Monday, December 11, 2006

The Lastest On Antidepressants and Suicide


[posted by dinah, not roy]

Don't forget to listen to the MY THREE SHRINKS Podcast! See link below.

Do SSRI's cause people to become suicidal? The question feels old-- I remember when these medications first came out and there were questions about whether the medications made people violent, seems that years later we still have the same questions.

It seems like this is something we should know-- it's been a while now, two decades in fact. It's easy if everyone who takes a medication gets an unusual symptom, harder if only a few people who take a medication have an adverse reaction, and harder still if the symptom caused by the medication is the same as the symptom caused by the disease the medicine treats!

In 2004, the FDA mandated that all the newer anti-depressants carry a black box warning stating that they may cause suicidal ideation in children and adolescents. The research is convincing that a small percentage of children (1 to 2 percent) who were not having ideas about suicide before they started medications, had them after they started, generally in the first weeks of treatment. No child in any study died of suicide, though this is such a rare event that it gets difficult to look at prospectively. Sorry, no links here, I've just heard a bunch of talks. Most recently (meaning last week) I heard Mark Riddle, the Chairman of Child Psychiatry at Johns Hopkins Hospital talk about treating adolescent suicide attempters: he noted that in any given year, 2.9 percent of adolescents have a suicide attempt requiring medical treatment. Think about this, it's a general population number: in a high school of 1000 kids, 29 will have a suicide attempt requiring medical intervention, many more will have suicide gestures and not get help. Completed suicides? 6 to 8 per 100,000 .... a rare event, but given that kiddy death is pretty rare, a significant cause of childhood mortality.

This coming week, the FDA will hear testimony about whether the Black Box warning should also include adults. See:
http://www.fda.gov/ohrms/dockets/ac/06/briefing/2006-4272b1-index.htm and press the link for the brief if you'd like to read all 140 pages in a pdf file. What will a Black Box warning mean? In my psychiatric practice, I don't think it will mean anything. Many of my patients feel helped by these medications. It may make some patients with depression less willing to try medicines, and more importantly, it may make some primary care docs afraid to prescribe them. I have to wonder why some symptoms get the dreaded Black Box warning and other's don't and why the designation has become so heated. Another post for another day.

So last week, the Archive of General Psychiatry published this
amazing study on a followup of over 15,000 patients in Finland. It included all patients admitted to Finnish hospitals for suicide attempts over a 7 year period (excluding those with psychosis) and followed their future behaviors noting whether or not they took medications and which medications they took. I tried to follow the charts and the data, but it was too confusing and too overwhelming. Please, if anyone out there could follow these statistics, please help me. I was left to just read the results and the conclusions and some of the thinking about it all. This study, however, is terrific in that in includes everyone in the country who attempted suicide, and they tracked whether the patients filled their prescriptions, so they had a fairly good idea of whether the patients were actually taking them.

And the findings? People taking SSRI's/SNRI's (eg effexor) had a markedly higher rate of serious suicide attempts. Now this could be because the people given/ or taking the medications were sicker-- there's not necessarily a cause and effect here. Furthermore people taking SSRI's/SNRI's had a markedly lower rate of both completed suicide and death from cardiovascular disease (Hey, didn't Roy talk about this on our podcast???). Some of the numbers surprised me: in their avg 3.4 year follow up of 15,390 , there were 1583 deaths-- could it be that nearly 10% of their suicidal patients died? 602 were suicides. The average age of their population was just under 39. The protective factor for cardiovascular mortality was huge: 30-40% reduction in deaths.

Other interesting facts: Paroxetine (paxil) was associated with a high mortality among the 10-19 age group with 4 deaths: 1 suicide, 1 drowning, 2 unintentional injuries. Venlafaxine was the only medicine associated with increased risk of suicide and Fluoxetine (prozac) was the only one associated with decreased risk of suicide. And I didn't see bupropion (wellbutrin) mentioned anywhere at all in the article. The strongest predictor of completed suicide was number of past attempts.