The Peter Attia Drive - #314 ‒ Rethinking nutrition science： the evolving landscape of obesity treatment, GLP-1 agonists, protein, and the need for higher research standards ｜ David Allison, Ph.D.

00:00:00.000 Hey, everyone. Welcome to the Drive podcast. I'm your host, Peter Atiyah. This podcast,

00:00:16.540 my website, and my weekly newsletter all focus on the goal of translating the science of longevity

00:00:21.520 into something accessible for everyone. Our goal is to provide the best content in health and

00:00:26.720 wellness, and we've established a great team of analysts to make this happen. It is extremely

00:00:31.660 important to me to provide all of this content without relying on paid ads. To do this, our work

00:00:36.960 is made entirely possible by our members, and in return, we offer exclusive member-only content

00:00:42.700 and benefits above and beyond what is available for free. If you want to take your knowledge of

00:00:47.940 this space to the next level, it's our goal to ensure members get back much more than the price

00:00:53.200 of the subscription. If you want to learn more about the benefits of our premium membership,

00:00:58.000 head over to peteratiyahmd.com forward slash subscribe. My guest this week is David Allison,

00:01:06.040 who returns to the Drive for a second sit-down. David is currently the Dean and Provost Professor

00:01:11.180 at the Indiana University Bloomington School of Public Health. He's authored over 500 scientific

00:01:16.540 publications and received many awards, and his research interests include obesity and nutrition,

00:01:21.540 quantitative genetics, clinical trials, statistical and research methodology, and research rigor and

00:01:27.620 integrity. In our conversation today, we discuss the relationship between nutrition, obesity,

00:01:32.700 and body composition, and how food affects body composition beyond caloric intake. This leads us

00:01:38.340 to a discussion around the complexity of nutrition research studies and how confusion continues to

00:01:44.520 remain with translating knowledge into practical outcomes, such as reducing obesity. We talk about the

00:01:50.140 public health efforts and policy and why they have failed historically in regard to obesity and why

00:01:55.900 there's such a trust problem with nutrition science. Next, we dive into the emergence of GLP-1

00:02:00.540 agonists in treating obesity and what is happening both socially and psychologically with drugs like

00:02:06.280 Ozempic and Manjaro. We end the discussion talking about protein intake and the adequacy of current

00:02:12.640 protein intake recommendations and the research gaps that lie between what we are told and maybe what is

00:02:19.000 actually known. Overall, this was a fascinating and philosophical at times discussion on the

00:02:24.400 evolving landscape of nutrition science, obesity treatment, and the impact of research. Without

00:02:29.160 further delay, please enjoy my conversation with David Allison. David, good to see you once again.

00:02:39.280 Good to see you, my friend. Lots to talk about today. The world of nutrition and health are always in the

00:02:45.980 spotlight, in particular around a class of drugs that no listener to this podcast will be a stranger to

00:02:52.720 called GLP-1 agonists. So I want to spend some time talking about those, but I think before doing so,

00:03:00.020 I want to just maybe go back and talk a little bit about what we know and maybe don't know about the

00:03:04.400 relationship between nutrition and obesity, which sounds like it should be obvious. So tell us what you

00:03:10.600 think is actually known about the relationship between food and body composition.

00:03:18.100 So I like the way you phrased the question and using the phrase body composition as opposed to

00:03:23.800 just obesity or weight. There are obviously three different things. Obesity implies a threshold,

00:03:28.680 you're too much. There's a judgment about the effects of the excess. Then there's body composition,

00:03:35.120 the tissue, how much is fat, how much is lean, where is the fat lean, what is the fat composed of,

00:03:41.720 what is the lean composed of, and then there's just weight, which is just your mass on this planet.

00:03:47.280 And those three things are highly related, but not identical. What we know indisputably,

00:03:54.420 and even people who sort of rail against something they call the energy balance model, which you and I

00:03:59.860 have discussed, whether it's really a model is unclear. It's really more of a constraint.

00:04:04.800 It's really a restatement of the first law of thermodynamics, which is the law of conservation.

00:04:09.840 Matter and energy can neither be created nor destroyed, but only converted. It is a constraint

00:04:14.400 by which all other descriptions of what happens with weight and mass and food intake and energy

00:04:21.160 intake and energy expenditure must operate. It's not a description or an explanation of what happens.

00:04:27.400 It just says, if you describe any proposed explanation of what happens, it's got to follow

00:04:34.360 that first law of thermodynamics in order to make sense. And that first law of thermodynamics in

00:04:39.560 the field of nutritional obesity often gets stated as something like changes in energy storage equal

00:04:46.460 changes in energy intake minus changes in energy output, or delta energy stores equals delta energy in

00:04:52.960 minus delta energy out. Food intake can affect those things. Alternatively, you could say that energy

00:05:00.820 intake is one of those things. So, it gets back to that descriptive thing. Now, one of the questions

00:05:06.600 becomes, how does all the other aspects of food besides the mere energy content of it affect the amount of

00:05:16.480 weight one gains or loses? The body composition, the tissues, where the mass is distributed, what types of

00:05:24.340 tissues it's in, composition of those tissues, and then, of course, whether or not one exceeds some

00:05:30.400 threshold. There's every reason to believe that many, many aspects of food, from the marketing and pricing

00:05:38.140 of it, which then can influence the intake of it, to the taste, the smell, the timing, what you eat it with, what

00:05:48.580 it's combined with, phytochemicals in it, micronutrients, macronutrients, all can affect energy expenditure, subsequent

00:05:58.000 energy intake, or nutrient partitioning, which is a fancy phrase for where you stick the energy that you store in the body. Do you stick it into

00:06:08.000 fat, fat, or muscle, or bone, or visceral fat, or subcutaneous fat, etc.? So, all those things can come into play. Now, what do we

00:06:15.760 really know? The truth is, I think what we know is modest, and partly that's because it seems to me to be very specific. That

00:06:27.040 is, we can do a study, and even when it's honestly done, and well done, and honestly reported, and we find that in this

00:06:34.300 species, with this delivery of this composition, in this way, this thing happens. And then when you look

00:06:44.080 in a different species, or a slightly different food, you get different results. So, there's many, many

00:06:50.980 studies saying, well, we got this with pea protein and casein, but not whey. Or we got it with whey, but not

00:06:58.260 casein. Or we got it when we fed it two hours before the test meal, but not one hour before. Or we got it

00:07:05.920 in men, but not in women. This makes me think we're talking often about subtle effects that may not be

00:07:13.440 that clinically reliable and meaningful. And so, the really big effect seems to be, how many calories do

00:07:23.200 you eat. But all these other aspects of food may then influence how many calories you eat, either of

00:07:31.300 that food, or in subsequent occasions. And those can seem to have big effects, but we're still sort of,

00:07:37.440 I think, trying to suss those out. Do you ever spend time interacting with physicists, or chemists,

00:07:45.200 biochemists, who sit on the sidelines and sort of look at the field of human energetics?

00:07:51.640 And wonder to themselves, why is there so much noise? And why is there so little understanding?

00:07:59.740 And I don't think anybody is standing around blaming the scientists and saying, well,

00:08:04.420 in physics, we have great scientists. In chemistry, we have great scientists.

00:08:08.080 In human energetics, they must be subpar, and that's why they don't know anything. I can't imagine

00:08:12.740 there's anybody that thinks that. What do you think it is at the meta level that explains the obvious,

00:08:19.320 but important observation that our knowledge in this space is woefully deficient relative to the

00:08:30.440 effort that has been put into elucidating truth. Just to restate that more poignantly, for how hard

00:08:38.780 the field of science has worked to try to get at the questions we're going to discuss today,

00:08:43.860 why do we know very little relative to the same amounts of effort that have gone into physical

00:08:50.460 sciences, for example? I think there are many reasons. Some are perceptual, and some are actual.

00:08:57.620 Some of the perceptuals, do we really know that much less? And we can argue about it. I think there's

00:09:02.780 still questions in physics where we say, gee, we really don't know that exactly how is it that

00:09:07.440 relativity and quantum physics are compatible, or is dark matter real, or what have you? I think

00:09:14.640 there are questions there. Though I would sort of just interject for a second, David, and say

00:09:19.280 another way to think of it would be if you look at the amazing progress that has been made that has

00:09:25.760 been enabled by the knowledge of physics and chemistry. If you just consider what's happened

00:09:30.500 in the last hundred years in terms of what we've been able to do, just look at computing power,

00:09:36.900 look at semiconductors, look at airplanes. I mean, look at technology that has been enabled by

00:09:43.020 engineering, physics, chemistry. We're multiple logs of advancement. The same cannot be said of

00:09:52.780 what we're talking about now. Our understanding of obesity a hundred years ago versus our understanding

00:09:59.580 of obesity today, while maybe greater, hasn't actually translated into a multiple log improvement

00:10:06.840 in the outcome of interest, which in this case might be a reduction of obesity, just as it might be

00:10:12.700 in the interest on the other side, which would be computing power. I think that's only half true.

00:10:18.040 I think we don't give ourselves certain credit for certain things. In physics, there's not a lot of

00:10:22.460 discussion in modern times of the power of Newton's universal law of gravitation. Those are pretty big

00:10:29.360 deal and a pretty big accomplishment, but we don't talk about it a lot. It's been figured out a long

00:10:33.080 time ago and we take it for granted that we know that now.

00:10:36.620 But do we even have the equivalent in energetics?

00:10:38.840 Yeah, I think we do. Some simple examples, both at the practical level. In this country and in most

00:10:44.720 industrialized countries, there's very little food shortage. That's a big deal. It is a big deal

00:10:51.160 that we know that alcohol contains calories. We take that for granted, but Wilbur Atwater,

00:10:56.540 who's the person who stated that, was vilified for it at that time by the temperance movement.

00:11:02.260 And he himself was a teetotaler, by the way, that alcohol had no nutritional value. And he said,

00:11:07.460 no, it doesn't. It's seven kilocalories per gram. So that's an example. Folate supplementation,

00:11:13.140 which has radically reduced spina bifida. Iodized salt. Micronutrient deficiencies being maybe not

00:11:21.720 eradicated in this country, but radically reduced to, among other things, to supplementation.

00:11:27.760 Greater food safety. So we've made a lot of practical progress. We've feeding a number of

00:11:35.380 people through nutrition and agriculture that back all the way to Malthus, but even more recently in

00:11:42.720 the 1970s, when we were told there was going to be a population explosion that would threaten our

00:11:47.880 ability as a species. But isn't that really more about agriculture than nutrition science?

00:11:52.700 It's agriculture. It's food science. But some of the nutrition science is more the micronutrients,

00:11:57.440 all the way back to eliminating scurvy through the work of James Lind and figuring out

00:12:01.980 eventually that it was vitamin C. They first thought it was just citrus in general. They didn't understand

00:12:07.080 it was the vitamin C to the folate and so forth. I think our notions, our understanding about LDL

00:12:13.480 cholesterol, which again, you know more than I about, is very important. The role of saturated

00:12:18.440 fats and that. We're still learning more, but we do know some things about that. So I don't think we

00:12:23.800 want to take for granted that we have learned a great deal. In obesity itself, until about, oh,

00:12:31.340 I don't know, maybe five years ago or a little more, when I would give talks about this, I would say,

00:12:36.840 we actually have learned a lot, but it's just not all that clinically relevant. And what's clinically

00:12:43.600 relevant is mostly truly in the clinic, not in the community and the population. So I said,

00:12:49.600 we've learned a lot about genetics and that's true. We have log orders, I would argue, magnitude

00:12:55.960 increase in our knowledge about the genetic underpinnings of obesity that we didn't have prior to 1980.

00:13:03.620 But until recently, we've had moderate improvements in the clinic and virtually no improvements in the

00:13:14.000 sort of public health community domain. If you allow me to be humored with an analogy though,

00:13:21.180 just because I'm going to keep pushing back on this a little bit. When the Wright brothers first put

00:13:25.500 an airplane into the sky, I don't think anybody would have said aviation is amazing. That was a proof

00:13:30.880 of concept. It was a wonderful example, but I think it's safe to say that almost monotonically

00:13:37.660 aviation has become safer and safer and safer over the past hundred years. And I think that allows us

00:13:45.020 to say our understanding of Newtonian physics, Bernoulli's principle, material science, all of the

00:13:53.160 things that enable aviation to be what it is today, relative to a hundred years ago, are probably

00:13:59.180 getting better. And we're also getting better at applying them to a real world problem.

00:14:04.180 Conversely, if the rate of airplanes falling out of the sky were increasing steadily over the past

00:14:10.540 50 years, such that in 1970, whatever, you know, 10% of airplanes fell out of the sky, but today 50%

00:14:18.580 of airplanes fell out of the sky. I don't think anybody would be walking around saying we're doing

00:14:22.860 really well. We understand much more about the physics of the airplane. Yes, it's true.

00:14:27.840 More of them are falling out of the sky. And yet I would argue that in the presence of all of this

00:14:33.800 knowledge that we have, we're getting fatter and we're getting sicker. So how do we reconcile the

00:14:40.600 fact that our knowledge is somehow increasing and we're so much more knowledgeable and yet the

00:14:45.900 actual problem that matters seems to be getting worse, not better?

00:14:51.160 Right. Well, what we don't have is, again, with a couple of exceptions, we're going to get to later,

00:14:56.320 I think. We don't have the sort of sea change, the real orbit jumps in knowledge of a utilitarian,

00:15:05.800 useful knowledge, knowledge that helps us change the way we do things now that lead to better outcomes

00:15:12.300 that we don't yet have. So we have useless knowledge?

00:15:16.740 What's the contrapositive of that?

00:15:18.160 We have knowledge that is useful for understanding and we hope we can build on to get to practical

00:15:26.140 knowledge. Steve O'Reilly gave a nice talk about this about two years ago at the Royal Society

00:15:32.960 meeting that he and I and others spoke at and hosted. And he said, as a physician, geneticist,

00:15:40.100 biochemist who works in the field, he looks at this and he also thinks about his early days in blood

00:15:46.460 pressure. And when he started his career a few decades ago, he said, we didn't really have a

00:15:51.240 lot of good drugs and blood pressure. And people kept hammering at the molecular biology and the

00:15:56.100 biochemistry and the physiology of blood pressure. And bit by bit, things started to break. And he

00:16:02.240 says, now we can treat blood pressure enormously better. And he said, I think that's where we're

00:16:07.400 going to go with obesity. And he said, but we're just sort of getting to the breaking point.

