David Ludwig, MD, PhD., is a practicing endocrinologist at Boston Children’s Hospital, professor of nutrition at Harvard Medical School, and author of the New York Times #1 bestselling book, Always Hungry? He recently joined Gary Taubes for a Heleo Conversation to discuss why we get fat, what the dietary culprit could be, and how to reverse the obesity and diabetes epidemics. Gary is an award-winning science writer whose most recent book, The Case Against Sugar, examines why sugar might just be the tobacco of our day—damaging, addictive, and backed by powerful corporations.
This conversation has been edited and condensed. To view the full conversation, click the video below.
David: The conventional view of weight control holds that all calories are alike. You just have to eat less and move more—it’s a question of personal discipline and willpower. That’s harder in our modern environment, with all of the tasty, high-calorie dense foods around us, but it’s basically a question of self-control.
So, if I locked you in a room and just gave you water and 800 calories a day of sugar, you’d certainly lose weight. What’s the matter with that way of thinking?
Gary: The question we want to know [the answer to] is: why do people accumulate excess fat? That’s a different question than how to remove excess fat. One of the things I learned during my scientific apprenticeship was that the question you ask has to be very specific, and the answer you get is very specific to the question you asked.
So, if you ask, can I starve people and get them to lose weight? Clearly. In fact, that’s one of the driving observations behind this energy balance hypothesis. When I lecture about this, I have a slide from prisoners in World War II concentration camps. You starve people, they lose weight. The point is, you can starve children and inhibit their growth. Nobody would say their growth is caused by excess calorie consumption. Why would we conclude, therefore, that excess calories drives the accumulation of fat to begin with?
David: We intuitively understand that linear growth of a child isn’t specifically driven by how much we eat. Yes, we need a certain number of calories, we can’t grow during times of severe starvation, but there’s an internal drive of linear growth. You’re making an analogy between that and the development of fat mass, arguing that something is programming fat mass to grow. Yes, you can temporarily arrest that—
Gary: Or reverse it temporarily by starving it.
David: So why not? Why don’t we just starve ourselves and solve the problem?
Gary: Because it doesn’t solve the problem. It reverses the process only while there’s an absence of calories available. We don’t live in an environment where there’s an absence of calories available, and on top of it, we can find populations in which malnutrition and undernutrition are clearly present, and yet there are high levels of obesity. At least in some populations, the fat tissue finds a way to grow despite the relative absence of calories.
David: So we can limit it by cutting back and eating less, but the body fights back with increasing hunger. We get hungry, and that is very difficult for most people to resist. There’s also metabolic changes.
“You’ve got sugar always at the scene of the crime in populations, you’ve got sugar always at the scene of the crime in the body. The mechanism is there.”
Gary: Invariably, these discussions get around to the aspects of what’s the best diet. The issue that I’ve been trying to approach is we’ve got these ongoing tragedies, these obesity and diabetes epidemics. Every population in the world that goes through a nutrition transition to a Western diet eventually manifests massive increases in obesity and diabetes. We’re trying to find a culprit for the crime committer of these epidemics. Sugar is the prime suspect because, first of all, it’s always at the scene of the crime. The nutrition transition for many populations—whether it’s the Inuit, half the population in Africa, or agrarian populations in India—you add sugar to the diet, you very quickly see a whole slew of chronic diseases that inevitably include obesity and diabetes.
Mechanistically, we can put it at the scene of the crime. These diseases are highly associated with insulin resistance. You could say that this epidemic of obesity and diabetes is an epidemic of metabolic syndrome and insulin resistance. Metabolic syndrome is a condition that precedes these two, and precedes heart disease and cancer to some extent. If that’s the case, we want to know what causes metabolic syndrome and insulin resistance.
There is a lot of evidence that implicates sugar, because the fructose moiety of the sugar molecule, as chemists like to say, is metabolized in the liver, and there’s a lot of evidence that insulin resistance starts with fat accumulation in the liver. Some of the fructose is converted into fat under the right circumstances.
Basically, you’ve got sugar always at the scene of the crime in populations, you’ve got sugar always at the scene of the crime in the body. The mechanism is there.
In my early book, I talk about sugar and refined grains, and this is the argument we always have about observational ecological data. I’m saying sugar and refined grains are the cause of obesity and diabetes, and somebody says, “What about Southeast Asia?” That’s a heck of a black swan. In Southeast Asia, they eat refined grains and have been doing it for thousands of years. Why don’t they have high levels of obesity and diabetes? Well, they didn’t consume much sugar. They had among the lowest sugar consumption in the world. Japan has been about a century behind us in sugar consumption, and about a century behind us in the diabetes rates. The simplest possible hypothesis to me has always been sugar.
