Wednesday, June 4, 2014

Calorie Intake and Body Fatness on Unrestricted High-fat vs. High-carbohydrate Diets

In recent posts, we've explored the association between calorie intake and the US obesity epidemic, and the reasons why this association almost certainly represents a cause-and-effect relationship.  I also reviewed the evidence suggesting that carbohydrate and fat are equally fattening in humans, calorie for calorie.

One valid objection that came up in the comments is that calorie-controlled diets in a research setting may not reflect what happens in real life.  For example, in a context where calorie intake isn't tightly controlled, diet composition can impact calorie intake, in turn affecting body fatness.  This, of course, is true, and it forms one of the central pillars of our fat loss program the Ideal Weight Program.

Some low-carbohydrate diet advocates argue that the obesity epidemic was caused by US dietary guidelines that emphasize a carbohydrate-rich diet*.  The idea here is that the increase in calorie intake was due to the diet shifting in a more carbohydrate-heavy direction.  In other words, they're hypothesizing that a carbohydrate-rich eating style increases food intake, which increases body fatness**.  According to this hypothesis, if we had received advice to eat a fat-rich diet instead, we wouldn't be in the midst of an obesity epidemic.

Fortunately for us, this hypothesis has been tested-- many times!  Which eating style leads to higher calorie intake and body fatness when calories aren't controlled: a carbohydrate-rich diet, or a fat-rich diet?

Short-term Studies

Short-term studies are quite consistent with one another: unrestricted high-fat dietary patterns lead to higher calorie intake than unrestricted high-carbohydrate dietary patterns (although in some studies, volunteers in both groups overeat) (1, 2, 3, 4, 5).  This is due in large part to the high calorie density and palatability of fat, because when studies control for those factors, the difference between fat and carbohydrate disappears (6, 7, 8).  Added fats like oils are particularly effective at increasing passive calorie intake while providing little added satiety.

Longer-term Studies

These studies lasted days to weeks, and essentially echo the short-term studies.  Unrestricted diets rich in fat tend to lead to higher calorie intake and fat gain relative to unrestricted diets rich in carbohydrate (9, 10, 11, 12, 13, 14, 15).  Again, the effect is probably due to the high calorie density and palatability of high-fat foods.

I also think it's worth noting that high-fat diets can be fattening in a variety of nonhuman species, including mice, rats, hamsters, squirrels, dogs, pigs, cats, and primates (16, 17, 18, 19, 20, 21, 22, 23, 24).  That effect also seems to relate to calorie density and palatability.  Usually, these high-fat diets are based on refined ingredients, but adding fat to unrefined diets also increases fatness to a lesser extent.  Contrary to popular belief, these diets don't have to be high in sugar, trans fat, omega-6, or artificial ingredients to be fattening in animals.  However, adding a bit of sugar does potentiate the effect.

Are you surprised to learn that the low-fat diet recommendations of the 1990s have a solid scientific basis?  These recommendations are often painted as ignorant and ideological, but that is simply incorrect.  Whether you agree or disagree with the dietary guidelines, you can't argue that the advice wasn't based on a reasonable interpretation of the evidence.

But What About the Low-carbohydrate Diet Studies!

Objection!  If this is true, why do people eat fewer calories and lose fat on low-carbohydrate, higher-fat diets?  Studies have consistently shown that overweight and obese people eat fewer calories and lose fat while adhering to a low-carbohydrate diet, and that fat loss is greater than on low-fat diets (25, 26, 27).  This effect is most pronounced during the initial "induction phase" of strict carbohydrate restriction, while the long-term effects of moderate carbohydrate restriction appear to be modest in most people (28).  Declining effectiveness over time could be partially due to declining adherence to the diet.

I think this is a good time to introduce a concept that may seem abstract, but I think is very important.  It's the difference between addition and subtraction.  To illustrate the concept, consider three scenarios:
  1. Sarah sits down to a dinner of beef stew, salad, and bread.  
  2. Sarah sits down to a dinner of beef stew, salad, and bread.  Before eating, she adds a stick of butter to the beef stew for flavor, and pours herself a plate of olive oil in which to dip her bread.
  3. Sarah sits down to a dinner of beef stew, salad, and bread.  She puts the bread back into the fridge and only eats stew and spinach.
In scenarios 2 and 3, Sarah's meal will be biased toward fat and away from carbohydrate.  Yet these two scenarios will lead to very different effects on calorie intake.  This is because in scenario 2, Sarah is adding concentrated fat to her otherwise unrestricted meal, while in scenario 3, she's subtracting one third of the foods she would normally have eaten.  Why do nearly all weight loss diets work to some extent, even diets that are based on diametrically opposed principles?  They all subtract something.

In scenario 3, Sarah's protein intake will also be higher.  Protein is the most satiating macronutrient by far, and recent research suggests that increased protein intake plays an important role in the ability of moderate carbohydrate restriction to reduce calorie intake and body fatness (29).

Extreme carbohydrate restriction may reduce calorie intake and body fatness by other mechanisms, perhaps involving the production of ketones.  We still have a lot to learn about ketogenic diets, and I believe continued research in this area will be genuinely informative.

Low-carbohydrate diet studies are interesting and valuable, but they have to be interpreted in the proper context.  Just because low-carbohydrate diets can be slimming doesn't mean it's slimming to put butter on your bacon in the context of an unrestricted diet.

What about Replacement?

We talked about adding and subtracting from a diet, but what about deliberately replacing one type of food with another?  One of my favorite studies did just that (30).  Overweight volunteers were randomized into three diets:
  1. Replacing 1/4 of daily fat intake with 'simple carbohydrates' (sugared foods)
  2. Replacing 1/4 of daily fat intake with 'complex carbohydrates' (both refined and unrefined starch foods)
  3. A control diet in which nothing was deliberately changed
None of the diets were calorie controlled or calorie restricted.  Over the course of 6 months, group 2 lost 9 pounds (4.25 kg), but there were no significant changes in group 1 or 3 (although group 1 did lose 0.6 lbs).  This suggests that sugared and fatty foods are equally fattening in the context of a typical diet, since substituting one for the other had no effect on body weight.  However, replacing fatty foods with starchy foods produced weight loss, suggesting that the most commonly eaten sugared and fatty foods are fattening relative to starch foods.

This is consistent with the results of a similar 14-day trial (31).


Research demonstrates fairly consistently that eating an unrestricted fat-rich diet, and/or adding isolated fat to foods, tends to increase calorie intake and body fatness in humans and other animals.  The US obesity epidemic probably wasn't caused by advice to eat a carbohydrate-rich diet, and wouldn't have been prevented by advice to eat a fat-rich diet instead.  The large increase in the use of added fats is probably one reason why we're fatter today than we were 50 years ago.

On the other hand, research also consistently shows that restricting the diet, including by subtracting carbohydrate, decreases calorie intake and body fatness in overweight and obese people.  This is particularly true if protein intake increases or if carbohydrate is restricted to ketogenic levels.

It's important to recognize that research typically focuses on averages, and doesn't always do a good job of reporting individual variability.  You may not react to a particular diet in the same way as the "average person".  In the end, I believe the best way forward is to develop an eating style that allows you to eat the appropriate number of calories to meet your health/weight/well-being goals while feeling satisfied.  That diet will vary from person to person, but there are general principles that apply to most people.

* Personally I find it highly unlikely that the USDA dietary guidelines contributed to the obesity epidemic.  If they did, it would have to be because people misunderstood or misapplied them.  From the very beginning, the guidelines recommended restricting refined sugar (including sweetened beverages), refined starches, and added fats.  Yet the intake of all three rose quickly in the ensuing decades.  The advice to focus on unrefined carbohydrate foods had little impact.  Did the low-fat message cause the obesity epidemic?  Unlikely, because our absolute fat intake didn't actually decrease.  My view is that people like to blame things on the government.  Sinister backroom machinations, or blind ideology with a side of incompetence at the very least.  It makes a good story, but in this case it's tough to reconcile with the evidence without a lot of intellectual squirming.

**  There's also the hypothesis that carbohydrate increases body fatness independently of calorie intake in humans, but we've already covered the evidence that refutes this.


Jeremy said...

Are you going to comment on the recent JAMA editorial (and related NY Times editorial) by Ludwig and Friedman? Ludwig is a well known academic researcher (he has a research center at Boston Children's Hospital) who is advocating a pretty standard low carb, Gary Taubes view.

Scott Russell said...

Two thoughts:

1. I think the distinction between addition and subtraction is important. People typically don't say that eating fat ad lib will lead to weight loss (hopefully), but rather that a high-fat, hypocaloric diet is preferable to a high-carb, hypocaloric diet. As a caveat, I think people avoid saying calorie-restricted when talking about low-carb diets for purely psychological reasons, because even thinking about diets makes people hungry.

2. While I think its fair to say that USDA guidelines probably aren't the first thing people think of when picking foods, I also think that the widespread belief that fat was bad led to the illogical conclusion that low-fat was automatically good. So people approached low-fat/fat-free foods as being "good," and thus justifying the addition of more of the foods that were "bad." We see this same effect with diet soda and weight loss supplements: tell people this is good for their diet, and they will make worse decisions elsewhere.

Rakesh C. Patel, M.D. said...

" In the end, I believe the best way forward is to develop an eating style that allows you to eat the appropriate number of calories to meet your health/weight/well-being goals while feeling satisfied. "

I completely concur, very well said.

The Ancestral Chemist said...

Hmm... maybe we should just try to eat better, rather than expend a ton of energy fighting endless carb vs. fat wars.
Buttering bacon is no better and no worse that chugging soda.

Kris said...

I agree with a lot in the article, but I disagree about the low-fat guidelines being evidence-based.

The main purpose of the guidelines (at least originally) was to reduce CVD risk.

But, as we know, the SFA-CVD link was never proven and still hasn't been (and probably never will be), as is covered in detail in many articles on this site.

Gretchen said...

"In scenario 3, Sarah's protein intake will also be higher." I think it's important to distinguish the amount of a nutrient and its percentage.

If I normally ate a burger and fries and then I reject the bun and the fries, my percentage of protein has gone way up, but the amount I'm eating is the same, and it's misleading to suddenly say it's a high-protein diet.

Gretchen said...

One thing many studies don't control for is the fact that people with metabolic syndrome or prediabetes may respond differently from those with excellent blood glucose control.

And, as you note, conclusions are based on averages.

Stephan Guyenet said...

Hi Jeremy,

I have no plans to comment on it in a post. The argument is similar to what Gary Taubes has advanced, and I think I've addressed the arguments adequately in previous material. Although if there's enough interest from the readership I might write up a post.

The main problem with the argument is that there's very little evidence to support it. It sounds good on a common sense level but it simply hasn't been supported by research.