00:16:11.960 I think that's what we're seeing now with the GLP-1 agonists, as well as some other drugs.

00:16:17.480 So in other words, we might get to a point in 30 years where we're sitting here and obesity rates

00:16:23.640 are back to the level they were 50 to 100 years ago. Virtually everybody will be on a drug,

00:16:30.380 which we may or may not understand the mechanism of action for.

00:16:33.940 I think we will understand more of the mechanism of action 30 and 50 years from now, but it is true

00:16:38.580 that today we don't fully understand the mechanisms of action.

00:16:42.840 I think that's a reasonable analogy comparing it to blood pressure or comparing it to lipid

00:16:46.540 management for that matter. Even 40 years ago, we didn't really have tools to manage lipids.

00:16:52.080 And where is the investment going? So there the investment was going not only, but heavily toward

00:16:58.880 biochemistry, molecular genetics, physiology, and pharmaceuticals. We are now seeing an uptick

00:17:07.260 in that. We've seen an uptick and we're seeing more of an uptick in that because some success is

00:17:12.540 being achieved. And the pharma companies, many of which who over the last few decades would be

00:17:17.560 tepidly in and out, they'd dip their toe in the water of obesity, wouldn't go so well,

00:17:22.100 they'd pull out. Now they're saying there's real success coming.

00:17:26.260 So there were a couple of recent Cochran collaborations that came out discussing the

00:17:32.640 success or lack thereof of public health initiatives around obesity. Do you want to say a little bit

00:17:39.340 about those and maybe also talk a little bit about the history of why, if I'm going to be blunt,

00:17:44.780 if I'm going to extrapolate from what we've just said, one would say that public health efforts to

00:17:50.400 curb obesity have been a failure and the future of obesity management will be pharmacologic,

00:17:59.240 not public health related. Is that a fair prediction?

00:18:02.640 I think it's a very reasonable prediction. I'm not sure it's one I will share completely.

00:18:07.600 I would share the first part that public health efforts to affect obesity in a meaningful way

00:18:14.580 have thus far been singularly unimpressive. And we'll come back maybe a little bit to

00:18:20.220 why and where that's going and where we should go with it. I do think in the present and even more

00:18:27.720 so in the not too distant future, clinical management, including surgery and pharmaceuticals

00:18:34.980 evermore will be ever more powerful, safe, effective, and utilized. I don't think they will

00:18:40.800 ever become the complete solution. And I don't think that there's no solution in public health,

00:18:46.440 but I think we've got to approach it differently. So let's go back in time a little bit.

00:18:51.780 When I started my career as a real professional, basically 1991, I come to New York, the New York

00:18:59.340 Obesity Research Center at Columbia University in St. Luke's Roosevelt Hospital. It's the only federally

00:19:06.060 or NIH funded obesity research center at the time. It's the first. It's run by Xavier Pissigny,

00:19:13.460 the legacy of Ted Van Italy. Across the park, you've got Jules Hirsch, Rudy Leibel, and that group

00:19:19.900 at Rockefeller. And it wasn't at the level of public interest that it is now the topic of obesity.

00:19:28.740 What you had is these very interdisciplinary groups, physiologists, geneticists, physicians,

00:19:34.800 psychologists, statisticians, nutrition scientists, et cetera, all working together on these problems.

00:19:40.740 Many had been working together for decades, very academic, but also clinical. And you had the

00:19:47.620 powerhouses that were in that region. You had Mickey Stunkard over at UPenn. You had

00:19:52.320 Marcy Greenwood and others at Vassar and so forth. And if a young person like me made some foolish 0.99

00:20:01.000 statement in a seminar about some aspect of physiology or medicine that showed that given my training,

00:20:07.840 I had no understanding of what the heck I was talking about, one more senior person would put

00:20:12.800 me in my place, but in a very constructive way and explain that I didn't know what I was talking about.

00:20:19.120 And the statisticians would argue with the physiologists and so on. So you had a depth

00:20:23.600 of knowledge and a real depth of expertise and an understanding. Then NHANES-3 data came out

00:20:30.840 and there was the sense of crisis, panic, public health.

00:20:34.760 That was what, 94?

00:20:35.800 It started to come out in the early nineties, the midpoint of it.

00:20:39.160 Ninety, 91 was just starting.

00:20:41.880 Tell folks what NHANES is and what the data showed.

00:20:44.440 This is the National Health and Nutrition Examination Survey. It was, at the time,

00:20:49.640 only done every few years. So first there was something earlier in the sixties called NEFIS,

00:20:54.280 I think National Health and something else. Then they developed the National Health and Nutrition

00:20:59.880 Examination Survey. They did two of them. And then the third one was done, I think, between 88 and 93,

00:21:08.360 maybe. So I think the midpoint data they released around 91, if my memory is right. And people started

00:21:17.080 using the word epidemic and they saw what looked like a jump. Whether it was a real jump or not,

00:21:22.680 around the late eighties or in the eighties is actually not so clear. If you look at skin folds,

00:21:27.800 you see less of a jump and you see the increase starting earlier. If you look just at BMI and

00:21:32.600 you look at increase, it's been going up for hundreds of years. The data from the Nobel laureate,

00:21:37.800 Robert Fogel, who won it in economics. He's since deceased, but he's a terrific, generous guy.

00:21:43.480 And he collected all these old data on British Naval cadets from the 18th century and French cadets and

00:21:50.920 Civil War soldiers and recaptured slaves during the Civil War and looked at these different groups.

00:21:57.880 And you see that obesity levels and BMI have been increasing for centuries. But they clearly

00:22:03.400 did seem to be an acceleration and that caused a panic. And then you had probably the most powerful

00:22:09.720 voice at the time in this domain was Kelly Brownell. Kelly had been a real devotee of Mickey

00:22:16.840 Stunkard. He was one of Mickey Stunkard's proteges and mentorees. And he was a behavioral psychologist,

00:22:23.880 still is a behavioral psychologist, doing behavioral treatment. As a grad student,

00:22:28.440 I'd go to his lectures and learn the mechanics of how to do behavioral treatment, cognitive behavioral

00:22:33.320 therapy for obesity. Meaning CBT to help people eat less?

00:22:38.280 Yes, eat less, exercise more, and so on. And then he had a change in the, I guess this would

00:22:45.240 have been the very late 80s, early 90s. He shortly thereafter switched to Yale. He got a MacArthur

00:22:51.800 Prize, the so-called genius award. And he started to look at maybe concerns about the negative effects

00:22:57.240 of obesity. And he was one of the most powerful, not the first, but one of the most powerful voices

00:23:02.120 to start raising questions about the effects of yo-yo dieting or weight cycling going up and down.

00:23:08.120 Are we doing more harm than good? Are we just building false hopes up for people because obesity

00:23:14.200 treatment is useless? And that started to change. And then he morphed into, it's the environment.

00:23:20.760 And he introduced, at least to me, the phrase toxic environment. We live in a toxic environment.

00:23:26.120 You can't drive down the street, he would say, without encountering a fast food restaurant.

00:23:31.080 And so, this is the problem. We need to stop the individual treatment. He sort of abandoned his

00:23:36.120 roots. We need to go to the public health treatment. Many others were grasping that idea,

00:23:41.880 inspired often by him, but others on their own. And the public health community rushed in.

00:23:46.920 And this was a community that was, up until that point, working on smoking or what types of-

00:23:53.080 Smoking, food safety, all kinds of things like the sanitation, vaccination, so on.

00:24:00.440 They rushed in. And I think there was a lot of sense of, this is simple. People eat too much,

00:24:06.760 they don't exercise enough. Eating less is good. Eating more, quote unquote, healthy food is good.

00:24:12.920 Some foods are considered healthy, some are not. And if you eat the healthy food, something magical

00:24:17.400 will happen. More exercise, of course, without any real understanding of this. I've had public health

00:24:23.560 people who said to me, one person wrote and said, well, I think if we got people to not walk with

00:24:30.280 their iPhones, then they would walk a little faster. And then they would expend more energy

00:24:36.200 while they're walking across campus. And that will help with weight loss. And what do you think?

00:24:40.520 And I thought, well, we're still going to cover the same distance. There's a non-linear relationship

00:24:45.080 between walking speed and energy expenditure, and the amount of energy is trivial, and et cetera,

00:24:51.320 et cetera, et cetera. And I just thought, nobody who understands movement science and energetics and

00:24:56.680 kinetics would make such a statement. But if you're a public health person, and you just think,

00:25:02.840 I just need clever ways of getting people to behave the way I already know they should behave,

00:25:07.080 then you come up with ideas like that. That's where if you're embedded in a group of people,

00:25:12.040 that doesn't happen. That wouldn't have happened at the New York Obesity Research Center. If I had

00:25:16.040 said that in 1991, I would have been immediately educated by senior people who had been thinking

00:25:22.920 about this. So in other words, the public health field wasn't really able to self-police

00:25:27.080 ideas that were not grounded in science. That's right. One of the things that we're very proud

00:25:30.440 of in our school, School of Public Health in Indiana University Bloomington, is that we have a

00:25:35.080 kinesiology department as in exercise science. And we're only one of four schools in the United States,

00:25:41.640 schools of public health, that have a named kinesiology department. Now, every school of

00:25:45.800 public health studies physical activity, but studying physical activity and being an expert

00:25:50.040 in exercise science are two completely different things. We have actual experts in exercise science

00:25:55.320 who understand this, who treat it as a science and take it seriously. We're very proud of that

00:26:00.600 and they do great work. So we got a lot of nonsense rushing in the field. We got a lot of things that

00:26:05.800 never would have, in the beginning, you wouldn't have predicted to work. But people tried them because

00:26:11.480 they sounded good, they felt good, gave people a positive feeling, vending machines, farmers markets,

00:26:18.920 walking trails, without really saying, all right, let's really work this out. How many people are going to

00:26:24.440 do it? If they do it, how much effect will it have? Will they compensate by eating more or less

00:26:30.520 or moving more or less later? Those things weren't done. So we've got a whole couple of decades of

00:26:36.280 lousy, uninformative research. But while the public health movement was taking hold,

00:26:43.960 what happened to the guard, the old guard, so to speak, that you referred to having learned the

00:26:50.200 physiology of obesity in the late 80s, early 90s, what was their response to this? Were they a part

00:26:56.760 of the movement? Were they distinct from the movement? I think it was a mix. As with anything

00:27:01.960 with many people in politics and money and careers and dominance and egos, all kinds of interesting

00:27:08.440 things happened. So some of the real strong behavioral people rushed in, did good science,

00:27:13.800 in the sense of things that were rigorous, but maybe not always well-conceived that they were

00:27:19.720 likely to be impactful, but there was grant money to be had and people went after it and still do and

00:27:25.080 so be it. So was this also just driven by funding? Was there a change in funding priority?

00:27:29.240 There was a lot more funding for obesity, still not as much as many people, including me, would like,

00:27:34.360 not as much as perhaps could have or should have been, but definitely big increases. There was the

00:27:40.120 Robert Wood Johnson Foundation, which didn't put a huge amount of money in, but put money in,

00:27:45.000 whether it was intentionally or not, in a very strategic way, meaning they put in,

00:27:50.520 relative to what NIH or pharma put in, they put in a small amount of money, but it sounded like a big

00:27:57.560 amount of money. It had millions. They made a lot of noise about it, very successfully. They got a lot

00:28:03.640 of careers started. They drew a lot of people into the field around public health, around community

00:28:08.360 intervention, around diversity issues, and that's all to the good. It's great. What I think often

00:28:14.120 happened is the amount of money they were dangling in, again, relative to NIH was small. And so many

00:28:20.520 people would rush in, get started that way, but then they'd go to NIH and get bigger studies. So it

00:28:25.800 did catalyze a lot of activity and that's good. What we've learned, you can make different arguments

00:28:31.240 about it. The very famous story with Edison, where his backers come to him and whether it's true story

00:28:36.120 or not, I don't know, but say all this time and money and you still have nothing to show

00:28:41.720 for your efforts toward making a light bulb. And he said, no, I now know a thousand ways not to make

00:28:45.880 a light bulb. These two new papers you referred to earlier were from Summer Bell. She was the senior

00:28:51.640 author, not the first author. They're just out in the last week. I just put up a LinkedIn post on them

00:28:56.440 about three days or so ago. What they do is one is in children, five to 12, I think. The other is in

00:29:02.760 adolescence. And they do systematic reviews, very thorough, very objective meta-analyses,

00:29:09.480 according to the Cochrane method. And what they find is that for both groups, there is no compelling

00:29:16.280 evidence of what you would call a consistent, reliable, long-term, clinically or public health

00:29:22.280 meaningful effect on preventing obesity in either children or adolescents. I included a lot of

00:29:28.680 adjectives in there and those are important. The most important one is probably preventing.

00:29:32.760 I didn't say treating, I said preventing. That doesn't mean that treatment of obesity in children

00:29:38.600 doesn't have any efficacy. The second thing is these are community diet exercise interventions.

00:29:44.280 By the way, how easily do we distinguish between prevention and treatment? Obviously,

00:29:48.040 conceptually it's trivial. Prevention is reducing the number of new cases. Treatment would be reversal

00:29:55.240 of. But can you give a sense of what reversal of obesity rates look like?

00:30:00.680 With public health stuff, I think it's, I don't want to say it's zero because you just always have

00:30:07.160 some spontaneous reversal. You said a moment ago that this article focused on the prevention side,

00:30:13.960 not the treatment side. Correct.

00:30:15.400 Was the implication of that, that public health treatment has been successful, but prevention has

00:30:20.840 not? No, no, no. I'm sorry.

00:30:21.640 I see. Okay.

