Did the FDA antidepressants warnings increased suicidality?

A new study just published in the BMJ makes this very point.

bmj-logo-ogChanges in antidepressant use by young people and suicidal behavior after FDA warnings and media coverage: quasi-experimental study

The title makes two claims:

  1. That the  2003 black box  US Food and Drug Administration warning about a possible increased risk of suicidality with antidepressants lead to a change in antidepressant prescriptions.
  2. Further, that there were changes in suicidal behavior following the said US FDAwarning.

While the title does not indicate the direction of the changes, the study conclusions carry little ambiguity:

Safety warnings about antidepressants and widespread media coverage decreased antidepressant use, and there were simultaneous increases in suicide attempts among young people.

This is obviously a big claim, with far reaching implications spreading from the level of primary care physicians who might consider changing their antidepressant prescription practices to the level of policy makers deciding on guidelines about antidepressants approvals and reimbursement.

In this post we will discuss some of the study limitations which, for unclear reasons, seem to have been all but ignored by the over-excited welcome that study received in the mainstream media.

Our goal here is to not discuss subtle academic limitations but rather obvious limitations which could have been picked even by a casual yet critical reader of the paper. We will then cursorily survey the media presentations of the paper and assess the quality of their review in terms of balance and fair criticism.

Let’s start with some of the study most overt shortcomings:

  1. The title is not entirely informative. The study reviews US data however the title seems to indicate world-wide findings.
  2. The study uses a quasi-experimental design assessing changes in outcomes after the FDA  warnings, controlling for pre-existing trends. However the reliability of the controls is not entirely convincing.
  3. Not much is known about the relationship (if any) between the individuals entered in the study because they were prescribed antidepressants and the individuals who took psychotropic overdoses.
  4. The study uses psychotropic drug poising data as a proxy for suicide but is that a good proxi?
  5. The study does not address the possibility that suicide has been on the raise for reasons having nothing to do with antidepressant prescriptions (such as the recent years economic crisis).

Big media enthusiastic welcome


First the Washington Post (Dennis) reports:

As a result [of the FDA warnings] antidepressant prescriptions fell sharply for adolescents age 10 to 17 and for young adults age 18 to 29. At the same time, researchers found that the number of suicide attempts rose by more that 20 percent in adolescents and by more than a third in young adults.

Comment from a non-study affiliated expert? YES

How many limitations are discussed? NONE

According to NBC News (Raymond) :

New research finds the warning backfired, causing an increase in suicide attempts by teens and young adults. After the FDA advisories and final black box warning that was issued in October 2004 and the media coverage surrounding this issue, the use of antidepressants in young people dropped by up to 31 percent.

Comment from a non-study affiliated expert? NO

How many limitations are discussed? NONE

Reuters (Seaman) reports that:

Antidepressant use decreased by 31 percent among adolescents, about 24 percent among young adults and about 15 percent among adults after the warnings were issued. At the same time, there were increases in the number of adolescents and young adults receiving medical attention for overdosing on psychiatric medicines, which the authors say is an accurate way to measure suicide attempts. Those poisoning increased by about 22 percent among adolescents and about 34 percent among young adults after the warnings. That translates to two additional poisoning per 100,000 adolescents and four more poisoning per 1,000 young adults, the researchers write.

Comment from a non-study affiliated expert? YES

How many limitations are discussed? ONE (limitation #5)

According to USA TODAY (Painter):

Warnings that antidepressant medications might prompt suicidal thinking in some young people may have backfired, resulting in more suicide attempts, new research suggests.

Comment from a non-study affiliated expert? YES

How many limitations are discussed? NONE

Forbes (DiSalvo) reports:

Antidepressant use fell 31 percent among adolescents and 24 percent among young adults after the FDA warnings, according to the study. Suicide attempts increased by almost 22 percent among adolescents and 33 percent among young adults in the same time period. Suicide attempts tracked in the study were largely the result of drug overdoses.

Comment from a non-study affiliated expert? NO

How many limitations are discussed? NONE

NPR (Stein) also reports on the story:

Antidepressant use nationally fell 31 percent among adolescents and 24 percent among young adults, the researchers reported. Suicide attempts increased by almost 22 percent among adolescents and 33 percent among young adults.

Comment from a non-study affiliated expert? YES

How many limitations are discussed? ONE (limitation #5)

Finally, the Boston Globe (Freyer) concludes that

instead of declining as hoped, suicide attempts over the next six years showed a “small but meaningful” uptick among people ages 10 to 29, according to a study published Wednesday in the journal BMJ. That increase followed a substantial drop in the use of antidepressants.

Comment from a non-study affiliated expert? YES

How many limitations are discussed? NONE

A few observations


To summarize: big media’s portrayal of the situation is straightforward. The FDA black box warnings were followed buy an almost 30% decrease in antidepressant use and an almost 30% increase in suicide rates for adolescents and young adults.The conclusion, while not stated directly, is implicit: antidepressants decrease suicide in this population.

The issue is that this conclusion is at odds with what the study really found:

1. The decrease in antidepressant prescriptions is not absolute but relative to a projected forward prediction of increased antidepressant use over the studied period. All is good if the prediction is accurate – but there in no way of actually assessing how correct these sort of predictions really are.

2. There is no absolute increase in suicides either. The increase rate is not for suicide for for psychiatric drugs poisoning, which might be an entirely different beast than suicide. Even if accepting the authors’ decision to use psychotropics poisoning as proxy for suicide the reported increase is again not absolute but relative to a predicted trajectory of in fact unknown accuracy for psychotropics poisoning.

3. Or maybe the suicide rate was on the raise anyway for reasons that have nothing to do with the FDA warnings or antidepressant prescriptions.  As discussed above the economic crisis of the last few years might have independently contributed to this state of affairs.

Which leads us to an interesting conclusion

Somewhat paradoxically, it turns out that a study that tongue-in-cheek points to the media’s uncritical coverage of medical news as a possible contributor to a public health issue receives an almost universal and equally uncritical acclaim from the same media that it rightfully criticizes.

In a past analysis of medical news reporting I stated that

[…] my hope is that members of the media who cover [medical] debate[s] will realize that “first do no harm” is not only the duty of physicians; it is also the responsibility of anyone trusted with giving health information to the public at large.

More than 2 years later I found that this conclusion still stands.

For my Storify discussion of this story – I’ll see you here.


4 thoughts on “Did the FDA antidepressants warnings increased suicidality?”

  1. In what population did the suicide attempts increase — those who kept taking the drugs (or started taking them) or those who stopped taking the drugs (or never started them)? It may well be that the increase in suicide attempts occurred for the population who took the drugs, as they may have been “inspired” by the warning label or felt that they had been given an excuse that could deflect blame and shame (i.e., it was the drugs not me). Alternatively, if people turned to psychotherapy/talk therapy instead, causing them to ruminate on their personal problems and critique relationships with family/partners/friends, then that could also account for the increase in suicide. Psychotherapy can be emotionally damaging, but biased researchers in the field are neglectful about tracking and reporting harm: http://www.ncbi.nlm.nih.gov/pubmed/24607768


  2. @Discussant: Good points. It is not clear if antidepressants were never started or stopped; also not clear if those ending with a psychotropic overdose were actually part of the not starting/stopping antidepressant population.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s