Introduction

The first published literature on the topic of antidepressants and suicidality dates back to the mid-1950s. Today, 10 years after the black-box warning was issued, this controversial topic is still debated. This article will review the data behind the black-box warning and its revision; address the subsequent impact the warning had on depression diagnoses, prescribing patterns, follow-up visits, and suicide rates in the United States; and summarize meta-analyses on this topic published since the revised warning was issued in 2007.

Methods

A PubMed database search using the MeSH terms Antidepressive Agents and Suicide was conducted from January 2012 through October 2014 to identify articles published on the impact the Food and Drug Administration's black-box warning had on health care at a national level and from January 2007 through October 2014 using the same MeSH terms to identify meta-analyses on the current research regarding the link between antidepressants and suicidality. Search results for both topics were further limited to those articles published in English on human participants with the age criteria Child: Birth – 18 years, Adolescent: 13-18 years, or Young Adult: 19-24 years.

Results

Due to the black-box warning, depression diagnoses and antidepressant prescription issuance rates declined in young patients. It is still debated whether the black-box warning had an untoward effect on suicide rates. Most of the meta-analyses identified a small risk of suicidality in the populations studied. Because of the varying nature of the meta-analyses and the major limitation of the limited availability of prospectively collected suicidality data, an absolute risk is still yet to be determined.

Discussion/Conclusion

More studies on antidepressants conducted to prospectively identify suicidality in patients with Major Depressive Disorder and other varying diagnoses need to be completed to truly identify the incidence of suicidal behavior and ideation when initiating antidepressants. Albeit seemingly low, there is a risk of suicidality when initiating antidepressants; there is also risk in not treating depressed patients with antidepressants in whom they are indicated. Clinicians must be cognizant of this risk and monitor high-risk patients per the Food and Drug Administration–recommended guidelines.

In 1990, Teicher et al1  described the first series of 6 case reports of adult patients taking fluoxetine who developed intense suicidal ideation; however, this was not the first published literature regarding antidepressants and suicidal ideation.1  In a psychiatry textbook published more than 50 years ago the authors state, “…the risk of suicide once more becomes serious as retardation fades.”2  This statement, referring to the initiation of antidepressant therapy, embraces one of the many theories that clinicians have regarding these commonly prescribed medications and the emergence of suicidal ideation when a patient's energy level normalizes but his or her mood remains depressed.

The case reports of fluoxetine causing suicidal ideation outlined by Teicher et al1  gained attention from the Food and Drug Administration (FDA) and the National Institute of Mental Health (NIMH), which ultimately led its manufacturer, Eli Lilly and Company, to reanalyze data from randomized controlled trials on fluoxetine.3  One year later, in 1991, this renewal of interest was put to rest when most of the Psychopharmacologic Drugs Advisory Committee (PDAC) of the FDA concluded that there was no clear association between the drug and the emergence of suicidality based on pooled data. During the next decade, however, much more attention was paid to reanalysis of existing data and new antidepressants being brought to the market as several groups of researchers looked to identify a possible causal link between antidepressants and suicidality.

In May 2003, GlaxoSmithKline reported issues concerning suicidality in pediatric patients taking a drug they manufactured, paroxetine, to the UK and US regulatory agencies after reanalysis of previous data.3  The safety and efficacy data from 9 clinical trials on paroxetine comprising more than 1500 young patients with Major Depressive Disorder (MDD), Obsessive Compulsive Disorder (OCD), and Social Anxiety Disorder (SAD) indicated that the drug was not any more effective than placebo, and that the rate of suicidal thinking and suicide attempts was 3.4% on the active drug compared with 1.2% on placebo (odds ratio [OR], 2.80; 95% confidence interval [95% CI], 1.25-6.25; P = .012).4 

In June 2003, the UK Medicines and Healthcare Products Regulatory Agency (MHRA) released a statement concerning the use of paroxetine in children and adolescents that explicitly stated that it should not be used to treat depression in those younger than 18 years, along with a mandated labeling change which made its use contraindicated in pediatric patients with MDD.3  Shortly thereafter the FDA issued a similar statement; however, no change in labeling was called for at that time. Then, in July 2003, the FDA sent a letter to the manufacturers of 8 additional antidepressants requesting summary data for the pediatric trails that had been completed. One month later, Wyeth released a Dear Health Care Professional letter regarding concerns of lack of efficacy and increased suicidality in pediatric trials of venlafaxine and voluntarily made labeling changes.5 

