Download Lectures on Biostatistics (1971).
Corrected and searchable version of Google books edition

Download review of Lectures on Biostatistics (THES, 1973).

Latest Tweets

There is no topic more widely discussed than what one should eat in order to stay healthy. And there are few topics where there evidence is so lacking in quality. This post isn’t about quackery, but about something much more important. it is about the real science (if it merits that description) behind dietary advice.  I’m not an expert in nutrition, but I do know a bit about the nature of evidence. I’m continually astonished by the weakness of the evidence for some things that have become received truths, and nowhere is that more true than in nutrition.

The BMJ used my review of Gary Taube’s book, The Diet Delusion, to start off their new Round Table feature [full text link to BMJ].

The published version had some big cuts so I publish the original version here.  Taubes was kind enough to send me a copy of the book after I’d mentioned his wonderful New York Times piece in my previous excursion into the murky world of diet and health, Diet and health. What can you believe: or does bacon kill you?

diet delusion cover

The biggest omission in the BMJ version was Taubes’ own ten point summary of his conclusions (on page 454).

"“As I emerge from this research, though, certain conclusions seem inescapable to me, based on existing knowledge

  1. Dietary fat, whether saturated or not, is not a cause of obesity, heart disease, or any other chronic disease of civilization
  2. The problem is the carbohydrates in the diet, their effect on insulin secretion, and thus the hormonal regulation of homeostasis – the entire harmonic ensemble of the human body.  The more easily digestible and refined the carbohydrates, the greater the effect on our health, weight, and well-being.
  3. Sugars – sucrose and high-fructose corn syrup specifically – are particularly harmful, probably because the combination of fructose and glucose simultaneously elevates insulin levels while overloading the liver with carbohydrates.
  4. Through their direct effect on insulin and blood sugar, refined carbohydrates, starches, and sugars are the dietary cause of coronary heart disease and diabetes.  They are the most likely dietary causes of cancer, Alzheimer’s disease, and the other chronic diseases of civilization.
  5. Obesity is a disorder of excess fat accumulation, not overeating, and not sedentary behaviour.
  6. Consuming excess calories does not cause us to grow fatter, any more than it causes a child to grow taller.  Expending more energy than we consume does not lead to long-term weight loss; it leads to hunger.
  7. Fattening and obesity are caused by an imbalance – a disequilibrium – in the hormonal regulation of adipose tissue and fat metabolism.  Fat synthesis and storage exceed the mobilization of fat from the adipose tissue and its subsequent oxidation.  We become leaner when the hormonal regulation of the fat tissue reverses this balance.
  8. Insulin is the primary regulator of fat storage.  When insulin levels are elevated – either chronically or after a meal – we accumulate fat in our fat tissue.  When insulin levels fall, we release fat from our fat tissue and use it for fuel. 
  9. By stimulating insulin secretion, carbohydrates make us fat and ultimately cause obesity.  The fewer carbohydrates we consume, the leaner we will be.
  10. By driving fat accumulation, carbohydrates also increase hunger and decrease the amount of energy we expend in metabolism and physical activity.”

It is on these bases that Taubes suggests that the increase in obesity is, in part, a consequence of the recommendation of a  low fat, and hence high sugar diet.

The Diet Delusion [ pp 601]

(published in the USA as Good Calories, Bad Calories)

Gary Taubes 2008

There is no topic more widely discussed than what one should eat in order to stay healthy. And there are few topics where the evidence is so lacking in quality. It is also a topic that is besieged by gurus, cranks and supplement hucksters.

You need to beware of misleading titles.  Dietitians are good.  Nutritionists are sometimes  good.  But titles like ‘nutritional therapist’ and ‘nutritional medicine’ are usually warning signs of alternative therapists and/or pill salespeople.

Gary Taubes is a journalist, but he is quite an exceptional journalist.  His account of the importance of randomisation for the establishment of causality is one of the best ever and it was published not in an academic journal, but in the New York Times [1].  His book, The Diet Delusion, is in the same mould.  It is more complete and more scholarly than most professional scientists could manage.  Not only does it cover the literature right back to Samuel Johnson, but it is also particularly good at unravelling what one might call the politics of science.  And by politics I don’t mean the vast lobbying industry that has built up with the aim of selling you unnecessary supplements, but the politics of academia.

Obesity sounds simple.  If you are fat it is because you eat too much or exercise too little, right?  Well no, it’s not as simple as that.  For a start, it has been shown time and time again that low fat diets, and exercise, have small and temporary effects on weight.  The problem with diet and health revolves round causality.  The law of conservation of energy is an inevitable truth, but says nothing about causality.  It could imply that you get fat because you eat too much, or equally the causal arrow could point the other way and “we eat more, move less and have less energy to expend because we are metabolically or hormonally driven to get fat”.  The assumption that positive caloric balance is the cause of weight gain has predominated since the 1970s and “this simple misconception has led to a century of misguided obesity research”.

At the heart of the problem is the paucity of randomised trials, which are the only way to establish causality.   Those that there are have usually shown that diet does not matter as much as we are told.  Taubes concludes

“It does little good to continue basing public health recommendations and dietary advice on association studies (the Framingham Heart study and the Nurses Health Study are prominent examples) that are incapable of reliably establishing cause and effect.”

I think it can certainly be argued that the problem of causality has been greatly underestimated. We are warned constantly of the dangers of processed meat, on the basis of very unconvincing evidence [2].

This is one reason why we still know so little about the causes of obesity, diabetes and heart disease.  For Taubes, a major villain was the US nutritionist  Ancel Keys (1904 – 2004).  His
forceful advocacy of the low-fat hypothesis in the early 1970s, was, says Taubes, based on ignoring the many studies that did not agree with the idea.  It seems that the sort of citation
bias, recently described systematically by Greenberg [3], resulted in great exaggeration of the strength of the evidence.    

It is quite possible that there was rather more to be said for the Atkins diet than was apparent at the time.  The fact that Atkins was not a university scientist, that his views were extreme and that he was so obviously out to make a lot of money from it, gave him all the appearance of being yet another profiteering diet guru. He was dismissed by the medical establishment as a quack.  Taubes points out that conflict of interest cuts both ways.  Atkins’ sternest critics at Harvard were funded by General Foods, Coca-Cola and the sugar industry. It adds up to a sorry story of a conflict of vested interests and scientific vanity.

Taubes’ final judgement is harsh.  He quotes Robert Merton’s description [4] of what science is, or should be.

“The organization of science operates as a system of institutionalized vigilance, involving competitive cooperation”
In such a system, scientists are at the ready to pick apart .and assess each new claim to knowledge. This unending exchange of critical appraisal, of praise and punishment, is developed in science to a degree that makes the monitoring of children’s behavior by their parents seem little more than child’s play".

He then comments

“The institutionalized vigilance, “this unending exchange of critical judgment”, is nowhere to be found in the study of nutrition, chronic disease, and obesity, and it hasn’t been for decades.”

It took Taubes five years to write this book, and he has nothing to sell apart from his ideas.  No wonder it is so much better than a scientist can produce.  Such is the corruption of science by the cult of managerialism that no university would allow you to spend five years on a book [5].  I find all ten points in his summary convincing.  But his most important conclusion is that you cannot have any certainty without randomised trials. 

The business of nutrition is greatly at fault for not having put more effort into organising randomised trials.  Until they are done, we’ll never really know, and we shouldn’t pretend that we do.

1.   Taubes G. Do we really know what makes us healthy? New York Times 2007 Sep 16.[full text link]  [pdf file]

2.   Colquhoun, D. (3 May 2009) Diet and health. What can you believe: or does bacon kill you?.

3.  Greenberg, S.A.. 2009  How citation distortions create unfounded authority: analysis of a citation network.   BMJ ;339:b2680 [pdf file].

4.   Merton, R. K. Behavior Patterns of Scientists . Leonardo, Vol.3 1970; 3(2):213-220. From Jstor [or pdf file]

5.   Lawrence PA. The mismeasurement of science. Curr Biol 2007; 17(15):R583-R585.PM:17686424 [pdf file]  [commentary]

If length had allowed, there should certainly have been a reference here to Robert Lustig of UCSF. He is an academic nutritionist who supports the main thesis of Taubes’ book. See, for example, his 2005 review, Childhood obesity: behavioral aberration or biochemical drive? Reinterpreting the First Law of Thermodynamics [full
text link
]. Lustig’s slide show, The Trouble with Fructose is available in the NIH web site.

There are a couple of other articles by Taubes that are well worth reading. The Scientist and the Stairmaster Why most of us believe that exercise makes us thinner—and why we’re wrong. Gary Taubes, in New York Magazine, and We can’t work it out, in the Guardian.

You can see Taubes in action on YouTube, for example in “on Cholesterol and Science Practices“, and “on Carbohydrates and Degenerative Diseases“. There is also a video of Taubes on medical grand rounds at Dartmouth-Hitchcock Medical Center in June 2009. You can see Robert Lustig on YouTube too: “Sugar: The Bitter Truth“.