00:30:22.360 Clinical treatment, I think there's some evidence for success, more so in adults,

00:30:29.000 but some evidence for success. Does clinical treatment include drugs and surgery?

00:30:33.080 It can include, but it doesn't have to. The idea of somebody coming to the clinic, they go to see,

00:30:39.320 let's say, a Len Epstein at Buffalo. They go to his clinic and he puts them in a study or

00:30:44.760 what have you for weight loss. I would call that a clinical intervention as opposed to Len Epstein

00:30:49.800 saying, I've got an idea. I'm going to go out to the public schools and set these programs up and we'll

00:30:55.560 try to get everybody to be less obese and see if we prevent obesity. That's how I distinguish those

00:31:00.760 things. Len, by the way, is one of the sharpest cats around and he's a very good skeptic and a very

00:31:06.840 good commenter on what we really know and don't know in that domain. But what Somerville shows

00:31:12.840 basically is not only is there no compelling evidence for effects, but there's reasonably

00:31:17.400 compelling evidence that given the methods we've used today, the effects are either zero or trivial.

00:31:26.200 And so, I think that's really important because I hear tremendous defensiveness now

00:31:33.720 among people who are not practitioners of pharmaceuticals or favorable toward pharmaceuticals

00:31:40.360 so much because of the great success of some of the drugs, especially the GLP-1 agonist related drugs.

00:31:46.440 In the same way as in the mid-90s, I heard tremendous defensiveness from the behavioral

00:31:53.880 psychology community and others about genetics because they weren't going to do genetics and

00:31:59.800 the Rudy Lybels of the world and the Claude Bouchards of the world were talking it up and it was going to

00:32:04.840 be big and they were starting to feel threatened and often didn't know much genetics but would try to

00:32:10.440 somehow minimize the role. And I think there's a lot of fear now among people who want community

00:32:16.680 intervention, public health intervention who say, I don't want our solution to obesity be

00:32:21.960 let everybody get obese or let two-thirds of the population get obese and then we'll give them

00:32:27.720 surgery or drugs for the rest of their life. It's too expensive, it's not my ethos, et cetera, et cetera.

00:32:34.360 And I think they're worried that people like me who say, look at the efficacy data, we need to think

00:32:41.800 about this more, are implying that we should shut down all the other stuff. And at least for me

00:32:47.080 personally, that answer is absolutely not true. But I think we need sub-paradigm shifts within

00:32:54.120 paradigms. And what I mean by that is I don't think the paradigm shift of saying, don't ever think about

00:33:00.440 nutrition anymore, only think about drugs and surgery is warranted. I don't think the paradigm

00:33:05.720 shift of saying, don't ever think about behavior, community intervention, family intervention is

00:33:11.320 warranted. Don't ever think about public health or policy, that's not warranted. I don't agree with

00:33:15.960 those at all. What I do think we need to do is to say within the paradigms of behavior, community,

00:33:22.200 family, policy, let's be honest. Let's look at Carolyn Summerbell's data and others and say,

00:33:30.280 there is no compelling evidence that any of this has had a meaningful impact. You can cherry pick

00:33:37.160 here and there. You can say this policy led to differences in how much of that food was purchased

00:33:43.480 in this context. Even if that's true, and sometimes those are a little shaky, those conclusions,

00:33:49.080 say, did it lower obesity rates? And those have never been shown.

00:33:54.360 How do we do that? I agree with you, by the way, and my own personal, because I think everybody has

00:33:59.000 to have a personal sort of bias if they're being honest. My personal bias is that so many of these

00:34:04.120 public health ideas on the surface just make a ton of sense. I can simultaneously hold true the

00:34:10.200 following truths, which is on the one hand, I can completely see why it was logical in the early to

00:34:17.480 mid nineties to say, we have to change the food environment. Richard Thaler's work, right? Another

00:34:22.840 Nobel laureate would suggest that that's the answer. You fix the environment, you make the default

00:34:29.080 environment better and people will opt into good choices. By the way, the default environment used

00:34:34.360 to allow people to eat in a way that was clearly ad libitum and obesity rates were not what they were.

00:34:41.240 So something about the environment 200 years ago or a hundred years ago, or even 50 years ago,

00:34:46.200 was significantly different from the environment today. It's not that our genes changed. Nobody

00:34:51.400 would argue there's been such a genetic drift that the reason that obesity rates are two thirds,

00:34:56.280 as opposed to 10% is due to a change in our species. An environmental trigger or a set of triggers

00:35:02.200 seems more likely and therefore public health solutions towards those seem very logical. So we can 0.94

00:35:09.720 hold that truth here. And then we have to be brutally honest with your assessment as well. The same as

00:35:15.560 Caroline's assessment, which is this has been an abject failure. I mean, if at the end of the day,

00:35:20.200 you're only measuring the outcome of interest, it hasn't changed. So we can say whatever we want, but

00:35:26.280 the outcome of interest hasn't changed. Either people smoke less or they smoke more or they smoke the

00:35:31.400 same. That's the only metric that matters if smoking cessation is what you're after. It's not, do we

00:35:36.600 collect more tax revenue? Are the commercials more or less favorable? Do people smoke less in restaurants

00:35:43.160 versus not in restaurants? No, we care if people as a society smoke less or smoke more. So given that,

00:35:50.040 how do we still say, and I'm not saying I disagree with this because again, my bias is there should

00:35:57.000 be solutions in public health, but how do we know after 30 years and billions of dollars with no effect,

00:36:07.160 that we should stay within the paradigm of public health solutions and just abandon all of the ones

00:36:14.840 we have when we don't really have a sense of why they failed? So we definitely don't want to only rely

00:36:20.200 on public health solutions. I would strongly oppose that. I agree with you that there is a superficial

00:36:26.520 sensibility to the public health arguments that were made for the various things tried, and it was reasonable to try

00:36:32.120 them. But I say superficial sensitivity or sensibleness because everything that's true makes

00:36:37.960 sense, as once we understand it. If we're wrong about something, then it didn't make sense. We just

00:36:43.080 didn't understand that it didn't make sense at the time. Some of that is assumptions, and it goes back

00:36:47.480 to that public health thing. I had a wonderful lunch with the most generous, interesting person,

00:36:52.840 Daniel Kahneman, who won the Nobel Prize.

00:36:55.080 Who recently passed away.

00:36:56.200 Right. Before he died. And he and his wife were gracious enough to allow myself and Michelle Cardell,

00:37:01.720 who now works at WW, was a former student with the group I led, to take them to lunch.

00:37:08.040 And we talk about obesity a little bit, and he's this great behavioral economist. And he says to me,

00:37:13.080 without artifice, he says, well, I think this nudge stuff is really good. So you could put things

00:37:18.520 on the menu and that would make people eat less. And I say, well, that's a good idea. And some things

00:37:24.120 like that are being tried and have been tried. And I said, but the big thing is compensation.

00:37:29.400 Yes, you can get a person to eat a little less in this context. But then if they go home for dinner,

00:37:35.240 and they just eat more at dinner, it goes away. And he looks at me without artifice. And he says,

00:37:41.160 hold it a second. So you're telling me that there might be mechanisms in people

00:37:46.040 that lead them to adjust for reduced calories. And I said, yeah. And this was a revelation. He said,

00:37:57.320 you've opened my eyes. And I was an economist. He didn't think about this. He's great with math,

00:38:02.520 and he's great with creative study designs. But this was again-

00:38:05.800 Yeah, he doesn't understand physiology, of course.

00:38:07.480 So I think that was a big part. A lot of things didn't make sense because they didn't take into

00:38:11.240 account compensation and many other factors. They didn't take into account magnitude of effect

00:38:16.760 and so forth. The second thing is the data themselves. People published a nice thing about

00:38:23.480 a meta-analysis of nudge type stuff in PNAS, Proceedings of the National Academy of Sciences,

00:38:28.760 a couple of years ago. Someone else just went in and redid it and said, if you adjust for

00:38:34.520 publication bias, it doesn't look like there's much holding up there. So often we're presented

00:38:40.040 with evidence. And we may want to come back to this when we talk about some other things,

00:38:42.920 like especially protein intake. We're presented with statements as though we confidently know these.

00:38:49.400 And yet when you really start to open the hood and peel things back, you say, hey,

00:38:54.280 there's not a lot of there there on the data. So the data that nudge works is actually shaky.

00:39:01.640 So that's the second problem. And the third is we seem to be unwilling to learn from the outcomes

00:39:06.360 of our studies. That is unwilling to say, we tried the school-based thing and it didn't get a big

00:39:12.440 effect. We tried it again. Fair enough. Let's try it a second time. Let's try it a third time.

00:39:17.320 At a certain point, we say enough. So if someone were to come to me, and I've been saying this for 20

00:39:22.200 years now, but I'll say it even more strongly today, if someone were to come to me and say,

00:39:26.040 we've got this opportunity to invest in these big school-based, community-based,

00:39:31.320 public health-oriented trials to reduce obesity levels in children or adults. And we have the

00:39:37.480 money available. We want to do good. Should we do it? And I would say, show me how this proposed idea

00:39:44.120 is radically different than what's been done for the last 30 years. And then let's talk. And if it's

00:39:50.440 not radically different, why are we wasting our time and money on that? So I think we really need

00:39:55.960 radically different public health paradigms. We need to stay in the public health paradigm,

00:40:00.280 but within the paradigm, we need a sub-paradigm shift to say, nutrition, education, modest physical

00:40:07.240 activity, build a little bit of a facility to allow people a little more activity. These have been tried.

00:40:13.400 They don't work. They don't have big meaningful effects. Let's try something completely different.

00:40:19.000 It's worth a try. That's what I think we need within the paradigms of public health policy,

00:40:24.360 and so on, radically different proposals. Now, if you were czar of the universe

00:40:29.160 and the ultimate resource allocator, what percentage of resources would you put into

00:40:34.840 a new and different form of public health, i.e. radically different approaches? And what percent

00:40:41.720 would you put into medical treatments for, such as surgery and drugs?

00:40:47.000 So first, I find it very entertaining to think about being the czar of anything since

00:40:52.520 my grandparents spent a lot of time successfully escaping the czars. It's interesting that what I

00:40:58.280 would say is probably a little more in the near term on the clinical treatment because I think we can make

00:41:05.320 more rapid gains in that while we need some slower, longer-term assessment of the others.

00:41:12.840 But also, I would amp up the non-pharmaceutical, non-clinical, non-surgery a little bit, the funding

00:41:18.920 from the government because I think a lot of that funding for those other things will come from

00:41:24.360 industry. So if you look at a budget of a Pfizer or a Lilly or a Novo Nordisk and what they put

00:41:30.280 towards certain areas, and then you look at what NIH can put to those areas, we're not talking about

00:41:35.960 NIH being this overwhelming big dog. And in fact, when you combine the pharmaceuticals on certain areas-

00:41:41.560 They presumably exceed, yeah. There's still something here that just philosophically doesn't

00:41:47.800 sit well with me. Not morally, so I want to be clear. I don't have a moral issue with the remarkable

00:41:56.520 success of the drug class that is now probably going to be the first thing that bends the arc

00:42:02.360 of this. I don't know when the next check-in will be, the next NHANES check-in.

00:42:05.800 Now it's annual.

00:42:06.840 Yeah. So it seems likely that very soon, if not already, we're going to see

00:42:11.160 for the first time in five decades, obesity rates going down. I hope we do.

00:42:15.640 But the reason that I'm still a little troubled is from a public health perspective, we don't have

00:42:22.360 the answer to the question, what was the, or what were the environmental triggers? I mean,

00:42:26.280 we think we know the answer, but every time we try an intervention against those things,

00:42:29.720 it doesn't work, which makes us call into question what the answer is. So clearly we did not

00:42:34.440 get obese because of a GLP-1 shortage that is now being ameliorated with GLP-1 drugs.

00:42:41.800 So clearly we have something that was causing the problem, again, multifaceted likely. And then you

00:42:47.320 have a totally different hack to work around the problem, which is why you're saying, I think,

00:42:53.720 what you're saying, which is we need to do both of these things. We still have to get back to this. But

00:42:57.640 if you had to speculate, what is it about the world in the early part of the 21st century that

00:43:05.160 makes obesity and by extension, type two diabetes, a problem that it wasn't, again, the year I was

00:43:11.320 born. It's a literally a log fold difference in type two diabetes, a log fold. That's hard to imagine

00:43:17.960 in 50 years.

00:43:19.080 I think there's multiple closely related factors. One is the food supply and its availability itself.

00:43:27.960 I think the second is kind of lagged intergenerational effects. Just for fun,

00:43:33.320 I'm going to try to rebut you on the genetics point, but only pedantically. I think that

00:43:39.640 we have seen genetic changes. Epigenetic changes or genetic changes?

00:43:44.440 Both. But certainly I'm going to put more of my direct knowledge and confidence

00:43:48.680 on the genetic as opposed to the epigenetic changes. And this is assortative mating,

00:43:53.320 differential mating. Do these fully account for the obesity epidemic? No, of course not.

00:43:59.400 Am I trying to say that they are the biggest influences? No, of course not. But I do think

00:44:04.680 it's important to push back and say, these are factors and they come in through migration,

00:44:09.720 through differential fertility, and through assortative mating. We've written papers about all

00:44:14.120 these and as have others. If you look in things like Framingham, you see that people in certain

00:44:18.520 BMI ranges have more children than people in other BMI ranges. And some will say, but obese people 0.70

00:44:24.680 have fertility problems. We're not asking about how good you are in theory at producing offspring.

00:44:30.280 We're asking how many offspring you produce. And so if richer, thinner people use more birth control

00:44:37.480 and have fewer offspring, and there's some genes for thinness, you're going to reduce their prevalence

00:44:43.320 and vice versa. So through migration, differential fertility, and then the other is assortative

00:44:49.000 mating, which doesn't change allele frequencies, but changes gene frequencies, which you get like

00:44:54.440 mates with like. But if you had to, again, all of those things make sense.

00:44:57.960 I just wanted to- They strike me as somewhat marginal though.

00:45:00.840 I had to be a professor for a minute and get the pedantic points out.

00:45:03.720 Fair. Okay.