Ludwig proposes that the fat tissue of some people just hoovers fat out of the blood due to elevated circulating insulin, making them hungry and causing them to gain fat, but like Taubes he provides no evidence that this actually occurs. In fact, we have fairly convincing evidence that it doesn't occur. The fat tissue of obese people actually has a harder time trapping fatty acids than the fat tissue of lean people. And elevated insulin levels do not predict future fat gain in observational studies.

Jeremy said...

Thanks for the reply, Stephan. Seeing a Taubes-like argument in a prestigious journal made me revisit the theory. I will be interested in their study ending in 2016 (results in 2017?).

Daniel Bradley said...

I agree that we shouldn't just blame everything on government but I'm curious about the connection between the adoption of dietary goals in the 70s and the subsequent rise in obesity. Maybe it's just coincidence. But I could the adoption of dietary goals setting this process in motion:

1. Based on official recommendations, Americans strive to eat low-fat and low-SAFA foods.
2. Food companies seize this opportunity to produce highly palatable foods that fit these guidelines, e.g. potato chips fried in vegetable oil (low-SAFA).
3. Having cut out a portion of calorie intake by avoiding whole milk, fatty meat, and butter, Americans subconsciously attempt to make up the balance by snacking on the highly palatable foods from (2), which they believe to be healthful because they are low-fat or low-SAFA.
4. Because these foods are so palatable and because they claim to be healthful, Americans end up not only making up the difference in calories but consuming an excess, leading to obesity.

I don't know that this chronology is completely accurate, but if so it seems like a possible framework by which the national dietary guidelines could have led to increased obesity in an unintended-consequences sort of way.

valerie said...

You write: "Added fats like oils are particularly effective at increasing passive calorie intake while providing little added satiety."

I think you mean "satiation" instead of "satiety."

You write: "According to this hypothesis, if we had received advice to eat a fat-rich diet instead, we wouldn't be in the midst of an obesity epidemic."

I never heard a low-carb proponent suggest that (they say we should have beed advised a low-carb diet, not a fat-rich diet). This is a strawman.

You write: "Are you surprised to learn that the low-fat diet recommendations of the 1990s have a solid scientific basis?"

If you're going to prescribe a diet to millions of people, you need long(ish)-term studies on humans. Anything less is irresponsible and certainly does not qualify as "solid scientific basis."

You write: "Personally I find it highly unlikely that the USDA dietary guidelines contributed to the obesity epidemic. If they did, it would have to be because people misunderstood or misapplied them."

Weren't we told explicetly to replace butter with margarine? Doesn't that increase risks of obesity in every experiment?

Weren't we told explicetly to replace full-fat milk with low-fat milk? Isn't that associated with higher obesity rates in epidemiological studies?

I don't see how you can say that people "misunderstood" or "misapplied" those very direct recommendations.

Anonymous said...

This is a great overview/analysis of those studies - so helpful! I find it incredibly frustrating to hear that "USDA nutrition guidelines made Americans fat" (and similar rhetoric) from the low-carbohydrate camp, because the guidelines are truly in line with scientific research. The real problem is that the general population has never been compliant with those guidelines.

So again, thank you! I've been a longtime reader, but don't believe I've ever commented before...yours is one of my favorite blogs for research-based health information.

Stephan Guyenet said...

Hi Jeremy,

I found the JAMA article and NYT op-ed frustrating, but I also look forward to the results of the study you mentioned.

Hi Daniel,

There's always the possibility of unintended adverse consequences with public health measures. As they say, the best experimental model for a human is a human, preferably the same human you want the results to apply to. There's no way to be 100% sure what the outcome of public health measures will be without trying them-- all you can do is experimentally model as well as possible and hope your results extrapolate.

Nevertheless, it doesn't seem at all obvious to me that the dietary guidelines caused the problem.

Hi Valerie,

Added fats do not provide satiation or satiety in proportion to the calories they provide. Differentiating between satiation and satiety in this context is splitting hairs.

You wrote: "I never heard a low-carb proponent suggest that (they say we should have beed advised a low-carb diet, not a fat-rich diet). This is a strawman."

I wish that were a straw man argument! Unfortunately, some leading LC advocates have stated that fat consumption does not contribute to body fat gain at all, only carbohydrate does, therefore they most certainly also believe that if our fat intake had increased instead of carbohydrate, we wouldn't have an obesity epidemic.

You wrote: "If you're going to prescribe a diet to millions of people, you need long(ish)-term studies on humans. Anything less is irresponsible and certainly does not qualify as "solid scientific basis."

Do you disagree with the research I cited? I'd say it provides a reasonable scientific basis for the safety of LF diets at the very least, particularly when added to the longer-term studies that have been conducted.

You said "Weren't we told explicetly to replace butter with margarine? Doesn't that increase risks of obesity in every experiment?"

I don't know of any head-to-head comparisons between butter and margarine, however rodents will get plenty fat on regular butter if you give them enough of it.

Regarding the full-fat vs. low-fat dairy recommendations, it's possible that those were counterproductive from an obesity standpoint, and yes that is a good example of an area where we did follow the guidelines. It remains to be seen whether the associations reflect cause and effect though.

Hi Dominique,


raphi said...

"In the end, I believe the best way forward is to develop an eating style that allows you to eat the appropriate number of calories to meet your health/weight/well-being goals while feeling satisfied."

I am confused - what is the primary goal you suggest people aim for? Are "appropriate calories" the main factor determining "well-being"?

valerie said...

Satiation vs statiety: you were citing short-term studies. Must be satiation then, no?

I don't know of any "leading LC advocate" who says that high fat intake never leads to weight gain. Most will say that the worst diet is the "high-everything diet": high fat intake in the context of high-carb diet leads to high insulin levels and easy, rapid weight gain.

Where do you cite studies showing the safety of low-fat diets for humans? I clicked on a few of your links, and I didn't see any. Would you be satisfied with the kind of proof you provide if we were talking about a medication instead of a diet?

donheff said...

I think Daniel's analysis of the impact of the USDA guidelines rings true. Many of us for decades simply cut fats (primarily because of the CVD horror stories)and paid no attention to the added calories in the substitutes. The net effect - a long slow accumulation of pounds. This occurred in the context of a massive influx of processed foods with reduced fats but added sugar and other refined carbs. The USDA guidelines may have been well intentioned but (in retrospect) the predictable impact was that the food industry would produce the junk we find in the grocery stores.

The reality is that counting calories is extremely difficult and does not seem natural. No animal in the wild achieves balance by counting calories, their body handles that for them. By simply cutting sugar and refined carbs many of us quickly reverse that slow weight gain trend. The important point here is that that reversal involves no conscious calorie counting, it follows naturally due to the changed dietary focus. I read your (Stephen's) blog because I believe you capture the science well. Because of your writings I safely added rice, potatoes, and a bit of bread back into my diet while continuing to restrict sugar and refined carbs. For three years that has been completely successful. The important factor to me is weight maintenance now comes naturally. I pay no attention to calories or fat. Based on your writings I would guess the effect can be attributed to the satiating effect of increased protein but the way it is consciously achieved involves simply restricting sugar, added sugar, and refined carbs.

donheff said...

I think Daniel's analysis of the impact of the USDA guidelines rings true. Many of us for decades simply cut fats (primarily because of the CVD horror stories)and paid no attention to the added calories in the substitutes. The net effect - a long slow accumulation of pounds. This occurred in the context of a massive influx of processed foods with reduced fats but added sugar and other refined carbs. The USDA guidelines may have been well intentioned but (in retrospect) the predictable impact was that the food industry would produce the junk we find in the grocery stores.

The reality is that counting calories is extremelly difficult and does not seem natural. No animal in the wild achieves balance by counting calories, their body handles that for them. By simply cutting sugar and refined carbs many of us quickly reverse that slow weight gain trend. The important point here is that that reversal involves no conscious calorie counting, it follows naturally due to the changed dietary focus. I read your (Stephen's) blog because I believe you capture the science well. Because of your writings I safely added rice, potatoes, and a bit of bread back into my diet while continuing to restrict sugar and refined carbs. For three years that has been completely successful. The important factor to me is weight maintenance now comes naturally. I pay no attention to calories or fat. Based on your writings I would guess the effect can be attributed to the satieting effect of increased protein but the way it is consciously achieved involves simply restricting sugar, added sugar, and refined carbs.

Cassandra said...

I've recently become immersed in the LCHF world and, out of 26,000+ people in a single online community, not one person has ever advocated for unrestricted caloric intake. Your basis for discussing this issue is valid, but you've painted some broad, heavy strokes on what people are actually saying about this particular dietary lifestyle. I will also echo other sentiments about USDA guidelines - less to do with the guidelines, more to do with the government being influenced by companies who stood a chance at major profit gains by recommending low cost ingredients from processed plant materials and then perverting the USDA guidelines to their benefit. You could actually say the same thing about Certified Organic now and how the organic guidelines are misleading people, while the government sits by and lets it happen. It would be foolish to assume the government has no hand in what their guidelines are doing to people. Anyone I have ever seen eat low fat will inevitably make it up in sugar or toxic chemical food substitutes, at which point their health falls apart. I've never seen a healthy person on a low fat diet, ever, including my extremely fat self who grew up on a low fat diet. Not even my no soy vegan bodybuilder cousin eats low fat, because when she did, she got endometriosis and ovarian cysts with less than 20% body fat.

Char said...

Good article!
During World War II wasn't butter rationed and margarine promoted? That's how some people started eating it and never switched back. Price was probably also a factor.

If a study shows something is "heart healthy!" then it's heavily promoted by whoever would benefit from that. An example is oat bran. When a study showed it had benefits then the companies that sold oat products and the Oat Growers Council (is there such a thing?) heavily invested in promoting that information. Now we have on the Cheerios box that they are "heart healthy!" and that instant oatmeal (full of sugars) is good for you!

I guess my point is that people don't use the USDA guidelines to decide their diet but we are influenced by advertising and what is available. A person eating instant oatmeal and low fat cookies thinks they are doing the healthy thing but they are not doing what the original studies had found were beneficial. And so it gets all confusing.

I think your advice to find and eat what works for you is very good.

Stephan Guyenet said...

Hi raphi,

Well-being is one factor in the equation. The calorie intake that maximizes health and weight outcomes may be one that makes you miserable. That's why, in the Ideal Weight Program, we define "ideal weight" as a balance between your personal health, weight, performance, appearance, and well-being goals. We don't decide what a person's "ideal weight" is, but we do help people define their own goals. We provide a BMI-based default recommendation that can be modified by the user.

Hi Valerie,

Satiation is feeling full within a meal. Satiety is feeling full in the hours after a meal. Many of the short-term studies measured both.

Hi donheff,

I basically agree with what you're saying re: establishing a dietary pattern rather than counting calories. Refined/concentrated carbohydrate foods do contribute to overeating, there's no doubt about it. Added/concentrated fats do as well.