22 December.2009,  Unlike the serious questions dealt with in the Diet Delusion, this concerns merely another bit of the ubiquitous nutribollocks that crops up in the media,  While writing this I was listening to the excellent early evening news programme, PM, run by Eddie Mair, when a diet-related topic came up,  it was nonsense about how a cocktail made with vodka, cointreau, acai juice and pomegranate juice would not give you a hangover.  I suppose it was meant as christmas fun but  whenever I hear the words ‘antioxidant‘ or ‘superfood; I feel an email coming on.  It seems that Eddie Mair liked the fact that the email contained the words ‘quack’ and ‘codswallop’  because the next thing I knew I was asked to give an interview on next day’s programme.  The mp3 is here.

Print Friendly, PDF & Email

70 Responses to The Diet Delusion

  • davehodg says:

    OK, so how do we lower our insulin?

    Lots of complex carbs. And…?

  • Muscleman says:

    The problem with Dr Atkins and his diet is not that he identified carbohydrates as the problem but that he threw out the baby with the bathwater by doing away with them completely instead of recommending a change to low GI carbs and getting more of them from fibrous vegetables instead of potatoes, rice, refined grains etc.

    Also Taubes’s little scheme with insulin is all very well but all such physiological schemes are inherently circular as anyone who has ever had to write essay answers on such control systems discovers. It is why I always drew flow diagrams at the head of all my essays so I could keep track of where I was.

    The question is thus how you get out of the wrong situation. The problem is that there has been too much focus on a dietary solution when it is clear exercise cannot be ignored. i fear this is largely because the medical profession, on both sides of the Atlantic, has given up on the idea of getting their patients to be more active. The decision to mass medicate the middle aged with statins is a de facto admission of this failure.

    BTW expending more calories than you consume only leads to hunger under certain circumstances. We have known since at least the mid ’80s when I did my undergrad degree that the appetites of the sedentary exceed their calorie requirements, the appetites of the moderately active closely matches their calorie requirements and the appetites of the extremely active are insufficient for their calorie requirements. In my youth when I ran 100miles a week and cycled 50 (to and from school) I had to force myself to eat. One Sunday morning I hit the wall after barely 5 miles after setting out to run 20. On reviewing what I had eaten over the previous 48hours it was clearly insufficient and not having very much body fat to fall back on my brain called a halt.

    I have lost a noticeable amount of work since I finally managed to put a continuous running program together (currently problematic, can’t run on ice) by eating a low GI diet, not taking second helpings and snacking on things like carrot sticks. I am currently hovering on the border of needing to ensure I eat enough which makes life a bit difficult.

    Oh and I can easily touch my toes, despite being a distance runner 😉

  • Felix says:

    The video of Gary Taubes on ‘Grand medical Rounds’ is excellent!

    Very dense, interesting, well presented and the site has 1 window showing him talking with another for the slides making it very easy to follow. (45 mins approx)

    Highly compelling, I watched 2 versions of Gary’s talk for a total length of 2 hours because it was so good.

    However, ‘Sugar: The Bitter Truth’ with Robert Lustig is a different animal!
    He talks like he is trying to sell you something, or put one over on you.
    There is no appreciable science for the first 40 minutes which is as far as I have managed to get. Rather lots of innuendo and hyperbole about sugar, industry, Coca Cola, Richard Nixon etc.

    A sample of his style:
    “Do you see that anywhere??!! He didn’t do it! He didn’t do it! He didn’t do the thing you need to do […] This was done before computers – we can’t check the work. He’s dead! {…} So we’re left with a conundrum – do we believe this study?”

    Does any one know of any recent studies in this area based on this reconsideration of the carbs issue?

    Also any well based any diet advice? 🙂

  • SueMcG says:

    There has been a (£2.7 million) trial – Project N02031 ran from 2004 to end 2007. However no final report has appeared on http://www.foodbase.org.uk

    Text of Food Standards Agency press release:

    “New study searches for hearty diets

    Wednesday 12 May 2004

    Today sees the launch of a large-scale study, funded by the Food Standards Agency, to find the best diet to prevent heart disease and type-2 diabetes.

    Led by the Medical Research Council, the £2.7 million study will take four years and involve scientists from the MRC’s centre for Human Nutrition Research at Cambridge, Imperial College London, King’s College London, the University of Reading and the University of Surrey.

    The researchers will look at the how types and amounts of fat and carbohydrate in the diet relate to the risk of people developing heart disease or type-2 diabetes.

    A total of 650 volunteers will have their diets modified and the results monitored. It is hoped that this will enable the Agency to recommend the sorts of diets that will reduce the likelihood of these health problems.”

  • David Brown says:

    For a while there I agreed with Taubes’ first conclusion, that “Dietary fat, whether saturated or not, is not a cause of obesity, heart disease, or any other chronic disease of civilization.” In fact, for many years I thought that excessive sugar intake was the major ingredient in the modern food supply responsible for heart disease. But after listening to a recent 37 minute lecture by Dr. Bill Lands entitled “Why Omega-6 Fats Matter for Your Health” I’m astonished at the lack of scientific interest in studying the effects of excessive omega-6 seed oil consumption on the health of humans.

    Taubes did not address this issue because there have been no scientific trials specifically targeting omega-6. For many years, all the attention was focused on saturated fats. Then scientists began studying the effects of trans fats. Most recently, there’s been a lot of research on the benefits of increasing omega-3 fatty acid intake.

    At this point Taubes seems to view the trans fat issue as a red herring and I’m inclined to agree. That’s not to say that it’s safe to consume trans fats. But it may turn out that a blend of omega-6 and trans fats is responsible for the observed effects.

    I urge all who read this comment to Google “Omega-6 Research News and explore Evelyn Tribole’s unique website. Through it you can access the video presentation mentioned above.

  • Dr Aust says:

    Errr…. …did you mean:

    “I have lost a noticeable amount of weight


    … though a lot of running might leave less time for working – not that there’s anything wrong with having less time for working…

  • @SueMcG

    The problem of registered trials that never result in final publications is quite a serious one.

    Did you write to the FSA or the MRC to find out what happened?

  • Muscleman says:

    Whoops! though to be correct I have no idea if I have lost weight, since we have no bathroom scales. It is entirely possible that am heavier, having swapped adipose tissue for more lean muscle and denser bones.

    I have not seen anything about omega 6, but I do know that plant based omega 3 and 6 are not chemically the same as fish based 3 and 6. Apparently only 1% of plant omega 3 is converted into the fish form, which is a lot of linseeds. It also implies that all those ‘enriched’ margarines are worthless. Eat a kipper instead.

    New Scientist reported recently that a number of agritech companies are trying to engineer oil seed plants to make the fish form of 3 and 6, both to improve the health of people and to conserve fish stocks.

    Also the message to eat more omega oils is supposed to be within the parameters of a healthy total intake of dietary fats. You are supposed to replace the bad fats with the good ones, not eat lots of ‘good’ oil on top.

  • […] to shift a few of those extra pounds that you’ve put on over the Christmas holidays? According to David Colquhoun’s article, which is essentially a glowing book review of Gary Taube’s ‘The Diet Delusion’, Taube claims […]

  • anoopbal says:

    Mr. Taube can only see studies dating back to 1950’s.

    Has Mr. Taube ever heard about a technique called “Doubly-labelled water” which can accurately measure calorie intake and has clearly shown obese people underestimate their food intake by 30 -50%. Taube’s main argument is that in the 1960’s people who were obese, ate less. Really?

    And since he is so into epidemiological studies, 70% of Asian diet is carbs. And I dont see too many obese Asians.Do you?

    Taube is good example of a science writer who badly needs some help with evidence based -knowledge.

  • anoopba
    Have you actually read the book? The fact that you say “since he is so into epidemiological studies” suggests to me that you haven’t. The constant underlying theme of Tuabes’ book is that you can NOT rely on observational epidemiology, that causality is crucial and that randomisation is the only way to establish causality (the fact that RCTs are hard to do doesn’t alter that inconvenient truth).

  • anoopbal says:

    Hi David,

    Agreed, I haven’t read his book, but I am talking about his book, “Good Calories, Bad Calories”.

    His conclusion it is not calories, but it carbohydrates which makes people fat.

    And his biggest evidence is how in the 1960’s people who ate less, were fat. This has puzzled scientists for along time. And it’s not surprising considering they did dietary recalls. Now after the doubly labeled water technique it is clear that obese people underestimate their calorie intake by 30 -50%.

    And people are fat because they eat more calories. There is plenty of RCT’s to show that in humans & animals.

    And I haven’t read his recent book. And epidemiological research do not stand on it’s own. And you don’t have to prove causailty everytime to help people. Cholera is case in point. Smoking is case in point. If it weren’t for epidemiological studies, most of us would be dead by now.

