Russ Roberts

Taubes on Fat, Sugar and Scientific Discovery

EconTalk Episode with Gary Taubes
Hosted by Russ Roberts
PRINT
Baumeister on Gender Differenc... Simon Johnson on the Financial...

Gary Taubes, author of Good Calories, Bad Calories, talks to EconTalk host Russ Roberts about what we know about the relationship between diet and disease. Taubes argues that for decades, doctors, the medical establishment, and government agencies encouraged Americans to reduce fat in their diet and increase carbohydrates in order to reduce heart disease. Taubes argues that the evidence for the connection between fat in the diet and heart disease was weak yet the consensus in favor of low-fat diets remained strong. Casual evidence (such as low heart disease rates among populations with little fat in their diet) ignores the possibilities that other factors such as low sugar consumption may explain the relationship. Underlying the conversation is a theme that causation can be difficult to establish in complex systems such as the human body and the economy.

Size: 38.0 MB
Right-click or Option-click, and select "Save Link/Target As MP3.

Readings and Links related to this podcast

Podcast Readings
HIDE READINGS
About this week's guest: About ideas and people mentioned in this podcast:

Highlights

Time
Podcast Highlights
HIDE HIGHLIGHTS
0:36Intro. [Recording date: November 7, 2011.] Book is about diet and health, heart disease, obesity; also about the challenge of being a truth seeker when you have an ideology or a pet theory. So we're going to talk about health, but underneath the discussion and occasionally right out in the open, we're going to be talking about economics--first because the body, like the economy, is a complex system where it's hard to measure cause and effect, and second, because epidemiology, the study of what causes diseases and what cures them, is very much like economics--a lot of data, a lot of statistical techniques trying to isolate the impact of one variable on another. So, Gary, you'll talk about the health, and from time to time I'll interject a little economics. Now your book starts with the idea, which was very prominent and commonly believed by a large group of people, that fat--eating fat and fat in your diet, particularly animal fat--isn't good for you and it leads to heart disease. How did that come to be accepted wisdom in the medical profession? First, let me say I think it's still commonly believed by most people, and the latest dietary guidelines are trying to get us to limit our fat intake, and limit our saturated fat intake. This is an hypothesis that grew out of the observations of one very zealous University of Minnesota nutritionist in the 1950s, a fellow named Ancel Keys, who came up with this idea that dietary fat raised cholesterol, and it was raised cholesterol that caused heart disease. At the time there was effectively no meaningful experimental data to support--I'll rephrase that: There was no experimental data to support that observation. It seemed plausible, though. It seemed plausible, compelling. Keys was a persuasive fellow. And by 1960 or so, the American Heart Association (AHA) got behind it in part because Keys and a fellow-proponent of this hypothesis, a cardiologist from Chicago named Jeremiah Stamler, got onto the AHA, got involved with an ad hoc committee, and were able to publish a report basically saying we should all cut our fat intake. This was 1961. Like I said, no data to support it; no experimental data at all. And once the AHA got behind it, it got a kind of believability. The attitude was: It's probably right, and all we have to do is test it. Or, we're going to believe it's true, but we don't have the data yet because we haven't done the tests yet. And researchers start doing the tests, experimental trials, taking a population. For instance, a famous study at the VA hospital in Los Angeles, where you randomize half of them to a cholesterol-lowering diet which is not actually low in fat, by the way--it's low in saturated fat and high in polyunsaturated fat. And then the other half of your subjects eat a control diet and you look for heart disease over a number of years and see what happens. And trial after trial was sort of unable to prove the hypothesis true. But the more we studied it, the more people simply believed it must be true. And meanwhile, the AHA is pushing it; other observations are being compiled to support it even though in order to support it you have to ignore the observations that don't support it. So, you pay attention to the positive evidence, ignore the negative evidence. One Scottish researcher who I interviewed memorably called this "Bing Crosby epidemiology" where you "accentuate the positive, eliminate the negative." Basic human nature. But this is what happened. And as the AHA gets behind it, the journalists see the AHA as honest brokers of information on this, so they have no reason to doubt the AHA. And the AHA was honest brokers--they just were bad scientists. Or they were not scientists. So, then the press gets behind it, and as the press gets behind it, politicians begin to think maybe we should do something about it, and a Congressional subcommittee gets involved, run by George McGovern, that had originally been founded in the late 1960s to address hunger in America; and they did a lot of good things with school lunch programs and food stamps. And by the mid-1970s they were running out of things to do, so they decided: Since we've been dealing with under-nutrition, which is not enough food, they would get involved with over-nutrition, which is a problem of too much food and obesity and diabetes and heart disease. And they had one day of hearings, McGovern's subcommittee, and they assign a former labor reporter from the Providence, RI, Journal to write the first dietary goals for the United States--the first document ever from a government body of any kind suggesting that a low fat diet is a healthy diet. And once McGovern comes out with this document, written by a former labor reporter who knew nothing about nutrition and health; now the USDA feels they have to get involved; and you get this kind of cascade or domino effect. To the point that by 1984 the National Institute of Health (NIH) holds a consensus conference saying that we have a consensus of opinion that we should all eat low fat diets, when they still don't have a single meaningful experiment showing that a low fat diet or cholesterol lowering diet will reduce the risk of heart disease, or at least make you live longer. Because a few of the studies suggested that you could reduce the risk of heart disease but you would increase cancer. And not one study--the biggest study ever done, which was in Minnesota, actually suggested that if you put people on cholesterol-lowering diets you increase mortality; they had more deaths in the intervention group than the control group. And in that case they just didn't publish the study because they got what they perceived as the wrong result. So, it's sort of 20 years of this must be true or why else would we be studying it.
7:26I wanted to go back a little bit. And again, I encourage the listeners to think about the analogies with economics. In economics, we don't usually have real experimental data with a control group. We have some experimental data in microeconomics, but when we talk about macroeconomics, the business cycle and what we might do about it, we don't have a control group. What we have are natural experiments, things that happen in the world that we assume are exogenous; and we can compare an after with the before; and we also have empirical studies using statistical evidence to try to tease out the individual effects of various variables by holding other things constant. And then we have logic, I should add, which is very important--often downplayed but very important. And of course it can lead you astray if you are not careful, on any issue. So on the case of this relationship between diet and particularly saturated fat and heart disease, there was a certain logic, which is we are pretty sure that heart disease is caused by arteries getting clogged up by plaque, and therefore if you ate things that changed your cholesterol level, it seemed logical that of course it would be the case then that you'd have a higher risk of having heart disease and then a heart attack. The evidence for that could be of two kinds--three kinds in the case of epidemiology. You could have historical incidents, and this would be the equivalent of trying to figure out what got us out of the Great Depression--and I want you to talk about that. And example that stands out in my mind from the book would be Japan: comparative epidemiology between Japanese citizens in Japan versus when the come to America, so you've kind of held genetics constant; so what would be left would be dietary effects. There's that kind of evidence; that encouraged people to think about this. They weren't just relying on the logic. Then you'd have real experiments, these control groups. Give us a little flavor of those two kinds of evidence--the before and after of the different populations. And then you talk about these failed experiments: there were no clinical trials that found some impact of diet on heart disease? And if so, what was the explanation of the proponents of the theory? They weren't unanimous. There were voices in the profession and the literature saying: This isn't 100% proven. Yes, those voices were often drowned out, but they were out there. So they had to respond to that. So first talk about the historical, comparative, what we could call a natural experiment; and then the actual data. The Japanese case you pointed out is a common one, because the Japanese had the highest longevity of any developed nation, and they had a very low heart-disease rate; and they eat very low fat diets. Therefore, you can hypothesize that high fat diets cause heart disease. When the Japanese come over to the United States--breast cancer is a classic example. Japanese women in Japan have very low rates of breast cancer. So when Japanese women come to the United States, by the second generation they have rates of breast cancer as high as any other ethnic group, and one possibility is it's because they come over here and they eat more fat. But the problem with those observational studies, those comparisons, is you don't know what you are looking at. So, you focus on fat because that's what your hypothesis is about--and this is an endemic problem in public health--and you just don't pay attention to anything else. So, sugar consumption is very low in Japan and very high here. So, maybe it's sugar that's the cause of heart disease, or the absence of sugar is the reason the Japanese are so relatively healthy; and if you don't look at sugar, you don't know. So, a common observation, one of the observations that sort of kicked this hypothesis up a notch, was that during the Korean War you had these young American soldiers who were killed in combat and they had autopsies; and you could already see formation of arterial sclerotic plaques in kids who are 18, 19, 20 years old. But they weren't there in the Korean soldiers. So, what's the difference between the Koreans and the United States? Well, the U.S. citizens eat high fat diets, the Koreans eat low fat diets. But there are a lot of differences between Americans and Koreans and American soldiers and Korean soldiers. For instance, Americans eat a lot of sugar; Koreans eat little. So, you don't know. All you can get out of this is a hypothesis. And you can very easily be mislead. And I had a cover story in the New York Times Magazine in 2007, using as my case study the single most well-known epidemiologic study in the country in public health, which is the Nurses' Health Study at Harvard. Every single time they've had a hypothesis of causation from their data that was tested in a clinical trial; without exception the trial failed to confirm the hypothesis. Doesn't mean the hypothesis wasn't true; but the trial found the opposite. When you get to these experiments, like I said, there are a few experiments that show that heart disease risk came down with low fat diets or cholesterol lowering diets, but simultaneously cancer risk went up. There was a study at a couple of mental hospitals in Helsinki, that was a really bizarre study, but that was one of the few studies to ever show that mortality actually increased [decreased? Econlib Ed.] on the cholesterol lowering diets. You had less disease and less deaths on the group getting the cholesterol lowering diet. So that was taken as evidence that the hypothesis was correct; but simultaneously a very similar study in Minnesota, the largest one ever done, found the opposite. They found a significant increase in deaths in the group that had gone on the cholesterol lowering diet. So if you like the hypothesis, that one is not reliable; and if you don't like it, the other one is not reliable. Exactly. And literally the investigators who did that Minnesota study--it was finished by 1973 and it was published in 1988 or 1989, which was a year after the principal investigator retired. And I am a journalist and I tracked him down and I asked him: Why did you wait 16 years to publish? And he said: Because we didn't like the way it turned out. A moment of honesty. The assumption is if you don't get the answer you expect, you did the experiment wrong. And that is still the case today. The classic example is the Women's Health Initiative, huge trial which had a lot of experiments in it; one of them was hormone replacement therapy; but one of them was a diet trial, and the estimate is that alone cost a half billion to a billion dollars. And 49,000 women are randomized into two groups; and 20,000 are put on a low fat, cholesterol lowering diet--you increase whole grains and fruits and vegetables and you eat lean meat instead of fatty meat. And they are followed for I think it was 7 years; and the other 29,000 women just go about their lives as usual. What's interesting is you would expect an intervention effect in a trial like this: you would expect to see a benefit from the low fat diet even if one didn't exist because your intervention group is getting an intervention. They are being put on a diet and they are getting counseling and the control group isn't, so it's not an equal intervention between the two. And despite that, you see no effect from this study. No effect on weight, on heart disease, on diabetes, on breast cancer. And so the largest experimental trial that I know of ever done, half a billion to a billion dollars, and the principal investigators who did this study, the National Institute of Health (NIH) that funded it, the World Health Organization, I think even the Center for Disease Control, all put out press releases saying this trial should have no effect on our dietary recommendations to eat low fat, low saturated fat diets, even though it showed that these diets were not beneficial, because the assumption is we did the trial incorrectly. And we've been telling people--women--to eat like this for 40 years; we must be right, right? If the study doesn't confirm that, we must have done the study wrong; and it's always easier to believe, with this kind of cognitive dissonance, that you did the study wrong than that you've just been giving the wrong advice. It's good here to remind the listener that Albert Einstein after the famous study that he had predicted that the gravitational field of the sun would bend starlight, and it was tested, beautiful experiment, and it was found to be true; somebody asked Einstein afterwards: What would you have done if the results hadn't confirmed your experiment? And he said: Well, I wouldn't have believed them because I know I'm right. And if you are the smartest person in the world--actually, it's even a little more dangerous when you are the smartest person in the world. But that's a common view that most scientists have. And of course many times they are wrong. It's the nature of science to make mistakes. Hypotheses don't get borne out by the data; there are things you didn't observe or imagine that could have other effects.
