weight and I feel better." There could be another 50 changes in my life. So, you don't know if the peanuts caused that, if the peanuts had no effect, if the peanuts actually worked against it, and it was in spite of the peanuts. There's no way to know that from an anecdote. But with these higher levels of evidence — cohort studies and randomized trials — there's a metric, a systematic effort to eliminate other variables or to minimize them at least, so that you isolate one moving part as much as possible, and you're confident that thing is probably causing this outcome, this result that we observe is maximized. That's essentially why they have that ranking.

SPENCER: I want to talk to you about diet and health. But before we go into that, I want to have a meta-conversation about what sort of way our foods are healthy or unhealthy. I think a lot of people think, "Well, there's healthy foods and there's unhealthy foods. Eating healthy foods is good, eating unhealthy foods is bad." But many things in biology don't work that way. It's not like the more you eat, the better you are. With many things, as you get more of the kind of good things, you start getting diminishing marginal returns. It kind of starts flatlining. So, they help you for a while, then it flatlines. And then if you have really crazy amounts, it starts becoming harmful. On the other hand, I think bad things work quite differently, where small amounts actually have no effect on you whatsoever, and then as you accumulate more and more, it starts becoming harmful, and then it becomes really, really, really harmful quickly after you kind of get past a sort of critical point. So, I'm curious to hear what you think about the idea of good foods and bad foods. Is that a good idea, or is that kind of problematic? Are we kind of missing nuance there?

GIL: There's some truth to it. I think a more empowering way to look at it is to think of it in terms of dietary patterns. This is how nutrition science essentially has moved, and dietary guidelines and public health guidance have moved more and more in that direction and are still moving more and more in that direction of talking less about individual foods and especially less about individual nutrients, which sometimes confuse the public, and talking more about dietary patterns. What does the general pattern look like? And less focus on one specific food and this idea of superfoods and toxic foods. Yes, some foods are undeniably more health-promoting than others, in general. As you touched on, the level of exposure matters absolutely. But a more empowering and simpler way to think about it is to think about what your general dietary pattern looks like. Bottom line: currently, the most validated dietary pattern is what's called a traditional Mediterranean diet. So, if you think about it, the postcard picture would be Southern Europe, by the beach, by the ocean, fish, seafood, fruits and vegetables, some whole grains, some dairy, there's not a whole lot of ultra-processed junk foods or fast foods. This would be kind of the olive oil and mostly unsaturated fats. That's kind of the picture to bear in mind.

SPENCER: And what does the evidence look like for that diet? Do we know enough to say that whoever you are, that's a good diet to eat, or is that sort of still a little bit too speculative?

GIL: There's always a nuance. There's always possibilities of individual variability. The evidence that we have at all levels, both from what we call best mechanistic levels — essentially looking at individual nutrients and kind of more lab experiments — all the way to cohort studies. So, looking at people who naturally eat diets that are closer to this pattern, they tend to be healthier than people who eat diets that are farther from this pattern. And then randomized trials, probably the most compelling level. There are at least three truly large randomized trials looking at Mediterranean-style patterns with thousands of people over years and looking at different outcomes like the number of heart attacks and even things like cancer. In general, the traditional Mediterranean diet is very powerful at improving these outcomes. But yeah, it's absolutely possible that there are individual variations, that some people may do less well. And also, we're talking about a kind of general pattern. But the details — the exact amounts and the exact foods in there — are where it all becomes less certain. So, I usually tell people that it's not a shackle, it's not something that binds you. It's just a rule of thumb, a general reference point. But within that, there's plenty of room for personalization. So, people who prefer to eat a bit lower carb or a bit lower fat or whatever their personal preference is, they can tweak that and personalize within that dietary pattern. But the basic idea is that they are foods that are more health-promoting, at least based on current scientific evidence.

SPENCER: I would imagine that the sort of typical American diet is so bad that there are just a vast number of diets that would improve upon it. Would you agree with that?

GIL: Yeah, that's a common trend in research. Basically, any one of these diets that you compare to the standard American diet, everything tends to look better. In fact, that's one of the first questions you ask: If you just have a trial that is structured to have diet X compared to no dietary intervention, you will often ask, "Okay, this is better than the standard of a Western population. But what about other healthy diets? How does that fare?" So, yeah, that's absolutely a concern.

SPENCER: Has the Mediterranean diet been pitted against other [quote] "healthy diets" and does it fare well, if it is?

GIL: Yeah, there's a lot of trials and a lot of cohort studies. In general, it does very well. No other diet consistently beats it. But again, when we're talking about these diets, I don't want to give the idea that it's an inflexible thing. When you talk about a Mediterranean diet, it's not necessarily in contradiction with low-fat or low-carb. You can have these diets designed in a way that the key health-promoting foods are present in all of them. So, it really depends on how you're putting them together and how you're designing them. It's important to understand that. But yeah, there's a number of trials looking at it. Even with things like mortality, the Mediterranean diet tends to look pretty good. Now, one caveat is that the Mediterranean diet has been tested more; there's just more tests on it, more literature, more attention. The spotlight tends to be on it historically, in part because it's something that already existed, and we had all these populations naturally eating it. So, we also have a lot of data from there. But I think it's entirely possible that in the future, another diet might be even better. I think that's a possibility. That's why I said, currently, with the evidence we have, it points to this general pattern. It does not mean it's the best diet for humans; we don't want to overclaim. And it also doesn't mean that you can't do it in one of these variations.

SPENCER: You mentioned macronutrients: fat, protein, and carbs. There's so much obsession with this right now. For a while, people were really into low-fat diets, and now people are into high-fat diets, and there's a lot of obsession over high-protein diets. What do we actually know about the health effects of getting different ratios of macronutrients? And are these even the right way to think about things? Are these overly broad categories?