Between 2003 and 2004, the FDA and the MHRA issued warnings regarding numerous antidepressants. On February 2, 2004, the PDAC and the Pediatric Advisory Committee (PAC) held a meeting to meet the legal obligations of presenting postmarketing adverse event data on paroxetine and citalopram, which were granted pediatric exclusivity in the previous year.6  This led to a formal FDA-solicited meta-analysis of 24 pediatric RCTs involving 9 different antidepressant medications to determine overall suicidality risk. The results of the aforementioned meta-analysis were the basis for the FDA's issuance of the initial black-box warning (BBW) regarding an increased risk of suicidality in children and adolescents taking antidepressants versus placebo.

This article will discuss the data behind the BBW and its revision; address the subsequent impact the warning had on depression diagnoses, prescribing patterns, follow-up visits, and suicide rates in the United States; and summarize meta-analyses on this topic published since the revised warning was issued in 2007.

In order to identify articles and/or documents pertinent to the review of the data behind the initial and revised BBW, information was obtained from the FDA's Web site in the form of briefing documents and meeting minutes. A PubMed database search using the MeSH terms Antidepressive Agents and Suicide was conducted from January 2012 through October 2014 to identify articles published on the national impact the FDA's BBW had on health care. Articles excluded from this section were (1) those regarding the impact the warnings had on countries other than the United States, (2) those regarding the impact at the state level, and (3) brief commentaries. Many articles pertinent for this review were identified through a bibliography search of a review article published by Adegbite-Adeniyi et al. A PubMed database search using the MeSH terms Antidepressive Agents and Suicide was conducted from January 2007 (4 months before the FDA announced the revised BBW in May 2007) through October 2014 to identify meta-analyses on the current research regarding the link between antidepressants and suicidality. Search results for both topics were further limited to those articles published in English on human participants with the age criteria Child: Birth – 18 years, Adolescent: 13-18 years, and Young Adult: 19-24 years. All abstracts were read by the author to determine appropriateness for inclusion.

Data Behind the Initial BBW

Briefing documents, meeting minutes, and articles used as sources of information for this section were obtained and/or identified from the FDA's Web page titled “Historical Information on Antidepressant Use in Children, Adolescents, and Adults.”7 

The FDA-solicited meta-analysis that led to the initial BBW was based on data from 23 trials conducted in pharmaceutical company–supported programs evaluating antidepressant efficacy in pediatric patients, and one large multicenter trial (the Treatment for Adolescents with Depression Study).8  Indications studied in the trials were MDD (n = 16), OCD (n = 4), Generalized Anxiety Disorder (n = 2), Attention Deficit Hyperactivity Disorder (n = 1), and SAD (n = 1). A total of 20 of the trials were included in the final analysis because there were no suicidality events in 4 of the trials. Nine widely used antidepressants were included in the analysis: fluoxetine, sertraline, paroxetine, fluvoxamine, citalopram, bupropion, venlafaxine, nefazodone, and mirtazapine.

Potential suicide-related events (SREs) were identified using a search algorithm with several text strings chosen to identify such events. The scale used by the experts to classify these events—the Columbia Classification Algorithm of Suicide Assessment (C-CASA)—organized SRE narratives into one of five categories as follows: (1) suicide attempt, (2) preparatory actions toward imminent suicidal behavior, (3) suicidal ideation, (4) self-injury with intent unknown, and (5) injury events with not enough information to determine whether they represented self-injury or other injury. Categories 1 to 3 were deemed to be most representative of suicidality and were used as the primary outcome suicidal behavior or ideation. The secondary outcome, possible suicidal behavior or ideation, encompassed all 5 categories. Suicide item score data were also evaluated for 17 of the 24 trials based on available data collected from the depression rating scales that were used (ie, Children's Depression Rating Scale-Revised, Hamilton Depression Rating Scale, and the Montgomery Åsberg Depression Rating Scale).

