Thursday, August 25, 2011

A Roadmap to Obesity

In this post, I'll explain my current understanding of the factors that promote obesity in humans.  


To a large degree, obesity is a heritable condition.  Various studies indicate that roughly two-thirds of the differences in body fatness between individuals is explained by heredity*, although estimates vary greatly (1).  However, we also know that obesity is not genetically determined, because in the US, the obesity rate has more than doubled in the last 30 years, consistent with what has happened to many other cultures (2).  How do we reconcile these two facts?  By understanding that genetic variability determines the degree of susceptibility to obesity-promoting factors.  In other words, in a natural environment with a natural diet, nearly everyone would be relatively lean, but when obesity-promoting factors are introduced, genetic makeup determines how resistant each person will be to fat gain.  As with the diseases of civilization, obesity is caused by a mismatch between our genetic heritage and our current environment.  This idea received experimental support from an interesting recent study (3).

Researchers have identified a number of genetic factors that influence obesity risk in industrial nations (an obesity-promoting environment).  These tend to be linked to genes most highly expressed in the brain, and particularly in the region known as the hypothalamus, which is central to the regulation of body fatness (4, 5).  In addition, they have found gene variants in the dopamine signaling pathway that associate not only with obesity, but also with susceptibility to drug and gambling addiction-- these genes influence reward sensitivity (6, 7, 8, 9, 10).  A plausible idea is that genetics influences 1) how well each person's hypothalamus resists the factors that promote obesity in the 21st century, and 2) how susceptible each person's reward centers are to hyperpalatable/hyper-rewarding food. 

This is why statements like "I eat junk food, and I'm not fat, therefore junk food doesn't cause obesity" aren't very convincing.  Due to individual variability, these anecdotes are not always useful for understanding population-level changes. Two different people can be exposed to the exact same food environment, but genetic differences will encourage leanness in one and obesity in the other.  This is part of the reason why I don't see obesity as a "moral failure". 


According to USDA data, Americans today eat an astonishing 425 more calories per day than they did in 1970** (11).  That is the reason for the obesity epidemic, plain and simple.  However, that fact doesn't tell us why we're eating more calories, so its usefulness is limited.

The increase in calorie intake has come primarily from refined carbohydrate, but even that doesn't get us very far, because why did we decide to eat more refined carbohydrate?  Probably because of the systematic efforts of commercial food manufacturers to increase the palatability/reward value and availability of processed food.  In the last four decades, the US has become saturated with hyperpalatable/rewarding commercial and restaurant foods including fast food, soda, french fries, chips, candy and other industrial products.  I've seen people claim that they ate these things just as much in the 1960s and 70s, but the USDA and National Restaurant Association data show otherwise (12).  The qualitative changes in the US diet have been swift and profound-- I'll be reviewing that evidence in future posts. 

There is a lot of research suggesting that food reward/palatability is an important factor in body fatness (13, 14, 15), but it isn't the only factor.  I regret calling food reward "a dominant factor" in my food reward series, because that phrase has been widely misinterpreted as indicating that I think food reward is the only factor in obesity.  Finding examples of people, or even whole cultures, that the theory doesn't explain, invalidates the hypothesis that food reward accounts for 100 percent of obesity.  Of course, that's not the hypothesis that I, or other obesity researchers, favor.  I'm sure there are exceptions in which food reward/palatability doesn't explain obesity, although I haven't found a whole population yet that is a clear exception, in the sense that major changes in palatability/reward do not correlate with changes in body fatness.  Food reward/palatability is the single most compelling explanation I've come across for the US obesity epidemic so far, but it does not exclude other factors. 


Another factor that contributes to obesity is inactivity.  The human body is designed to engage in physical activity, and if that doesn't happen, we don't function optimally.  Exercise maintains leptin sensitivity in the hypothalamus, and the rodent literature suggests that without it, leptin sensitivity declines and fat mass increases (16, 17).  This offers an explanation for why overweight people don't necessarily compensate for the calories they burn while exercising by eating more (18).  If energy expenditure is increased, but leptin sensitivity is increased, the balance between these two factors will determine hunger and food intake.  Calorie compensation is highly individual, and accounts for the variability in the ability of exercise to cause fat loss between people (19).  Treadmill-type exercise isn't a very effective way of reversing obesity for most people, however it can increase the effectiveness of dietary interventions, and it may work better for preventing obesity than for reversing it (20, 21). It's also good for general health, up to a point. 

Other Factors

There are probably other factors involved as well.  Animal experiments have taught us that inflammatory signaling in the hypothalamus can cause leptin resistance and increase body fatness (22, 23).  Anything that increases inflammatory tone in the brain is a potential contributor.  The potential causes are murky, but may involve altered gut flora, insufficient dietary fiber, insufficient polyphenols, poor omega-6:3 balance, psychological stress, insufficient sleep, micronutrient deficiency, and physical inactivity.  Fiber and exercise are the most solid, because they have been shown to protect against fat gain in rodents and humans (24, 25, 26, 27, 28).  The rest are just possibilities, according to my current knowledge-- nothing to hang your hat on. 

There's an additional factor that I've come to believe may be an "elephant in the room" when it comes to insulin/leptin resistance and chronic inflammation, and that is, ironically, energy excess.  Glucose and fatty acids, the body's main two fuels, are toxic when present in the bloodstream in excess.  When someone eats too many calories, his body has to deal with the excess.  The healthiest way of doing this is actually to shunt the excess energy into fat tissue where it is inert.  If the fat tissue does not have a sufficient affinity for the excess fat, free fatty acid levels in the circulation may rise, and tissues and cells may accumulate fat and fat metabolites (e.g., triglycerides, diacylglycerides, ceramides and fatty acyl CoAs).  This is what we see in people who have a disorder of fat tissue called lipodystrophy-- since their fat tissue can't absorb much fat, they accumulate fat where it isn't supposed to be and develop profound insulin resistance (29).  This is presumably because of the insulin resistance-promoting and inflammatory effects of some of these fat metabolites (30, 31).

Even in people whose fat tissue does have a high affinity for the excess circulating fat, which is most common at least initially, if fat mass increases enough, fat cells become insulin resistant, release more fatty acids into the circulation and fail to clear fatty acids from the circulation after a mixed meal (32, 33, 34).  Essentially, fat tissue loses its formerly high affinity for excess fat, and these undesirable fat metabolites accumulate in lean tissues in a manner reminiscent of lipodystrophy.  This contributes to insulin resistance and glucose intolerance by the same mechanism described above, creating an excess of circulating glucose as well, which together with the excess of fatty acids can enhance chronic inflammation, further insulin resistance and damage the insulin-secreting pancreas. 

Consistent with this idea, diet-induced obesity in rats is accompanied by the abnormal accumulation of lipids in the hypothalamus*** (35).  This has been shown to inhibit leptin and insulin signaling in the hypothalamus via a protein called PKC theta, which is similar to how fatty acid accumulation in the periphery is thought to inhibit insulin signaling (36, 37, 37B).  Suppressing PKC theta activation in the hypothalamus attenuates diet-induced obesity in rats (38).

Therefore, it's possible that an excess of circulating fatty acids (and perhaps glucose) itself acts to raise the setpoint through the gradual accumulation of fatty acid metabolites and inflammation in the hypothalamus, promoting leptin resistance and creating a "cascading failure" of energy balance regulation, glucose metabolism and inflammatory signaling.  This would explain why people in affluent societies have trouble staying lean as they age, as well as why obesity is so difficult to treat.  I think this is likely to be a late stage process, occurring after significant body fat accumulation and essentially "cementing" the increase in body fatness.  The early stage that causes the initial rise in body fatness probably has more to do with food reward/palatability/availability, although that should remain a factor even after obesity is well established. 