GIL: I think they're overhyped on social media. I don't think they're a pillar of nutrition. I think there's room for these things with personal preference — different people prefer different macronutrient balances. And for certain diseases, there might be a role to play as well. But if you look on the internet and on social media, this is all you'll see. And one other thing that's interesting is, I don't know if you've noticed this, but most of these diet fads or diet trends are rooted on things that you're not supposed to eat. So, there's the low-fat diet, and there's the low-carb community, and there's a gluten-free diet, and you should eat low salt, and you should avoid lectins, and you should cut out the seed oils. The focus is on prohibition. The focus is on the thing that you're not supposed to go eat, instead of talking about what foods you should be favoring and what foods you should be populating your diet with. It's kind of a weird concept. Can you imagine selling, say, a car by talking about all the cars you should not be buying, instead of talking about the thing that you are representing? I think it's because of a couple of reasons. One is simplicity. If you identify one thing that is the devil, it's very easy to remember; it sticks. Also, because it's a negative emotion. It's based largely on fear. And fear is the most powerful human emotion. So, it's very gripping if you can instill fear in someone; you have their undivided attention. In fact, you see this with the verbiage that's used on social media. A lot of times, it's negative emotional words like, "It's poison," "It's toxic," "It's going to give you this," "It's going to give you that." And when you look at scientific evidence and dietary guidelines, people who have scientific training in this area rarely talk like that. It's not fear-based. It's much more nuanced and much more positive and less of a sense of urgency and fear. So, yeah, we have this trade-off between fear being very powerful and gripping; you definitely have somebody's attention. But then, it kind of blocks our reasoning ability. So, this creates a problem in society because these fads take over the imagination, but then it doesn't really teach people how to eat a healthy diet. So, people jump from fad to fad. And over time, what we see is the population isn't getting any healthier. If anything, they're getting unhealthier because they go to one fad that didn't work, jump to the next, and they all share this problem with the fear and the finding of a devil. So, for example, we see people who are low-fat, they fear fat, they're terrified of fat. But then a lot of times what we see is they end up eating more of the wrong type of carbohydrate to compensate for removing the fats. We see the same thing with low-carb. Low-carbs are petrified of carbs. They go and eat more of the wrong types of fat and other foods. So, maybe a more empowering mindset is to focus on the positives, focus on the foods that have the highest quality. So, what are the healthy sources of fat that you can have in your diet? It's not trans fats, it's not Twinkies or McDonald's. It's going to be things like salmon, seafood, olive oil, walnuts, Greek yogurt. There's no need to fear those fats. What are the healthiest sources of carbohydrates? It's not going to be candy or soda or artificial syrups with a lot of refined sugar added. It's going to be things like eating a mango, eating a pear, eating a bowl of lentils, some quinoa, some broccoli sautéed in olive oil, and mixing the best carbs with the best fats. There's no need to fear those carbs or those fats. So, basically, a positive focus and focusing on foods that are health-promoting and trying to populate your diet with those.

SPENCER: This reminds me, one time I was at a restaurant with some friends, and an acquaintance was there. I'm about to take a bite of my veggie burger, and the person says, "You're eating poison." I was like, "What?" It turns out they were talking about the bun. They thought the bun was poison. It's just so dramatic. It really does get your attention. You're like, "I'm eating poison? Oh my God." So, you could see why that would be very effective as a marketing tactic.

GIL: Even the things that are admittedly not the most health-promoting foods. Everybody agrees that refined carbohydrates are not great. I'm not afraid of eating a cookie once in a while. It's not a staple of my diet. It's not a daily thing, it's not even a weekly thing. But if I feel like having a cookie once in a while, I do it. I'm not worried about it because I'm focused on the healthiest foods that are populating my diet. And these things around the edges that I do once in a blue moon, they're not what's going to sway your health.

SPENCER: So, is it fair to say that there's a healthy way to do all the different combinations of macronutrients? Obviously, you need at least a little bit of every macronutrient, but is there a healthy way to do low-carb, a healthy way to do high-carb, a healthy way to do high-fat, and a healthy way to do low-fat?

GIL: I think that's what most of the evidence is telling us. Yes, you can probably put together a low-fat diet or a low-carb diet if that's your personal preference, and you can design it in a way that probably can support all your aspects of health by doing exactly what we talked about, by favoring the high-quality fats and the high-quality carbs. But that's often not done, mainly because the knowledge and the attention are not there, because people are focused on the thing they're fearing, which is not really the crux of nutrition health.

SPENCER: A diet I see that's pretty popular is the super low-carb, like ketogenic-type diets. And I've always wondered about that because, as I understand it, you're literally putting your body into an unusual state that's kind of a starvation mode. What is known about the health effects of that?

GIL: There's lots of research depending on the outcome. All the way from weight maintenance and weight loss, cardiovascular outcomes, effects on people with diabetes or Type 2 diabetes, insulin resistance, prediabetes. There's lots of research. Things that are conserved that you see are weight loss. It's pretty common to see, not in every individual, but consistently you'll see it. Basically, you'll see weight loss with any diet that eliminates a lot of foods that people normally eat. The more radical the elimination, the more consistently you see weight loss. The problem is you have a trade-off between the power of weight loss, but also how difficult it is for most people to stick to these diets long-term. So, we see that consistently in trials, whether it's a very low-fat diet or a very low-carb diet. Some people do very well and love it, but those seem to be a minority. And those people become the evangelists that love the diet and talk about it in high praise, and that's great; it works for them. The problem is when we look at unbiased studies that just randomize people, the majority fall off the wagon within months to a year. They can't stick with these diets. And then they put the weight back on, or they lose the gains. Sometimes if they start out with diabetes, they improve after six months, and you look at them after a year or two or three, and they're back where they started because they focused on the one thing to eliminate, but they didn't find a pattern that was sustainable for them. So, it's something that we have to bear in mind when making these recommendations for the general population. It can work very well for some people. It can have transformative effects for some people that do well on it and can stick with it. It's very good for weight loss, good for control of Type 2 diabetes and glucose levels. Again, if done with attention to the quality of these foods, the low-carb diet, the keto diets get a bad rep for raising cholesterol and raising these things — people being at cardiovascular risk. I think most of the evidence is telling us very clearly that it's because of the choice of the types of fats. If you have the same attention as every other diet — this goes for every diet — mainly unsaturated fats, fatty fish, seafood, olive oil, walnuts, you can actually improve your lipids on a low-carb diet. So, you can have your cake and eat it too.

SPENCER: So, is it fair to say that if you try a ketogenic diet and you find it pleasant and easy to stick with, and you're thoughtful about what those foods are that you're actually eating, trying to eat healthier fats and so on, that it could be a good lifestyle choice? Whereas if, in order to do the diet, you feel like you're constantly depriving yourself and you're making unhealthy choices within the ketogenic category, then maybe it's not a good lifestyle choice?

GIL: Yeah, in that latter case, you're probably not going to stick with it. Most people are not going to stick with something for years that is a chore, that feels like suffering. So, that's a pretty good indication. Sometimes there's a transition period, and people struggle in the beginning, and then they get used to it. So, that's something to consider as well. I know lots of people who tried to eat lower carb and they just couldn't stick with it, and they're disciplined people too. So, it really depends on the individual. But yeah, I think this is possibly the biggest factor for public health and for individual choice: finding something that marries the scientific evidence with your personal inclinations, something that not just that you don't find suffering, but that you enjoy. Living for years and years with something so personal and so important as food without having any pleasure is no way to live an entire life. So, you want to find something that you enjoy. And yeah, it seems to vary a lot from person-to-person. I think there's also something to be said about the uncertainty of most of these diets long-term. So, if we talk about a diet that's very low-fat or very low-carb, there's quite a bit of evidence short- and medium-term, up to maybe a couple of years. But we don't really know what happens to people if they're on a ketogenic diet or an extremely low-fat diet or many of these popular diets for 10 or 20 years. We don't know; we just don't have the data. That's just because there aren't enough populations doing that consistently that have been studied systematically. It's probably going to change in the future, but it should be said that there is a level of uncertainty there. But the evidence we have short- to medium-term is that they can probably be pretty healthy if designed with these factors in mind.

SPENCER: Is that an advantage the Mediterranean diet has at least currently, that it's just been studied for a lot longer?