17:15Talk about Ancel Keys's 7-nation study, because that's another example--a beautiful example, very similar to a lot of studies we see in economics. This was a study that began to shift the paradigm. It's interesting how these work. Remember, Keys comes up with this hypothesis; he believes it. I think as early as 1952 he was telling everyone that he believed that all Americans should eat very low fat diets, even though he was admitting that there wasn't evidence to support the hypothesis. So, he could say those things simultaneously. Like Einstein did: I'm sure I'm right and we should act on it, even though I admit that we have no evidence. But we will; it's just a matter of time. The word "yet" shows up all the time in these studies. You can see it in newspaper articles: We don't have the data yet. One of the things I'm always arguing to people in these fields is you may never have the data. Never use the word "yet." It's a hypothesis. As soon as you start thinking in terms of "yet," you know that you've become biased and you are not going to be able to do good science. Slippery slope. Keys, because he's the leading proponent of the idea, even though he's done one bad study after another and he's actually been spanked by one report that came out of the AHA back in 1956, 1957 where they had a team of cardiologists, bio-statisticians look into this and say there's just no evidence to support it. But because he's the leading proponent, he's the one you give money to test the study. And this is how bias can interject itself into research from the very get-go. So, instead of saying: That's a very interesting idea, Ancel; why don't you come up with a study on the idea and we'll have a group of very smart people design it with you and then we'll fund somebody else who is unbiased to do it--instead you say: Here's $200,000, you do the study. And Ancel's not that interested in refuting his hypothesis. There's not a Popperian bone in his body. But he's a doctor and he cares about people. I said that with irony. But academics, doctors get the benefit of the doubt and it's presumed that they are unbiased, that they only care about the truth. And of course they care about money--one thing--but they also care about their reputation, the glory, the fame; they don't want to be embarrassed. There's all kind of things that get in the way of truth seeking. That's what I've argued. I think industry is far less of a factor in determining how these sciences play themselves out than the kind of issues you just talked about--the ego, funding, fame, building a career. Once you bet your reputation on a hypothesis, you cease to do science. You cease to test it, and now what you do is you look for data to confirm it. You are an advocate. It's true. I now have an alternative hypothesis and I'm an advocate for that hypothesis; and I hope it's right because it's going to be very hard for any data to come along and convince me it's otherwise. I recently blogged at Cafe Hayek, my blog, about Keynesianism; and it would be equally true about the alternative to Keynesianism, whatever flavor you want. I said: What evidence would dissuade you? If you were a Keynesian, if you believe that government spending creates prosperity in the face of a recession, which many, many very smart people have been advocating, what evidence would dissuade you from that belief? And of course we know that predicting 8% unemployment after a stimulus package is passed and getting 10% is not going to dissuade anybody. No Keynesian, in the face of the alleged or apparent or whatever you want to call it stimulus package changed their mind. They just said: Well, it was worse than we thought, or we didn't spend enough. Sorry, I've got you off track again. Talk about Keys. Here we're dealing with preventive medicine. The problem is you are preventing these diseases from occurring--in theory. So, it's not like you have symptoms that you could address immediately and see if the symptoms resolve if you do intervention; although what this does is it turns high cholesterol into a symptom. Even though you don't even know if cholesterol is the cause of disease, but what we've done is create a whole world of doctors now who think high cholesterol causes disease. And then they can treat it with statins, and they think they are treating symptoms when they have no idea whether they are actually preventing death or real disease, hard endpoint, which is what you want to do. But you can never tell from your actions whether what happens in an observation. There's a famous study in Finland where Finland changed their diet and heart disease rates came down in this one part of Finland. And ergo they got everyone eating lower fat diets; but they also were quitting smoking and also doing other things. Then if you look at that part of Finland you say: This proves that our intervention works. And then other research came along and said: Wait, you've got the same decreases in heart disease rates all over Finland, in places where you didn't do the intervention. Slightly awkward.
22:49But go back to the 7-nation study. What did Keys do there? Keys chooses 7 nations to study the diet and then follow them for a decade and study them with regard to heart disease rates; and he picks nations in advance--he rationalized it, where this is where he has connections, he knows people, so it's cheaper to do it. He's created bonds with researchers; they can help him on it. So, he picks Crete and like, Yugoslavia, and Japan--a couple of places in Japan. All over the world. It's a nice representative survey of different types of developed nations, less developed, Asian, Western. All nations where he has a pretty good idea what the heart disease rate is and what the diet is. Awkward. So, maybe he's stacking the decks. It would have been interesting had he picked France, since it's a country with a high fat and high saturated fat diet and low heart disease rates. Or Switzerland. But you are not going to do that. So, he picks countries where he knows in advance, kind of what the answer is; and then he gets the answer he expects for the most part. So, saturated fat tracks with heart disease, ergo so does cholesterol, ergo saturated fat raises cholesterol, ergo saturated fat causes heart disease. I'm going to bring in sugar again because I can't help myself; but sugar actually tracks just as well with saturated fat, because basically affluent nations eat a lot of saturated fat and a lot of sugar. And Keys is a believer of the fat hypothesis, not the sugar hypothesis; that was a British nutritionist named John Yudkin. He just says sugar is irrelevant. Obviously people who eat a lot of sugar are obviously people who smoke a lot of cigarettes. Cigarette smokers, because they drink coffee and they put sugar in their coffee; so you could rationalize away the sugar. And now we've spent--this is the most expensive observational study ever done, so the findings have to be true. And we're going to believe it by virtue of the fact that it has to be truth. Actually, when I've lectured to epidemiologists about this problem that there's no causal information contained in an observational association, their argument still to this day is: But this is the best we can do. It's the same in economics. And the similar thing in economics is: that it's better to do something than nothing, and so, although we're not sure 100% about this dietary thing, we can't just stand by and let people die of heart disease. We've got to do something, and the best thing we can do given what we know is to reduce fat. What's the problem with that? It seems like better safe than sorry. It seems like the precautionary principle is invoked. We have to act; we don't have perfect information. It's funny--a line that I would get a lot and that you'd find in the literature: We don't have time to dot the i's and cross the t's; we don't have time to get definitive scientific data. People are dying every day. This is the way physicians act. Tell me you are in a car accident, they come in and they throw the body down in front of you--you don't have time to get scientific data; you've got to act. This is what physicians think they have to do. The problem is, from the scientific perspective, and I think it's one of the most profound problems, paradoxes, I see in this field, is if you just take the perspective of a hard scientist, not having time to get definitive scientific data means you don't have time to know if you are right or not. It's that simple. It's like saying we've got to get a man to the moon before the Russians. Although we don't really know where the rocket's going, we'll do the best we can. We have to act, we've got to beat them, but again: How do you rationalize this? How do you make the scientific viewpoint work? Because we know that in science, if we don't have definitive scientific data, you don't have compelling evidence yet it means there are holes that alternative hypotheses could explain what you've seen. It's a tossup at best whether you are right or not, whether you consider the odds against you 50-50 or infinity to 1. But you have to give something to the view, with your point about don't use the word "yet"--I agree with that--but I think the counterpoint has some merit, which is: You use the word "definitive." We rarely if ever get definitive proof in science. Almost by definition it's not definitive. It either supports or contradicts or raises questions. But eventually you lose the ability to come up with hypotheses that can explain the data other than the one you've got. I grew up in physics, a hard science, where you could reproduce experiments perfectly. So, you could have much more confidence about whether or not another hypothesis is going to come along that's much better in explaining what you've seen. It's a fundamental problem. And indeed, you do have to act. I think the key point--and I think this is why macroeconomics is so muddled, macroeconomic policy--in the area of diet and nutrition we do have some hope that we will understand things at the micro level that would inform us about the macro effects. We do make advances; we do get better; we do reject some theories; we do know what's not true. And I think in economics it's much, much harder to be optimistic that we are going to get there. But in the case of epidemiology, diet, and health, we've made some genuine advances in understanding the causal mechanisms rather than just looking at correlation. Is that true? Yes; you can learn from these epidemiologic studies. High blood pressure is associated with an increased risk of heart disease. But you can't learn from an observational study what causes high blood pressure. What dietary, lifestyle pressures actually increase blood pressure. That, you have no idea. So, when it comes to: you believe if you could reduce blood pressure you could reduce heart disease and you do experimental trials to show that you could do that with drugs; and ideally you would do experiments, trials, to show you could do that with dietary interventions; but what you find there is you can get closer to the truth in nutrition and health, but what you find is how closely, how much we hold onto what we've already believed up until now. So, as I've written, there's precious little evidence to support the idea that sodium restriction will reduce blood pressure, salt restriction will reduce blood pressure and hypertension, and so heart attacks; and yet with each year that goes by the effort to restrict salt in the diet gets greater and greater. Made by the Center for Disease Control and various organizations. And the data just get worse. The evidence gets worse, but the efforts to do something about what is a perceived problem get more vigorous. It's a kind of crazy situation, where each time a paper comes out that doesn't support the hypothesis, which is virtually every paper now, the establishment responds by pushing harder to get the hypothesis translated into action. Obviously there's a lot of psychological reasons for that. My favorite tragic example of this is maternal mortality before we understood hygiene and bacteria and puerperal fever. There were two groups. There was one group that said it was something in the air and we had to keep the windows closed. There was another group that said, no, actually doctors should wash their hands before they come from the morgue when they deliver a baby. And it tragically took a long time to establish which of those is correct. But we did advance on that. We figured that one out. But surely if you had been telling doctors for a long time you didn't need to wash your hands and it was really something in the air and it was okay to go to the morgue, the idea that you had been killing women for decades is unbearable. So, you are going to have a very hard time conceding the statistical evidence that shows the contrary. And they did. And that's a very human response. The press release that the AHA will put out some day to flip their nutritional paradigm: We just want to apologize that for the past 50 years we've been telling you to eat low fat, high carbohydrate diets and we now realize, as we should have known 49 years ago, that these diets will actually increase the risk of heart disease, diabetes, obesity, perhaps even cancer; and we apologize for killing your loved ones. Really sorry. And on top of that, now that we've got our heads around right, you can trust us, that the advice we are giving you today is correct.
