Malgus, with all due respect (and I’ve developed a considerable amount of that for you in my relatively short time here), I can only say, “Wow! Where to start?” I apologize up front to you and the group for the length of this response, but given what you posted, I think it’s necessary. I’ve at least broken it into two posts, as explained at the bottom (and in the 2nd post).
First, you’re breaking no “bad news” to me at all. While I’m retired and no longer reading extensive journal articles such as this one, nothing in this study has me particularly surprised. You quoted the line that says, “The increased risk we found in young people needs validation in other studies.” Yet as part of the same sentence that contained that quote in your response, you concluded, ” this pretty much clinches it for me.”
I don’t mean this as a personal attack, just a statement of what appears to be a pretty solid conclusion. They say, “We don’t have enough information here to call this conclusive,” yet you disregard that and say you’re convinced by their extensive analysis that didn’t convince them. I don’t have an logical argument for that.
I have no way of knowing what your background in statistical analysis is, but I’ll admit that while mine used to be substantially solid, it’s almost completely rusted out with age and disuse. That doesn’t mean, however, that I haven’t retained a small fraction of what I once knew sufficiently well to get an A+ in statistics in graduate school, when very few even got an “A” of any variety in that course from the hated department chairman. I don’t know why I felt very comfortable with that course while others that I considered at least my equals or better intellectually, struggled mightily with it. But that just happened to be the way it was.
So anyway, with that vague, highly rusted knowledge, I did manage to see some things that are important to understand. First, and foremost, is the fact that the ONLY age group for which a level of concern should logically exist, is the 15-24 year age group. Everybody older than that comes in at statistically insignificant probabilities that the results were anything other than pure chance.
Many may not follow all the letters and numbers in that abstract you quoted. I admit I had to reach really far down into aging memory to remember what some of it was saying for certain, such as “CI” (“confidence intervals”). I even pulled out my old copy of my grad school stats text book just to do a quick brush up. But out of all that, one number is very, very important to understand. It’s the “p”robability number – the number that reads “p < 0.001″ or other levels depending on the age range. That particular number (p < 0.001) basically means that the “p”robability of the results being purely by chance is less than 1 in 1000. That’s considered a highly statistically significant result (generally, no study would consider results significant with levels worse than p < 0.05, meaning only a 5 in 100 chance the results were due to random chance). All results for age ranges higher than 24 are considered statistically insignificant in this study.
Translated for any that aren’t familiar with that nomenclature, the probability of the results being due simply to random chance are 1 in 1000 for the 15-24 age group, but the probability increases for the 25-34 age group to a 12½ % chance that the results were simply due to random chance. Then moving into the 35-44 age range, the chance that the results were just random were a full 67% (666 chances out of 1000). And in the “old folks” (45 and up), it was still extremely highly likely that the results were random chance – slightly over 50/50 (59.4% chance).
Therefore, the use of SSRIs – IF they are really a problem – is only a problem for folks under age 24. And just how high a problem does it appear to POSSIBLY be, according to the study? It says the “absolute risk” is still only 3%. So if I’m understanding their use of the term “absolute risk” correctly, they’re still saying that the study INDICATES that 97 out of 100 people in the 15-24 year age range are STILL not going to be affected by SSRIs to an extent that causes them to go out and shoot up a classroom. But there’s even more important information right on the main study page from which you quoted that is of great significance:
What Do These Findings Mean?
These findings show an association between SSRIs and violent crime that varies by age group. They cannot, however, prove that taking SSRIs actually causes an increase in violent crime among young people because the analytical approach used does not fully account for time-varying risk factors such as symptom severity or alcohol misuse that might affect an individual’s risk of committing a violent crime (residual confounding). In addition, some people who committed a violent crime might have subsequently taken SSRIs to cope with the anxiety and stress of arrest (reverse causation). The lack of a significant association between SSRIs and violent crime among most people taking SSRIs is reassuring; the association between violent crimes and SSRIs among individuals younger than 25 years is worrying. However, this finding needs confirming in studies with other designs undertaken in other settings. If confirmed, warnings about the increased risk of violent behavior among young people when being treated with SSRIs might be needed. But, note the researchers, it might be inappropriate to restrict the use of SSRIs in this age group because increases in adverse outcomes associated with poorly treated depression, such as suicide, might outweigh the public health benefit accruing from decreases in violence.
Note that very last sentence: even these researchers say it may create more problem than it solves by not giving these meds to that age group. Given the high number of kids taking them, you might actually have a higher rate of suicides if they WEREN’T taking them, making the number of deaths higher for unmedicated suicides than the number of medicated classroom assassinations. Tell me: would you rather have more suicides, or more murders? Is either group of victims any less valuable as human beings than the other?
Because of the length of this response, I’ll
attack (hee hee – just kidding – couldn’t resist) discuss the other part of your response in a separate post in a moment. The bottom line here, particularly if you read the last large block quote from the same study, is that even the researchers weren’t comfortable enough with their own results to draw solid conclusions from it, and they even went further to point out that the alternative to the medications (i.e. no meds) in many cases might cause MORE deaths than the possible violence induced by the meds themselves.
So my conclusion stands firm. If the vast majority of people on SSRIs aren’t even statistically likely to commit more violence than anybody else over age 24, and those between 15 -24 are only slightly more likely to do so, then that leaves a massive number of SSRI users that are simply not the risk, and therefore should not be the “target” of focus. It’s a liberal smokescreen to keep us from looking at the REAL issue. We have a Constitution that says people have the right to carry guns if they so choose. The anti-gunners are looking for any excuse – however invalid it may really be – to restrict the number of people “allowed” to have that freedom to the absolute minimum number short of deleting the 2nd Amendment altogether (and they’d love to do that if they could).
I also reiterate that I am exceptionally concerned about the degree to which SSRIs are passed out, and even more concerned about the horrid lack of proper patient management and education by the people who prescribe them or otherwise work with those patients. It’s the PRACTITIONERS that ought to be looked at more than the drugs themselves. And I say that as a licensed professional within that community!
Stand by for my comments on your completely inappropriate linking of SSRIs and medications for ADD/ADHD, and the discussion you posted around that linking.