76 replies to this topic

[brianmdelaney]

Michael,

First, thanks for the amazingly useful post. I'd been lazy about my supp. regimen, and looking at yours (which I've down far too cursorily in the past) has proved to be the best way for me to review my own.

I decided to start taking creatine (because of the likely benefit to cognition in vegetarians), and I'm wondering about the volume to mass equivalence of creatine. I'm currently without an extremely accurate scale.

1/2 tsp (~2.5 g) CreaPure Creatine monohydrate (do not buy commodity creatine! See below).

My container of creatine monohydrate (brand: "Skip" -- not CreaPure; I was impatient, though I've ordered a batch made with CreaPure creatine) says 2 tsp = 5 g.

Obviously, the volume to mass conversion could be very different for different brands. Still, just curious: have you actually weighed the amount of creatine you're taking?

Another question --

Creatine: As noted, I'm vegetarian, and our levels are lower than omnivores' and trials seem to show that vegetarians' brains work better on some tasks when given a low-dose supplement(18).

Here's what the paper says about how the creatine was administered:

2. MATERIAL AND METHODS
[....]
(b) Study design
Subjects were asked to consume this supplement at the same time each day for the next six weeks and received advice on how best to take this supplement to ensure maximum solubility and absorption.

Irritatingly non-informative! Before I write the principle author, just curious: have you per chance written her and learned what this "time" and "advice" were?

Thanks
Brian

18: Rae C, Digney AL, McEwan SR, Bates TC. Oral creatine monohydrate supplementation improves brain performance: a double-blind, placebo-controlled, cross-over trial. Proc Biol Sci. 2003 Oct 22;270(1529):2147-50. PubMed PMID: 14561278; PubMed Central PMCID: PMC1691485.

[Michael]

A couple of follow-up question regarding lysine (and arginine) as AGE scavengers. Would one's body weight and/or caloric consumption affect the dosing protocol?

Likely, yes, but I don't know of any data on this and don't have a good idea of how to do so.

You appear to consistently eat thrice daily. If one were eating five time daily would you advise dosing before each of those five meals?

Yes, but I wouldn't advise eating that often.

1/2 tsp (~2.5 g) CreaPure Creatine monohydrate (do not buy commodity creatine! See below).

My container of creatine monohydrate (brand: "Skip" -- not CreaPure; I was impatient, though I've ordered a batch made with CreaPure creatine) says 2 tsp = 5 g.

Obviously, the volume to mass conversion could be very different for different brands. Still, just curious: have you actually weighed the amount of creatine you're taking?

I trust you have the good stuff now? Really, I'd not advise taking commodity creatine. Last I checked, your brain already worked pretty well.

IAC: no, I hadn't measured. To my annoyance, I see that in reality, 1/2 tsp is ~1.8 g. I'm actually OK with that: there is no mention of any pharmacodynamic experimentation in the Rae et al trial, and I would be very surprised if they'd done any; from these three sources, it appears that typical omnivorous dietary intake is ≤ 2 g/d, and some of that will be unabsorbable due to cooking, unlike high-quality supplemental creatine. So this errs on the side of caution.

Here's what the paper says about how the creatine was administered:

2. MATERIAL AND METHODS
[....]
(b) Study design
Subjects were asked to consume this supplement at the same time each day for the next six weeks and received advice on how best to take this supplement to ensure maximum solubility and absorption.

Irritatingly non-informative! Before I write the principle author, just curious: have you per chance written her and learned what this "time" and "advice" were?

No; I expect they followed the now-standard protocol of taking it on an empty stomach with juice (or glucose), as I did, tho' that would drive much of it into muscle, not brain. However, because omnivores get their creatine with meals (and bound up within the food matrix), and since (as I noted in my original post) "it's not at all clear that this [1/3 C pom juice] is anything like enough glucose to meaningfully affect this, nor that the extra boost to tissue reserves as vs. water coadministration during the loading period is sustained with ongoing administration)," I am not overly concerned that this is going to affect the matter IAC.

Reference
Rae C, Digney AL, McEwan SR, Bates TC. Oral creatine monohydrate supplementation improves brain performance: a double-blind, placebo-controlled, cross-over trial. Proc Biol Sci. 2003 Oct 22;270(1529):2147-50. PubMed PMID: 14561278; PubMed Central PMCID: PMC1691485.