32:46Let's move on to some of the alternative hypotheses, which I find myself remarkably sympathetic to--which, it scares me. And to get there I want you to talk about what I thought were interesting parallels to economic modeling. You talk about two ways of approaching many of these diseases. One is: each one has a unique cause. There's high blood pressure; the establishment says: That's caused by salt. There's heart disease: That's high fat diet, a saturated fat diet. There is diabetes: That's sedentary behavior. There's obesity: That's you eat too much and you don't exercise enough. So they all have their own, unique--and I love the way you point this out: They're all complicated, so of course we expect the explanation to be complicated. But there's an alternative view, which is that many of these things are connected; maybe there's actually a simple hypothesis that explains it. So, talk about what that simple hypothesis might be, which is sugar and refined carbs in your diet, and what the evidence for that is. The idea is obesity, diabetes, heart disease are all associated. So, if you are overweight you are more likely to be diabetic, more likely to get heart disease if you are diabetic. This is Type 2 diabetes we are talking about, not Type 1. They cluster together in patients, and they cluster together in populations. Hypertension is another one that goes with them, so the current explanation is hypertension is caused by salt; obesity is caused by eating too much. Gout, by the way, is another disease that is associated with hypertension, obesity, and diabetes; so gout is caused by eating too much meat and drinking too much wine; and diabetes is caused by sedentary behavior--because we have to explain we. We can't explain by eating too much, because we have to explain why lean people also get Type 2 diabetes. But the idea that they cluster all together is connected by the idea that there is a state, a condition now called metabolic syndrome. And metabolic syndrome is a disorder of various metabolic abnormalities that are all in effect caused by or associated with this effect caused insulin resistance. So, your cells become resistant to the hormones that your pancreas secretes; and as your cells become resistant, your pancreas responds by secreting more insulin to do the job. And your insulin levels go up. And it turns out that Type 2 diabetes is a disorder of insulin resistance, of elevated insulin levels. And obese individuals turn out to be insulin resistant and have high levels of insulin; and this metabolic syndrome can be thought of as the primary risk factor for heart disease now. So, if you have a heart attack, almost assuredly it's not going to be because you have high LDL cholesterol, which is the conventional wisdom, but because you have these metabolic abnormalities called metabolic syndrome. And they are all tied together by insulin resistance, and high levels of insulin; and sort of disregulation of insulin signaling. So, you could make the argument that whatever it is that causes insulin resistance causes all these diseases. And more importantly, if you could stop doing whatever that is, you could improve all these things. As opposed to, say, losing weight by a frantic exercise program. If you don't change your insulin resistance problem, you are not going to change your risk of a heart attack. Um, yeah; and the idea is that if you do lose a little weight, you will improve your insulin sensitivity, which will then have positive effects on all these associated diseases; but you can't show that anyone can actually lose weight by exercising more. So, now we go to, there's a sort of historical train, context of this as well, which is that going back into the 19th century, the British had these colonial and missionary physicians all over the world, all over the British empire; and a common observation they would make is that you didn't see any of these diseases that clustered together--obesity, diabetes, heart disease, cancer. That they didn't exist in populations that didn't eat Western diets.
37:23So this grew into the 1960s when a British naval surgeon, researcher named T. L. Cleave wrote a book, he called it The Saccharine Disease [?] because the biggest difference, when populations moved from eating their traditional diets to Western diets, the first thing that happened was they started eating sugar and white flour, because you could ship sugar and white flour all around the world without its being spoiled because effectively there are no nutrients in it, so even the rats aren't interested in it. And so whenever population went from eating their traditional diet, whatever it was--an agrarian population, a meat-eating population like the Inuit, or the pastoralists in Africa like the Masai that lived on the blood and the milk and the meat from the cattle they herded, in their traditional diets, none of these diseases. And then when they started eating sugar and flour you'd see these diseases occur. And with a pretty consistent--like first, obesity, then diabetes and heart disease, and then cancer. And now you've got the whole cluster. Of course, you have to be careful. It could be, what you call it--the Disease of Civilization? It was known as the Disease of Civilization, and then that was considered a demeaning term, implying that these people weren't civilized. So then it became Western Diseases. But there's a very commonly believed, and now a consensus of opinion today that these diseases are consistent with Western diets and lifestyle. The Japanese, for instance, having less of a Western diet, have less of these diseases. But there's a measurement problem, of course. When they get developed or they move to Western civilization, they often live longer, so you have to make sure there isn't that problem. There is a measurement issue, so you have to make sure if they didn't diagnose it correctly. Into the 1970s, 1980s, when researchers are doing carefully controlled studies between populations--like even with the Alzheimer's, the famous study where the same researchers diagnosed Alzheimer's in African-Americans, I think it was Cincinnati, somewhere in the mid-West, and then they went to Nigeria and did it there. So you had the same researchers with the same diagnostic equipment and the same diagnostic criteria doing the diagnoses, and you got significantly higher rates of Alzheimer's here than you got in an African population in Nigeria. So, pretty much a similar ethnic background, separated by 300-400 years. One study. Than any specific age you had there. So what would it be about the lifestyle that would affect that. And like I said, you tie this with the sugar and flour idea. So, that's Cleave. So what's Cleave's punchline? Well, Cleave's punchline is you add sugar and flour to a population, you get diabetes, heart disease. Obesity--and while Cleave is doing that, in the 1960s, you have a revolution in the understanding of the accumulation of fat and fat tissue. It took a couple of technologies, one that was discovered in 1956, to measure fatty acids, the concentration in the blood stream, and then the radioimmunoassay to measure hormone levels accurately in the blood, which was published in 1959-1960 by Rosalyn Yalow, ended up winning the Nobel Prize. Berson had died by then. With these technologies you could determine what regulates the accumulation of fat in fat cells; and it turns out to be the hormone insulin. And as you study insulin--in fact one of the first measurements that Yalow and Berson did was insulin levels in Type 2 diabetics and obese subjects and it turns out that lo and behold they had high levels of insulin. We expected that Type 2 diabetics would have low levels just like Type 1 diabetics do. But they don't. And it's the Type 2 diabetics that tend to be obese. So now you've got this hypothesis that all you have to do is increase insulin levels and you store calories in your fat tissues. And in fact independent of how much you are eating or how much you are exercising. So, just like if you had a child that was growth deficient because they are lacking growth hormone, you could give them growth hormone and they would grow--you will accelerate their growth and cause a sort of compensatory increase in appetite because they have to bring in the calories they need to build all these new body structures, the idea was all you had to do is increase insulin levels and you'd increase fat formation; and this taking in more energy than they expend would be a compensation rather than the cause. One of the arguments I make in my book is that in any growing system--in nature, in effect, growth is the cause and this taking in more energy than you expend is the effect. So if you think of a growing child, the child gets taller, heavier; and it has to take in more calories than he or she expends. But it's not taking in more calories--she's not growing because she is taking in more calories than she expends. She's growing because she is secreting growth hormone, and the growth hormone is stimulating the secretion of insulin-like growth factor, and that's driving growth and the sort of positive energy balance is the effect, not the cause. If you have a tumor growing--when you have a tumor growing it's taking in more calories than it expends because it's getting bigger. But that's not the quiet growth. It's growing because it's got all these defects and tumor-suppressor genes that are driving it to grow, and the positive energy balance is an effect. And the argument I make is this is always the case. And so by the mid-1960s, you should have had a hypothesis of obesity that it is caused by this disregulation signaling that is resulting in elevated levels of insulin driving fat accumulation. And the problem is that by that time we had decided that dietary fat causes heart disease. And see, insulin is secreted primarily in response to the carbohydrate content of the diet. The more carbs, the more refined carbs, and sugars in particular seemed to cause this thing called insulin resistance. So, you've actually got a hypothesis that really nicely fits with what Cleave documented 80 or so years that documented research in these populations--that you change the nature of the carbohydrates they consume; they start secreting more insulin; they become insulin-resistant; they get fatter; they get Type 2 diabetes; they get heart disease. But it couldn't be reconciled with this idea that dietary fat caused heart disease. And it couldn't be reconciled with what most lean people sort of naturally believed to be true--that they are lean because they eat in moderation and exercise a lot. Which can be true. Well, actually, I don't think it's ever true. And that's the point. I would argue that body types are genetically determined, and the thing that can change the genetic determinant of obesity would be how you respond to the carbohydrates in the diet, how you well you could tolerate them in effect. So a lean person is a person who can tolerate a carb-rich diet and has the energy to exercise; and a fat person is someone who can't and because of the insulin signaling they end up with too much fat tissue and now they don't have the energy to exercise because all that energy is being sucked up by their fat tissue.
45:34So, I want to cut to the chase, because we have a lot more to talk about. This is, in a way it's a punchline of the first part of the conversation, which is that you come out and you wrote a beautiful, interesting NYTimes article on this, particularly on the role of sugar. So you come out and you say: Look, for a long time we got seduced or waylaid or whatever it is: we got too focused on this fat as the source of our ills, in our diet. And in fact it's not fat. It's sugar and it's refined carbs. And I have to confess--and my problem is not as bad as yours--when I hear that, there's something so elegant and beautiful about it. So, although there is something rich and interesting that each disease has its own cause, this idea that there is one cause, and all I have to do--if you are listening out there, there is an incredible power to this--all I have to do is cut back on sugar and refined carbs and I'm going to reduce my risk of obesity; I'm going to reduce my risk of heart attack; I'm going to reduce my risk of cancer, of diabetes. That's a very seductive idea, particularly because for the last 10 weeks I personally have started cutting back on sugar. And carbs, refined carbs particularly. I want to believe that was a good sacrifice. I was doing the right thing. And I have lost a lot of weight, by the way. So, this small data set, one observation--me--I'm thinking: This might be true. And now I read your book and I think: Wow! This is the truth! But of course, you, as the author really have a problem because you have a huge stake in this argument right now. What's your evidence that you are not just falling prey to the Ancel Keys and other folks who have made the same mistake? Um, well this is what's interesting. One of the problems with this hypothesis--you know, quacks always argue what you just describe. One thing, I'm going to sell you a book about, by the way. And I've got tapes, too. I've got DVDs. You get a t-shirt. Yeah, yeah, yeah. And one of the problems with this has always been--you know, physicists love unified theories. That's all they want is a unified theory of gravity, quantum physics, the universe, beautiful; everybody's in the same business, that's why nature works. In medicine, it's quackery. We don't want a unified theory of disease. And actually one of the responses you get, I get all the time, is obesity is a complicated, multi-factorial disease. And if I argue that you should begin with Occam's razor--don't complicate hypotheses beyond necessity, they'll argue that we're far beyond that. That only means matters in physics. You are not a scientist. Come on. You are a journalist. You are making it so easy. That's the easy way out. It's deeper; it's more complicated. It's interesting. Because of