GIL: I think that's fair to say. It's a feather on its cap that we have longer-term evidence just because there are populations that eat it naturally. So, you can look at those populations over generations, over lifetimes, and you know what their health status is. I think it's undeniable that that is a factor. You're going to basically factor all this in, all the information you have, also what works for you, what you can optimize your health with, and then make an informed choice.


SPENCER: Let's talk about meat because, on the one hand, meat is often demonized. On the other hand, now you have all these pure carnivores. I think Jordan Peterson's daughter, Mikhaila, advocates this pure meat diet. What is known about the health effects of meat, and to what extent is meat really even a category we can talk about, or do you really have to go to the level of specific meats and how they're processed and so on?

GIL: You do need to go to the specific categories because we do see a lot of difference between, for example, processed meats and unprocessed meats. We see a difference between red meat and other types of meats, for example. And we see a difference depending on the level of intake. So, all of these things matter. And we also see a difference depending on the type, even within unprocessed red meat, whether it's fatty or lean. We have data that it matters also, for example, for blood lipids. So, that's all very relevant. The other question that people often ask is, "Okay, what if it's grass-fed versus CAFO-fed, like factory farms?" We don't have long-term data. And by long-term, I just mean outcome data in terms of risk of heart disease or risk of any of these hard outcomes for these more specific products like grass-fed beef. So, we don't know if they make a difference or not. That's the only honest answer. In general, for meat, as always on the internet, you'll see this polarization. People will exaggerate. One camp will say meat is poison, you touch it, you're dead. The other camp will say there's no risk whatsoever, and every scientist is wrong. Both are exaggerations. What we have is evidence suggesting that above a certain level, the risk of cardiovascular disease and certain types of cancer tends to rise. But there's levels of uncertainty there as well. There's certainly individual variability. It probably does matter what else you're eating, and it almost certainly matters what type of meat you're eating in terms of, for example, fatty versus lean. So, yeah, there's levels of uncertainty there. The guidelines are essentially giving you some information so you can make an educated choice. But the idea that having any red meat in your diet is unhealthy is not really compelling. But based on the evidence we have, I think there's a very real possibility that, for example, having a diet that's very high in fatty meats, to the point where it's elevating your risk factors (your ApoB, for example), in all likelihood, that will raise your cardiovascular risk. That's one reason why one way, by the way, to address this individual variability is to ask, "How do I know if I am susceptible to this thing or not?" Your blood work is one way. Your family history is another way. If you're eating a diet with a bit more meat than the next guy but your blood work looks amazing, I would worry less. And if your family history is also unremarkable for those things, it's also a sign that you may have a stronger resistance if you tolerate it better.

SPENCER: Would you say the evidence says right now that the healthiest way to eat meat is to eat lean cuts that are relatively low in saturated fat and that are less processed?

GIL: Yeah, absolutely. Less processed forms, leaner cuts, whiter meats, fish, which essentially is a type of meat. And fish has better health outcomes, probably because of the types of fat that it contains. That's what I would favor: the fish and the seafood, maybe some white meats, and then red meat for people who want to include it more sparingly, maybe once or twice a week or something like that.

SPENCER: There's this narrative that the world kind of went anti-fat but it kind of went too far. And then actually, a bunch of the evidence against saturated fat was kind of inflated, and it turns out saturated fat isn't that bad for you, or maybe it's even fine for you. What is the actual evidence on saturated fat?

GIL: A lot of the debates come from the fact that there is context dependence. So, again, it's because people will oversimplify and say saturated fat is bad for you. And it's very easy to find studies or contexts that contradict that. So, one thing that we were talking about, just to give you one example — not to put people to sleep — but one example is "instead of what?" This has been looked at a lot. If you compare saturated fat to, for example, refined carbohydrates, you don't see a difference in cardiovascular risk. If you compare it to trans fats, saturated fat even trends to looking better. If you compare it to unsaturated fats or whole carbohydrate sources, you tend to see a bit higher risk for saturated fat.

SPENCER: For heart disease?

GIL: Yes, coronary heart disease. And this is isocaloric comparisons.

SPENCER: So, that means two people would be given the same exact number of calories in their diet. Is that what you mean by that?

GIL: Right, two populations getting the same amount of calories from these different foods. You tend to see higher risk for saturated fat than some of these foods and no significant difference compared to refined carbohydrate, for example. This is one big source of confusion or controversy because there's a lot of studies that just don't specify what the replacement is. They just say, "Oh, here's populations eating different levels of saturated fat, and we don't see a difference in heart disease risk. This contradicts all of science." No, if you look at what scientists are actually saying, it's to favor foods that are richer in unsaturated fats. It's not to avoid saturated fat like the devil and go eat whatever you want instead, which is what Western populations tend to do. So, this explains a lot. And then there's other details like types of saturated fat. Also, some types of saturated fat don't raise cardiovascular risk, like the saturated fat in dark chocolate, for example, which is called stearic acid. So, these nuances, if you will, kind of complicate the issue. But as we were talking about, if you zoom out and we move from this level of focus on the nutrient to thinking about foods and thinking about dietary patterns, this is all kind of rectified, and we move away from the sources of confusion, really.

SPENCER: People talk about processed meats as being especially risky, either for cancer or heart health. Is this true? And if it's true, why would processed meats be worse? What about the processing makes them worse?

GIL: It's the same types of evidence that we have for unprocessed meat, but the outcomes look worse. So, the hazard ratios are higher so the risk of diseases looks higher. It's more compelling; it's also more consistent in different studies. The level of confidence is higher for, say, bacon or beef jerky — these types of processed red meats. The why is a bit more speculative. There are several possibilities: things like heme iron and nitroso compounds, and Neu5Gc; all these things that are created during the processing that are thought to increase, for example, colorectal cancer risk and also maybe cardiovascular disease. That's kind of why it's said that those are higher risk. I think it's a reasonable and compelling argument that those are higher risk and the different possibilities as to why that would be.

SPENCER: When we get to biological mechanisms, I often hear people on podcasts and so on go on at length about biological mechanisms. To me, this is actually not a great sign because I think of them as pretty weak arguments in favor of things. Not only do I think of them as weak in terms of how much evidence they provide, but also they tend to be very, very hard to follow for a lay audience. You kind of just get blinded by science. You're like, "That sounds really sciencey, and I can't really follow the argument." How should we think about when someone says, "Well, we know that this causes this because of this biological mechanism?"

GIL: Great question. I think you're right. I think it does get overused in podcasts because it sounds sciencey, and it sounds smart to say the name of the molecule that people haven't heard. Also, it sounds novel, and that's a huge factor for virality and all these things. If you tell people a name that they haven't heard before, you're the guy who came up with it. The first time lectins came up, it was like, "What is this lectin thing? I've never heard of it." So, that's a big factor. Now, let me just say that I'm not biased against mechanisms. If you look at my history of scientific publications, actually, most of my work is in this mechanistic realm. So, if anything, I like mechanistic work, but I understand the limitations. Mechanisms are informative for two things, essentially. One is they're hypothesis-generating. So, you find a mechanism, you do an experiment in a mouse or in a test tube, and you find that a molecule activates a pathway. That's an interesting finding. You can then formulate an experiment in humans to go test if that pans out in a human being eating a food. So, that's one important function. The other important function is supportive for what we call outcome data. Outcome data would be looking at a population of humans eating food X: do they have more or fewer heart attacks? Do they lose weight or gain weight? Do they blah, blah, blah? That's outcome data. If you have a mechanistic understanding, that can increase the confidence. Not only do we see the effect, but we know how it might occur. So, that's another role that they play. What they don't do is exactly what you were alluding to, which is throwing out the mechanism and claiming cause and effect and never going to outcomes. That's the pattern of what you see with social media communications, a very common red flag. And once you learn to recognize this, you can tell immediately, especially with these flashy claims that are very counter-current, "Oh, everything you've ever heard is wrong. It's actually the opposite because here's this molecule, this biochemical pathway." You can spot that a mile away once you get used to it.