The primary result of this pediatric meta-analysis showed that the overall risk ratio (RR) of suicidal behavior or ideation was 1.95 (95% CI, 1.28-2.98) when comparing active drug with placebo.8  When conducting separate analyses for suicidal behavior and ideation the RRs were 1.90 (95% CI, 1.00-3.63) and 1.74 (95% CI, 1.06-2.86), respectively. The secondary result, possible suicidal behavior or ideation, also showed an increased risk of the active drug compared with placebo (RR, 2.19; 95% CI, 1.50-3.19). Meta-analysis also revealed that there were inconclusive results regarding the suicide item score for drug compared with placebo in terms of worsening of suicidality (RR, 0.92; 95% CI, 0.76-1.11) or emergence of suicidality (RR, 0.93; 95% CI, 0.75-1.15).

Although there were no completed suicides in any of the clinical trials evaluated, this meta-analysis revealed a modestly increased risk of suicidality associated with antidepressant treatment in children and adolescents (4%) compared with placebo (2%).9  Findings were presented at the September 13-14, 2004, meeting of the PDAC and PAC. All 27 voting members unanimously agreed with the statement, “The data in aggregate indicate an increased risk of suicidality, as previously defined, in pediatric patients. Although there is variability in the results, we are unable to conclude that any single antidepressant agent is free of risk at this time.” On October 15, 2004, the FDA directed manufacturers of all antidepressants to revise labeling and alert health care providers and subjects about an increased risk of suicidality in pediatric and adolescent patients who take these medications.10 

Data Behind the Expanded BBW

Briefing documents, meeting minutes, and articles used as sources of information for this section were obtained and/or identified from the FDA's Web page titled “Historical Information on Antidepressant Use in Children, Adolescents, and Adults”7 

Two months after the pediatric BBW was issued, the FDA sent letters to the manufacturers of 12 different antidepressant drugs requesting data from all double-blind, randomized, placebo-controlled trials in adults for any indication in an attempt to further elucidate this increase in risk.11  In addition to the 9 antidepressants studied in the pediatric meta-analysis, escitalopram, duloxetine, and olanzapine/fluoxetine were included in the adult analysis.12  In order to conduct this meta-analysis, Stone et al12  were provided with data from 406 trials. After exclusion criteria were applied, data from 372 clinical trials (295 trials for psychiatric indications), which included 99 839 patients, were evaluated.

Possible SREs were similarly identified searching clinical trial databases for text strings; however, the drug companies themselves identified the events. The primary outcome of the study was defined as suicidal ideation or worse (categories 1-4 below), which corresponded to the primary outcome of suicidal behavior or ideation in the FDA's pediatric meta-analysis. The secondary outcome was defined as suicidal behavior (categories 1-3). The 7 mutually exclusive categories by which the events were classified were: (1) completed suicide, (2) suicide attempt, (3) preparatory acts toward imminent suicidal behavior, (4) suicidal ideation, (5) self-injurious behavior, intent unknown, (6) not enough information (fatal), and (7) not enough information (nonfatal).

The OR for the primary outcome of suicidal ideation or worse in those taking antidepressant medications for the whole data set, which included all trials, ages, and indications, was 0.85 (95% CI, 0.71-1.02).12  The OR for the secondary outcome—suicidal behavior—was 1.12 (95% CI, 0.79-1.58). These results were in contrast with the pediatric meta-analysis that did show significant results in the primary and secondary outcome measures. However, when data for the trials in adults with psychiatric disorders (295 trials) were broken down categorically into age range subgroups, significant results were realized. In patients younger than 25 years, the primary outcome (ideation or worse) did not reach statistical significance, with an OR of 1.62 (95% CI, 0.97-2.71); however, there was a statistically significant increase in the secondary outcome (suicidal behavior) in the antidepressant-treated group, with an OR of 2.30 (95% CI, 1.04-5.09). There was no appreciable effect on those ages 25 to 64 years, and there was a significant protective effect on suicidality (primary and secondary outcome measures) in patients older than 65 years.