    And I find it is funny he talks against epidemiological studies, but his first book was everything about it and he ignored every basic science study which shows it’s calories that matters most in obesity.

  • anoopbal
    If you had read a bit more carefully, you would have seen that Good Calories, Bad Calories is the name used in the USA for the book known as The Diet Delusion in the UK

    I should have looked at your web site the first time, because I realise now that you make your living from exercise. Although your writings there seem to make a real attempt to stick to evidence (our views on anti-oxidants don’t differ greatly, for example), I have to say that I feel you have missed the whole point of Taubes’ book.

    Nobody is denying the first law of thermodynamics. That isn’t the point. The point is that you can’t use the information sensibly until you know which way causality lies, and simply repeating “it’s calories that matters most in obesity” tells you nothing whatsoever about that critical question..

  • anoopbal says:

    Thanks for the comment and checking my website.

    Let’s keep it simple: So you really think that it’s not calories. it’s carbohydrates that make people fat?

    I heard Taube talk about the book in a 2 hr presentation and I didn’t feel the need to buy the book. His conclusion is that it’s not positive energy balance that makes you obese, but its carbohydrates. He ignores all the control studies whic shows eating more calories makes you fat and eating less cause you lose weight and only look at epidemiological studies dating back to 1960’s to come to his conclusions. His main argument is how obese people at less which has now been clearly shown to be wrong with “doubly labelled water techniques”

    Read the review by Bray GA about Gary Taube’s book. Gary is a good example of ignoring opposing data to drive home his point.

    Click to access bray-review-of-gcbc.pdf

    And just a suggestion: It would be good to have a preview button to see what you wrote and maybe make the comment submit area a bit bigger.

    And nice blog!

  • anoopbal

    Thanks for the review, which looks pretty good. But I still think you are missing the main point. People, whether they are fat or thin, are in (astonishingly precise) energy balance for almost all of the time.. What matters is what controls the ‘set point’, the more or less constant weight achieved in the steady state. In order to know what to do about it, it us crucial know which direction the arrow of causality points. In the absence of RCTs there is NO sure way to find this out Consequently there is no sure way to be sure why obesity appears to be getting more common. I think Taubes makes a good argument for his view of the problem, but in the end it is a matter of not pretending to know things that we do not know. If we did know, obesity would not tbe the problem that it is.

  • anoopbal says:

    I will wait for your thoughts on the Review I gave you.

  • anoopbal

    I have read the review by Bray, It would have been rather fairer if you had also provided a link to Taubes’ response to Bray. In it he says, for example,

    “Bray also consistently confuses associations – the obese eat more than the lean; the obese are in positive energy balance as they fatten – with causes and effects. Do they get fatter because they overeat, as Bray continues to imply, or do they overeat because they’re getting fatter. The goal of science is to correctly determine causality. In these two competing hypotheses, the causalities are diametrically opposed.”

    Personally I find his response quite convincing.

  • anoopbal says:

    Taube talks a lot about causality and what does he conclude: not calories but it’s carbs that makes you fat. Isn’t that causality?

    And where is his RCT that proves his point. I can understand if he refutes the calorie theory fr the lack of highly controlled RCT’s and direct causative studies. But saying there is no causality and then coming with a causality statements is just ridiculous

    And honestly, do you really think all these obesity researchers got it all wrong all these years?

    Here is a recent RCT on diet macro nutrients:

    N Engl J Med. 2009 Feb 26;360(9):859-73.
    Comparison of weight-loss diets with different compositions of fat, protein, and carbohydrates.

    Sacks FM, Bray GA, Carey VJ, Smith SR, Ryan DH, Anton SD, McManus K, Champagne CM, Bishop LM, Laranjo N, Leboff MS, Rood JC, de Jonge L, Greenway FL, Loria CM, Obarzanek E, Williamson DA.

    Department of Nutrition, Harvard School of Public Health, Boston, USA..

    BACKGROUND: The possible advantage for weight loss of a diet that emphasizes protein, fat, or carbohydrates has not been established, and there are few studies that extend beyond 1 year. METHODS: We randomly assigned 811 overweight adults to one of four diets; the targeted percentages of energy derived from fat, protein, and carbohydrates in the four diets were 20, 15, and 65%; 20, 25, and 55%; 40, 15, and 45%; and 40, 25, and 35%. The diets consisted of similar foods and met guidelines for cardiovascular health. The participants were offered group and individual instructional sessions for 2 years. The primary outcome was the change in body weight after 2 years in two-by-two factorial comparisons of low fat versus high fat and average protein versus high protein and in the comparison of highest and lowest carbohydrate content. RESULTS: At 6 months, participants assigned to each diet had lost an average of 6 kg, which represented 7% of their initial weight; they began to regain weight after 12 months. By 2 years, weight loss remained similar in those who were assigned to a diet with 15% protein and those assigned to a diet with 25% protein (3.0 and 3.6 kg, respectively); in those assigned to a diet with 20% fat and those assigned to a diet with 40% fat (3.3 kg for both groups); and in those assigned to a diet with 65% carbohydrates and those assigned to a diet with 35% carbohydrates (2.9 and 3.4 kg, respectively) (P>0.20 for all comparisons). Among the 80% of participants who completed the trial, the average weight loss was 4 kg; 14 to 15% of the participants had a reduction of at least 10% of their initial body weight. Satiety, hunger, satisfaction with the diet, and attendance at group sessions were similar for all diets; attendance was strongly associated with weight loss (0.2 kg per session attended). The diets improved lipid-related risk factors and fasting insulin levels.

    CONCLUSIONS: Reduced-calorie diets result in clinically meaningful weight loss regardless of which macronutrients they emphasize.

  • Yes, I think that if you were involved in research yourself, you would realise that it is entirely possible for researchers to get the story wrong over many years. That is particularly true in an area like this where so much of the data are based on unreliable observational epidemiology.

    I quite like the Sacks paper that you cite. I discussed it in my earlier post about the strength of the evidence that processed meat causes colorectal cancer (conclusion, very weak). What you fail to mention is that none of the diets were very successful in the long run, and you might also have mentioned the criticisms of the Sacks paper by Katan (2009).

  • Dr Aust says:

    The link to “Taubes’ response to Bray” seems to be broken.

  • Oops, thanks Dr Aust. The link to Taubes’ response to Bray is now fixed.

  • mjmanning says:

    This raises an interesting issue: What should the public do in cases like this? Generally speaking, it’s sensible for non-specialists to defer to the scientific establishment. If there’s anything like an orthodox opinion it makes sense for non-scientists to accept what they say (taking “acceptance” to mean something less than 100% degree of belief). We shoud align our beliefs with those of the experts. As Ben Goldacre points out, it’s always possible to find some academic somewhere who will support any claim you like. Often it’s easy enough for even someone like me to spot that they’re peddling dubious nonsense.

    Just to be clear, I’m not criticizing either Taubes or your defence of him (I also haven’t read the book, but I’ll certainly do so). The point is that, lacking any background, I’m not in a position to adjudicate. It’s interesting to ask what the public should do when someone comes along and says the whole academic diet industry is plain wrong. Even if Taubes is right, wouldn’t it be irrational for people to diverge from the standard dietary advice?

  • Muscleman says:

    It seems to me that everyone is talking about causality of different things, as Bray says in his review. There is great sloppiness in confusing: how do people become overweight; how do they then become obese; how do people, underweight, normal, overweight, obese regulate their weight; do they regulate their weight long term; and finally: can you keep weight off by diet alone.

    David you responded to Anoopbal’s point by changing to: but do dieters keep the weight off? which is again shifting the locus of the question being asked.

    It is perhaps not obvious enough that studying the already obese will not necessarily tell you how they became obese. You must study a normal weight/overweight population over time and watch as some become obese. This however introduces a plethora of problems both procedural and ethical. Not to mention recruitment, retention and funding it all. Hence the debate will rumble on, though privileging people like Taubes just because they are taking on the establishment is not helpful. Bray does a good job of pointing out particularly wrt double labelled water studies that the obese both eat more and underreport calorie consumption. Since Taubes relies on the obese eating less based on reporting his hypothesis falls down at base. Bray’s point about hyperinsulinema is good one too. Taubes is wrong, insulin is necessary for obesity but it is not sufficient. Put both together and the whole thing fails. Even if he is right and calories do not matter, he is wrong about the mechanism which means his corrective is unlikely to work.

  • Muscleman says:

    I should point out that the idea that you can’t keep weight off with a calorie controlled diet is only true in an environment where calories are not in fact limited. Otherwise nobody would ever starve.

    Anyway all this still ignores the effect of inactivity. We know from studies using accelerometers that we are becoming ever more inactive in daily life as we drive or are driven everywhere, use escalators and lifts and spend our leisure watching TV or sitting arguing on the ‘net 😉

    But we don’t talk about that because we have as a society essentially given up on getting people to be more active except by massive restructuring of cities which have only moderate effects. Again the fact of mass medication with statins is a tacit admission of this though it is never presented as such since we don’t even want to hear the exercise message any more.