00:45:04.440 All right. So now that's out. I think that it is largely, but not exclusively, the increased

00:45:11.960 availability of a greater variety of foods, of highly palatable foods, of foods that are relatively

00:45:20.440 modest in cost, foods that are easy to acquire, the control of ambient temperature, which makes it

00:45:27.160 easier to overeat foods. You don't want to overeat a lot if there's no air conditioning and you live

00:45:31.880 in Austin, Texas, and it's 110 degrees out. But if there's air conditioning, the buffet's okay.

00:45:37.560 And then I think there's some intergenerational lag effects that we, or at least I, don't fully

00:45:42.920 understand. If you look at the Danish data, Torkel Sorensen and others have written about this.

00:45:48.840 They, for over a hundred years, conscripted, if that's the right word, every 18-year-old healthy

00:45:56.680 male into the Danish army. And they have not only heights and weights of each one naked, kind of

00:46:04.040 weirdly, they have photographs of each of them naked. And what you see in these BMI levels is you'll 0.82

00:46:10.280 see a period where it's flat for a little bit, approximately, then you'll see a steep acceleration

00:46:15.800 or steep increase, and then it'll flatten out again a little bit, and then you'll see a steep slope.

00:46:20.680 This has happened three or more cycles, I think. I don't think anybody exactly understands why.

00:46:26.440 Diana Thomas's mathematical model, she's a professor at West Point, studies obesity. Her 1.00

00:46:31.880 mathematical models predict some of that. I don't fully understand how that works, but we might ask

00:46:38.040 her. It does suggest to me, even culturally or behaviorally, there could be some lags,

00:46:43.400 whereby the weight of your parents or grandparents is affecting you.

00:46:48.360 Socially or genetically? Both. So the oocyte

00:46:52.120 that formed you was formed in your grandmother. So potentially through epigenetic things you've

00:46:57.400 mentioned or others, that could be affecting you. Then there's the cultural part. I think about it,

00:47:04.840 when I was a kid and we went out to dinner with my dad, and we weren't poor, but we weren't rich,

00:47:09.320 we were decidedly lower middle class, creeping up. If we went out at the local Italian restaurant or

00:47:15.800 something, order shrimp, you had to ask dad about that. The chicken parm you could order

00:47:20.600 without asking. Shrimp you had to ask because shrimp was expensive. You can get shrimp by the bucket now

00:47:26.680 at the local buffet for next to nothing. So I'm prepared to eat a lot more shrimp than my dad ever

00:47:34.360 would have thought of ordering or sitting down because of our changing economic times and so on.

00:47:41.320 Now, my kids think nothing about ordering dinner in from DoorDash every night, where I still think,

00:47:48.920 even though I could afford to do it, as well as my kids could because they're spending my money some

00:47:53.080 of the time, which is great. I'm glad they're doing it. But I think, oh, that just seems excessive to me.

00:47:57.480 You know, it seems too indulgent. So I think there may be sort of levels at which one ratchets, culturally,

00:48:04.040 as well as physiologically or anatomically. So I think all of these things can be in play.

00:48:10.360 I also think we need to change some of the attitudes. This is speculative on my part.

00:48:14.360 I have no proof that this is true. But I think one of the bad things that the nutrition field has done,

00:48:20.600 including very much the public health community, which talks about, I used earlier,

00:48:25.800 I said the healthy foods that have magical effects. But I also think the low carb advocates

00:48:30.680 and zealots who came up through the late nineties and still exist at present and have very powerful

00:48:36.360 voices and yet others still, I think there's the sense that there's a right way to eat.

00:48:42.600 Nobody agrees on what the right way is, but there is an underlying supposition that there is a right

00:48:48.840 way to eat. And if you just ate that right way, then you would maintain the weight you want to

00:48:55.080 maintain and the fat level you want to maintain without ever feeling lack of satiety or dissatisfaction

00:49:04.440 or what have you. You and I were talking about our personal diets.

00:49:07.560 What's interesting about that is that's actually philosophically not that different from a drug

00:49:12.360 approach. In other words, if you constructed a lot of parallel universes, it's certainly possible

00:49:19.160 that if you put everybody on a perfectly adherent version of diet X, Y, and Z on each of those

00:49:26.120 parallel planets, you would eradicate obesity. And by the way, one of those planets, you might say,

00:49:31.800 well, we're also going to put everybody on terzepatide. So you now have multiple different

00:49:37.960 dietary treatments when perfectly adhered to that will dramatically improve obesity. One of those will

00:49:45.160 be just a drug. Maybe two of them will be a drug. Another one will be a gastric bypass, etc.

00:49:49.560 It still doesn't answer the question, what triggered the problem, right? It still doesn't

00:49:55.240 answer the question. I don't know that we want to spend too much more time on that because these are

00:49:58.760 unanswerable questions. What is the right diet to fix it doesn't mean that the absence of that diet

00:50:05.400 is what caused it. I agree. The point I was trying to make is that by saying to people,

00:50:11.000 there is a right way to eat. We may foster a delusion. That is, the real debate perhaps is not

00:50:17.720 between the low-carb guy and the non-low-carb guy as to what this thing is or the eat locally or

00:50:24.200 whatever. The real debate may be, is there a right way to eat compositionally or behaviorally or time

00:50:31.240 of day or something that will satisfy you, not make you feel deprived in the real world we live in,

00:50:37.720 not a parallel universe we could construct? And the answer may be no. And yet by continuing to sell

00:50:44.200 that idea, we may continue to have people searching in the wrong spot. Instead of searching for,

00:50:50.520 how do I control or overcome my incomplete satisfaction with eating only this amount?

00:50:58.520 And instead, they're looking for, what's the way to eat that I don't have that dissatisfaction?

00:51:04.120 And I think what we may have to accept at some point is that for most of us, there are exceptions,

00:51:10.760 but for most of us to maintain a truly thin or lean body composition, if that's what we want,

00:51:17.800 and I'm not saying everybody should want it, but for those who do want it, that we may have to accept

00:51:22.280 that either we're going to have to alter our desires in part through pharmaceuticals,

00:51:28.520 or we're going to have to accept that we don't get to meet all our desires at times,

00:51:33.720 as opposed to continue what may be the charade, that there is a way that you can just eat a certain

00:51:41.560 kind of food or certain type of diet or eat in a certain way that will lead you not to ever feel

00:51:47.800 dissatisfied. So I think that's an important stoic approach, right? A little more stoicism.

00:51:53.960 Let's go back to something you were asking though about evidence earlier. And I do want to make a

00:52:00.280 point about this, that we also need to increase the quality of the evidence and the standards we

00:52:08.040 hold. We spend too much of our research budget on lousy evidence. So in the childhood obesity field,

00:52:15.400 my group, for example, often will write letters to the editor. Another paper was retracted last week,

00:52:21.640 because we found statistical errors in it. And if you think about it, this was a randomized control

00:52:26.280 trial of a treatment for obesity-related or nutrition.

00:52:29.240 Behavioral?

00:52:29.800 I think so. We find it in diet, behavior, et cetera. Not usually drugs. And what we see is,

00:52:36.040 if you think about a randomized control trial, at the low end, a randomized control trial is usually

00:52:42.120 over $100,000 to conduct. At the high end, it's tens of millions. The ones we're looking at that

00:52:48.280 often we find these mistakes in, and many cases are retracted, especially in childhood obesity,

00:52:54.280 are probably in the multiple hundreds of thousands of dollars, occasionally millions. And then you

00:52:58.840 think that's all wasted if they misanalyzed and misreported the data and got the wrong answer.

00:53:05.000 So we kind of feel like we're rescuing those dollars in some sense by getting the wrong answers

00:53:09.240 out and the right answers in. So we think it's an important service. But I think we need to hold

00:53:14.200 our field's feet to the fire much more strongly on doing research that answers new questions,

00:53:20.200 that answers questions well, that honestly reports the data.

00:53:23.560 Do you think that that problem, which I'm quite aware of, of course,

00:53:28.280 is disproportionately present in this field? Or do you feel that it's both acknowledged and

00:53:36.680 demonstrated at the same frequency in all fields of medicine? Do you feel that we have a brighter

00:53:42.200 spotlight on it here, thanks to certain individuals? How do you think this stacks up?

00:53:46.360 All of the above. What we know is that there are many anecdotal statements by leading thinkers,

00:53:54.680 like Stuart Ritchie, as just one example, or Gary Taubes, our mutual friend, who say nutrition is

00:54:00.920 singularly bad. And there are some of these all wonderfully colorful statements. Johnny Aniti said,

00:54:06.440 we need to accept that nutrition epidemiology is a dead science and bury the corpse. That's a quotation.

00:54:12.280 Those are opinions. Those are not bits of data. If we go further and we look at the Pew Charitable

00:54:17.240 Trusts- But hang on, that's, I mean, one could agree with that on some pretty objective facts.

00:54:22.760 But the question I'm asking is more on the challenges of experimental research that you're

00:54:28.040 talking about, where real dollars are being thrown at experiments that are being done incorrectly or

00:54:34.520 being analyzed incorrectly, or where the questions that are being asked are incremental, useless,

00:54:39.880 uninteresting, and unlikely to add meaningfully to the fund of knowledge. Like, let's just forget

00:54:45.240 about nutritional epidemiology, but I want to talk about this other, what seems to be more distressing

00:54:50.440 problem based on both the dollars that go into it, but also I think the confusion that it sows and

00:54:57.160 the noise that it creates.

00:54:58.280 Right. So it's clearly created that confusion, noise, and that's what the Pew Charitable Trusts have

00:55:03.080 shown that in surveys, now we're talking data, in surveys of representative samples of American

00:55:09.320 population, people trust nutrition experts, clinicians, purveyors of knowledge, more than

00:55:16.840 they trust nutrition scientists. And they trust nutrition science less than they trust other forms

00:55:22.600 of science. So that's a fact. We do have a trust problem in nutrition science. Now let's go to the

00:55:30.520 last stage, which is, is our research really better or worse? Harder to pin that one down. There's not

00:55:36.520 enough concrete, strong comparisons to other areas. We're trying to start some in our group, but there was

00:55:43.000 a recent paper that came out in economics. It's not a one-to-one comparison, but in an economic journal

00:55:48.840 looking at reproducibility. Reproducibility and replicability are not quite the same. Reproducibility

00:55:55.000 is, can I get your original data, run exactly the same analysis you said you ran, and get exactly the

00:56:02.840 same result? If I can, I've reproduced your research. It doesn't mean your result was right. Maybe you ran

00:56:07.480 the wrong analysis, but at least I could do what you said you did. We do that in nutrition and obesity,

00:56:12.840 and we find we don't have exact numbers. It's not a random sample of papers, but we find what seems

00:56:19.080 to be a not infrequent errors, irreproducibility, or what we call verification problem. Meaning we

00:56:27.160 could reproduce your result, but it was wrong. It was wrong because you ran the wrong analysis. We

00:56:30.920 write the right analysis, get a different conclusion. And you might do how many of these a year?

00:56:35.800 More than a dozen. Of the dozen you run a year, how many turn out to be not reproducible or not

00:56:41.880 verifiable? I would say probably, again, these are all approximations,

00:56:46.840 maybe half, but keep in mind we're not randomly sampling. Understood. What's the criteria upon

00:56:51.560 which you select besides size of study? It's usually one of two things. It's

00:56:57.160 interestingness or it's something doesn't look quite right. So if it's something doesn't look

00:57:02.360 quite right- Your pretest probability is higher.

00:57:04.680 Then we take a closer look at it more often. Or if it's just very interesting, we say,

00:57:09.480 that's really interesting. And it was published in Nature and that could be paradigm changing.

00:57:14.040 I'd be curious when you have enough data to know if you take out the, that looks fishy

00:57:19.400 sample and just said, hey, when we looked at the, this is interesting, if half of those are coming up

00:57:24.360 unverifiable, that's a crisis. Yeah. I would say we should do it. I hope

00:57:29.800 there's a funder out there listening who will want to fund it. NIH, as you might imagine, may not be always

00:57:34.440 so keen on having us answer this question and it's hard to get that through. But yeah,

00:57:39.560 we'd like to do that and hope we can do some more. We're doing little spot checks in the area.

00:57:45.640 My sense is even within obesity, if you look at pharmaceutically done randomized controlled trials,

00:57:52.040 and I'm not trying to say that people at pharmaceutical companies or pharmaceutical companies

00:57:55.720 somehow morally superior or not. They're people. They're just responding to their environment as well.

00:58:00.520 But their environment is a very strong regulatory authority called FDA that holds their feet to the

00:58:05.560 fire and so on. Do you think that's the reason that drug studies tend to be very rarely found

00:58:13.640 to require retraction? In modern times, yes. In modern times, and if somebody said to me,

00:58:19.880 do you trust randomized controlled trials coming out of the pharmaceutical industry

00:58:24.840 more or less than academia? Infinitely more from the pharmaceutical industry.

00:58:29.080 Yeah. Think about that for a moment. Let's just reflect. That's a big statement.

00:58:32.600 I agree with you, but I think it's not intuitive to the average person listening to us. Many people

00:58:36.920 listening to us would say, what? The data coming out of Pfizer are more trustworthy than the data

00:58:44.280 coming out of Harvard? But the point here, the key point is that Pfizer has to answer to somebody,

00:58:52.440 the FDA, who will bring down a much greater and swifter punishment if issues are discovered in

00:59:00.200 methodology, statistical analysis, reporting, et cetera. Whereas the academic community doesn't

00:59:05.640 have that degree of policing basically. And the funding. Often people say,

00:59:11.960 oh, the industry is so much more efficient than academia because they have the profit motive. I think it

00:59:16.840 depends what you put in your denominator of efficiency. If you say output per unit time,

00:59:23.640 no question, industry in general and pharma in particular, blow academia away. But if you say

00:59:29.400 output per unit dollar, academia probably blows industry away because we know how to stretch every

00:59:34.840 penny. Right. You guys are working on a shoestring budget.