These data, presented at the December 13, 2006, meeting of the PDAC along with a second independent analysis of the same data, prompted the FDA to extend the BBW to include younger adults. Additionally, verbiage was added to address the inherent risk of poor outcomes in untreated depression, because this was a caveat of the voters in favor of the extended warning.13 

Depression Diagnoses, Prescribing Patterns, Follow-Up Visits, and Suicide Rates in the United States

The PubMed search returned 95 articles. A review article published in 2012 that summarized topical literature regarding national data on rates of depression diagnoses, prescribing trends, doctor visits, and suicide rates before and after regulatory advisories from 2001 to 2012 helped identify many of the articles pertinent for this review.14  Additionally, an original research article published on US national data after the aforementioned review article was identified and included in this section. In total, 25 articles and additional suicide data from The Centers for Disease Control and Prevention (CDC) were identified for this section of the review.

Rates of depressive symptoms and a formal MDD diagnosis, antidepressant prescription issuance, follow-up visits, and suicidal behavior and ideation were all thought to be affected by the regulatory actions of government agencies across the world.14 

In the United States, pediatric depression diagnoses rose steadily from 1999 to 2004 (3 to 5 per 1000) and then declined to 1999 (3 per 1000) rates in 2005 after the warnings were made known to the general public.15,16  These data, gathered from a national claims database, showed a significant deviation in pediatric diagnoses of depression from the historical trend. Because of the fact that diagnostic rates were affected, one may expect that prescribing patterns shifted as well.

Before health care providers and the public were privy to the warnings, antidepressant prescribing had been increasing in the United States among children and adolescents; however, this pattern shifted after the warnings.15-26  The number of new antidepressant prescriptions for children and adolescents decreased significantly as well.13,16,25,27,28  Data analyzed from the Agency for Healthcare Research and Quality under the US Department of Health and Human Services revealed that new antidepressant prescriptions decreased 47% in treatment-naive patients ages 5 to 21 years in the United States from 2002 to 2006. Notably, prescription rates in those previously exposed remained the same.25  Changes in prescription rates appeared to vary among prescribing clinicians, most profoundly for nonpsychiatrists. Researchers found a significant reduction in antidepressant prescribing by nonpsychiatrists and non–primary care physicians in patients ages 6 to 17 years just 1 year after the initial warning.29  Another group of researchers found that 11.5% of pediatricians and 3.9% of family medicine practitioners were no longer prescribing antidepressants to young people at all; this was the case in less than 1% of psychiatrists.30  In addition to affecting prescribers, guardians of children/adolescents and children/adolescents themselves were also seemingly influenced by the warnings; prescribers reported that 14% to 22% of guardians and 9% of child/adolescent patients refused treatment with antidepressants.30,31  It is also interesting to look at antidepressant prescribing rates by individual agent as warnings began to surface about particular drugs before the class warnings were issued. When the paroxetine warning began to surface, some researchers reported that prescribing rates of nonparoxetine selective serotonin reuptake inhibitors (SSRIs) increased, whereas others reported a significant decrease.24,29  Regarding paroxetine itself, a ~33% reduction in prescribing was observed in patients ages 6 to 17 years following the 2003 public health advisory. On the other hand, a relative increase was observed in prescribing rates of fluoxetine; this was confirmed by a claims database that showed that the percentage of children prescribed fluoxetine doubled between 2003 and 2005.25,28  Similar prescribing trends were observed in Canada and parts of Europe.14 

Follow-up visits to prescribers following the warnings were urged by the FDA so that high-risk patients could be closely monitored following antidepressant initiation.14  Specifically, physicians were to meet with the patients on a weekly basis for the first month, every other week for another month, and then again at 12 weeks; however, little change in visitation frequency was noted among providers.14,30,32  Morrato et al32  found that around 5% of patients met with their physician per the FDA criteria before the warnings, and no substantial difference was realized when compared with the postwarning period. Bhatia et al30  found that nearly one third of physicians did meet with patients more frequently, but only 7.5% of the providers met the recommended weekly visits for the first month following initiation. In Canada, 30% of pediatricians who heard about the FDA guidelines increased their visit frequency.33  It should be noted that the United States and Canada have major differences in health care systems, which may account for the disparities noted above. Due to the passage of the Patient Protection and Affordable Care Act (ACA), access to mental health care for those in the United States will increase. The implications of this act, signed into law in March 2010, remain to be elucidated on a national level because the provisions took effect on January 1, 2014. The new legislation should have the greatest impact on those who were not insured prior to the implementation of the ACA.34 