SPENCER: Yeah. I think one of the challenges is that a mechanistic claim is really hard to refute, but it doesn't mean it's true just because we don't know that it's false. There could be many, many possible mechanisms. The human body is so ridiculously complicated. And often, our mechanisms are wrong. For example, my understanding is there's this giant hypothesis about Alzheimer's disease and how to treat it by targeting amyloids. It all seemed very reasonable, and a lot of experts said it was true. But then we look at the outcome data, and we're like, "Hmm, it doesn't seem like these drugs really help. Or if they help, it's only a very small amount."

GIL: That's actually a great point. Something that I didn't mention that I should have is that mechanistic insight is also crucial to develop therapeutics. So, a lot of medication is based on mechanistic understanding. But a lot of times, you come up with drugs based on something mechanistic, and then they don't work. And so, there's crosstalk. It starts with mechanistic: you develop a drug and test it in humans. Sometimes it does work; it increases your confidence that it's working through that mechanism. And sometimes it doesn't work; it lowers your confidence. So, this is all back and forth. But what you're saying is absolutely true. So, the reason that mechanisms are not very compelling by themselves in the absence of outcome data, or especially if they go against outcome data, is basically due to two reasons. The first is that a mechanism may not pan out in a human being. So, you might see something in a mouse or in a test tube, and the same thing just does not happen in a human being, either because the dose that you used was different or there's some feedback mechanism that shuts down that process. This happens all the time. The other one is what you were touching on. You can have a mechanism that does pan out, so it happens in the test tube, it does happen in the human being, but a thousand other things happen at the same time. So, the net effect goes in the opposite direction. For example, with exercise. Exercise has lots of effects that, if you looked at them in isolation knowing nothing about exercise, you might conclude, "Oh, exercise is bad for us. It raises your heart rate, it raises your blood pressure, your glucose levels go up, inflammation markers go up, it causes oxidation of different proteins." If you just look at the mechanisms, you might conclude it's terrible for you, it's toxic. But when you look at the net effects and the outcomes... So mechanisms are not useless, they're not trash. We want to avoid these exaggerations. They are useful scientific evidence, but it's the overclaim and taking something that's at the bottom of the hierarchy and trying to plug it at the top and erase everything else.

SPENCER: Let's talk about protein because I feel like everywhere you go now, there are protein bars and protein shakes and so on. I've gone online and looked at different protein calculators, and every calculator I find tells me I should be eating three times more protein or five times more protein than I am. How much protein do humans actually need? Are people getting enough? Should we be eating more protein? What's going on here?

GIL: From my understanding from my interviews with the protein specialists on the channel, most Westerners are getting all the protein they need. They don't need to worry about getting more; they're actually getting more than they need. Protein needs vary by age and by level of physical activity. So, one kind of rule of thumb, one number that's thrown out there a lot and I think based on good evidence, is 1.6 grams of protein per kilo of body weight. And that's kind of a threshold where they see muscle growth being maximized. That's more for people who are actively doing resistance training, who are lifting weights, who are trying to put on some muscle. That's one number that's out there. Some people will put brackets around that, so a little bit more, a little bit less, but the ballpark maybe is 1.8 or maybe it's 1.4. But that's one good reference number, I think. And then for people who are just looking for maintenance, 1.2 is another number that Stuart Phillips, one of the premier protein investigators that I've had on, suggests also, in general, for example, for an older population that sometimes needs a bit more protein because there's a certain anabolic resistance. So, sometimes you need a stronger protein stimulus to get muscle to be built, to activate muscle protein synthesis. So, 1.2 is another number. These are all grams of protein per kilo of body weight. And this is all in reference to the RDA, which is 0.8 grams of protein per kilo of body weight, which some evidence suggests is enough for many people but might not be enough for some. It probably won't get you there if you're trying to maximize muscle growth (probably won't do that). And for some people, especially above a certain age, it might also be below your optimum. So, those are some reference points.

SPENCER: Do you think of that 0.8, that's the RDA, as being sort of a minimum?

GIL: Yeah, that's Stuart Phillips' position, that that should be the minimum. And he thinks that should be kind of the bottom floor that we talk about, and then from then on up. So, for some people, it might be fine. For others, they might need more. Yeah, I think that's reasonable.

SPENCER: One of my loved ones is kind of a super athlete, and she eats, I think 200 grams of protein a day. Is there anything to worry about? Because I know some people have said, "Okay, it could be dangerous, maybe it's bad for your kidneys," and so on. If someone's eating tons and tons of protein, but they feel really good, should they worry about that?

GIL: I don't think the kidney damage theory has much evidence behind it; it's mainly been debunked. Maybe one caveat would be if someone has kidney disease already, if they have problems with their kidneys, or maybe they have a family history of kidney problems. So, I don't go crazy on the protein. But for the majority of the population, the idea that eating a lot of protein will damage your kidneys, I think most of the evidence has pointed against that. Most of these ideas have not been supported. There was a recent paper that came out that got a lot of action on the headlines about protein giving you heart disease. It turned out to be mainly a mouse work. Yeah, I haven't seen anything compelling that protein per se will cause much harm. Now, as we were talking before, going from nutrient level to the foods and the dietary patterns, it also depends where you're getting your protein from. If you're getting your protein from certain sources and not others, getting protein from fish and seafood and lentils and things like that in a Mediterranean dietary pattern is one thing. Getting protein from bacon and whatever else is not the same case at all. So, I would make that caveat about the source more than the protein itself. I am not convinced that protein itself is harmful. There are several investigators that disagree with me, especially in the aging field, there is an idea that limiting protein might be better. I suspect it's more about the source of the protein. So, I haven't seen any compelling evidence that protein itself is damaging in humans, but I'm open to it if the data comes.

SPENCER: When you say before that you think that older people might actually need more protein, but you're saying that some people might think that they need less for optimal health. Is that right?

GIL: Good question. So, yeah, in the aging field, there is an idea among certain researchers that not going overboard with protein is better for longevity. But actually, in some studies, for example, Valter Longo from USC, one of the researchers that spearheads this, has some studies where he has this divide at around age 65. So, under 65, it's better to be moderate with protein, but then if you're older, you don't see the detriment anymore. He does make that distinction that once you get older, it might benefit you more to go a little higher on protein. And when you're younger, moderation might be more beneficial. But personally, I don't even necessarily believe that this whole effect is due to protein. So, we might just disagree. I don't know. I'm open to arguments and evidence. I might be wrong.