00:59:38.040 Exactly. But that means often not much rigor. Whereas the big pharma company who's going to

00:59:44.280 put their registration trial in is checking and double checking and having professionals check

00:59:49.080 and so on. Now, there may be more, I hate to use the word bias because it's not clear.

00:59:54.120 By the way, it might be worth also explaining to folks that when people talk about,

00:59:57.880 quote unquote, a Pfizer study or a J&J study, they're hiring CROs to actually do the study. I think

01:00:05.080 sometimes people are under the impression that when Lilly is doing a study on a drug,

01:00:11.000 it's like the whole Lilly team doing the experiment as opposed to Lilly providing the agent,

01:00:17.000 helping think about the experiment, but basically having a clinical research organization actually

01:00:22.280 do it and having independent folks do the analysis. That's right. And I think that's important.

01:00:28.120 Now that doesn't mean that there's no, and again, I was about to use the word bias. I don't have a

01:00:32.600 better word right now to use, but I use that one hesitantly, but there may be more bias in some ways

01:00:39.640 in the industry funded work. And that's often in the question asked.

01:00:43.080 I was just about to say, it's how the question is asked, which determines how the study is designed

01:00:48.120 to look for a particular answer for sure. Exactly. So an industry group might say,

01:00:53.080 I'll compare my new drug to the worst old drug in class.

01:00:57.880 Right. If a university guy did it, he or she might say, no, I'm going to compare it to the best drug

01:01:03.400 out there. Exactly. But once they've decided on the question, then the design, execution,

01:01:09.800 and reporting of the study seems to be enormously more rigorous in pharma. Now that's not true if you

01:01:15.960 said, what about dietary supplement industry? Different game. We've got a more complex answer there.

01:01:20.520 So back to this, I think we need better data. I think we need to assess this. I think within

01:01:27.560 the non-industry funded stuff, typically, like the public health, the school-based stuff,

01:01:33.880 the child obesity trials, it's going to vary a lot. So the cluster randomized community school-based

01:01:40.600 childhood obesity trials tend to be quite poor. And I think the non-verifiability rate is very high.

01:01:48.360 Whereas if you went to certain other kinds of trials, the NIH-funded clinical management of

01:01:53.880 obesity trials will tend to be better. So it's going to vary a lot. And just hopefully I'm not

01:01:59.320 biased, but I'm sure someone will think I am and that's okay. They're entitled to their opinions on

01:02:03.160 this. I'll disclose that I have funding from all these groups. So I've got most of my funding is

01:02:08.280 government and NIH, but I have funding and the school I lead is funding from industry, including

01:02:14.120 many of the pharmaceutical companies to think about clinical trials design and biostatistics.

01:02:19.480 We're funding from food industry at times, commodity groups. So I just want to disclose all that.

01:02:24.280 Let's pivot now and kind of talk about the current state of obesity, which is really seeing a success

01:02:30.280 it's never seen. And it's been a relatively short period of time. I think three years ago,

01:02:36.680 very few people knew what semaglutide was or even Ozempic, which is the trade name given to the

01:02:44.760 diabetes version of that drug. Whereas today, I can't imagine too many people haven't heard

01:02:51.800 the words Ozempic or some of its derivatives. I think Ozempic might be one of the most recognized

01:02:57.560 of these drugs. It's pretty remarkable. It's also worth noting that these are not new drugs,

01:03:06.760 semaglutide and trisepatide are newer drugs, but they've been around for a while. At least

01:03:12.520 semaglutide has and liraglutide and others have been around for at least a decade and they've

01:03:18.840 successfully treated people with type two diabetes. And like all things, or it's often the case,

01:03:25.160 you sort of notice something in treating one subset of patients that gives you an insight into

01:03:30.120 treating another. And so basically as people with type two diabetes were treated with this class of

01:03:36.920 drug, you notice that it wasn't just improving their diabetes. They were also losing weight.

01:03:42.280 And that led to what became a set of dedicated experiments to test the efficacy of these drugs

01:03:47.720 in non-diabetic obese patients. And the rest is history. Talk a little bit about

01:03:53.480 what you think is socially and psychologically happening at the moment. Why? Why are people

01:03:59.640 so interested in this drug? It's fascinating. I think people are interested for the obvious reason.

01:04:05.240 The obvious reason is lots of people want to lose weight and lots of people want to help other

01:04:10.120 people lose weight. And for the first time in history, as you've noted, we have drugs that are now

01:04:18.040 powerfully effective and appear to be reasonably safe. We've had drugs that were powerfully effective

01:04:24.360 before, but would kill you. And we've had drugs that were reasonably safe before, but at best,

01:04:30.120 modestly efficacious. We now have ones that are powerfully effective and appear safe, reasonably safe.

01:04:36.840 Safety is a social judgment, not a factual determination. Risk is a factual determination.

01:04:41.880 Safety is a social judgment. And so it invites all kinds of interesting speculations about cause.

01:04:49.080 What is the role of GLP-1 in causing obesity? And is there a role? Just because things involving GLP-1

01:04:55.960 treat it doesn't mean it's involving the cause. What's the effect on stigma? If we can treat it,

01:05:01.960 does that reduce stigma in the same way that Viagra changed many things around erectile dysfunction?

01:05:11.240 And interestingly, I sort of didn't predict the full cultural impact of that, which shows you it's

01:05:16.840 hard to predict these things. People didn't predict what Viagra was for. It was being used for something

01:05:20.840 else. They noticed erections as a side effect, and then they started working on it. And in the early

01:05:25.560 90s when I went and visited one of my buddies who's a biostatistician at Pfizer, and that individual

01:05:32.200 told me they were working on this new thing and explained what it was to me. I laughed at it and

01:05:37.000 I said, why are you wasting your time on something so ridiculous and unimportant? Why don't you do some 0.98

01:05:41.880 important research? Shows you what I know. So I think here we're learning that, again, that we get

01:05:47.720 surprised in science. We're seeing a moral panic. This is subjective on my part, but this is something

01:05:54.200 I'm noticing. A lot of old arguments that had kind of gone semi-dormant, at least in the academic

01:06:00.440 community over the years of, well, if you give people a drug for obesity, it doesn't teach them

01:06:06.360 anything. And therefore, when you stop the drug, the weight just comes back. And this was said as a

01:06:11.960 criticism, as opposed to saying, well, who said it had to teach them anything? Who said that was the goal?

01:06:18.200 And for many drugs, anti-seizure medications, if you have seizures, anti-hypertensives,

01:06:24.680 anti-diabetes drugs, et cetera, you're going to take those for the rest of your life if you're in the

01:06:29.560 right class for that. We don't say, but the person with schizophrenia shouldn't get the drug because

01:06:36.120 if we stop giving it to them, the schizophrenia symptoms come back. Say, no, schizophrenia is a

01:06:41.000 serious disease. We need to give it to them. With obesity, this has come up again. It sort of seemed to

01:06:46.920 be put down a few, that idea a few years back. And now I'm seeing, I'm hearing it again, this kind

01:06:53.080 of moralistic judgment about that. We're also hearing the moralistic judgments come about

01:06:57.880 motivation. It's okay if you're motivated for health. It's not okay to get the drug if you're

01:07:04.360 motivated for something other than health, which implies that assuming we have the same health issues,

01:07:09.560 assuming the person would equally benefit from their health, we make a moral judgment about

01:07:14.600 your motivation. But there is no evidence that I know of that people who are motivated for health

01:07:20.520 to lose weight do better than people who are motivated for cosmetic or any ego, business,

01:07:26.920 any other reasons. So I think we need to get over some of that moral panic.

01:07:31.720 Once we get past the safety, the cost, and the availability issues, and I don't want to trivialize

01:07:35.800 those. The safety, the cost, and the availability issues are big issues. The safety issue is really,

01:07:41.240 and in that sense, I'm defining safety in the sense that sometimes the FDA defines it,

01:07:47.160 which is safety involves risk and risk involves uncertainty as opposed to being risk involving

01:07:53.080 known factor. I don't just mean the probability that you get this. I mean the fact that we don't

01:07:58.120 know what happens if you take it for 40 years. So there is some safety issue, some open questions. No

01:08:03.800 one's taking it for 40 years, so we don't know what happens if you take it for 40 years.

01:08:06.680 Right now it's very expensive. Our country is divided on how healthcare should be paid for.

01:08:12.360 There's a lot of different opinions. And also there's an availability problem. But let's just

01:08:17.160 fast forward to a time when we say we've learned the safety.

01:08:20.920 By the way, say a little bit more about the availability problem. I mean, I only realize it

01:08:25.720 because you see compounding pharmacies now making semaglutide and terzepatide, which when I first saw

01:08:33.080 that I couldn't understand how they were doing that legally because that's pretty clearly not

01:08:38.120 within the statute of what a compounding pharmacy can do. A compounding pharmacy can't make an

01:08:42.360 existing FDA approved drug. They have to make a variation of that drug. For example, they have to

01:08:48.520 change the delivery mechanism if they make something topical that would only be available orally or

01:08:53.800 something of that nature. Unless, and one of the exceptions to the rule is, if the FDA approved drug

01:08:59.960 can't be produced in sufficient quantities, then a compounding pharmacy can create the exact same

01:09:07.160 drug that is available through the FDA label. Presumably that is happening. Do we have a sense

01:09:12.920 of why it's happening? What is the manufacturing bottleneck? Obviously demand is outstripping supply,

01:09:18.680 but the question is why is supply not able to meet demand? And then secondly, do you have any insight

01:09:25.000 into whether the quality control at the compounding pharmacy level matches that of lily or novanortis?

01:09:33.560 So with respect to the first part, why is there an availability problem? I don't know the technical

01:09:39.000 mechanics of it, but my understanding is that the technical process by which these drugs are produced

01:09:45.720 is different than some other drugs. And the technical process is a slow one. And so until they ramp up

01:09:52.520 more and more production sites, they just can't do it fast enough, but they are ramping up more and more

01:09:59.640 production sites. That's good. Novo just bought Catalan, which happens to have a plant in my backyard

01:10:07.560 in Bloomington, Indiana. So we'll probably see more of that ramping up. The second thing is about the

01:10:13.480 compounding pharmacies. So when I first heard about it, I'm far from an expert in compounding pharmacies or

01:10:18.920 the legal aspects. But I too was skeptical, is this okay? Was the quality control? And is this a kind

01:10:27.000 of shady thing? And I started to hear a lot of reports about this described as though it was a

01:10:32.520 very shady endeavor. And again, that moralizing came in again. Then I've talked to some other

01:10:38.360 people who are experts in it and who are using these. And I've said, admittedly, again, this is

01:10:43.320 their business. So they have a motivation, but they have said, well, when we do it and they've

01:10:48.040 described, and I said, tell me your process, who do you use? How do you do it? What quality control?

01:10:53.000 And then they've gone through, say, for this compounding pharmacy that I use, we use it in

01:10:58.200 this way, this degree of quality control. And I say, wow, that sounds to me, I have not physically

01:11:04.520 inspected the plants. I'm not an expert in it, but it sounds to me like some very rigorous quality

01:11:09.160 control. So I don't think we should be dismissive of the concerns around compounding pharmacies,

01:11:16.040 but I also don't think we want to paint everybody with the same brush. The question becomes, as with

01:11:21.240 anything, is show me your data, show me your evidence on your quality control, your procedures.

01:11:27.000 And if they're good, they're good and let's use them. Let's get over the moral panic.

01:11:30.440 I don't know that I think of that as a moral panic. I think the bigger moral panic is less about

01:11:35.160 the source of the drug, but the use of the drug. And so you brought up an interesting

01:11:38.920 distinction, which is let's take an individual who is medically obese and by the way, metabolically

01:11:45.560 unhealthy. So that's the key point I want to get out here. So this is a person whose health is

01:11:50.280 compromised by their weight, both from an orthopedic perspective and metabolically. And then let's take

01:11:55.800 another individual who's overweight, but if you're looking at them objectively, you don't see the

01:12:01.320 metabolic signs of overweight. They're not suffering physical and orthopedic issues associated with it.

01:12:08.680 So both of these people, let's just assume, have a desire to lose weight. One of them

01:12:12.360 to primarily ameliorate the medical conditions and also the aesthetic conditions. And then the

01:12:19.400 latter person just for the aesthetic conversions, right? Okay. We probably look at those people

01:12:25.000 differently. When I say we, I mean society might make a different moral judgment on those two.

01:12:28.680 That's right.

01:12:29.320 You're arguing that's a false dichotomy.

01:12:32.600 It's a legitimate dichotomy to see the situations as distinct situations,

01:12:39.000 but not necessarily implying distinct recommendations coming from those.

01:12:45.560 Let's refine it to a two by two. We've got people, let's just say four individuals come to you and

01:12:52.360 we're going to say that you're the objective all knowing agent.

01:12:55.320 Meaning I determine who goes on the drug?

01:12:58.040 No, you determine their state of being.

01:13:00.760 Got it.

01:13:01.560 Half of the people are objectively at medical physical risk because of obesity and would

01:13:09.800 be objectively medically helped by losing weight on this drug. Half of the people are not at

01:13:16.120 objectively medical increased risk and would not be predicted to have a major medical benefit.

01:13:22.120 Within each of those groups, half of them think they have a medical problem,

01:13:28.280 regardless of whether you objectively determine they do and think they would benefit. And half of

01:13:33.240 them aren't interested in that. They want to do it for cosmetics, income, other opportunities, 1.00

01:13:40.600 et cetera, stigma reduction, quality of life. The question is how should those four groups be treated?

01:13:46.440 Now, it seems to me from an obvious point of view, if we're concerned about expense and the expense is

01:13:54.440 borne by society, not the individual coming, or if there's shortages and we're going to take it away

01:13:59.960 from someone who's genuinely medically needed, then going to the non-medically needy people is

01:14:04.920 questionable. But if we get over those problems, if the person says, I can afford to pay it for it

01:14:10.120 myself, and the availability is there, and we think there's no big safety problem, or even if there's

01:14:16.600 some safety problem, but we've told them, fully consented, take the libertarian view, it's their

01:14:23.640 choice, it seems to me. It's hard to imagine any reasonable person could argue with that position.

