Well, I just spent fifteen minutes trying to find a single peer-reviewed scientific article by Ann Blake Tracy to no avail. So when she talks about her “research,” it seems pretty clear that she means “opinions.”
I actually find it really frustrating when people try to dismiss scientific findings, especially with the argument “correlation is not causation.” The people who make these statements almost universally do not understand statistics and do not understand the literature they criticize, if they even bothered to read it.
That said, I think what you have here is a brilliant, shining, perfect example of a situation in which correlation is not causation. I would be hard-pressed to find a better real-life example of two things that are highly correlated that almost certainly do not have the causal relationship suggested.
Here’s why: suppose you have a random group of people who are suicidal, and a random group of people who are not suicidal. If you then survey those two groups of people to see how many are taking antidepressants, which group of people do you think has a higher rate of antidepressant use?
The suicidal people, obviously. The suicidal people are taking antidepressants because they are suicidal. They are not suicidal because they are taking antidepressants.
So you might be asking, well, nikipedia, how do you know the directionality of the causation? There’s a simple answer: science! If you properly design your experiment, you can actually begin to draw conclusions based on data and statistics rather than… opinions!
One way to properly control for this might be to survey rates of suicidality in people with depression who are taking a placebo versus antidepressants. That study was performed in 2000 by Khan et al. and published in the Archives of General Psychiatry, and again in 2003 in the American Journal of Psychiatry. Another way to study this might be to see if, over time, as more people take antidepressants, rates of suicide go down. That study was performed in 2003 by Olfson et al. and was also published in the Archives of General Psychiatry, and again by a different group in Australia that published their findings in BMJ.
So no, I don’t think there is any reason to believe antidepressant medications of any kind increase the risk of suicide, and SSRIs are the very least likely culprit. The opinions of a woman whose myspace page is in her top 3 google hits are not convincing.
What convinces me that we have no real cause for concern is not the FDA’s ruling one way or another, but the decades of research performed by hundreds of scientists on thousands of research participants around the world.