course if you think that the planets have to travel around the sun--and this is going to get really tricky now because, and I'll explain--but if you believe that the planets travel around the sun in perfect circles then you can't explain any observations of retrograde motion and other things; and now you have to start adding epicycles; and you say, it's complicated. And it's complicated because your fundamental theory is wrong. Now here's where it gets tricky. We didn't talk about this. My first book, I looked at Cern, the big physics laboratory in Geneva. This was back in 1984, 1985; and I watched some excruciatingly smart high-energy physicists discover some non-existent particles. And then I watched the experiment sort of schism into two groups, those who were concerned with covering themselves and proving they were right all along; and the other group that felt they actually had an obligation as scientists to get the truth out there, even if it made them look silly--and they would actually look better if they said: Here's how we screwed up, and we apologize and if they never said that. And then I wrote a book called Cold Fusion, on the great scientific fiasco of the 1989-1990; and when I came out of that I said: Look, anyone, when I was talking to journalists--one of my rules of writing now is whenever someone invokes Galileo as a personal role model, he's a quack. Meaning? Give me an example. Well, the joke is in effect I just invoke Galileo by using epicycles and perfect circles. But the argument is literally the paradigm is wrong. And as long as your paradigm is wrong, if you believe obesity is just about eating too much and expending too little, this energy balance idea, it's all calories-in/calories-out, you can't get the right answer. And the counter argument is pretty simple. The counter argument is: Look, we are just looking at hormonal enzymatic regulation of growth, just like we would on any biological system. And if we just pay attention to the hormonal enzymatic regulation of fat tissue, then your problem is carbohydrates; and that should be your null hypothesis that you would have to refute. In defense of what I've done, the way I got into this, I was doing this piece for the NYTimes that I pitched back in 2001 on what caused the obesity epidemic. At that point in time, the obesity epidemic was new enough that our consciousness of it was 3 or 4 years old and it was new and interesting; and you could imagine what caused it; and I had come out of doing what was basically a year-long investigation of science about the evidence supporting the low-fat-is-good-health notion. And there were two things that changed back in the late-1970s to the mid-1980s, viable hypotheses why obesity levels started shooting upward just at that period. And one was that high-fructose corn syrup came in. And actually, our total caloric sweetener--high fructose corn syrup is just another kind of sugar--and our total sugar consumption started to increase. In the past, whenever sugar consumption had gone up significantly, so had obesity and diabetes rates. That had been from about the 1870s to the 1920s, was the last time we'd seen a significant increase in sugar consumption. So, that was one hypothesis. And the other was that we institutionalized this idea of a low-fat diet being a healthy diet. When you reduce the fat, you replace the fat in the diet with carbohydrates: the food guide pyramid. The base of the food guide pyramid, the staples of our diet, are pasta, potatoes, bread, rice, starches. Cereals. All these foods that my mother's generation believed were inherently fattening. But because of the low-fat dogma they had been transformed into heart-healthy diet foods that we were supposed to eat all the time. So, simultaneously, while we start increasing our sugar consumption, because we don't realize high-fructose corn syrup is sugar, something the corn refiners worked hard to make us think and they succeeded, we also started eating less fat and more of these formerly-believed-to-be-fattening carbohydrates.
52:56So now, I'm doing this story and I'm looking into this and I come upon five clinical trials that were done and had been published yet, even though the results had been reported in conferences--so I could discuss them. And they were all clinical trials of the Atkins diet. The Atkins diet being a low-carb diet. The Atkins diet is a low-carb, high fat, high saturated-fat diet. In which you can eat as much as you want, by the way--just don't eat carbs. So, it's called an ad libitum diet--you can eat as much food as you want, so long as you don't eat carbohydrates. It's lunch; I'm going to have a chicken. A whole chicken. Lobster Newburg, a 2-pound steak; and breakfast is going to be 5 eggs with bacon and sausage. And we had two hypotheses going in here. One is that obesity is caused by eating too much. So, if you want to lose weight you have to eat less, exercise more. And the other is that heart disease is caused by a high fat, high saturated fat diet. So, now we do a clinical trial where we put half our subjects on an AHA Step One diet--this is a low-fat, low calorie, low saturated fat diet. So, basically it's a small portion of skinless chicken breast with some brown rice for lettuce and lunch; and breakfast is a bowl of Special K with skim milk and a banana. The routine--we all lived like this for decades. Speak for yourself! And the other is this high fat, high saturated fat diet, this Atkins diet where you could eat as much as you want of all these foods that are supposed to kill us. And in every one of these 5 studies, not only did the Atkins diet group lose more weight--remember, they are allowed to eat as much as they want--but they lost more weight than people who were calorie-restricted, who were told only to eat 1400 or 1500 calories. Their heart disease risk factors improved as well. So, the diet that's supposed to kill you and make you fat, if you believe these twin hypotheses that we had all come to believe in fact was the diet that made you lose more weight and seemed to increase your risk of living long enough to see your great grandchildren graduate from college. The counterpoint to that, though, as someone who has tried a zero-carb diet--not what I'm experimenting with now--but it's not sustainable. Your body gets real upset that it doesn't have many carbs, and then you eat one potato chip; and then the next thing you know you've finished off three bags and it's over. In fact, you usually end up weighing more after you are done with that. So, those 0-carb diets, you have to be on them forever and we can't do it. So, maybe it's not such a good idea. Not only can [can't?] we do it, but we've only run our tests out for two years, so the dietary fat, saturated fat's going to give you heart disease, and maybe it's going to give you cancer, and we've only done the trials for two years; so the best we can say is these are short-term weight loss diets, at which point you are supposed to add the carbs back in. But the counterargument for me is: You are supposed to quit smoking. I'm a writer and a freelance writer; I'm a neurotic, I used to smoke; I tried to quit every day for about 20 years before I finally succeeded. I had periods where I succeeded for 3 months, 6 months, but nobody who quit smoking for 3 months and then falls off the wagon and goes back to smoking--it doesn't affect the argument that cigarettes cause lung cancer. Good point, fair enough. But in this world, it does, because we're going to use every argument we can to shoot these down. So, a common argument you hear is: It's hard to stay on the diet, therefore, carbohydrates don't make you fat. And that's like saying: It's hard to quit smoking cigarettes, therefore, cigarettes don't cause lung cancer. It's back to: We have to do something, and since this doesn't help us do something, we may as well ignore it. You find all these rationalizations. I think diet is really like a religious, maybe economics too. It is. We talk about it all the time. It's dogmatic. It's religious in a certain dimension. In diet we also have these issues--meat-eating is an issue, abusing animals is an issue; vegetarians and vegans who argue that these arguments; I mean now it's environmentalists who are involved with it. Supposedly meat-eating contributes as much to global warming as driving 1969 Chevy Camaros without tailpipes. The argument again is I know meat-eating does. But that's not the debate. The debate is literally what makes us fat, what causes heart disease, what causes diabetes; and if you are already obese and diabetic--like, let's take an obese and diabetic mother. And she's going to give birth to children who are predisposed to be obese and diabetic. What do you want her to do for her kids? Because starving them and getting her kids to run around the track an hour a day is just torturing the kids, and it's not actually reversing the problem, because the problem isn't that they eat too much and exercise too little. Those are just the facts. Imagine you could find cancer patients who are bedridden and don't exercise at all. You are not going to run them around the track because you think sedentary behavior must be the cause of their cancer. You've got to give them the right; you have to diagnose the disease correctly; you have to identify the cause correctly; and you have to act. And if the act is getting rid of the carbs that cause the problem, then unfortunately these people are going to have to eat more meat. More animal products. Because when you say you've experimented with this low-carb diet, you can't do that in a vegan, vegetarian diet. You can't do it in a vegan world. It's virtually impossible, if not impossible in a vegetarian world. You can drink olive oil all day. Soy. And soy. It's tough. You are pretty much stuck with animal products. It becomes this ethical issue, this religions issue, this environmental issue. And it's fundamentally the issue--I argue, let's get the health right. Like, if somebody knows they are going to doom their kids to a life of obesity and diabetes because they are going to make them vegetarians or vegans, then that's fine so long as they understand that they are not doing their kids any favors. That's right.
59:43The question, though, is: Is that true? Let's talk, we have a little time left, about where you think we stand with respect to this hypothesis. So you said: Well, persuasive to you. Five studies, clinical trials, of the Atkins diet, and overwhelmingly to show that it wasn't calories in or calories out, the energy balance model. It was whether you ate carbohydrates or not. Now, it's true that we all are prone to confirmation to bias; it's true that scientists have their pet hypotheses; it's true that it's hard to admit you are wrong. But there's 5 clinical trials. Well, actually, there were 5 in 2001, 2002. There's dozens now--probably there might be 100 now. Now they are done regularly. So, the question is: there's a nice post on your blog about a speech you gave in front of epidemiologists where they said: Do you think we're all idiots? So, are serious scientific people who had the--as you call them in your post, the wizards, people who have got the right credentials and who appear to be scientific at least--aren't they moving in your direction? Well, some of them. I'd say probably 9 years ago I probably had a thousandth of a percent of the medical research community on my side; and now I might be up to 1% or 2%. That's a pretty dramatic increase. So what do the other 98% say when you wave your evidence? Well, the other 98% don't pay attention to my evidence. It would be an interesting question. As a journalist, the problem I have is what I call the: It's crap; I haven't read it problem. This was after I wrote this book on high-energy physics. It turned out to be an exposé because I ended up watching these physicists discover non-existent particles and the Nobel Prize winning head of the experiment trying to cover it up. And one of my friends in the physics community asked one of the chief kind of crony of this Nobel Laureate what he thought of my book. And this fellow said it's crap; I haven't read it. It's the parallel explanation for when you put forward a hypothesis and people say: It's obvious and it's wrong. It's a fancy way of saying it makes me uncomfortable. From their point of view: I know enough about it, I know enough about what this fellow Taubes thinks; and if I know who Taubes is, and most of them do by now and I'm kind of an annoyance, I know enough to know what Taubes thinks; I saw that magazine article he did 10 years ago and I don't have to know any more. You are just a fancy version of a quack. A journalistic version of a quack. And I've had this argument with people. I've said to people, it's similar to think about it: Imagine you are an atheist, or a devout Catholic and somebody suggested that if you read Richard Dawkins's book. It's going to convince you that you are wrong. I'm an atheist, I read a book, some journalist has written a book that's very compelling evidence that God exists. Even if I was open-minded and said: Okay, I'm going to read it, because I'm open-minded and it has gotten good reviews and people I respect think highly of it. And I would bet you I am going to get maybe a couple of chapters in and I'm going to think he's not telling me the whole story; that he's twisting this and spinning that, and this is a rhetorical game; and I'm not going to get through it. I'm going to close it. The funny thing is I have a friend at Yale, happens to be a born-again Christian. I said this to her, and she said: Oh, you should read this book written by a journalist, compelling evidence that God exists. I said: I should; and I put it on my Amazon shopping list, which basically means to consider buying when my 3-year-old goes off to college in 15 years. I can't do it. Even though this is a very smart woman, married to a very smart researcher; respect them both. Well, because you've made a big investment in your worldview; you've thought about it a lot, perhaps--I'll assume you have--and you like it. And there's another piece of this, which we haven't talked about. I happen to be a religious Jew, in reality; and I'm very aware of the role of doubt and belief; and it all makes me think about economics as well, because we have the same issues as we talked about earlier. And it's part of our identity, too. It's not just: Well, I weighed all the facts and I came to religion; or I weighed all the facts and I came to sugar and carbs. It becomes part of the people you hang out with, the uniform you wear out in the world, the club you are in, your self-identity, your self-perception. And it's extraordinarily costly to change those things. And it's painful. And most of the time we don't do it. Before we started this interview, we talked briefly about Francis Bacon. We are not wired to do that. Once those perceptions are created in our brain, for whatever reason, we're wired to in effect reject evidence against it. I don't actually know what evidence could be presented at this point of the existence of God that I wouldn't preferentially interpret as me having gone crazy. Right. I understand that. And that's why we are not going to debate it. And we don't debate it generally. I don't say to you: Let me make the case for; and you say: Let me make the case against. And we yell at each other. We both realize that's not a fruitful experience. But it's interesting to me that in economics, or in epidemiology, we are doing the same thing, really; but we are masking it--and this is the Hayekian idea of his condemnation of scientism--we are masking it in the language of science. We are using data. And as a result, what we are really debating often in these cases is dogmatic. Or ideology, or philosophy. But we pretend we are debating about which study is the right study. And my claim is not to say: Oh, therefore evidence doesn't matter. That's not the point. My point is: Have the debate, but be aware in yourself and in your opponents' possibly being correct, and in possibly being misled and wrong, be aware that you've got these deep-seated biases, priors, etc. I love your quote of Richard Feynman, which is: The first principle is you must not fool yourself, and you are the easiest person to fool. I think a thoughtful person, whether you are dogmatic about carbs and sugar, whether God exists--all those things--you've got to look into your soul. Or your mind, or your heart--whatever you want to call it. And admit that you are not a truth-seeker in some ideal, romanticized way. You've got this baggage. We all have it.
