T Nation

0.82g of Protein?


#1

Has anyone seen the article on Bayesian Bodybuilding arguing that you only need 0.82g of protein per pound of lean body mass ?.
It cites studies done on both natural beginner lifters and experienced and elite lifters (guessing enhanced) and claims increases in protein for both natural and enhanced lifters showed no positive effect. This number has been rounded up to be safe too, apparently.

Any opinions on this? Seems quite a rational breakdown.


#2

0.82 is a rounded number? Cool.

Wait a month or two and another study will come out with different number. As no two bodies are the same, just do what is best for you. Also, look at the kind if subjects they use. Also look at the target audience. Etc, etc.


#3

[quote]JFG wrote:
0.82 is a rounded number? Cool.

Wait a month or two and another study will come out with different number. As no two bodies are the same, just do what is best for you. Also, look at the kind if subjects they use. Also look at the target audience. Etc, etc. [/quote]

When I say rounded up I meant this (from the article):

Based on the sound research, many review papers have concluded 0.82g/lb is the upper limit at which protein intake benefits body composition (Phillips & Van Loon, 2011). This recommendation often includes a double 95% confidence level, meaning they took the highest mean intake at which benefits were still observed and then added two standard deviations to that level to make absolutely sure all possible benefits from additional protein intake are utilized. As such, this is already overdoing it and consuming 1g/lb â??to be safeâ?? doesnâ??t make any sense. 0.82g/lb is already very safe.

The picture below summarizes the literature. As you can see, 1.8g/kg (0.82g/lb) is the point at which additional protein intake ceases to yield any benefits.


#4

[quote]JFG wrote:
For sedentary people, sure.

look, it’s been going around for a long time. At the end of the day, eat enough to grow. And stick around, another study will be out soon enough with another number. It’s the cycle Of the study.
[/quote]
Damn phone…


#5

[quote]JFG wrote:

[quote]JFG wrote:
For sedentary people, sure.

look, it’s been going around for a long time. At the end of the day, eat enough to grow. And stick around, another study will be out soon enough with another number. It’s the cycle Of the study.
[/quote]
[/quote]

That is gram per kilogram. Not per lb. The studies in the article the mods removed seemed pretty set in stone. They found no benefit in exceeding 0.82 in both beginner and experienced lifters.


#6

Not finding an effect is different from finding, reliably, that it does not exist.

Yes, some authors would compare for example two training methods, one of which is good and the other poor, and find no “statistically significant” difference between the methods. Because there was so much variation between subjects.

Unfortunately, many such authors would write that they had found there was no difference.

Studies that fail to find a difference should not be taken as evidence of there being no difference, until looking at how powerful the study was at resolving differences.

As an example of what I’m saying, let’s take a look at your first reference. It says,

"Subjects in AL [higher protein intake] experienced a 22% and 42% greater change in 1-RM squat and 1-RM bench press than subjects in RL [moderate protein intake], however these differences were not significant. "

Measuring 22% and 42% more improvement, but then discovering that due to high variation among subjects and the limited power of your study your data is not statistically significant, is a heck of a lot different than finding there is no improvement!

I hope the distinction is clear? (It’s often confusing.)

Yet a further point relevant to many other studies in this category is that it’s not necessarily the case that muscle synthesis rate post-exercise tells all. Why should it tell all, and what’s even a trace of evidence that it does?

I’ve read a lot of studies of these sorts over the years, but generally don’t remember titles or authors and so cannot say just by glancing at your long list of references.

Rather than have anyone need to track down and examine every single one of them, could you pick the one best shot in there of proving the point of no benefit past this number? Then we can look at that one closely.


#7

Thanks Mr Roberts. Better said then I ever could.

I have just seen this before so many times and it has never been conclusive.

(OP) At then end of the day, do what is better for you. Give it a try at .82 and report back in 6 months. I would be interested in your findings. I prefer to err on the side of caution and go higher then that.


#8

I googled the source article. The authors seem to have settled on a value of 0.82 g/lb/d–not 0.82 g/pound of LBM/d as stated in the OP. In other words, the value falling out of their meta-analysis (0.82 g/lb/d) dovetails pretty well with the broscience standard of 1 g/lb/d. Score one for the bros.


#9

Nah Bro’s are on that 1.5g per lb, just to be safe so they don’t go catabolic when they walk Brah.


#10

[quote]JFG wrote:
Thanks Mr Roberts. Better said then I ever could.

I have just seen this before so many times and it has never been conclusive.

(OP) At then end of the day, do what is better for you. Give it a try at .82 and report back in 6 months. I would be interested in your findings. I prefer to err on the side of caution and go higher then that. [/quote]

I will do. All aboard the experimental gain train!


