Posted by Larry Hoover on April 19, 2008, at 13:54:05
In reply to Re: OOOOOOOOOOOOooooooooo, posted by linkadge on April 19, 2008, at 10:30:17
> I think the study that Betula brought up is very important, becase it is infact *not* just one study. It is a meta analysis of 47 other studies and the largest one of its kind to date. As mentioned it includes data that the drug companies conveniently left out which increases its validity beyond any one single study or any previous subgroup of *more positive* released trial data.
It is not data the drug companies left out. All of it was submitted to the FDA. Some of it was not published. That issue was already discussed at length, and you cannot lay the responsibility on just the drug companies. In the era in consideration, publication of negative studies almost never occurred. Moreover, negative studies teach us nothing, as we cannot discrimate between failed methodology and failed drug.
> Saying that Kirsh is biased only goes so far. If you are going to make a substantial argument that he has somehow biased this data than do so. I have not seen one person pose a good reason why his methodology is flawed in some way.
Funny, I thought I did all right.
> He was using standard methods of statistical analysis as far as I know, can you or anyone point to his flaws?
See my post entitled Re:debate.
> The clinical trial is unfortunately the only real way to scientifically establish the efficacy of antidepressants.
And Kirsch found statistically superiority of drugs over placebo, p <.001, including the "hidden" data. The new bar he held the old data up to was a novel and arbitrary standard of clinical significance, a test the original studies were not designed to meet. We don't take older cars off the road because they don't meet current emissions/efficiencies standards, so why should we retrospectively reassess these drugs based on data collected for other purposes? What he did is called data mining, and it is scientifically frowned upon. IMHO, he fished around until he found the statistics he liked, and he published those and those alone. A quick reference to the NICE document would show that to be the case.
Lar
poster:Larry Hoover
thread:823248
URL: http://www.dr-bob.org/babble/20080412/msgs/824276.html