Carbs vs. fat: the final answer? [Study review]

A new study has emerged that spells the final verdict on the everlasting carbs vs. fat debate.

 

The study is titled “Calorie for calorie, dietary fat restriction results in more body fat loss than carbohydrate restriction in people with obesity“. As the title suggests, the researchers got a group of obese people to reduce their energy intake by 30% by either cutting out a lot of carbs or fats from the diet. So the protein and calorie intakes of both diets were the same but one group ate a low fat diet and the other group ate a low carb diet.

 

Now as I feared on my Facebook here’s where people are getting confused. Unlike what the title suggests, the study did not actually measure directly that the people in the low fat group lost more fat. In fact, they didn’t. Body composition was measured with DXA scans and there was no significant difference between the groups in the loss of total fat mass or body fat percentage. The statement from the title about the greater fat loss instead comes from the results of the indirect calorimetry measurement from the metabolic ward.

 

Wait, what? A metabolic ward is basically a science lab where oxygen and carbon dioxide consumption and production by your body are measured. The amounts enable us to predict someone’s energy expenditure and the ratio allows us to predict whether fat or carbohydrate was burned (oxidized).

That just leaves protein, but this can be corrected for by measuring nitrogen excretion in the urine (protein contains nitrogen). So when you combine this data, you can estimate how much of all the macronutrients were burned. Which is what the researchers did. And the estimate was that the low fat group lost more fat.

 

Not only was this estimate not supported by actual measurement of fat loss, the diet periods were only 6 days with only 3 measurement days in the metabolic ward. And there were only 17 subjects with reliable data. However, these limitations do not explain the greater fat loss (again, estimated from indirect calorimetry) in the low fat diet group. (See this article about why a low sample size does not negate a study’s findings and may actually strengthen them.) Even or especially in this scenario, indirect calorimetry is more sensitive to detect small changes in fat imbalance than the DXA scan.

 

So how can we explain the greater fat loss in the low fat group? It wasn’t the insulin fairy. It wasn’t because ‘fat is easier to store as fat, bro’. It wasn’t the magical macro ratio. It was good old physics, thermodynamics to be precise.

Many people know the energy balance principle: weight loss is the result of the balance of energy intake from food and energy expenditure.

However, what many people don’t realize is that this equation is dynamic. Energy expenditure changes based on energy intake and as the diet progresses, it also changes due to your body’s metabolic adaptation. But the study authors knew this very well. In fact, one of the study’s researchers Kevin Hall is somewhat of a statistician legend in the scientific community and I reference several of his works in my Bayesian PT Course in the topic on human metabolism. As the researchers put it in this study, “Changes in whole-body metabolic fluxes, thermic effect of food, and body composition generated by isocaloric variations in carbohydrate and fat were responsible for the simulated differences in energy expenditure.”

 

Specifically, energy expenditure decreased more in the low carb group than in the low fat group. Now before you start screaming about how low carb diets kill your thyroid, take a deep breath and let the following fact sink in: the difference in daily energy expenditure between the groups was 48.1 calories. In a group of people with an average weight of 234 pounds (106 kg). And yes, the study included women (*insert big mama joke*).

 

This tiny difference could have been the result of many things.

  • The subjects were performing an hour of incline treadmill walking every day. For these people, that’s basically endurance training. If the carbohydrates helped fuel this training, it could explain the difference in energy expenditure.
  • The subjects maintained their habitual food choices on the diets. It’s safe to conclude their food selection wasn’t the healthiest. Since different fat sources differ greatly in their thermic and metabolic effects (think trans fat vs. coconut oil), this could explain the difference in energy expenditure.
  • The low fat diet was really low fat with 7.7% calories coming from fat. Such a drastic reduction in fat intake could trigger a period of metabolic inefficiency. Considering the short study period and the minor difference in energy expenditure, this is plausible.