Putting it All Together

The basic idea is that in genetically susceptible people, excessive food reward/palatability/availability and inactivity cause overconsumption and an increase in the body fat setpoint, followed by the eventual accumulation of fat metabolites and inflammation in the hypothalamus, which exacerbate the problem and make it more difficult to treat.  Other factors, such as micronutrients, gut flora, fiber, fat quality, polyphenols, sleep and stress, may also play a role.  I think this is a reasonable working hypothesis of why obesity has increased so rapidly in the last 30 years, and is so difficult to treat once established.  I believe these ideas are broadly consistent with the research and opinions of senior obesity researchers I respect. 

* More precisely, these studies measure the heritable component of body mass index variation.

** Estimates I've seen range from 250 (NHANES) to 570 kcal/d (39), but everyone agrees there's been a large increase.

*** Palmitoyl and stearoyl CoA specifically.  These are lipid metabolites on their way to being burned by the mitochondria.  If the cellular influx of lipid exceeds what the mitochondria can burn, fatty acyl CoA and other lipid intermediates accumulate.


Laura said...

I would also speculate that there is some cumulative effect, since I observe a large number of people that maintain leanness as children or young adults but eventually succumb to overweight while eating essentially the same foods.

Tucker Goodrich said...

Interesting. So one would think that training your body to burn fat would be a preferred strategy under these circumstances. Occasional fasting or aerobic training, for instance.

How do you think the different "neolithic" foods play into this scenario, since avoiding certain foods seem to pretty reliably influence weight loss?

David Pier said...

Except for the omission of "epigenetics", I think this is a nicely balanced synopsis/abstract. In line with the quality that attracted me to your blog in the first place. I feel that several of your more recent posts have been posted a bit hastily, so that their wording was less than ideal. I'm glad to see the old Stephan back.

David Pier said...

Besides epigenetics, I would also add "fetal environment".

Christopher said...

It seems that the benefit of exercise is to ensure that fat mass remains close to set point (by increasing leptin sensitivity in hypothalamus).

So one could say that exercise is "necessary but not sufficient" for fat loss from the obese state (beyond the presumably noise-level fat mass of leptin resistance)?

Also it seems that exercise wouldn't prevent obesity.

Matt Lentzner said...

Stephan, This is all very interesting although my own experience seems to counter some of what you are saying.

About 3 years ago I was pudgy (not fat ~20% BF). I started lifting and put on about 15 lbs of lean mass, but my body composition did not change. After switching to a Paleo diet I dropped 15 lbs of fat. I ended up at the same weight I started at, but at about 13% BF.

Thing is I now eat much more than I ever did, 3000-4000 kcals per day and I enjoy it. My observation is that milk and sugar cause me to put on fat while meat and vegetables do not. There are a lot of people like me out there.

I think it's obvious that my genetic predisposition to obesity is low, but I still seem to be able to buck the trend. I think what's missing in all this is any data on how a person with well-developed muscle mass reacts to food. IMO the average American is not only overweight, but also underweight in lean mass - far below the biologically desirable level.

Anonymous said...

When I am lectured by anti-carbers about insulin and blood glucose, I figure my high exercise lifestyle is highly protective of metabolic issues, and I don't really worry about it. I'm not into low-fat or low-carb, and I'm very comfortable with my diet, few cravings, not much hunger. I don't believe that people get fat or diabetic from fruit or brown rice, which IMHO, is not the same as cotton candy or soda. Anyway, not completely related to this post, but I've been reading a few of them, so they all kinda meld together.

My weight stays stable (for a few weeks anyway), whether I exercise a lot or just a little, but I think it does wonders for my mood, and great things for my hormone and other signaling, most of which I don't understnad, but I don't have to.

@NaturalWorkwear said...

Stephen, I was wondering about the 425 excess calories per day…

If I understand this correctly, 425 kcal/day * 365 days/year ÷ 3600 kcal/pound(fat) = 43+ pounds/year? Perhaps this can be countered by increased caloric expenditure, but if sedentary behavior is increasing…

Stephan Guyenet said...

Hi Tuck,

I don't think it matters whether you burn fat preferentially or not, what matters is whether your total calorie intake is appropriate for your expenditure.

I think certain industrial foods play into that because they're highly palatable/rewarding and easy to overconsume.

Hi David,

Yes, I tried to use the word "heritability" to represent genetic and epigenetics, but it was too cumbersome in some places.

Hi Matt,

I don't think your experience is at odds with what I wrote, in fact I think it agrees nicely. If you lost fat, that means your energy expenditure exceeded your energy intake for a while. You accomplished that by exercising and reducing food reward (paleo), which presumably lowered your setpoint. You may be eating a lot but you're also expending more, if you're weight stable.

Hi Fraz,

The calculation isn't that simple unfortunately, because energy expenditure increases as body mass increases. Thus, energy "flux" is increased in obesity. This paper goes into detail on that:

Toivo Ellakvere said...

"I think this is a reasonable working hypothesis..."
You may be absolutely right (or wrong as a matter of fact) but that doesnt help people to loose weight.

Exercise more, eat less?

Matt Lentzner said...

Just to clarify something. If I eat more calories than used to, but maintain a lower fat setpoint this could still be caused by a lower food reward factor?

The thing that I don't feel that gets accounted for in the calorie model is the high variability of your body's baseline metabolism. There are all sorts of crazy ways your body can burn excess energy. As well as many ways to downregulate it to conserve.

Admittedly, maintenance calories for me is much higher than it used to be because my lean mass is much more metabolically active. In general, I eat more than I "want" to to create a caloric surplus to support further growths in lean mass. It doesn't seem to stick on me unless I stray from my diet.

The case of children is what I come back to. A lot of kids eat terrible diets and they remain very lean, at least to adulthood. When don't they get fat? My anecdotal observation is that they burn it off with increased activity, while adults tend to store it. What has changed?

Stipetic said...

The Massas study you posted a few weeks ago demonstrated that humans (the Massas, specifically, of course) appear to be capable of burning over 1000 kcal above their maintenance levels when overfed (they should have gained over 100 lbs given their caloric intake, but only gained 37--and this was with a decrease in activity, I believe). If it's only about calories, as your theory is predicated upon, shouldn't we be able do burn off a measly extra 400 or so calories?

I'm wondering when factoring in the human body's incredible ability to burn off excess calories, and factoring in the other stuff you said about our the toxic modern environment, wouldn't the following testing hypothesis be just, if not more, appropos: Essentially, our genetic heritage is not well matched to our current food and lifestyle environment, which encourages the overconsumption of neolithic agents of disease, and encourages inactivity, leading to fat gain in genetically susceptible people.

Pål Jåbekk said...

It is problematic that the only parameters normally measured in the metabolic syndrome are those related to nutrient partitioning and the classical risk factors for diabetes and heart disease. We need more info on the brain and hypothalamus in these people. Is there a way to measure hypothalamic inflammation and leptin action in vivo?

gunther gatherer said...
This comment has been removed by the author.
Rob said...

Stipetic - I think I can answer your question, obviously Stephan correct me if I'm wrong. Most people's compensatory mechanisms aren't geared to burn off huge amounts of calories via NEAT (non-exercise activity thermogenesis). This makes sense as storing body fat easily would make survival more likely among our ancestors when food was limited.

Thus with passive overeating that occurs in our modern environment and with a less active lifestyle, those seemingly measly 400 calories aren't burned off.

But there is a huge variance among people in regards to NEAT, some lucky folks expend hundreds of calories as you mentioned unconsciously and don't gain a significant amount of fat compared to others.

gunther gatherer said...

Hi Stephan, the part about overeating ties in nicely with what Eric Stice says about its effects on D2 sensitivity:

Just eating a huge meal alone, whether paleo, LF, LC or junkfood, will cause desensitization of dopamine receptors. Maybe you could include this in your list of overeating's physical effects. Our portions have become enormous in the last 20 years, there's no doubt about that.

As a weird anecdote, after watching this clip, I started eating smaller meals spaced throughout the day as opposed to the 2 huge ones I normally ate.

I have always had a small lump of fat on my scapula about the size of a golf ball. Not surprisingly, it appeared three years ago when I started intermittent fasting and feasting.

