A new study appeared in the American Journal of Clinical Nutrition today showing that people who ate the most protein, especially animal protein, had the least abdominal obesity.
Danish researchers studied 22,570 woman and 20,126 men aged 50-64 years for five years to determine changes in waist circumference as a function of macronutrient intake. At the start of the study subjects filled out food frequency questionnaires (FFQ) to determine their intake of specific foods. Readers of this blog know I’m critical of FFQ, but in this case the researchers validated the FFQ with 7 day dietary diaries, which, I believe, adds more strength to the study than if FFQ were solely relied upon. Researchers measured the waist circumference of the subjects then rechecked them 5 years later. After correcting for smoking and activity level, the intake of fat, protein, carbohydrate and alcohol was correlated to increase in abdominal circumference.
The findings were interesting.

This study showed that total energy or energy from each of the 3 macronutrient groups–carbohydrates, fat, and alcohol–were not associated with subsequent DWC [difference in waist circumference] in either women or men, whereas an inverse association with protein intake was observed.

An inverse association with protein intake. An inverse association means that the two variables move in opposite direction. In this case, it means that as protein intake went up, waist circumference went down. In other words, those subjects who reported eating the highest amount of protein had the least increase in waist size over the following five years.
The researchers divided the macronutrients into different groups to see if there were any specific sub-groups that appeared to cause or prevent waist size increase. The idea for doing this came from the notion that perhaps there were offsetting effects. Since there was no overall correlation with carbohydrate intake, for example, maybe those subjects who got their carbs from vegetables had a decrease in waist circumference while those who got the same amount of carb from sugar had a comparable increase offsetting the decrease from the former.
As the researchers drilled down into the data that is exactly what they found.

When the macronutrients were divided into subgroups on the basis of the contributing food sources, significantly different associations were seen for some of these groups. There was an inverse association with animal protein and a direct association with vegetable fat.

So, the correlation with animal protein was even stronger than with just protein in general. And there was a direct association with vegetable fat intake, which means that the subjects consuming the most vegetable fat had the largest increases in abdominal size over the next 5 years.

In women, a partition model showed that DWC was inversely related to intake of carbohydrate energy from fruit and vegetables, whereas intake of carbohydrate energy from food sources with simple sugars or from added sugar was positively associated with DWC. A clear and significant positive association with intake of carbohydrates from refined grain and potatoes was observed…

So, there you have it. At least in women, carbs from vegetables and fruits equate to smaller waist; those carbs from potatoes, sugar and refined grains make the belly expand.
An interesting aspect of this study, to me, at least, was the fact that there was NO correlation with an increase or decrease of waist size with caloric intake. It wasn’t the number of calories consumed that caused the change, it was what those calories were made of. So, that makes this another one of the growing list of studies that throws a wrench into the works of the a-calorie-is-a-calorie set. As we all know, it’s not just the caloric content of the food we eat, it is also what those calories do to us metabolically.
To anyone who understands that calories have an effect beyond the amount of energy they provide, this study makes perfect sense. For those who can’t see beyond the energy content, this study is a real head scratcher.
At some point someone in academia is going to realize that the old calories in equals calories out model doesn’t work. I read a paper recently from a researcher at the University of California at San Francisco who is sniffing around the edges, but based on the conclusions he has drawn, the depth of his dumbth is of cavernous proportions. I’m saving his paper for a future post.
The take home message from this Danish study is that if you want to minimize the growth of your abdomen over time, you should eat meat, avoid vegetable oil, eat non-starchy fruits and vegetables and avoid sugar and refined grains. Hmmm, I’ve heard that somewhere else.


  1. Hey, a study of 45,000 people over 5 years and I haven’t heard a thing about it on the news. I wonder why (roll eyes).
    Sometimes, I feel so jaded by all the lies and deceptions, I just throw up my hands and say “let them all die eating that crap.” Less people, less pollution, etc. But if I didn’t care, I wouldn’t get mad. Ok rant over.
    Hi Hellistile–
    Not only is it a study of 45,000 over 5 years, it was published in the world’s most prestigious nutritional journal. With all the hoopla over low-fat diets out there, one would think that this would qualify as a real man-bites-dog news story. I guess those rules don’t hold for nutritional reporting. I can pretty much guarantee that if the study had shown that eating a lot of carbs and no meat correlated with an abdominal decrease, it would have been newsworthy.