[Matt79]

Thanks for posting Michael, I see you take NAC eyedrops. How well is this supported? Have you noticed any effects?

[Matt79]

Creatine has been associated with hairloss.

[Michael]

Thanks for posting Michael, I see you take NAC eyedrops. How well is this supported?

Moderately well -- far better than most supplements, but not as well as I'd like. See this article (which is, of course, marketing material, rather than a disinterested and balanced assessment of the evidence -- but marketing material by someone that I like to think had his BS detector and scruples functioning at the time -- see my discl0sure). There has been more evidence of roughly similar quality since, but no long-term or very large studies -- and this guy said at SENS4 that his later studies had been less positive, and don't seem to've been published.

I might consider dropping them, on the balance of evidence alone, but (to answer your second question) anecdotally, I have apparently gotten some benefit out of them. I had floaters through most of my twenties and thirties, and thought nothing of them until doing the research required to write the op cit article, and started taking them. They resolved within a month or 2, IIRC, and haven't been back since.

[Michael]

Creatine has been associated with hairloss.

Do you know of any direct evidence for this? I see it asserted or asked about all over the 'net, but no studies indicating such. I believe this idea comes entirely from (1), reporting that after "25 g/day creatine with 25 g/day glucose ... for 7 days followed by 14 days of maintenance (5 g/day creatine with 25 g/day glucose...) ... serum T[estosterone] levels did not change. However, levels of DHT increased by 56% after 7 days of creatine loading and remained 40% above baseline after 14 days maintenance (P < 0.001). The ratio of DHT:T also increased by 36% after 7 days creatine supplementation and remained elevated by 22% after the maintenance dose (P < 0.01)."

It's a rigorous study design, so deserves to be taken seriously. However, the loading-phase dose is an order of magnitude more than I'm taking, and even the maintenance dose is (as I recently learned (thanks Brian!)) almost triple my normal daily dose. And, even after 14 d on the lower dose, they might yet have had levels higher than on long-term maintenance. So I'm not too worried about it, unless you (or anyone) know of more evidence -- and DHT is troubling for prostate cancer as well as hair loss. Watch This Space.

Reference
1: van der Merwe J, Brooks NE, Myburgh KH. Three weeks of creatine monohydrate supplementation affects dihydrotestosterone to testosterone ratio in college-aged rugby players. Clin J Sport Med. 2009 Sep;19(5):399-404. PubMed PMID: 19741313.

[nowayout]

Creatine has been associated with hairloss.

Do you know of any direct evidence for this? I see it asserted or asked about all over the 'net, but no studies indicating such. I believe this idea comes entirely from (1), reporting that after "25 g/day creatine with 25 g/day glucose ... The ratio of DHT:T also increased ..."

I think one can discount this study. In my opinion, they showed nothing of the sort.

First, it was one small crossover study with 20 people only.

Now, the creatine group started off with an average baseline DHT about 25% lower than the placebo group. Furthermore, in the placebo group, DHT fell by 20% after three weeks. Since in a crossover study, the placebo group are the same people as the creatine group, it is clear that random statistical variation in this case was very large and favored their chosen hypothesis by pure chance and regression to the mean.

Furthermore, since the "two" groups are in fact the same people, we can also test their hypothesis another way: The placebo group at baseline had a DHT of 1.26 +- 0.52 nmol/L, while the creatine group after 21 days supplementation had a DHT of 1.38 +- 0.45 nmol/L. Given that the two groups are the same people crossed over, we can also test their hypothesis by comparing these two numbers. Given the error bars, they are identical. Looked at this way, I'd say an equally valid conclusion would be that creatine doesn't raise DHT.

Reference
1: van der Merwe J, Brooks NE, Myburgh KH. Three weeks of creatine monohydrate supplementation affects dihydrotestosterone to testosterone ratio in college-aged rugby players. Clin J Sport Med. 2009 Sep;19(5):399-404. PubMed PMID: 19741313.

Edited by Michael, 17 July 2012 - 12:11 AM.

[Michael]

I believe this idea [that Creatine has been associated with hair loss] comes entirely from (1), reporting that after "25 g/day creatine with 25 g/day glucose ... The ratio of DHT:T also increased ..."