It is now shrinking at one week since this eating pattern change. This could be signs of a correction of the lipodystrophy you discuss above.

Stipetic said...
This comment has been removed by the author.
test said...
This comment has been removed by the author.
test said...

Dear Dr. Guyenet,
I see that food reward is a factor in obesity and most probably in general food choice and intake in everyone. (And I know reward is not the same as palatability.)

If we focus on reward, I think it is vital to consider its determinants. One could speculate that it is determined by the composition of the food - and that’s where to me it gets interesting. My guess is that in terms of macronutrients, both carbs and fat have a positive main effect on food reward (beyond mere caloric value). Further, I speculate that there is a strong positive interaction effect between fat and carbs (their effect is more than additive). At the same time, I think the main effect of carbs is stronger than that of fat (think of the average person’s drive to eat large amounts of candy vs. a stick of butter). All of this may be (partially) mediated by taste, but I am not sure about that. (I am not about sure protein, but it certainly also plays some kind of role.)

All of the above, if true, could potentially explain junk food “addiction”. Also, as an example, I think “emotional eaters” in stressful situations preferentially go for foods that have both fat and carbs, then mainly carbs and only then mainly fat (keeping calories constant, if I remember some research I once read correctly). They probably seek reward, so it fits.

This would allow for both low fat and low carb diets to “work” (amongst other things they reduce the reward value, with all the downstream effects that follow). It also could explain why some studies that find a difference between these diets seem to slightly “favor” the low carb (ad libitum) diets.

Of course, there are a million other variables involved (satiety, hormones, etc.) and this only is concerned with what goes “in” and how much of it. I just wanted to ask what you think about this and whether there is research to prove or disprove this.

Kindke said...

Right, so its all because people eat too many calories. Fat people are gluttonous pigs and need to exert some willpower,


Calories Calories Calories, thats all that matters.

Your body burns an EXACT amount of calories everyday and any change in energy intake is is EXACTLY compensated for by changes in fat mass.

omgwtfbbq GENIUS!

Alex said...

Remember the increase of 425 calories per day is an average, it probably ranges quite widely.

Gretchen said...

It would be helpful if you'd list your references at the end in addition to providing links. It's too time-consuming for people with slow Internet connections to follow the links, and often one can look at a citation and think, "Oh yes. That paper" without having to revisit it.

Anonymous said...

Based on my own experience with low-carb and low-glycemic dieting, I second everything "test" said, but I'd also add that fat salt is the key to fat palatability. Ever had a steak, pork chop, or a strip of bacon without any kind of salt or marinade? Ever had avocado or butter without salt? None of these foods are very palatable, and in fact if you've ever eaten the fat of a pork chop or steak, you're not going to feel like eating again for quite a while. I've been experimenting with Stephan's earlier recommendations on reducing food palatability. I didn't notice much change in appetite except for the difference that excluding salt from meat made. From a reward point of view, salt makes a very good culprit. It was not likely a big part of the prehistoric diets, and once people figured out how to harvest it, it became highly prized, and even fought for.

WK said...

Okay. So it's back to calorie counting, like my mother did back in the sixties, with her "Pocket Guide to Calories."

In other words, if you are fat, you are eating too much.

To lose weight, eat less. Which you can do by eating stuff that isn't rewarding (which is not the same as not palatable) although otherwise it doesn't matter what you eat.

Clear as mud to me.

Bill said...

Why does everyone always bag on Stephan when he says calories play a role? THEY DO.

It has been shown over and over again that modern humans eat more and move less then their ancestral counterparts and even those who lived just 100 years ago.

Whether you like it or not, our food is more calorie dense, we eat out more, we move less, we sleep less than we should, we stress more, we eat more pufas, more wheat, etc etc... but in the end, calories do matter. If they didn't everyone would be able to effortlesslty maintain sub 10% body fat for males and sub 18% for woman on 4000 cals a day sitting on their ass. Sorry folks, but it just doesn't work that way.

I think people get offended by the calorie theory, as if we are pointing a finger at them for eating too much. What Stephan puts forward with the food reward idea is that it's not anyones fault that they eat more... the food is designed so that they do and continue to crave it. It is more flavor and calorie dense and thus, when we overeat it, we put on fat. WHy is this so offensive and controversial to people? It makes so much sense to me. This is why Diet Coke actually stalls weight loss for some and can even cause weight gain.

Calories DO matter folks. There are degrees to which they matter and matter less, but they DO matter.

Unknown said...

Are you assuming that 100% of the excess 425 calories taken in, is absorbed into the body?

Steve said...

"A plausible idea is that genetics influences 1) how well each person's hypothalamus resists the factors that promote obesity in the 21st century,"

Does mindfulness meditation improve the hypothalamus's ability to resist obesity?

luckybastard said...

i appreciate this post, stephan. what i'm getting from this is that we have been consuming in excess of 400 calories/day in recent decades and this has led to our obesity epidemic. i also interpret that you are placing some- likely a large part- of the blame on the hyper-palatability of processed foods. i can see your theory complementing that of the one that the big 3(fructose, high o6 oils(yes, i still think these do much more harm than animal fats and if the right studies are conducted, it will bear that out), and gluten are responsible for a good number of diseases of civilization. i would even go as far as to say that the big 3 are the main contributors to that hyperpalatability and increased food reward- though i know you would throw in salt and all fats.

the calories in/calories out nature of this is just a matter of fact, like saying that if a carburetor runs rich(pardon me, i still love 40 year old vehicles)the car will burn more gas. the question isn't whether the carb is running rich, but why is the carb running so rich? i personally think that the big three muck up satiety mechanisms in the hypothalamus and other pathways(i'm not a scientist so i'm not well versed in these things)and leads to the mindless consumption of 400+ calories in excess a day.

what seems to have worked for me and many others is just to avoid the big bad 3 and eat real food- nothing processed. going low carb in the beginning also seems to have some kind of healing affect on certain pathways(not to mention ketosis helping people control hunger initially). just stating where i'm at from what i've read so far.

Susan said...

Actually he said that as a society, we all eat more than our ancestors. Even our recent ancestors. He also said that some people, because of genetic factors as well as stress and lack of sleep and excercise, react differently than others. Some become obese while others do not.
Try this experiment: go to a garage sale or second hand store and buy a dinner plate from the 1950s, or 1960s. Take that plate to your favorite dinnerware store, and see if you can find one even roughly the same size. You will, but it will be called a salad plate.

Susan said...

Sorry for 2 posts.
I also agree that salt seems to change the palatability factor. I googled salt consumption through time, however, and came up with an article in USA Today from 11-23-2010 that claims:
"The researchers thought they would find that salt intake had increased over time because Americans eat more processed foods today than in 1957. But decade after decade, people consistently consumed about 3,700 milligrams of sodium a day, the data showed. Current sodium guidelines advise up to 2,300 milligrams (about one teaspoon) a day for adults, and 1,500 milligrams for those who have or are at risk for high blood pressure.

The study appears in the November issue of the American Journal of Clinical Nutrition."

So hmmm. I would not really call what I did research, but it is interesting.

Anonymous said...

Well nice and detailed post.. I give this post 10/10 from The Surgical Blog

dim sum massacre said...

A couple of points:

1. In any sort of obesity hypothesis, calorie consumption should be (primarily) viewed as an effect, not a cause.

2. The obesity/diabetes epidemic didn’t start until at least 1970, and therefore shouldn’t be considered one of the ‘diseases of civilization’. It’s really in its own separate category…

luckybastard said...


he's basically saying what guys like mark sisson and robb wolf have been saying for a few years now but with his specific science background has proposed mechanisms that point part of it toward food reward which is fairly novel, if not unique. the core of what stephan's been saying is something that is not far afield from what most paleos try to practice. the first i ever heard about epigenetics was reading mark sisson. Kurt Harris recommended several meditation books that i picked up and started to read. listening to robb wolf turned me on to TS Wiley and the importance of sleep. i focused more on diet aspect- hyperpalatability and food reward- because that's the novel part of Stephan's approach that is "new" to me.

allison said...

