Posted by Larry Hoover on September 13, 2007, at 9:26:06
In reply to Re: STAR*D study, 33% sucess with first AD-sam123, posted by jhj on September 13, 2007, at 1:27:43
>
> Please do not mention about STAR*D study here.You read the comments made on STAR*D study here in reponse to another thread.
>
> "Without a placebo arm, the results are kind of meaningless.
>
> Linkadge"I missed the earlier discussion, but I'll join in this one.
The methodology for this study is entirely different than that used for placebo-controlled clinical efficacy trials.
> Then,here is one by another very knowledgable person to the same thread.
>
> "Exactly - what waste of time, money and effort. It's also not clear what samples or populations the percentages refer to"
>
> If you want to know about the person and the quality of that person,here it is.
>
> "I'm a statistician so I'm hopefully not talking out of my a**e Fred"First, this methodology is a within-group analysis, not a between-groups analysis. One homogenous population was selected, given a single treatment, and then proceeding to differing treatments according to displayed heterogeneity. The methodology is to demonstrate heterogeneity within the subject population, not heterogeneity in the treatment used.
The study, furthermore, is not yet complete. Published data, so far, are based on very limited samples of what will yet emerge. All we have, so far, are hints.
Lar
poster:Larry Hoover
thread:781684
URL: http://www.dr-bob.org/babble/20070911/msgs/782635.html