David: So your two-minute argument to the jury is that [sugar] is at the scene of the crime. There’s a smoking gun, although we haven’t really seen sugar holding it yet.
Gary: It should be the prime suspect. If I was a prosecuting attorney, I could get an indictment. I don’t think I can get a conviction based on the evidence as it is now.
David: Let’s dive a little bit more into the nutritional science of carbohydrates. All carbohydrates are made of, or break down into, sugars. There’s really only three different kinds of sugars: glucose, fructose, and galactose. Galactose really only shows up as milk sugar.
Anything that comes from a grain product or potato product, these starches were the base of the food guide pyramid in 1992. If processed, they will digest very quickly into glucose, raising blood sugar and insulin, whereas virtually all of the common sweeteners have a different composition. They have typically half glucose and half fructose. It’s the fructose component that adds the real sweetness to sugar. Glucose by itself tastes anemically sweet. Conversely, pure fructose is cloyingly sweet. It seems that the sweet spot for humans is that 50/50 combination.
Is fructose as opposed to glucose uniquely harmful, or are they equivalently harmful, if highly processed? This is a critical question to public health. If fructose is uniquely the problem, we could get rid of sugar, replace it with white bread, white rice, potato products, unsweetened cookies, crackers and the like. Have our cake and eat it too. Can we feel good about those other processed carbohydrates, or is it a question of picking your poison and we need to take a more global look at processed carbohydrates? What’s your opinion?
Gary: Well, I can give you a firm maybe. There are a lot of different issues here that you’ve touched upon. One of the things I’ve thought about in the course of doing the research for [The Case Against Sugar] is maybe the fundamental problem is liquid carbohydrates. Maybe it’s the delivery system.
“Clearly, the enormous amount of sugar we consume relative to what we consumed 200 years ago appears to cause metabolic changes.”
I imagine a world in which, instead of drinking 24 ounces of sugary beverages, half glucose/half fructose, we are making our sugary beverages with fructose only. As you put it, glucose alone is anemically sweet, as Gatorade used to be. When I was a kid, Gatorade tasted like swamp water, but you drank it anyway. Now, if we drink the same amount of glucose water that we drink of fructose-sweetened water, what are the metabolic effects? The answer is, I don’t know. We don’t do it. It’s like when people ask me, is honey as bad as sugar? To quote Lustig, the dose makes the poison.
Clearly, the enormous amount of sugar we consume relative to what we consumed 200 years ago appears to cause metabolic changes. Honey is simply not as useful a food. You can’t blend it with cold drinks, and much of our consumption of sugar is in cold drinks. If we could consume 80 or 90 pounds of honey per capita, it is likely to be as harmful as sugar, but we can’t do it. Per capita consumption of honey has been a couple pounds a year for a century.
I’m putting my money on fructose. In part, it’s the fructose-glucose combination, because that’s all we know, because that’s what we consume.
David: To be fair, we don’t have definitive studies. Some of the studies that have made fructose look very bad have used just fructose alone, which is not a way that humans would ever consume it—at astronomically high doses, more than double the 95th percentile of consumption.
Hopefully they’ll have more definitive research that puts fructose into the proper context. After all, we eat a lot more glucose than we do fructose, because of all the grain products and potato products.
Let’s move on to a related concept, glycemic index (GI): it reflects how fast carbohydrates turn into glucose, the ultimate breakdown product. Most processed grains and potatoes turn into glucose very quickly in the digestive tract. In fact, they raise blood sugar faster than sugar. These things are going to raise insulin more.
Insulin promotes calorie uptake in fat cells. If you just give an animal insulin, it gets hungry, it gains weight. Even if you restrict its calories, it still becomes overly fat.
The fast-acting glucose takes a hit on the fat cells. The fast-acting fructose takes a hit on the liver by leading to increased inflammation. If the liver has a metabolic problem, you’ve got a metabolic problem. Conversely, slow-acting forms of these carbohydrates are much less concerning, whole fruits or low glycemic index, minimally processed.
Gary: That was the thesis of my first book, and it’s still indirectly the thesis of the second one. We’re confounding two issues: one is the cause of the obesity epidemic. What’s the trigger of the obesity? We know that nutrition transitions tend to increase the glycemic index. They increase refinement and sugar content. They might even decrease the total carb count, because they also increase fat and protein content. Populations become wealthier, they can afford to eat more animal products. But the dominant changes are sugar and higher GI. From that observation, both of them are the case.