I was born in 1962, so I am old enough to recall the prevalent dietary habits of the 1960s and 1970s. I can say without doubt that processed food has proliferated since about 1980. Growing up in a middle class home, we rarely are out. We had the occasional trip to McDonalds or Taco Bell. However, the store shelves were not stacked with processed foods the way they are today. That entire center portion of the grocery store where frozed and refrigerated processed foods are kept did not exist back then.

Eric said...

Just one question about the whole calories in-calories out theory. So if I eat a 500 calorie steak, how many calories are excreted as waste? How would one measure it?

Rob said...

Kindke - Of course your body doesn't burn exactly the same amount of calories everyday but we can roughly estimate our intake and expenditure. In the long run +-50 calories isn't gonna matter much. With that said if you really wanted to get a nearly exact figure on how much you're burning day to day, get a bodybugg.

Eric - Steak usually has a good amount of protein so it's TEF would be higher however you I think you should still regard it as 500 calories. No need to get super neurotic about it. And as I mentioned in my response above, measuring your intake is always an approximation.

And sorry Stephen, I know these questions and posts are directed at you.

Greg said...

I second the notion for a mention of epigenetics- some aspects of which have nothing to do with heritability, and others which could be inherited in the short period of time of the obesity epidemic.

What do you think of the theory that abnormal gut flora contributes to obesity (along with many other health problems)? The obesity epidemic seems to coincide with an increase in antibiotic use and reduced breast feeding.

While you have convinced me that food reward is a factor in susceptible individuals, I think there is too much evidence that healthy individuals can eat as much rewarding food as they want, so I don't understand how it could be the "trigger" of obesity. It seems more plausible that an environment containing some aspects of our modern society (increased stressors and toxins, and nutritional deficiencies) is a necessary pre-condition that degrades the body's ability to regulate itself to the point that food reward can cause obesity.

Paleo Phil said...

Yet another marvelous post, Stephan. If I understand you correctly, I think you are saying that Eaton and Konner's 1985 proposed theoretical framework of evolutionary nutrition (and medicine by implication), with biological discordance as the broad mechanism, is still the most likely candidate for overarching theory of everything under which obesity, counterproductive food reward signals, diseases of civilization, and other undesirable phenomena fall, yes?

Paula said...


The last sentence in your post was "I believe these ideas are broadly consistent with the research and opinions of senior obesity researchers I respect."

Would you mind naming these senior obesity researchers whom you respect?

Thanks in advance.

David Pier said...

I just looked for, and failed to find, any study of the heritability of obesity in donated embryos. That would help disentangle the contributions from fetal environment vs. genes and pre-gestational epigenetic changes. Plenty of research on adopted children showing the heritability of obesity, and the small effect of family environment.

Perhaps there is an equivalent study in rodents, at least?

Dave Tate said...


Very intriguing article. Most of us are familiar with the U.S. and the foods that are available to us when it comes to palatability. Is there any data to show similar calorie increase in other societies with the same type of foods being available and being overconsumed? I recall Taubes purporting that there was a link with increased sugar consumption with many of the countries that were plagued with the diseases of civilization. I am pretty sure John Yudkin made the same argument but Keyes obviously carried the day with the lipid hypothesis.

Jonathan Passey said...

Hi Stephan,

I'm sure you've read the interesting research about larger containers/vessels for food leading to unconscious overeating. And just from a logistics standpoint I think prepared/ready-to-eat foods may play a huge role in the obesity epidemic.

Recently my wife and I have been experimenting with eating only whole foods prepared at home (and no refined carbohydrates). I track my eating and my calories have dropped dramatically, maybe in part from food reward (although much of what I make is delicious and full of fat and salt) but I think the reduction is largely due to access. There isn't any food in my house that is easy to eat. Maybe vegetables raw. But often when I previously would have 'snacked' (non-hungry mindless eating) on something I now just ignore the desire.

I speculate that food-access is as much of a culprit in obesity as food-reward. Or perhaps the two together are the perfect storm. Only recently (40 years?) has there been such a huge number of ready-to-eat and instant foods. I think this has paved the way for emotional eating/binging and mindless eating.

What do you think?


Anonymous said...

If you chart the rise in obesity it DIRECTLY correlates with the industrial revolution and mechanized travel. It is very very simple. We move a lot less than we did before and our metabolic flexibility is impaired.

Go a step further and take a look at sugar consumption per person in pounds since the 1700s.

1700 AD 4
1800 AD 18
1900 AD 90
2000 AD 145

That's about 7,000 calories in 1700 and 260,000 calories in 2000. Were we REALLY substituting that amount of calories for something else? I highly doubt it.

Anonymous said...

Now we're talkin'. I'm glad I keep reading this blog. Cheers.

Alan said...

>> A lot of kids eat terrible diets and they remain very lean, at least to adulthood. When don't they get fat?

A lot of kids take EXCELLENT care of their baby teeth..... but they always all fall out.

What part of juveniles-are-not-the-same-as-adults don't you understand?

Sue said...

Alan, kids are different to adults - enlighten us with your thoughts (from Matts comment).

David Moss said...

Hi Stephan. I know you've talked about the benefits of fibre in terms of butyrate before, but could you point to something else that backs up your claims about fibre being one of the most solid extra factors (more so than gut flora, omega-6:3, psychological stress, sleep, micronutrient deficiency etc.). The studies you cite seem to be about oligofructose in pretty high doses, rather than fibre per se.

CarbSane said...

Nice summary Stephan.

I'm always a bit amused at the reaction to simply acknowledging that we eat more on average than we did 30-40 years ago. This is an undeniable fact folks.

As one of those who frequently points out that growing up in the 60's & 70's we had many of the same foods. In my opinion this does vindicate macronutrients, as cause. However as you've pointed out previously we're eating more of it. This is, frankly, mostly cultural. More single parent homes and homes w/o one stay-at-home parent means less home prep and more easy processed food.

An average of 400 cals sounds like a lot, but that's not so difficult to accumulate just eating at the "healthy" salad bar! I still cannot get over the calorie count of mashed potatoes at TGIFridays. The regular side order is over 500 calories and "loaded" is like 900! A SIDE order. I don't see how someone even makes such a concoction at home. We are also a nation obsessed with drinks -- be it sugary and fatty Starbucks or energy drinks, sodas, fruit juices, whatever. I think the latest numbers show that Americans are getting on average 30% cals from liquid sources - and we're not talking slimfast!!

It's been a long time since I read the NYT Big Fat Lie article, but Taubes said right there that we WERE eating more, and drinking more of our calories.

David Pier said...

Regarding heritability, here is an interesting discussion that effectively makes the point that attempting to attribute a particular fraction of the obesity problem to genetics only makes sense given a particular environment or range of environments. Obesogenic genotypes will manifest to different degrees under different environments:

David Pier said...

And so: maternal and paternal contributions, both genetic and epigenetic (and inherited gut flora, as well), to obesity will be different in different environments, such as different countries, different socioeconomic classes, and different eras.

Raju Ahamad said...

I think you are really right on it.
there are several thing that our we can not do. Keep it up.

Matt Schoeneberger said...

Stephan, your second paragraph sticks out to me. Just because people know what foods they should be eating doesn't mean their equipped to make the correct choices when it counts.

Couple that with the fact that pretty much everyone underestimates their caloric intake (Hill), and you've got a lot of people headed in the wrong direction.


greenbee said...


What role do you think stomach acid plays in obesity? Since strong stomach is critical to digestion/absorption of nutrients- especially b-vitamins, zinc, etc? And most foods, drugs, etc. today reduce acid and therefore reduce nutrition-- and proper digestion?

Katie said...

Thank you very much for this condensed review of contributing factors.

