Posted by larryhoover on December 6, 2013, at 22:27:45
In reply to Re: Irving Kirsch, placebos and antidepressants » larryhoover, posted by doxogenic boy on November 11, 2013, at 12:06:25
> I have found a meta analysis in fulltext in Canadian Medical Association Journal:
>
> http://www.cmaj.ca/content/178/3/296.long
>
> Excerpt from article:
> Effectiveness of paroxetine in the treatment of acute major depression in adults: a systematic re-examination of published and unpublished data from randomized trials
>
> Corrado Barbui MD,
> Toshiaki A. Furukawa MD,
> Andrea Cipriani MD
>
> [...]
> Interpretation: Among adults with moderate to severe major depression in the clinical trials we reviewed, paroxetine was not superior to placebo in terms of overall treatment effectiveness and acceptability. These results were not biased by selective inclusion of published studies.I had not seen this article before, and frankly I don't know whether to laugh or to cry.
They have decided that the the primary criterion for antidepressant efficacy is not the actual change in depression score (which actually significantly lies with antidepressant treatment), and instead substitute their own measure, drop-out rate.
In the abstract, they clarify that significantly more subjects dropped out in the treatment arm because of side effects (paxil is notorious for side effects), but they fail to account for the fact that the net drop-out rate was equivalent between the groups (RR = 0.99), yet they don't account for what might have balanced the two groups overall (e.g. lack of efficacy in the placebo group, perhaps?).
I've been kind, so far, to consider the report on face value, so far. But have you ever seen an antidepressant rated by its drop-out rate against placebo? Do you think that the medical regulators would accept this as a criterion of efficacy? Do you think that scientific studies should be re-assessed on parameters that were not primary controlled measures of the methodology? No, No, and No.
This is a garbage article. I am aghast.
Lar
poster:larryhoover
thread:1052457
URL: http://www.dr-bob.org/babble/20131115/msgs/1055611.html