Hi Cassandra,

Are you serious? A number of LC advocates insist that calories are totally irrelevant to body fatness so you can eat as many non-carb calories as you want. Go to Gary Taubes's blog and see what you find there. Several of his posts are attempts to argue that calories are completely irrelevant, and only carbohydrate intake matters. His books also make the same argument.

If the people you interact with online think total calorie intake matters, that's wonderful, but in my experience there are many people in that community who don't accept that fact. I want to point out that I did not make any broad statements about the LC community in my post. It's a diverse community with many different viewpoints represented. I specifically argued against a particular viewpoint within the community, not against the community as a whole.

raphi said...

“calorie intake that maximizes health and weight outcomes may be one that makes you miserable”
1) “maximised health” and “misery” don’t (& shouldn’t) really belong together. It is unnecessarily & excessively reductionistic to view “mental health” & “physical health” as independent variables. Any diet/health plan assuming this false dichotomy is kinda ol’ school… [Of course, this is true for all diet recommendations guilty of this, not just yours.]

J. Stanton @ says “[…] palatability is like pornography. We all know it when we see it—but we just can’t seem to give it a rigorous definition.”
I agree & would love to see you address this explicitly (pretty please!).

Your response to Valerie regarding defining satiation & satiety is quite distinct from definitions given by J Stanton. IMO, his definitions fit better within the nutritional marraige of biochemistry and behavioural theory.

What do you think of it [not him!]?

- “Hunger is not a singular motivation. It is the interaction of four different clinically measurable, provably distinct biochemical processes:
• Satiety: Our body’s nutritional and metabolic state. It includes both our biochemical response to the absorption of nutrients, and our access to stored nutrients.
• Satiation: An estimate of future satiety, based on the sensory and cognitive experience of eating.
• Hedonic impact (“likes”): The pleasure we experience from an action. “Palatability” is the hedonic impact of food.
• Incentive salience (“wants”): Our actual motivation to obtain something we “like”. It is largely, but not exclusively, a product of the other three motivations.
Two more factors interact with hunger to modulate our food intake:
• Availability: How difficult it is to get something we want.
• Willpower: The conscious overriding action of the forebrain, known as “executive function”.

Thanks for the reply Stephan!

spughy said...

I think it's still fair to blame the government a little bit - it seems obvious in retrospect that government agricultural policies that resulted in a surplus of cheap grain commodities drove the development of more processed (and highly palatable, and cheap) foods. The food pyramid played into that to encourage the consumption of commodity grains, in whatever form - and people's own fallible brains led to the choices of those in junk format. The pyramid should have had *vegetables* forming the biggest part - but vegetables are not (for the most part) commodities upon which fortunes are made.

(And also, you shouldn't store bread in the fridge, it goes stale much faster.)

Sergio said...

Let me start by saying that I enjoy your blog a lot, thank you!

My puzzle: as far as I know, no wild mammal eats a diet with ~60-65% of the calories as digestible carbs (as is/has been the guidelines recommendations). If the intake of plant food is high, a lot of the calories come in the form of SCFAs from fibers.
Chimps eating only wild fruit have a carb intake of only ~30% of the calories, a gorilla eating grass/leaves ~20%. Wild carb sources usually have a much lower ratio of digestible-to-nondigestible carbs (fibers) than domesticated whole grains or fruits: a modern apple has ~88% carbs, brown rice ~80%, modern raspberries or blackberries (semi-wild fruit?) have ~50-55% carbs. Such a huge amount of fibers in wild plant food gets converted into SCFAs (I assumed a yield ~1.5cal/gr to get the numbers above) and should not be counted as carbs.

Another point is that human milk has ~39% carbs, and adults have smaller brain and much larger muscles than an infant (relative to weight): this suggest that an adult diet should contain less carbs and more proteins.

In conclusion, it seems to me that "ancestral" evidence suggests that any recommendation of eating >40% of the calories from carbs has to be wrong... does this make sense?

Grinch said...

I believe Jimmy Moore eats sticks of butter by themselves, which nobody would ever do if they actually believed calories from fat counted towards weight gain.

Stephan Guyenet said...

Hi raphi,

J. Stanton strives to make things unnecessarily complex, which is a common tactic used by people who have no expertise to try to wow an audience.

His definition of hunger is fundamentally confused and I'm not sure where he got those ideas. I've never encountered that definition in the scientific literature and it certainly isn't widely accepted. It seems to be a ham-handed mash-up of Kent Berridge's work with some other unidentified sources. I spoke with Berridge recently and I'm certain he wouldn't agree with Stanton's definition of hunger.

Referring to incentive salience as a "biochemical process" is kind of like Deepak Chopra explaining the relevance of quantum mechanics to philosophy or religion. Biochemistry is involved in all of our body's functions, of course, but you don't study incentive salience using biochemistry. This is another attempt to wow the audience with big words that are out of context, and it reflects a lack of understanding of the subject.

Hunger is simply a motivational state and a physical sensation. Eating because the food tastes good ("hedonic impacts") is not the same thing as eating out of hunger. You don't eat pumpkin pie at the end of a Thanksgiving meal because you're still hungry. You eat it because of a separate, non-hunger motivation, which is the reward/palatability value of the food.

Stanton and others have tried to critique the palatability concept because it's subjective. The best way to determine if something is palatable is to ask the person eating it if they like it. By that same logic, we could criticize the study of depression or PTSD. Is depression a worthwhile thing to study? Is it important? Of course it is. Can it be studied scientifically? Of course it can, and it has been.

Palatability can be measured in humans, and it has been measured countless times in a research setting. Typically this involves a "visual analog scale" measure where you ask people to quantify how much they like a food. It's subjective, but quantifiable, and if the experiment is properly designed you can do stats on it, come to reliable and replicable conclusions, etc. just like any other measure.

When you boil it all down, J. Stanton and others are simply flailing around at scientific concepts that challenge their personal worldviews.

aluchko said...


The fibre that wild animals get certainly helps, but it's not really ancestral evidence for humans. Humans have been cooking and preparing our food for the critical portions of our evolution, no other animal can claim the same.

And breast milk is designed to grow babies into toddlers, I'm not sure why we'd expect it to be a basis for a healthy adult diet.

As for ancestral populations there's lots of thin non-industrial populations that eat a lot of carbs. More relevantly if a rice-heavy diet did cause obesity we'd expect to see high obesity rates in Japan and China.

John said...

Look around at the fat people in restaurants, or watch people who claim to be dieting but aren't losing weight through the day.

They eat plenty of fat and carbs.

The addition vs subtraction comparison is particularly useful for illustrating this. When governmental recommendations suggested low fat, if that contributed to obesity its because the future fat people started eating low fat fare in addition to the fat they were already eating.

People ponder endlessly on the mysteries of macronutrient contribution to bodyweight. I think I remember Jimmy Moore's experimentation with "safe starches" involving 3 tablespoons of butter.

I remember when I, looking for easy and healthy alternatives to the low fat diet I followed that got me below 10% body fat. Switching to calorie unrestricted low carb, my ability to ignore disappearing abdominal vascularity, decreasing muscular definition, and decreasing gym performance, was astounding. Almost as fascinating was my ability to push from my mind "if carbs are so innately fattening, then why was I so comparatively lean eating 60% of my calories as carbs, following a bodybuilder style diet used by countless lean people?"

I know that I sure loved eating tons of fat on my steaks and vegetables, though. And getting to believe that this meal of the fat man's fantasy was actually the epitome of healthy eating was like living in a wonderful dream.

Well once I got back to reality and realized macronutrient managed hedonism wasn't magical, either with or without calorie restriction, I was able to figure out a diet that actually kept me lean, while allowing me to eat with other humans in normal social ways.

At least I don't think saturated fat is per se bad anymore!

WilliamS said...

"Go to Gary Taubes's blog and see what you find there. Several of his posts are attempts to argue that calories are completely irrelevant, and only carbohydrate intake matters. His books also make the same argument."

I don't believe GT has ever argued that calories are irrelevant to weight gain. I would like to see such a citation. In fact, quite the opposite, he has always insisted that notion is incompatible with the laws of thermodynamics.

What he does hypothesize (and only that—NuSi is dedicated to testing the hypothesis) is that the normally assumed direction of causality (calories in/out determine weight) is reversed. Perhaps instead, weight determines calories in/out, i.e., the body regulates appetite, metabolism, and activity to make calories in/out match hormonally-driven body mass changes. GT hypothesizes that industrial carbs, possibly mainly sugar, is the main factor that has shifted the hormonal milieu toward fat deposition via elevated insulin, thus necessitating eating more and moving less, just as hormonally-driven *vertical* body mass increases during normal childhood growth result in ravenous teenagers. Who would argue that kids grow because they eat too much? Clearly, they eat a lot because they are growing.

You many not put much stock in this hypothesis, which GT always notes that he revived but did not create, but it is quite the opposite of claiming calories are irrelevant to body mass changes. Anyone making such an argument would have to upend thermodynamics, which would require much more hubris than even GT can muster.

R said...

I've heard an idea floating around that our bodies may up-regulate fat gain in response to inflammation.

I can't remember where I heard it, which may strongly suggest it's a load of mumbojumbo. But it certainly stuck in my head.

Any thoughts?

Unknown said...

Hi Steven,

Nice piece. I just want to add a comment regarding government corruption that occurred. If you have not read about Luise Light and her story, here is a link to an article on my site about her (

Additionally, I remember growing up and looking at the food pyramid. It was in school, it was at the store, it was everywhere. Can we please recall with me what was on the bottom row?

I just looked at it again (Cereal, Pasta, Bread, Rice and crackers). Doesn't that sound a little processed to you? Where would be get those processed products from if not for big food?

It just so happens that big food did exactly what the government asked them to do, eat our grains and ensure it was low in fat.

Did big food hijack the guidelines of was our government complicit in their recommendation?

Unknown said...

Stephan, I am so sorry I spelt your name wrong in my first post, please correct if possible. Sorry again.

raphi said...

Hi Stephan.

Grand! We agree on Deepak Chopra & “pseudoscientific” being quasi interchangeable. However, I do not wish to give those in the practice of spewing nonsensical pseudoscience too much of a bad name…

Not to be provocative, but have you never smoked a joint? Straight from the paper of your friend Kent Berridge (Advances in the neurobiological bases for food 'liking' versus 'wanting' - PMID: 24874776): “Related evidence has demosntrated endocannabinoid stimulation in the NAc hotspot and even GABAergic herperpolarizations in the same hotspot can also enhance 'liking' reactions to sweet tastes.” - Biochemistry, no? (amongst other methods).
Moreover, “the precise relation between intra-cellular mechanisms and 'liking/'wanting' effects still remains to be clarified.” - Hence they ARE studying intra-cellular mechanisms that AFAIK are still largely biochemical processes.

Furthermore, you say “you don't study incentive salience using biochemistry”. Why not?

