Hey, ass-hat with the abstract link. Yeah. You.

[Edit: More than a few readers gave me feedback about my use of a derivative of the r-word, which has gone from a medical term still used in medical journals to a pejorative one when used in a derisive context. A small portion of my patients are developmentally delayed and understanding the power of language (being more than moderately opposed to, “That’s so gay,” referring to anything more offensive than a bus of drag-nuns–which really isn’t offensive at all since the Sisters are AWESOME (honestly, I’ve been to their parties); so I guess”gay”=”awesome”), I’ve replaced it with “Ass-hat”. I have Googled its etymology and can find no reason not to use it. Now, onto the post.]

The rate at which psuedo-information flies around has now reached epic proportions. And not in a good way.

Listen folks. Yeah, I know science is great. Broscience bad. PubMed will save us all. STOP READING ABSTRACTS. Bored at work? LetterPress is a great game to take up. Need a tweet to keep your Klout score high? Post a shirtless photo. (No really, they get more retweets and hits. If you’re hot, I promise to re-tweet it with gratitude.) If you must read them, STOP pretending you know anything about the study unless you actually went and read it. Don’t have time? Don’t have access? TL;DR? Quit eff’ing passing off the abstract link and your dumb-ass conclusion as though you did.

Remember the Green Lantern movie? Remember how amazing that trailer was? Wasn’t it great? Remember seeing the Green Lantern movie? In 3D? Yeah. Shirtless Ryan Reynolds couldn’t save it. And that’s saying a lot.

Soenen S et al. Normal protein intake is required for body weight loss and weight maintenance and elevated protein intake for additional preservation of resting energy expenditure and fat free mass. Journal of Nutrition, DOI: 10.3945/jn.112.167593, 2013.

Here’s a link to the touted abstract. http://www.ncbi.nlm.nih.gov/pubmed/23446962

This is like the trailer to the Green Lantern movie.

What actually happened?

Seventy-two people with BMI>25 were randomly put on a “normal” protein diet, vs a “high” protein diet with calorie restriction. During the study time, protein was provided to the subjects in the form of meal replacements. The authors do not state if the subjects were able to eat protein from other sources or not (but it sounds like the subjects were NOT allowed to eat other sources of protein until the last two months of the study from the sentence, “…During the last 2 mo, different protein sources were also allowed to smooth the transition to a diet possible to maintain during habitual daily life…”)

There was no structured physical activity component to this study. Subjects were told to do what they normally do.

For the first 2 weeks, subjects ate at their calculated daily caloric requirement level. Then, for the following 6 weeks, they ate at 33% of their calculated daily caloric requirement level with the same protein level as they had been assigned (this was described as the “weight-loss” phase.) This was then followed by “weight-maintenance” phase for 17 weeks, which was 67% of the ORIGINAL calculated caloric requirement with the same amount of protein. The original calculated caloric requirement was done using the Harris-Benedict equation with a 1.5 activity modifier.

To those of you who have bizzare ideas of what “normal” protein and “high” protein actually IS (most likely something like 1g per POUND of bodyweight), this study defined it as 0.8g per KILOGRAM of body weight for “normal” and 1.2g per KILOGRAM of body weight for “high”.

The subjects were measured pre-diet, after the weight-loss phase and then after the weight-maintenance phase. I’m not going to go into everything that was measured because the main question here is fat-free mass.

Fat-free mass was essentially measured using a fairly old method of calculating total body water content (which involves radiolabelled water and equations for dilution of said water) and dividing total body water by 0.73. I’m also not going to get into the niceities of this method. For argument’s sake, let’s just assume this is reliable and accurate because if it isn’t, then the whole study is useless (and therefore linking it is mind pollution), and if it is, it still doesn’t make a difference in my final conclusion. Resting energy expenditure was also measured in a fairly standard fashion (sampling for 30 minutes.)