Indeed, just by sharing kitchen space with other people one may notice that peculiar lustful twinkle in the eye of someone who on further reflection will show various signs of higher sensitivity to hyperpalatable foods. A closer observation on that same person will reveal several relevant psychological/mental factors including IQ, narrow scope of interests, low self-awareness and self-esteem, a past of abuse or excessive parental control, and a disinterest in recreational activities of low intensity (e.g., choosing to watch a visually and acoustically intense blockbuster at a big movie theatre rather than a more intellectually demanding movie at home or in a smaller theatre).

This profile I've sketched above is not intended to denunciate or blame, of course, but rather to attempt addressing the psychological side of obesity with regard to possible treatment alongside the consideration of biochemical aspects. I strongly believe that help could come from CBT specialists and vigorous work with the way obesity-prone personalities deal with themselves when alone and not busy with work, i.e. at leisure. True, in contemporary culture many people are overworked. When too tired to engage in distracting and fun activities, the fun box is ticked with the uncontrolled enjoyment of food. In addition, it is probably cheaper than even a game of bowling.

I am very interested to know if there is any relevant psychological research into the exact moments when the demographics described above find themselves at leisure and are planning next activity. I suspect intervening/helping that process could be of help.

Dragan said...

Hey Stephan,

I've been following your "food reward" notion for a while and I either I'm getting better at understanding it or you're getting better at explaining it (I'd wager it's both). It's difficult for most academics to explain their ideas in a non-technical but rigorous manner so I appreciate the effort. Looking forward to more.


Anonymous said...

What happened to the stuff about fatty liver, sugar, and choline? I was expecting it to play some role here, but it's completely absent from this "roadmap".

Gordon Rouse said...

Is this consistent with the findings of Dr Ethan Simms. He took non-obese prison inmates and asked them to over consume to put on weight.

All of them returned to their original weight after the experiment.

Were none of them susceptible to obesity, that they didn't maintain it?

I have a feeling that you are just looking at factors here that stretch the weight a little from its setpoint - but are not really permanent changers.
Dr Simms experiment was trying to test if set-points could be altered by overeating, and he failed to get this result?

I am surprised you didn't look more at the liver.

Dr J Friedman's experiment shows that maternal diet of junk food can create offspring far more susceptible to obesity. (in primates)

His findings seem to relate to an effect in the liver, and, in personal communication to me, he said that the effect correlated with high Omega6/Omega3 ratios. His anti SFA stance is off-putting, but his results are still intriguing.

Surely this is a gift-wrapped finding against your favourite nemesis?

I know you are no longer trying to pin obesity on LA, but surely you could re-open the case here?

The point is - if you are looking for evidence that a nutrient causes weight gain, ignore experiments with adults and look at the womb environment.

Abigail said...

Obesity is a serious yet common health disorder that most of them face. Only a proper diet practice followed by regular exercise can help fight obesity. Nice sharing. Stay in touch and keep posting.

online medical doctor

James Kimbell said...


General nutrition and food reward theory question:

What would you recommend for someone trying to gain weight? Not unusual, unsustainable bodybuilder amounts of weight, not 100% muscle and no fat, just the kind of weight gain that will bring an underweight person to a better body size and composition. I'm thinking especially of food reward here, and how it could be used to boost intake in a positive way, if possible. I know there probably aren't specific recommendations you can make with certainty, but some guiding principles would be great.

psychic24 said...

One of the most rewarding aspects of eating at, say mcdonalds, is the fact that I can order 2000kcals of food for a measly $5 or $6 and be satiated all day from it; compared to eating healthy animal products and starches, of which I can barely fit in 1000kcals per meal, even if I try super hard-- just too satiating. Now I know I may sound stupid but I was more convinced with the homeostatic regulation of caps in and out, rather than reward or even energy density( which is the 3rd or 4th tier of your reducing food reward plan). I know when I eat healthier and am unable to eat enough cals I become fatigued, experience orthostatic hypotension..etc. Anyways.... My question is why do these two systems seem to be at odds; how can you be full and unmotivated to eat, yet at the same time feel like crap because you haven't eaten enough cals? Any chance that it suggests low energy dense foods should not be the main constituent of a healthy diet?

Josh said...


I have have been using some strategies loosely based on my interpretation of one aspect of food reward to increase healthy caloric intake. Being slightly on the lean side (BMI-20.5) I find I feel better when I eat more and if I don't eat enough I am generally grumpy, low body temp and averse to physical activity. Judging from studies, people seem to voluntarily reduce caloric intake when eating a larger percentage of calories as protein or deliberately skewing the fat/carb ratio in either direction. I have found I don't eat enough if I have too much protein or have an imbalance of starch, fat and fruit sugar and this is usually due to eating what I think I "should" be eating and ignoring healthy cravings. If I get the balance about right and eat to appetite I feel great. Whenever I have consciously tried to increase or decrease any food group based on some nutritional paranoia from something I have recently read e.g. fructose is the devil, i need more chicken breast for B6, coconut milk has healthy MCT's etc I end up eating an unnaturally skewed macronutrient ratio which leads to reduced calorie intake. Choosing a wide range of 'healthy' starches, fruits, fats and protein and paying attention to appetite and healthy cravings tends to lead to optimal nutrient intake. That's my take on it anyway.

dim sum massacre said...

After reading this post a second time, I realized the most interesting part was the section dealing with free fatty acids and build up of fat metabolites. As far as I can recall this is the first time Stephan has written about this… Carbsane has written a lot about this, although more in the context of diabetes than obesity. Very interesting stuff.

It seems to me like another important factor to consider when building a ‘roadmap to obesity’ is mitochondrial function (see section 3 here:

James Kimbell said...


Thanks for the response. Your description of the problem fits me - the caloric intake targeted by my appetite is consistently lower than the caloric intake that gives me the most vitality. I'll take your advice about not skewing the general ratios.

psychic24 said...

Quick quote i wanted to share. As with blood lipids, they mostly reflect the nature of the diet (ie. high in carbs- higher trigs, higher in fat-higher hdl and ldl) until you take a more detailed look at the actual lipids (ie.vertical auto profile to detirmine size of lipoproteins..etc); the same goes, in my opinion, about the increased caloric intake in the US. "In most instances, energy intake can be interpreted as a crude measure of physical activity"~Walter Willet and Meir Stamfer, Harvard 1998 Nutritional Epidemiology

What if the reason our caloric intake has increased (at least in large part) due to increased recommendations to exercise more and more, coupled with the mechanism to overeat after exercise in most people (i.e. i deserve this chocolate cake for my efforts today), seems like it could explain a lot.

p.s. quick question for anyone willing to share some wisdom: If by eating junkfood i was actually more satiated than by healthy foods (ie eating junk i could eat 2k cals and be satisfied, while it would take 2500 cals for me to be satisfied eating healthy foods) and there would be evident weight gain from eating more whole foods compared to less junk, which would be better, healthwise? I ask because i know eating over calorie needs causes inflammation, but then again, so do the toxins in yea, any feedback is appreciated

skarab said...


I just wanted to second Josh's advice. For quite a while I thought paleo 'didn't quite work for me'. I was trying to control my macronutrient ratios and force myself to eat specific foods, but I just ended up undereating. I felt like crap and was very averse to physical activity. Since giving up all that tinkering and just following my appetite all my problems have gone away and I have great energy.

mem said...

Interesting information from the Lancet re: obesity pandemic -

Sonya said...

Hi Stephen, it's been awhile.

If you can accept that there's a mismatch between environment and genetics, I don't understand why you can't accept that there's a mismatch between current forms of carbohydrates and some people's genetics. Especially given the racial breakdown of obesity and type 2 diabetes, not to mention the incredible race bias in the BMI categories.

Furthermore, whereas I can understand not being compelled by the insulin theory, are you not at all impressed by the fact that adipose tissue is a complex endocrine organ, and energy storage may not be its primary purpose at all? If your thyroid doubles in size, is absolutely now contains more calories, calories by definition, you did not burn. But so what? That's not WHY it doubled in size. Given the incredibly complexity of adipose tissue with regards to leptin, adiponectin, TNF-alpha, etc, I don't understand why regulation of these factors aren't part the analysis for WHY this organ enlarges (which would require alterations in appetite, to support the enlargement).