I fail to see how it can be both relevant (“Biochemistry is involved in all of our body's functions, of course”) and yet remains unused [“to study incentive salience”] (or ‘should’). It certainly isn’t the only lens one can use to study incentive salience but it is a decent lens nonetheless.

Moreover, we use biochemistry to study ’subjective’ aspects of human behavior as well as what we tend to call the mind. Look at psilocybin research by Dr. Robin Carhart-Harris & Dr. David Nutt at Imperial College London clearly demonstrating “that entheogenic properties appear to be specific to 5-HT2AR agonist classic psychedelics and this suggests a key role for this receptor in their genesis” (PMID: 24550805). They use “arterial spin labeling (ASL), an fMRI technique” & and “Positron emission tomography (PET) studies in humans found positive correlations between 5-HT2AR binding and trait neuroticism” (yes yes correlation not causation).
But you get my point - how can biochemistry NOT be involved?

You say: “Hunger is simply a motivational state and a physical sensation”. It’s intent maybe be stupid “simple” (EAT) but how it comes to manifest in our consciousness as well as physiology is faaaar from it.

I’m not disputing the merits of AVAS. I would use it. I wouldn’t however put all my eggs in an AVAS or DVAS basket. I’m also aware of its signal-to-noise ratio compared to other more modern methods [The use of visual analogue scales to assess motivation to eat in human subjects: a review of their reliability and validity with an evaluation of new hand-held computerized systems for temporal tracking of appetite ratings. PMID: 11103211]: “VAS correlate with, but do not reliably predict, energy intake to the extent that they could be used as a proxy of energy intake. They do predict meal initiation in subjects eating their normal diets in their normal environment […] It appears that VAS are best used in within-subject, repeated-measures designs where the effect of different treatments can be compared under similar circumstances. They are best used in conjunction with other measures (e.g. feeding behaviour, changes in plasma metabolites) rather than as proxies for these variables.”

It’s disappointing to see you hold the work of GT and J Stanton in such poor regard. Hopefully you’ll start noticing how some of their work is not entirely mutually exclusive to yours. For the record, GT has always said calories count but other things count more. Hopefully this can be mutual ground on which to stand on?

Mark said...

The argument that I have seen for unrestricted low carb eating is based on the idea that, while calories may not be irrelevant, the satiating effect of protein and fat, and the lack of blood sugar spikes, makes it nearly impossible to overheat on such a diet. While this may have some truth - nobody, as Dragline said, can eat 50 eggs - I think it is not borne out in practice. It is exceedingly easy, for example, to eat several hundred excess calories a day, without diminishing mealtime appetite, by grazing on calorie-dense palatable high-fat items like nuts.

Kim said...


Taubes has also stated that people who are eating low amounts of calories can remain obese (ie. the obese Pima woman with her starving baby example), but this idea has been thoroughly debunked in metabolic ward experiments. Taubes' response there is that its the reduction in calories that naturally occurs during calorie reduction that actually causes the weight decrease.

Also I've heard plenty of low carbers (not taubes) suggest that laws of thermodynamics are irrelevant to the human body because its an open system. However this is ludicrous because experimentation has uncovered all the reasonable paths of energy expenditure and ruled out the idea that we can throw away the laws of thermodynamics.

Stephan is absolutely right when he says the low carb community largely thinks calories are irrelevant. At the same time though, i would say calorie reductionism is nonsense, as calorie intake needs to sufficient for weight change to occur, but its not enough to explain why some people eat a lot of calories verses others when they both come from a similar environment.

raphi said...

What the "LC community" says or does is not equivalent nor can it be substituted with the science exploring LC eating styles. Conflation here is unhelpful. This applies to any 'diet community', of course.

Stephan Guyenet said...

Hi raphi,

What Kent Berridge does is psychology, pharmacology, and neurobiology, not so much biochemistry. You wouldn't find someone in a biochem department studying incentive salience (or rarely). Biochem is more about how enzymes work, the crystal structure of proteins, the chemistry of metabolic pathways, that sort of thing.

People who study incentive salience, hedonics, etc. tend to be in the dept of psychology. Berridge himself is a "professor of psychology and neuroscience". Again, biochemistry underlies everything, so on some level everything is biochemistry, but the effects of marijuana for example would be studied primarily as pharmacology and psychology, which is a different set of techniques.

You asked why one wouldn't use biochem to study incentive salience. There may be some instances where it would be relevant, but generally the problem is that the scale is wrong. To illustrate the principle, consider this slightly more extreme example. You can try to use quantum chemistry to study psychology as well, but it won't get you very far. Yet everything, including psychology, is ultimately the outcome of subatomic physics (quantum chemistry). You have to use tools on the appropriate scale to study the phenomenon in question. Quantum chemistry is great for understanding how atoms and simple molecules behave, but it has little to say about why we're drawn to cake.

Similarly, researchers don't refer to incentive salience or hedonics as a "biochemical process" any more than they refer to it as a "quantum chemical process".

Sorry my opinions on GT and Stanton disappoint you. I think if you were in my position, watching the life's work of some of the most brilliant people in the world get mangled by people with no background or expertise in the area and an ideological axe to grind, you'd feel the same way.

Stephan Guyenet said...

PS- I just Googled "incentive salience" and "biochemical process". The only two links I found that referred to incentive salience as a biochemical process were J. Stanton's website, and this comment thread, haha.

raphi said...

Hi Stephan,

I think your Google machine is broken. Try with “biochemistry”. I got about 4,430,000 results.

- The incentive salience theory by Kent Berridge
Unsurprisingly, he is intermingling behavioural theory/psychology, pharmacology, biochemistry, neurology and so on.

- From prediction error to incentive salience: mesolimbic computation of reward motivation (Berridge KC) PMID: 22487042
“[…]Here I discuss how mesocorticolimbic mechanisms generate the motivation component of incentive salience. Incentive salience takes Pavlovian learning and memory as one input and as an equally important input takes neurobiological state factors (e.g. drug states, appetite states, satiety states) that can vary independently of learning. NEUROBIOLOGICAL STATE CHANGES CAN PRODUCE UNLEARNED FLUCTUATIONS OR EVEN REVERSALS IN THE ABILITY OF A PREVIOUSLY LEARNED REWARD CUE TO TRIGGER MOTIVATION. Such fluctuations in cue-triggered motivation can dramatically depart from all previously learned values about the associated reward outcome. Thus, one consequence of the difference between incentive salience and learning can be to decouple cue-triggered motivation of the moment from previously learned values of how good the associated reward has been in the past. Another consequence can be to produce irrationally strong motivation urges that are not justified by any memories of previous reward values (and without distorting associative predictions of future reward value) […].”


- Incentive Salience and the Transition
to Addiction,%20Robinson%20&%20Berridge%202013%20Incentive%20salience%20&%20transition%20to%20addiction%20chapt.pdf

And so on.

You say: “You have to use tools on the appropriate scale to study the phenomenon in question.”
I agree. One describes single electron transfer in biochemical redox reactions using particle analogies - because it is convenient and appropriate. Yet, more cumbersome but accurate models exist to do so in terms of wave function properties [obviously I know next to jack sh*t about quantum mechanics]. Biochemistry remains right at the befuddling intersection of quantum effects & ‘up’. I mean - what is PET scan after all…? Data obtained via quantum mechanical probing used to inform biomedical therapies.

Look at van der Waals’ ‘Ideal Gas Law’. It is neat, nimble and accurate enough in simple circumstances. The non-ideal one has more variables & thus complexity to it but remains accurate where the ideal one fails.
Similarly, health or weight loss mechanisms require additional variables than “a unit of energy” to describe them more accurately. And yes, thermodynamics is real ALWAYS.

Stephan Guyenet said...

Hi raphi,

Maybe I missed it, but where in your Googling did you find a researcher who described incentive salience as a "biochemical process"? I wasn't able to identify a single instance, and you didn't cite one either, so I have to think my Google machine is working just fine after all.

There is nothing in the quote you posted to suggest that Berridge refers to incentive salience as a biochemical process. In fact, nothing in that quote is remotely connected to biochemistry and I don't see the relevance of that quote to any aspect of this discussion.

All the quote says is that an organism's current state can influence its motivations, even those that have been previously learned. Berridge's classic example, which he explained to me when we spoke, is that rats that have been taught to avoid (unpalatable) super salty water will suddenly relish it if you make their body think it's salt-depleted. Makes pretty good sense to me. What's your point in quoting this?

Stephan Guyenet said...

While I'm on the subject, J. Stanton also doesn't understand what "incentive salience" means. He states that incentive salience is a factor in determining hunger, but that is so wrong it makes my head hurt.

Incentive salience ("wanting" as Berridge calls it) is basically just a fancy word for motivation. Hunger is a TYPE of incentive salience in which the incentive is food. When you're hungry, a food INCENTIVE is very SALIENT (motivating, relevant, attention-grabbing). A simpler way of saying the same thing is that hunger is a type of food-directed motivation.

If you're thirsty, that is a type of incentive salience in which the incentive is water. In other words, thirst is a type of water-directed motivation. Etc.

There are several different motivational states that can lead to eating or not eating, and hunger is one of them. Social factors, palatability, and cognitive goals are other motivators that can influence food intake. These interact with hunger/satiety to determine overall food motivation, but they are also partially independent of hunger.

J. Stanton confidently tells us what causes hunger, but his explanation bears no resemblance to anything I've seen in the scientific literature. I don't think any researcher would say we know exactly what causes hunger, because there's still a lot to be learned about it. These are the best candidates I'm aware of:

1. Circadian entrainment. You become hungry around the time when you're used to eating. This may relate to an increase in ghrelin.

2. Low energy status, based on your brain's estimate of your needs vs. recent intake and fat stores.

3. Exposure to food cues like smells and tastes of things you like.

raphi said...

Hi Stephan,

I wasn’t looking for researchers describing incentive salience as a biochemical process. It isn’t. It is more like a property of motivational states.

Yeah, that salt-depletion example is pretty cool.

My point is to highlight the importance of biochemical processes underlying Incentive salience and other things which have to do with reward and behavior in general.

Although incentive salience is not described in biochemical terms, biochemical events significantly modulate it. [Hence my inclusion of the capitalised sentence taken from Berridge’s abstract]. Part of what determine these “neurobiological states” are things we quantity (like calories) and ones we assess on the basis of qualities (mood affects stemming from endogenous microbial population dynamics). They’re fundamental biochemical considerations, forcing us to conclude that biochemistry at least partially illuminates certain mechanisms governing motivational states.

,You say: “I don't think any researcher would say we know exactly what causes hunger, because there's still a lot to be learned about it”. J Stanton makes no claim to having ‘closed the case’ on hunger. He is providing a conceptual framework that doesn’t ignore metabolic processes and states for example, known to affect mood, behaviour, performance and the million other variables out there.

