Effects of antidepressants on existing suicidality

[hamstergang]

So antidepressants have the black box warning due to studies in the pediatric population showing an increase in the number of subjects on the active medication reporting suicidal thoughts or behaviors compared to subjects on placebo. A fascinating discussion on its own, but that's for other (already existing) threads.

I'm wondering if anyone knows of any studies examining pediatric patients who are already reporting suicidal ideations or self-injurious behaviors, and then looking at the impact of antidepressants vs placebos on the suicidality of them? Basically, if I have a 13 y.o. patient with MDD and SI, is there any evidence that starting an SSRI could be associated with this patient having an increased frequency or intensity of SI, increased chance of self-injury, or increased chance of suicide attempt/completion?

---

Members don't see this ad.

 

[FlowRate]

Moreover, did suicidal depressed patients buy more volvos than non-suicidal depressed patients?

 

[thoffen]

The answer for me for this has always been that combined med+therapy treatment ameliorates the risk of med treatment alone. I make sure my young adults get therapy as well (I don't do kids).

 

[MacDonaldTriad]

I read somewhere that the addition of the black box warning about antidepressants and kids made us so stingy with SSRI use, the completed suicide rate for kids went up for a while. I'm not a child psychiatrist, but it seems like unintended consequences happen in kids a lot.

 

[thoffen]

I read somewhere that the addition of the black box warning about antidepressants and kids made us so stingy with SSRI use, the completed suicide rate for kids went up for a while. I'm not a child psychiatrist, but it seems like unintended consequences happen in kids a lot.

There's that whole correlation causation thing to contend with here, but I think more importantly not treating depression at all is bad. We can't do placebo like they do in studies. Validation and expectation of treatment response is probably pretty important.

 

[nitemagi]

There's that whole correlation causation thing to contend with here, but I think more importantly not treating depression at all is bad. We can't do placebo like they do in studies. Validation and expectation of treatment response is probably pretty important.

It's epidemiological data, but still.

 

[DD214_DOC]

I read somewhere that the addition of the black box warning about antidepressants and kids made us so stingy with SSRI use, the completed suicide rate for kids went up for a while. I'm not a child psychiatrist, but it seems like unintended consequences happen in kids a lot.

I'm not sure about the specific statistics, but these warning labels have certainly made it more likely that primary care will not initiate or manage a pediatric depression patient on an SSRI and will instead refer to child psych, which significantly delays appropriate treatment -- sometimes for months. I would imagine this would play some part in any alleged increase in completed suicides, but that's just me making stuff up.

I, personally, have never clinically seen an increase in SI that I really believed was solely due to an SSRI. I have seen disinhibition/behavioral activation many times, though.

As mentioned above, expectation management is extremely important for not only kids and parents but also adult patients. Many patients believe they are submitting to the will of the pill and everything will be magical and wonderful on their Zoloft, so they simply sit back and wait for wonderful things to happen. I like to crush their unrealistic dreams from the very beginning and make sure they understand they still have an active role to play in their treatment.

 

[Crayola227]

is there any evidence that close follow up has any impact on negative outcomes in these scenarios? or we just hope it might because it might make sense?

 

[MacDonaldTriad]

It makes a lot of sense for the retrospectascope. My favorite is a general attitude that if we just wouldn't have discharged this person, they would still be alive. This can be stretched to reach months even. Managing our risk is what we do, but there are often no absolute right answers so bad things will continue to happen. When they do, fingers get pointed.

 

[Crayola227]

I think my point was more, if these drugs increase suicidality initially, what if any benefit, is there to warning the children and parents, and providing close f/u?

Is the conversation just ass-covering, then?

Not sure this is totally analagous to a d/c of a hopitalized patient or a scenario months out. I'm talking outpt in an acute setting.

 

[thoffen]

It makes a lot of sense for the retrospectascope. My favorite is a general attitude that if we just wouldn't have discharged this person, they would still be alive. This can be stretched to reach months even. Managing our risk is what we do, but there are often no absolute right answers so bad things will continue to happen. When they do, fingers get pointed.

I think it would be very, very difficult to convince me that there could ever be a preponderance of evidence that psychiatric care or lack thereof is responsible for any suicide. Of course, I really doubt I'll ever end up sitting on any jury, much less one for psychiatric malpractice.

 

[DD214_DOC]

I think my point was more, if these drugs increase suicidality initially, what if any benefit, is there to warning the children and parents, and providing close f/u?

Is the conversation just ass-covering, then?

Not sure this is totally analagous to a d/c of a hopitalized patient or a scenario months out. I'm talking outpt in an acute setting.

The benefit of warning and close f/u is that it's part of informed consent and risk management. Probably more, "risk of being sued", than actual risk for the patient. I am really never very concerned with the black box label, and as part of my canned IC spiel, I mention it but also make a point to tell them I've never actually seen this happen clinically.

I am, however, legimately concerned for behavioral activation, which is something I've seen several times. I have them f/u within the 2-3 week timeframe more for this reason than any other, assuming I have the availability for it.

I also reassure them that, no matter what happens -- whether the medicine works but has side effects, sort of works, or doesn't work at all -- there is always something we can do about it.

 

[PistolPete]

I'm not sure about the specific statistics, but these warning labels have certainly made it more likely that primary care will not initiate or manage a pediatric depression patient on an SSRI and will instead refer to child psych, which significantly delays appropriate treatment -- sometimes for months. I would imagine this would play some part in any alleged increase in completed suicides, but that's just me making stuff up.

I, personally, have never clinically seen an increase in SI that I really believed was solely due to an SSRI. I have seen disinhibition/behavioral activation many times, though.

As mentioned above, expectation management is extremely important for not only kids and parents but also adult patients. Many patients believe they are submitting to the will of the pill and everything will be magical and wonderful on their Zoloft, so they simply sit back and wait for wonderful things to happen. I like to crush their unrealistic dreams from the very beginning and make sure they understand they still have an active role to play in their treatment.

I love your use of “crush”! Really adds that emphasis!

 

[Merovinge]

So antidepressants have the black box warning due to studies in the pediatric population showing an increase in the number of subjects on the active medication reporting suicidal thoughts or behaviors compared to subjects on placebo. A fascinating discussion on its own, but that's for other (already existing) threads.

I'm wondering if anyone knows of any studies examining pediatric patients who are already reporting suicidal ideations or self-injurious behaviors, and then looking at the impact of antidepressants vs placebos on the suicidality of them? Basically, if I have a 13 y.o. patient with MDD and SI, is there any evidence that starting an SSRI could be associated with this patient having an increased frequency or intensity of SI, increased chance of self-injury, or increased chance of suicide attempt/completion?

Keep in mind that the "damning" data that generated the BBW even shows that when you isolate Prozac response, kids had a lower rate of SI. The fact that they aggregated data and include Paxil, which should basically never be prescribed for kids, is what skewed the whole data set. Activation is definitely a concern, and that can lead any given pt to an increased risk of self-injury or suicidal behavior, but the population as a whole that you are treated should not have an increased rate.

 

[hamstergang]

So while this discussion can certainly be interesting, I'm asking a different question than the black box warning addresses. I'm guessing based on my lack of success with searching and the lack of direct answers here that the real answer is "we don't know," but just in case: anyone know if there's any study examining the effects of antidepressants on kids who are already reporting suicidal thoughts (some statistic that excludes kids without suicidality at baseline so as to not skew the data for the population I'm interested in)?