While I appreciate the dangers of using correlations to demonstrate a
causal relationship I think that we should consider these data
carefully. If you are suggesting that it is only a spurious relationship
then what would you propose as a logical "third variable" to account for
the apparent connection?
-Don.
Paul Brandon wrote:
At 8:02 AM -0500 12/12/06, [EMAIL PROTECTED] wrote:
Last weeks issue of Time Magazine, with cover date 14/4/06, has an
article The year in medicine A to Z. One of the entries is
DEPRESSION
Researchers still don't understand why severely depressed teenagers
are more likely than adults to commit suicide while taking
antidepressant drugs like Paxil, but a major study out of UCLA
concluded that the drugs do more good than harm. Starting in the
early 1960s, the annual U.S. suicide rate held fairly steady at 12 to
14 instances per 100,000--until 1988, when the first of a new
generation of antidepressants, the selective serotonin reuptake
inhibitors, was introduced. The suicide rate has been falling ever
since, to around 10 per 100,000. The investigators estimate that
nearly 34,000 lives have been saved.
Since much of the discussion has been focused on the fact that
antidepressants don't work, how would you explain this result?
Facetiously:
Since the time period in question coincides with the birth of the
InterNet, one could just as easily say that an increase of virtual
suicide by internet addiction is responsible.
In other words -- a correlation is still a weak argument.
Were the authors receiving drug company support by any chance?
--
Don Allen
Department of Psychology
Langara College
Vancouver, B.C., Canada
V5Y 2Z6
604-323-5871
---
To make changes to your subscription go to:
http://acsun.frostburg.edu/cgi-bin/lyris.pl?enter=tips&text_mode=0&lang=english