Furthermore, you also says: “his [J Stanton] explanation bears no resemblance to anything I've seen in the scientific literature”. Unfortunately the literature on this topic is more akin to a grave-yard with dead-theories roaming about it and certainly has yet to offer much in the way of effective therapies. So, I don’t interpret a missing echo from the field as a thumbs down.

Consider Sam Harris’s words @ :
“[…] My interest is in the nature of reality—what is actual and possible—not in how we organize our talk about it in our universities. There is nothing wrong with a mathematician’s opening a door in physics, a physicist’s making a breakthrough in neuroscience, a neuroscientist’s settling a debate in the philosophy of mind, a philosopher’s overturning our understanding of history, a historian’s transforming the field of anthropology, an anthropologist’s revolutionizing linguistics, or a linguist’s discovering something foundational about our mathematical intuitions. The circle is complete, and it simply does not matter where these people keep their offices or which journals they publish in.”

I think you’re leaving some valuable conceptual avenues untapped due to preconceived notions about what fields should study what and how.

aluchko said...


I think you misinterpret Sam Harris's essay. He's arguing against researchers being restricted to their disciplines. This makes some sense as a problem may not be best approached by its traditional discipline. For instance a biologist may have great difficultly building a model of a biological process, a computer scientists might find that same task routine by using standard computational modelling approaches.

But you're claiming something different. That an outside researcher can outdo a discipline at its own game, and that they are correct even after researchers from the discipline itself has rejected the hypothesis.

This is more like a physicist arguing with a cardiologist over diagnosing a heart problem. In this case the physicist is probably wrong because the don't have the necessary background information. The physicist might know enough to make a convincing argument, but they lack the expertise to know which parts of the argument are wrong.

Gretchen said...

Stephan, What is the minimum number of excess calories that could be detected (via weight gain or loss) in a metabolic study in a reasonable amount of time?

It would be interesting to tell people to keep on their current diets (using people with different dietary approaches) and give them a pill they were told was a nutritional supplement but was actually concentrated calories, either fat or carbs or both depending on their current diet.

Whether or not you could get the XS calories in a capsule would depend on how many you'd need to be able to detect a difference in weight.

Richard said...

As someone who has been on a low carb diet for about five years... It works. But if calories are kept constant, it may not work. The idea is, as I understand it, to eat fewer calories based on the satiety provided by eating fats, as compared with larger amounts (by volume and calories) of carbohydrates required to get the same sense of "fullness." Second, there is more to this than just obesity: There are coronary and heart and liver issues caused by not eating SFA.

However, after five years, I have gained weight, after an initial loss. I am not distressed, because as far as I can tell, it's mostly muscle. Even at age 69.

I have continued to exercise 4-6 times a week with both some aerobic and some anaerobic (HIIT) exercise, almost always with as much intensity as I can manage. The result is about as would be predicted. I gained a fair amount of muscle.

My thought on a lot of this is related to "Ancestral Diets" in the sense that almost all such diets contain a significant consumption of fermented foods, such as sauerkraut, natto, fermented fish or shrimp, even ketchup. That is a big part of what has been lost in modern diets, and the result is arterial and other calcium deposits caused by insufficient Vitamins K2 and D. Proper supplementation seems to remove those deposits. It's not too late.

raphi said...

raphi says:

Hi aluchko,

You says: “He's arguing against researchers being restricted to their disciplines”
I am not misrepresenting Sam Harris’s essay. I am saying that Stephan’s field of research would likely benefit from such an approach because it has been failing for a long time to provide practical applications. Yes, it is a general statement but should not be so controversial.

The argument that cardiologists are better at diagnosing/treating heart issues than physicists doesn’t really…say much. Also, it naively ignores how technologies stemming from physics research benefits the ‘cardiologist’ immensely (insert obvious examples).
On the other hand, the cardiologist nowadays can only provide primitive emergency solutions to the physicist (generally, of course).

Again, not all fields are doing equally well! Lets take that as a hint.

Historical scientific breakthroughs were often multidisciplinary, precisely because each discipline tends to bring less bias with it to intractable issues. It is not magic bullet but the scientific process really seems to like such approaches (historically, at least).

Jane said...

J Stanton did make one very good point, which is that palatability changes according to micronutrient status. He found an old Roger Williams paper showing that rats on a typical human diet with lots of refined carbs ate much more of the sucrose they were offered than rats on the same diet supplemented with micronutrients.

This result changes things. It means that we can't blame the obesity epidemic on manufacturers increasing the palatability of their products, at least not solely. Micronutrient removal from carbohydrate foods must take some of the blame. I cannot understand why scientists working on obesity and palatability take no interest in Roger Williams' remarkable result.

Sergio said...


I'm puzzled by Stephan statement that "the low-fat diet recommendations of the 1990s have a solid scientific basis".

These recommendations
contain the (in)famous food pyramid, whose base reads "Bread, Cereal, Rice, and Pasta", and suggest ~65% carbs, ~10% proteins.

These numbers are in stark contrast with what paleolythic human have been eating (20-40% carbs, ~20-30% prot), and it's not just about meat, even vegetarian gorillas (~16%carbs, ~25% prot) and quasi-frugitarian chimps (~25-30% carbs, ~20-25% prot) have similar percentages. Actually no mammal wild diet is a high carb diet, as far I know.

The point I wanted to make in my post is that evolution is part of science, and tells us that such a high carb, low protein recommendation is going against our genes and Nature, in my opinion, such a recommendation doesnt really have a "solid scientific basis", it looks more like a large scale experiment.

Stephan Guyenet said...

Hi raphi,

You said "Stephan’s field of research would likely benefit from such an approach because it has been failing for a long time to provide practical applications". This is a common misconception.

Obesity research has uncovered a tremendous amount about the mechanisms of obesity. This understanding has led to a number of effective therapies, which validate the knowledge that has been gained. Bariatric surgery is extremely effective at reversing obesity, and we have a number of effective anti-obesity drugs. Sadly, most of these drugs have unacceptable side effects and are not available for medical use, but the fact that they work validates the existing models we have.

The problem with behavioral approaches is that researchers can't put a gun to your head and make you behave in a healthier way, and they can't control your food environment. That's why this argument against obesity research, used by Taubes and co., falls flat. Do they expect that researchers are going to magically solve their problems by putting something in the water supply? And if they aren't able to do that, obesity research has failed?

Let's assume for the sake of argument that Taubes/Lustig are right, our current "paradigm" is all wrong, and sugar is the main cause of obesity. Do you think that knowledge will stop people from eating sugar and we'll all become lean? Of course not. People already know sugar is unhealthy, yet they continue to eat/drink large quantities of it. Researchers are already telling people to eat less sugar. Researchers can't put a gun to your head and make you behave in a healthier way. Researchers can't stop the onslaught of processed food, restaurants, vending machines, commercials, etc. Holding researchers to that impossible standard is just another way that Taubes and co. dismiss an entire field of research rather than taking the time/effort to understand it.

Hey, this would be a good topic for a post.

aluchko said...

ralphi, I don't disgree about the value of interdisciplinary research, but I disagree on the reason it works. It has very little to do with bias except that the insider researchers stick to the tools of their trade. It works because the outsiders are experts

The outsider has a vast field of expertise lacked by the insiders. And the problem in question needs that outside expertise. It works because the cardiologists ran into a physics problem and didn't have the expertise to recognize it as a physics problem.

What you're arguing with J. Stanton is the opposite, he has no outside expertise or discipline. The only thing he brings to the table is a strong belief in paleo which is the worst thing to have in science.

I may be mistaken but I think Stephan has shown papers that indicate micronutrients don't affect hunger in the past. The fact that Stanton has to dig up a possibly erroneous result from 1976 is telling.


I pointed out previously the wild animal data may not be relevant because we have evolved on cooked food.

Otherwise I don't know how accurate the paleo macronutrient ratios are or how much they varied among populations. But the fact we have different ratios now doesn't necessarily mean that's a bad thing. Stephan has pointed out many thin and healthy pre-industrial high carb populations in the past. Doesn't that indicate the ratios aren't the problem?

raphi said...

Hi Stephan,

First off thanks for the back & forth. Secondly, I read your papers whereas you do the the actual research. This gives you unique insights, granted. With that in mind, it seems you’re maybe missing the forest for the trees (si je me permet).

For the sake of discussion, lets say the existing models in your field (which you have in mind) are well validated by these drugs, most of which have “unacceptable side effects and [that] are not available for medical use”. What are we left with, in terms of practical applications that have the slightest hint of potential success? Nothing much beyond your 2 examples; bad drugs & hard-core surgery!

I may put more stock in the value of ‘mass education’ - especially in this Internet age. It comes down to this: telling people to avoid sugar is certainly a better practical application for impacting obesity than are weight-loss drugs & bariatric surgery (so far). Hopefully we share this conviction.
On the other hand, is it a 100% technically accurate, single statement precisely identifying all causal factors in this obesity epidemic? No, of course not. It was never intended to be so.

You point to methodological difficulties encountered when attempting to test your theories. Fair enough.
However, either these limitations are overcome or another hypothesis & test need be sought. Taubes (& others) simply believe answers are more readily obtained elsewhere. And consequently that some of the questions are best reframed in terms adequate to this alternative approach. You may disagree and you may well be right. Nevertheless, until something more substantial from your field emerges, I think this "we know better" vibe feeds narrowed thinking.

Regardless, ”It doesn't matter how beautiful your theory is, it doesn't matter how smart you are. If it doesn't agree with experiment, it's wrong.”
Richard P. Feynman

PS: The only thing I won't stand for is you not confirming you're a fan of Feynman! :)

raphi said...

Hi Aluchko,

"a strong belief in paleo which is the worst thing to have in science".

IF you're referring to the social "Paleo" movement then there is some truth to it. Although you're exaggerating entirely.

IF you're referring to evolutionary theory applied to biological systems, then I might as well be debating a FlatEarth-Creationist and will definitely abstain.

Good day, sir (,d.d2k&psig=AFQjCNE7lGj6zCRrGXA373CFdwtIm5OCgQ&ust=1402255085966321)

Stephan Guyenet said...

Hi Sergio,

I'm not arguing that the 1990s USDA dietary recommendations were optimal. I'm simply arguing that there was a reasonable amount of evidence to support the concept that a lower-fat diet would be more effective for calorie control than a higher-fat diet. Recall that Loren Cordain's original Paleo diet called for lean meats. Paleolithic humans in Africa probably ate fairly lean diets, and certainly not high in animal fat because most African animals are extremely lean. The only African hunter-gatherer group I'm aware of that has a higher-fat diet is the !Kung San. They get a fair amount of fat from nuts, although the mongongo fruit/nut is mostly carbohydrate by calories (a fact that was recently pointed out to me by Staffan Lindeberg).