01:14:28.280 Well, one of the big statements that got in some news was a very reputable entity, major player in

01:14:35.800 mainstream medicine who has an interest in actually promoting this. A three-step statement was made.

01:14:42.760 Step one is, the drugs were intended and designed and studied for this use, meaning treatment of

01:14:49.400 medically needy people. Second, the drugs were approved for that use. Third, therefore they

01:14:56.520 should only be used for that. And the third part is a moral judgment, not a factual judgment.

01:15:02.040 The first and the second are true. And what they really tell you is, therefore the cost-benefit

01:15:08.440 analysis has to be viewed through the lens of that patient population. In other words,

01:15:15.320 when you ask the question about risk and benefit, you have to at least acknowledge that the long-term

01:15:23.320 risk, long-term benefit are studied in that population. Correct.

01:15:27.560 And as such, this is what the data are. These are the risks, these are the benefits,

01:15:33.480 make your judgment. Conversely, if you ask the question, hey, for a person who is subjectively

01:15:41.800 10 pounds overweight, like me, you could argue I'm 10 pounds overweight, nobody knows but me basically,

01:15:48.280 but hey, should I be taking this drug? So let's take an analogy. Patient comes to you,

01:15:54.920 they're very wealthy. They're in good physical health. They have a house. They have a car. They

01:16:01.640 have all the material things they need. They have a family. Family loves them. They don't engage in

01:16:06.840 violence. And they'd say, I feel miserable. I'm anxious all the time, or I'm depressed all the time.

01:16:13.960 You might try a few things, explore it, but assume you've explored it, it's real. Maybe you

01:16:18.200 tried some cognitive behavioral therapy, didn't seem to work. You might say, yeah, an anti-anxiety

01:16:25.160 drug or an antidepressant might be for you. FDA approves those things. We take the person's

01:16:30.760 quality of life and their feelings into account. Why is it that the person who says, I feel too fat,

01:16:40.040 and I want to be 10 pounds thinner and look good in my bathing suit, or I want to get this job as the

01:16:46.440 leading actor in that film, or I want a promotion in my environment and I think I'm more likely to

01:16:52.120 get it if I'm thinner, or I'm hungry all the time and I don't plan to lose weight. I just want to stop

01:16:58.760 being hungry all the time. Why are that person's feelings or non-medical desires any less valid than

01:17:06.280 the person with depression? Or for that matter, the person with an unusual, but not health damaging

01:17:14.520 physical feature, you know, an unusual nose or something who says, I just feel like I'd be judged

01:17:19.720 better. I don't think it is. I guess the only thing I would suggest as the backstop to that is when

01:17:27.720 the person who doesn't like their nose goes to the ENT surgeon or the plastic surgeon to have the

01:17:34.920 completely non-essential but emotionally beneficial procedure, if they're seeing a good surgeon,

01:17:43.480 the surgeon can tell them with unambiguous clarity what the probability of negative outcomes is.

01:17:51.720 And I think the same is true in the case you described at the outset about the individual

01:17:56.920 with depression or anxiety. A very good physician can explain to them what the risks are,

01:18:04.600 and by the way, as you know well, very few physicians would give you a medication for

01:18:12.280 anxiety or depression without also prescribing in parallel to it psychotherapy. The data are pretty

01:18:19.400 clear that medication by itself is nowhere near as effective as medication coupled with psychotherapy.

01:18:25.320 So you have two things going for you that make this analogy not apples to apples, which is in the case

01:18:31.000 of depression, we can say much more about the long-term side effects and we're combining it with a

01:18:37.720 behavioral therapy that aims to improve the efficacy. Again, I'm not suggesting that the person who wants to

01:18:44.920 lose 10 pounds doesn't have a legitimate concern. I think my concern is we don't know enough about the

01:18:53.800 long-term risk to tell them for their relatively minor health compromise, is it potentially worth it?

01:19:03.320 Is the trade-off worth it? I think we could probably say that with a higher degree of certainty for the

01:19:08.760 individual with significant obesity. Because even if we would have kind of a small bracket of understanding

01:19:15.160 the downside potentially of the drug, we really know the downside of having a BMI of 40. Being insulin

01:19:21.960 resistant, having type 2 diabetes, having a BMI of 40 has such a clear downside that the other side

01:19:28.920 of that bet is a pretty easy one to take. So I think that to me, so again, for me it's not a moral

01:19:34.120 question at all when I'm confronted with this question, which I am all the time. Every week I

01:19:39.560 probably, or every two weeks at least, interact with a patient who fits the exact description you're

01:19:44.360 talking about, which is, I'd love for this to be easier. And again, I don't think there's anything

01:19:49.800 wrong with wanting something to be easier. But my hope is we get to a point where we could give

01:19:54.920 them the same degree of clarity around risk that the plastic surgeon can give the patient who wants

01:20:02.360 to undergo a rhinoplasty. Right. And I agree with you on that. And I think the moral questions come in

01:20:07.720 around how do you conceive of the role of FDA, society, physicians in regulating choices.

01:20:16.360 And by the way, to be clear, that's why I'm not taking one of these drugs. I'd love to be 10 pounds

01:20:20.760 lighter. I would love to be 10 pounds lighter. I would love to never be hungry. All of the things

01:20:26.600 that these drugs do, by the way, they improve glycemic control. All of those things are appealing

01:20:30.840 to me. But the truth of it is for somebody who is quite a risk taker, and I am quite a risk taker.

01:20:38.120 You are.

01:20:38.680 I am. When it comes to my health, I would argue I'm quite a risk taker.

01:20:43.160 But I've watched countless patients take these drugs. And as I've shared with you and others,

01:20:49.000 without exception, the resting heart rate overnight goes up about 10 beats per minute.

01:20:53.800 And I don't know what it is about that fact and the fact that heart rate variability goes down

01:20:58.200 slightly that just has me asking the question, for me personally, is it worth a trade-off?

01:21:04.280 Is there some underlying sympathetic, parasympathetic imbalance that results from this drug

01:21:09.880 that is doing a whole bunch of other good things vis-a-vis my appetite, potentially?

01:21:14.360 But you know what? Over the arc of my life, is it worth it? Maybe if it were 40 pounds and it was

01:21:20.280 medically a problem, I'd say, oh, I'll take the heart rate bump any day of the week.

01:21:24.680 So informationally, I'm with you 100%. And in terms of the morality of the honest communication,

01:21:30.280 I'm with you 100%. Well, by that I mean, informationally, we have a fair bit of data that

01:21:36.040 allowed FDA to make its decisions on the use of these drugs for particular indications in patients

01:21:41.800 who are judged to be, quote unquote, medically needy of those drugs. And we don't have a lot of

01:21:47.640 data on the person who's thin, but who says, I just want it to be easier. Or the person who's thin,

01:21:53.080 but says, I'd like to be 10 pounds thinner. And I think any treatment or provision of something to

01:21:58.360 people without a full disclosure of what you know, and an honest disclosure, is not right.

01:22:03.000 So I think if I were in your shoes, I'm not a physician, I don't prescribe drugs,

01:22:06.920 but if I were in your shoes, and that person came to me, my bare minimum is that I've got to say to

01:22:11.320 them, I want you to be aware that I have no data on this over many decades. We only have a few years.

01:22:18.840 I want you to be aware that it was only tested thoroughly in these populations, which is not your

01:22:24.760 population. And you need to know that there are, as Rumsfeld famously said, the unknown unknowns.

01:22:30.120 Then I think there's an issue of choice. There are lots of things that I think it's

01:22:36.040 acceptable that our society permits, but I don't personally want to do them. Think freedom of

01:22:41.720 speech. I think it's perfectly acceptable and necessary that we allow certain people to come

01:22:47.000 out publicly and make certain statements. But I'm not sure I want to make all those statements.

01:22:51.320 And I can imagine you saying, I think it may be acceptable that somebody provides this drug to

01:22:56.760 this person under these circumstances, but that's not what I want my career or life to be. And I

01:23:01.400 think you should have that choice. So I think these are things we ought to do. And it comes down very

01:23:06.200 much, I think, to this sense of after we have the inputs, we can agree on the facts, or we should be

01:23:12.360 able to agree on the facts. Then what we do with those facts, we can disagree because we have different

01:23:17.160 values. And I think that's where it's how much of a paternalist is one. The FDA is very paternalistic.

01:23:23.560 They're going to decide which drugs are good for whom, or how much are you a libertarian where you

01:23:30.120 say, we'll tell you about the effects to the extent we can of this drug or this treatment,

01:23:35.960 but how good it is, whether you should do it, whether you want to do it implies values. And you

01:23:42.360 make that decision as long as it's a fully informed decision. And those are different views of how we

01:23:47.800 should proceed. All right. Let's consider one more zinger on this topic. You are now in charge

01:23:54.280 of both WADA and USADA. So world anti-doping and US anti-doping agencies.

01:23:59.960 Thank you for defining those for me. You have an obvious and clear hard line

01:24:04.600 against drugs that improve performance. An athlete cannot take testosterone or growth hormone

01:24:10.600 or EPO or anything that boosts performance. Now, if you think about it, a lot of sports

01:24:17.400 have their performance improved when the athlete is lighter. Weight management is a big part of many

01:24:24.600 sports. Cyclists, runners, gymnasts, if you think about it, rowers, any sport that is cardiac output

01:24:35.640 versus body weight, those athletes, and I used to be one of them, you are just as focused on weight

01:24:42.600 management as you are cardiac output. Should these drugs be banned by WADA and USADA? Are they indeed

01:24:51.560 performance enhancing drugs? Great question. I hadn't thought about that until you asked it. Great

01:24:56.600 question. Because it introduces a whole different set of interests. Prior, we were talking mainly about

01:25:04.280 the individual persons taking the drugs interest, and a little bit about the provider's interest,

01:25:09.240 you, a little bit about society, cost, FDA, so on. Here, you've introduced a fourth party,

01:25:17.640 and that party is the sport. All the spectators, the people who own it, the other participants.

01:25:24.440 The sport has rules. Sport is very different than some other things where there's an arbitrariness to

01:25:31.640 it. Why does the baseball bat have to be this long and not that long? Why does the tennis racket

01:25:35.560 have to be within these dimensions? Well, that part's arbitrary, but what's not

01:25:39.080 arbitrary is we want it to be equal. We want everybody to have the same chance. So in other

01:25:44.120 words, we don't spend too much time worrying about the length or weight of the baseball bat. We worry

01:25:50.280 far more that you didn't screw into yours and put cork in there and change the weight of it. That's the

01:25:56.920 thing we care about is fairness. Because that's the rule. But the rule,

01:25:59.880 we even change the rules about the intrinsic things. So we change the rules about, in some

01:26:05.720 places, we don't condition on age. In others, we have age brackets. Some boxing, we have weight

01:26:10.760 brackets. Wrestling, we have weight brackets. We don't have height brackets in basketball. Some

01:26:15.880 colleagues and I are trying to write a whole paper on, mathematically, what is bias? What do we mean by

01:26:19.880 that? And we use basketball as an analogy. And I use myself as the example and say,

01:26:25.560 if I try out and I don't do well for the basketball team because I'm short, I don't call that bias

01:26:31.160 because intrinsic to the idea of basketball is these are the rules. We don't have springboards

01:26:35.880 for shorter guys. We could, but we don't. We don't have height classes. And so that's not biased.

01:26:41.640 In contrast, if you asked me to try out to be a biostatistics professor and the book is on the top

01:26:48.360 shelf that you want me to lecture from and there's no step stool, I would argue that's biased because you

01:26:53.880 could have put a step stool there and it's not intrinsic to biostatistics professor performance

01:26:59.160 to be able to reach tall things. And so we need to look at the sport and say, what do you want it to

01:27:04.840 be? And if somebody says, I want it to be things where part of the sport is being able to maintain

01:27:11.080 your weight. And so I don't want anybody to have a performance enhancing drug, then to me, so be it.

01:27:17.160 I could also alternatively turn around and say, we just want you to be able to get the basket in the

01:27:22.440 hoop, or we just want you to be able to row the boat. And if you do it by having more money and

01:27:27.960 hiring a better coach and you do it by taking Ozempic and you do it by having good genes,

01:27:34.120 all is fair. I don't think there's a right answer there from the sport point of view.

01:27:37.640 But given that the sport has already made several decisions, they've already said,

01:27:42.680 you can't take a drug that increases the number of red blood cells that you have.

01:27:48.680 That's EPO. You can't take a drug that increases the rate at which your muscles repair themselves

01:27:57.480 after hard training. That would be testosterone. Go on and on and on. You can't take a drug like a

01:28:03.560 diuretic that takes body weight away from you. This is not a philosophical question about drugs.

01:28:09.800 It's a practical question about this class of drugs, whose efficacy is, as you said, profound

01:28:17.800 and its safety, at least in the short term, unquestionable. Are we going to basically see

01:28:24.200 at the Olympics this year in France, if they were drug testing for it, what fraction of athletes would

01:28:30.440 be taking GLP-1 agonists of the sports where body weight regulation is a key? I don't expect many

01:28:37.560 shot putters to be taking it. Fascinating. No.

01:28:39.400 But I do wonder how many boxers and rowers and runners and cyclists will be taking it.

01:28:43.560 Really interesting. We should do that study. Let's work on it. So I don't know the answer.

01:28:47.720 Haven't heard about that before. I think your speculation is apt. I think that as a formalist,

01:28:54.040 I would go and say, well, what is these groups that have said you can't take testosterone and this and

01:28:58.680 this and that, they probably put out some underlying principles. They probably said,

01:29:03.160 you cannot take a drug that enhances performance unless you have a medical need. I don't know if

01:29:09.880 they've said that, but if they have, then it could get really tricky because now you say, well,

01:29:16.280 who defines the medical need? That's right.

01:29:18.360 What about now is it fair if we take the person who's just below the threshold for needing it,

01:29:23.960 who says, I don't get to take the drug, but the person who's just above the threshold,

01:29:28.280 who does? You have then also this idea of a fairness, a disabilities issue. If I have obesity,

01:29:34.680 particularly I've got a strong genetic predisposition to it. I can't manage to be

01:29:39.160 not obese without the drug. Do I effectively have a disability? And is this now prejudicial or violation

01:29:46.840 of the Americans with Disabilities Act or something like that, or different countries have different

01:29:51.560 variants, but is there a fairness issue? And again, I don't know that there's a right answer.