The second question is: what are we going to do with patients? What advice would you give obese and diabetic individuals? There’s another twist—what advice are you going to give lean, healthy people to prevent them from developing obesity and diabetes?
David: So, if somebody is on the verge of Type II diabetes, do you want them on a no-fructose diet, a low-carb diet, a ketogenic diet?
Gary: In general, it would depend on their weight. I would remove sugar first because it’s an easy target. Then I would restrict the high-GI carbs just like you would. Then, what dose of the intervention do we need to make the person healthy? My dose is removing carbs and replacing them with natural animal fats, which is a new synonym for saturated fats that we’re not too scared of. This would be the ideal treatment.
David: Do you think the safest thing is a minimal carbohydrate diet?
Gary: We don’t really know. It needs to be studied. I hear from people all the time who have reversed diabetes on a low-GI diet. An example of this is the pediatric epilepsy case. The original treatment was a very strict ketogenic diet that was like drinking eggnog all day long without the alcohol. Then people moved on to modified Atkins diets and realized that that did a very good job of inhibiting seizures.
David: Just to clarify, a ketogenic diet is the granddaddy of all low-carb diets. The typical American might be eating 50-55% carbohydrate. Mediterranean diets are in the 40% range. You start thinking a low-carb diet when you get down to 30%, and very low carb at 20%. Once you get below about 15%, the body does not have enough carbohydrate to feed the brain directly from glucose, so you shift into ketosis. That’s a natural pathway that humans have evolved to feed the brain during either times of starvation or during long winters when there were no carbohydrates at all.
Gary: Or during the hundreds of thousands of years in which many of them did not eat carbohydrates.
“Processed carbohydrates, especially during the low-fat years, were given a blessing by some of the most preeminent leaders in nutrition. These cause this flood of carbohydrate to enter either the general circulation or the liver much too fast, faster than would ever be the case except if you were eating honey as a hunter-gatherer.”
David: That’s right. In fact, carbohydrates are the only totally dispensable nutrient. You need protein, you need fats, you can do without carbohydrate entirely. When you get to those very low levels, then ketones come very rapidly out of fat cells.
The idea is that it’s a very potent fuel for the brain. It might help with epilepsy, it might help take the stress off the pancreas for somebody with diabetes. Do you like that kind of a diet?
Gary: A lot of people find it remarkably effective. Would I advise it to an entire nation? Not on the basis of the research that exists so far.
What are your prime suspects [for the obesity and diabetes epidemics], and what are we going to do about it?
David: My prime suspect is fast-digesting carbohydrate. That’s the most strictly controlled metabolic fuel in the bloodstream. The body cares a lot about keeping blood sugar in a very tight range. We have one hormone, insulin, to lower it. We have a handful of stress hormones to raise it. You don’t want blood sugar too high, it’s highly reactive and it’s lost in the urine. You don’t want blood sugar too low because the brain suffers.
These processed carbohydrates, especially during the low-fat years, were given a blessing by some of the most preeminent leaders in nutrition. These cause this flood of carbohydrate to enter either the general circulation or the liver much too fast, faster than would ever be the case except if you were eating honey as a hunter-gatherer. That stresses the liver, it causes de novo lipogenesis and inflammation. It raises insulin that promotes fat deposition and fat tissue.
For most people, lowering processed carbohydrates will give them most of the benefit. I don’t think most people necessarily need a very low carbohydrate diet. I do believe that for people with either Type I or Type II or with severe insulin resistance, that these very low carbohydrate and ketogenic diets have extremely promising positive findings.
David: [Regarding protein, some people ask if] there is an advantage of plant versus animal protein—specifically how much. In a ketogenic diet, protein actually raises insulin and can be converted into glucose via gluconeogenesis, but fat can’t. So very high amounts of protein prevent ketosis. Do you have a recommendation on how much protein [to consume]?
Gary: No, and traditionally these diets are always high-fat. The idea that they should be high-protein emerged in the 1980s and 1990s because it was so politically unacceptable to push a high-fat diet. Everyone agreed protein was good for you, so suddenly we had terrific low-carb books like Protein Power. When I advise people privately to eat these diets, I remind them that they should be high fat diets. You shouldn’t be eating lean protein.
David: So, you’ve taken away sugar, what about artificial sweeteners? What about the ones that were made in a laboratory?