SPENCER: Let's talk about sugar because that's another food that's obviously extremely demonized. Is it true that everyone should try to reduce their sugar intake, or is it more like as long as you're eating only a modest amount, it doesn't really matter?

GIL: I think it's true that most Westerners would do well to reduce their refined carbohydrate intake. If you look at what's recommended by the guidelines and what people are actually eating in terms of whole grains, almost no one meets the recommendation in the US particularly; it's like poultry. And then with refined carbohydrates, people are eating crazy amounts. So, most Westerners, especially in countries like the US and most of Europe, would do well to reduce their refined carbohydrate intake because they're having a ton. Why exactly sugar, refined carbohydrates, is problematic isn't entirely figured out. There are lots of possibilities. One obvious thing is just the caloric density; they are less satiating. So, if you're eating a lot of refined carbohydrates, a lot of cookies and ice cream — although some of these foods have a lot of fat too, so it's kind of a tricky comparison — but if we look at candy or cookies or crackers that might be more carb-rich and low-fat, those foods are going to be more calorically dense and more calorically concentrated than whole forms of carbohydrates. Also, they have lower fiber. So, all in all, you're going to tend to overeat calorie-wise before you're full, before you're satiated. That is probably one of the big reasons that refined carbohydrate is problematic. Is it the whole story? I don't know. There are other models out there that suggest it causes inflammation, that it changes glucose metabolism. But it's not really controversial that eating a substantial amount of refined carbohydrate is problematic, especially if that leads you to overeat calories.

SPENCER: It sounds like most people would be better off if they switched their refined carbohydrate to a whole grain carbohydrate. In terms of terminology, sometimes you'll see "whole grain," sometimes you see "multigrain." What exactly should they be looking for if they want to do that switch?

GIL: The ideal is to look for whole foods. So, whole grains as close to nature as they are found: steel-cut oats, quinoa, corn on the cob. You can't go wrong with those. Now, I realize that not everybody is going to be doing that exclusively. So, people want to have some bread or they want some pasta. Normally, I'm going for the whole grain. 100% whole grain is a safer choice; basically, it means nothing was removed; all the components are still there, it just got ground down. So, that's generally thought of as superior to the more refined options.

SPENCER: Does it actually make a difference whether you're eating sugar out of a box of sugar or you're eating white bread? At the end of the day, is it kind of the same for the human body?

GIL: Probably doesn't make a whole lot of difference, no, depending on the amount. One interesting thing about satiety is that if you eat pure sugar — nobody that I know binge-eats pure sugar. It's always a combination of things. Same with white bread, people tend not to eat slices of dry white bread. It's usually with something on top, something in between — so, this is an interesting aspect of satiety that these foods that are more craveable tend to mix different registers. So, they might have sugar, they might have some fat, they might have salt. So, usually, you need at least two of these key combinations to have a perfect storm. Just one thing, no one binges sugar out of the tin, no one binges just oil out of the bottle. But if you mix them together, if you have sugar and fat with ice cream, that's delicious. If you have starch with salt, potato chips, very craveable. So, the combination is usually the tricky part.

SPENCER: So, ironically, it could actually be that people might be better off eating pure white bread with no salt, no butter, because chances are they wouldn't be able to eat nearly as much of it.

GIL: You probably would eat a lot less. I don't know if anybody would be able to. You propose to somebody just eat dry white bread, it would be a very difficult diet to even imagine. It reminds me of Walter Kempner's white rice diet (I don't know if you heard about that). This is from the 60s or 70s. For the white rice diet, he took people who were very overweight with Type 2 diabetes and malignant hypertension, and he put them on this crazy diet that was mostly white rice, just pure sugar, and some fruit juice. That was most of it; it was 95% carbs, very little protein. And people lost crazy amounts of weight, crazy fast. Most people improved their diabetes parameters, blood pressure came crashing down. Nobody would recommend this as a healthy diet in the long run, but you get these improvements in metrics in the short term because you get this weight loss rapidly, because it's not a diet that you're going to crave, even though you're getting 4000 calories from it. It's not something that we would call a healthy diet. It's an illustration that the principle of satiety is very powerful. We were talking about elimination diets earlier; this would be an example of an ultimate elimination diet. You're eliminating almost everything. And even though you're left with foods that are not very healthy in and of themselves, the fact that you're forcefully reducing your caloric intake gets you these short-term improvements. It doesn't necessarily mean that you're going to be healthier in the long run; there's a lot of question marks there.

SPENCER: It's really interesting how you've got this kind of effect of weight or excess fat, and then you have this effect of the diet itself. Presumably, if you ate that pure rice and sugar diet for 20 years, it probably wouldn't be great for you. But if in the short term you're losing weight, you might see a bunch of metrics improve. So, there's this idea in society now that maybe it's okay to be obese from a health standpoint. Obviously, people who are obese get a lot of hate thrown at them, a lot of prejudice, which is just awful. But then there's this sort of scientific question of are they putting their health at risk being at a higher weight, and people are increasingly pushing back on that. So, how much do we know about that? Is that a slam dunk where we absolutely know that excess body fat is harming you?

GIL: The distinction here is between population level and individual level, and this is a question that people always raise with any risk factor. So, when we talk about these things and risk of disease, we're always talking about population level. At a population level, we have abundant evidence with Type 2 diabetes, with mortality, with cardiovascular disease, that as people become more overweight and obese, the risk of all these things goes up, even cancer. So, populations exposed to that risk factor — I would say obesity is a strong risk factor — have more incidence of these diseases. That does not mean that every obese individual will suffer from those diseases. This is the case with every risk factor. Some smokers don't get lung cancer because there's always a factor of individual susceptibility. So, you can find people who are obese who don't have Type 2 diabetes, who have never had a heart attack, who don't have any disease and seem reasonably healthy, at least at present. But it's always a statistical question. And then it's a question of making a personal decision based on the information out there: I can choose to smoke or have my blood pressure very high or have my cholesterol super high or be obese, knowing that statistically the risk is higher. Is the lightning going to hit me? Once it happens, you know it's you. But prospectively, the only educated guesses or the hints are the things that we talked about. So, your blood work, if you are obese but your lipids are really good, your glucose levels are good, all these things, your blood pressure is good, it's a sign that you have a stronger constitution; you tolerate it better than the next guy whose blood work is a nightmare. If your family history is solid, your family members are also obese and they live to an older age. Okay, maybe you guys have won the genetic lottery. But population-wise, you'll have more diseases with exposure to any of these risk factors.

SPENCER: And in the randomized controlled trials where people lose weight, how much do they tell us about the benefits of weight loss?

GIL: A lot. Again, populationally. A lot of these findings come from trials, so things like all the way from preventing things, like heart attacks, to putting diseases in remission. So, people with Type 2 diabetes that lose weight put their diabetes in remission, so they effectively no longer have Type 2 diabetes. This can be achieved relatively fast — in a matter of a couple of months — with a dietary change and some weight loss. This has been shown in large, respected randomized trials that are landmarks in this field of diabetes. The direct trial by Roy Taylor has shown this. It's a very powerful data: you take a population of people who are overweight/obese and have some of these problems, and just by losing weight, with all the metabolic improvements that come with weight loss, you see a lot of these improvements at the population level.


