Posted by Larry Hoover on February 3, 2007, at 8:43:36
In reply to Re: Report: Comparative Effectiveness of Antidepres..., posted by mindevolution on February 2, 2007, at 2:05:50
The type of report referenced here is the result of a very conservative methodology. By definition, it excludes the bulk of all published research or reports, based on clearly defined criteria that are laid out on page 41. In total, 2,099 citations were identified, but they ultimately included 293 in the final detailed analysis. See the discussion on page 46, and the diagram on page 47.
> key question 2a page 23 of 167: efficacy and effectiveness of sgad (sic) for maintaining response or remission? (i.e preventing relapse or reoccurence)
> what?! they prevent depression from occurring???
The subject is clearly expressed on page 33. In fact, this report is the first evidence-based analysis which assesses long-term outcomes in this manner. That's why I thought it worthwhile to post the link.
> severe adverse events on page 26 of 167
> there is no data for suicide, seizures or cardiovascular events or any other severe adverse event! what is going on with that study, is it sponsored by all the drug companies combined?
No. Nothing of the sort. In an evidence-based analysis, the quality of the data made available is also assessed. I would paraphrase the conclusions expressed as: "There is little available data, and the data published are of insufficient quality to make firm conclusions."
The methodology excludes case reports, or other anecdotal evidence. Absent any control subjects, it is impossible to analyze such cases by statistical means.
Clearly, the published data do not adequately address concerns vis a vis adverse events. That is not the fault of these authors.
> from pages 60 - 73 they always suggest there is no difference between most of the ADs! the ADs vary significantly in chemical structure, something smells a bit about all the "meta analysis" going on.
Frankly, I find the statistical analysis in this section to have been superbly done. Moreover, I cannot find even a hint of bias in the entire report.
> on page 74, the graph showing the summary shows that the only drug that is persecuted by the study is fluvoxamine which has been discontinued as a brand in the US! by chance, no way!
I'm not exactly sure what you're driving at. Fluvoxamine did seem to be the underdog in comparator studies. I think that might be a relevant detail. If I recall correctly, it is often preferred for treatment of PTSD. I was unaware of the branding issue. It is still marketed, though, isn't it?
> I stopped reading at p75, i suspect drug company involvement to the point of it being a waste of time reading the document.
The research authors all had no drug company connections. All documents were accessed from publicly available resources, except 3 articles supplied by drug companies. (page 40)
I too condemn the publication bias arising from private funding of the bulk of pharmaceutical research. Roughly 2/3 of the documents considered in this study had direct pharmaceutical company funding. In more than 20% of the cases, funding could not be determined. Only 6.7% were funded by government.
This published review could only consider the literature extent. It would be of great public service to require Phase IV reviews for all drugs. Unfortunately, we cannot make up for the lost historical opportunities to have done so. The drug review process is constantly evolving. It was not so very long ago that a drug could be developed and marketed without any oversight whatsoever.
Lar
poster:Larry Hoover
thread:728734
URL: http://www.dr-bob.org/babble/20070201/msgs/729287.html