Hi Aluchko,

Regarding outside expertise, well said. We like to think that anyone can do anything if he just has enough gumption, but it simply isn't true. The likelihood of a fantasy novel author making a scientific breakthrough on his Paleo blog is exceedingly small, particularly when he fails to grasp basic concepts. Similarly, the likelihood of a neurobiologist being able to pilot the space shuttle into orbit is exceedingly small.

Regarding the effect of nutrient deficiencies on appetite, they decrease food intake and body weight in most studies-- the opposite of what some people claim. The study Jane/Stanton cited showed that rats drink different quantities of sucrose under different dietary conditions, but the study did not attribute the effect to changes in palatability or even attempt to test that hypothesis. The authors didn't even discuss the possibility that the result was relevant to palatability. In other words, we don't know why the rats drank different quantities of sucrose, so this study is not directly relevant to the idea that micronutrient status influences palatability.

It's an isolated study and the results are interesting but quite difficult to interpret. Apparently they just sprinkled micronutrient powder onto the rats' food, which could have made it less appealing. I don't know if you've ever bitten into a multivitamin pill but it tastes gross.

For example, in the methods they state: "magnesium oxide was replaced with magnesium carbonate after 8 days, because it appeared to cause slow darkening of the food, an elevated pH, diarrhea, and reduced palatability and weight gains".

They only measured food intake over less than 10% of the experiment, so it's impossible to know what overall food intake was. However, the micronutrient-supplemented group ended up heaver than the unsupplemented group, suggesting that their overall food intake was probably higher. They didn't report body composition.

Overall, an interesting study but it seems to have spawned some wild speculation. It would be useful to measure food intake over the entire experimental period, measure fat mass, and measure the palatability of the different diets and sucrose. Then perhaps the result would be relevant to the claims that are being made about it.

aluchko said...


The problem is that J Stanton isn't investigating nutrition to better understand nutrition. Instead he's investigating nutrition to find evidence supporting paleo. As a result he's going to find evidence supporting paleo whether it's there or not.

That's what I meant when I said his strong belief in paleo was the worst thing he could bring to the table. Anytime you approach a question heavily invested in an answer you're going to find evidence supporting that answer.

To throw another Feynmen quote back at you:

"The first principle is that you must not fool yourself — and you are the easiest person to fool"

Jane said...

The reason Roger Williams' experiment is so important is that it could provide an explanation for the inflammation found in the hypothalamus of obese people and rodents on an obesogenic diet. In the rodents, it appears before obesity does, suggesting injury to the hypothalamus is what causes the obesity.

If the inflammation is indeed due to injury, and the injury to oxidative stress, which also seems likely, micronutrients with antioxidant activity might prevent it.

I'm not talking about antioxidant vitamins, which don't work because they lower ROS too low. I mean the ones that keep ROS at just the right level, which we know is critical for the proper functioning of the satiety system.

These particular micronutrients are also important in maintenance and repair, which obviously isn't working properly if the brain is inflamed.

Edward said...

I think you are being too generous to the early USDA guidelines regarding sugar. In the early 1980s, my wife worked as a biochemist at a pharmaceutical company. Her well known employer evangelized the USDA nutrition recommendations, holding lunch time nutrition recommendation talks, presented by professional nutritionists. The employer encouraged staff to invite spouses and family to attend these talks (so I was in attendance too).

The overwhelming message was that we must cut back on fat consumption, sharply (in retrospect, they assumed all of us were consuming too much fat without considering our actual individual consumption).

A person sitting next to me asked, "So what you are saying is that except for tooth decay, we do not need to worry about sugar consumption?"

The reply from the nutritionist was "The major health hazard from eating sugar is tooth decay. As long as you brush your teeth, sugar consumption is not a serious problem". That is how the USDA guidelines were interpreted by the professionals. That answer was followed again by another statement on the importance of avoiding as much dietary fat as possible.

This web site ( has a scan of the 1980 USDA Dietary Guidelines written for the general public, which says the same thing.

Regardless of what the science or the USDA intended to tell us in the early 1980s, the actual message delivered to the public was that consumption of sugar was not something for us to worry about.

Related to this was that were also told that we needed to avoid consumption of saturated fat (meat), cholesterol (eggs), leaving us to eat grains, sugar and seasonal vegetable.

Only in recent years, thanks to high energy consumption food transport across huge distances, do we eat fresh vegetables year around in northern latitudes. Frozen foods were mostly non-existent until WW II when frozen food was expanded for delivery food to the forces overseas.

The actual message, and as applied in practice, was to eat a grain-based, high starch diet, with little meat or eggs. And not to worry about sugar as long as we kept our fat intake down.

Again, regardless of what the USDA intended, the above is the message delivered by professional nutritionists in the early 1980s.

Anonymous said...

"I also reviewed the evidence suggesting that carbohydrate and fat are equally fattening in humans, calorie for calorie. "

NOOOOOOOOOOOOO!!!!! That's an extreme over-interpretation of that evidence. What it really shows is that high carb moderate fat and high fat moderate carb diets are equally fattening. When there are studies showing that overfeeding with pure carbohydrate or fat causes identical body fat gain, then you're entitled to make that statement. But those studies will never exist because you yourself have conceded that de novo lipogenesis is a minor pathway in humans. That makes fat more fattening than carbohydrate on a calorie for calorie basis. It amazes me how few obesity "experts" including yourself eschew the most obvious diet for combatting obesity, which is a very low-fat one.

Mikko Järvinen said...

I wonder why the high carb advocates keep arguing that calorie is calorie? Yes, it most likely is in a lab setting, but in real life we see that many formerly obese have been finally able to lose weight with low carb, which was previously impossible.

I'm not obese, but I always start gaining weight, if I start to eat "normal" amount of carbs. Yes, I do overeat carbs, but it's just so much easier to over eat carbs than fat and protein. I can easily stuff a bag of crisps and liter of ice cream in my face, but getting the same amount calories from fat and protein from meat, avocados, coconut oil, olive oil is nearly impossible.

The whole argument is like saying: You get equally drunk drinking beer and moonshine. Yes, if the amount of alcohol is the same, then yes. But in practise, most people would not like to drink moonshine as much as they would drink beer. Also, moonshine is not available in all restaurants, bars and at a friends fridge.

I'm not saying that meat and fat would be repulsive, thus people eat it less. But you just cannot get a decent low carb meal from McDonalds, so you skip it. The result is that you just end up eating less, because the low carb food is not so readily available, on top if the fact that you generally do not even want eat fat and protein as much as you would eat carbs.

Also, when one eats carbs, instead of being in ketosis, it is a well known fact, that one experiences hunger much more often than when low carbing.

In the end you might say: so carbs and fat are equally fattening, you just need to control your carb intake.

Yes. If you are able to control your carb intake, you most likely are lean to begin with and to not need to low carb or be in ketosis. But if you simply have poor ability to refrain from indulging yourself with carb loaded foods, it IS easier to lose weight and stay lean by low carbing.

Sergio said...

Hi Stephan, thanks for your reply.
You certainly know much more than me about this topic, nut let me reiterate my point. I understand that tropical animals are very lean, but eating them "nose to tail" would still provide ~60%fat-40%prot (in calories) from animal food. Most contemporary hunter-gatherers obtain significant more energy from animals than from plants.
Another point is that wild plant food (except for seasonal honey) seems to contain about the same amount, in grams, of digestible carbs and fibers. This seems to imply that digestible carbs percentage can never go above a maximum of ~~40% of the energy: Say we have a 2500cal/day diet. 40% --> 250gr of digestible carbs/day. It's very hard with wild plant food to get 250gr of carbs with less than ~~200gr of fibers/day. Such an amount seems huge to me for humans, that do have a small colon (w.r.t. quasi-vegetarian primates for instance). It's like trying to get a high carb diet eating ~5kg (11lbs) of vegetables/day.

Also, we know as a fact that human milk provides a bit more calories as fat than carbs, shouldn't this count as scientific evidence that a reasonable guideline should recommend, if anything, a bit more fats than digestible carbs?

Aluchko, as far as I understand cooking only turns resistant starch into digestible starch, but all starch is already counted as "digestible carbs". Please correct me if Im wrong here.

Robert said...

Many factors likely bear responsibility for our obesity/diabetes epidemic but I believe that the USDA does compose a "slice" of the responsibility pie. The original USDA guidelines contributed to a whole generation of Americans eating in an unbalanced and culturally novel fashion -- 6-11 grain portions a day, many refined, is culturally novel as far as I'm aware, just as Atkin's is likely biased too far towards fat/meats for many people. And I do think the relative focus on fat reduction over sugar reduction has driven the average American to become sugar-dependent (as added sweeteners are in ~80% of all processed foods acc. to Popkin). This added excess sugar calories to the American diet we didn't need. And as you've explained on Youtube, Dr. G, vegetable oil consumption has skyrocketed over the past century, partially due to the USDA's CVD focus, which kept processed fat intake stable while ref. grain and sugar intake rose.

So the average American eats less whole foods than 100 years ago partially due to the USDA convincing Americans that "nutritionism" (popularized by Pollan), and the processed foods they support, are part of a healthy diet based on the nutrients they contain. EAting based on common sense, food culture, and evolution got tossed out the window. Which is why my Mom, who struggles with her weight, thinks hyper-processed protein bars are as healthy as a salad with nuts because both have fiber and protein... Don't get me started on the 5000+ novel synthetic food additives, mostly from Chinese factories, in processed foods that lack good, long-term human data...

So the USDA doesn't get a free pass in my book.

Stephan Guyenet said...

Hi Yvonne,

I wish it were that simple. Among the carb vs. fat overfeeding studies I cited, Lammert et al. used a 11% fat diet for its carb overfeeding arm. That is an extremely low-fat diet and difficult to achieve in real life. Yet it caused the same amount of body fat gain as fat overfeeding when calories were controlled.

The reason is because de novo lipogenesis scales to meet the body's fatty acid needs. The lower the dietary fat intake, the higher DNL goes to compensate. That is why, in your words, "obesity "experts" including yourself eschew the most obvious diet for combating obesity, which is a very low-fat one".

Hi Sergio,

I don't think many African animals are 60% fat by calories. I believe the only fatty organ in most tropical animals is the brain (with a few exceptions, like hippos). Marrow is probably lean in extremely lean animals, just as it is in lean temperate animals in spring.

In fact, there are data from Loren Cordain's research that speak directly to this. In his (free access) paper "Plant-animal subsistence ratios...", he reports the percentage of calories obtained from fat from the nose-to-tail consumption of animals with different percentages of carcass fat.

For animals with 2-5% carcass fat, which would be similar to most African game, 18-35 percent of meat calories would be delivered as fat, assuming the brain was eaten.

In their seminal paper "Stone Agers in the Fast Lane", Eaton, Konner and Shostak stated that "Because game animals are
extremely lean, Paleolithic humans ate much less fat than
do 2Oth-century Americans and Europeans, although more than is consumed in most Third-World countries." Fat intake was estimated at 21% of total calories, based in part on carcass analysis of 43 species of African game animals.