Gary: My feeling on the artificial sweeteners was captured by Philip Handler, president of the National Academy of Sciences, when they had a forum on sweeteners back in 1975. He said, “The critical point is, how much sweetener do I have to consume to replace the 100, 120, or 140 pounds of sugar I would normally consume? Which is worse?” It’s not whether or not sweeteners are going to cause cancer in one rat, because we both think to some extent the sugars and refined grains, for whatever reason, are also going to increase cancer risk. You use a little bit because it’s so much sweeter than sugar. Is that better or worse than sugar? I can’t believe it’s worse.
David: Would you drink it?
Gary: No. I used to, but I don’t anymore. I think virtually everybody who advocates changes to a healthier diet would say: consume water, lose your sweet tooth. I worry that, by keeping the sweet tooth alive, it keeps the cravings alive. There’s some evidence that it could have metabolic effects that are deleterious, but it’s not very convincing evidence.
David: [A question I’m often asked is] how can society support medical research and nutrition that is not controlled by big Pharma? Where is the best research being done and how is it funded?
Gary: [Personally,] I define the best research by whether it’s asking the right questions and whether it has the possibility to answer the right questions. I believe the right questions are very different than the mainstream of nutrition and obesity research.
“We have epidemics, and although we think we know why they’re happening, the fact that we failed completely to control them in any way suggests that our fundamental understanding is incorrect.”
I do think you guys are doing the best research because you’re one of the few researchers out there asking the same questions I’m asking, and who thinks in a similar, albeit nonidentical way, about how those questions have to be answered.
David: What’s the big view? Where do we have to go as a society, if it takes a billion dollars to bring one drug to market for just one complication of one diet-related chronic disease?
Gary: Not only that, remember that diet-related chronic disease—obesity and diabetes together—cost the healthcare system about a billion dollars a day.
David: And it’s going up. Type II diabetes alone is predicted to hit half a trillion dollars a year in costs.
Most nutrition studies are done on a shoestring budget with poor quality control and an inability to ask questions definitively. How do we deal with this?
Gary: The way they fund studies is a fundamental issue. There’s a little bit of funding for thousands of researchers none of whom have enough money to do anything meaningful even if they wanted to. The first step, like getting off alcohol, is acknowledging you have a problem. We need the National Institutes of Health and the Public Health Authorities to say we have epidemics, and although we think we know why they’re happening, the fact that we failed completely to control them in any way suggests that our fundamental understanding is incorrect.
Then we need them to say, what are the alternatives to the way we understand the cause of these epidemics? Because if we don’t understand the cause, we’ll never successfully prevent them. Then we have to fund research to identify what we might have missed.
In an ideal world, you get the money from independent philanthropists who recognize that there’s an enormous problem and want to do something about it. It’s hard to imagine the federal government stepping in with the right mindset—which of course is a synonym for my mindset.
David: You’re honest to acknowledge that.
We can’t forget that scientists have egos, too, and we get stuck in our paradigms and our careers can be affected. There’s always been ego in science. What’s different now is massive industry money, which can bias what we research in various directions.
“The history of industry funding, be it for tobacco, traffic fatalities, sugar, is that these relationships tend to benefit the industry and undermine public health much more than advance science.”
Gary: Let’s say Coca-Cola comes to you and they say, we’re really worried about our products. We want to give you $20 million because we know you’re a critic. We’re going to give you $20 million to design a study, and we’re going to have the best scientists in the world review that study, and we’re going to really find out if we’re killing people.
Are you going to turn them down?
David: Yes. I have, and the history of industry funding, be it for tobacco, traffic fatalities, sugar, is that these relationships tend to benefit the industry and undermine public health much more than advance science.
For my last question, I’d like you to envision you’re at the end of your career, and you’re looking back. What would you see if your life’s work turned out to be completely successful?
Gary: We used to discuss in NuSI (Nutrition Science Initiative) how we would define success. At the moment, we have failed approaches to treating obesity. We know it doesn’t work because people just keep getting fatter and fatter. That leaves the door open for everyone to have an idea—from journalists to diet doctors, cardiologists, you name it, everyone’s got a theory and there’s this noise being generated all the time. People don’t know what to do.
In my ideal world, everyone involved knows what the right answer is. Everyone involved is giving the right advice. Pediatricians are giving the right advice, mothers are giving the right advice, schools are designing themselves in a way that foods that the kids are getting are right. The food system has changed so that whatever foods shouldn’t be [consumed] are harder to find. Insurers know what the right advice is, the doctors, hospitals, and researchers know what they should be studying to perfect that advice.
I want to make sure that every obese child in the world is going to get the right advice and the help they need to maximize their chance in being a lean and healthy adult. That’s the goal.