Daytime Pattern of Protein Intake

[alan aragon]

^ I'd better eat 60 meals a day to speed things along. Look how your words can get twisted on here.

In this cute little short-term study, 8 meals dun gooft compared to 4, so 60 meals per day would probably turn you into a skeleton in a couple weeks.

[Quelly]

Yes, because the success of any given program (muscle gain or other) is determined by multiple variables that vary in magnitude. The large variables would be the consistency of proper training, recovery, and total nutrition. Each of those major variables has minor sub-variables that comprise it. Meal frequency/distribution is a minor variable within total daily nutrition. Variances in this aspect have a minor impact on the large scheme, as long as the other major variables are on point. Furthermore, why would 4 meals versus 2 meals likely yield a small edge? Because this "edge" is virtually invisible in eucaloric & hypocaloric conditions. There's no plausible reason to believe that everything dramatically changes in favor of a higher frequency during hypercaloric balance (which would actually raise the chances of evoking the refractory response, lawl). The 3 meal minimum for optimization is merely a safe speculation, and it should be re-emphasized that we're groping for a very small edge if it exists at all given all other program variables firing on all cylinders.

Hah, no. It is only revealed in dreams.

To be more concrete with a highly common example, let's imagine I'm wrong about 3 protein feedings being the minimum optimal distribution, and that 6 protein feedings a day increases muscle gain by a whopping 10% more than 3 a day. This means that the 6-a-day guys will reach their muscular potential after 5.4 years of bodybuilding instead of 6 years for the 3-a-day guys. Is that worth the extra meals/effort/pain in the ass for someone not inclined to eating 6 meals a day?

Might seem off topic but its not, but when talking about small differences over the long term and studies trying to discern minor differences, really the issue is the statistics used in analysis. ie something being statistically significant vs clinically significant.
Context is huge here. In a sports performance setting where a .05sec difference in race time determines gold or bronze, or in this case, where minute differences in protein accretion MIGHT add up to be significant over time, statistical significance might not tell us what we need to know.
Studies based on statistical significance often deem results that might be clinically significant in terms of performance as statistically non significant.

For example, in a study you're probably quite familiar with Alan:
Coingestion of carbohydrate with protein does not further augment postexercise muscle protein synthesis by Koopman et al.
http://www.ncbi.nlm.nih.gov/pubmed/17609259

if you look at figure 5 you can see that fractional synthetic rate increases linearly from protein alone, to low carbs + protein, to high carbs + protein, however, the results are deemed statistically non significant and given a P value (vs a magnitude based inference in comparison to a norm that would tell us clinical significance). But really, we don't care if they are statistically significant, we care if those minute changes will make some sort of difference over time in the variable we care about. Based on the statistical model they used, we simply can't know that from this study.

But anyway, I'm not really trying to suggest that THIS study is wrong and that carbs do matter pwo. It's not really even directly measuring muscle growth or anything in an applied manner and the entire length of the intervention is 7 hours, you CAN'T make a long term conclusion based on this study no matter what type of stats you use. However, the conclusion and results might have reported these changes as clinically vs statistically significant with a different analysis...something to think about.

more here if you are interested Alan, and honestly I think this would be a great thing to talk about in the AARR at some point.
http://www.sportsci.org/jour/0103/inbrief.htm (just the first part)

[alan aragon]

Might seem off topic but its not, but when talking about small differences over the long term and studies trying to discern minor differences, really the issue is the statistics used in analysis. ie something being statistically significant vs clinically significant.
Context is huge here. In a sports performance setting where a .05sec difference in race time determines gold or bronze, or in this case, where minute differences in protein accretion MIGHT add up to be significant over time, statistical significance might not tell us what we need to know.
Studies based on statistical significance often deem results that might be clinically significant in terms of performance as statistically non significant.

For example, in a study you're probably quite familiar with Alan:
Coingestion of carbohydrate with protein does not further augment postexercise muscle protein synthesis by Koopman et al.
http://www.ncbi.nlm.nih.gov/pubmed/17609259

if you look at figure 5 you can see that fractional synthetic rate increases linearly from protein alone, to low carbs + protein, to high carbs + protein, however, the results are deemed statistically non significant and given a P value (vs a magnitude based inference in comparison to a norm that would tell us clinical significance). But really, we don't care if they are statistically significant, we care if those minute changes will make some sort of difference over time in the variable we care about. Based on the statistical model they used, we simply can't know that from this study.

But anyway, I'm not really trying to suggest that THIS study is wrong and that carbs do matter pwo. It's not really even directly measuring muscle growth or anything in an applied manner and the entire length of the intervention is 7 hours, you CAN'T make a long term conclusion based on this study no matter what type of stats you use. However, the conclusion and results might have reported these changes as clinically vs statistically significant with a different analysis...something to think about.

more here if you are interested Alan, and honestly I think this would be a great thing to talk about in the AARR at some point.
http://www.sportsci.org/jour/0103/inbrief.htm (just the first part)

Thanks for the link, Eric. I agree that this distinction needs to be made. I'd also add that there's a dual-directional implication with stat significance vs clinical significance (or what I like to call real-world relevance). Sometimes a statistically significant difference is too miniscule to matter in the real world. Conversely, sometimes a lack of statistical significance can cause the hasty dismissal of a small yet important effect on certain goals.

[HDMiBarry]

Well this thread became pretty useful in the end. Thanks to PM for spurring that discussion.

Also effect size ftw.

[greekmanman]

Thanks for the link, Eric. I agree that this distinction needs to be made. I'd also add that there's a dual-directional implication with stat significance vs clinical significance (or what I like to call real-world relevance). Sometimes a statistically significant difference is too miniscule to matter in the real world. Conversely, sometimes a lack of statistical significance can cause the hasty dismissal of a small yet important effect on certain goals.

yea...see studies done on statins where there is statistical significance yet the NNT(number needed to treat) is over 9000 lololol

[PerpetualMotion]

Also effect size ftw.

Statistical power ftw.

[HDMiBarry]

Statistical power ftw.

Lifting power ftw.

So now we've come full circle