Psycho-Babble Medication | about biological treatments | Framed
This thread | Show all | Post follow-up | Start new thread | List of forums | Search | FAQ

Re: OOOOOOOOOOOOooooooooo » linkadge

Posted by Larry Hoover on April 19, 2008, at 15:30:15

In reply to Re: OOOOOOOOOOOOooooooooo, posted by linkadge on April 19, 2008, at 14:48:09

> >In the era in consideration, publication of >negative studies almost never occurred.
>
> There you go.

So, how is that to be blamed on drug companies? The article that revealed this publication bias qualified its results by saying they could not determine if in fact that bias arose within the publication industry, or if the studies had never been submitted. Why do we never hear about this significant qualifier?

> >Moreover, negative studies teach us nothing, as >we cannot discrimate between failed methodology >and failed drug.
>
> Oh, I see, we only pay attention to postitive trials. That makes a lot of sense.

That's not what I said. A finding of the null hypothesis could be due to there being no true difference to be found, or it could be a failure of the study methodology to detect a true difference. They cannot be distinguished from each other. As a result, such a study is considered to be a failed study, which serves only to guide further development of hypotheses and methodology. It doesn't show us anything else.

> >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?
>
> Well, if that level of significance is enough for you then great.

Putting data to tests for which they were not collected always raises the issue of unexamined confounds. Again, post hoc analysis is only appropriate in further hypothesis development and testing, but not re-evaluation of the meaning of the data themselves.

> >IMHO, he fished around until he found the >statistics he liked, and he published those and >those alone.
>
> If that is your take.
>
> Linkadge

Yes, it is. The NICE document showed a variety of tests of the antidepressant data, and Kirsch picked the weakest finding. For example, NICE found that: "There is strong evidence suggesting that there is a clinically significant difference favouring SSRIs over placebo on increasing the likelihood of patients achieving a 50% reduction in depression symptoms as measured by the HRSD (N = 1719; n = 3143; RR = 0.73; 95% CI, 0.69 to 0.78)." Not only did the statistic reach *clinical significance*, it was also rated as "strong evidence". I dismiss Kirsch both for what he said, and what he didn't say.

Refer to NICE for further definitions: http://www.nice.org.uk/guidance/index.jsp?action=download&o=29617

Here's an example of cherry-picking from NICE. This statistic dismisses the efficacy of CBT (cognitive behavioural therapy), when compared to *doing nothing at all*.

p. 123
"There is insufficient evidence to determine if there is a clinically significant difference between CBT and wait list control on increasing the likelihood of achieving remission as measured by the BDI (N = 1; n = 24; RR = 0.70; 95% CI, 0.41 to 1.20)."

Acccording to this analysis, you are going to have a similar likelihood of achieving remission whether you receive CBT, or you remain on a wait-list for that therapy.

Of course, NICE presents a lot of other data, and statistics. I present this limited example to demonstrate how easy it is to bias conclusions. A critical analysis of the external validity of this finding would reveal that there is more to the story. And that's all I did when I trashed Kirsch.

Lar

 

Thread

 

Post a new follow-up

Your message only Include above post


Notify the administrators

They will then review this post with the posting guidelines in mind.

To contact them about something other than this post, please use this form instead.

 

Start a new thread

 
Google
dr-bob.org www
Search options and examples
[amazon] for
in

This thread | Show all | Post follow-up | Start new thread | FAQ
Psycho-Babble Medication | Framed

poster:Larry Hoover thread:823248
URL: http://www.dr-bob.org/babble/20080412/msgs/824304.html