Lastly, I'm just flatly not buying the food reward thing. I find meat hugely more "rewarding" than starchy, sugary foods. The "reward" is never reached with starches. There is no "off signal" with these foods. That's not a great enjoyment. That's scratching an itch and making it worse.

And there's really no difference between saying, "You're eating because you're an immoral glutton" to "You're eating because you're wire up to be just so in love with the taste."

Pagg Supplements said...

Thats the interesting post, i had ever read, burning fat is necessary to prevent ourselves from the obesity until it will leads to deadly diseases
Pagg Stack Review

bentleyj74 said...

@ Adipocyte

It's ironic that you don't buy the food reward theory because this

"The "reward" is never reached with starches. There is no "off signal" with these foods. That's not a great enjoyment. That's scratching an itch and making it worse."

is more or less my functional description for distinguishing between foods that have high food reward and foods that do not. If it's NOT satiating and IS like an itch you can't

Stephan Guyenet said...

Hi adipocyte,

I do think there's a mismatch between certain forms of carbohydrate and some peoples' genetic makeup, I just don't think it happens through insulin. Also, it's about more than just carbohydrate.

Fat tissue is indeed a complex endocrine organ, but that does not change the fact that its size is regulated by the brain.

No offense, but you don't seem to understand the concept of food reward. Feeling satisfied after eating something is not reward. If you keep eating a starch food beyond what's appropriate, that is probably because it has too much reward/hedonic value for you. Opioid signaling, implicated in hedonic processing, shuts off satiation signals in the brain and may also increase the setpoint. Dopamine signaling, implicated in reward, can strongly influence food intake and also seems to be able to increase the setpoint.

bentleyj74 said...

@ Stephan,

I don't think I have completely grasped the concept of food reward either. Still some lingering confusion re cause and effect.

Also wondering about adaptability.

I know plenty of people who can overeat on meat while I find it mildly repulsive in general. It has to be perfectly seasoned and completely devoid of gristle for me to truly enjoy it. Likewise fats. I'd possibly rather have my tongue nailed to the floor than endure the taste and mouthfeel of unseasoned fats.

However I can eat boiled plain potatos more enthusiastically then I expected and legumes as well.

I have no problem climbing aboard the good ship low food reward because it gives me carte blanche to eat the things I already like most. Is that truly low reward for me still? I'd likely feel just as reluctant as some if I believed that my future included a flat grey hell of unseasoned ground beef stretching far into the horizon.

allison said...

I'm curious how fiber can prevent obesity. How exactly does a nonutritive, indigestible substance do that? Or is it simply a matter of fiber displacing other more deleterious foods from the diet? Does fiber cause some sort of satiety signal?

psychic24 said...

What exactly do you replace your new non-rewarding eating patterns with...? I mean if we don't get reward from food we must be getting it from somewhere, right? Long day at the office, bills piling up, house is leaking due to the hurricane, i don't think a run around the block will relieve any of the tension....a big mac on the other hand...

bentleyj74 said...


I think I remember a few "reward" suggestions from the healthy skeptic interview which should still be available.

Mohammed said...

I have been following your blog for quite some time now and I am happy to see a grand conclusion of your research culminating in the post: Roadmap to Obesity.

I appreciate your critical review of evidence ranging from low carb to high carb societies and respective insulin resistance indices. I am insulin resistant myself and I have tried low carb. After about ten days (30g carb/day), I was unable to continue low carb due to my preference for carbs. In those ten days, I did not keep track of the amount of calories I consumed but I am pretty sure that they were greater than 3000Kcal/day. During that time, I lost weight and felt quite hot (sweating). I think my thermogenesis increased. My question to you is that how is it that by going low carb & high fat, I did not gain fat even though I was consuming a lot of calories. I suspect that for controlled trials for isocaloric diets, low carb will come out superior in subjects which have already been insulin resistant before the study.

On a different note, the african study of fattening up and losing weight over a period of three years; I am slightly critical of that study. I assume that only thin subjects were chosen to fatten up. The environment was non industrial. I think that if a person is thin in a hyperpalatable food environment e.g. an industrial society, force/over feeding him to make him fat will work and then when he will slowly lose weight (exponential decay). In other words, it does not matter what the environment is, a thin person will return back to being thin in the same environment after a brief period of over feeding.

I say all of the above because my mother is insulin resistant and she only prepares whole food. Rarely anything is from outside the house.

The correct study would be to take somebody who is already fat and insulin resistant from an industrial society and put them in a non industrial society and see if they lose weight. That will be the final solution to reverse obesity by reducing hyperpalatibility.

As far as raw foods are concerned, I find them quite hyperpalatable too. Wholemeal bread is nice and it is easy to overdo calories. So is brown rice, grapes, dates, organic whole milk, smoked mackerel etc. All are calorie dense. It is very easy to overdo raw food as it is very tasty. When one changes from processed food to raw food, I suspect that initially there is a drop in calories but soon raw food becomes very tasty and over feeding resumes.

I have changed my diet a fair bit and now it is all home prepared, with little processing, raw food and I find it hard to cut calories.

Steve said...


As I understand it, fiber can improve gut health, which in turn seems to improve leptin signalling in the brain.

psychic24 said...

Does Heritability have something to do with obesity? I'm sure it does, i see tons of people who just can't seem to lose weight meanwhile they're eating healthy foods...Well maybe that's a bit of a stretch, but i'm sure I see at least 1 or 2 people like that everyday...maybe 4 if it's a busy day ;3

Does rewarding food have something to do with it? Sure it does, we're not going after those plain potatoes or eggs stored in the vending machines. I definitely think beyond all the leptin signaling, all the talk of gut permeability..etc, I think it's as simple as saying we live in a more stressful time, or we live in a time where we have access to much more rewarding "things"--leather couches, air conditioning, and hersey's chocolate bars. I'm almost positive that 9 out of 10 people would do great following the USDA's nutritional guidelines. How do i know? Well I don't for certain, but I can logically deduce.... Seeing plenty of thin people walking around these days, I'm sure most of them follow these USDA guidelines, just like they exercise and do everything else that is politically disseminated--I'm operating under the premise that at least 90% of people know the crux of the president's state of the union address a day or so after it is given, that vegetarians live longer than meat eaters (that many paleo eaters cite study-wise) and all the "healthy" concomitant factors that make this true and give us enough peace to sleep at night after eating an animal rich diet. Then, of course, are the deviations at both ends of the curve, the genetic freaks in both directions, and those of us who include varying degrees of scrumdiddlyumptious foods in our diets. In all honesty, the USDA recommends a pretty bland diet, and even though i don't have distinct proof that eating fruits, vegetables, grains, protein, and dairy in non-"oversized portions" would promote leanness, but in the words of Justice Potter Stewart, "I shall not today attempt further to define the kinds of material but I know it when I see it". I just think by invoking leptin signaling and our metabolically complex adipocytes, we're just making excuses for why we aren't losing weight (instead of just driving past the McDonalds) and why we shouldn't burden ourselves with exercise.

Which brings me to a muse on exercise. Seems to me after indulging in overly rewarding foods and driving our caloric intake into excess of what we need, exercise is a proactive attempt at correcting the imbalance. But then again, exercise is a stressful task, just like any stress, good or bad, and will likely be coped with in a manor similar to however usually cope with stress--hundred years ago we'd do it with the more socially appropriate outlet of friends and family, nowadays we go to food and tv...with that same family, maybe...

psychic24 said...

I don't mean to be a downer, or to deride any of this scientific talk, I just think that it's much simpler to explain this "epidemic", or just our individual journey to this obese state of no returns. I think going out for drinks and restaurant meals is fun, not because it's providing a much needed source of calories, but because it feels good. I don't think we avoid exercise because we don't have the energy, via insulin or leptin or whatever, but because we get more enjoyment from sitting back on our couches, watching tv or playing games.