  • dikken says:

    D.C.’s original post and the comments are exactly the reason that so many people are so confused about science and dietary advice. I understand that the human body is complex, and that we are still surprised by new discoveries, but why can’t we discover what we are supposed to eat to be healthy?

    No wonder so many articles in magazines are about food – no one knows what is good or bad.

    I am confused and absolutely disgusted about what should be, by now, established science. Can anyone explain why this is so difficult?

  • Asclepius says:

    Muscleman, you appear to be making the case here for:

    “level of obesity = calories in – calories out”

    This belief is tackled by Taubes in the book. ‘Calories In’ and ‘Calories Out’ are not independent variables. You cannot change one without affecting the other.

    If you increase calories out by doing more exercise, your body will make you eat more – it used to be called ‘working up an appetite’. Similarly if you reduce ‘calories in’ then the body can make you lethargic to reduce calories out – and thus maintain the balance.

    That is one of the great insights of Taubes work – that the at a cellular level, the appetite and energy expenditure of the thin and obese IS in balance. The problem is that unlike the thin who can access/cycle their fat stores due to lower levels of circulating insulin, the obese cannot sequester their fat stores and so are driven to eat more.

    You cannot escape the weakness of the “eat less, do more” model – no one can resist hunger for long and the body will go to extreme measures to maintain homeostasis.

    Part of the problem is that intense exercise would appear to upregulate insulin sensitivity thus giving the appearance that you have burned off the fat – but what you have actually done is reduced circulating insulin levels which change your hormonal environment to one conducive to the release and subsequent burning of fat. It is the ‘release of fat’ that is the important bit.

    Still not convinced? Think of it this way. If we were going round to a friends house for a meal with all the trimmings, and the friend told you to ‘bring your appetite’, what would you do to compel yourself to eat more? Most of us would ‘do more’ exercise (increase calories out), or ‘eat less’ by skipping a meal or eating a smaller breakfast (thus reducing calories in).

    So we are in an ironic position that the very things we tell people to do to lose weight are exactly the same things we do to compel ourselves to eat more.

  • RobertD says:

    Dear David, Thank you for this review. I have ordered a copy of the book – reviews do work! I can make better comment after reading it. Given the obsession with ‘diet’ and health I am always surprised that there have not been better studies.

  • Beth@IDblog says:

    Taubes makes a point at the end of the Dartmouth medical grand rounds video that I think is important:

    “I’m not trying to convince you that it’s true, I’m trying to convince you that it should be taken seriously.”

    What Taubes is saying in his presentations and his book is that the conventional explanation for obesity (we eat too much and move too little) does NOT explain ALL the data available. This includes both modern and early 20th century populations who could not be eating too much because they couldn’t GET too much.

    In particular, I thought the pointer he gave to Benjamin Caballero’s Nutrition Paradox in the NEJM re obese mothers and underweight children compelling. But then, perhaps the thermodynamics folks think it’s plausible the mothers are underfeeding their kids while overfeeding themselves?

    So I definitely agree with Taubes when he says:

    “If you want to prevent obesity and diabetes, and the chronic diseases that associate with it, you have to get the cause of obesity correct.”

    Whether or not you buy his premise that carbs are the root of the problem, I think he does lay out a good case that obesity is not adequately explained by the conventional wisdom. And given the role insulin plays in fat metabolism, it’s hard for me to ignore dietary carb as a concern wrt obesity.

    That said, I too wonder how populations like the Japanese and Kitavans fit in with Taubes’ hypothesis (Lustig argues that, unlike standard Western diets, these diets are very low in fructose).

    Me, I suspect it is more complicated than just dietary carb intake, and certainly a lot more complicated than eat less, move more. I’m watching Stephan Guyenet tackle this over on Whole Health Source. Maybe it’s fructose, maybe it’s lectins and other anti-nutrients in wheat, maybe it’s all the problematic omega 6/omega 3 balance in our diet, maybe it’s vitamin D deficiency.

    Whatever it’s cause (and I very strongly doubt there’s only one single cause), obesity is very clearly a wicked problem: http://en.wikipedia.org/wiki/Wicked_problem

    BTW, re science, Jonah Lehrer has a great piece up on Wired on how scientists study things, and why it’s harder than you think: http://www.wired.com/magazine/2009/12/fail_accept_defeat/all/1

  • Dangerous Conventional says:

    Interesting debate.I think on a previous link I discussed the benefits of metformin vs insulin/glibenclamide in type 2 diabetes (UKPDS). Taubes hypothesis could explain why metformin reduces micro and macrovascular complications and tight blood glucose control with insulin and glib. doesn’t. The last two drugs feeding the obesity? Nice to see conventional meds doing more harm than good, again? There is no evidence for tight blood glucose control in type 2 so why the hell does the NHS have targets.
    Just had an interesting discussion with a respiratory specialist today about why the hell we give patients with COPD steroid inhalers. Now how many millions are we wasting there (and extra cases of pneumonia). Makes the debate on homeopathy look rather pathetic.

  • Nash says:

    Does central heating contribute to obesity? In the UK, prior to the 1970’s central heating was not a common feature. How many calories do we burn off just keeping warm? One of my friends turned off their heating at night and lowered the temperature on the Thermostat to save energy 2 years ago and he and his wife have both lost weight since. They didn’t make any changes to their diet or exercise more. But both lost about half a stone each over 3 months.

  • Thanks especially to Asclepius and to Beth@IDblog for some very perceptive comments.

    I’m not so sure about the piece in Wired though. That seems to conclude “Experiments rarely tell us what we expect. That’s the dirty secret of science,” Nobody who has actually done experiments would describe that as secret; It is just the normal slog of everyday science. Neither is there anything the least “dirty” about it. On the contrary, it is how progress is made.

    You say

    “I am confused and absolutely disgusted about what should be, by now, established science. Can anyone explain why this is so difficult?”

    Yes I think it is quite easy to explain why. It is because it is genuinely extremely difficult to do long term randomised controlled trials of diets. That isn’t just a technical detail but rather it is at the very heart of the problem. It is, by now, well known that observational studies can easily give results that are the opposite of the truth, and cannot tell you what causes what,

    This is nobody’s fault. It is just an unfortunate fact of life. If there is fault anywhere, it lies in not saying more openly that we don’t know.

    It is really rather chastening to realise that enormously difficult problems in phsyics and maths can be solved, and we can walk on the moon, yet what sounds, at first hearing, like a very simple question turns out to be very hard indeed to answer.

  • Beth@IDblog says:

    David, sorry, I didn’t add any context to my link. I do find Jonah Lehrer’s stuff to be thought-provoking. In this case, I thought it interesting as a response to the question that anoopbal asked about all these obesity researchers getting it wrong.

    I’m not a scientist, but I thought the example he gave of the e coli researchers having a tougher time because they tended to look at things the same way as each other had a curious (tho perhaps not compelling) parallel as far as the adoption of the lipid hypothesis.

  • Beth@IDblog says:

    Oh, one other thing. Re your comment to dikken, the challenge of doing the science with humans and RCTs is certainly true. And what you can measure with participants in a lab when you feed them and measure their activity versus what they self report (and/or comply with) are two very, very different things!

    But I think it’s also true that obesity is not like gravity or the speed of light. It’s not like there is one phenomenon to be discovered … even one that’s difficult (a la Schrodinger’s cat).

    This is why I liken obesity to the story of the blind men and the elephant. Depending on where you are, it looks very different!

  • anoopbal says:

    Taube’s claims of his paradoxical explanations are observed by many. Only problem is that he missed the invention of “doubly-labelled water” which clearly explained all the puzzles observed by scientists. And rightly so, it was one of the greatest 5 discoveries of the century nominated by by more than 100 nutritionist and biochemist’s. And that was Taube biggest evidence.

    As Gray writes in his review”Producing positive energy balance by asking human
    subjects to eat extra food has routinely produced fat gain (20,21). Similarly, inducing energy deficit by asking human
    beings to exercise more or by restricting food intake has routinely produced fat and weight loss (24). If the author of
    Good Calories, Bad Calories wishes to challenge these fundamental controlled experiments, he needs to provide data showing that they are wrong.”

    And if indeed it is carbohydrates and not calories is the culprit, why has 100’s of studies done with low carb and high fat diets not shown any impressive numbers?? As Gray points out, why are people on insulin not obese? And Taube has no real clue about the De novo iploysis cycle. He doesn’t even mention in his book and Bray says that.

    And Acclepius, what you are saying is adaptive thermogenisis has been studied to death. It is not Taube’s new revelation?. And as Gray points, in the long term these homesotasis mechanism are overridden. And that’s exactly the reason 100’s of studies have shown people putting on weight even with all the homeostatic mechanisms functioning.

    From what I have seen, most people are not familiar with diet and metabolic studies.So people go by logic and common sense and anecdotes and Taube had a lot of these.