1:07:32Well, this is what's interesting. Because in one of the arguments I would make: it's just--I don't have to convince people I'm right. I just have to convince people it's worth testing rigorously. And in one of my challenges, I think one of the reasons we got into this situation in nutrition, is because it's been a pseudo-science since the 2nd World War. That in effect there was a culture of science that had grown up out of Europe prior to the second World War that was very highly honed and rigorous and skeptical; and it kind of vanished with the war. And it feels like physics, we embraced all these European physicists who got chased out of Europe; we had atom bombs to build and a cold war to fight. And actually many of the leading figures in high-energy physics in the mid- to late 20th century were European émigré or their students. Because they had this culture of science. But in other fields, particularly in public health and probably economics as well, we didn't want anything to do with these people. So, the ideas of rigorously testing hypotheses, of not acting on faith, not allowing an idea or an assumption to sneak into your world view masquerading as a fact because once it does it's going to pollute everything that comes afterwards. I have to convince people first of all that this idea is worth testing; and then you've got to test it right. Because if you don't test it rigorously, you are going to get the wrong answer again. Are you going to come up with these kind of vague, nonsensical bodies that you could interpret any way you want, like Rorschach tests. It's actually worse than that in economics. I believe it. Because I can imagine how to test. I hope some day to raise the money. We're in the process of creating a non-profit to raise money to do the kind of studies I think are necessary. But in economics, the more certain you are, the more confident you are, the more dismissive you are of your ideological enemies, the bigger your platform. You talk about the change in the culture--in 1970 there were two economists who could make a living, making some money at all, talking about policy differences and ideology. They were Paul Samuelson and Milton Friedman. They each had a column in Newsweek and they yelled--not yelled, they were very polite people--but they disagreed each other and they were able to have a nice platform because there was room for two of them, maybe. Now there's room for hundreds because of the blogsphere, because of the explosion in information. And the competition is to be loud and sure of yourself. That's how you get to the top. And of course, you could still be right. I don't want to suggest that being loud and sure of yourself proves you are wrong. It doesn't. But it does change the culture of the profession. The very first book I wrote about physics, Nobel Dreams, the physicist I was writing about who ran this experiment was a very Machiavellian Italian physicist who--he taught at Harvard and he worked in Geneva and he commuted between the two weekly; and he had been on on virtually everything he had done. But he'd just about demonstrated how he could move up to the very top of his field by claiming a discovery and then kind of moving on to the next experiment and leaving better scientists to clean up the mess after him. And this has been a common theme in everything I've written about. In making declarative pronouncements based on preliminary data, and fighting viciously to get people to believe in you, you can do very well for yourself in these careers. Nobody moves forward by spending their career checking other people's work to see if it was right or not. And actually one of my favorite lines from physics was from this Nobel Prize winner Sam Ting at MIT, who said to me, if I can get this right: To be first and right is good; to be first and wrong is not so good; and to be second and right is meaningless. That's deep.
1:12:04I want to ask you two more things. I think there's an important difference between economics and epidemiology, which are this cell-based science where you can actually see what's going on and eventually, it's imaginable that the causal relationships will be irrefutable, or at least powerful and persuasive. Do you think that's going to happen? Do you think it already has happened? In nutrition? Yes. I don't know. That's a good answer. Again, as long as the paradigms are wrong, what they do at a cellular level doesn't really matter. And you've still got these huge--I'm thinking just a week ago there was an editorial in the New England Journal of Medicine, by this Walter Willett of Harvard, who is kind of the biggest name in nutritional epidemiology, and this fellow David Ludwig, whose claim to fame more than anything is having written a column in the Journal of the American Medical Association (JAMA), and now he's perceived as kind of an authoritative voice. And they are arguing that the dietary guidelines have got it all wrong. And they are saying: We have to do much better science. And what they consider good science is these long-term, observational perspective studies considering capable of determining cause and effect. And the same studies--like I said, if you generate a hypothesis out of them and then test the hypothesis in a real experiment, you get a different answer. So, right now, the field of people who are professing to be the sort of voices of reason, I would consider pseudo-scientists. So, we are moving in the wrong direction. But it's not so much even with nutrition doing cellular experiments, being able to see what's happening at a cellular level. You can do if you think hard enough and you are willing to spend the money, pretty rigorous controlled trials with humans. That's what I want to raise money to do. If you can treat humans like laboratory rats it's not easy; it's expensive. But you can do them where in effect you feed them specific foods to make them eat it and see what happens over a long enough time period that you could at least get significant changes in well-established risk factors for disease. You've also got to monitor them, because you've got to make sure they don't sneak out and get that ice cream fix. Well, that's one of the things that makes it expensive. You have to do it in kind of a closed environment so you could demonstrate so they are eating what you told them, what you gave them. This is not the culture of the field at the moment. Like I said, it hasn't been for 50 years. I think that can change, but there I'm probably being naive. But we'll see. The last thing I want to ask you is somewhat like economics, but not really. There's the administrative, institutional, government driven, funding driven, reputational driven beliefs that come out of the scientific community. And we've talked today about how they've swung back and forth over time, sometimes ways we think are okay; sometimes ways we think are encouraging. But there's also what everybody believes down on the ground, people like me. So, if I like your theory, I might ignore all the so-called institutional wisdom or so-called experts, and I'm just going to keep going with my low-sugar, low-refined-carb diet, and I'm going to get my own little test; and it's one observation, and it's flawed. But I'm going to make my own decision. And I've got the placebo effect working for me, too in the background, by the way. So, even if it doesn't work, the fact that I believe it might help me anyway. I've got nothing against a good placebo effect. It's awesome. And you are selling a lot of books, I'm sure. People like this; and as you point out in the beginning of the book, which I found utterly fascinating, consumption of carbohydrates and weight loss has been folk wisdom for a long, long time and people believed it and acted on it and lost a lot of weight; and maybe you'll win anyway. So, what's your reaction to that? That's one possibility. I think this idea that refined carbs and sugars are bad has already been accepted popularly. When people do diet trials now, low fat diets, they'll still give advice: Don't eat white breads, don't eat starches; you are supposed to eat whole grains and what are called low-glycemic index carbohydrate foods--these are foods you digest slowly so that it's all in effect suppressing insulin levels. I think will be accepted. I've had people say: Oh, you are not saying anything we don't already know. There are bad carbs and good carbs, bad fats and good fats. The battle is getting accepted that fats are not only good, but they might be necessities; and they might be the more you eat, the better. And that's going to take more time. But it is true that everyone can do one experiment here. When it's coming to weight--that's how I did this. I did an experiment 10 years ago; there was actually an economist named Andrew Lo [?], MIT. He runs a laboratory for financial engineering, and I was doing this story on dietary fat for Science; I was also doing a story for Discover on the economics of the mathematics of the stock market; and he said: Oh, if you are doing fat, you've got to try the Atkins Diet. You know, it's a high fat, high saturated fat diet; he had a colleague at Wharton whose father lost 200 pounds on it. You know, he said he lost 40 pounds giving up white rice; and I should do it. And I tried it; and you lose weight effortlessly. You are not hungry. Anyway, the question is you don't know if you are going to kill yourself. You don't know about the longer run. And that's what you need the clinical trials, just to have this faith that while you are sitting there eating your eggs and bacon in the morning you are not indeed clogging your arteries. Or doing something else. You are losing weight but there's an unintended consequence, which is what happens. And you just never know. I'm a 55 year old man; I have ailments, and I never know. Are they caused by my diet? Or would I be in worse shape if I ate what I used to eat? Just like stimulus. It either made the economy worse, or it just wasn't big enough. Hard to distinguish between the two. Unfortunately, that's for our country. For you as an individual, I care about you, too, but we ought to be making those decisions on a somewhat scientific basis, it seems to me. One of the arguments I've made when I lecture in medical schools is: Look, we've got this obesity and diabetes epidemic going on; I mean, diabetes rates, diagnoses, have tripled since 1980. In 30 years. That's an enormous increase. No matter how much of the diagnostic effect or changing diagnostic criteria, something has changed. And it's changed during this period that you've insisted you know what causes obesity and what causes diabetes and what kind of diet we should eat. Can't you accept that maybe you got it wrong? And then they say: Yes, maybe we got it wrong, but it's multifactorial and complex. Do you find it challenging to this idea, by the way, that despite these rises in diabetes, etc., lifespan keeps getting longer? Every year. It's an interesting finding. The thing with lifespan is it's so, you know if you think about the number of bypass surgeries you do every year, the number of, the extraordinary increases in--what you have to establish in effect is that you are actually, you know, how much of that is lifestyle. You need a counterfactual. And you don't know. I've often wondered if I'm going to blame obesity, heart disease, diabetes, cancer, etc., on sugar and refined carbs, what are you going to die from if you don't eat it? And then my theory is: Okay, you are in perfect health, you go to your grandkid's third birthday when you are 75, you have one ice-cream cone and your heart blows up. Because you are not inured to these things. And then everyone blames it on your low carb diet that you've been eating until then. But you still have to do a randomized, controlled trial. The fact that we are living longer could be just the wonders of medical care and hospitals and emergency room, emergency medical services. Yes. But economists tend to look at fundamental nutrition; could go the other way; they tend to explain lifespan by the fact that nutrition has improved dramatically. Just in terms of raw calories. We are fat because food is cheap. We use Occam's Razor, too. Then again, the argument I make is: Yes, food is cheap; but why is it that poor people are fatter than rich people, and why is it that you can find populations where levels of obesity like we have in the United States today that are unimaginably poor. Fair enough.