SPENCER: What are the strongest markers of health when it comes to obesity? Some people critique BMI, which is essentially weight divided by height squared. But then there's also percentage body fat. What should we actually be looking at?

GIL: Yeah, BMI is good at a population level. Individually, it's tricky because of the muscle. Some people are just more muscular. When you look in the mirror, it's hard to argue with that. Also, if you compare yourself to when you were younger, although now people are getting overweight and obese younger and younger so that breaks apart a bit more. If you were leaner when you were in college or high school, and then when you're 40 or 50, you put on 20 or 30 pounds, for most people that's not pure muscle, unless you're a strength athlete. For most people, it's weight around the waist, legs, and buttocks. That's one way to tell if you put on weight. You can do more fancy tests like DEXA scans and things like that. They're expensive, but they try to estimate your body fat percentage. These biomarkers give you an idea of how your physiology in general is being affected: your blood pressure, glucose levels, lipids like cholesterol, and triglycerides which is another one that consistently goes up. All these things tend to go up with substantial weight gain, and they consistently come down with weight loss. The triglycerides come down. The blood pressure comes down. The glucose levels, if people have them elevated, come down. Inflammatory markers are another one; it's consistently elevated in people who are overweight and come down with weight loss. And then, things at the level of symptoms like pain in lower limbs. There's a lot of these things that tend to improve or even go away with weight loss alone.

SPENCER: In terms of the specific biomarkers you want to measure, let's say you go to the doctor and want to get a measurement to see how you're doing. I've heard that ApoB can be a preferable measure over things like LDL and HDL. Is that true? And what is it really measuring?

GIL: Yeah, in general, it's a more validated, more rigorous metric of heart disease risk than LDL cholesterol and HDL cholesterol. The measures that we normally get — HDL cholesterol, LDL cholesterol, and triglycerides — are basically the types of fats being carried around in our blood. They are reflections; it's almost like looking in the mirror or at a shadow of the things that really cause disease. So, they're useful; they are not useless. They are red flags, but they can get things wrong and miss the picture quite often, actually, depending on the population. ApoB measures the number of what's called a lipoprotein, which is essentially the vehicle that carries the fat around in the blood. Most of the evidence in the cardiovascular research field points to ApoB as the number of these lipoproteins being the main metric of risk or the main cause of risk, not the content of fats. So that's why ApoB is a more reliable measure of risk than LDL cholesterol. For example, you can have somebody whose LDL cholesterol is a bit high but whose ApoB is normal. Population-wise, they're not at higher risk. And vice versa, if their ApoB is high and LDL cholesterol is normal, they are at higher risk. So LDL cholesterol can sometimes throw us off; it's good to bear in mind.

SPENCER: What about measures of glucose? What do you want to actually measure, and what's known about how that's related to health?

GIL: Glycemia or the blood fasting glucose is one metric. You can also measure your glucose after meals. One thing to bear in mind, that a lot of people don't realize, is you can have problems with your glucose metabolism for years before your glucose goes up. So, you can have insulin resistance building up. Similar to what we said about LDL cholesterol being this useful but imperfect metric, that doesn't catch everything. Sometimes you miss the picture by looking at LDL cholesterol. The same thing happens with glucose. It's useful. It's undeniably a red flag if your fasting glucose is high. But if your fasting glucose is normal, you could still have a problem with your glucose metabolism and just not have gotten to that point yet. Basically, your body will try to hold things together, sometimes for years, by producing more insulin to get your glucose out of your blood. It will be able to compensate. But then, you get to a certain point where you lose that ability. You can no longer produce that compensatory bolus of insulin, and you get hyperglycemia. That's when you get into prediabetes and diabetes. So that's something to bear in mind. There are different ways to measure insulin resistance. There are different metrics. There's HOMA-IR, which is basically a fraction: you measure your glucose, and you measure your fasting insulin levels. We have a video going over those formulas, and there are several calculators on the internet. There's another one, the TyG (triglyceride glucose) index, which is a bit easier because you use fasting glucose and triglycerides, which more people have. People usually don't have a fasting insulin measurement. Those are some ways to get an idea of your insulin resistance state.

SPENCER: An idea I've increasingly heard people talk about is that you shouldn't have your insulin spike throughout the day. I hear about people wearing these real-time monitors, and they say, "Oh, when I eat an almond, this is what happens to my blood sugar." Obviously, we know there's a link between blood sugar spiking and things like diabetes. But I think they're making a bit of a leap there saying, "No, you don't want it to spike throughout the day, and if it spikes a bunch throughout the day, maybe this could lead to long-term health effects." What is known about that? Is it true that you should be trying to control your spikes throughout the day?

GIL: I think there's a lot of exaggeration. A lot of time, people have this idea of high glucose because it's diagnostic for diabetes. The diagnosis of diabetes is having high glucose. And also, if you have high fasting glycemia, that's undeniably a metabolic problem. Sometimes this is misunderstood. People say, "Well, after eating, my glucose went up. I'm going to have diabetes." This is a misunderstanding. Your blood glucose levels will oscillate after eating, after exercise, and during phases of sleep. These oscillations, or excursions of glucose, are completely physiological. What's not physiological is if they go up and stay up very high for hours because your body is not able to bring them down. That can also be another diagnosis criteria for diabetes: two hours after eating, if it's above a certain range (like 140 milligrams per deciliter in the US), you also have criteria for Type 2 diabetes. So, the distinction between fasting hyperglycemia, which is bad, and postprandial elevations in glycemia, which may be completely normal. Bear in mind, if it's hours after eating and it stays very high, you might have an issue. There is some debate over peaks that are unusually high and whether that is a harbinger of problems or not. But just this key distinction: high glucose in a fasting state is a problem; glucose going up after a meal is physiological.

SPENCER: It sounds like we're still talking about correlations, though. Either way, is it known whether, if you were to have a diet where you make sure that your insulin never spikes too high by choosing foods that don't cause the spikes, is that actually a sign that you're eating in a healthier way, or will you have better long-term health outcomes on average?

GIL: That's a great question. It's hard to do a randomized trial where you specifically change the glucose spikes. Usually, trials that involve intervention that look at people who already have metabolic issues, like diabetes for example, and they go in with medication to try to reduce glycemia. There are several variables there. There are some experiments, we have a recent video that was trying to tease this apart, these peaks of insulin after eating, they took people who had higher peaks of insulin postprandially, and over the long run, when they corrected for other variables and tried to minimize those, these people actually seemed to have a lower risk of developing diabetes over the long run. They seemed to be protected, possibly because the higher production of insulin after a meal is signaling that these people have healthier pancreases, so they're able to last longer, whereas people who have lower insulin peaks may have pancreases that are already kind of giving out. So I think it's fair to say there is some uncertainty there, but I don't think there's any compelling evidence. I think there is some exaggeration there and fear-mongering also with CGMs (continuous glucose monitors) — by the way, I'm not against CGMs. I think CGMs can be useful and very important for people with Type 2 diabetes. And I'm not against somebody who doesn't have any disease if they want to use a CGM just for kicks. I don't think that's necessarily a problem.