    Please read that GA Bray counter point against Taube.Bray is an obesity researcher who even drafted the section of the book which Taube cites as a major evidence for his revolutionary idea.

  • Asclepius says:

    Anoopbal – it is painfully obvious that you have not read GCBC. I don’t mean to be rude, but your comments show that you are ignorant of both Taubes position and the wealth of evidence he cites for it.

    This also makes discussing GCBC awkward as whilst I may discuss GCBC, you are discussing what others have told you about GCBC.

    Taubes might well be wrong, but we should at least take what he says seriously – and address the quality of his argument/evidence for it on objective grounds.

    Despite these shortcomings, let’s look at a practical examples. Have you seen a type one diabetic? They are skinny. Very skinny. Regardless of what they eat. They might put on weight, but only under medication with insulin.

    Now go find a type two diabetic. They are often obese. In fact even some cursory research will show you that there is an association between obesity and type two diabetes.

    T1 is caused by an inability to produce insulin. T2 is caused by poor insulin sensitivity (and so large and increasing amounts of insulin are required for a given response).

    Here we have two quite contrasting examples of body mass/composition, BOTH of which can be logically and CONSISTENTLY explained with a single theory, when considered via the role of insulin in fat storage and mobilisation.

    By contrast could you tell me how an increase in calories can explain this? By your logic it would appear that T1 are undereating and T2 are ‘overeating’.

    “Aha”, you might say, “but Aslcepius is confusing a hormonal issue with one of calorie consumption and expenditure”. But all this means is that whereas I can explain a single symptom (obesity), with a single theory (insulin), you are having to pump out multiple custom theories to explain the same single symptom.

    Occam’s razor!

  • Beth@IDblog says:

    Re exercise and/or restricting calories to produce weight loss, yes, they do: over the short term.

    Ref The Handbook of Obesity: “Dietary therapy remains the cornerstone of [obesity] treatment and the reduction of energy intake continues to be the basis of successful weight reduction programs … [the results] are known to be poor and not long-lasting.” Or Katan in the NEJM: “It is obvious by now that weight losses among participants in diet trials will at best average 3 to 4 kg after 2 to 4 years and that they will be less among people who are poor or uneducated, groups that are hit hardest by obesity.”

    Re comparing diets, different studies show different things. I’d encourage you to check out Christopher Gardner’s diet trial:

    Gardner is a Stanford researcher who compared four different diets (Ornish, LEARN, Zone, and Atkins) over a year. It shows some of the same challenges that Katan refers to (problems with compliance, drop-out, lack of weight loss or even gain), but there are some interesting findings nonetheless.

    The Atkins folks started out doing very low carb and were doing more moderate carbs by one year. Even so, these dieters had lost more weight those following different diets. In addition, the Atkins participants had more favorable changes in other metabolic areas (blood pressure, HDL, triglycerides) compared to some of the other diets.

    One very interesting finding Gardner had was that success on a lower carb or higher carb diet had a lot to do with the participant’s level of insulin sensitivity. Those who were sensitive did better with a higher carb diet; those who had some degree of insulin *resistance* did better with a lower carb diet.

    Gardner has been a vegetarian for 25 years, so acknowledged being taken aback by the results of his own findings. Well worth the watch.

  • dikken says:

    Thanks for the response, David. I had not realized that unraveling dietary requirements was so difficult. It does explain why there is so much contradictory information about.

    The complexity of the subject seems to mean that we get only glimpses of insight and then find out that it is more complicated than we can imagine. And since science doesn’t stand still, it only gets more complicated.

  • anoopbal says:

    Hi Asclpeius,

    I haven’t read Taube book and I wrote it before. And I don’t see the need to buy the book by just looking at his conclusions and I have watched his 2 hr video full of anecdotes, “he said that”, and his studies from 1960’s without never showing any data which goes against his view point.

    And for your practical example.

    1. First, people have Typ1 diabetes take insulin . If not, they are DEAD! So what you see are just almost normal people.

    2. Type 2 diabetes means the muscles & fat are resistant to insulin. So according to Taube logic (and your logic) they should never be obese. And when they take drugs, they should be more insulin sensitive so they get more fat according to your logic.

    3. You don’t take examples of pathological conditions to explain obesity. I brought the diabetes thing up because Taube talks about it in his book and Bray replies to his in article.

  • anoopbal says:

    Thanks for the you tube video.

    And as he rightly says, it’s all about adherence. And we know the best diet is the one which you can stick with.

    People who are insulin resistant usually do well with low carb diets which is known. But now we have some scientific evidence for it too. I have written it in an article which I wrote 2 yrs back.

    check this: http://www.exercisebiology.com/index.php/site/articles/the_best_diet_for_fat_loss/

    Unlike the Sack’s study which I posted, this study is a comparison of how well people do on popular diets and not a controlled low carb and a high carb study. As the author mentions, you cannot just extrapolate it to a comparison of high carb and land low carb diet. But it has certainly a practical value because that’s how people diet.

  • Asclepius says:

    Anoopbal, you may have “watched his 2 hr video full of anecdotes” but I really do recommend you read his several hundred page book that is full of references.

    There is a lead time up to diagnosis of T1 and T2. You would not mistake a T1 with a T2 in this time.

    I beleive that T2 are often awash with insulin. There problem, at least initially, is their lack of sensitivity to insulin.

    How do you explain GT’s example of the distribution of obesity in the native American popluation both pre and post adjustment to a diet high in refined carbohydrate?

  • Asclepius says:

    *apologies for the spelling mistakes – I was in a rush!

  • anoopbal says:

    I will read his book just to write a review on it.

    And not about anecdotes, he had studies, but only which goes with his conclusions and studies dating back to the 1960’s. The hallmark of a good researcher is to say both sides of the story.

    Lead time just basically means that the pancreas are still producing little bit insulin and will shut down soon. If insulin was all that needed, all Type 1 & Type 2 diabetic individuals will be morbidly obese. Insulin is necessary but not sufficient for obesity.

    If insulin was all that needed, every weight LOOS diet study with a high carb study would show weight gain than weight loss. Insulin only works with calories.

    And can you further explain the native american example?

  • Asclepius says:

    Anoopbal, if you go ahead and read his book, then I think you will be genuinely and pleasantly surprised. It is to your credit that have reconsidered and taken this step. I look forwards to your review!

    The native american example involves the Pima (p237 in GCBC). In 1846 a US army battalion passed through the Pima lands. The battalion’s surgeon noted that the Pima were ‘sprightly’ and in ‘fine health’. He also noted that they had an abundance of food and saw several well-stocked storehouses.

    Within a few decades a wagon route had been opened up through the Pima lands. New settlers hunted the local game to extinction and the river (a source of fish), dried up due to upstream irrigation projects.

    This plunged the Pima in to a state of poverty and dependency. By about 1895 the Pima were living on government rations (the Pima called this the ‘years of famine’). Subsequently an anthopologist (and physician) called Hrdlicka passed through the Pima area (around 1905). He noted the Pimas transition to the ‘white-mans diet’ and specifically their fondness for sugar, coffee and canned goods. He also noted the high levels of obesity amongst (predominantly) Pima women.

    What is specifically interesting here is that the Pima women performed much of the manual labour and worked ‘considerably harder than the men’.

    So this example throws up a few paradoxes:

    1) Obesity was not present initially in the Pima, despite an abundance of food.
    2) Obesity appeared in an environment of poverty, following the ‘years of famine’ on a government diet featuring highly refined foods.
    3) Obesity was predominant amongst those that performed the majority of the manual labour in the Pima tribe.

    Odd that obesity should accompany poverty, manual labour and refined foods. Stranger still that it should be scarce in an environment of abundance!

    There are other travellers, frontiersmen, ethnographer and so forth who chart a similar story concerning the spread of both obesity and ‘wester diseases’


  • andrew says:

    @ Asclepius

    Here is a more recent (2000) paper on diet & exercise in the Pima Indians of Arizona & North West Mexico :-


  • Asclepius says:

    First an update – my last post above should read ‘Western Diseases’.

    Andrew – thanks for that link. I don’t normally like to get drawn too deeply in to these dicussions as I often end up pretty isolated when flying the low-carb flag and it usually results with a range of folk against me – all evidently more qualified and armed with various scientific papers, which I am then expected to begin dissecting!

    Having said that, let me put a few thoughts together from which (I hope), you will be motivated to then go on to read Taubes’ book.

    Leaving aside whether the two Pima are actually the same tribe (“The degree to which the Mexican Pima and the USA Pima Indians are related is uncertain”), this paper comes over a century after the documented events Taubes identifies with the Pima. No matter how robust the paper, it does not make the observation by Hrdlicka for example (that the obese Pima women worked much harder than the less frequently obese men), any less valid and thus requiring explanation. In GCBC, GT gives other examples of this paradox of hard labbour and obesity.