Comments and Sharing



TWITTER: Follow Russ Roberts @EconTalker

COMMENTS (36 to date)
Nipper writes:

Gary Taubes has done a lot of great deconstruction, but his alternative carbohydrate hypothesis is somewhat lacking - counterexamples can easily be found. Consider the mice who have massive circulating insulin levels and are not obese, or the Peruvians and Kitavans who eat massive amounts of high-glycemic index tubers and are remarkably free from western disease. The macronutrient ratio is not too important, and healthy-fat eating populations can be found as well. The Inuit, certain Pacific island cultures, obtain much of their calories from saturated fat and are also free from western disease.

The Weston A. Price / Paleo 2.0 approach toward studying the differences between western/ancestral diets is more fruitful. (See Kurt Harris's Archevore diet.) The macronutrient ratio is not the key difference. These diets are far richer in micronutrients than the typical western diet and minimize fructose (Taubes is right about this), vegetable oils, and nutritionally poor, improperly prepared grains that contain high levels of antinutrients and substances that provoke autoimmune responses. They are also low-reward.

Weston A. Price/Paleo diets happen also happen to be high in fat and animal products, but not because carbohydrate is inherently sickening.

Dallas Weaver, Ph.D. writes:

Russ,

I like your analogy between macro-economics and political nutrition.

In nutrition, there is a huge amount of real science that can actually generate somewhat optimal diets for animals. The real complexity is huge and multi-factorial. The solutions often use linear programing to solve several thousand simultaneous equations with detailed analysis of all the fatty acids, amino acids, all types of carbohydrates, all minerals of all available feed ingredients and the availability of each component as a function of processing (cooking), along with the cost of each feed ingredients. There are published international feed tables providing data (complete analysis) on everything from corn protein extract to feather meal and 50 types and grades of fish meal.

The level of science is advanced enough to create totally vegan diets for carnivorous marine fish, where the fish actually grow faster on less feed than on fish based diets.

However, the field of human nutrition seems to be one of black magic with only glimmerings of knowledge, tons of myth and little utilization of the vast amount of animal information, where we do have hard experimental evidence. It gets even worse when we try to go from basic nutrition science to political nutrition.

The problem of human nutrition is multi-factorial and complex and the field is filled with simple solutions that probably aren't true, including the "refined" carbohydrate/sugar hypothesis. To give an idea of the complexity, starting with raw food material, their composition will change with time (fatty acids will oxidize, for example flax seed oil -- a good food -- will change into linseed oil -- a toxic wood preservative). Then we cook the material to detoxify some of the anti-nutritional factor and increase the digestibility, while denaturing proteins and making some very complex and sometimes toxic or carcinogenic chemicals. After we eat it, our digestive enzymes turn it into smaller components. Table sugar becomes almost the same ratio of glucose and fructose as high fructose corn syrup and a potato becomes glucose. Then the microbiome (bacteria) in your gut work things over, producing a whole range of chemicals with variable impacts. You will get different results with the same diets with different gut bacteria.

From an individual human standpoint, your individual metabolism depends upon your genes and your epigenetic status that helps determine what gene is and what is not turned on. The epigenetic status will change with nutritional history. For example, one of the low cal life extension diets that take your BMI down to the 18 range and makes you ask whether the added life is worth living, with no energy or good food, will change the regulation of hundreds of genes. The Auschwitz diet with vitamins will improve your metabolism and health.

We do understand how the Atkins diet works. You screw up the metabolism enough and you won't get growth or even maintenance. We also know that a small dose of long chain fatty acids before eating a bunch of carbs will turn on GLP-1, which will shut down glucose production in the liver and result in a lower peak glucose from the carbs (this is how the diabetes drug Juvenia works, but a 1/2 teaspoon of peanut butter 30 min. before eating is a lot cheaper).

My interest in human nutrition and interaction with exercise is the result of a 4X bypass and metabolic syndrome. Having spent three decades raising aquatic animals, I was forced to understand a small part of the complexity of animal nutrition and develop a lot of contacts with researchers in the field.

Brit writes:

I have to question some things about this podcast. For example, he says that, in colonial times, incidence of obesity, diabetes, heart disease and cancer were not found in native populations. This seemed like quite a claim to make. While I can understand obesity and diabetes (since they seem to be related). I do question how often colonial-era doctors were diagnosing diabetes, heart disease, and cancer.

Also, there's this: "The word cancer came from the father of medicine, Hippocrates, a Greek physician." ... "The world's oldest documented case of cancer hails from ancient Egypt, in 1500 b.c. The details were recorded on a papyrus, documenting 8 cases of tumors occurring on the breast."

Obviously the ancient Greeks and ancient Egyptians weren't eating a Western diet.

Jacob writes:

Nipper, I agree that the macronutrient ratio is not too imporant. However, I'd like to point out that the tubers the kitavans consumed was not "high glycemic index" foods like potatoes or bread. Nor did the Kitavans have a high caloric intake.

Another point that must be observed is that both the Kitavans and the Japanese have ingested a lot of long chain omega-3 fatty acids, especially DHA, from seafoods. It is conveivable that a reason for insulin resistance is simply that the cells in the body is too stiff due to too high intake of saturated fats AND sugar. Excess sugars, especially fructose, is turned into saturated fats. Since DHA is super fluid, just 1000-2000 mg per day of DHA may suffice to change this condition even in the presence of lots of saturated fats and fructose. Unsaturated olive oil, may then provide no advantage over butter.

Personally I think the high glycemic index theory is somewhat questionable, but I do feel grains are very problematic, especially whole grains, even if prepared properly. Potatoe, being a root vegetable, I think is far better, at least if the outer skin is removed. Even Adam Smith points out the superiority of potatoes to oat meal (in terms of strenght and beauty) in "the wealth of nations".

An interesting study showed that over time almost all the rhesus monkeys in a colony developed diabetes when they had unlimited access to monkey chew. Yet in nature periodically these same monkeys have free access to fruits and do not develop diabetes. The monkey chew contained lots of whole grains, synthetic vitamins etc, not unlike the typical "healthy" breakfast cereals, and was very high in fiber. However the fiber was the insoluble type. Tubers and fruits provides mostly soluble fibers, and traditional people remove the outer toxic skin with the anti-nutrients and the insoluble fiber.

Staffan Lindeberg's "diabetes study" published in 2007 (http://www.springerlink.com/content/h7628r66r0552222), showed that diabetes was practically reversed over 3 months on a diet with unlimited lean meat, fruits and vegetables. Lindeberg noted that those that ate the most fruits (up to 1,5 kg daily) had the most weight loss. More interestingly Lindebergs research shows that calorie intake is spontaneously restricted on such a diet and that it does alter the leptin hormone that controls appetite. So I would recommend that this approach is treid instead of reducing carbohydrates.

Finally, one may say a lot of things about Keys. But at least he lived to be 100 years old. Now Dr Wolfgang Lutz who for decades promoted a 72 gram carb per day diet, did also live to be 97 years old. But Atkins only lived to be 72, and Blake Donaldsson, the proponent of a meat only diet, lived only to be 62 years old. We also know that inuits are fairly short lived. Perhaps these figures do not mean anything, but my feeling is that too little carbohydrates reduces longevity.

Nathan writes:

@Brit,

A "Western diet" essentially means a grain-based diet. I'm not sure if ancient Greeks and Egyptians were refining sugar, but they surely were eating a lot of bread.

The headwind that Taubes and the paleo movement is fighting isn't just the late-20th century fat aversion championed by the United States govt; rather, it is an increase in quantity and decrease in quality of food (particularly for poorer people) starting ~10,000 years ago with the move away from hunter-gatherism and towards agriculture. Kind of a nutritional Gresham's law.

JRo writes:

Thanks, Gentlemen. Well done.

I was reminded of a 2009 BBC interview of Wallace Broecker, who supposedly coined the term 'global warming', in which he told how he had gotten his doomsday predictions of global cooling spectacularly wrong in the 1970s. Apparently, instead of modifying or dampening his predictions, or recognizing any limitations in himself or in his science, he then went exactly 180 degrees out and spent the rest of his career pitching a doomsday warming scenario.

Now, think I'll have some guilt free ice cream.

James Gambrell writes:

One point that wasn't brought up was the issue of effect size. Even when effects are found for diet, the effects are usually small. Even if I knew for certain that forgoing french fries, potato chips, etc would increase my lifespan by a couple years, would I really give them up? No, its far too high of a daily cost. I think most American's feel this way, the effect of a dietary change needs to be quite large for them to change.

midnightmagic writes:

The Inuit are not short-lived, except lately, because their diet is no longer their traditional diet. The other commenters here are neglecting to mention or perhaps just ignorant of the fact that the fat content in high-fat diets must be moderated by fat type. The Inuit ate whale blubber, with is about 70% monounsaturated, and 30% omega-3 polyunsaturated. Inuit eat large amounts of seal as well. Seal blubber is also a source of omega-3 polyunsaturated.

Why do I keep mentioning polyunsaturated fats?

These studies appear to suggest that diets with high levels of saturated fats (such as those found in fat-heavy steaks as you recommend) are associated with increased cardiovascular events. Specifically, these studies and other related ones I've been reading in my evenings suggest that high-fat caloric-unrestricted diets are effective for losing weight, but that a significant portion of calories from saturated fats in specific should be replaced with poly-unsaturated fats for a healthier heart.

1. Samaha F. Effect of very high-fat diets on body weight, lipoproteins, and glycemic status in the obese. Current Atherosclerosis Reports. 2005;7(6):412-420.

"Conclusion: In summary, the available prospective studies suggest that weight loss may be as successful with caloric restriction on a diet that allows a high proportion of fat as one with a low proportion of fat, at least over a 1-year period. Nevertheless, dietary fat composition yields significantly different effects on lipoprotein levels and on cardiovascular events. Diets high in saturated fat fail to decrease total cholesterol, even with significant weight loss. Both saturated and unsaturated fat increase HDL cholesterol levels. However, in clinical tudies, increased saturated fat consumption is still associated with an increased incidence of cardiovascular events. In contrast, substituting unsaturated fats for saturated fats yields the combined benefit of lowering total serum cholesterol levels and raising serum HDL cholesterol. This effect, together with favorable effects on markers of insulin resistance and inflammation, and potential antiarrhythmic effects of marine-based n-3 fatty acids, may partly explain why unsaturated fat consumption is associated with reduced cardiovascular events. Current guidelines from the NCEP now recommend increased proportions of unsaturated fat (25% to 35% total fat with less than 7% from saturated fat) [2]. However, a remaining question for future investigation is whether there is any beneficial effect of diets containing even higher proportions of unsaturated fat (eg, > 35% total fat with

2. Jakobsen MU, O’Reilly EJ, Heitmann BL, et al. Major types of dietary fat and risk of coronary heart disease: a pooled analysis of 11 cohort studies. The American Journal of Clinical Nutrition. 2009;89(5):1425-1432.

"Results: During 4–10 y of follow-up, 5249 coronary events and 2155 coronary deaths occurred among 344,696 persons. For a 5% lower energy intake from SFAs and a concomitant higher energy intake from PUFAs, there was a significant inverse association between PUFAs and risk of coronary events (hazard ratio: 0.87; 95% CI: 0.77, 0.97); the hazard ratio for coronary deaths was 0.74 (95% CI: 0.61, 0.89). For a 5% lower energy intake from SFAs and a concomitant higher energy intake from carbohydrates, there was a modest significant direct association between carbohydrates and coronary events (hazard ratio: 1.07; 95% CI: 1.01, 1.14); the hazard ratio for coronary deaths was 0.96 (95% CI: 0.82, 1.13). MUFA intake was not associated with CHD. No effect modification by sex or age was found. Conclusion: The associations suggest that replacing SFAs with PUFAs rather than MUFAs or carbohydrates prevents CHD over a wide range of intakes.
Am J Clin Nutr 2009;89:1425–32."

3. Astrup A, Dyerberg Jã¸, Elwood P, et al. The role of reducing intakes of saturated fat in the prevention of cardiovascular disease: where does the evidence stand in 2010? The American Journal of Clinical Nutrition. 2011;93(4):684-688.

