John's Blog

The Blog of John Gibson, PhD


Antidepressants Don't Work...Can this be True?

Magazine cover

According to research published last month in THE JOURNAL OF THE AMERICAN MEDICAL ASSOCIATION, antidepressants are being challenged as an effective treatment for depression.

The cover story for NEWSWEEK (Feb 8, 2010) gives an excellent report on these new findings.

For mild, moderate, and severe depression, antidepressants are virtually no better than a placebo. However, antidepressants have been found to be effective for very severe depression.

These conclusions were reached using meta-analysis, a statistical technique that allows researchers to put the effects of any given study on a common metric so that an overall effect size can be calculated. (Don't let the jargon bog you down. Meta-analysis simply puts separate studies on the same yardstick so they can be compared.)

In this case, Irving Hirsh used 48 studies, taking pains to gather all of the studies he could find. He went so far as to use the Freedom of Information Act to obtain unpublished studies by drug companies that were available through the Food and Drug Administration.

The end result? For the most part, antidepressants are no better than a dummy pill.


The drug companies have some explaining to do. Incidentally, they are not disputing the findings of their own studies, but they are trying to point out the depression is an individualized illness and that patient responses may vary. Still, we're talking about a multi-billion dollar industry here. We're talking about big news.

No doubt, we're bound to here more on this story in the comming months. In the mean time, if you are taking an antidepressant, do not stop abruptly. This can be dangerous. Consult the person who prescribes them to talk over what's best for you.

One last point: psychotherapy has been show to effective in treating depression. And compared to antidepressants, the relapse rate is lower, too.