01:29:57.560 I think these would be tough political and moral questions, but it's really particularly tough

01:30:03.240 because you bring in the interest of the sport. And then you're going to get also, it's going to

01:30:07.800 reflect back when you get into the health interests of the individual, just as with many sports,

01:30:13.880 we might say it's in the interest of the team or the coach or the sport itself to have this person at

01:30:20.760 greater risk. But of course, it's not in the interest of their situation. And yet we somehow

01:30:28.360 accept that we allow people to play football, even though there's concussion risk and we allow people

01:30:32.600 to box and many other things. But are there some limits where we might say, we're not comfortable

01:30:39.160 with your putting yourself at risk for this? We need to protect you as much as the sport.

01:30:45.000 All right. Let's pivot to something a little bit easier to talk about. You've already alluded to

01:30:49.960 protein. It's a huge interest of yours clinically, personally. It's a topic I've addressed a number of

01:30:56.200 times. What do we know about protein? And what do we, at least in your view, what do you think we think

01:31:01.320 we know that we don't know? So you're as much or more of an expert on the physiology and biochemistry

01:31:06.440 of it. But I will venture a few things. So with respect to what we know, we know some very basic

01:31:13.800 things. We know you can't live without protein, without consuming some protein. We know that the

01:31:18.920 body is made, not totally, but heavily of proteins. They're essential for functioning. We know that

01:31:26.360 proteins are made up of amino acids. They're different amino acids that have different effects.

01:31:31.320 Some amino acids can be synthesized in the body. Some can't be. I think you did a podcast with

01:31:36.760 Luke Van Loon recently, which I found enormously educational. And so I'd refer people to that one.

01:31:42.760 And he knows a lot more about protein than I do. So I think we know that we need protein.

01:31:47.960 We need a certain amount. We need certain amino acids and we can get them from various foods or

01:31:53.880 combinations, animal-based foods. We can pretty much get all the proteins we need from them.

01:31:59.720 If we only eat plant-based foods, it's not impossible with the exception of maybe taurine,

01:32:04.280 but whether we have to consume taurine or not. If you're a cat, you have to consume

01:32:09.800 taurine. But if you're a human, maybe not. You could drink Red Bull and still eat plant-based

01:32:14.840 protein and you're fine. So then we get into, are there known things about the amino acids in

01:32:20.440 terms of long-term human health? I think modest. So we see certain things about leucine being important for

01:32:27.960 skeletal muscle growth, anabolic effects. We see some things about isoleucine in mice maybe not

01:32:35.240 producing longevity. We see taurine supplementation in mice and some other species appearing to prolong

01:32:45.480 life in Vijaya Dev's work. We see methionine restriction in Rich Miller's work prolonging life.

01:32:52.600 Thionine is related to taurine. A lot of confusion. What will really prolong life in humans is unclear.

01:33:00.200 Whether the same things that will prolong or shorten life in terms of macronutrient composition

01:33:05.080 in mice will do the same thing in humans is unclear. And there are different outcomes. This is, again,

01:33:10.360 part of why I railed against the idea of healthy foods so-called or unhealthy foods so-called. Healthy for

01:33:16.200 what? You might want to be 10 pounds thinner. I might want to be able to lift 10 more pounds on

01:33:22.120 the bench press. That person wants to live 10 years longer. The three diets for those things may be

01:33:28.440 different. So I think that after that, after the idea that we need some protein, we need some minimal

01:33:34.760 amount, we need the amino acids, I think it gets shaky then. In other words, minimums and maximums were

01:33:41.080 not necessarily a part of what you just described with much certainty. Right. I think there's

01:33:46.120 reasonable confidence, and you've been a great progenitor of this idea, that the old school

01:33:52.280 recommendations for this much is enough were probably too low. 0.8 grams per kilogram body

01:33:59.960 weight is the RDA. Right. And I think many people think that's too low, that you can survive on. It's

01:34:06.600 not that you can't survive, but can you thrive? And that's sort of, I think, a big point of your book

01:34:11.400 and other people's lives, your book Outlive, which is, yes, we can think about treating diseases,

01:34:18.120 we can think about preventing diseases, but neither of those are equivalent to optimizing our lives and

01:34:24.440 our health. And different people have different ideas of optimal. Is optimal optimal comfort? Is it

01:34:28.760 optimal length? Is it optimal ability? Performance, yeah.

01:34:32.760 Whatever it is, it doesn't seem that that's the level, the LRDA, 0.8 grams per kilogram,

01:34:39.960 is the optimal level for health or longevity or anything else. So it's probably somewhat higher.

01:34:46.200 Next question is, are there minimal thresholds at any sitting? So Don Lehman and others have argued

01:34:53.480 there are. Sometimes you hear 20 grams, sometimes you have 30 said. So 20 to 30 grams in a particular

01:35:01.480 sitting is the minimum to get anabolic. Is that true? When I as a statistician hear this, I'm like,

01:35:09.400 really? A threshold in biology? You're telling me there's a step function and you know it? Now,

01:35:15.880 I don't really believe there's a step function, but maybe it's sigmoidal. Maybe it's sort of a

01:35:19.160 little flat and then it goes up steeply and then it's flattens out a little bit. And then I say,

01:35:23.000 how much sample size and how many different doses would you need to really get a fix on that

01:35:28.920 and test whether it's there? And then you look at the studies done and you go,

01:35:34.280 you've got to be kidding me. We talked earlier about pharmaceutical company studies. Think about

01:35:39.640 the numbers of people on which we tested COVID-19 vaccines. Think about the numbers of people we've

01:35:46.040 tested statins and now GLP-1 agonists. Now think about the numbers of people used in randomized controlled

01:35:54.520 trials from the nutrition community to look at protein needs. And not all of us need to

01:35:59.880 take a statin. A lot of us do. Not all of us need to take GLP-1 agonists. We've talked about that.

01:36:06.520 All of us need to eat protein. And yet the quality of evidence and the quantity of evidence we have

01:36:12.200 is tiny. It's dust compared to what we have on these pharmaceuticals. And so we really need to ramp

01:36:19.560 this up. I would say, I don't think that we really know that you don't get anabolic until you hit 20 or

01:36:25.720 30. But again, just to be clear, I mean, the mechanisms that are described on those are based

01:36:30.840 on small studies. And they're really small by necessity, just based on funding and complexity

01:36:36.440 of doing these studies. I mean, these are amino acid labeled tracer studies where they give people

01:36:42.920 various doses of protein and they look at muscle protein synthesis. I'm not here to say that we

01:36:49.320 shouldn't be doing bigger, better studies, but some of the studies that have been aimed at elucidating

01:36:54.760 this are quite rigorous in terms of their mechanistic insights. And so I guess the question is,

01:37:02.760 isn't it at least biologically plausible that there is a threshold? And I agree, it's very unlikely a

01:37:08.680 step function. It's more likely a sigmoidal shaped curve, but it seems at least biologically

01:37:14.600 plausible, which doesn't make it right, that at low doses, at 10 grams of amino acids,

01:37:20.360 the liver itself might just prioritize gluconeogenesis. And there's a saturation point

01:37:26.120 at which it says, oh, well, okay, we have excess nitrogen now. Let's go off and do this other thing.

01:37:31.000 I think it's entirely plausible, but we talked many times.

01:37:34.760 Yeah. Lots of plausible things turn out to be wrong.

01:37:36.440 Exactly. So we need to do the studies. I'm not putting anybody down for these studies.

01:37:39.880 No, no, I get it. Yeah.

01:37:40.920 Some of them, when I hear about what Don Lehman's done and what Luke Van Loon has done and others,

01:37:45.640 is really impressive and rigorous. But as a statistician who's saying, do we know the answer?

01:37:50.600 I say, not really. We're also interested in long-term effects. And so there's that old saying,

01:37:55.720 there's many a slip twixt cup and lip. What's the saying?

01:37:58.760 There is many a slip twixt between cup and lip, drink and you spill. You think it's a done deal. If I've got the

01:38:06.200 cup and I'm moving it through my mouth, I get the drink, but maybe not. The tracer studies are

01:38:11.080 important, but-

01:38:12.360 They're indirect.

01:38:12.920 Right. What we really want to know is if you do this for a year, are you stronger? Are you bigger?

01:38:17.960 Are you, you know, et cetera. So I think we don't really know that unequivocally.

01:38:21.960 You did a great discussion recently of a study in which Luke Van Loon was one of the authors

01:38:26.920 that used up to a hundred grams and looked at the other end of the threshold. Is there an asymptote?

01:38:32.200 Is there a level in which you don't get any more benefit?

01:38:34.600 Right. Where conventional wisdom was 40 grams-ish was the ceiling. And I think the study from Luke

01:38:41.080 suggested, that might be true for a very rapidly hydrolyzed protein, such as whey. But with casein,

01:38:48.680 at least his data suggested maybe not. Time release protein might be a value. And meals,

01:38:53.880 like a steak might be closer to casein kinetics than whey kinetics.

01:38:58.760 Right. So I think we don't know that there's an upper limit. We don't know the full duration.

01:39:04.520 Some people said you're only stay anabolic for two hours after eating the protein. I think that

01:39:09.240 Luke's study shows more. So I think there are those things.

01:39:12.600 What about the maximum amount of protein? Again, traditional thinking here is three grams per

01:39:17.880 kilogram is the maximum. And if you consume more than three grams per kilogram,

01:39:23.880 as a healthy individual, you risk kidney damage.

01:39:26.760 Right. This is something that's been intriguing to me. When I look at this and I haven't done a

01:39:31.800 complete thorough check, but I'm sort of in the process of working through it, I hear it's going

01:39:37.560 to reduce bone mass or could reduce bone mass, excessive protein intake that is, could read to

01:39:43.320 kidney function problems, some other unspecified problems. There's even this old thing called rabbit

01:39:49.320 starvation, which you can find papers going back on this at least a hundred years. And they talk about

01:39:54.440 hunters and survivalists and so on out in the woods who can shoot a lot of rabbits and eat their fill

01:40:00.680 of rabbits and yet starve to death because they don't have enough fat and carbohydrate to properly

01:40:06.440 digest. But if you say, now let's go back and find the trials that show this. So you find a paper and it

01:40:13.240 says, here are the limits. Steve Heimsfield and Sue Shapps has just had a wonderful nutrition 101

01:40:19.560 commentary in New England Journal of Medicine. And they talk about some of the upper limits and they

01:40:23.640 cite some papers and they say, some bad things can happen if you eat too much protein. You go back to

01:40:27.960 those papers. They're review papers. They're not trials. Those papers say, the limits are like this

01:40:34.280 and they cite a few things. And you keep going back and we're unable to find trials where people do it.

01:40:39.400 All you seem to get to is somebody said, well, but there was this group of hunters in this population

01:40:47.640 who ate this many grams and they were okay. So don't go above that. But nobody said, if you go above

01:40:54.520 that, something bad happens. There's this study, and I say loosely of one or two guys who ate nothing

01:41:00.680 but meat for six months to a year and they were fine. Interestingly, there's a corresponding study from 1928,

01:41:07.400 two Polish scientists put two Polish people on a diet for six months of nothing but potatoes,

01:41:14.680 fruit, and a little bit of fat to cook the potatoes in. And the idea was, can you get enough

01:41:18.600 protein and nitrogen out of potatoes? And the answer was yes, at least for six months and they were fine.

01:41:23.480 So people have been fine eating nothing but meat and no plants for six, 12 months. These are

01:41:29.720 semi-anecdotes. They're intervention studies, but they're not big randomized control trials.

01:41:34.840 Or nothing but potatoes for six months as a protein source. And they've all been fine.

01:41:40.440 Do you remember in any of those studies how much weight was lost in each group?

01:41:44.120 In the potato study, there was neither weight lost nor gained.

01:41:47.640 Really? Yeah.

01:41:48.760 How did those people not lose weight?

01:41:51.400 They're probably thin to begin with. If you go back to the classic studies of Ted Van

01:41:56.040 Italy and Sammy Hashim from 60s, I think, where they would bring in, as they described at the time,

01:42:04.280 lean Columbia University students versus obese adults. And they gave them MetraCal,

01:42:10.520 which was sort of the boost or insure of 1960. And through a tube where they could get unlimited supply,

01:42:18.120 but they couldn't really see how much they were eating. And what they found is that the obese people

01:42:23.080 generally lost weight. They didn't fully compensate or the monotony made them reduce intake. Whereas

01:42:29.640 the lean students all maintained weight. So it's probably that the effects of diet on weight change

01:42:36.600 vary a great deal depending on where you're starting. Anyway, so the potato eaters neither gained nor lost

01:42:43.000 weight and they have beautiful nitrogen balance. From what I've understood, I haven't studied those papers

01:42:48.760 as carefully, but all meat eaters, at least for six, 12 months were fine. So I think either way,

01:42:54.520 it can be done. What I have not seen is somebody who said, to test this rabbit starvation thing,

01:43:01.640 we brought a bunch of healthy adults in and we fed them nothing but cooked rabbit for six months and

01:43:09.000 something good or bad didn't happen. I've not seen somebody say, we fed enormous levels of protein

01:43:15.320 to normal adults and we saw leaching of bone mass. And by the way, do you think that study would need

01:43:22.280 to be rabbit because it's so lean or could it be ribeye, which is equally void in carbohydrate,

01:43:29.560 but at least is high in fat. So from a macro perspective, you're dividing things up.

01:43:34.680 I think it depends on what you think the mechanism of action is. And it's not crystal clear to me that

01:43:38.680 people have specified a crystal clear mechanism of action. There is some speculation. And again,

01:43:44.840 there's some nice recent papers on this, that especially for people who want anabolic effects,

01:43:50.440 bodybuilders, weightlifters, that when you eat protein, you should have some carbohydrate with it

01:43:56.760 that will enhance the anabolism. Through the insulin. Exactly.