Prior to regulatory agencies' warnings, antidepressant prescription rates were increasing and suicide rates were decreasing in children and adolescents in the United States.22,35,36  The CDC reported that suicide rates declined during an 18-year period from 1991 to 2009 (from 9.24 to 7.21 per 100 000) among those ages 10 to 24 years.37  Conversely, since 2005, several other studies have suggested increased rates of suicide due to the known decreased rate of antidepressant prescription issuance. Gibbons et al35  reported a steady decrease in suicide rates in children and adolescents during a 15-year period from 1988 to 2003 (from 4.4 to 2.83 per 100 000). During this same time period, SSRI prescription issuance to children and adolescents was increasing. Suddenly, between 2003 and 2004, the rate of suicide increased 14% among children and adolescents. Interestingly, it was found that rates of suicidal ideation actually decreased from 1995 to 2005 (12.7% to 10.9%) in children and adolescents ages 12 to 17 years.38 

A recently published, NIMH-funded, nationwide cohort study on suicide rates after the 2004 FDA warnings offers insight into the effect these regulatory actions have had in the United States.39  The study looked at 11 different geographically distributed US health care organizations that provide care to a diverse population of more than 10 million people in 12 states; the cohorts included were adolescents (ages 10-17 years), young adults (ages 18-29 years), and adults (control group; ages 30-64 years). The authors looked at all patients dispensed an antidepressant and used the poisoning by psychotropic agents international classification of diseases, ninth revision, clinical modification (ICD-9) code to identify suicide attempts, because this is a reliable representation of population-level suicide attempts. Because of the prolonged nature of the FDA warnings and actions, the study was broken up into three distinct periods covering 10 years: prewarning was Q1 2000–Q3 2003, phase-in was Q4 2003–Q4 2004, and postwarning was Q1 2005–Q4 2010. Antidepressant use decreased significantly among adolescents (−31%), young adults (−24.3%), and adults (−14.5%) in the second year following the warnings. At the same time, there were relative increases in psychotropic drug poisonings among adolescents (21.7%) and young adults (33.7%); these percentages equated to 2 and 4 additional poisonings (suicide attempts) per 100 000 people. The adult population did not show a significant increase.

The study's findings are consistent with the existing body of evidence that shows reduced antidepressant use in young people after the warnings and provide the first evidence that suicide attempts increased rather than decreased after the warnings. The authors added that the findings are supported by previous studies that did not show compensatory increases in the use of treatment alternatives due to decreased antidepressant prescribing rates. Additionally, this was the first study to look at suicide attempts over a long period of time. Despite analyzing administrative data from millions of patients, there were no observed changes in suicide completion rates after the warnings. The fact that completion rates did not suddenly change around the time of the warnings provides some evidence that the FDA warnings may not have had an effect on suicide completion rates. There were also limitations to the study that warrant attention.

Due to the study's focus on psychotropic drug poisonings requiring medical attention, the impact of the warning was underestimated, because only one third of suicide attempts are medically treated and psychotropic drug poisonings only account for 38% of all attempts. Another limitation was the inability to track E-codes for deliberate self-harm because of inconsistent reporting across study sites. Additionally, a validated algorithm (eg, combination of injury and psychiatric diagnosis, which has a positive predictive value of 87.8%) was not used to identify suicide attempts; however, this was done to avoid the introduction of ascertainment bias because the rate of depression diagnoses declined after the warnings. Other limitations were that the sample largely included insured (public and private) populations, the analyses were not stratified by race/ethnicity or socioeconomic status, and that population-level data cannot be used to make inferences about patient-level outcomes.

Meta-Analyses

Initially, the search returned 310 articles. Upon further review, there were 5 meta-analyses that were published on this topic since the FDA issued warnings on antidepressants causing suicidality in children, adolescents, and young adults (Table).

