Wednesday, January 18, 2012

Bray On The Loose


The article linked below examines a study of what happens if you feed folks 1000 kcal/day more than you think they need.  To test the question, the participants were combined to metabolic ward.   The results and study design itself are quite interesting.

-Exercise was not part of the study, and their physical activity was controlled
-The low protein variant result in lean muscle mass loss and fat gain; but the total weight gained was less than that of the other variant
-The carbs levels were kept the same, it was protein levels and fat levels that varied

The author didn't seem to understand the study and essentially parroted the party line:
Whether you are just starting a New Year's diet or struggling to maintain a healthy weight, a provocative new study offers some timely guidance. It isn't so much what you eat, the study suggests, but how much you eat that counts when it comes to accumulating body fat.

"That's a very important message," said Francisco Lopez-Jimenez, an obesity researcher at Mayo Clinic, Rochester, Minn., who wasn't involved with the study. "Weight gain depends primarily on excess calories, regardless of the composition of the meal."

These are funny conclusions to reach when the study shows that calories do not dictate weight gain – the groups all ate the same excess calories, but the low protein group gained as much fat while losing muscle mass. 

The study author is well known – some might say notorious – for his advocacy of the concept that might be know as “a calorie is a calorie.” 
Dr. Bray said the more than 50,000 extra calories were roughly equivalent to the excess calories the average American consumes over a decade.
And he is of course right, as regards oxidation of macronutrients in a bomb calorimeter.  However, his own test shows that calories are not calories when we’re talking about a human subject:
After eight weeks, all participants in the study gained weight. The 16 men and nine women made similar gains. The low protein-diet group gained about seven pounds, about half the 13.3 pounds added on by the normal protein participants and 14.4 pounds put on by the high protein group.
In other words, sometimes the body takes extra protein calories and makes (or sustains) muscles from them, and sometimes, the calories become fat. 

This is a great study for illustrating how truly difficult it is to do good science on humans.  For example, one of the common observations about folks that are leaner is that they are more active in unconscious ways than heavier folks are.  Their bodies adapt to higher food intakes by expending more energy as activity.  Likewise, folks that are starving tend to be lethargic and to feel cold, as if their internal feedback systems find ways to conserve.  In short, there are plenty of reasons to believe that the body is “not a closed system” and therefore, simplistic application of the first law of thermodynamics to humans is problematic. 

One very important element that was not included in the write up from the article would have been an assessment of what the subjects’ fasting lipid profiles were like.  It would also be interesting to see what would happen if the carb levels were not held constant – cause I’ll bet you 1000 kcal per day extra of a 60% carb diet would indeed have had a measurably greater fat gaining effect.  But that’s part of the problem; if you can’t hold constants constant, the study cannot determine causality.  Of course, there’s the other problem – if you hold constants constant, you are not examining how the human body responds “in the wild.”


Luckily, you don’t have to subject yourself to the vagaries of the scientific method with a bunch of other folks being force fed high calorie diets, with their activity constrained, and their calories counted.  You can just eat meat, vegetables, nuts and seeds, little fruit or starch, and no sugar/wheat, and prove for yourself whether or not it is good for you. 

PS - Here's Fat Head's review: Another Bad Study, Badly Reported

No comments:

Post a Comment