Antidepressants and Suicidal Ideation

TCPR: Dr. Fassler, you are a child psychiatrist and you testified at the original FDA hearings leading to the black box warnings on antidepressants. Can you give us some insight into how that process played out?

Dr. Fassler: Sure. I testified at the FDA hearings in February and September of 2004 dealing with the use of SSRI antidepres- sants. The hearings were held to address the question of whether or not these medications increase the risk of suicide in chil- dren and adolescents.

TCPR: Why did the FDA hold these hearings in the first place?

Dr. Fassler: The FDA had been receiving reports raising questions about possible suicidal thoughts and behaviors in young patients taking antidepressants. They decided to do a large scale review. They went to the pharmaceutical companies and asked for data on suicidality in children from all 23 clinical trials which had been completed. However, it quickly became clear that the limited information available would be difficult to interpret due to the varying methodologies employed. The FDA then asked researchers at Columbia University to develop a new scale and perform a retrospective review of all available information from the trials. The goal was to try and extract data on suicidal thoughts and behaviors which could then be ana- lyzed.

TCPR: What was your initial impression of the data?

Dr. Fassler: The first thing I realized was that we had never heard of most of these studies. There were 15 studies on depression and eight on anxiety. Of the 15 studies on depression, only four had ever been published: three showed efficacy, and one showed a positive trend. But what we had not known before the FDA released this data was there were also 11 unpublished studies: three showed a positive trend, and eight were negative. So there was clearly a significant problem with publication bias, meaning that as physicians were trying to keep up with the literature, and trying to understand how to best help kids, we weren’t getting a full picture of the studies that had actually been done.

TCPR: And what did these published and unpublished studies have to say about suicidality in kids?

Dr. Fassler: The 23 studies included data from 4300 children. The researchers found no significant differences between med- ication and placebo in the categories of suicidal thinking or suicidal behavior for the eight studies dealing with anxiety. Similarly, there were no significant differences in the combined 15 studies dealing with depression. However, when they com- bined the depression and anxiety studies together, children taking SSRI antidepressants were approximately twice as likely to have reported at least one episode of suicidal thinking or behavior. Specifically, the incidence among children taking medica- tion was approximately 4% while the incidence among those on placebo was approximately 2%. Fortunately, there were no deaths attributable to suicide in any of the studies reviewed.

TCPR: So how did you interpret these numbers? Do they mean that antidepressants actually cause suicidality in kids?

Dr. Fassler: I remain concerned about the accuracy of the data. Out of the 23 studies, 17 actually included systematic assess- ments of suicidal ideation, and there was no significant difference between medication and placebo in these studies. Furthermore, from a clinical perspective, the finding was puzzling since we’re well aware that suicidal thoughts are actually quite common among children and adolescents. For example, according to the most recent Youth Risk Behavior Survey, 14.5% of high school students have thought about suicide in the past 12 months. Among teenagers with depression, the inci- dence is considerably higher. The much lower rates reported in the clinical trials may be attributable to the fact that the researchers’ analysis was based primarily on spontaneous reports. It’s quite possible that medication increases the likelihood that a child or adolescent tells someone about their suicidal thoughts. And this may actually be a good thing, since it can lead to intervention.

TCPR: Nonetheless, the FDA voted to require a black box warning about an increased risk of suicidal thoughts or behaviors.

Dr. Fassler: Yes, the hearings took place in an emotionally charged environment. Parents testified, telling powerful and tragic stories about kids who had committed suicide. There was extensive media coverage. Based on the data presented at the FDA hearings, there was no evidence that these medications actually increase the risk of suicide in children and adolescents, but the media painted a different picture. For example, The New York Times reported, “FDA Links Drugs to Being Suicidal.” The headline in the Washington Post read, “FDA Confirms Antidepressants Raise Children’s Suicide Risk.” In the end, the Advisory Committee voted 15 to 8 in favor of imposing a black box warning on all antidepressants, and the FDA subsequently followed that recommendation.

TCPR: Why do you think the FDA approved the warning, given the evidence was so ambiguous?

Dr. Fassler: It’s my sense that the FDA recognized that the data on suicidal thinking and behavior was weak, at best. I believe they acted, in part, because of the relative lack of efficacy data for these medications, particularly when used in the treatment of depression. The data on use in anxiety disorders was more convincing.

TCPR: Do you think antidepressants are helpful in children?

Dr. Fassler: I think these medications can be helpful for some, but not all, children and adolescents who suffer from depres- sion or anxiety disorders. They’re generally most effective when used as a component of a comprehensive treatment plan. I also think that it’s difficult to extrapolate from short term trials to actual clinical practice, where children and adolescents typ- ically have multiple comorbidities and are often treated for a more extended period of time.

TCPR: How has the black box warning affected clinical practice?

Dr. Fassler: There was a fairly dramatic shift in practice patterns throughout the country. The FDA initially recommended a monitoring schedule that was pretty unrealistic: Once a week for 4 weeks, every two weeks for next four weeks, then at 12 weeks. This didn’t fit into reality of pediatric practice, and pediatricians and other primary care physicians became particularly reluctant to use these medications. The APA and the Academy of Child and Adolescent Psychiatry went back to the FDA and argued that there was no data to support this specific schedule, and it was eventually removed. Nonetheless, the decline in prescriptions was precipitous. Studies suggest that in 2003, 60% of kids diagnosed with depression received an SSRI, and by 2005 the figure had dropped to below 30%.

TCPR: Does this mean that these kids are being referred to therapy instead of getting medications?

Dr. Fassler: Unfortunately, no. The research actually suggests that fewer children and adolescents are being diagnosed with depression and referred for treatment of any kind. Of significant concern, we’ve also seen the first increase in the adolescent suicide rate since the early 1990s, an increase which corresponds to approximately 300 actual deaths per year, in the U.S.

TCPR: Is this caused by the decline in antidepressant prescriptions?

Dr. Fassler: It’s too early to tell if the increased suicide rate is a direct result of the reduced use of SSRI antidepressants. There are other potentially intervening variables, including access to firearms, enhanced reporting, and the overall availability of mental health services. However, the trend is clearly of significant concern. Given the information currently available, it would seem appropriate for the FDA to reconvene the Advisory Panel to review recent studies and to consider the public health consequences of the black box warnings. I think the FDA has an obligation to monitor the impact of their regulatory decisions and to make modifications, as appropriate, based on subsequent data and research findings.

TCPR: At this point, what are your monitoring recommendations for psychiatrists starting children on antidepressants?

Dr. Fassler: There’s no one schedule or approach that’s appropriate for all kids. For some, it’s best to see them weekly or biweekly until they’re relatively stable. For others, it may make more sense to see them less frequently, but to have regular telephone contact with their parents and other members of the treatment team. Ultimately, the approach to monitoring should be based on your clinical assessment and individualized to the needs of the child and family.[Ed. Note: The American Psychiatric Association and the American Academy of Child and Adolescent Psychiatry have pro- duced a free guide to help answer parents’ questions about SSRI antidepressants. The guide is available at www.ParentsMedGuide.org.]

General Psychiatry