Miguel Roig wrote: > A study by Irving Kirsch and others on the efficacy of anti-depressants > was recently discussed in TIPS. Those of you who were following that > thread may be interested in reading an editorial in the most recent > issue of BMJ in which the authors,Erick Turner and Robert Rosenthal, > state "In contrast to Kirsch and colleagues´ conclusion that > antidepressants are ineffective, we concluded that each drug was > superior to placebo". The paper is available for free at: > http://www.bmj.com/cgi/content/full/336/7643/516.
Chris Green replied: >Yes, that is a direct quotation, but reading the rest of the item, it is > really a debate about the proper interpretation of effect sizes -- the > effect size measures coming from the two studies were nearly > identical. T&R go on to say that the amount of "superiority" over > placebo was quite small, even by their lights. So we return to the > question of whether the actual effectiveness of SSRIs really warranted > all the hype (and cost!) that attended (and still attends) them<snip> Yes, both studies actually are very close. Turner and Rosenthal say that in their report they found an effect size of 0.31 while Kirsch et al reported 0.32, a negligble difference. The disagreement is what interpretation to put on an effect size of that allegedly "medium" magnitude. Even so, the interpretations are not that different. While T & R did find that anti-depressants were superior to placebo, they also said "Our main finding was that antidepressant drugs are much less effective than is apparent from journal articles". Here are a few more things to throw in the mix. 1) The data set in both cases was rather limited, consisting only of FDA trials submitted for the licensing of the drugs. As the drugs in question were approved between 1987 and 1999, it means that all of the trials are rather old They also were short, most only six weeks long, as is usually for licensing trials. Also, the type of patients and the severity of their condition for licensing drug trials are not necessarily the same as for patients presenting in a typical clinical practice. Now that there is much more experience with the use of these drugs, it would be of interest to see what more recent and long-term studies are reporting on their effectiveness. 2) The main concern is that these drugs may be little more effective than placebo. This is an important issue, particularly as it relates to an understanding of the physiological mechanism involved. For example, if a placebo works just as well, then the effectiveness of an SSRI can hardly be explained as due to its ability to selectively suppress the re-uptake of serotonin. But the depressed person couldn't care less about that. All he or she wants is to feel better. And it seems that anti-depressants do a good job in doing that. Kirsch et al, for example, give an effect size change of 1.24 for drug and 0.92 for placebo, which they characterize as "both of extremely large magnitude according to conventional standards.". So anti- depressants do make people feel better, which is what they're supposed to do, but it may be largely a placebo effect. Is this so bad? Given what we've just learned about higher-priced placebos working better than cheap ones, perhaps we should be thanking drug companies for ripping us off! 3) There's one further problem here. Change scores for placebos don't show purely the power of the placebo. As some depression is short-term and goes away on its own as people find their own ways of dealing with their problems, placebo effects are inflated by effects due merely to the passage of time. To truly evaluate the effect of placebos, expensive or otherwise, we really need information on a group given no treatment at all, a suggestion I believe we heard from Michael S. not too long ago. Unfortunately, I imagine ethical considerations would make this data now impossible to obtain. Stephen ----------------------------------------------------------------- Stephen L. Black, Ph.D. Professor of Psychology, Emeritus Bishop's University e-mail: [EMAIL PROTECTED] 2600 College St. Sherbrooke QC J1M 1Z7 Canada Subscribe to discussion list (TIPS) for the teaching of psychology at http://flightline.highline.edu/sfrantz/tips/ ----------------------------------------------------------------------- --- To make changes to your subscription contact: Bill Southerly ([EMAIL PROTECTED])
