Controversy Over Suicide Risk in Children and Adolescents Taking Antidepressants: Lessons Learned

Jay M. Pomerantz, MD



Whether certain antidepressants are effective in the pediatric population—or whether they make some children and adolescents potentially suicidal—is the subject of controversy sparked by 2 recent British reviews of prior studies. In the first study, Whittington and colleagues[1] undertook a meta-analysis of data from randomized controlled trials that compared an SSRI with placebo in participants aged 5 to 18 years. The researchers looked at results published in peer-reviewed journals and compared those results with unpublished data on file with the United Kingdom's Committee on the Safety of Medicines.

In the analysis by Whittington and coauthors,[1] results of 2 published trials suggested that fluoxetine had a favorable risk-benefit profile, and unpublished data supported this finding. Published findings from 1 trial of paroxetine and 2 trials of sertraline suggested equivocal or weakly positive risk-benefit profiles. In both cases, the addition of unpublished data indicated that the risks associated with the prescribing of paroxetine and sertraline outweighed the benefits. Data from unpublished trials of citalopram and venlafaxine also showed unfavorable risk-benefit profiles.

In the second review, Jureidini and associates[2] looked at the collective results from 6 studies involving 941 children who were assigned to receive fluoxetine, paroxetine, sertraline, venlafaxine, or placebo. Effectiveness of the antidepressant was determined only by doctors' ratings, as contrasted with 10 other measures that were based on reports from patients or parents. Serious adverse events were more common among active drug recipients than among those receiving placebo.

In all of these data from pediatric clinical trials of SSRIs and related antidepressants, it is important to remember that we are not talking about completed suicides. Fortunately, suicide is a relatively rare event, and there were no reported suicides among any of the 4100 children and adolescents enrolled in the trials.[3]

In anticipation of the publication of these reports, the FDA issued a Public Health Advisory on March 22, 2004, cautioning physicians, their patients, and patients' families about the need to monitor closely all persons being treated with antidepressants.[4] The FDA noted an association between the use of 10 commonly prescribed antidepressants—buproprion, citalopram, fluoxetine, fluvoxamine, mirtazapine, nefazodone, paroxetine, sertraline, escitalopram, and venlafaxine—and an increase in symptoms of anxiety, agitation, panic attacks, insomnia, irritability, hostility, impulsivity, akathisia, suicidality, hypomania, and mania.

More recently, Jick and associates[5] published the results of a case-control study that included 2 SSRIs (fluoxetine and paroxetine) and a tricyclic (amitriptyline); all 3 antidepressants were compared with the reference anti depressant (dothiepin, also a tricyclic antidepressant and widely used in Europe). The analysis used data from the UK General Practice Research Database, a highly accurate, detailed, and complete body of medical information on more than 3 million persons; data had been collected continuously since 1987.

Using information about patients treated between 1993 and 1999, Jick and colleagues[5] concluded that there was no substantial difference in effect of the 4 drugs on persons aged 10 to 19 years with regard to the risk of suicidal behavior. For each of the drugs, both in children and adults of all ages, suicidal behavior was more common within a few days to a month of starting the antidepressant than later on. The authors thought the most likely explanation of this finding is that antidepressant treatment is not immediately effective and that the medication may be prescribed just when the patient's symptoms and suicide risk are greatest.

The final verdict is not yet in on the newer antidepressants and the appropriateness of their use in children. It may be that these medications provide benefits for some children and not others. One problem related to the prescription of newer antidepressants is inadequate monitoring of patients by physicians. It is possible that partial or early response to medication might result in some patients' energy levels increasing before their suicidal ideation clears, leaving them able to think about or enact a suicide plan, whereas before they were too incapacitated to do much of anything. Undiagnosed bipolar illness may also be a culprit. Prescribing antidepressants for bipolar patients, if they are not concurrently taking mood stabilizers (eg, lithium, valproate), can induce mania or cycle acceleration and otherwise increase symptoms.[6,7]

More data are needed before one can know whether the newer antidepressants are effective or ineffective or whether they cause harm in children and adolescents when appropriately prescribed and monitored. That being said, there already have been some lessons learned from the controversy.