SPENCER: CGM is Continuous glucose monitors, right?

GIL: Yeah, right. For people who are not familiar, it's a device where you're constantly measuring your glucose levels. It's actually not in the blood; it's interstitial. But you can have an app on your phone that gives you your glucose reading in real-time after a meal. You can have this nice graph at very regular intervals. I think there is a lot of fear-mongering and over-interpretation with those. People who have never seen glucose variation after eating might eat a bowl of oatmeal, and they're metabolically healthy, and they see their glucose go up and come down again, and they go, "Oh my God, my glucose spiked. I'm going to have diabetes; I'm going to have this or that." So, just be aware that glucose excursions are normal. If you have pathological changes, like fasting glucose, where your glucose stays high for a very long time, that's a problem. Like I said, there's some debate over people who have unusually high spikes and who may sometimes have symptoms. Some people feel lightheaded after eating a very carby meal, and they look at their CGM and see a very unusually high spike. And then, they change their diet a bit and feel better. I don't see a problem there. I think it's a reasonable course of action. But yeah, just being aware that changes in glucose level after eating are not disease.

SPENCER: It seems that losing weight is one of the most powerful things we can do for health if we're overweight. But if you look at these diet studies, one of the problems with them is it seems like in almost all diets, most people gain the weight back. Is it true that like 70% or 80% of people will eventually gain it back almost regardless of what diet they do?

GIL: Unfortunately, yes. Most trials we do show that if the trial is long enough and with follow-up, people tend to put the weight back on. I think this is seen more consistently the more radical the diet is and the more the diet deviates from the standard diet people are used to. So with all these very low carb, very low fat, and entirely plant-based diets, we see people falling off the wagon. As soon as the trial is over, a lot of times we don't see a difference. People stop doing it and they put the weight back on. So yeah, that's a consistent observation. With something like a Mediterranean diet, there is some evidence that it is more doable for a Western population because it's closer to what they're used to. It's not super low fat or super low carb. You're not completely eliminating entire departments of your diet, so it's a little easier. There's variety as well. Another factor is the social aspect; people want to have dinner with each other and go out with their friends. When you're on a very restrictive diet, it's very difficult to go to a restaurant and find something to eat when you're away from home. When you're with friends, you're the annoying person who always needs special food. So, yeah. There are different ways around that. Some people will "cheat" a little bit when they're with friends or when they go out to eat, which may actually be a good idea and may work. But yeah, that's another reason people often fall off the wagon: they go out to eat with friends, they start eating other foods, and then they just don't go back to their pattern. So absolutely, finding something sustainable that you can stick to long-term, and that is pleasurable to you. At the end of the day, it has to be also something you enjoy.

SPENCER: That makes sense because if you're not going to stick with the diet, you'll eventually gain the weight back. So it almost doesn't matter what diet you do. The first thing is it has to be something sustainable for you, that you can make a permanent part of your life, not just a temporary part of your life. But what is known about different dietary foods? Are there some that seem to promote weight loss better, or does it really not matter as long as you're reducing the number of calories?

GIL: There are foods that are more satiating and nutrients that are more satiating. We have a whole video about satiety and the different triggers. One is protein. In general, higher protein foods are more satiating. Fiber is another trigger, probably less potent than protein but still a factor. Caloric density or energy density is another factor. So, foods that are very energy-dense are less satiating. And this is per calorie, which is a critical metric for weight gain or loss. Obviously, if you eat more of the food with more calories before you're satiated, it's going to be easier to gain weight. Whole foods, foods that are not ultra-processed, are usually a good rule of thumb because they are less calorically dense. Foods that contain more water, foods that contain more fiber, and foods that contain more protein. It's possible to do a dietary pattern around these triggers for any of the popular diets, whether it's plant-based, low carb, keto, or low fat. Another interesting thing is you don't need to hit all the triggers. As long as you hit a couple, you can find a satiating diet. For example, a plant-exclusive diet can sometimes be lower in fat and protein (depending on how it's been put together) but higher in fiber and very good in terms of being calorically dilute. So, that can work for some people. Others might have a keto diet that's lower in fiber but very high in protein and mostly whole foods. So, that might work for them in terms of satiety. You can pick a couple of triggers that you can hit with your favorite diet, and it can work with different diets. It's not not that limiting.

SPENCER: At the end of the day, with weight loss, is approximately all that matters the number of calories you eat? Sometimes people talk as though you can eat certain foods as much as you want and still lose weight. Is that just because it turns out people don't overeat that food so they're not going to eat that excess calories?

GIL: I think that is the main factor. Most of the time when you hear on social media, "It's not about the calories. You can eat X as much as you want." It's an exaggeration or misinterpretation. You can gain weight by eating almost anything if you eat enough of it, but some foods are much harder to overeat than others. That's a key factor. I'm not completely convinced that calories are 100%. I don't want to oversimplify or overclaim. I think there are factors that can modulate. First of all, it's not just about the calories you eat. There's all kinds of factors downstream: How much of that is absorbed, how much of that is then metabolized? All of these factors will matter. You can eat the same amount of food and metabolize it differently. Obviously, you can exercise more. So, I think there are other factors beyond the amount of calories that you eat. Time of eating is one thing that people ask a lot about. There is some evidence that eating the same amount of foods and the same foods actually( the same meals) at different times of the day can have a different effect in terms of how much weight you gain and also some of these parameters, like insulin resistance. I don't think it's entirely nailed down. It's a little controversial still, but there is some evidence that this (what's called) "chronobiology" — basically, at what time of the day you're eating or sleeping — does have an impact on your weight gain, possibly through modulating absorption and metabolism of these foods. So, I think that it's possible that that's a piece of the puzzle. But yeah, I think it's very hard to deny that the big, big, big piece is the amount of calories that you're able to eat at the end of the day.

SPENCER: Does chronobiology suggest eating earlier in the day or later in the day might be better?

GIL: In most cases, for most people, earlier in the day is better. So the same meal earlier in the morning or early afternoon is better metabolically than late afternoon and night.

SPENCER: Before we wrap up, I just wanted to ask you a few questions about sort of the other big category, which is exercise. Is it just sort of uncontroversial? Like, exercise is great, we should all be doing it. It's wonderful for our health.

GIL: With the same caveats as before for patterns of diet and everything, yeah. The evidence is very consistent. We have evidence going back decades that exercise provides all kinds of benefits even beyond calories. So even with populations that are not losing any weight, you still see benefits of exercise added on top in terms of insulin resistance, for example, and cardiovascular health. So it's not really controversial that exercise is beneficial at a population level. Again, at an individual level, you can ask, "Can one person get away with being a couch potato, and the other one not?" Sure, depending on your genetics. And then levels of exercise also, not everybody can tolerate the same types of exercise in the same intensities. And for some people, just walking more or doing some light physical activity — doesn't even need to be exercise per se — might actually be better. Sometimes, for example, for people who are trying to lose weight and who have a history of not exercising a lot, starting very slowly can actually be better. Just walking for 10 minutes around the block daily can be much better because there's less risk of injuries, less risk of pain, and muscle soreness. You're going to ramp up to it, and you're going to be much more likely to stick with physical activity than if it's something that's excruciating and that you're not going to sustain. So yeah, that level of personalization is very important. But in general, I don't think there's any controversy that moving our bodies is better statistically than not.