"ABSTRACT
Current dietary recommendations advise reducing the intake of saturated fatty acids (SFAs) to reduce coronary heart disease (CHD) risk, but recent findings question the role of SFAs. This expert panel reviewed the evidence and reached the following conclusions: the evidence from epidemiologic, clinical, and mechanistic studies is consistent in finding that the risk of CHD is reduced when SFAs are replaced with polyunsaturated fatty acids (PUFAs). In populations who consume a Western diet, the replacement of 1% of energy from SFAs with PUFAs lowers LDL cholesterol and is likely to produce a reduction in CHD incidence of 2–3%. No clear benefit of substituting carbohydrates for SFAs has been shown, although there might be a benefit if the carbohydrate is unrefined and has a low glycemic index. Insufficient evidence exists to judge the effect on CHD risk of replacing SFAs with MUFAs. No clear association between SFA intake relative to refined carbohydrates and the risk of insulin resistance and diabetes has been shown. The effect of diet on a single biomarker is insufficient evidence to assess CHD risk. The combination of multiple biomarkers and the use of clinical endpoints could help substantiate the effects on CHD. Furthermore, the effect of particular foods on CHD cannot be predicted solely by their content of total SFAs because individual SFAs may have different cardiovascular effects and major SFA food sources contain other constituents that could influence CHD risk. Research is needed to clarify the role of SFAs compared with specific forms of carbohydrates in CHD risk and to compare specific foods with appropriate alternatives. Am J Clin Nutr 2011;93:684–8."

4. Hite AH, Berkowitz VG, Berkowitz K. Low-Carbohydrate Diet Review. Nutrition in Clinical Practice. 2011;26(3):300-308.

"The shift in metabolism that occurs on a LC diet heralds a shift in our current dietary paradigm. Changing the macronutrient content of the diet changes the metabolic profile. Although increasing fiber and decreasing saturated fat intake may be of concern to those whose diets are high in carbohydrates, they may be of less concern to those whose diets are not. Increasing vegetable consumption and decreasing consumption of foods low in nutritional value is a dietary goal cited numerous times in the 2010 Dietary Guidelines recommendations.87 Reducing dietary carbohydrate can accomplish this goal while improving glucose control, insulin response, atherogenic dyslipidemia, and other cardiovascular risk factors, in
addition to reducing caloric intake without hunger. This makes carbohydrate reduction a sensible dietary approach to achieving and maintaining health."

George writes:

The paper by Mercier and Sperber, Why do humans reason? Arguments for an argumentative theory, is probably applicable to this podcast's general question of why we focus more on proving ourselves right rather than focusing on finding the right answer. Here is a bit from the paper's abstract:

Reasoning is generally seen as a means to improve knowledge and make better decisions. However, much evidence shows that reasoning often leads to epistemic distortions and poor decisions. This suggests that the function of reasoning should be rethought. Our hypothesis is that the function of reasoning is argumentative. It is to devise and evaluate arguments intended to persuade...

For those who haven't already read the paper, it is probably worth your time.

Brit writes:

I was reminded of a 2009 BBC interview of Wallace Broecker, who supposedly coined the term 'global warming', in which he told how he had gotten his doomsday predictions of global cooling spectacularly wrong in the 1970s. Apparently, instead of modifying or dampening his predictions, or recognizing any limitations in himself or in his science, he then went exactly 180 degrees out and spent the rest of his career pitching a doomsday warming scenario.

First, good job hijacking the thread. Second: Global cooling was never a big topic in the 1970s among scientists, although the media really jumped on it, which is why the public is confused about it. And third, apparently, Broecker was a global warming proponent back in the 1970s when you say he was "[getting] his doomsday predictions of global cooling spectacularly wrong in the 1970s.": "In 1975, Broecker inadvertently coined the phrase global warming when he published a paper titled: “Climate Change: Are we on the Brink of a Pronounced Global Warming?”" Fourth, watch this video series (shown on the right panel): http://www.youtube.com/Potholer54#p/c/A4F0994AFB057BB8

emerich writes:

With all the clashing arguments and claims about the role of fat in the diet, it's at least interesting that there's close to a consensus about one thing: refined carbohydrates, also known as junk food, are bad. Taubes fingers refined carbos in his book, but advocates of low-fat diets also counsel avoiding the stuff. Just sticking to that rule in your own diet will probably get you a lot closer to the healthy end of the spectrum, whatever the poroportion of fat in your diet. As Taubes himself argues in the book, many of the indigenous peoples who got fat and diabetic didn't just adopt a "western diet," which arguably would include meat and eggs. They started eating refined white flour and sugar--junk.

Taubes makes another compelling point that doesn't require complete agreement with the "fat is OK" argument: when Americans cut fat from their diets starting in the 70s, they tended to add sugar and refined carbos, partly because it got added to all the "low-fat" prepared foods. If they had cut down on fat and added apples, berries, calliflower, pumpernickel bread (however unlikely) would obesity have exploded the way it did?

Jacob writes:

@emerich

I think it is more complicated than simply "refined" versus "unrefined".

For example, the now longest living people on the planet, the Japanese use only white rice. And people in the mediterranean - also amongst the longest living in the world - use mainly white bread and pasta for their carbohydrates. I doubt that these people would live even longer if they started to eat breakfast cereals with wheat bran, soybean flour, synthetic vitamins and so forth.

Another example: there's several studies showing a benign effect of honey, such as reducing risk of heart disease, improving memory, reducing anxiety, relieving infections and so forth. Beekepers are suggested to be very healthy and live longer than the general population. White sugar and high fructose corn syrup, on the other hand may increase risk of heart disease, increase anxiety, reduce longevity etc. Apparently honey does not lead to weight gain, unlike the white sugar and fructose corn syrup. Yet honey is basically 99,5% water and sugars, so in one one sense it is not much different.

Human milk could be considered a "refined" food, in that it provides over 50% of energy from fats (of which half saturated) and another 40% of energy from lactose/sugar. Human milk is very low in vitamins and minerals, providing only 30-60% of the RDA for most of those per 2000 kcal. Essentially if one only look at vitamins, minerals, carbohydrates, fats and proteins, there's not so much difference between human milk and a McDonald's meal.

I think instead of focusing on refined versus unrefined, we should look at natural versus unnatural. We should also consider that some foods are designed from nature not be be eaten and therefore have a lot of anti-nutrients in them (such as grains, seeds), while other are actually designed to be eaten; the plant wants its fruits and berries to be eaten so that the seeds can be spread and grow out of the excrement. The seeds however are sometimes poisonous or at least cause digestive problems, hormonal problems and so forth.

Perhaps the McDonald's meal would not be so unhealthy if a mixture of coconut and olive oil was used for the fries, if very little salt was added (and only natural sea salt), and instead of the giant coke you would have 1/2 cup of natural honey in some herbal tea etc. And maybe add in 1/2 teaspoon of cod liver oil also. But by many standards it would still be an almost equally refined dish.


@midnightmagic

As regards the inuits being short lived, I have not studied it properly, I was merely referring to Stefansson's article from 1936 where he concludes that:

"Perhaps it may be considered that meat is, overall, a stimulating diet, in the sense that metabolic processes are speeded up. You are then living at a faster rate, which means you would grow up rapidly and get old soon. This is perhaps confirmed by that early maturing of Eskimo women which I have heretofore supposed to be mainly due to their almost complete protection from chill - they live in warm dwellings and dress warmly so that the body is seldom under stress to maintain by physiological processes a temperature balance. It may be that meat as a speeder-up of metabolism explains in part both that Eskimo women are sometimes grandmothers before the age of twenty-three, and that they usually seem as old at sixty as our women do at eighty."

For sure I agree that omega-3 fatty acids is beneficial. I would take Icelandic people as an example - they have the highest life expectancy after the Japanese. Iceland have the highest per capita fish intake in Europe, and they also do dwell in a quite harsh climate.

However, the Icelandic people also eat a lot of saturated fat from dairy products and lamb/beef. So I am suggesting that the mix of the overall fat intake - the balance of saturated with polyunsaturated - is of more importance than one or the other per se. Human milk is also an example of this. No-one get more saturated fat than the child.

Too much omega-3 relative to saturated fats/carbohydrates I think is not a good idea. (At least this is why I believe Japanese and Icelandic people live longer than the inuits.)

John writes:

So, does this mean that the government is doing us a favor by having a high tariff on imported sugar?

Luke writes:

Enjoyed the podcast, but it would have been better without the attempts to draw economic parallels.
We like the topics in economics, but it is not necessary to milk those principles with everyone.

Trevor writes:

This was a great podcast for skeptics like myself.

I am an 'Austrian' because I do not have faith in the higher economic powers to be able to do the right thing (Gov't - Central Banks etc). I also have no faith in a Godly power, which makes me an Atheist.

It's made me wonder if someones thoughts on God had any relevance on their thoughts on the economic theories or their ability to accept things like fad diets on faith.

Floccina writes:

Gary Taubes had me up until you shifted to his alternative hypothesis.

I am sure that fat helps us feel satiated but barring extremes or diets lacking in certain amino acids or vitamins, IMO we still do not know if one diet is healthier than another.

Also:

I lecture in medical schools is: Look, we've got this obesity and diabetes epidemic going on

Hispanics are much more likely to have type 2 diabetes and this needs to me taken into account. Also there is a could be long term effects for less healthy people living long enough to reproduce in modern society.

Gary Noreen writes:

I was diagnosed Type 2 diabetic 18 years ago and quickly found that a very low carbohydrate diet "cured" me. Not only did my blood glucose levels become normal without medication, but my risk factors for heart disease (cholesterol, etc.) dramatically improved - in spite of eating high quantities of fat and saturated fat. I've been on a low carb/high fat diet ever since. Yes, low carb/high fat CAN be followed over the long term.

I am convinced that Taubes is right, and that there are enormous economic consequences. Consider the savings in health care costs if diabetics followed very low carbohydrate diets (under 30 gm/day) rather than the Institute of Medicine Recommended Daily Allowance of at least 130 gm/day of carbohydrates. Diabetics would need far less medication, or in many cases no medication, to "cover" the carbs that IoM says they should eat. Diabetics would need fewer doctor visits to control their blood glucose. There would be fewer costly diabetes complications.

Low carb/high fat diets may lead to massive health care savings not only from reduced need for and cost of diabetes treatment, but also from the many other diseases linked to metabolic syndrome, including obesity, cancer and even Alzheimer's (sometimes called "diabetes of the brain").

Given the many diseases linked to metabolic syndrome, and the spiraling cost of health care, the promotion of low fat diets by the US government may be costliest error it ever made.

Corey writes:

Perhaps this is all just a systemic problem from government funding research and development. If R&D came only from private firms then we wouldn't have these distortions.

wow writes:

[Comment removed for supplying false email address. Email the webmaster@econlib.org to request restoring your comment privileges. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

David Blair writes:

Hello Russ:

Discovered your podcast a few weeks ago and have made it all the way back to about mid-July in my listening. Thanks for the truly excellent discussions.

I have question rather than a comment, and it's not directly related to this episode, but may be something you could discuss with Simon Johnson. Reading the Krugman column linked below, I started wondering whether tax increases could be considered a form of "forced dis-saving" and, when paired with increased government spending, could be a way to make currently underemployed capital more productive?

