Posted by Larry Hoover on January 13, 2008, at 11:19:28 [reposted on January 15, 2008, at 3:12:46 | original URL]
In reply to Re: placebo response, posted by SLS on January 8, 2008, at 16:00:40
> > > We don't know how many people the drugs actually helped, just that more remitted while on them.
> > Although sometimes, it may even be the other way around. For example, in the SJW trial that was published in JAMA in 2002, the remission rates at 8 weeks were 31.9% for placebo and 24.8% for sertraline.
> > I have the article in front of me right now. LOL, I still giggle when reading it. The way they conclude that the study does not support the efficacy of SJW, but they say nothing about Zoloft, even though SJW and Zoloft both failed to outperform placebo.
> > The full, original article can be accessed at
> > http://jama.ama-assn.org/cgi/reprint/287/14/1807
> Technically, this is a "failed study". Since it has already been well established that sertraline is more effective than placebo, this comparator acts as a fail-safe to judge the validity of the trial.
> - Scott
I'm trying to recall, but I do believe this is the only study I've come across where the placebo group came out on top. However, I have to agree with Scott, that this can only be viewed as a failed study. The inclusion of an active comparator that also failed implicates the methodology, rather than the investigative drug. It's useful for generating new hypotheses, but worthless insofar as answering any.