SPENCER: Is it known where it starts hitting diminishing marginal returns? If you exercise seven days a week, is that better than three days a week?

GIL: I don't know if there's a clear cutoff there. It really depends a lot on the population. I doubt that you see much of a difference between five days and seven days by the time you're getting to those ranges. Also, because when you look at Western populations, it tends to be enriched for the lower rungs of physical activities. A lot of times you see the biggest effect when you go from nothing to something. And then there's a bit of a plateau. So I don't know if it continues to go up. I imagine it'll probably plateau after two or three sessions a week. But again, this will vary depending on the intensity and everything.

SPENCER: I believe it's Peter Attia who claims that Zone 2 exercise is sort of the one we should be targeting for health. As I understand it, it's not super easy to carry a conversation with Zone 2. It's kind of challenging, but you could do it. Whereas Zone 3, it would be very, very difficult. And Zone 1 would be very easy to have a conversation. Do you believe that there's evidence that Zone 2 is actually a preferable form of exercise?

GIL: I don't know if it's been teased apart if it's just because you're doing more of it, and so you're getting more cardiovascular capacity, more weight loss. I don't know if that's been teased apart. One thing that has emerged that I think is very compelling is the types of exercise. So historically, cardio got all of the spotlight in guidelines and everything. And more and more over the years, the evidence has been mounting on resistance training: lifting weights and challenging your muscles. Even the American Heart Association recommends resistance training. So, bearing in mind, combining these two — trying to have some resistance training in your program and adding that to cardio — I think is a great idea for most people. And one mental barrier: sometimes people think, "Oh, weights are just for 20-year-olds or for dudes who want to get jacked." None of that is true. That's a big kind of limiting belief. Lifting weights can be very beneficial for women, for people of more advanced age, even people who have some diseases or some limitations. Of course, you're going to personalize the program. The types of exercises you do, the loads are not going to be the same. But challenging your muscles can be very important, especially sometimes after a certain age. We talked about this anabolic resistance, even if you're 60 or 70, you want to have some form of resistance training to maintain your muscle mass to avoid sarcopenia with age.

SPENCER: Do you think it's additive where cardio exercise plus resistance is better than either one on its own?

GIL: For some outcomes, it is. It depends on the outcome that you look at. For some things, you basically see that they're equivalent. But for some outcomes, you do see an added benefit. For insulin resistance, for example, you do. So yeah, I think if you do both, you are getting gains that you're not if you're just doing cardio. It also depends a bit on the type of individual. For example, people who are very overweight or obese tend to have more muscle mass under the fat. So the priority there oftentimes is to lose some of the fat mass because you want to find this balance of muscle-to-fat. So doing cardio and losing some of the fat mass is more urgent. And when you lose weight, you always lose some muscle mass also, but you want to minimize that. And if you have a lot of muscle mass under the fat and you lose some of that, but you lose a lot of the fat, that's still a net positive metabolically. But when you're (what's called) skinny fat, you have too much fat, but you don't have that much muscle to begin with, you don't want to lose more muscle. Sometimes even if you're losing fat mass, that can be a net negative metabolically. So particularly for those body types, you want to make sure that you: a) have an appropriate protein intake, which we've talked about; and b) do resistance training to minimize muscle loss or even gain some. It is possible to gain muscle at the same time as you lose fat. It's not the easiest thing to do, but it is possible.

SPENCER: So that would suggest that while dieting, it might be a good idea to have a reasonably high protein intake, and maybe do resistance training, to help just reduce the amount of muscle loss that can naturally occur from dieting, is that right?

GIL: Absolutely. So the protein has a double whammy there. It helps with satiety, and it helps with body composition. So diets with higher protein are very popular for weight loss, and I think for good reason. It helps you lose weight, helps you feel satiated with fewer calories, and it helps you maintain your muscle mass, which is crucial.

SPENCER: Gil, final question for you. Nutrition as a science gets a lot of hate, where people say, "Oh, well, nutrition scientists don't really know anything. The whole field is just a bunch of bunk." Why do you think that there's this belief? And what's your response to people with these claims?

GIL: First of all, I think it's a natural reaction. Most people's exposure to nutrition is not actually directly to the science, it's to headlines and social media content. So I think it's very understandable because the headlines and the social media content are artificially polarized. So these headlines, doing a 180 every week, and giving people whiplash, it's done by design to get people clicking. But it doesn't really reflect the progress of science. So to some extent, this idea that scientists can't agree on anything, they're always changing their mind is a misperception of nutrition science. Don't get me wrong, there is some nuance, there are lots of unknowns, scientists are working on it, but the fundamentals have been there for decades. If you look at the guidelines going back 40 years, they are very similar on the fundamentals. That's something to bear in mind. Another thing is, a lot of science that comes out on nutrition is annoying. It's inconvenient. It's kind of a party pooper. All the foods that we like seem like we can't eat them or we have to moderate them. So, it is a natural reaction there. But notice also that people's resistance to the guidelines is very selective. When the guidelines say don't eat a lot of junk food, nobody disagrees with that. People might not follow them, but nobody says the guidelines are wrong on that. When the guidelines say don't smoke, exercise, nobody disagrees with that. It's very specific things that people disagree with. So a lot of times the guidelines don't match people's personal preferences. And I think we don't need to justify ourselves if people want to eat differently from the guidelines. It's their prerogative to do it. Guidelines are not laws. You're not breaking a law by eating differently. The guidelines are there to assist us, to help us make an educated choice, but at the end of the day, it's your call.

SPENCER: Gil, thanks so much for coming on. It was a fascinating conversation.

GIL: Thanks, man. My pleasure.


JOSH: A listener asks: "What issues in politics do you think are most misunderstood?"

SPENCER: It's a tough question. But one that comes to mind for me is this idea of dark matter in politics. And that's the concept that — We see a conversation happening about politics, right? It's the conversation we see in the newspaper. It's what's being talked about on social media, and YouTube videos, and so on. But there's a huge amount that goes on in politics that we're not aware of. The private backroom conversations, the deals being struck, and so on. And so I think one of the most misunderstood things is that people confuse the surface level conversation for the real conversation. Not to say that there's no relationship between the two. Obviously there is a connection. Sometimes the surface level conversation and the deeper conversation behind the scenes are related. But they aren't necessarily that related. And there can be a lot going on that you're just not aware of. And so I think there's this show that we get to watch, and then there's the real stuff. And it's difficult not to confuse them with each other.




Click here to return to the list of all episodes.


Sign up to receive one helpful idea and one brand-new podcast episode each week!

Contact Us

We'd love to hear from you! To give us your feedback on the podcast, or to tell us about how the ideas from the podcast have impacted you, send us an email at:

Or connect with us on social media: