All Diets Are Reduced Carbohydrate Diets

I just finished reading Gary Taubes' second blog post entitled "Calories, fat or carbohydrates? Why diets work (when they do)."   Distilled, he pointed out that when you reduce calories, you automatically reduce carbohydrate, which then begs the question, does a reduced calorie diet work primarily because of reduced energy intake, or because of reduced carbohydrate intake?


Gary gives some illustrations of this principle, and I want to expand upon them.  If you've been eating 2500 calories daily with the typical 35% as fat, 15% as protein, and 50% as carbohydrate, you've been getting 875 calories from 97g of fat, 375 calories from 94g of protein, and 1250 calories from 313g of carbohydrate.  


If in seeking weight loss you reduce calories by 500 and fat to the widely recommended 30% of calories, while keeping protein at 15% of calories, your "diet" will now consist of 2000 total calories, with 600 calories from 67g of fat, 300 calories from 75g of protein, and 1100 calories from 275g of carbohydrate.  


This "diet" has reduced fat by 275 calories, protein by 75 calories, and carbohydrate by 150 calories.  Thus, this diet is a reduced calorie and reduced fat diet, but it is also a reduced protein, reduced carbohydrate diet.

 Now, if you do lose body fat, how do you know what caused it?  Was it the reduction of total calories, the reduction of fat calories, the reduction of protein intake, or the reduction of carbohydrate intake?  

Many people will assume that it is the reduced caloric intake that produced the change, or that the reduction of fat intake is "most" responsible since the calorie intake from fat was the most reduced.


In fact, from this experiment, there is no way that you can tell which of these reductions is actually responsible for the reduction of your fat mass, because all of these things have changed.

Researchers continue to act as if reduced calorie diets are NOT reduced carbohydrate diets, when in fact all reduced calorie diets are also reduced carbohydrate diets.  

Taubes also discusses a study published by Shai et al in the New England Journal of Medicine: "Weight Loss with a Low-Carbohydrate, Mediterranean, or Low-Fat Diet [full text]."  He posted this table from the article:

As Gary says, we have to take this data with a shaker of salt since it is from diet records that are notoriously misleading.  Let's set aside the fact that the data collected probably gives us a poor picture of the actual dietary practices of the groups.  If you study it, you find that the so-called "low carbohydrate diet" group reduced their intake of carbohydrate from 50% of energy to 40% of energy over 24 months.   Meanwhile subjects in both the "low fat diet" and the "Mediterranean diet" groups got 50% of energy from carbohydrate.  

This means that the "low carbohydrate" dieters consumed 80% of the amount of carbohydrate as the "low fat" and "Mediterranean" dieters.  Assuming an intake of 2000 calories, the "low carb" group took an average of 800 calories/200g of carbohydrate daily, while the "low fat" and Mediterranean" groups took an average of 1000 calories/250g of carbohydrate daily.  


I don't know anyone in the "low carbohydrate" diet camp who would consider 200g of carbohydrate daily a low carbohydrate diet.  I don't consider a diet low in carbohydrate unless it contains no more than 25% of calories from carbohydrate.  Of interest here, the authors apparently consider 30% of energy from fat to be "low fat," while 40% of energy is "low carbohydrate." 

So, if you are interested in producing a study that shows little or no difference between low carbohydrate diets and low calorie diets for fat loss, all you have to do is to allow the "low carbohydrate" group to eat a diet high in carbohydrates, then call what they did "low carbohydrate" when you write up the results.

Despite this quite obvious lack of rigor, which by the way got past all the peer review process (illustrating what that is worth), the study actually did show that the people eating the "low carbohydrate" diet had the greatest fat loss over 24 months.  According to the authors, "of the 272 participants who completed the intervention, the mean weight losses were 3.3 kg, 4.6 kg, and 5.5 kg" respectively for the "low fat," "Mediterranean," and "low carbohydrate" diets.

Think this through.  An average reduction of only 50g of carbohydrate daily (about 3 slices of bread, two 4-oz potatoes, or 2 pieces of fruit) gave the "low carbohydrate" group a weight loss 67% greater than the low fat group, and 20% greater than the "Mediterranean" group. 

Meanwhile, the "low carbohydrate" group ate more fat than the other groups.  The "low carbohydrate" is reported to have gotten 39% of energy from fat, the Mediterranean 33%, and the "low fat" 30%.  That would translate to 780 calories from 87g fat for a 2000 calorie intake for the "low carbohydrate" group, 660 calories from 73g fat for the "Mediterranean" group, and 600 calories from 67g fat for the "low fat" group.  This data clearly suggests that the more fat you eat, the more fat you will lose; completely opposite to the "low fat" mantra.  While eating 20g more fat per day, the "low carbohydrate" diet group lost an average of 67% more body fat than the "low fat" group. 

In reality, this study compared three different high carbohydrate diets, each with a low fat intake, and found that those who ate the most fat and least carbohydrate had the greatest reduction of body fat.  Yet the popular press presented this as if it doesn't matter whether you reduce fat or carbohydrate, so long as you reduce caloric intake.

Bad science.  As usual.