The normal protein group weighed, on average 90kg (SD 14), and the high protein group weighed, on average 90kg (SD 14) at the beginning of the study. Fat-free mass was 54kg (SD 9.8) for the normal group, and 54kg (SD 9.6) in the high protein group. Resting energy expenditure (REE) was 7.02MJ/d (SD 1.1) for the normal group and 7.02MJ/d (SD 1.0) in the high protein group (7.02MJ/day is about 1670 calories).

After the weight-loss phase, the normal protein group had lost an average of 1.3kg (SD 0.2kg) of fat-free mass, while the high protein group lost 0.6kg (SD 0.2kg) of fat-free mass. That’s a difference between the two groups of 0.7kg, or 1.5 pounds. This was a statistically significant difference. In total, the normal protein group went from 90kg (SD 14kg) to 84.1kg (SD 12kg), while the high protein group went from 90kg (SD 14kg) to 85kg (SD 13kg) over 6 weeks. That’s about a 12-13 pound total weight loss over 6 weeks.

MEASURED REE dropped in both groups after the weight-loss phase: 6.76MJ/d (SD 0.9) for the normal protein group, and 6.74MJ/d (SD 0.9) in the high protein group. CALCULATED REE was 6.86MJ/d in the normal protein group, and 6.94MJ/d in the high protein group.

After the weight-maintenance phase, the normal protein group had a fat-free mass of 53.2kg (SD 8.3) and the high protein group had a fat-free mass of 53.9kg (SD 8.0). Measured REE was 6.48MJ/d (SD 0.9) in the normal group and 6.82MJ/d (SD 0.09) in the high protei group. The authors failed to detect a statistically significant difference between the two groups in this phase for both fat-free mass, but did find a statistically significant difference between the DROPS in REE between the two groups (normal protein -0.5MJ/d (SD 0.02), high protein -0.2MJ/d (SD 0.01). ) That’s a difference of 0.3MJ/d on average, which translates to 72 calories.

There’s a whole other section of how REE was then expressed as a function of fat-free mass by regression analysis, and how the linear regression curves were different for the high protein and normal protein groups.


The abstract states at the end that, “A NPD of 0.8g/kg BW/day is sufficient for BW management, where as a HPD of 1.2g/kg BW/day is necessary for preservation of REE and a stronger initial sparing effect of FFM and lowering of DBP.” BW is body weight, FFM is fat-free mass, DBP is diastolic blood pressure (which I’m not getting into). NPD is normal protein diet, and HPD is high protein diet.

There are three main points I want to make here:

1) For most of you reading this, 1.2g of protein per kilogram of body mass makes you want to cry. That’s actually 0.54g of protein per pound of body weight. That’s the definition of high protein in this study. Not YOUR definition of “high protein”. Not bodybuilding.com’s definition of “high protein”. So before you start spouting off that you need a “high protein diet” to preserve muscle mass on a weight-loss diet because this study said so, keep in mind what “high protein diet” really means here. Or even “sufficient protein diet”. Because that’s 0.36g of protein per pound of bodyweight in this study.

2) The subjects in this study had their daily caloric intake cut not BY one-third, but TO one-third in the weight-loss part of this study. On average, that’s 900 calories per day. Just let that sink in a bit.

3) Points 1 and 2 aside, the challenge I see here is taking the “statistically significant” and making it “important”. We CAN definitively say that the average loss of fat-free mass was less in the high-protein group than then normal-protein group. That’s not a coincidence. However, I would challenge you to tell me that a 1.5 pound difference in fat-free mass loss has any actual importance over 6 weeks. We can also say that the drop in REE was different in the normal protein group compared to the high protein group. By 72 calories. Per DAY. How eff’ing protective is that?

What happens after 6 weeks? Does that difference get bigger? Smaller? This study doesn’t answer that question. For all we know, it’s linear, and the 1.5 pound difference and 72 calories burned per day just stays constant and consistently irrelevant. Or maybe you die. We just don’t know.

The Bottom Line: Unless you have actually READ the full study and can interpret the numbers, stop polluting other people’s minds. Seriously, it’s like mental second-hand smoke.

Click Here to view the Full Version of our Website