#11

[quote]EyeDentist wrote:
I googled the source article. The authors seem to have settled on a value of 0.82 g/lb/d–not 0.82 g/pound of LBM/d as stated in the OP. In other words, the value falling out of their meta-analysis (0.82 g/lb/d) dovetails pretty well with the broscience standard of 1 g/lb/d. Score one for the bros.[/quote]

Here is the “take home message”


#12

I’ve done a significant amount of reading trying to figure this out. What I finally realized is that protein intake has two main functions:

  1. Provide substrate for muscle protein synthesis (building blocks)

  2. Provide activation of mTOR via Leucine which turns on muscle protein synthesis (signalling)

From a substrate perspective 1.8g/kg of body weight is indeed the correct amount for natural trainees. (Those on assistance can build more muscle and thus need more substrate)

Most proteins contain roughly 8% Leucine, and maximum mTOR signaling occurs at approximately 0.05g/kg of bodyweight. This signalling is on a per meal basis. The longer between meals the stronger the signal but obviously if you eat less times per day this means less signalling opportunities. The optimum signalling for most people is then going to be at roughly 4 times per day with each meal at least four hours apart. Three or five meals are acceptable but not quite as good from this perspective.

Ok, so if we have a 200lb trainee then his substrate requirement is roughly 164g of protein. On the other hand, to get maximum signalling he’s going to have to eat four meals each with 57g of protein, for a total daily intake of 228g of protein. The standard bodybuilding recommendation would be 200g which you can see is either more than necessary from a substrate perspective or insufficient from a signalling perspective.


#13

Protein is critical but overrated by most people. I don’t believe you need nearly as much as a lot of people recommend. There really isn’t all that much protein even in a pound of muscle. If I?m a really serious natural trainer and have put on 25 pounds of muscle, that’s like 2500 actual grams of protein over a lifetime of training. I don’t think building of muscle is all that demanding on protein stores. Mostly it’s going to repair and rebuild, which I’m sure is something, but it isn’t 200+ grams (2+ pounds of muscle tissue) worth a day. I think most people, even athletes, are probably fine with as little as .5 g/lbs as long as other macros are meeting their energy needs.

That said, eating more as insurance isn’t going to kill you.


#14

[quote]DoubleDuce wrote:
Protein is critical but overrated by most people. I don’t believe you need nearly as much as a lot of people recommend. There really isn’t all that much protein even in a pound of muscle. If I?m a really serious natural trainer and have put on 25 pounds of muscle, that?s like 2500 actual grams of protein over a lifetime of training. I don?t think building of muscle is all that demanding on protein stores. Mostly it?s going to repair and rebuild, which I?m sure is something, but it isn?t 200+ grams (2+ pounds of muscle tissue) worth a day. I think most people, even athletes, are probably fine with as little as .5 g/lbs as long as other macros are meeting their energy needs.

That said, eating more as insurance isn?t going to kill you.
[/quote]

Never a truer word spoken


#15

[quote]YamatoDamashii92 wrote:
Has anyone seen the article on Bayesian Bodybuilding arguing that you only need 0.82g of protein per pound of lean body mass ?.
It cites studies done on both natural beginner lifters and experienced and elite lifters (guessing enhanced) and claims increases in protein for both natural and enhanced lifters showed no positive effect. This number has been rounded up to be safe too, apparently.

Any opinions on this? Seems quite a rational breakdown. [/quote]

I agree with the estimate for the most part.

Take a 200 pound athlete.

I have seen many studies that show that .7 grams per pound (140) will keep 90% of athletes in nitrogen balance, and .85 (170) will maximize protein synthesis and that all protein consumed above this amount can be demonstrated to yield ammonia. I have found only 1 study that shows that more than .85 produces any extra muscle gain over a long period and the specific study used .85 and a whopping 1.7 grams per pound (170 and 340 hypotheticals), and saw a small but significant increase with 1.7, but there is no reason for me to believe that is would not have happened at a much lower number like .9, .95. 1.0 etc.

The plausible reasons that the extra 170 grams made a small difference (even though the extra 170 grams all netted ammonia) is that it stimulated insulin secretion and added calories, but those using 340 grams were actually more insulin resistant at the end of the 6 month period. I can’t find it, but it was a Russian study. Other than that, I can find nothing that shows more protein synthesis or muscle gained by more than .85 grams per pound of bodyweight, except you get DoggCrapp guys claiming that you need 300 grams if you are not growing. Again since protein stimulates insulin (and will use available insulin to enter cells) it doesn’t sound like a good long term strategy to me.


#16

I’ve never really understood the persuasiveness of the nitrogen excretion, etc, arguments.

For example, Ellington Darden is a very smart guy, and related his experience with this. While an exercise science student, he believed in the bodybuilding value of high protein, but at least one of his professors did not, and wanted to perform a nitrogen excretion experiment with him.

Darden agreed, and to his (sophomoric) shock, found that the extra protein he was eating was being excreted, or rather metabolites were. This should not have been a surprise, however.

Somehow he thought, then anyway, that this proved that the extra protein was useless.