Stephan Guyenet said...

Hi Sergio,

One other point. There are many wild plant foods that contain more digestible carbohydrate than fiber. For example, the mongongo fruit, which is a primary source of food for !Kung hunter-gatherers, contains 3.5g fiber and 70.2g carbohydrate per 100g. It has been described as similar to a date fruit.

Another primary source of food for the !Kung is the tsin bean. It provides 1.0g of fiber and 23.2g of carbohydrate per 100g.

Baobab is another commonly eaten fruit/nut among African hunter-gatherers. It provides 10.7g of fiber and 51.4g of carbohydrate per 100g.

In fact, browsing through the nutritional composition of wild plant foods commonly eaten by HGs in Southern Africa, nearly all of them provide much more carbohdyrate than fiber, and none provide more fiber than carbohydrate. These data are available in appendix D of Richard Lee's book "The !Kung San".

Personally I think there have been some very misleading arguments about macronutrient ratios coming from prominent people in the Paleo community. The goal seems to be to rationalize the low-carbohydrate high-fat diet using an evolutionary argument, but the African ancestral human diet was probably not high in fat.

Anonymous said...

The paleo/primal folks grossly misrepresent the diet of even the 18th and 19th centuries. Yes, the wealthy did eat a lot of meat (and suffered from cardiovascular disease), but they were a really small percentage of the population. Bread, beer, pasta, rice, potatoes and other starches made up the bulk of the diet, especially in the winter, because that was what could be stored in the pre-refrigeration era. Canning didn't become possible until the 19th century.

Every culture had its characteristic carbohydrate source around which meals were built. They needed those cheap calories because they did nothing but physical labor, needed to stay warm during the winter, walked as their primary transportation (only the rich owned horses) and had little access to fresh food for a good part of the year.

A USDA "serving" is really tiny. A sandwich, for instance, made with two slices of bread form the oversized loaves sold now might easily represent 3-4 USDA servings of grain. The USDA pyramid shows sweets sitting at the tip and advises that Twinkies, etc. be eaten "sparingly" rather than advocating their consumption as the paleoistas claim.

The Neqnderthal cooked and ate grain.

Woodcat said...

"I think this is a good time to introduce a concept that may seem abstract, but I think is very important. It's the difference between addition and subtraction. To illustrate the concept, consider three scenarios:

1 Sarah sits down to a dinner of beef stew, salad, and bread.
2 Sarah sits down to a dinner of beef stew, salad, and bread. Before eating, she adds a stick of butter to the beef stew for flavor, and pours herself a plate of olive oil in which to dip her bread.
3 Sarah sits down to a dinner of beef stew, salad, and bread. She puts the bread back into the fridge and only eats stew and spinach.

In scenarios 2 and 3, Sarah's meal will be biased toward fat and away from carbohydrate. Yet these two scenarios will lead to very different effects on calorie intake. This is because in scenario 2, Sarah is adding concentrated fat to her otherwise unrestricted meal, while in scenario 3, she's subtracting one third of the foods she would normally have eaten."

You missed scenario 4, my preferred option, Sarah sits down to a dinner of beef stew, salad, and bread. Before eating, she adds a stick of butter to the beef stew for flavor, and bins the bread.

Why do we only get the extra fat if we have the bread? Personally, I think without the fourth option, the argument is unfairly skewed.

(I eat more calories on a higher fat diet than I ever did on a high carbohydrate diet, yet I'm losing weight, my health is improved, my susceptibility to viruses decreased, and my cognitive function is significantly improved. Bread makes me tired, as does calorie restriction)

Deirdre said...

Hate to break up the bromance between Stephan and Raphi ;) , but had to make a quick comment on this …

"People already know sugar is unhealthy, yet they continue to eat/drink large quantities of it."

Oh, consumer confusion is alive and well, I'm afraid. As long as we continue to let our food marketers be our food educators, "people" will basically know nothing.

The food industry is quite clever at rebranding sugar in the forms of "healthy foods" - the whole grain movement and fruited Greek yogurt come to mind.

You're my "go-to", Stephan, but you lost me on the whole USDA "No Government role in obesity" stance. They are as thick as thieves.

Tsk, tsk ..

Richard J D'Souza Hypnotherapy Cardiff said...

Interesting research. This is the type of article that the mass population need to read. Do some obese people just ignore the advice given to them anyway?

Mikko Järvinen said...

There also seems to always be the remark that our ancestors did consume a lot of carbohydrates, which is then supposed to dismiss all of the paleo, primal, atkins etc. Like the paleo folks would have failed to realize one important thing that the high carbers have realized.

But do the high carbers realize, that while our ancestors diet might have been low in fat, their activity levels and the availability of food and carbs was entirely different.

So again to the point. As most people cannot seem to be able to control their carb intake or get adequate amount of exercise to compensate, the result is obesity.

Now there are many people that have at least semi-permanently lost weight on low carb, whereas there are only a few people that have done that in high carb.

Yes, there are the people with optimal metabolism, appetite control and hormones in general that can either control their carb intake or get adequate amount of exercise to compensate, but those are a rarity.

I'm not saying that either of these is correct or the best way to eat, but in our current environment where carbs are in each vending machine, in all fast food joints and served at your grandma's birthday, it is just a fact that carbs do make people obese.

Now if the culture were to change, that instead of whoopy pies and doughnuts people would serve beef jerky and avocado salad with olive oil and vending machines would contain bulletproof coffee and high fat pemmicans, some folks would most likely be obese even in that setting. But I still doubt that as many people as are now obese due to carbs, would be able to get obese by eating low carb.

So really, why keep arguing that if you just control what you eat, you lose weight and maintain weight, when we see that it just does not work in practise.

People need to start avoiding highly palatable foods to lose weight. Foods that contain a lot of carbs just are much more palatable than fat and protein.

I also suspect that if our ancestors would have had cornucopia of doughnuts and marshmallows, they would have been fat and miserable, as we modern folks are.

To this note - do we have any evidence that our ancestors that had an abundance of fat (coconut, olives, fatty meat, fatty fish) would have been fat?

Sergio said...

Hi Stephan, thanks a lot for your detailed reply. I have to reconsider my position that plant food has 1:1 ratio of carbs:fibers. I also have to agree that the official paleo view is skewed toward high fat.
As a side note, I also looked/googled at baobab and I found that baobab powder is 1:1 in carbs:fibers (and they claim that the powder is simply crushed fruit). Wild tubers do have a 1:1 ratio but, at least for the Hazda, the tubers are chewed and then spit out, so what goes down is a mostly starch, and only a bit of soluble fibers.

I read the interesting 2000 paper from Cordain, Eaton et al you linked. Here Im still puzzled by your numbers. Let's assume african HGs ate 50% animal food - 50% plant food (the data in that paper say maybe slightly more than 50% animal). If those animal are really that thin (25%fat-75%prot), and the plants are 62%C-24%F-14%P as they claim, the diet would be
31% carbs - 44.5% prot - 24.5% fat

The same paper explains at pag 688 "we calculated the mean maximal protein intake for an 80-kg subject to be 250 g/d (range: 212–292 g/d). For a 12552-kJ [=3000kcal] energy intake, the mean maximal dietary protein intake would be 35.1% of energy". This is because of rabbit starvation. Their solution of the puzzle is the african HG "2) hunted larger animals because percentage body fat increases with increasing body size; 3) hunted smaller animals during the season in which body fat is maximized; 4) selectively eaten only the fattier portions of the carcass, including lipids boiled from the cancellous tissues of bones, and discarded the rest". This implies that the effective contribution of animal food is more ~ 50% fat- 50% prot, and for the overall diet we get
~ 31% carbs - 32% prot - 37% fat

Im not sure what is the best name for a diet like like this (superbalanced? :), but it's definitely very high protein both from the point of view of official guidelines (that suggest 10-15%prot) and also from "official paleo guidelines" (that suggest 15-25% prot).

Now I really wanna ask you a question. In your years of blogging you seemed to carefully avoid the issue of protein intake, I would be very interested in some discussion by you about optimal/maximal protein intake for modern humans: both quantity (in gr/kg/die and in %age of calories) and quality (for instance, should we eat a ton of collagen since whole animal protein is 50%collagen?)

Stephan Guyenet said...

Hi Mikko,

I know it's commonly stated that people who exercise can tolerate a higher carbohydrate diet, and people who are sedentary should eat less carbohydrate, but what's the evidence to support that statement? I've heard personal trainers say that, but I've never seen any research on it. I respect the experience of personal trainers if they feel they've had success with that approach, but I have a hard time accepting it as a general principle without some research to back it up.

Hi Sergio,

It's possible that different analytical methods report different amounts of fiber. However, the first research paper I found on baobab indicates that it contains a lot more carbohydrate than fiber:

Protein intake can be balanced by carbohydrate and/or fat to avoid rabbit starvation. There are sources of fat in some parts of Africa, but animal fat tends to be scarce. The fattest animals, such as hippos and elephants, are often extremely dangerous to hunt with stone age tools. Most of the fat in the !Kung diet, for example, comes from mongongo nuts.

I think the average African Paleo diet would probably have been moderately low in carbohydrate compared to modern diets, but also moderately low in fat, and low in animal fat. There would have been large variability between groups, however, and all of them would probably have been selected based on their ability to survive over a wide range of macronutrient ratios. Sometimes tubers or nuts are all that's available, and if you can't survive on them for a while, you get weeded out.

Anonymous said...

Here's Denise Minger's review of wild type tropical fruits, all of which are quite high in carbohydrates without the benefit of selective breeding:

Liz said...

I very much agree with the conclusion. Ive never been able to stick with either low carb-hi fat diets, or low fat vegan diets. Also the only times I've lost weight have been through eating whatever I wanted but in much less quantity. So it was calorie restriction. I eat a lot less when I dont try to force myself to eat these extreme diets. I find a plate of mixed vegetables, fish, white rice and tomato sauce very very satiating. Ive played around adding fat to meals to see if it helps me feel satiated faster and it never works. Some foods cause me to over eat because they are very tasty and not satiating bad combo. I would put pasta, white bread, ice cream, pizza, and cookies here. The fact that pizza and ice cream are very high fat (and high carb) doesnt make them more satiating than a cookie or pasta when matched in calorie. I imagine Ice cream has more protein than cookies and thus is more satiating, but pizza often is drenched in fat (dripping in olive oil) and I can still eat the whole box, and then some. So Ive foud the key for my satiation is just eating enough at each meal, having meals be a nice mix of protein and carb, preferably with some vegetables, avoiding highly palatable, unsatiating foods (binge foods), and forgetting about fat intake, in fact just using the amount needed in cooking, or a small amount for flavor. I dont think it would be smart to just put a stick of butter on top of all your meats, and have your tubers and salads swimming in olive oil. lol Specially if you cant low-carb. Low carbing is a special situation -I find myself never entering unless forced-.

crispy said...

