Posted by Larry Hoover on May 9, 2006, at 13:56:01
In reply to Statistical question on SSRIs, posted by Squiggles on May 6, 2006, at 6:54:51
> Would anyone here know the statistical
> rates of AD-caused suicides, comparing
> different classes of drugs; or where I
> could find studies in such assertions?
> I am searching for evidence regarding
> the proposed unique ability of SSRIs to
> cause suicide, in exclusion of other
> causes and other classes of antidepressants.
>
> Thanks
>
> SquigglesI don't know of any such evidence. Clinical trials are not of much use, as the variables are controlled in such a way as to try to eliminate such an effect, if it existed.
What we need are studies of whole populations. Naturalistic or observational or ecological studies. There are very few studies which even try to answer that question. There was one issue of BMJ (I think) that was dedicated to articles attempting to answer that question, just a few months ago. As I recall, there was no such signal found. Healy, and a bunch of other "names", they all took a look, and found nothing to support that hypothesis. {with very specific exceptions, not germane to this review}
One of the limitations of studying populations is that you can't determine which independent variables are truly responsible for your supposedly dependent measurements. You must assume that you know what you're doing, I suppose.
If we refer to epidemiological findings, we may yet have the answer to your question. In the following study, the entire population of a country (Sweden) was being studied. In that study population, we presumably include all manner of people. Those at high risk for suicide, and those with low risk. People being given SSRIs for depression, but also those being given SSRIs who have comorbid conditions (generally excluded from clinical trials), those being given the drugs "off-label". But also, we include untreated depressed people not in contact with medical support, and so on. We include the lost souls, too. There are no restrictions on the study population, other than that they are all Swedes.
What was found was that there was a "significant change in slope" (a reduction) of the suicide rate, following the introduction of SSRI meds. A change in slope can only be caused by a change in the independant variables. An independent variable was added, changed, or removed. In this case, the changed variable was (assumed to be) the introduction of SSRI meds. It is problematic to assign the nomenclature of experimental design to ecological data, so it may be more accurate to refer to the introduction of SSRIs as being a predictor variable, or a factor, rather than a true independent variable.
In real life, not in the artificial environment of a clinical trial, suicides decreased significantly when SSRIs became available in Sweden. It remains a possibility that some other independent variable or factor also changed at the same time as SSRIs were introduced, and that it is a coincidental finding to see the suicide rate change like this. Even with such a coincidence, though, we can still confidently say that SSRIs did not increase the rate of suicide in Sweden. As sales (and presumably consumption) increased, the suicide rate did not.
Pharmacoepidemiol Drug Saf. 2001 Oct-Nov;10(6):525-30.
Antidepressant medication and suicide in Sweden.
Carlsten A, Waern M, Ekedahl A, Ranstam J.
Department of Social Medicine, University of Goteborg, Sweden.
anders.carlsten@telia.comOBJECTIVE: To explore a possible temporal association between changes in antidepressant sales and suicide rates in different age groups. METHODS: A time series analysis using a two-slope model to compare suicide rates in Sweden before and after introduction of the selective serotonin reuptake inhibitors, SSRIs. RESULTS: Antidepressant sales increased between 1977-1979 and 1995-1997 in men from 4.2 defined daily doses per 1000 inhabitants and day (DDD/t.i.d) to 21.8 and in women from 8.8 to 42.4. Antidepressant sales were twice as high in the elderly as in the 25-44-year-olds and eight times that in the 15-24-year-olds. During the same time period suicide rates decreased in men from 48.2 to 33.3 per 10(5) inhabitants/year and in women from 20.3 to 13.4. There was significant change in the slope in suicide rates after the introduction of the SSRI, for both men and women, which corresponds to approximately 348 fewer suicides during 1990-1997. Half of these 'saved lives' occurred among young adults. CONCLUSION: We demonstrate a statistically significant change in slope in suicide rates in men and women that coincided with the introduction of the SSRI antidepressants in Sweden. This change preceded the exponential increase in antidepressant sales.
You can never prove the absence of something. You can't prove what didn't happen. However, we can show that there was not a population-wide increase in completed suicides that can be attributed to SSRI medication. We see no such signal in a broad population. One that is completely and thoroughly documented by its government. For whatever reason, these people write everything down. That provides us with an historical record of a population as it undergoes changes. And when we look at that record, we find no evidence for your hypothesized effect.
One possible explanation for that failure, though, is that the observed suicides are actually all SSRI-induced suicides. That would be very difficult to demonstrate unambiguously. That we have all these new and induced suicides, nested into the background rate, and yet the rate itself has not increased proportionately. Another possible limitation to this study is that Swedes might not be like other people.
I don't see the signal you seek evidence for. I've looked, and I can't find it.
That does not invalidate anecdote. That there are unambiguous cases of SSRI-induced suicide is not something that I am trying to refute. The evidence suggests that it is a fairly uncommon occurrence.
I'd tell you if I knew of the evidence you seek. I've looked, and I've looked hard. I can't find it anywhere.
Lar
poster:Larry Hoover
thread:640557
URL: http://www.dr-bob.org/babble/20060504/msgs/641794.html