And I really think that just as with the seeds oils revisitation, there are many bright lights that can be shined into what previously seemed an opaque hole of ill health (i.e. wheat/grains, fructose, and seed oils); something meant to avoid at all costs, without really glancing at what was inside (unfortunately a bunch of in vitro studies with amounts of toxin that would be hard to come by in the context of real food). A quote that seems to summarize this, roughly, is one by Matt Lalonde, "[what athletes would] be eating on a quote unquote paleo diet, you’ve got things like yams and cassava. and if you look at the antinutrient content, it’s the same order of magnitude… so if you’re going to tell someone, ‘hey, you should not eat grains and legumes because they contain anti-nutrients’ a biologist – a plant biologist – is just going to look at you and say, ‘wow, this guy’s a moron". Wording around a new slogan to eat whole feeds and sit on your ass less just seems to be a tenable roadmap to solving the obesity crisis.

Steve said...


I don't think what you're saying is unreasonable--that certain behaviors may have gotten us into this mess.

But the main problem seems to be what to do about it now. Once obese, the body seems to defend the new setpoint. So doing the opposite--eating less/better and exercising more--doesn't seem to reverse the weight gain as easily. Forcing the issue through caloric deficit causes further metabolic damage and it's unclear if you're any healthier for doing so, nor is it clear if it's maintainable.

Brad Marshall said...

Stephan, previously you wrote about the effects of fructose consumption and wheat consumption:

Now, in the second post, you say "As usual, my microscope is pointed directly at industrial food", but the google archives show that you've recently edited that phrase. It used to say, "As usual, my microscope is pointed directly at wheat."

You keep arguing that food reward is not the ONLY factor in obesity, yet you are rapidly backpedalling from alternative explanations such as:

-fructose causing direct metabolic damage to the liver/other organs
-lectins in wheat interfering with leptin signalling
-omega 6 imbalance causing inflammation of the hypothalamus

All of these seem like plausible explanations backed by some science and are ideas that you've supported for a long time. I'm not sure why you're suddenly running from them so quickly.

For instance, in the case of the glucose vs. fructose based soft drinks, the fructose based drinks led to increased visceral fat and other markers of damaged metabolism compared to glucose based drinks. Do fructose based drinks have higher food reward? That seems doubtful, yet they directly led to increased insulin resistance, which you seem to be claiming is merely an effect of leptin resistance, which is caused by food reward. When I look at the study, though, the fructose seems to have directly done the damage independent of food reward.

Stephan Guyenet said...

Hi Brad,

It's a bit of an exaggeration to say I'm "rapidly backpedaling" or "running from" my previous ideas.

I used to believe that sugar, white flour and seed oils promote obesity, and I still believe that. The question is, why? Is it due to an inherent harmful property of those foods, or because they're used to make foods with a high reward/palatability value, or both? I had my tentative hypotheses back then, but I've moved on from them because I've found something better.

I never came to any strong conclusions about fructose consumption being inherently obesity-promoting in the normal range of intakes. There are several reasons for this. 1) in rodents and some other animals, sugar is not inherently obesogenic under controlled conditions. 2) the fructose drinking study you cited found an increase in abdominal adiposity, but not total adiposity, so it's hard to argue that it promoted obesity per se. 3) the doses used in that study and others were very, very high. Lower doses don't show that kind of effect.

There may be some inherently toxic properties of fructose if it's consumed in excess, particularly if it's refined, but that doesn't make it inherently fattening. I think refined seed oils are probably inherently unhealthy in excess, although I no longer have any reason to believe they are inherently fattening. Wheat is definitely inherently unhealthy for some people, but what percentage of the population that applies to is unclear, and its inherent ability to fatten is very questionable. On the other hand, many commercially processed wheat foods are highly fattening because they're among the most rewarding/palatable foods (pastries, donuts, etc).

Basically, I've encountered a hypothesis that is a much more convincing way to explain the same observations I described earlier on this blog. I feel that my overall understanding of obesity has expanded greatly since I wrote those posts, so I am de-emphasizing some of my earlier, tentative hypotheses.

Brad Marshall said...

Thanks for the response, Stephan.

Another comment I have about this post is you state that caloric consumption has increased by between 400 and 600 calories. But the USDA data you site is "food disappearance", not what people actually ate. So another reason for the big increase is that we've become more wasteful.

The NHANES data, an indication of actual intake, I believe, shows that male caloric intake went from 2450 daily to 2618 between 1971 and 2000, an increase of "only" 168 calories. Still an increase, but not nearly as big a jump as you suggest. Interestingly the increase was greater in women - over 300 calories.


Stephan Guyenet said...

Hi Brad,

You were looking only at the 20-39 year old age group. NHANES found the overall increase over that time period for adult men plus women was 250 kcal/d. That is on the lower end of estimates, which range up to 570 kcal/d (also based on direct measurement rather than food disappearance).

The USDA food disappearance is adjusted for food losses between manufacture and consumption. See this page:

Brad Marshall said...

Nope, I was right. From the link:

"For men, average energy intake increased from 2,450 kcals to 2,618 kcals (p<0.01), and for women, from 1,542 kcals to 1,877 kcals (p<0.01). For men, the percentage of kcals from carbohydrate increased between 1971--1974 and 1999--2000, from 42.4% to 49.0% (p<0.01), and for women, from 45.4% to 51.6% (p<0.01) (Table). The percentage of kcals from total fat decreased from 36.9% to 32.8% (p<0.01) for men and from 36.1% to 32.8% (p<0.01) for women."

Which actually means fat calories consumed by men decreased by about 50 calories or so between 1971 and 2000 (according to NHANES). Increasing carb consumption made up for this drop and then some.

Brad Marshall said...

Even more interestingly, in men ages 20-39, an age period when many men get fat, caloric consumption went from 2784 daily to 2828, an increase of only 44 calories. It was older men who showed a big increase in caloric consumption.

Brad Marshall said...

OK, not to beat a dead horse, but if you believe the NHANES numbers, men ages 20-39 between 1971 and 2000 actually decreased their fat calories from 1030 per day (37% of 2784) to 840 (32% of 2828) while increasing carb consumption from 1174 to 1309, basically just replacing fat with carbs.

Also, if we're talking about trends, caloric consumption was unchanged from 1971 to 1980 (in men and women). Caloric consumption only jumped during the 1988-94 period when carbohydrate consumption as a percentage of calories increased from about 42% of calories to 48% of calories (in men).

So I think we can all agree that the obesity epidemic took off while everyone was doing low fat in the early nineties. Caloric consumption increased as dietary fat percentage decreased and carb percentage increased. The question is whether it's because snackwell's (the early nineties low fat snack of choice, remember) had high food reward or because a lower fat diet (by caloric percentage) was less satiating or some other mechanism.

sharon baker said...

Nice to find it. I like your blog site very much. You have done very appreciative work. Good job! Keep posting...

kamagra jelly

Sam Knox said...

"According to USDA data, Americans today eat an astonishing 425 more calories per day than they did in 1970** (11).  That is the reason for the obesity epidemic, plain and simple."

An excess of 425 kcal/day is 155,125 calories in a year, or roughly 38 pounds if stored as fat.

That seems...high.

Mickey said...

Very nice post, and it covers a lot of different aspects of the obesity problem. I am also interested in the effect of epigenetics, and I do wonder why hardly anyone wants to talk about female hormonal changes spurring weight gain, especially during pregnancy and menopause. The formerly slender young woman who gains weight with every pregnancy is almost a cliche.In my own family, going back 5 generations that I know of, all the women got fat when they hit their 40's. No HFCS, no junk food, no McDonald's, no automobiles, no TVs in those days, and they still got fat.

Reijo said...

I think this was a quite balanced view. Thanks! Someone mentioned increased access to food. I think this might be very important on top of the continously increasing variety of foods. We simply have to many stimulus and sensory food signals around us.