I think one can discount this study. In my opinion, they showed nothing of the sort. ...

Now, the creatine group started off with an average baseline DHT about 25% lower than the placebo group. Furthermore, in the placebo group, DHT fell by 20% after three weeks. Since in a crossover study, the placebo group are the same people as the creatine group, it is clear that random statistical variation in this case was very large and favored their chosen hypothesis by pure chance and regression to the mean.

Furthermore, since the "two" groups are in fact the same people, we can also test their hypothesis another way: The placebo group at baseline had a DHT of 1.26 +- 0.52 nmol/L, while the creatine group after 21 days supplementation had a DHT of 1.38 +- 0.45 nmol/L. Given that the two groups are the same people crossed over, we can also test their hypothesis by comparing these two numbers. Given the error bars, they are identical. Looked at this way, I'd say an equally valid conclusion would be that creatine doesn't raise DHT.

I'll have to take your word on the numbers, as I've not seen the full text -- but if those are right, and if there was not a significant washout period prior to the crossover (please inform!), then I suggest that this isn't the most likely reading of the results. As you say, in a crossover study, the "two" groups are in fact the same people. This means that one half of the creatine-administered subjects were previously not exposed to any intervention at baseline, and the other half had been administered a placebo; while one half of the placebo subjects were previously not exposed to any intervention at baseline, and the other half had been administered creatine.

If creatine boosts DHT levels -- and if, as the study seems to show, there is a significant lag before levels return to basesline -- then unless there was indeed a sufficient washout period, the placebo-administred group would be half-composed of people with residual elevated DHT from their prior period of creatine supplementatioin. And their levels would fall as they returned to true, unsupplemented baseline (which is another kind of "regression to the mean" .

Per contra, the creatine-administred group would be entirely composed of people with either their naive, unsupplemented DHT levels (which would then be lower than those of the placebo group), or at most would have a few folks with residual elevations caused by some manner of placebo effect (but of corse, teh placebo-administered group would also have this). The baseline level for the group as a whole would thus be lower than in the placebo arm as a whole -- and would rise, in response to creatinie administration.

But feel free to blow this out of the water by quoting the paper to indicate that they had a one-month washout period .

[Matt79]

Creatine has been associated with hairloss.

Do you know of any direct evidence for this? I see it asserted or asked about all over the 'net, but no studies indicating such. I believe this idea comes entirely from (1), reporting that after "25 g/day creatine with 25 g/day glucose ... The ratio of DHT:T also increased

Thanks Mike, that was the study I had in mind. Note I didn't say Creatine causes increased DHT. I'm not sure one study is enough, but I do think it's cause to be concerned.

I hate anecdote in general, but I was into gymming for a while and visited a few gyms. It seemed to me that body builders tended to have more hairloss than the average population of similar age.

That of course is pure speculation and observation on my part.

Edited by Michael, 17 July 2012 - 11:24 PM.

[Matt79]

The placebo group at baseline had a DHT of 1.26 +- 0.52 nmol/L, while the creatine group after 21 days supplementation had a DHT of 1.38 +- 0.45 nmol/L. Given that the two groups are the same people crossed over, we can also test their hypothesis by comparing these two numbers. Given the error bars, they are identical. Looked at this way, I'd say an equally valid conclusion would be that creatine doesn't raise DHT.

Interesting, this is the first evidence I've seen discounting the results of the SA Rugby study, based on statistical reasoning. I would love access to the actual paper to be able to check up on this.

Edited by Michael, 17 July 2012 - 11:23 PM.

[nowayout]

But feel free to blow this out of the water by quoting the paper to indicate that they had a one-month washout period .

The washout was six weeks, and

All subjects were residents at the Institute and
the same diet was given to all subjects. During part of the
washout period, all subjects had a short winter break and did
only maintenance training. Therefore, at the beginning of the
second phase of the study, subjects were in similar condition as
the start of the study and not fatigued from consecutive weeks
of match play.