01:44:00.760 Or you inject insulin. But my understanding, and again, I'm just really entering in this,

01:44:06.920 but I've read so far, there is no compelling evidence that that is true. That is that you get more

01:44:12.840 anabolic effect if you eat carbohydrate with your protein than if you eat protein alone.

01:44:18.120 So another presumption or myth. So rabbit starvation, bone loss, kidney problems, and

01:44:24.600 you must have carbohydrate with it. All of these things are, I think, these things that are presumed

01:44:28.920 known and readily talked about, but I don't think demonstrated. So I've not seen any trial data yet

01:44:35.800 in normal adolescents or adults that suggests a negative health problem, not conjectured, but

01:44:45.080 observed as a result of too much protein. I'm not saying there isn't such a study, but I have not

01:44:50.040 yet found those studies. If anybody else knows them, please send them to me.

01:44:53.240 Yeah, that would be interesting. Let's just assume that those studies don't exist, in fact. Or if they do,

01:44:59.000 they're very, very small and therefore probably not worth extrapolating to the ends of the universe on.

01:45:05.720 What is the probability that such basic questions like this will be answered in the coming decade

01:45:12.920 of nutrition science? What is the appetite, no pun intended, for this type of clinical investigation,

01:45:19.560 especially in light of everything else we've spoken about, which is, hey,

01:45:23.640 the name of the game in nutrition science now is pharmacology. It's not these mundane questions

01:45:32.040 about macronutrients. For obesity treatment and closely related things, diabetes treatment,

01:45:37.800 prevention, I think you're right, the name of the game. For other areas, I think there'll be more

01:45:43.480 interest. So longevity promotion, and you'll get something like the Evolution Foundation weighing in,

01:45:48.600 which could conceivably do big studies. But even there, nothing's unlimited. They may say,

01:45:55.000 well, we're going to focus a big trial that'll be definitive and really give the answer, but it's

01:45:59.640 going to give the answer only in this age group or something like that. I think NIH will fund some,

01:46:04.920 I think industry will fund some, but NIH industry are likely to fund, in most cases, things that are

01:46:10.680 small enough that we're not going to know the answer about every dose at every period of time.

01:46:18.280 In every race, age, sex, and health status group. What we'll know is pocket answers. So I think what

01:46:24.760 we could get, for example, is a study funded in which we very, very thoroughly looked at ordinary

01:46:32.280 healthy adults over 60 years old who want to increase strength and muscle mass. And we'll look

01:46:40.680 at protein intake and we'll look at other upper limits. I think that could conceivably be done and we

01:46:45.720 could probably nail that answer. But then you might come back and say, well, you've shown it safe or

01:46:49.960 unsafe for a 60 year old. That doesn't mean it is for a 20 year old. And we'll say, that's true. And 0.86

01:46:57.000 you said, when you showed it with casein whey, but not with pea protein and whatever, that's true too.

01:47:04.200 So how compelling do you find the data that high protein diets reduce longevity? There are many

01:47:11.000 proponents of this view out there. Often I suppose within the plant-based community, although again,

01:47:16.920 I don't think those are necessarily an overlap, although that just seems to be where I notice most

01:47:22.680 of the lower protein is better rhetoric. But how do you assess the strength of that claim?

01:47:29.400 Very low. I think it's going to depend on species and that's important because it leads to the

01:47:34.360 extrapolation issue. If you're a butterfly, I think it probably does reduce lifespan. And I think a

01:47:42.200 higher carbohydrate diet may increase lifespan more if you're a butterfly.

01:47:47.080 We do have some butterfly listeners of this podcast, but I'll be honest with you. Our efforts

01:47:53.320 to increase butterfly listenership have largely been, I would just say they've been less successful

01:47:58.200 than I would have enjoyed. We'll keep working on that. But I think in humans,

01:48:02.680 there's no compelling, in my view, no compelling evidence. I would even say there's some evidence

01:48:07.720 and reasons to believe the contrary. Some of that's going to be tied to wealth. Wealthier people

01:48:12.360 eat more protein than less wealthy people, including within our country. So it's hard to tease all this

01:48:17.640 apart, epidemiology. But if you look at the association studies, even there, I don't find it compelling.

01:48:23.720 Then you can say, well, do you accept the association studies? Not all that much. If you look at the

01:48:29.080 mouse studies, I'm not sure if there you see the full translation, but also I'm not sure they're

01:48:35.000 all that compelling, that low protein. So I think nothing that I know of would say to me there's

01:48:41.000 very strong reason to believe, even if not definitive RCT, that higher protein will lead to

01:48:46.280 less longevity. And if anything, I think there's more compelling reason to the contrary. Now, my friend

01:48:52.760 and IU alumnus, Barry Sears, talks about the zone diet. And the idea of the zone is that there's

01:48:58.600 not too much on this end. You shouldn't be too high on this end. Don't be too low over here.

01:49:03.800 There's a zone of things in the body and in diet that are right. You got to find the right spot.

01:49:09.320 And he very strongly believes that you want to upregulate AMPK to live longer,

01:49:14.280 and you want to not upregulate mTOR too much to live longer. And I know you've talked about rapamycin,

01:49:21.880 which sort of has effects that would go along with what he's saying. So he might argue from that

01:49:28.280 point of view that too much protein would reduce longevity. And again, I can't say he's wrong. I think

01:49:35.480 it may also depend on at which period of time. So what's good for you to do or eat early in life

01:49:43.000 to prolong life may not be what's good for you later in life. Sometimes we call that

01:49:48.040 antagonistic pleiotropy. And here we might say, and John Hollisey found this in rats. I don't know

01:49:54.280 if it holds up as one study, but found that exercise in rats reduced mortality rate in the first half of

01:50:01.240 life, but it increased mortality rate in the second half of life. And so if we accept that as causal and

01:50:08.600 valid and replicable, then maybe the same things might be true for protein. It might be good to eat

01:50:13.960 more when you're younger and less when you're older, or less when you're younger and more when you're

01:50:17.720 older. I don't know. But I think that we also need to think that longevity is only one factor. We talked

01:50:24.200 about this a little bit the other day, and there's no right answer to this. But if somebody were to say

01:50:29.320 to me, live this way, and our best guess is you'll die two years earlier. But until you die, you'll be 0.86

01:50:37.880 stronger, you'll feel more energized, you'll look better. I'd say, I'll make that trade. Now, 0.67

01:50:43.480 someone else might say, I won't make that trade. And who's to say who's right?

01:50:46.680 David, one last thing before we go. You serve as sort of the editor of a newsletter that comes

01:50:52.280 out every Friday, Obesity and Energetics. I've been a subscriber for, I guess, a decade,

01:50:57.480 maybe a bit more, right? When did it start? It started kind of organically all the way back

01:51:02.840 when I was just in grad school, basically, or getting out. And then there was no formal letter

01:51:08.040 at the end of the internet. I would hand my professors papers I had read and said,

01:51:12.440 hey, maybe we could talk about this or something. Then I moved away, I would mail them. And then I

01:51:16.840 mailed to a couple of people. And it kept going. And soon people started asking to be adding to my list.

01:51:23.000 Then it became electronic. Then it became a formal web thing.

01:51:26.440 So now it goes out to over 100,000 people worldwide. We don't charge anything for it.

01:51:31.160 It's free. It's called Obesity and Energetics Offerings. We don't accept any commercial support

01:51:35.880 for it. It contains usually about 100 or a few more links to mostly scientific papers,

01:51:43.240 sometimes popular media articles commenting on things in virtually every category related to obesity,

01:51:50.360 energy, metabolism, nutrition. Well, I'm a big proponent of it. My team all subscribes to it.

01:51:57.960 And it's one of the not too many newsletters that I rely on. Again, people ask me, how do I stay up to

01:52:04.120 date on things? And the truth of it is I have to rely on other people doing a lot of the aggregation.

01:52:09.880 And then I'll kind of go where my curiosity goes and sometimes go a bit deeper. But

01:52:13.720 anyway, I just wanted to make a plug for people to subscribe to obesity and energetics offerings.

01:52:19.560 It's great. And one of the fun things that I think is great for people learning is one of the sections

01:52:25.320 is always called Headline Versus Study. And I just think that if folks listening to us now

01:52:31.800 are not going to read any of the subheadings there, just read that one. Because it gives you a great

01:52:38.520 sense of how misleading the traditional media can be. Not necessarily because they're nefarious. I don't

01:52:45.960 think that's the case. I think it's scientific ignorance and a misalignment of incentives.

01:52:51.880 So one, they're simply not qualified. They don't have the scientific literacy

01:52:55.880 to understand what a study shows. And secondly, they're really incentivized to get you to read a

01:53:00.360 study and click through something. And they have to come up with a headline that makes that appealing.

01:53:05.880 And as you point out every week, there's a great example of one where the headline is patently false

01:53:13.000 at worst and at best, so misleading as to be useless. So what I think is valuable for folks is to

01:53:19.960 get into the habit of checking that once a week and seeing, hey, that was a headline. I could see

01:53:25.080 that headline. I could see how I'd get fallen. I'd get duped for that. But oh, there's the study.

01:53:30.200 The headline could be something as outrageous as women are so much more likely to outlive their 1.00

01:53:36.200 partners if they have sex three times a week. And then you look at the study and it's about this

01:53:40.520 rare species of fruit fly that sometimes mate with male fruit flies that die a little premature.

01:53:47.480 You know what I mean? Exactly. It can be so ridiculous.

01:53:50.200 Well, I really appreciate your pointing that out. And anybody can subscribe for free. Just 0.88

01:53:53.640 type into Google obesity and energetics offerings. You'll find it. If not, email me. And I hope Andrew

01:53:59.160 Brown, who's a professor and a former mentor of mine at University of Arkansas now is listening.

01:54:05.560 He took over handling that category a few years ago. He does a beautiful job with it. And so he sets

01:54:11.000 these things up and often finds these interesting things. And even got to the point where others picked

01:54:16.200 up on his use of the phrase in mice as kind of like a standard thing, which is just as often

01:54:22.040 a way of saying, well, we found this in mice, but in mice is often left off in the headlines.

01:54:27.800 How many people are involved in curating that list each week?

01:54:31.320 In any one week, it's about five people, including me. So there's me, there's an editor,

01:54:37.160 then there's Andrew who cleans everything up. And then there's Colby Vorland who cleans everything

01:54:41.720 up and someone else who posts it. That's a lot of work for five people.

01:54:44.840 That's a lot of work. But it rotates. So the editor rotates

01:54:48.040 every week, but the rest of us, including me are on.

01:54:51.320 Thank you for never asking me to be an editor on that.

01:54:54.760 I hadn't thought about it until now. Maybe I will.

01:54:56.600 No, no.

01:54:57.800 It is a lot of work, but it's fun. It's a labor of love.

01:55:00.280 Well, David, thank you so much. This was a super fun discussion. And I know that folks are

01:55:04.360 going to get a kick out of it. Thank you, Peter. Great to be here with you.

01:55:07.400 Thank you for listening to this week's episode of The Drive. It's extremely important to me to

01:55:12.840 provide all of this content without relying on paid ads. To do this, our work is made entirely

01:55:17.960 possible by our members. And in return, we offer exclusive member-only content and benefits above

01:55:24.280 and beyond what is available for free. So if you want to take your knowledge of this space to the

01:55:28.600 next level, it's our goal to ensure members get back much more than the price of the subscription.

01:55:33.320 Premium membership includes several benefits. First, comprehensive podcast show notes that

01:55:39.640 detail every topic, paper, person, and thing that we discuss in each episode. And the word on the

01:55:45.480 street is nobody's show notes rival ours. Second, monthly ask me anything or AMA episodes. These

01:55:53.080 episodes are comprised of detailed responses to subscriber questions typically focused on a single

01:55:58.280 topic and are designed to offer a great deal of clarity and detail on topics of special

01:56:03.160 interest to our members. You'll also get access to the show notes for these episodes, of course.

01:56:08.440 Third, delivery of our premium newsletter, which is put together by our dedicated team of research

01:56:13.720 analysts. This newsletter covers a wide range of topics related to longevity and provides much more

01:56:19.640 detail than our free weekly newsletter. Fourth, access to our private podcast feed that provides you with

01:56:27.080 access to every episode, including AMA's sans the spiel you're listening to now and in your regular

01:56:33.240 podcast feed. Fifth, the qualies, an additional member only podcast we put together that serves as a

01:56:40.200 highlight reel featuring the best excerpts from previous episodes of the drive. This is a great way to

01:56:45.720 catch up on previous episodes without having to go back and listen to each one of them. And finally,

01:56:50.760 other benefits that are added along the way. If you want to learn more and access these member only benefits,

01:56:57.080 head over to peteratiamd.com forward slash subscribe. You can also find me on YouTube,

01:57:03.160 Instagram, and Twitter, all with the handle peteratiamd. You can also leave us review on Apple

01:57:09.260 podcasts or whatever podcast player you use. This podcast is for general informational purposes only

01:57:15.560 and does not constitute the practice of medicine, nursing, or other professional healthcare services,

01:57:20.360 including the giving of medical advice. No doctor patient relationship is formed. The use of this

01:57:26.380 information and the materials linked to this podcast is at the user's own risk. The content on this

01:57:32.500 podcast is not intended to be a substitute for professional medical advice, diagnosis, or treatment. Users

01:57:37.960 should not disregard or delay in obtaining medical advice from any medical condition they have, and they

01:57:43.440 should seek the assistance of their healthcare professionals for any such conditions. Finally, I take all

01:57:49.520 conflicts of interest very seriously. For all of my disclosures and the companies I invest in or advise,

01:57:55.500 please visit peteratiamd.com forward slash about where I keep an up-to-date and active list of all disclosures.

01:58:03.300 Thank you.

01:58:04.300 Thank you.

01:58:33.300 Thank you.

The Peter Attia Drive - August 19, 2024

#314 ‒ Rethinking nutrition science： the evolving landscape of obesity treatment, GLP-1 agonists, protein, and the need for higher research standards ｜ David Allison, Ph.D.

Episode Stats

Summary

Transcript