    Another issue to consider is that the paper alludes to the obese Pima living in an environment that “provides unlimited opportunities for the consumption of excessive food.” Well, if you read my post above you will see that “[in] 1846 a US army battalion passed through the Pima lands. The battalion’s surgeon noted that the Pima were ’sprightly’ and in ‘fine health’. He also noted that they had an abundance of food and saw several well-stocked storehouses”. So here is another conundrum. The US Pima have been noted as being in ‘fine health’ in an environment of ‘abundant food’ prior to the papers’ observation they are obese in an environment of abundant food.

    Now back to the paper itself and the comparison of the two tribes. I cannot see any breakdown of macro nutrient composition. Given that GT says carbs drive insulin, drives fat storage, and given that refined carbs in particular promote secretion of insulin, this is a massive hole in the use of the paper to refute GT’s notes on the Pima I quoted above.

    I just want to reiterate one point – perhaps the most difficult one to really get a grip on in this whole business. The VERY first line of the paper starts with an assumption ‘Obesity is caused by an imbalance between energy intake and energy expenditure.’ This is exactly the issue Taubes deconstructs. Let us look at the basic equation again:

    Body Mass = Calories In – Calories Out

    I think that we can all agree on this equation – reflecting as it does the first law of thermodynamics. However these researchers have waded in with an assumption that there is an ‘imbalance’ between the RHS of the equation that is causing the LHS to increase. But where is the arrow of causality? You might say that of course if you increase ‘calories in’ then body mass will increase. Or that if you reduce ‘calories out’ (as implied with the Pima), then body mass will increase.

    But there are instances where the LHS of the equation drives the RHS; where body mass directly affects ‘calories in’ (by forcing an increase in calorie uptake), and reducing ‘calories out’ (through lethargy). Anyone who has teenage children will know exactly the scenario I describe! They eat loads and sleep all the time. Similarly pregnant women feel very tired in the first trimester (so a reduction in calories out), as nutrients and so forth are directed to the developing foetus and placenta.

    Hormones govern the LHS of the equation and can drive the RHS. Thinking about energy at the cellular level clarifies this point. If insulin in elevated in the blood stream then fat cannot be released from storage. Thus at a cellular level, your body is starving and so you are compelled to eat REGARDLESS of whether you are fat or thin.

    There is not only the LHS/RHS issue to consider, but also the belief that you can change ‘calories in’ and keep ‘calories out’ constant (or vice versa).

    If you cut ‘calories in’ via a traditional calorie-restricted diet, then your body can compensate and cut ‘calories out’ by inducing lethargy. You have no control over this – it is way downstream of your control. If you increase ‘calories out’ through exercise and maintain your levels of ‘calories in’, your mitochondria are now working hard and they need fuel. You are working up an appetite. If insulin is elevated in the body then those juicy fat stores are a no go. The fat is unavailable and so at the cellular level you are starving! Something has to give and to restore balance you will be made hungry or fatigued (to reduce energy out).

    Do you see that by either reducing ‘calories in’ or increasing ‘calories out’ you end up hungry and eventually fatigued?

    I was dismayed to find that it was rebranded on release in the UK. Any book with ‘delusion’ in the title was going to come away as some lightweight tome reliant on an association (any association) with RD’s ‘The God Delusion’. Putting ‘diet’ in the title was further going to alienate the intellectual readership as it has and continues to allow many to dismiss ‘Good Calories, Bad Calories’ (aka The Diet Delusion) as ‘simply a diet book’.

    GCBC/TDD is NOT a diet book. It is a well researched scientific tome covering a broad range of subjects from ethnography and biology to politics and cytology. It is difficult to pigeon-hole.

    My god, I have already written a bloody essay. I hope I didn’t bore you. It is late and I hope I have conveyed a sensible argument without too many errors. GT says all this in much more detail and more eloquently than I could ever do. I cannot recommend his book highly enough.

    Hope that helps.

  • Muscleman says:


    You wrote:
    “Now go find a type two diabetic. They are often obese. In fact even some cursory research will show you that there is an association between obesity and type two diabetes.”

    Yet the two people I know who are type 2 (unrelated) are skinny. Obesity is neither necessary nor sufficient to get Type 2. Which may mean that there are at least 2 ways to become insulin resistant or that Type 2 is not a single disease. Anyway your point is over simplistic and as such does not support your argument.

    You may have missed this work, but there is now increasing evidence that obesity induced Type 2 is due to fat in the muscles. Not between fascicles but actually inside the muscle fibres, which store both fat and glycogen. When internal lipid stores rise above a certain level the muscles actively refuse to take up glucose. A quick PubMed search shows the molecular mechanisms of this are now under investigation.

    This has several interesting consequences. It says that it is not obesity per se that predisposes to Type 2, it is inactivity combined with obesity that does so. It explains how aerobic exercise both helps to manage Type 2 and reduces your chances of getting it as once you have a reasonable amount of fitness your muscles will burn fat at even moderate levels of activity and also at rest and after exercise. In addition it implies that it might be possible to be overweight and healthy with it, if you are also active. The Americans are doing this experiment increasingly, I have met a number of them online who are overweight and active. They tend to choose non weight bearing sports like cycling and swimming, but numbers of them are working out and not losing weight by eating prodigiously.

    Again Taubes is undoubtedly right that things are more complex than ordinarily presented but that does not mean he has got his science right. Again, Bray shows where he ignores large parts of the literature that blow his hypothesis out of the water. This is not good science, if your idea only stands up if you ignore large amounts of the evidence then you must reject or modify it.

  • Muscleman says:

    “How do you explain GT’s example of the distribution of obesity in the native American popluation both pre and post adjustment to a diet high in refined carbohydrate?”

    Sigh, that was the only change was it? It couldn’t have been that instead of an active hunter-gatherer or small scale slash and burn agriculturalist they became sedentary welfare consumers wracked by alcoholism and disease, surely?

    This is just typical of Taubes’ approach, he picks one thing, carbs and ignores everything else.

    I’ll say it again, we are obsessed by diet because that ‘seems’ like an easy thing to do. Much easier than getting people on the wagon and doing regular sufficient exercise. In addition there is a pile of research money available from the big food producers, especially in the States. Not to mention the vested interests of the big diet promoters. Have you seen the stuff for people on the Atkins diet? That is big business now.

    I am not trying to come over all conspiracy theory here, I’m just trying to explain why and how it is that we are obsessed with the seemingly easy diet route and ignore, wilfully everything else.

    @ Beth@IDBlog

    You cannot write:
    “Re exercise and/or restricting calories to produce weight loss, yes, they do: over the short term.”

    and then cite a study that looks only at diet. This is just eliding the point, yet again. Yes, diet alone rarely produces long term weight loss. But add exercise and keep up the exercise and the weight comes off and stays off.

    Also you cannot counter the meta analysis of diet in the Bray review with a single study. That is equivalent to anecdote. Either show do your own meta analysis with more (validated) studies showing that Bray got it wrong or suck it up.

  • Asclepius says:

    Muscleman – some good points and much I would agree with.

    W.r.t T1 and T2, my observations were generalised and I acknowledged that. Don’t forget that under medication and dietary control, both groups would move their body composition closer to the norm.

    Regarding to changes in the Pima diet – yes there are lots of changes in diet and lifestyle that really should be controlled for. Nevertheless, notice the point about the obese women working ‘much harder’ than the more infrequently obese men. Don’t you think that this is interesting and of significance to any explanation for obesity rooted in a sedentary lifestyle?

    You do seem to push the case on the importance of exercise (increasing calories out) over diet. But this again misses the point that the obese and the thin may well be matching calories in and calories out – but at a cellular level and controlled by appetite and the availability of fat from ones fat stores.

    I think we can both/all agree that there are some very sophisticated mechanisms involved and the research is patchy. GT may not have all the answers but he has rattled the cage – which science needs on ocassion.

    Finally I just want to consider your comment that “this is just typical of Taubes’ approach, he picks one thing, carbs and ignores everything else. “. Have you read his book?

  • anoopbal says:

    Hi Asclepius,

    From your posts,I can see where you are confused: You are asking about the calories in vs out equation and how people manage to balance both sides of the equation.

    1. It’s called the set point theory and we have known this for years. Friedmann discovered the hormone Leptin which seems to mainly controls this feedback mechanism.

    2.This explains why obesity has a big genetic component. obesity is 70- 80% genetics.This was shown in studies using twins. Also the very reason why most of obese people lose weight gain it all back.

    3.If this is what Taube’s saying, this doesn’t prove anything or refute anything. Does it? If Taube says by eating less carbs,you can change your GENES,then I see his point.

    4.And we are in an age here even researchers are supposed to blinded to their experiments so that they don’t knowingly or unkowingly bias the outcome.

    And you basing the whole of your argument based on the “anecdote” a battalion surgeon who saw this in 1800 and said they were “healthy”.

    Anytime when you make ground breaking claims,the problem is that people expect some ground breaking evidence too. And the burden is on you to prove it and not for us to disprove it.