 

As for the difference in fat balance, it’s important to note that the low carb diet resulted in a greater difference in carbohydrate loss. That is, the low carb group lost more glycogen, your body’s stored form of carbohydrate. This isn’t a bad thing, as when you switch back to maintenance or a bulk or a higher carb intake, you will get this back (or it will happen anyway as your body adapts) and you will gain less fat than people on a high carb diet. So whether you lose more carbs or fat, it doesn’t really matter. In the end, your actual fat loss will still be determined by the energy deficit you created. Which brings us back to the above points.

 

So before you get on the low fat diet hype train, consider that this study is 1 limited and ambiguous finding. There is a large literature on the topic of carbs vs. fat for weight loss and, if anything, that literature shows that low carb diets are equally effective as or even more effective than low fat diets. Moreover, which diet is most effective for you depends on your personal physiology (‘carb tolerance’), not to mention psychology. There is no such thing as 1 magical macronutrient ratio that works best for everyone and this study certainly isn’t proof that low fat diets reign supreme. On the contrary, it primarily confirms that thermodynamics and not the carb-to-fat ratio of your diet dictate weight loss. This study just affirms the subtle fact that energy balance is dynamic.

10 Comments

  1. Hemming says:

    Good write up Menno! I think you make an unbiased review. To me, the researchers are not claiming that LF should be the default diet for everyone. They wanted to test two diets in a very controlled environment, and that’s what they did. I do agree that the title of the study could have been phrased better.

    The whole point is as you mention. Not everyone needs to be on the exact same diet. I still don’t understand why some people feel that we should all eat exactly the same.

  2. Chris says:

    Hi Menno,

    thx for the article!

    Glycogen: examine.com´s argument is exactly yours: LC group would use less glycogen on the treadmill. Were there any info about the wattage/velocity of the BMI36 participants? With low wattage (i.e. they were simply walking), I guess there wouldnt be much of such an effect.

    Your linked statistical article: http://bretcontreras.com/powerful-stats/
    I hammer down those points of your article every time a statistical layman cries “sample size”. But where they are right (even if the dont know the mechanism) is, that small sample sizes are bad: theyre prone to sampling errors that lead to type 1 errors. This may be less the case with intervention studies where you know about a host of influencing variables and can get baseline measures for them to ensure good randomization. Other than that, large sample sizes are always better. If you can interpret the results in terms of effect sizes and not just look at significance.
    Just trying to caution not to make ppl believe small sample sizes are good, as I noticed this tendency with myself when trying to clear up that common misunderstanding about small sample sizes.

    • There wasn’t much information about the intensity or wattage, no, but given how out of shape the participants were, an hour of anything is likely to use a significant amount of carbs.

      I agree completely about the sample size argument. Low sample sizes definitely aren’t a virtue, but they’re frequently heralded as an argument to invalidate a study without merit.

  3. Dillon says:

    In summation, overall energy balance still trumps individual macronutrient ratios. Long-term adherence from the dieter is also rather important, lol. Good read, Menno!

  4. Alistair says:

    Hi Menno – another quality well written & balanced article.

    Good to see Cybernetic Fitness up also.

    I have heard Børge Fagerli mention on at least one podcast in recent months that the majority of one’s fat intake should be eaten during the day, and that the more fat eaten in evening is unfavorably oxidised? Does this align with your thoughts and should we really be making a conscious effort to eat the majority of fat during the day and focusing on lean protein & carbs in eve?

    As always, I appreciate context is everything and we do not all need to eat the same diets. However, I am keen to understand your perspective on this given your focus on nutrient timing.

    Best regards,
    Alistair

  5. Matthew Niedbala says:

    Except we know for a fact that humans followed a more plant-based diet 6000 years ago. This diet was just plants and bugs and no we weren’t nutavores. What is natural is almost always more healthy and the fact that foods that contain more carbs tent to be much more healthy. Fact is: you can go plant based and be healthy and lose weight, but a low carb diet makes you stupid, unhealthy, and often ends in deficiency, cancer, heart disease, and horrible digestion that sometimes causes “ballooning” of the intestines as a low carb diet has little fiber and is unhealthy. Anyways if you cut out the sugar (for the most part) and increase the fiber it rarely ever makes fat on a 2500 calorie diet.

Leave a Comment