Of course, I've listened to enough of your podcasts to know that the answer is, at best, "maybe, maybe not," but I'd find the discussion interesting, anyway.

Looking forward to your next episode,
David

Russell writes:

I suppose I would point out that the "eat anything" aspect of the Atkins Diet was also to be tempered by an "eat when you are hungry" proviso. The idea was that going on a low-carb diet would mean you're not hungry as often so your intake would also decline as well. It seemed to work though I didn't stick with it for unrelated reasons and I should have. My doctor would be happier if I weighed 20 -30 pounds less than I do.

The beauty of Atkins, well any diet really, is that you could gorge on carbos (or other forbidden foods) at a meal and that's okay as long as you get back into the routine. It's not like you utterly ruined the diet by doing the "bad" thing, if you get back on it. So have that big ice cream sunday or huge bag of chips (or eight slices of pumpkin pie :)), just don't make a frequent habit of it.

Brian Cosgrove writes:

Russ - love your show for how you explore all markets - hope your diet works out for you. Speaking of food - over Thanksgiving a friend recommend I read RAJ PATEL http://rajpatel.org/ for some new insights on the food markets. I would enjoy learning more about how food trades, is priced, etc.
Thanks for all your great work
Brian

Gary Noreen writes:

In response to Corey -

Private industry has been no better than government in the nutrition arena. A case in point :

Anthem Blue Cross, my insurance company, regularly sends me a newsletter entitled “Living Your Best Life With Diabetes.” One such newsletter featured a recipe for “Delicious Oven French Fries." This recipe has 52 grams of carbohydrate in the form of baked potatoes, which raises blood glucose more than table sugar. This is typical - every diabetes newsletter I receive from Anthem features a low fat/high carb recipe. Anthem nurses call me regularly and tell me to follow a low fat/high carb diet.

I can’t imagine anything worse for a diabetic like me to eat than baked potatoes. This would cause my blood glucose to skyrocket, requiring me to take a large dose of insulin (I have managed to avoid the need for insulin on my low carb/high fat diet). The insulin would be hard to control, making it difficult to avoid large changes in my blood glucose and likely leading to hypoglycemia (low blood sugar) and, ultimately, the many diabetes complications that I have, until now, avoided.

Not only is Anthem's nutrition advice appallingly bad for its diabetic clients, but it leads to much higher costs for Anthem due to:

1) Increased medication to “cover” the high level of carbohydrates Anthem recommends,

2) increased medical supervision to control the effects of these higher levels of medication, and

3) increased complications due to the hyperglycemia (high blood glucose) induced directly by ingestion of carbohydrates, as well as from hyperinsulinemia (high insulin levels) induced by the combination of Type 2 insulin resistance and increased insulin needed to "cover" excessive carbohydrates.

What is the economic incentive for insurance companies to tell diabetics to eat more carbohydrates? Are they just incredibly stupid, or am I missing something?

I wrote a letter to the President of Wellpoint, the parent of Anthem Blue Cross, a year and a half ago complaining about this, yet I still get Anthem diabetes newsletters pushing high carb recipes.

NonEntity writes:

[Comment removed for policy violations. Email the webmaster@econlib.org to request restoring your comment privileges.--Econlib Ed.]

vic writes:

Wonderful podcast- perhaps because it reinforced my own biases.

A few points I would like to make.

1. I am of asian indian origion. Indians have one of the highest cardiovascular risks in any ethnic population. risk is high in india. goes up in us immigrants. it just so happens that a very large proportion are vegans. and the indian died is extraordinarily rich in carbs. There is no other ethnicity I know of that will eat a meal of bread or rice with potatoes as the entree.

while you ponder over this.
Here is something else.

Pfizer and Merck developed the following drugs
torcetrapib and Anacetrapib respectively. they beong to a new class of drugs called CTEB inhibitors.

"In 2004, a phase I study of people taking that Pfizer drug, a CETP inhibitor called torcetrapib, reported that it could lower bad cholesterol and improve good cholesterol.

But a large outcomes study later indicated that the drug increased the risk of death by 59 percent and heart attacks by 25 percent. After spending more than $800 million to develop the drug, Pfizer ended the project in 2006."

http://prescriptions.blogs.nytimes.com/2010/11/17/mercks-new-cholesterol-drug-shows-promise/

So to my mind- here are two drugs that dramatically lower cholestrol, LDL cholestrol and increase HDL cholestrol---- all of what is what you want biochemically--- right
but mortality and cardiovascular mortality goes up.

classic scientific method (Popper/ kuhn both)says that if observable facts cannot be explained by working hypothesis- question the hypothesis..Where are the scientists questioning this????

I am a Pediatrician who stopped being a memeber of the American Academy of Pediatrics many moons ago as I concluded that they are intellectually challanged hacks at best and con men at worst. I wrongfully thought that this phenomena was restricted to the AAP. Now I realize that the AHA is equally intellectually challanged. Perhaps this is a problem endemic to professional academies/ societies.

I find Taube's comments on the state of science post WWII interesting. One of the things that occured post WWIII was "big science" funded by federal grants - NIH/ DOD/ DOE etc. Science was no longer about stochaistic tinkering by dilitantes, but now was all about grants and federal funding.

Create an idiotic system - you get idiots as a natural consequence.

Most practicing scientists of my acquaintance are frankly no better than witch doctors/ shamans. In my charitable moments I refer to them as technitians.

vic writes:

in the interest of maintaining equipoise
sometimes " experiments of nature " tell us a lot
a lot more than clinical observational studies
so deal with this

the incidence of coronary artery disease is extraordinarily high in people with a genetic disorder charecterized by very high levels of Cholestrol- familial hypercholestrolemia.

this would be very strong evidence for a link between CAD and cholestrol. However mechanisms might be different ( there goes my equipose!)

matt writes:

Age adjusted mortality from heart disease has declined significantly (60%) over the past 40 years in the US. Fat consumption has gone down but sugar consumption has gone up. Seems to be a big strike against sugar.

Seth writes:

Thanks for having Taubes on. I enjoyed the podcast.

I also wanted to thank Taubes for a couple things.

First, I lost weight 11 years ago. I thought it was due to a number of things I changed - like balancing my macro nutrient intake and meal timing.

In "Why We Get Fat" Taubes points out that, like me, most people mistakenly attribute lots of things to their weight loss, but a reduction in carbs/refined sugars is usually the main factor.

At the time I read that book, I had 5 stubborn pounds that I had been holding for about a year and a half. Reading his book made me realize that I had also gradually increased my sugar intake. I cut back on sugar and those 5 pounds came off within a week. I told a couple family members about it, who were also complaining about some stubborn pounds. They cut their sugars back and also saw immediate reductions in weight.

Second, for years my cholesterol numbers have been out of whack. Mainly, my HDL was less than it should be. Taubes' book "Why We Get Fat" convinced me to try bucking the advice my insurance company gave me (eat less cholesterol) and try eating more. It seemed to work. At my last health assessment, I expected my cholesterol numbers to be off the charts because of my change in eating habits. I was surprised when they all had improved and my HDL for the first time was in the recommended range.

So, for the sample size of 3, his advice seems to work.

Sebastian Hagen writes:

Gary Taubes's take on nutrition related in this episode reminded me of that of Robert H. Lustig, whose position is that the cause of metabolic syndrome isn't either dietary fat or carbohydrates in general, but specifically the carbohydrate (component) fructose, which makes up about 50% of both sucrose and HFCS, and is extremely commonly used as a sweetener in processed foods of many kinds.

Lustig presents his case in the video Sugar: The Bitter Truth; besides analyzing the historical trends, he also gives a causal model for why fructose causes metabolic syndrome and glucose doesn't. I highly recommend giving it a watch for people interested in this topic.

George writes:

Interesting & timely NYT Op-Ed piece by Bill Keller that echoes very closely Russ' own observations on economics punditry.

http://www.nytimes.com/2011/11/28/opinion/keller-the-politics-of-economicsthe-politics-of-economics.html

AHBritton writes:

I want to preface this message by saying that I am far from an expert on this topic. And beyond that (although I think this is an important, interesting, and worthy topic) it is currently rather low on my list of things I am actively involved in researching.

That being said, I have looked into this topic somewhat in the past and have am few general criticisms of Taubes.

Despite much of his rhetoric Taubes' seems to paint things as very black and white (scientists said fat is bad) and without nuance. In addition he seems very interested in emphasizing his "controversial" stance.

In debates and responses to Taubes, many seem more than willing to accept some of his claims, even failing to see what is controversial about them.

Many even will agree with his basic methodological and epistemological points regarding imperfection of studies, distortionary effects of research funding, need for humility when assessing the strength of claims, etc.

The main contention seems to be over the fact that Taubes' seems (in my opinion) incredibly insistent and sure of himself in his claims that exercise has little to nothing to do with healthy weight loss and that heigh protein/very low carb diet without portion control is the only/best method for healthy weight control.

Despite his rhetoric THIS seems to me to be the hubristic, uncompromising, myopic, and unsupported position in the debate.

Again, I may be entirely wrong, however this is the impression I get not only from watching him interact with trained medical professionals but also from reading about his position and those critical of it.

A resource critical of his position can be found here:

http://weightology.net/?p=265

http://en.reddit.com/r/Fitness/comments/gf8zy/a_thorough_debunking_of_gary_taubes/

Sean Schubert writes:

You might want to check out Seth Robert's website. He'd make a great interview. He wrote The Shangra-La Diet, and is very interested in personal experimentation as a counterbalance to the locked-in scientific consensus.

One of his examples is acne. Through personal testing he found diet mattered, but his dermatologist was completely uninterested since it didn't involve "the Doctor" being the "expert" that dispensed drugs and sage advice.

http://blog.sethroberts.net/2011/08/31/the-continued-existence-of-acne-reveals-the-perverse-incentives-of-modern-medicine/

Bubba writes:

[Comment removed for supplying false email address. Email the webmaster@econlib.org to request restoring your comment privileges. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

L Taylor writes:

Discovered an significant conflict between text and audio @ about 13:30.

Audio says..."that are incapable of"....

Text says...."considering"....

[Hmmm. First, thanks for being so attentive. But second, I re-listened and around 13:30 there is a significant mistake in the text Highlights (I missed when Russ corrected himself from "owed" to "had lent to"), which I've corrected. I can't, however, locate where you say the text says "considering." Feel free to email me at webmaster@econlib.org or to post another comment here.--Econlib Ed.]

Richard Bates writes:

one note from the end of the story is that the zero carb diets cannot be continued for a long time. The Atkins diet is NOT a zero carb diet. Unfortunately that is the way it has been portrayed. However the first week is 20 grams of carb but you start to move back in carbs on the diet as the weeks go on until you reach your carb balance. And a large percentage of the carbs are from the foundation items vegetables etc. So it is sustainable if done right. The all meat no carb diet is not. Just FYI. I am off to buy his book though. Great article! I have been screaming this stuff for years. I however like most of us, waited until this year to act on it. Loosing 58 lbs. so far. Thanks for a great interview.

Jim in StL writes:

Regarding longevity increasing steadily. A major driver is improved survival of infants and children. Eliminating mortality from infectious disease in the first 4 years of life drastically increases average life expectancy over the life of the first generation. Conversely, the life expectancy falls relatively slowly if infant and child mortality increases until multiple cohorts have entered the population.

Comments for this podcast episode have been closed
Return to top