  • Asclepius says:

    Anoopbal – I know that one of the known side effects of insulin on diabetics is weight gain. Would you suggest that insulin makes diabetics increasingly hungry (so they increase calories in), or increasingly lazy (so they reduce calories out)? Next time you see a large city marathon you might also want to look out for the number of ‘larger’ individuals who take part!

    Your point that obesity is 70-80% genetic may well be true but this is way too vague and ignores the specific role of hormonal/molecular components of fat storage. It also fails to explain why obesity in the west has increased in the rates and distribution it has. In fact you could look at what we call ‘middle age spread’. Do you think that this is genetic? Taubes makes no case for gene expression (AFAIK).

    It is best that you pursue your intention to read his book rather than MY (admittedly limited) attempts to try to explain his position. The example of the Pima is simply one that came to mind – GT provides many more.

    I find it odd that a thread that has developed in response to a book review should be so animated by individuals who have not read the book!

    Lest we lose sight of it, let us just reflect upon the last paragraph of DC’s review:

    “It took Taubes five years to write this book, and he has nothing to sell apart from his ideas. No wonder it is so much better than a scientist can produce. Such is the corruption of science by the cult of managerialism that no university would allow you to spend five years on a book [5]. I find all ten points in his summary convincing. But his most important conclusion is that you cannot have any certainty without randomised trials. ”


  • Muscleman says:

    Hi Asclepius
    “But this again misses the point that the obese and the thin may well be matching calories in and calories out ”

    Did you read Bray’s review of the book? If so did you not notice the point that double labelled water studies show that obese people eat more than normal people do? More in fact than their increased metabolism due to the demands of their large frames?

    Taubes says that obese people eat less, based on self reported diet diaries. Yet those have been shown to be inaccurate, as Bray points out and as I as a scientist have known for some time. Diet records are notoriously inaccurate. It is hard enough, surprisingly monitoring how much a lab mouse eats, I know because I tried and gave up. I gave up because I realised the little blighters were playing with their food and weighing their food hoppers daily was therefore a complete waste of time as differing amounts of food ended up on the floor in different cages.

    You can do it, in the very same room another group were in fact doing just that, on one mouse, for a few days in a calorimiter. It would have been theoretically possible for us to do it, but practically with the numbers of mouse we had utterly impractical. I can however tell you that some mice in my care lost, and kept of prodigious amounts of weight under particular conditions. My experience may only be lab based on mice, but the area under discussion is not alien to me.

    In addition I am a prime candidate to agree with Taubes, since carbs are a big problem for me. 20 years ago I was diagnosed by exclusion diet as gluten intolerant. I agree that carbs are a big problem in our diets, but only because we eat and drink them in too highly processed forms that are too high GI.

    Now this again may make me prone to agree with Taubes since it implies a role for insulin. But no, insulin is a culprit only because it lowers blood sugar from a high GI meal so quickly. This efficient lowering, makes us feel hungry, so we eat. The solution is not to follow Atkins* and remove carbs, but to switch to consuming them in the form of fruit, vegetables and more wholegrain forms with much less refined sugar. With a little ingenuity and home baking (I make my own gluten free muesli for eg) along with modern GF flours I can eat this diet easily. Which has two benefits, firstly I don’t need to eat between meals and secondly I have much better control of my calorie needs as a moderately high long distance runner where getting it wrong has noticeable consequences. You can stave off hitting the wall with dextrose gels but only if you also eat right, one without the other does not work and results in a long, slow, painful slog home.

    *Atkins diets are not magic, they are simply low calorie diets like all of the diet alone schemes. It is difficult to eat more under an Atkins regime as fat and protein are very rich foods. However we know that Arctic peoples who eat an equivalent diet, die quite young. The men drop dead from heart attack and stroke in late middle age despite their active lifestyles. But we are doing the experiment and you can bet the epidemiologists are keeping the pathologists on their toes on this one. We might have the answer, in 20 years.

  • csrster says:

    ‘The VERY first line of the paper starts with an assumption ‘Obesity is caused by an imbalance between energy intake and energy expenditure.’ This is exactly the issue Taubes deconstructs. Let us look at the basic equation again:

    Body Mass = Calories In – Calories Out’

    That should be “Change of Body Mass”, not “Body Mass”. One point which I don’t think Taubes addresses but which Nash hints at above is that the “Calories Out” term is actually something like

    Calories Out = Work done + Calories Radiated + Calories Excreted

    I assume that the caloric content of healthy humans’ bodily waste is very small but I have no idea how significant radiated heat is. How much does the body adjust the amount of energy it radiates to maintain homeostasis?

  • Asclepius says:

    It shouldn’t surprise anyone that in most instances the obese eat more than the thin – I mean they are carrying a heavy weight around. This of course does not go any length to contradicting the point that “the obese and the thin may well be matching calories in and calories out”.

    I agree with much of what you say but must again recommend that you read GCBC/TDD. You will see that GT is not simply advaocating ‘Atkins’ and he also has something to say about fructose (and hence why anyone wanting to shed fat might want to limit fruit).

    Cheers for the response.


  • Asclepius says:

    A nice little set of films (around 15 mins in total), that illustrate the ‘common sense’ widom of the cause of obesity in the 1960s:

    This is part one (from where you can see links to parts two and three):

  • anoopbal says:

    For heaven sake, stop making all stupid associations about diabetics, marathon runners and what someone said in 1980’s to make your point about obesity.

    If you have to make a point, pick a a recent study and try to discuss rather than posting you tube vidoes from the 1960’s. Nutrition science has advanced a lot in recent years. I guess that’s news for you.

    And if you have no background or interest in discussing scientific studies, pls try to stay away from discussing these topics.

    Taube started from a stupid assumption of how obese people ate less. The rest of his arguements are bending and ignoring data to fit his hypothesis.

    And if people still freakin go back to how Taube said, it’s all carbs and not calories, please explain.

    1. Why in all these RECENT low carb and high carb studies, the high carb group still LOST weight

    2. Why the low carb group DID NOT LOSE any incredible amount of weight compared to high carb group.

    And please stop going back to the 1960’s and diabetic examples if you don’t have anything to say.

  • Asclepius says:

    Anoopbal – There is no argument that in the short-tem a calorie restricted diet will work. For the body to adjust ‘calories out’ in response will take time.

    As this thread is in response to a review of Taubes’ GCBC then it wholly appropriate that we discuss the book and some of the themes in it. I hope you are not going to go back on your promise to both read and review his book?

    Your attitude is rather patronising and disrespectful. Those who have read GCBC are will be more than mildly amused by the fact that you cannot get Gary Taubes’ name right (it is ‘Taubes’ not ‘Taube’), and have latched on to a hopelessly inaccurate idea that he has started “from a stupid assumption of how obese people ate less”.

  • anoopbal says:

    And I called your associations stupid, not you. And thank you correcting my use of his name.

    1. Can you please be more specific to the questions above.

    2. Can you say why Taubes assumption of obese people ate more is right when doubly labelled water studies shows otherwise?


  • Asclepius says:

    This thread is a response to a book review. Let’s discuss the book. When you have read it, we can discuss it.


  • anoopbal says:

    And discuss what?

    You have the book and your haven’t written anything besides the pima indians in 1800, the diabetics, and how the calorie theory is an assumption. If that’s the best you could get out of the book, I have nothing to discuss.

  • Asclepius says:

    This thread seems to degenerating. It would be much easier to discuss the themes GT introduces if YOU have read the book. You may wish for me to explain GTs position here, but I simply have not the time to paraphrase a 600 page book.

    You have claimed that you “will read his book just to write a review on it.”

    I hope this statement is not hot air. Please do not let us down.

  • anoopbal says:

    I will read the book. And I will most probably write a review if I get the time. But trust me, I don’t have the patience to discuss things with people who have no clue about an evidence- based approach. And are not even bothered to learn something about it either.

    And you don’t have to buy each and every diet book and read it to see is if it makes sense. You just have to take a look at how he reaches his assumptions and conclusions.

    And here is Steve Novella, the founder of the Science-based medicine blog write about low carb diets. He is someone who really knows the science based approach:


  • Asclepius says:

    “I don’t have the patience to discuss things with people who have no clue about an evidence- based approach” – Once again with the disrespectful comments eh?

    We both agree that “you don’t have to buy each and every diet book and read it to see is if it makes sense”. But, if you had read what I have written above you would see that I emphasise that GCBC is NOT a diet book.

    You are attempting to tackle GT’s work without even attempting to understand it. Looking at YouTube or trying to tackle it via proxy, through some of my posts, is frankly never going to cut it.

    Existing research is patchy (despite which, you seem quite sure of your conclusions), and GT recommends further research to ‘directly’ test the carbohydrate-hypothesis (p457 onwards). Again, your ignorance of the book means you are poorly positioned to evaluate such thoughts.