Something that was missing and might contribute: decreased share of complex carbs. Complex carbs include a fair share of low digestable carbohydrates (LDC). LDCs contain energy only 2 kcals/g and are fermentable in colon. Fermentation of LDCs produce short chain fatty acids that improve satiety. Consumption of LCDs has decreased since 1970s while usage of refined carbs has increased at the same time.

Stephan Guyenet said...

Hi Brad,

Yes, it looks like I misread the chart. Still, the overall NHANES finding was an increase of about 250 kcal/d, as I said. See this post:

Overall, total fat intake in grams has remained quite stable, possibly declining slightly. Vegetable fats, mostly soybean oil, have partially replaced animal fats.

Hi Sam,

That's not how it works. As body mass increases, energy expenditure increases, so you will not keep gaining fat mass indefinitely. A person will reach a plateau when energy intake matches expenditure. You can't just make calculations based on the number of calories in a pound of fat mass.

Craig said...

How does the type II diabetes epidemic fit into your food reward theory? Your theory would seem to require obesity and a broken metabolism as a precursor to type II diabetes. And yet, I am frequently reminded by people with expertise in this area that not all type II diabetics are fat; there are plenty of thin people who get diagnosed with the disease. Are these all LADA types? Or it is possible to break you metabolism even without getting fat? If so, why does a modern diet break some people's metabolisms even in the absence of obesity, and why doesn't that happen in primative peoples following a more traditional diet and life style. Or is the absence of type II diabetes in such populations over exaggerated?

Apolloswabbie said...

A proposition on what the purpose of a metabolic system is:
-fueling the brain is the highest priority of the metabolic system, the most critical element being oxygen, second is glucose/ketones IOW a "brain specific fuel."
-possibly the next most critical function is hydration
-not as time critical, but critical, is prevention of an excessively high level of blood glucose to prevent damage to the nervous system
-after that, the priorities are to distribute fuel to keep the body moving, and then to ensure long term viability via distribution of essential nutrients to support growth, healing, and the mechanisms of long term health

In other words, for a model to explain a condition like obesity, or how to reverse it, I would expect it to reflect the primary drivers of metabolic purpose. I understand how the carb/insulin/fat model aligns with the priorities as I understand them - that model explains the role of insulin in defense of excess blood sugar, it explains why we get the ketogenic response in the absence of carb intake (brain fuel source redundancy, in combination with the fact that most cells exhibit metabolic flexibility to either help control glucose levels from exceeding healthy levels, but easily burn fat for most purposes when glucose is not abundant), it explains why we observe very high blood concentrations of free fatty acids in the absence of fat intake with high carbohydrate intake (liver production of palmitic acid from glucose), and in this model we observe that in the obese, many indicators of metabolic derangement can be reversed via carb restriction alone in just a few days (and less rapidly so with non-carb restriction of calories).

Apolloswabbie said...

Part 2:
IOW, the carb/insulin/fat model is compelling for me because it fits within a coherent model which satisfies the important "WHY" question: why do we have a metabolic process.

I am working through an understanding of the model you advocate - is there a post in which you've highlighted how that model also meets the metabolic purpose? IOW - why is it good that the body as a "set point"? What purpose does it serve, what survival/adaptability function is enhanced, via "set point/leptin signaling?" I can see how leptin signaling serves us in the carb/insulin/fat model, but the light is not yet on for your advocated model.

As I read the comments, I'm surprised by how many of the commenters on "both sides" (and Taubes, but he likely didn't read it) can't see the significance of this part of your post: "According to USDA data, Americans today eat an astonishing 425 more calories per day than they did in 1970** (11). That is the reason for the obesity epidemic, plain and simple. However, that fact doesn't tell us why we're eating more calories, so its usefulness is limited." (my italics/underline/bold)

I'm also curious, in your experiment w Kresser, how you will define palatability in a way that differentiates that quality from carbohydrate in general, sugar specifically, in real food.

For me, as I have become habituated to carb restriction, regular, non-sweetened/salted food tastes much better. My "addictive" behaviors as regards sweets have been regulated. My pleasure in eating sweets now is decreased. I cannot remember the last time I "ate big", and I don't feel the desire to eat big. My body fat has modulated between 10% and 20% over the last 25 years (everyone else in my family spend at least half their lives overweight; I don't have a genetic super-metabolism). Prior to age 20, I couldn't eat enough to gain weight, ever since, it has required significant experimentation to find a strategy to maintain a healthy weight, despite hours of weekly exercise. I can explain all of that - to my satisfaction - via the carbs/insulin/fat model. I'm still a long way from that level of understanding in your model.

Lastly, thank the heavens the debate has moved beyond "saturated fat bad, plant based food good, exercise more and control your portion size."

Apolloswabbie said...

Have you made a post about when the change in model from the carb/insulin/fat model to the Guyanet/pallatability model results in a change in clinical implementation? It appears that carb control (~100g/day) is practically the same as palatability control.

Apolloswabbie said...

Dr. William "Wheat Belly" Davis makes the point that the wheat we eat today, like many foods, isn't even like the wheat we ate 100 years ago, and on his own body, eating the new wheat has a far stronger impact on blood glucose. Is there a way to use this phenomenon to understand your view of the impact on obesity?

Apolloswabbie said...
This comment has been removed by the author.
Coach said...


Not to be party-pooper, but categorizing macronutrients is the work of man - as is the prioritizing of body functions.

Nature doesn't exactly work as our abstract thinking would like it to, and this may just be the largest obstacle in our understanding of her. So asking questions such as these might be, for lack of a better term, irrelevant.


TedHutchinson said...

Re Nature doesn't exactly work as our abstract thinking would like it to, and this may just be the largest obstacle in our understanding of her. So asking questions such as these might be, for lack of a better term, irrelevant.
Those who agree with the above may enjoy these two full text articles.
Technology, Diet, and the Burden of Chronic Disease
Commentary. The big issue is ultra-processing‘Nutrient profiling’, and ‘fortified’ soft drinks for Africans

shannon said...

Stephan, I love your blog! I just found you about 2 days ago and I can't get enough. I also have a website, not for profit like yours that is just for sharing information about fat metabolism, healthy eating and recipe sharing. May I please share some of the information you have posted here and give you credit for it all? I will be happy to refer people back to your site whenever I quote you. My email is thank you so much.

Unknown said...

You said, "Exercise maintains leptin sensitivity in the hypothalamus, and the rodent literature suggests that without it, leptin sensitivity declines and fat mass increases (16, 17).  This offers an explanation for why overweight people don't necessarily compensate for the calories they burn while exercising by eating more (18).  If energy expenditure is increased, but leptin sensitivity is increased, the balance between these two factors will determine hunger and food intake.".
Then we have this:

Using animal studies to propose novel hypotheses can be fun. However, we cannot extrapolate findings directly. This is important when discussing metabolic pathways and the effect of different diets. For instance, mice have a basal metabolic rate that is 7 times greater than humans, so a 40% calorie restriction in mice mimics a therapeutic fasting in humans. There is evidence that mRNA levels of several lipogenic enzymes are different between rats and humans.

What you say happens in rodents does not happen in humans. The hypocretin neurons are evolutionary quite young in mammals phylogentically speaking. We only have 50k of them and they can replenish as far as we know today. They also have been shown to control the depolarization of the dopaminergic reward tracts. How does this science fit into your reward series. Right now it appears a completely unfinished theory based upon the hypothalmic wiring and neurochemistry we do know today in 2011.

Unknown said...

Its preservative ingredients triggers heart illness

It sets the scene for obesity as it includes trans-fat and saturated fat; the fact that it is fried may cause both hearth illnesses and cancer; consuming too much salt and sugar might cause hypertension and diabetes. As it does not include fiber and sediment may cause digestion illnesses and various cancer types. Due to preservative materials, allergic reaction can be observed in some people.

Keep them away from children

How your child is being fed is important for his brain development. Studies show that IQ of children consuming junk food very often is low and their learning capacities are affected. Besides, this type of food makes the child more aggressive, causes concentration problem, hyperactivity and sleeping disorders.

Say “no” to Junk Food