Results:

TABLE 2. Testosterone, Dihydrotestosterone, and
Dihydrotestosterone to Testosterone Ratio Before, After
Loading, and After Maintenance Doses of Creatine with
Carbohydrate or Placebo with Carbohydrate


Testosterone (T; nmol/L) Day 0, Day 7, Day 21

Placebo 17.09 +- 3.42, 17.02 +- 4.11, 17.04 6 5.25
Creatine 14.44 +- 2.95, 16.08 +- 2.86, 16.69 +- 4.61

Dihydrotestosterone (DHT; nmol/L) Day 0, Day 7, Day 21

Placebo 1.26 +- 0.52, 1.09 +- 0.40, 1.06 +- 0.43
Creatine 0.98 +- 0.37, 1.53 +- 0.50*, 1.38 +- 0.45*

* indicates P < 0.001

My remark was that if you do a cross check by comparing placebo day 0 with creatine day 21, they are essentially identical, calling into question their claimed result.

Edited by viveutvivas, 17 July 2012 - 01:07 PM.

[Michael]

But feel free to blow this out of the water by quoting the paper to indicate that they had a one-month washout period .

The washout was six weeks, and [...] if you do a cross check by comparing placebo day 0 with creatine day 21, they are essentially identical, calling into question their claimed result.

Well, then.

Here is my response.

Thanks!

[Michael]

Do you know of any direct evidence for [Creatine having been associated with hair loss]? I see it asserted or asked about all over the 'net, but no studies indicating such. I believe this idea comes entirely from (1), reporting that after "25 g/day creatine with 25 g/day glucose ... The ratio of DHT:T also increased

Thanks Mike, that was the study I had in mind. ... I hate anecdote in general, but I was into gymming for a while and visited a few gyms. It seemed to me that body builders tended to have more hairloss than the average population of similar age.

That of course is pure speculation and observation on my part.

Well, in addition to "anecdotal evidence" being a contradiction in terms, I'd say this is an exceptionally uninspiring anecdote . There are a lot of differences between bodybuilders and the general population, and of course not all bodybuilders use creatine (tho' I'd be surprised if many didn't these days). Aside from the very brief acute increase in T immediately after intense resistance exercise, which might (don't know the science here) in turn lead to an increase in DHT (and maybe a longer-lasting one), bodybuilders are much more likely to be using anabolic steroids than the general population; and I imagine (again, don't have the data here) that they have higher pre-bodybuilding T and/or T responsiveness to exercise than the general population even, as this would (a) make them more "aggressive" about pursuing their exercise, (b) directly make them more successful, and © provide positive reinforcement because of that greater success, which would push them to keep lifting iron and pushing themselves harder, in a nice little positive feedback loop. Any of these would be expected to increase the incidence of male pattern baldness in bodybuilders.

At minimum, if it were possible, you'd want to compare creatine-using to non-creatine-using bodybuilders -- not bodybuilders to holders-down-of-couches.

[Matt79]

Well, in addition to "anecdotal evidence" being a contradiction in terms,

http://en.wikipedia.org/wiki/Evidence

Think you might be thinking too narrowly in this instance.

Types of evidence

• Anecdotal evidence
• Intuition
• Personal experience
• Scientific evidence
• Testimonial

Edited by Matt79, 18 July 2012 - 01:27 AM.

[Matt79]

At minimum, if it were possible, you'd want to compare creatine-using to non-creatine-using bodybuilders -- not bodybuilders to holders-down-of-couches.

Why? If for example 80% of bodybuilders use creatine at the gyms, then a generalization is useful. Inferential probalistic reasoning does not seem to sit well with you.

[niner]

Dihydrotestosterone (DHT; nmol/L) Day 0, Day 7, Day 21

Placebo 1.26 +- 0.52, 1.09 +- 0.40, 1.06 +- 0.43
Creatine 0.98 +- 0.37, 1.53 +- 0.50*, 1.38 +- 0.45*

* indicates P < 0.001

I fail to see the problem. Perhaps the randomization was sub-optimal or the population a bit too small, but that's still a large rise in DHT in the treatment group; P<0.001. Within the error bars, DHT is identical at day 0 in both treatment and placebo group. I'm afraid I can't dismiss the study on that basis. As I've mentioned in these fora recently, I have an anecdote regarding creatine and hair loss that I respect more than the usual anecdote. It was my own hair going down the drain. Shortly after I started creatine, I had a major WTF level of hair loss. A bit of googling turns up lots of people with the same story. When I stopped the creatine, the hair loss stopped very quickly. I'm probably more susceptible to DHT-driven hair loss than the average guy; I've responded very well to finasteride in that regard.