    This is a thread about GCBC/TDD. If you wish to tackle this book and expose weakness in the arguments therein, I suggest once again that you go ahead with your repeated intentions and read it.

  • Beth@IDblog says:

    Whoa! Asclepius, you sure have more patience than I! Anyways, just tuned back in to provide a couple of recent additions relevant to this thread.

    First, Toban Wiebe of Higher Thoughts has just published his notes from a re-read of Taubes’ book. Fans (or haters) may want to download while the publisher allows it:


    Second, Mike Eades (of Protein Power fame) just published a post brutalizing the Sacks et al study that anoopbal mentioned in comment 19.

    First, Eades notes that the diets weren’t really different enough in carb composition (the lowest was still 35% carb, much higher than recommended by VLC folks like the Eades) to determine the usefulness of low carb diets. Second, that said, the researchers noted that HDL goes down as carbs go up. Interesting finding, no?


    BTW, re exercise and weight loss, I think the Time article pretty much says it all:


  • anoopbal says:

    First, nobody is hating anything. I don’t care if it’s low carb or high carb for weight loss. But I don’t see the data supporting low carb diets that’s my only problem.

    Second, it’s not Eades’ criticism. It’s already been commented in the journal and the author already replied to it. Really shows how much of science reading goes with Eades’ time.

    Anyway, this is Sacks reply to the comment about diet not being low carb:

    Boling and colleagues raise
    a valid point. We did not test very-low-carbohydrate intake because we had a concern — based whether extreme or moderate, do not consistently
    result in more weight loss than other approaches. Moreover, our findings confirm that despite best efforts, studies that compare diets for weight loss have not shown large differences in dietary macronutrient composition.

    And Eades & Taube try to pick apart every study which shows no difference in weight loss with high carb & low carb diets. (and there are a lot mind you). And where is their impeccable RCT study proving their point?

    My question is can Taubes or anyone PLEASE show me a meta analysis which even remotely indicates people on a low carb diet were losing incredible amounts of weight or a high carb diets showing people putting on weight and slowly moving towards being obese? If carbs are THE cause, why isn’t none of this showing up in studies??

    Here is a recent review: Weight-reducing diets: are there any differences?
    Nutr Rev. 2009 May;67
    Foreyt JP, Salas-Salvado J, Caballero B, Bulló M, Gifford KD, Bautista I, Serra-Majem L.

    This paper compares the efficacy of two widely used weight-loss diets differing in macronutrient composition – a low-carbohydrate diet versus a low-fat diet. Although “a calorie is a calorie” under the controlled conditions of a metabolic unit (i.e., only the level of calorie intake matters and not the source of calories), we conclude that these interrelationships are far more complex in the free-living situation. The different diet-related factors that condition energy balance, including total energy intake, satiety and hunger sensory triggers, and palatability, must be considered when assessing the efficacy of weight-reducing diets of different macronutrient composition.

  • anoopbal says:

    I thought this was interesting. Taubes answers a question about asians and why they dont get fat with a high carb diet:

    The first one, curiously enough, concerns China: “How do Asians and others living a seemingly high-carb existence manage to escape the consequences?” Taubes’s answer:

    There are several variables we have to consider with any diet/health interaction. Not just the fat content and carb content, but the refinement of the carbs, the fructose content (in HFCS and sucrose primarily) and how long they’ve had to adapt to the refined carbs and sugars in the diet. In the case of Japan, for instance, the bulk of the population consumed brown rice rather than white until only recently, say the last 50 years. White rice is labor intensive and if you’re poor, you’re eating the unrefined rice, at least until machine refining became widely available. The more important issue, though, is the fructose. China, Japan, Korea, until very recently consumed exceedingly little sugar (sucrose). In the 1960s, when Keys was doing the Seven Countries Study and blaming the absence of heart disease in the Japanese on low-fat diets, their sugar consumption, on average, was around 40 pounds a year, or what the Americans and British were eating a century earlier. In the China Study, which is often evoked as refutation of the carb/insulin hypothesis, the Chinese ate virtually no sugar. In fact, sugar consumption wasn’t even measured in the study because it was so low. The full report of the study runs to 800 pages and there are only a couple of mentions of sugar. If I remember correctly (I don’t have my files with me at the moment) it was a few pounds per year. The point is that when researchers look at traditional populations eating their traditional diets — whether in rural China, Japan, the Kitava study in the South Pacific, Africa, etc — and find relatively low levels of heart disease, obesity and diabetes compared to urban/westernized societies, they’re inevitably looking at populations that eat relatively little or no refined carbs and sugar compared to populations that eat a lot. Some of these traditional populations ate high-fat diets (the Inuit, plains Indians, pastoralists like the Masai, the Tokelauans); some ate relatively low-fat diets (agriculturalists like the Hunza, the Japanese, etc.), but the common denominator was the relative absence of sugar and/or refined carbs. So the simplest possible hypothesis to explain the health of these populations is that they don’t eat these particularly poor quality carbohydrates, not that they did or did not eat high fat diets. Now the fact that some of these populations do have relatively high carb diets suggests that it’s the sugar that is the fundamental problem.

    Go to a south Indian home today and you will 70% of diet is from carbs. And white rice is the staple food. We make food from white rice for breakfast, lunch & dinner. I know this for a fact because I was was born and brought up there.

  • anoopbal says:

    I messed up the reply by sacks. I didn’t copy the whole thing. Here is it again:

    Boling and colleagues raise a valid point. We did not test very-low-carbohydrate intake because we had a concern — based on previous studies — that participants would not be able to sustain such a low intake beyond the initial 2 to 4 months, even with continual instruction and assistance from the research staff of dietitians and behavioral psychologists. For example,
    Foster et al. found that after 3 months there
    was no difference in levels of urinary ketones between participants who were assigned to the Atkins diet, which starts with an intake of 20 g or less of carbohydrates per day and increases to an intake of 50 g or less per day (approximately 10% of the daily requirement of kilocalories), and participants
    who were assigned to conventional diets.1 Foster and colleagues opined that long-term adherence may be difficult, and other studies support this view, showing that adherence to carbohydrate goals deteriorates within 2 to 3 months and that
    intake increases to 37% at 6 months2 and to approximately 34 to 40% at 12 months.3-5 The Atkins approach may work well for some persons, but the data from randomized trials provide support
    for the view that low-carbohydrate diets,whether extreme or moderate, do not consistently result in more weight loss than other approaches. Moreover, our findings confirm that despite best efforts, studies that compare diets for weight loss have not shown large differences in dietary macronutrient composition.

  • Bogusman says:

    Well this is all very interesting. I’m sorry to have been away for the last two weeks.

    Firstly, thanks to DC for starting the discussion with the review. I have been looking forward to seeing it ever since you mentioned it was coming over on BadScience.

    I don’t think that as an IT-worker with a thirty year old degree in Physics I can add much weight to the debate but I do have a couple of observations.

    Firstly, Anoopbal, please do read the book. Your assertions about its arguments are almost unrecognisable in the context of what is actually written in its pages. In particular Gary Taubes makes no statement that obese people eat less than others. In fact he often refers to them commonly eating more. The issue is the direction of causality ie do they become obese because they eat too much or do they eat more because they are obese? I am sure that your reference to it as a “Diet Book” is based purely on the title. If you read it you might understand that it is as far from a conventional diet book as I am from being a premier league footballer (and for the avoidance of doubt, that’s a long way indeed).

    Secondly I think that the distillation of this whole argument down to low-carb versus low-cal as some have done here does the book a great disservice. As someone said elsewhere, we need to recognise that it might be a bit more complicated than that.

    As Taubes says in the book and in the Dartmouth lecture, the simple energy imbalance hypothesis would imply that we all maintain a fantastically accurate balance between energy in and energy out in order to remain at a more or less constant weight. Unless of course there are physiological mechanisms at work to do this for us. I have a hypothesis, (based purely on personal experience so I recognise it as being rather weak) that the body adapts over a period of time to any change of regime and gradually reverts to its “set point”. So if my body is 1.9 metres in height and “wants” to be 120 kilos in weight it will get there even if I assiduously follow a strict low-carb or low calorie or high-exercise program. I would love to know if there have been any studies that investigated the effectiveness over time of following one program or another. Clearly one effect is that over time compliance with the program will reduce due to simple boredom / lack of motivation or whatever. I would like to know if there is any evidence for reduced effectiveness even if levels of compliance are maintained.

    Most of all I would like to see some real science done to test the fundamental assumptions around weight regulation. After all, until recently we all knew that stomach ulcers were caused by stress didn’t we?

  • […] Professor David Colquhon’s discussion on Gary Taube’s book “The Diet Delusion&#822… […]

  • […] in a debate I should believe. Occasionally debates crop up that illuminate this effect. Here is one such example. What am I, as an observer, supposed to make of this […]

Leave a Reply

Your email address will not be published. Required fields are marked *


This site uses Akismet to reduce spam. Learn how your comment data is processed.