This site is not maintained. Click here for the new website of Richard Dawkins.

Comment

← James Randi Explains - Homeopathy

Luciani's Avatar Jump to comment 9 by Luciani

I noticed on the site that someone links to JB explains Homeopathy better than James Randi: http://www.youtube.com/watch?gl=US&hl=hi&v=4Zo-yUJ_2aI

JB then responds to a challenge of producing a trial that has a positive outcome over placebo. He responds as such:

[quote]You ask for one scientifically positive-proven survey about homeopathy. I will do better than that. I will give you 5 META ANALYSES:
Cucherat etal 2000* 16 Hi-Qt studies POSITIVE.
Linde& Melchart 1998* 32 Hi-Qt studies POSITIVE.
Lindeetal 1997* 89 studies POSITIVE.
Boissel etal 1996 15 Hi-Qt studies POSITIVE.
Kleijnenetal 1991 105 studies POSITIVE.
Will you allow your prejudice or scientific mind to win out. You have lost the argument but your prejudice is much more comfortable, isn't it?[/quote]

Linde & Melchart 1998: "when the analysis was restricted to the methodologicall y best trials no significant effect was seen ... The evidence, however, is not convincing because of methodological shortcomings and inconsistencies ."
I also noticed it was published in J Altern Complement Med.
Not exactly a Cochrane meta-analysis (a very professional institution specializing in meta-analyses). It would also be interesting to compare these results to a funnel-plot test of the trials selected to exclude publication bias.

On another note: I could not find any of the other mentioned articles on pubmed (the recognized database portal for ALL clinical trials). Links, good sir, links.

Something people also tend to forget: statiscally speaking in the scientific community, we recognize a trial as significant when there is less than a 5% chance, that the result is because of pure chance.
When there have been conducted trials for ca. 200 years there are bound to be quite a few trials that have positive results based on chance.

That is why it is exceedingly important that meta-analyses are based on stringent, transparent criteria so as to avoid cherry-picking of even the most well-conducted trials on homeopathy.

We all ridicule the fundamentals of homeopathy (with damn good reason) but what to do when they respond with trials back? By critically evaluating them as with every other article.
Do not be bullied by references to absolute crap articles when scrutinized. There is a reason the scientific community condemns this sort of quackery. And especially do not be put down by a doctor or scientist endorsing it.

A quote comes to mind, which I am para-phrasing:
"There isn't a hypothesis stupid and ridiculous enough that you can't find at least one Ph.D. to endorse it" Frighteningly true. Which is why authority is no substitution to the scientific method which ANYONE can practice.

Thu, 18 Feb 2010 12:45:00 UTC | #442107