Posted by linkadge on December 30, 2006, at 17:38:33
In reply to Re: After the hype » linkadge, posted by Klavot on December 30, 2006, at 11:32:37
I see your logic.
Although, I think my reasoning was even more flawed in that you would never have a certain percentage of trials showing a specific sucess rate. In fact it would be hard to find any two trials that show the exact same sucess rate.
In addition, this method only works if one assumes that each clinical trial has the same number of people in it. So even saying that more than 50% of trials fail doesn't say much, as the failed trials might have significantly fewer participants than the trials that are sucessfull (ie it should be weighted towards the sucessfull trials)
I'm just going to try and forget what I've started here :)
Ie. I don't think you can find the percentage of people helped by antidepressants without knowing the exact number of people helped and not helped in each clinical trial.
If you wanted to go into even more detail you could factor in the degree to which each person felt they were helped :)
I'll shut up now.
Linkadge
poster:linkadge
thread:717262
URL: http://www.dr-bob.org/babble/20061224/msgs/717635.html