I started to believe that fat does NOT make you fat when I read BIG FAT LIES (Can't remember the author) . The book was referenced and the author quoted studies to prove his point.
I wonder if this author read the book and what he thinks of it.
Gary Taubes clearly showed that adding calories does not make you gain fat in the corresponding amount: if you add to your daily intake a few calories ( +11.2 cal) you will not gain 1 pound of fat per year for life!!
Also worth noting (Gary fails to mention this)gaining 454g of fat means gaining a lot more weight than 454g as fat is surrounded by tissue and water...
By the same argument , removing 11 cal (2g of carbs for ex) from your diet will not make you lose 454g of pure fat by the end of the year!!
I hope I got my figures right because i thought it was 20g of carbs not 2g/day.
454g fat x 9cal/g = 4666cal
divided by 365 days we end up with 11 cal/day
5 cal per g of carbs so about 2 g of carb (there are 20g of carbs per slice of bread)
So if the calorie-in calorie-out theory worked, reducing one's intake by 1/10 of a slice of bread would yield a loss of 10 lbs of aft in 10 years (probably 20lbs of actual weight)
By the same logic removing 1 slice (100cal or so) would yield a loss of 200 pounds over 10 years (100pounds of pure fat).
Where is the flaw??

crispy said...

I read reference (30) which shows complexe carbs are better than simple carbs yet in Wheat Belly, i think the doctor shows that whole wheat bread has a higher glycemic index than white bread and about same as sugar
I guess that is different as white bread is already a complex carb but again glycemic index are high.
I'd guess glycemic index is only part of the story.

James said...

Stephan, although I take no issue with the general drift of the post and realize that I have come across the argument before, I must at the same time admit that it doesn't seem to work for me, over a longer period of time. Since I started leaving out a lot of the processed stuff, including wheat products, but not cheeses and yogurt, beginning the day invariably with a couple of eggs-sunny side up- and bacon and a salad, I have lost quite a bit of weight, gained some later, plateaued, and then slowly lost again. At 72 standing 6 feet I am weighing 66kg and I can load up on fat and protein without gaining. I do some strength training and am fairly active. I realize this is just N=1 but it does puzzle me what mechanism could be involved when people do gain on high fat/low carb. James

Dr Bob said...

Your references do not include very low carbohydrate long term weight loss studies. Very low carbohydrate diets consistently result in spontaneous reduction in caloric intake. High Carb diets require calorie restriction. (ATOZ study, see several studies here:
I am disappointed in your selective references.
Your conclusions are based on selective quotation. As a clinician I see the results of a low carb approach and understand the endocrinology involved.
Jenkins DJ, Kendall CW, McKeown-Eyssen G, Josse RG, Silverberg J, Booth GL, Vidgen E, Josse AR, Nguyen TH, Corrigan S et al: Effect of a low-glycemic index or a high-cereal fiber diet on type 2 diabetes: a randomized trial. JAMA 2008, 300(23):2742-2753.
Westman EC, Yancy WS, Mavropoulos JC, Marquart M, McDuffie JR: The Effect of a Low-Carbohydrate, Ketogenic Diet Versus a Low-Glycemic Index Diet on Glycemic Control in Type 2 Diabetes Mellitus. Nutr Metab (Lond) 2008, 5(36).
Gannon MC, Hoover H, Nuttall FQ: Further decrease in glycated hemoglobin following ingestion of a LoBAG30 diet for 10 weeks compared to 5 weeks in people with untreated type 2 diabetes. Nutr Metab (Lond) 2010, 7:64.
Gannon MC, Nuttall FQ: Control of blood glucose in type 2 diabetes without weight loss by modification of diet composition. Nutr Metab (Lond) 2006, 3:16.
Gannon MC, Nuttall FQ: Effect of a high-protein, low-carbohydrate diet on blood glucose control in people with type 2 diabetes. Diabetes 2004, 53(9):2375-2382.
Forsythe CE, Phinney SD, Feinman RD, Volk BM, Freidenreich D, Quann E, Ballard K, Puglisi MJ, Maresh CM, Kraemer WJ et al: Limited effect of dietary saturated fat on plasma saturated fat in the context of a low carbohydrate diet. Lipids 2010, 45(10):947-962.
Jakobsen MU, Overvad K, Dyerberg J, Schroll M, Heitmann BL: Dietary fat and risk of coronary heart disease: possible effect modification by gender and age. Am J Epidemiol 2004, 160(2):141-149.
Siri-Tarino PW, Sun Q, Hu FB, Krauss RM: Saturated fat, carbohydrate, and cardiovascular disease. Am J Clin Nutr 2010, 91(3):502-509.
Int J Cardiol. 2006 Jun 16;110(2):212-6. Epub 2005 Nov 16. Effect of a low-carbohydrate, ketogenic diet program compared to a low-fat diet on fasting lipoprotein subclasses.
Westman EC, Yancy WS Jr, Olsen MK, Dudley T, Guyton JR.

Mol Cell Biochem. 2007 Aug;302(1-2):249-56. Epub 2007 Apr 20.Beneficial effects of ketogenic diet in obese diabetic subjects. Dashti HM, Mathew TC, Khadada M, Al-Mousawi M, Talib H, Asfar SK, Behbahani AI, Al-Zaid NS.

Stephan Guyenet said...

Hi Dr. Bob,

Sorry to disappoint you. I'm afraid you may have missed the point of my post though. The post is about unrestricted high-carb vs. high-fat diets, not restricted diets like low-carb diets. I touched on low-carb diets briefly and made essentially the same points you made in your comment, but that wasn't the focus of the post so I don't see why I should pepper it with a bunch of references that are scarcely relevant.

Dr Bob said...

Thanks, Stephan. But you state:
"The US obesity epidemic probably wasn't caused by advice to eat a carbohydrate-rich diet, and wouldn't have been prevented by advice to eat a fat-rich diet instead." I disagree. Advice to reduce fat intake (USDA, AHA, ADA etc) results in substitution of carbohydrates. 1/3 of Americans are overweight and 1/3 obese. 2/3 of our population would likely benefit from significant carbohydrate restriction in combination with elimination of processed foods particularly refined grains and refined "vegetable" oils. As a clinician I focus on that message because I treat disease and view our messages to the public as important. By stating that fat rich diets would not have prevented our obesity epidemic you confuse much of the public about the flip side, what is the science related to weight loss? When discussing this topic, from a public policy point of view I consider it essential to represent the science of weight loss. Our public has received too much bad advice.
The advice to restrict fat and substitute carbohydrate has been a disaster. Now in order to treat the disaster we must advocate for carbohydrate restriction in combination with ancestral foods. It is clear to me that the unscientific message from governmental and non-governmental agencies has significantly contributed to our obesity epidemic. That point is much more important and relevant. Avoidance of animal fat has resulted in disease. Now to treat disease we must restrict carbs. For the 1/3 of our population who are not obese or overweight a whole foods ancestral diet (variously defined) is just fine. Most folks can tolerate a variety of carbs coming from fruit and fresh vegetables. But once the body is broken more extreme measures must be taken.

Tobin said...

Hi Stephan,

Great post, and interesting discussions in the comments. One point that I feel is relevant--the longer-term low carb studies you mention (Samaha, Foster, Direct trial, A to Z Diet study etc, references 25-28) are not truly low carb vs. low fat. They are mostly low carb/unrestricted calories vs. low fat/restricted calories.

Given that design, you would think the restricted calorie arm would win. It is striking that the "low fat" arm did not result in more weight loss in ANY of these RCTs (and there are a dozen more at least). Even by chance you might think the "low fat" arm might win once or twice.

While I agree that weight loss likely comes down to caloric balance, these results suggest (to me anyways) that restricting dietary carbohydrate even while not consciously restricting calories is a more effective way to reduce caloric intake than the combination of restricting dietary fat AND consciously restricting calories.

I don't know how to reconcile that conclusion with the short-term studies you cite, but to me it seems like an unavoidable conclusion. Do you think that is a reasonable interpretation?

Stephan Guyenet said...

Hi Tobin,

Yes, it is a reasonable interpretation and I agree with you. Telling people to eat fewer calories of a low-fat diet isn't a very effective fat loss aid. Telling people to eat a low-carb diet without giving them any calorie advice has modest weight loss effects as well, but clearly superior to LF/CR for the "average person" (at least up to a year).

When researchers measure calorie intake using accurate methods, they find that LC diets lead to a lower overall calorie intake than calorie-restricted LF diets. As you said, there's nothing here that's incompatible with our current understanding of the relationship between calorie intake and body fatness.

The way to reconcile this with the short-term studies I cited is to recognize that the intervention is quite different. In the studies I cited, people were eating unrestricted high-fat diets, not restricting carbohydrate. In other words, they were eating unrestricted diets that focused on high-fat foods. This leads to a different pattern of intake than you get when you tell a person to restrict carbohydrate intake. For example, pizza and ice cream are high-fat foods that could be part of an unrestricted high-fat diet, but a person on a low-carb diet would typically restrict them. Adding more butter to food would be part of a high-fat diet, but a person on a low-carb diet wouldn't necessarily do that (and people in these diet trials were probably not encouraged to add a large amount of fat to their foods).

A shift toward an unrestricted high-fat diet is what happened to the US diet over the last half century, while a low-carb diet is a restrictive diet that people use to lose fat.

raphi said...

They use it to regulate their weight, appetite & health. In our current context this often involves losing excess fat.

Tobin's analysis is spot on. Your concern seems to be with the question "how can I show that ultimately the laws of thermodynamics hold" even though the answer is a patently obvious resounding YES.

"How do 'calories in' influence 'calories out' & vice-versa?" goes beyond the questions implicit in your thinking. The better questions arise from a systems thinking model. Who's to say that understanding the complexity of energy regulation within us isn't comparable to confidently predicting not only weather patterns but long-term climate cycles? Ones has to accommodate for the stochastic nature of biological processes within any proposed model as ultimately, it must conform to experimental results.

Physics doesn't lie & no one said it did.

Timbo said...

Any chance of some commentary around all this, and The China Study?

Btw, I wish people would stop referring to carbs and vending machines. Anything coming from a vending machine is likely to have loads of fat and simple sugars. HELLO insulin spike.

It's absolutely silly to compare a big plate of rice and veggies to "carbs" from a vending machine (or by using a comparison to McDonalds).

Let's get real and compare a true high carb, fibrous meal.

DaveHPT said...

Thank you for this very helpful, objective and intelligently considered article.

Unknown said...

Being overweight can make it more difficult for you to stimulate the natural HGH production. Dealing with excess weight means that you have relatively high levels of insulin. As a result, your HGH production is interrupted. this hypergh14x review