  2. Not surprising to many of us. Protein also makes it much easier to stay on plan. I believe our bodies are happiest living on protein, with a good dose of healthy fats to make it tasty!
    Hi Cindy–
    Right on!

  3. Sir vegetable oil…i assume they don’t mean huile d’olive ?
    Hi Simon–
    The study didn’t say specifically, but I’m assuming that it did not mean olive oil.

  4. I am an engineer. Calories are the only thing that matter if you are designing a boiler. In something more complex like the combustion engine in your car, calories are not everything. For instance, ethanol is more efficient, calorie by calorie than gasoline (I live in Brazil, we know those things). Diesel is more efficient than gasoline too. The thing is, there are other factors like compression rate so as to make things more complicated than simple calories.
    The human body is certainly much more sophisticated than a car engine.
    The use of calories by biologists probably started as an useful simplification. Raising it to the altar it is now was certainly a mistake.
    Hi Mauro–
    In another life, I, too, was an engineer. How well I remember dealing with steam tables and adiabatic expansion in my thermo classes.
    I think you’re right about the calorie issue. It probably was used initially as a simplification and took on a new life from there.

  5. The information about the difference between vegetables and fruits vs potatoes, grains, and sugar was interesting. But isn’t it true that if someone ate the same volume of those two categories, they’d be getting fewer carbs in the vegetables and fruits category, and wouldn’t that contribute to the effect?
    Also, is vegetable oil an independent variable, or is it likely that those who used a lot of vegetable oil did so because they were following low-fat guidelines with the predictable adverse results?
    Hi Chuck–
    Good to hear from you.
    The researchers didn’t do a particularly stellar job of explaining how they arrived at the numbers they did. They did report using the Danish equivalent to the USDA food tables, so I’m assuming that total carbohydrates were counted without removing the fiber component, which would, as you point out, skew the results.
    I would bet that you hit the nail on the head about the vegetable oil as well. I haven’t been in Denmark in 30 years so I don’t know what they eat these days, but I would imagine that they’ve fallen for the low-fat fallacy in the same way everyone else has. If so, your observation about the veg oil would indeed hold true.

  6. More fascinating stuff Dr. Eades.
    I had a quick question for you.
    Do we need antioxidants for health and longevity? I’m guessing if you consider cholesterol an antioxidant, the answer would be yes, but other than that, do we in your opinion? In other words, if we don’t need to eat veggies and fruit to be healthy (assuming we instead were eating high quality, pasture-raised animals, including their organs and marrow, with some of it being raw) what’s all the fuss about “phytonutrients” and antioxidants (a fuss I’ve always made as a naturopath)?
    I personally place tremendous value on these sorts of nutrients, and, of course, studies always show better outcomes for people with higher fruit and veggie intake, but in light of the well-known studies of Inuits, and the Maasai, I found myself questioning the common recommendation for “more fruits and veggies”. Maybe we just need them if we eat crap otherwise….
    Your thoughts are eternally appreciated.
    Daniel Chong
    Hi Daniel–
    I think for the most part that antioxidants are overrated. In my opinion as long as one eats plenty of good quality meat one probably doesn’t need a whole lot more.
    Having said that, there are a number of phytonutrients that are probably valuable since most people don’t eat a diet with plenty of good quality meat. And some phytonutrients–circumin is one of my favorites–have therapeutic properties and can be used in place of medications that always seem to come with some side effects.

  7. European food labels don’t include fibre as part of the carb count. Unless the Danes had some reason for adopting the US system I would have thought they’d stick with the normal European convention and that fibre wouldn’t have been included in the carb count. (The UK editions of low carb books such as PP don’t mention this fact, and leave PP’ers subtracting fibre twice until they work it out for themselves.)
    Hi Janet–
    Thanks for the clarification.

