A TIPSter has kindly pointed out that the supposed URL I gave for the 
article below was actually a link to a file on my PC! It was a file 
copied from my previous PC, and appears with the file link I cited on 
the new PC. Evidently I saved the article when it was online some years 
ago, but it is no longer available, so here is the full reference:

Prevention & Treatment, Volume 1, Article 0006c, posted June 26, 1998
Copyright 1998 by the American Psychological Association

"Listening to Meta-Analysis but Hearing Bias"
Donald F. Klein, M.D.
Department of Psychiatry, Columbia University,
Department of Therapeutics, College of Physicians and Surgeons, and the 
New York State Psychiatric Institute

Allen E.
------------------------------------------------------------------
Re:[tips] Clinical training: Boulder and Denver
Allen Esterson
Wed, 14 Sep 2011 02:34:40 -0700

The URL for "Listening to Meta-Analysis but Hearing Bias" didn't come
up as a link in my last post, but can be accessed by copy-and-paste
into Google. (TInyURL doesn't work, probably because the URL doesn't
start with http or www.)

file:///C:/Users/dell/Documents/C%20Files/MISC%20PSYCHIATRY/Drugs%20and%2

0psychiatry/KleinListening%20to%20Meta-Analysis%20But%20Hearing%20Bias.ht

m

Abstract

Kirsch and Sapirstein (1998) present a meta-analysis of 19 studies,
attempting to define the relationship of placebo to antidepressant drug
effect. They conclude that the substantial majority of drug effect is
due to placebo effect and the rest is either measurement error or
active placebo effect. The article is criticized because it derives
 from a miniscule group of unrepresentative, inconsistently and
erroneously selected articles arbitrarily analyzed by an obscure,
misleading effect size. Further, numerous problems with the
meta-analytic approach, in general, and Kirsch and Sapirstein's use of
it, in particular, go undiscussed. The attempt to further segment the
placebo response, by reference to psychotherapy trials incorporating
waiting lists, is confounded by disparate samples, despite Kirsch and
Sapirstein's claim of similarity. The failure of peer review and the
opportunity provided by an electronic journal for rapid discussion is
emphasized.

[…]

A major problem with meta-analysis is that the appearance of
statistical rigor can lull the usual reader, who cannot be expected to
retrieve and analyze the original sources and is not statistically
expert, into an uncritical, complacent mode (especially if the
conclusions are congenial). This suspension of disbelief is encouraged
by the conviction that peer review has carefully vetted both the source
and meta-analytic articles for misstatements, distortions, poor
inferences, and statistical malfeasances. Kirsch and Sapirstein (1998)
provide a trenchant example of a tendentious article whose departures
 from any critical standard has not precluded publication and has been
foisted on an unsuspecting audience as a "peer reviewed" contribution
to the literature. For instance, it seems evident that these peer
reviewers did not retrieve and critically review these articles. As for
some of the source articles, the adequacy of peer review is clearly
questionable.


N.B. A useful discussion of the issues involved appeared in "Scientific
Review of Mental Health Practice" in 2003:

http://www.srmhp.org/0201/media-watch.html

Allen Esterson
Former lecturer, Science Department
Southwark College, London
[email protected]
http://www.esterson.org



---
You are currently subscribed to tips as: [email protected].
To unsubscribe click here: 
http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=12647
or send a blank email to 
leave-12647-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu

Reply via email to