Well, if there’s about 90 g of protein in a pound of muscle, did anyone think a pound of muscle per day was going to be added by consuming say 240 g per day versus 150 g? No, that was never the question.

Proving that that doesn’t occur, that instead the great, by far, majority of that is excreted as metabolites, does absolutely nothing towards saying whether improved body composition and muscle mass over say a year’s time, does or does not occur.


#17

Looking back at some of my notes, I was a little off.

Again with a 200 pound athlete

.7 or 140 will put 90% into nitrogen balance. Keep in mind that this is fine for competition, but not for anabolism.

Protein synthesis does not go up above about 125% of the amount that put athletes into nitrogen balance. This would be about the .85 level.

However there are really 4 caveats to this.

  1. it is possible/likely that athletes who train for strength and muscularity fall above the 90% who only need about .7 grams for nitrogen balance.

  2. it is possible that weight trained athletes are in the upper level of the population of athletes for protein synthesis signalling.

If we use the 90% mark for the first value and assume that strength athletes will be closer to the 1% mark on a normal curve we basically DOUBLE the z-value. This would mean that if the curve is normal, and 90% of athletes are in nitrogen balance with .7 grams per pound, that 99% of athletes will be in balance with double that, or about 1.4 grams per pound for nitrogen balance and 1.7 for maximum protein synthesis. This matches the Russian study that showed elite athletes could retain about 3 kilograms more after 6 month on 1.7 grams per pound than .85 grams per pound. So 90% of 200 pound athletes may not benefit from more than .85, but 10% will, and 1% will benefit from going up to double that mark, and it probably has to do with holding on to more muscle and training harder.

  1. BCAAs can reduce cortisol and stress hormone induced autophagy. In other words they prevent CATABOLISM even if they don’t raise protein synthesis.

  2. Protein and carbs together have a greater insulin stimulating effect. This suggests that some protein with carbs will increase muscle glycogen stores.


#18

I believe some of some of the posts above is way over-thinking, by which I mean, taking the precision of thinking far beyond the precision of the evidence used for the thinking.

Is a single one of the studies involved sensitive enough to tell the difference between a diet that with the same training, etc, yield say 3 lb better condition after many months than the comparison diet?

I doubt it (of course, anyone who has the study whose statistics show that it could resolve this instead of finding it “non-significant,” that would be excellent to post.)

But yet guys who train regularly and have done so for years will notice and value such differences.

I’m not saying that the above, super exact 0.82 grams figure is a bad dietary figure.

I am saying that I think that to argue (if anyone is) “This exact number and not 1 gram per lb” is reading way too much precision into the findings of these studies…

Remember, the first-cited one measured a 42% increased improvement in bench press with the higher protein diet, but the study wasn’t sensitive enough for that to be statistically significant! Geez, wouldn’t you like that much improvement? It might well have been real, but the study didn’t have the power to establish that a 42% improvement, if real, was real. So it gets called nothing.

How can an exact conclusion be arrived at from things like this? It cannot, or at least not validly so.

Another study looked at recently on the forums here, comparing pea protein to whey, was a good example. The sensitivity of the study was low enough that it was unable to tell the difference between administering whey protein (for people on a low protein diet) vs nothing. So if the study can’t tell that, how can it be thought able to say with any accuracy that pea is as good or better than whey? Common sense please! (Or some care with the statistics and the correct meaning, which is not what people commonly attribute!)

Various failures to find are not the same as proving non-existence.

I would much rather people try for themselves: it won’t mean much if in a condition where further gains come easily no matter what, but if training well for years and being in a slow-gains sense, you’re a much better laboratory than these studies that measure 42% better improvement but the studies (not the substance tested, but the studies) fail to be statistically significant.

I suggest, every time where a paper says “was not statistically significant” insert “Our study.”

The more accurate version now reads, regarding the point in question, “Our study was not statistically significant.”

Now was the measured effect real, caused by the treatment? We don’t know from the study, but that’s all. The non-significance of their study is NOT evidence as to whether the effect is real or not.


#19

Research measuring protein synthesis has many problems too. Lifetime trends are hard to establish with short term instantaneous measurements.

The other thing people fail to realize when looking at studies is that sometimes the average outcome is meaningless, even though it’s what many people focus on. Many studies show large variation among individuals. So, while one study may show a very sound average 10% increase in muscle mass (or whatever) with X diet, it could still be a terrible diet for you. It could be that 90% of people gained 20% muscle mass and 10% of people LOST 50% muscle mass. And if you are one of those 10%, following this “scientifically proven” diet might be horrible for you.

Or conversely, A diet might give 50% of people a 50% increase and 50% a 50% decrease. Then a study could show an average of 0% effect of the diet. BUT for half of people it could be a really good way to go.

Science can give you things to try, but you should only keep the things that work.


#20

Agreed completely, could not have said it better. Very key points.