On 27 February 2008 Chris Green wrote in response to previous posting on this thread: >The issue is not, as you say (but attempt to impute to me) as >simple as antidepressants "work" or "don't work."
A search of my posting does not bring up the "work" or "don't work" dichotomy directly in relation to anything specific that you wrote, so I'll presume you are referring to the equivalent wording, where I wrote initially that "the question of the efficacy of antidepressants" is not a straightforward right/wrong issue *specifically* in response to your saying in relation to articles I cited: "These two articles are ancient history now." Since the articles are both critical examinations of meta-analyses of studies on the efficacy of antidepressants, I took you to mean that critical analyses of the methodology of studies, and of meta-analyses, were "ancient history". (How else could I take it?) The clear implication seemed to be that there is no need to spend time examining the methodology of antidepressant efficacy studies any longer. If that is not what you meant - though, as I say, I can't see how else one could take it - I apologise for misinterpreting your position. Chris wrote, quoting me first: >Allen Esterson wrote: >> I fear that this is going to be a lengthy posting,... >And it was, but I think the critical move was here: >> It is certainly the case, as Chris writes, that trials with significant >> results are more likely to be published than those with negative or >> neutral results, and this applies to medical studies in general >> (and no doubt on a wider scale in scientific research). >This is not a simple matter of the "file drawer" phenomenon. As was >revealed a few months ago, studies that did not get the "correct" >results appear to have been systematically kept from the public by the >FDA. The only reason that most of the negative studies in the Kirsch, et >al. meta-analysis appeared at all is because he used Freedom of >Information legislation to force the FDA to release them. We can, of course, have different views about what was "the critical move" (whatever that may mean) of the several points I raised in my article, but let's try to clarify what was going on. You previously wrote: >...With the recent >revelations that the published body of articles is itself a highly >biased sample of the research that has actually been conducted >(essentially, if you didn't get a positive effect, you didn't get >published), there is little doubt that the "new" anti-depressants have >a serious credibility problem now. I didn't want to make my posting interminably long (what's stopped you before? you ask) so I didn't try to dissect this in any way. Had I done so, I would have pointed out that "the published body of articles" now extends far beyond anything that the pharmaceutical companies have instigated under their own auspices, and that there are many scores of studies of the efficacy of antidepressants for which it is not simply the case that "if you didn't get a positive effect you didn't get published". It was in relation to this - that the published body of articles is far more extensive than seemed to be indicated in your comment - that I alluded to what you referred to as the "file drawer" phenomenon. In short, the problem from my point of view was that I couldn't see where you distinguished between studies initiated by the pharmaceutical companies and the published body of articles as a whole, and I short-circuited discussion of this by taking up a generalized point about published studies that I thought we would agree on. I apologise for not making this clear. > It does appear to be simply the case however, that they do not >work nearly as well, nearly as reliably, or nearly as widely as we >had been led to believe by the pharmaceutical industry and those > in the government agencies that are charged with protecting us. We can agree on that, as (I hope) on some of the kinds of remedies for the current situation as suggested by Quitkin et al (2000) quoted in my previous posting. Incidentally, I would make one point that is rarely mentioned. It not infrequently happens that when a patient does not respond to one drug the practitioner prescribes a different one, and on occasion this does produce a positive response. This kind of positive result of drug treatment will not show up on efficacy studies of single products. This relates to the wholly inadequate understanding and categorising of different presenting symptoms, alluded to an article I incorrectly titled in my previous posting: see especially last paragraph in the section "Why Change the Current Unitarian Diagnostic Paradigm?" in G. Parker (2000): "Classifying Depression: Should Paradigms Lost Be Regained?" http://ajp.psychiatryonline.org/cgi/reprint/157/8/1195.pdf Allen Esterson Former lecturer, Science Department Southwark College, London http://www.esterson.org --- To make changes to your subscription contact: Bill Southerly ([EMAIL PROTECTED])
