Don't Panic Folks. Antidepressants Still Do Work. 2/28/08

http://www.medscape.com/viewarticle/568939_print <--- That is the link to the print-friendly version (easier on the eyes to read) of an article regarding the literature review in question that has been popularly and erroneously reported by the press with inflammatory headlines such as "Gasp! Oh noes! Anti-depressants do not work."

[N.B. http://www.bugmenot.com/view/www.medscape.com will yield several user names and passwords that one can utilize to view the article.]

According to the Medscape article Study Identifies Bias in Favor of Publishing Positive Antidepressant Trials written by Marlene Busko, "The team identified the phase 2 and 3 clinical-trial programs for 12 antidepressants approved by the FDA between 1987 and 2004, which involved 12,564 adult patients. They also determined whether the FDA judged the studies to be positive or negative with respect to primary end points."

Actually, the literature review (which was published in The New England Journal of Medicine on January 17, 2008) maintains that S.S.R.I.s are more effective than placebos; however not so much more effective as the public has been led to believe. In other words, this review says yeah the stuff works but maybe the difference between the drugs and the placebos are not as great as the public has been led to believe.

a).The FDA uses the data that is favorable to the purposes of the FDA (in deciding to license a drug for use).
1). Negative results are sometimes used by the FDA as showing positive results.
2). Negative results are sometimes ignored by the FDA.

b). A bias may exist in reporting or people do not always have access to negative research study results.
1). Peer review journals may not accept negative research studies' results for publication at a similar rate to the acceptance of positive research studies results.
2). Some researchers for whatever reasons may not submit those studies.

In terms of the numbers of test subjects, in many cases more people were used in the discounted research studies than in some of the studies that were reported/factored into the FDA decisions to license a drug. The discounted studies often had a higher reliability factor than the ones that the FDA based their licensing decisions on.

This particular study was a review of the literature available rather than its' own research study using . The reviewers found that a bias exists in the public reporting of results rather than that the "antidepressants don't work" as incorrectly reported by the popular press.



The team was quoted in the Busko Medscape article as saying:
"Each drug, when submitted to a meta-analysis, was superior to placebo. On the other hand, the true magnitude of each drug's superiority to placebo was less than a diligent review of the literature would indicate."

There was no mention in the original literature review of the distinction between those who are clinically depressed and those who are experiencing situational depression. The purpose of the original literature review was not to take a side on the shrinks rule and N.A.M.I. mommies know best vs. the anti-psychiatry debate.

The reporting of this literature review by the popular media (e.g. newspapers) has been absolutely horrible and inaccurate.


radical sapphoq

cross-posted to some other places.