  8. Hey Mike! The study is a mouthful if you don’t have all your senses into statistical analysis. Thank you for summarizing the main points.
    I too thought about the fiber not being discounted (or at least not made a separate subgroup within carbohydrate intake). However, it may not be absolutely necessary. Since the inverse association between DWC and intake of fruits/green vegetables is opposite to that of potatoes/refined carbohydrate intake that would at least suggest something. Fruits/green vegetables have more fiber than potatoes/refined carbohydrate/simple sugar-containing foods. So fiber may be skewing the results but toward the message ‘food with more fiber tend to have less carbohydrate that can affect waist circumference’.
    In support to a previous post to this blog entry, when you explain the ‘calorie is a calorie’ concept to an engineer, a physical chemist, or a biochemist, they understand that the concept just doesn’t make any sense at all. I would seem that only nutritionists and dietitians (an physicians for that matter) have bought into that idea. The roots, as explained by Richard Fienman who has published about it, could be traced to the fact that the first law of thermodynamics (or book keeping type of law) is the only flag waived to support the calorie is a calorie idea. Fine if we were closed systems… Like you said once, there is a lot of danger in a little knowledge.
    Unfortunately, people in the Academia who understand this are not always involved in nutrition research. Even those with the potential to understand remain so biased or influenced that just can get passed that we don’t use nutrients in the same way, no matter how much evidence is in front of them.
    I may have a shot at doing this kind of research and hopefully I’ve gathered enough evidence that supports the need for it as its useful and practical outcomes. However, the most difficult part will be to get passed the biases that research committees may have, which normally tend to negatively influence their decision. So… wish me luck!
    Hi Gabe–
    Thanks for the commentary. And good luck!

  9. “the depth of his dumbth is of cavernous proportions.” What a cool line.
    I so enjoy reading your blog, particularly the new research you highlight.
    Hi Carol–
    Thanks for the kind words. I’m glad you enjoy the posts.

  10. I wish They would have done a better job of differentiating the sources of the oils. The implications of this study are very different if the vegetable oils are corn or soybean (esp if hydrogenated), or olive, or coconut or palm.
    Hi Kevin–
    I agree. I would suspect, though, that the oils are corn, soybean, etc. because not that many people in the general population use coconut, palm and/or palm kernel oils. Also, I would imagine the researchers would be aware of the purported health benefits of olive oil and not include it with the vegetable oils.

  11. Hi
    I’m eating less & less carbs, I’ve been doing some IF (although I can’t go for a full 24 hours without feeling crap) I’m often in ketosis (measuring urine) but the weight is not coming off – how long does it take?
    Hi Lynne–
    Based on some of the other comments, weight should start dropping in a few days. And it depends on how much you eat on your eat days. If you double up, weight loss will be slow.

  12. I enjoy traveling to exotic destinations and sampling exotic foods. The result is, I am battling an expanding waistline, something that surprised me and reminded me that I am not as young as before, when I can gulp anything and not put on an inch.
    Trying to reduce my waistline, I discovered your book Protein Power. I have just put myself on a high protein dietary plan, starting yesterday. Wish me luck! One question: is there a difference between what you teach and what is taught by Dr Atkins?
    Hi Tim–
    I think there is a significant difference between our philosophy and that of Dr. Atkins.  Check out the discussion forum on our website and see what all the members there have to say.

  13. I certainly beg to differ with this. I spent most of my life eating a lot of lean meat, especially chicken. I always had that round apple-shaped figure and hated it. Then this January I became a vegetarian. Not only did I lose 30 lbs, but my waist became remarkably smaller in comparison to the rest of my body. And I eat a lot of vegetable fats like all natural peanut butter, olive oil and avocados. So you guys can go ahead and try this “diet” if you like but I’m sticking with what works for me.
    Go for it, but I’ll take good, solid research over anecdotal reports any day.

Leave a Reply

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