[Matt79]

Dihydrotestosterone (DHT; nmol/L) Day 0, Day 7, Day 21

Placebo 1.26 +- 0.52, 1.09 +- 0.40, 1.06 +- 0.43
Creatine 0.98 +- 0.37, 1.53 +- 0.50*, 1.38 +- 0.45*

* indicates P < 0.001

I fail to see the problem. Perhaps the randomization was sub-optimal or the population a bit too small, but that's still a large rise in DHT in the treatment group; P<0.001. Within the error bars, DHT is identical at day 0 in both treatment and placebo group. I'm afraid I can't dismiss the study on that basis. As I've mentioned in these fora recently, I have an anecdote regarding creatine and hair loss that I respect more than the usual anecdote. It was my own hair going down the drain. Shortly after I started creatine, I had a major WTF level of hair loss. A bit of googling turns up lots of people with the same story. When I stopped the creatine, the hair loss stopped very quickly. I'm probably more susceptible to DHT-driven hair loss than the average guy; I've responded very well to finasteride in that regard.

Your story is similar to mine too. I forgot to mention that, apart from observing others at the gym, my own experience was one of rapid thinning over the period I did weights (creatine on and off - about 12-18 months). Since I stopped all that nonsense and went on Finasteride with Ketoconazole (2% shampoo) as a backup, thinning has ceased to be an issue.

Look I'm also skeptical Creatine causes increased DHT and therefore hairloss, all I'm saying is it appears to be implicated at this stage and more trials are needed, and hence the precautionary principle works for me in this case, ie avoid for now.

Edited by Matt79, 18 July 2012 - 08:33 AM.

[Michael]

Well, in addition to "anecdotal evidence" being a contradiction in terms,

http://en.wikipedia.org/wiki/Evidence

Think you might be thinking too narrowly in this instance.

Types of evidence

• Anecdotal evidence
• Intuition
• Personal experience
• Scientific evidence
• Testimonial

Rather, Wikipedia and the culture at large is thinking too sloppily in this instance .

At minimum, if it were possible, you'd want to compare creatine-using to non-creatine-using bodybuilders -- not bodybuilders to holders-down-of-couches.

Why? If for example 80% of bodybuilders use creatine at the gyms, then a generalization is useful.

I just explained, I think pretty clearly, why this is exactly what makes the anecdote so exceptionally weak.

Dihydrotestosterone (DHT; nmol/L) Day 0, Day 7, Day 21

Placebo 1.26 +- 0.52, 1.09 +- 0.40, 1.06 +- 0.43
Creatine 0.98 +- 0.37, 1.53 +- 0.50*, 1.38 +- 0.45*

* indicates P < 0.001

I fail to see the problem. Perhaps the randomization was sub-optimal or the population a bit too small, but that's still a large rise in DHT in the treatment group; P<0.001. Within the error bars, DHT is identical at day 0 in both treatment and placebo group.

Yes, but one really would be forced to take the possibility of regression to the mean pretty seriously on these numbers, even in a simple, linear controlled trial -- let alone in a crossover trial, as in this one.

I have an anecdote regarding creatine and hair loss that I respect more than the usual anecdote. It was my own hair going down the drain. Shortly after I started creatine ... I've responded very well to finasteride in that regard.

Your story is similar to mine too. I forgot to mention that, apart from observing others at the gym, my own experience was one of rapid thinning over the period I did weights (creatine on and off - about 12-18 months). Since I stopped all that nonsense and went on Finasteride with Ketoconazole (2% shampoo) as a backup, thinning has ceased to be an issue.

Well, again, the finasteride is a pretty substantial confounder. I can't reasonably ask you to try it, but a much better (albeit still anecdotal, n=1)l test would be what happened if you quit creatine and did nothing else.

Look I'm also skeptical Creatine causes increased DHT and therefore hairloss, all I'm saying is it appears to be implicated at this stage and more trials are needed, and hence the precautionary principle works for me in this case, ie avoid for now.

I certainly can't argue with that, especially if there are a lot of anecdotal reports: when I Googled previously, it seemed mostly to be either rumors, worries based on the rugby player study, or hearsay ("It happened to my cousin Joey ..."). I agree that it'd be nice to have more and larger trials on this (and also some attempt at a case-control study).

[nowayout]

I'd be more inclined to take this seriously if some biological explanation for why creatine should modify DHT could be offered.

[niner]

Dihydrotestosterone (DHT; nmol/L) Day 0, Day 7, Day 21

Placebo 1.26 +- 0.52, 1.09 +- 0.40, 1.06 +- 0.43
Creatine 0.98 +- 0.37, 1.53 +- 0.50*, 1.38 +- 0.45*

* indicates P < 0.001

I fail to see the problem. Perhaps the randomization was sub-optimal or the population a bit too small, but that's still a large rise in DHT in the treatment group; P<0.001. Within the error bars, DHT is identical at day 0 in both treatment and placebo group.

Yes, but one really would be forced to take the possibility of regression to the mean pretty seriously on these numbers, even in a simple, linear controlled trial -- let alone in a crossover trial, as in this one.

I don't understand. The two arms of the study seem to be regressing to different means. If it was mean regression, shouldn't they be tending toward the same value?

I'd be more inclined to take this seriously if some biological explanation for why creatine should modify DHT could be offered.

Well, I know how you feel; it always helps to have a plausible biochemical story behind it, but those are often wrong, and may be misleading. I think that sometimes it's better to let the data be the "facts on the ground" that you have to explain.
I don't actually know if there's a biological explanation for why creatine enhances muscle growth- is that well understood?

[nowayout]

Dihydrotestosterone (DHT; nmol/L) Day 0, Day 7, Day 21

Placebo 1.26 +- 0.52, 1.09 +- 0.40, 1.06 +- 0.43
Creatine 0.98 +- 0.37, 1.53 +- 0.50*, 1.38 +- 0.45*

* indicates P < 0.001

I fail to see the problem. Perhaps the randomization was sub-optimal or the population a bit too small, but that's still a large rise in DHT in the treatment group; P<0.001. Within the error bars, DHT is identical at day 0 in both treatment and placebo group.

Yes, but how do you reconcile with their hypothesis that the first number on placebo is essentially the same as the last number on creatine (given error bars)? (And that the last number on placebo is essentially the same as the first number on creatine?) Remember, the placebo group is the same people as the treatment group.

Their data seems to say that placebo increases DHT as much as creatine does, except the effect of the former is backwards in time.

Edited by viveutvivas, 18 July 2012 - 07:40 PM.

[niner]

Dihydrotestosterone (DHT; nmol/L) Day 0, Day 7, Day 21

Placebo 1.26 +- 0.52, 1.09 +- 0.40, 1.06 +- 0.43
Creatine 0.98 +- 0.37, 1.53 +- 0.50*, 1.38 +- 0.45*

* indicates P < 0.001

I fail to see the problem. Perhaps the randomization was sub-optimal or the population a bit too small, but that's still a large rise in DHT in the treatment group; P<0.001. Within the error bars, DHT is identical at day 0 in both treatment and placebo group.

Yes, but how do you reconcile with their hypothesis that the first number on placebo is essentially the same as the last number on creatine (given error bars)? (And that the last number on placebo is essentially the same as the first number on creatine?) Remember, the placebo group is the same people as the treatment group.

Within the error bars, every number in the experiment is the same as every other number. And yet there's a statistically significant signal. I think we have to go with what the statistics says.

[nowayout]

And yet there's a statistically significant signal. I think we have to go with what the statistics says.

Many results of statistical significance turn out not to be real effects, as we have seen in many medical trials much larger than this one in recent decades that found significant effects that diminished or disappeared in subsequent trials.

Given that the initial DHT level in the treatment group is an outlier, as we see when comparing it to all the placebo DHT measurements (in the same people), this presumably random fluctuation would have unfairly favored their hypothesis by at the very least causing the magnitude of the effect (if any) to be overestimated. Can we agree on that much?

Furthermore, I believe the outcome would not have been significant if the day 0 treatment value weren't a statistical outlier, for the following reason: Consider the average of all non-treatment DHT values - it is about 1.1, which would have been a better starting value for average untreated DHT in the creatine arm, instead of the statistical outlier that was taken. In that case, the rise would have only been 0.28 points instead of 0.40. But if you look at all the nontreatment values, 0.28 points is the range of the random variation of DHT anyway, and thus cannot be statistically significant.

Edited by viveutvivas, 18 July 2012 - 11:19 PM.

[niner]

Hmm. I am wondering how they arrive at P<0.001, but have error bars of 30-40%. On the face of it, that looks odd. They got it past the reviewers... that doesn't mean it's right, but it's something. Anyway, If we're going to talk about outliers, it's actually the day 0 of the placebo arm that is the most different. If you throw it out, and average the rest of them, then you get 1.04 as a putative starting value. I don't think that's methodologically justifiable, but, whatever.

[maxwatt]

Would cognac have the same, or some of the same benefits as wine? It should contain many of the same polyphenols despite the distillation process.

[niner]

I would think that distillation would remove most of the polyphenols, since they aren't particularly volatile. It would at least have the benefits of the alcohol content, which I suspect is a lot of the benefit. There was just a paper posted here a few weeks back claiming beer was a good for you as wine.

[Matt79]

Furthermore, I believe the outcome would not have been significant if the day 0 treatment value weren't a statistical outlier, for the following reason: Consider the average of all non-treatment DHT values - it is about 1.1, which would have been a better starting value for

Hmm. I am wondering how they arrive at P<0.001, but have error bars of 30-40%. On the face of it, that looks odd. They got it past the reviewers... that doesn't mean it's right, but it's something. Anyway, If we're going to talk about outliers, it's actually the day 0 of the placebo arm that is the most different. If you throw it out, and average the rest of them, then you get 1.04 as a putative starting value. I don't think that's methodologically justifiable, but, whatever.

Thanks guys, I feel horribly rusty being 10 years out of varsity, so won't be able to contribute to the discussion just yet. While those error margins look horribly large, I can't say for certain that it makes a P value of < 0.001 impossible, which is frustrating.

On a more positive note, I enrolled yesterday in an online refresher course of Stats 101. :p

Edited by Matt79, 19 July 2012 - 05:57 AM.

[brianmdelaney]

Here's what the paper says about how the creatine was administered:

2. MATERIAL AND METHODS
[....]
(b) Study design
Subjects were asked to consume this supplement at the same time each day for the next six weeks and received advice on how best to take this supplement to ensure maximum solubility and absorption.

Irritatingly non-informative! Before I write the principle author, just curious: have you per chance written her and learned what this "time" and "advice" were?

No; I expect they followed the now-standard protocol of taking it on an empty stomach with juice (or glucose), as I did, tho' that would drive much of it into muscle, not brain. However, because omnivores get their creatine with meals (and bound up within the food matrix), and since (as I noted in my original post) "it's not at all clear that this [1/3 C pom juice] is anything like enough glucose to meaningfully affect this, nor that the extra boost to tissue reserves as vs. water coadministration during the loading period is sustained with ongoing administration)," I am not overly concerned that this is going to affect the matter IAC.

Reference
Rae C, Digney AL, McEwan SR, Bates TC. Oral creatine monohydrate supplementation improves brain performance: a double-blind, placebo-controlled, cross-over trial. Proc Biol Sci. 2003 Oct 22;270(1529):2147-50. PubMed PMID: 14561278; PubMed Central PMCID: PMC1691485.

I finally got around to writing the princ. auther. Her response was that the researchers asked the subjects to dissolve the powder in hot tea and to make sure it was all dissolved as best as possible.

That was it!

[snikeris]

Michael,

Why do you take Pyroglutamic acid more often than Piracetam? It seems that there are more studies demonstrating the safety of Piracetam than there are of Pyroglutamic acid.

[nowayout]

Would cognac have the same, or some of the same benefits as wine? It should contain many of the same polyphenols despite the distillation process.

One would hope so. Also, at least some of the benefits seem to be from the alcohol itself, not the polyphenols.

As for myself, I like gin, so I kind of hope that also counts. I do like red wine but it gives me a heavy, headachey feeling (maybe it's the sulfites they so idiotically insist on here in the U.S.) whereas a good quality gin gives me a light and airy high with no consequences ever...