The first lesson is that the medical community finally understands that industry research support (without third-party oversight) has undermined the fairness and completeness of the data available to clinicians and regulators. The antidepressant controversy has led to the demand that data from all clinical trials, not only positive ones, be made public. The AMA has approved a policy to urge the US Department of Health and Human Services to develop a public registry of all clinical trials and their outcomes.[8] This initiative grows directly out of the concerns raised by the American Academy of Child and Adolescent Psychiatry and others over the unpublished studies on antidepressants and potential suicide risk.

In addition, editors of prestigious medical journals, such as the Journal of the American Medical Association, The Lancet, and The New England Journal of Medicine, all support such a requirement. The editors are studying the feasibility of linking prior public registration to the possibility of later publication in their journals. Institutional review boards, which already monitor the treatment of participants in research studies, could also require that researchers publicly register their studies and agree to publish findings, whether positive or negative, with regard to the clinical effectiveness of the medication or procedure being tested.

Several large pharmaceutical companies, such as Merck, have agreed to the idea of a government-run database that would track all late-stage clinical drug trials from start to finish.[9] GlaxoSmithKline plans to establish its own database, linked to its corporate Web site, to provide results of its sponsored trials of marketed medicines.[10] That policy may become an industry standard with Eli Lilly and Company having just made a similar pledge.[11]

The second lesson learned from the controversy is that antidepressants are not trivial medicines and their use has to be carefully monitored by the prescriber and other clinicians. The National Committee for Quality Assurance has focused on this issue for several years, but only for patients older than 18 years. The Health Plan Employer Data and Information Set requires that participating health plans report the percentage of time that a clinical guideline—3 or more follow-up practitioner visits within 12 weeks of an initial antidepressant prescription for major depression—is met.[12] National HMO/POS plans currently show that patients follow their medication regimen, on average, only approximately 20% of the time.[13] That statistic is obviously not very good and may indicate that practitioners do not understand that patients with depression must be monitored carefully and frequently, especially if the depression is severe enough to warrant treatment with antidepressant medication. Hopefully, a benefit of the new concern about the safety of antidepressants will be increased vigilance and monitoring for all newly treated patients.

The third lesson has been an increased sensitivity to the importance of depression and suicide in children and adolescents. Depression is a major risk factor for suicide, but depression also carries with it an increased likelihood of substance use disorders, early pregnancy, poor academic performance, and impaired psychosocial function. Up to 6% of adolescents have symptoms of sadness, apathy, lack of energy, irritability, and social isolation that may denote major depression at any given time.[14] Despite the prevalence of these symptoms in young persons, fewer than 1% of children and adolescents (but mostly adolescents) receive specialty mental health treatment for depression.[15] Even that treatment, usually psychotherapy, is very short-term, with a mean number of 7.8 visits per year, and many persons are seen only 1 or 2 times. In a little over half of the cases, the treatment includes antidepressant medication.

Against this background of prevalence and scarce treatment, we note that suicide is the third leading cause of death among teenagers in the United States. Unsuccessful suicide attempts are also extremely common. Data from the 2003 Youth Risk Behavior Surveillance System (a national school-based survey conducted by the CDC) showed that 8.5% of high school students admitted to having attempted suicide in the 12 months preceding the survey.[16] Teenage suicide, however, seems to be on the decrease, even as antidepressant use rises somewhat in that age group.[3]

It will be interesting to observe how the controversy about the use of antidepressants in children and adolescents finally is resolved. Will we eventually see better awareness of pharmaceutical manufacturers, clinicians, and the general public regarding the importance of proper treatment of depressed children and adolescents? Will publication of all clinical studies regarding marketed products become a reality? The increased scrutiny of how depression in children and adolescents is managed will be worthwhile if the result is better recognition, leading to improved care.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: