Antidepressants and Suicide
Collection of articles published in
Child & Adolescent Psychiatry Alerts
Click here for a printable version
| Antidepressant/Suicide Relationship | |
AMA Recommends Antidepressant Labeling Review
The American Medical Association has concluded a causal link has not been established between antidepressant use in children and adolescents and suicide. As a result, they are recommending further study and have urged the FDA to examine how the new label warnings will affect the ways in which the agents are used. They also remind clinicians that the warnings are precautionary and should not prevent patients, who could benefit, from receiving the agents. According to an AMA spokesperson, the decision to use antidepressants in young patients must be based on individual risks and benefits and should include consideration of the severity of symptoms.
AMA urges review of antidepressant labels. Wall Street Journal. June 22, 2005:B4.
From August 2005 Child & Adolescent Psychiatry Alerts
Antidepressant/Suicide Relationship
Antidepressant use did not significantly increase risk of attempted suicide in a nationwide sample of adolescents with depression. There is some evidence that adolescents who complete a recommended course of antidepressant treatment may be less likely to attempt suicide.
Methods: Data were analyzed from a U.S. managed care database for >24,000 adolescents diagnosed with major depressive disorder (MDD) between 1998 and mid-2003. Suicide attempts were compared between patients who received antidepressants (n=6806) and those with MDD who were not treated with antidepressants (n=17,313). Among treated patients, the majority (68%) received SSRIs, other agents included TCAs (<1%), other antidepressants (7%), or multiple antidepressants (25%). Because the study was a retrospective cohort analysis, patients were not randomly allocated to specific therapies. To correct for nonrandom allocation, "propensity scores" that reflected the likelihood (based on age; gender; concomitant diagnoses; provider specialty; and other factors) of that patient receiving a specific agent were generated for each patient. These propensity scores were included in the regression analysis used to estimate the effects of antidepressant treatment on suicide attempt.
Results: Antidepressants were prescribed at the initial visit with increasing frequency over the study years, 5% in 1998 and 37% in 2002. Of all adolescents with a diagnosis of MDD, 1.4% attempted suicide. In a crude analysis, the risk of suicide was elevated in adolescents receiving an antidepressant. However, the relationship dissolved when the propensity score was included in the analysis, suggesting the apparent association was the result of confounding factors rather than antidepressant use.
Rates of attempted suicide did not differ significantly among patients receiving SSRIs (hazard ratio, 1.59), other antidepressants (hazard ratio, 1.03), or multiple agents (hazard ratio, 1.43). No patient receiving a TCA attempted suicide but the agents were used by only 45 of the >6800 patients. Antidepressants viewed as activating (e.g., fluoxetine and venlafaxine) were not associated with greater risk than less activating agents. Suicide risk was decreased in adolescents who received drug treatment for ³6 months, compared with those treated for <8 weeks.
Discussion: The results of this analysis suggest previously found associations between antidepressants and suicide were caused by factors other than antidepressant use. The finding that patients who complete their antidepressant course are less likely to attempt suicide is of concern because according to other research, most adolescents with depression are not treated for the full recommended 6 months.
Valuck R, Libby A, Sills M, Giese A, et al: Antidepressant treatment and risk of suicide attempt by adolescents with major depressive disorder: a propensity-adjusted retrospective cohort study. CNS Drugs 2004;18 (15):1119–1132. From the University of Colorado, Denver; and other institutions. This research was not funded.
From February 2005 Child & Adolescent Psychiatry Alerts
Drug Trade Names: fluoxetine—Prozac; venlafaxine—Effexor
Adolescent Depression: Risk of Doing Nothing
The seriousness of the possibility for antidepressants to increase suicidality must be recognized, but according to David A. Brent, md a member of the Psychopharmacologic Drugs and Pediatric Advisory Committee that recommended the black box warning, the concern is based on overestimation of the risks and underestimation of the benefits of treatment. The greatest risk to children is in doing nothing.1
The rate of adolescent suicide has been declining for over a decade and a portion of this decrease may be attributed to improved detection of pediatric depression and to the use of validated treatments. Evidence of an association between increased SSRI prescription use and a decrease in adolescent suicides has been presented,2 and there is indisputable evidence supporting the efficacy of fluoxetine (Prozac) in pediatric depression and mixed evidence for other antidepressants. The risk for treatment-emergent suicidality with antidepressants is real but small and according to Dr. Brent, acceptable in a risk–benefit analysis. The Treatment for Adolescents with Depression Study (TADS),3 found CBT inferior to fluoxetine at improving depression but the rate of suicidal events was significantly increased nearly 5-fold with fluoxetine treatment. However, in patients treated with fluoxetine, depression improved 4 times as often as suicidality developed, suggesting acceptable risk.
The ease of use and favorable side-effect profile of SSRIs may have led to an overly relaxed attitude about prescribing for pediatric depression but banning or seriously limiting pharmacological treatment of pediatric depression would leave patients and their families with nothing but hope for the development of effective treatment in the future. By discussing the risks and benefits of pharmacotherapy, educating families about how to recognize emergent adverse effects, and closely monitoring patients treated with antidepressants, perhaps a balance can be achieved between the risk of suicidality and that of doing nothing.
1Brent D: Antidepressants and pediatric depression—the risk of doing nothing. NEJM 2004;351 (October 14):1589–1601. From Western Psychiatric Institute and Clinic, Pittsburgh, Pa.
2Olfson M, et al: Relationship between antidepressant medication treatment and suicide in adolesents. Archives of General Psychiatry 2003;60:978–982. See Child & Adolescent Psychiatry Alerts 2004;6 (January):1–2.
3March J, et al: Fluoxetine, cognitive-behavioral therapy, and their combination for adolescents with depression: Treatment for Adolescents with Depression Study (TADS) randomized controlled trial. JAMA 2004;292:807–820. See Child & Adolescent Psychiatry Alerts 2004;6 (October):55–56.
From November 2004 Child & Adolescent Psychiatry Alerts
Notes From The FDA Advisory Committee Meeting
Michael Powers
Executive Editor
The FDA Psychopharmacologic Drug and Pediatric Advisory Committees met September 13–14, 2004 to discuss the occurrence of suicidal ideation and suicide attempts in younger patients who participated in antidepressant clinical trials.* Concern about the risk first surfaced when the FDA reviewed safety and efficacy data from 8 pediatric SSRI studies and noted adverse event reports suggestive of a possible association with suicidal behavior.
The FDA requested additional data from the pharmaceutical study sponsors and an Advisory Committee met on February 2, 2004 to review the adverse event data. At this meeting it became clear that inconsistent data had been used to describe events in the clinical studies and that some adverse events appeared to have been misclassified. An example of the latter was a case labelled "medication error" involving a 14-year-old who impulsively took 11 tablets of study medication and then went to school. To address this problem the FDA commissioned the Columbia Reclassification Study, the results of which were presented at the September meeting.
As an attendee of both the February and September meetings, I thought it might be useful to share some impressions, provide some comments, and try to relate my general understanding of this highly important subject.
- The plan and execution of the Columbia Reclassification Study was exquisite and rigorous: all adverse event reports were reviewed and classified by an expert panel blinded to report status of whether it was associated with drug or placebo.
- The reclassified data demonstrated a clear association between antidepressants and suicidal ideation and suicide attempts. The risk appears to be about 2–3%.
- No patient in any of the trials committed suicide.
- Based on the committee's recommendations, it appears the FDA will request manufacturers' of all antidepressants to add a "black box" warning to the label about the risk of suicidal tendencies. Individual antidepressants will not be singled out for the warning. In addition, the FDA will request adding an instructional guide for patients to the prescription package.
- The Advisory Committee recommended that antidepressants not be contraindicated in this country because of their overall value in treating depression in young people and adults.
* See “Reanalysis Confirms Antidepressant Risk” Child & Adolescent Psychiatry Alerts 2005;6 (September):49.
From October 2004 Child & Adolescent Psychiatry Alerts
Paroxetine in Completed Suicides
Toxicological analysis was undertaken of persons under age 18 years certified as suicide or accident by the Office of the Chief Medical Examiner of New York City for the 6 years following the introduction of paroxetine. Of 66 suicides, the analysis excluded 8 with no toxicology data and 4 in which the youth survived longer than 3 days after the attempt (because paroxetine levels would have been undetectable at the time of death). Paroxetine was not detected in any of the toxicology analyses. Imipramine and fluoxetine were each detected in 2 youths.
To determine whether any paroxetine-related suicides were misclassified as accidental deaths, toxicology results were also examined for 407 accidental deaths in the same age group. Again, paroxetine was not detected in any of the analyses but 2 were positive for amitriptyline and 1 for fluoxetine.
These findings are consistent with the FDA advisory stating that there were no instances of paroxetine-related completed suicide in the drug's clinical trials.
Leon A, Marzuk P, Tardiff K, Teres J: Paroxetine, other antidepressants, and youth suicide in New York City: 1993 through 1998. Journal of Clinical Psychiatry 2004;65 (July):915–918.
From Weill Medical College of Cornell University, New York, N.Y. Funded by the NIH.
Drug Trade Names: amitriptyline—Elavil; fluoxetine—Prozac; imipramine—Tofranil; paroxetine—Paxil
From October 2004 Child & Adolescent Psychiatry Alerts
Reanalysis Confirms Antidepressant Risk
A new FDA analysis of pediatric clinical trial data has confirmed an association between suicidal behavior and antidepressants.
The findings of a previous analysis of 25 antidepressant trials in children and adolescents were unpublished because they were considered by the FDA to be premature and based on ambiguous data. The same studies were reviewed in depth by researchers from Columbia University to determine which of the reported incidents accurately reflected suicidal tendencies. The reclassified data was then analyzed by an FDA reviewer who prepared a report. The document has not yet been made public but a draft form was reviewed by and reported in the Wall Street Journal. An FDA spokesperson declined to comment on the analysis before the advisory committee meeting scheduled for September 13–14, 2004.
The reanalysis generally agrees with the initial findings of an association between antidepressants and suicidal tendencies but the second analysis differs in the way suicidal behavior and self-harm are classified. The present analysis found that children and adolescents taking antidepressants had 1.78 times the risk for "suicide attempt," "preparatory actions towards imminent suicidal behavior," and "suicidal ideation" as patients taking placebo. Patients taking antidepressants had a 2-fold increased risk of a far broader category of incidents, including self-harm not necessarily associated with suicidal intent and incidents that "couldn't be defined," according to the article. When only data on SSRIs were included, a statistically nonsignificant 1.4-fold increase in risk for suicidal intent was found. The link was strongest with paroxetine and venlafaxine, while other drugs appeared to have little or no effect. No patient in any of the trials committed suicide.
Mathews A: FDA revisits issue of antidepressants for youths: new analysis may pressure agency to set limit on use because of suicide risk. Wall Street Journal. August 5, 2004:A1.
Drug Trade Names: paroxetine—Paxil; venlafaxine—Effexor
From September 2004 Child & Adolescent Psychiatry Alerts
Antidepressant Drugs in Children
Do antidepressants really increase the risk for suicidal behavior in some children with depression? The FDA is reviewing the question and has requested changes in the labeling of SSRIs and related drugs warning clinicians to monitor patients for worsening depression and the emergence of suicidality. Meanwhile, according to Vitiello and Swedo from the NIMH, one way to improve the risk/benefit ratio in children with depression is to reserve antidepressant use for patients with persistent or recurrent illness that does not respond to other therapies.
Evidence from at least 2 independent trials is required to establish antidepressant efficacy. Fluoxetine (Prozac) is the only agent labeled for treatment of depression in children, but off-label use of other agents occurs. For fluoxetine, the 2 clinical trials showed modest efficacy compared with placebo but the studies were characterized by high placebo response rates, nonresponse to active medication in one-third of patients, and higher rates of agitation and manic symptoms in patients receiving fluoxetine. Although it is theoretically possible for these adverse behavioral events to increase patients' risk of self-injurious acts, rates of suicidal behavior were the same for fluoxetine and placebo in the 2 trials. No completed suicide has been reported in the more than 4100 patients enrolled in pediatric SSRI trials.
Placebo-controlled trials typically exclude patients at high risk for suicide, and other sources of data also have limitations. Epidemiologic studies have observed a significant decrease in child and adolescent suicides accompanying the increased use of SSRIs in this age group. However, this temporal association does not establish that SSRIs caused the reduction in suicides. Adverse event reporting systems such as MedWatch are of limited usefulness because suicide can be a symptom of depression or a distinct drug-related adverse effect.
Nonpharmacological interventions such as cognitive-behavioral therapy, behavioral or environmental change, or emotional support are useful in children. Such interventions, followed by an early follow-up interview to establish the persistence of depression, may be indicated before introducing antidepressant medication.
Vitiello B, Swedo S: Antidepressant medications in children. NEJM 2004;350 (April 8):1489–1491. From the NIMH, Bethesda, Md.
From June 2004 Child & Adolescent Psychiatry Alerts
Canada Issues SSRI Advisory
In light of recent safety concerns regarding SSRIs and associated suicide or suicidal ideation, Health Canada is urging all patients under the age of 18 years who are receiving bupropion; citalopram; fluvoxamine; mirtazapine; paroxetine; sertraline; or venlafaxine to contact their physician to determine if the benefits of treatment continue to outweigh the risks. They also have requested a review of the safety data in pediatric patients from the manufacturers. In Canada, SSRIs are not approved for pediatric populations but off-label use occurs. The advisory does not apply to adult patients, for whom use is approved.
Health Canada Advisory: Health Canada advises Canadians under the age of 18 to consult physicians if they are being treated with newer anti-depressants. Posted at
www.hc-sc.gc.ca/english/protection/warnings/2004/2004_01.htm.Drug Trade Names: bupropion—Wellbutrin; citalopram—Celexa; fluvoxamine—Luvox; mirtazapine—Remeron; paroxetine—Paxil sertraline—Zoloft; venlafaxine—Effexor
Child & Adolescent Psychiatry Alerts
Antidepressants Reduce Suicide
Reductions in suicide rates correlated with increases in antidepressant use were identified in a retrospective study of adolescents. This observation supports suicide prevention policies based on identifying and treating patients with antidepressant-responsive psychiatric disorders.
Methods: U.S. regional suicide rates and antidepressant use were examined in each 1-month period for the years 1989 and 2001. Included in the study were prescription data from a large pharmacy benefits plan; suicide data from the CDC; and socioeconomic and health care data for the geographic regions.
Results: When comparing time trends, increases in antidepressant use were significantly associated with decreases in suicide in males, in adolescents aged 15–19 years, and in lower socioeconomic status regions. Similar associations were suggested but did not reach statistical significance in females and in areas with a high median income. Cross-sectional data from 2 of the 24 observation periods showed that areas with high rates of antidepressant use also had high rates of suicide. No association was observed in children aged 10–14 years.
Discussion: The larger decrease in suicide in male teens may be attributable to gender differences in suicide attempt lethality. Results of this study are open to question because the association was ecological rather than individual. However, suicide is extremely difficult to study with classic experimental methods.
Olfson M, Shaffer D, Marcus S, Greenberg T: Relationship between antidepressant medication treatment and suicide in adolescents. Archives of General Psychiatry 2003;60 (October):978–982.
From the College of Physicians and Surgeons of Columbia University, New York, N.Y.; and the University of Pennsylvania, Philadelphia. Source of funding not stated. The authors have disclosed financial relationships with Bristol-Myers Squibb; Wyeth Pharmaceuticals; Eli Lilly & Co.; and GlaxoSmithKline.
From January 2004 Child & Adolescent Psychiatry Alerts
Antidepressant Public Health Advisory
The FDA has completed a preliminary review of reports of suicidal thinking and suicide attempts in pediatric patients with major depressive disorder treated with citalopram; fluoxetine; fluvoxamine; mirtazapine; nefazodone; paroxetine; sertraline; or venlafaxine. Although the data do not clearly establish an association between antidepressants and suicide, they do not rule out the possibility. As a result, the FDA has issued a Public Health Advisory and a meeting is scheduled for February 2004 to promote public discussion of the data and appropriate regulatory action.
FDA Talk Paper: FDA issues public health advisory entitled: reports of suicidality in pediatric patients being treated with antidepressant medications for major depressive disorder (MDD). Posted at: www.fda.gov/bbs/topics/ANSWERS/2003/ANS01256.html.
Drug Trade Names: citalopram—Celexa; fluoxetine—Prozac; fluvoxamine—Luvox; mirtazapine—Remeron; nefazodone—Serzone; paroxetine—Paxil; sertraline—Zoloft; venlafaxine—Effexor
From December 2004 Child & Adolescent Psychiatry Alerts
Pediatric Safety of Sertraline
The FDA has concluded that pediatric sertraline (Zoloft) does not differ from adult use in that it is associated with little risk of suicide beyond that associated with depressive disorder. The agency recommended that clinicians supervise patients closely during initiation of antidepressant therapy and to prevent overdoses they should prescribe sertraline in as few tablets as possible while maintaining good patient management.
The review was mandated as part of the Best Pharmaceuticals for Children Act and was based on spontaneous reports to the Adverse Event Reporting System beginning in February 2002. Nearly 700,000 prescriptions for sertraline were written for patients aged £16 years in 2002. Depression was the leading indication, accounting for 36% of all prescriptions in this age group.
Reported adverse events were similar to those described in adults, and included psychiatric events (e.g., hostility, hallucinations) as well as extrapyramidal and movement disorders. Congenital anomalies, possible neonatal withdrawal syndromes, and memory impairment were also reported. Most pediatric adverse event reports involved other drugs in addition to sertraline or confounding medical disorders. Adverse events resulted in 4 deaths, including 2 suicides; 19 hospitalizations; and 26 other serious incidents requiring medical intervention. The suicides both occurred in adolescents, 1 with depression who began sertraline treatment 1 week prior to death, and another with no history of depression who consumed several doses of sertraline without a prescription, over a 10-day period. The relationship of sertraline to the suicides was unclear. A premature child of an HIV-infected mother who used sertraline and multiple other medications died 15 days after birth from apparently unrelated causes, and a school-aged child died of multiple drug toxicity. There were 5 nonfatal overdoses, 2 of which were accidental and 3 deliberate.
Center for Drug Evaluation and Research, Food and Drug Administration presentation: One year post-pediatric exclusivity postmarketing review: sertraline (Zoloft). Posted at www.fda.gov/cder/pediatric/presentation/ac6-03si/default.htm.
Paroxetine Advisory
Based on data recently released by the manufacturer, the FDA and regulators in the U.K. are urging doctors to not prescribe paroxetine (Paxil) for children.1,2 The results of clinical trials involving >1000 patients suggested that mood swings, increased crying, suicidal thoughts, and potential suicidal behavior occurred twice as often in patients treated with paroxetine (3.2%) as in those receiving placebo (1.5%). In addition, in 3 controlled trials in pediatric patients with depression, the FDA determined that paroxetine was not more effective than placebo. The agent is not approved for use in pediatric patients. Despite the concerns about its safety, the FDA advisory cautions that paroxetine should not be abruptly discontinued.
1FDA Talk Paper: FDA statement regarding the anti-depressant Paxil for pediatric population. Posted at: www.fda.gov/bbs/topics/ANSWERS/2003/ANS01230.html
2Hensley S: U.K. urges doctors not to give Paxil to kids. Wall Street Journal. June 11, 2003:B1, B7.