3 ms·

Almost all the studies used in the first meta-analysis, and all the studies in the second, were investigating the secondary preventative effects of Ω-3 suppleme

by rosser 12y ago

Almost all the studies used in the first meta-analysis, and all the studies in the second, were investigating the secondary preventative effects of Ω-3 supplementation on CVD. That means the people being studied already had heart disease. Supplementation didn't demonstrate a definitive impact on all-cause mortality, or cardiovascular events (though in both cases, there appears to have been a slight reduction in cardiovascular deaths) in that population. That's hardly surprising, and proves nothing at all, except perhaps something we already knew: people with heart disease have heart attacks and die.

scott_s 12y ago
"Proves" is a complicated word. What they did was provide evidence supporting the hypothesis that Omega-3 supplementation does not make a significant difference for people who already have heart disease. While not ground-breaking, it's something. And if you gather enough meta-studies showing "Supplementation of x in y circumstance had no significant effect", then it starts to become reasonable to claim supplementing x in general is not worthwhile. (I'm not saying they have done that, but I'm pointing out that this is part of the corpus of data we need to test a supplement in general.)
- joe_the_user 12y ago
  And if you gather enough meta-studies showing "Supplementation of x in y circumstance had no significant effect", then it starts to become reasonable to claim supplementing x in general is not worthwhile Woah, that's a horrible way to approach things. As far as I know vitamin C only prevents exactly one ailment, scurvy. By that token, you could test against a hundred other random conditions and claim you'd proved it was worthless. Of course, you test substances for ailments you have a reason to think they will prove effective against and any other tests are beside the point. Also, whatever the merit of the study mentioned, for its purse, the GP citing here is kind of misleading or irrelevant. People don't take fish oil to cure their existing heart disease, they take supplements to prevent heart disease (which it may or may not do regardless of whether or not it does other things).
  - scott_s 12y ago
    There is no evidence that vitamin C supplementation has any benefit. I have seen studies on this, and they don't show any benefit. Please note the difference between supplementation and deficiency. I am using "supplementation" in the same way that I understand the medical community uses it: for strictly when you are supplementing a normal diet, and you are not already deficient in that vitamin or compound. If you are deficient in vitamin C because you are forced to eat foods without it for a long period, then you should take vitamin C pills to prevent scurvy. But if you are not deficient in vitamin C, and you do get it from your normal diet, then no one has established any benefit to its supplementation. In that context, I think that the meta-study is relevant. People supplemented their normal diet with omega-3s, and it didn't help. Yes, there are other circumstances we have to test. We have not "proved" that omega-3 supplementation is not beneficial. But we need to amass a large corpus of evidence to address the question, and that is part of it.

by rosser 12y ago

scott_s 12y ago

"Proves" is a complicated word. What they did was provide evidence supporting the hypothesis that Omega-3 supplementation does not make a significant difference for people who already have heart disease. While not ground-breaking, it's something. And if you gather enough meta-studies showing "Supplementation of x in y circumstance had no significant effect", then it starts to become reasonable to claim supplementing x in general is not worthwhile. (I'm not saying they have done that, but I'm pointing out that this is part of the corpus of data we need to test a supplement in general.)

joe_the_user 12y ago
And if you gather enough meta-studies showing "Supplementation of x in y circumstance had no significant effect", then it starts to become reasonable to claim supplementing x in general is not worthwhile Woah, that's a horrible way to approach things. As far as I know vitamin C only prevents exactly one ailment, scurvy. By that token, you could test against a hundred other random conditions and claim you'd proved it was worthless. Of course, you test substances for ailments you have a reason to think they will prove effective against and any other tests are beside the point. Also, whatever the merit of the study mentioned, for its purse, the GP citing here is kind of misleading or irrelevant. People don't take fish oil to cure their existing heart disease, they take supplements to prevent heart disease (which it may or may not do regardless of whether or not it does other things).
- scott_s 12y ago
  There is no evidence that vitamin C supplementation has any benefit. I have seen studies on this, and they don't show any benefit. Please note the difference between supplementation and deficiency. I am using "supplementation" in the same way that I understand the medical community uses it: for strictly when you are supplementing a normal diet, and you are not already deficient in that vitamin or compound. If you are deficient in vitamin C because you are forced to eat foods without it for a long period, then you should take vitamin C pills to prevent scurvy. But if you are not deficient in vitamin C, and you do get it from your normal diet, then no one has established any benefit to its supplementation. In that context, I think that the meta-study is relevant. People supplemented their normal diet with omega-3s, and it didn't help. Yes, there are other circumstances we have to test. We have not "proved" that omega-3 supplementation is not beneficial. But we need to amass a large corpus of evidence to address the question, and that is part of it.