TABLE:

Published meta-analyses since the revised black-box warning

Published meta-analyses since the revised black-box warning
Published meta-analyses since the revised black-box warning

In the meta-analysis by Carpenter et al,40  which evaluated all trials involving paroxetine in adults across all indications, there was no difference in suicidality between antidepressant and placebo. However, there was a higher incidence of suicidality in young adults (ages 18-24 years) on antidepressants versus placebo across all indications. When trials for MDD were grouped together, there was a greater incidence of suicidality in those being treated with antidepressants versus placebo; this was largely explained by the higher incidence in young adults. In the meta-analysis by Hetrick et al,41  which evaluated published and unpublished trials of newer-generation antidepressants in children ages 6 to 18 years with a depressive disorder, there was a statistically significant increase in suicidality for those taking antidepressants versus placebo. In the meta-analysis by Julious,42  which evaluated trials on antidepressants in children and adolescents with varying diagnoses, there was a trend toward increased suicidality in those treated with antidepressants versus placebo; additionally, this trend remained when trials were grouped by indication. In the meta-analysis by Gibbons et al,43  which evaluated trials on fluoxetine (in youth, adult, and geriatric populations) and venlafaxine (in adults) for patients given a diagnosis of MDD, there was no difference in suicidality between fluoxetine and placebo in the youth trials. In the adult and geriatric trials, there was a decrease in suicidality in patients on antidepressants versus placebo. In the meta-analysis by Bridge et al,44  which evaluated published and unpublished trials of newer-generation antidepressants in children 18 years and younger with varying diagnoses, there was a statistically significant increase in suicidality in those being treated with antidepressants versus placebo. When trials on MDD were grouped together, there was a trend toward an increased risk of suicidality in those on antidepressants versus placebo.

Not surprisingly, with trials involving varying disorders, methods, and results reported, researchers are still unable to reject a risk associated with using antidepressants in children, adolescents, and young adults.

The inability to discount the small risk of suicidality in children, adolescents, and young adults treated with antidepressants is echoed by Andrew C. Leon, PhD, the biostatistician who sat on the PDAC and cast one of the six votes in favor of extending the BBW, who wrote, “My vote to extend the ‘black-box' warning to young adults was based on concern that risk of suicidality could not be ruled out and, given the widespread antidepressant use, even a small risk must not be ignored.”45  One of the most important limitations to note about the meta-analyses that prompted the warnings was that the suicide-related events were identified retrospectively. In addition, the information was not collected systematically, and researchers relied on data that were spontaneously described through adverse event reporting systems. Today, clinical trials are being conducted in a manner in which data on suicidality are collected prospectively. The standard of care in clinical trials on antidepressants in young patients is the administration of the Children's Depression Rating Scale–Revised, the Columbia Suicide Severity Rating Scale (C-SSRS), and reporting of treatment-emergent adverse events. In contrast with the C-CASA used in the FDA's retrospective meta-analyses, the C-SSRS is a validated instrument designed to prospectively identify suicidal ideation and behavior.46  The FDA recommends use of this tool to directly categorize suicidality into 1 of 11 different categories:

  • Suicidal ideation:

    • 1.

      Passive

    • 2.

      Active: nonspecific

    • 3.

      Active: method, but no intent or plan

    • 4.

      Active: method and intent, but no plan

    • 5.

      Active: method, intent, and plan

  • Suicidal behavior:

    • 1.

      Completed suicide

    • 2.

      Suicide attempt

    • 3.

      Interrupted attempt

    • 4.

      Aborted attempt

    • 5.

      Preparatory actions toward imminent suicidal behaviors

  • Self-injurious behavior, no suicidal intent

This scale is now a regular part of assessing suicidality in trials involving antidepressant and antiepileptic drugs for psychiatric and nonpsychiatric indications as well as other drugs with central nervous system activity. Although there is now guidance for the industry going forward to prospectively evaluate suicidality in clinical trials, the issue of the antidepressant BBW is still debated today. Some argue that the BBW should be removed altogether based on the effect it has had on a national level, whereas others argue that the BBW has had no appreciable effect on suicide rates.47,48 

Two commentary articles recently published in the New England Journal of Medicine provide insight into the current opinion and its variability among providers. Dr Richard A. Friedman, MD, a professor of clinical psychiatry at Weill Cornell Medical College, argues, “…the FDA should consider removing the [black-box] warning entirely. At the very least, the medical profession should have a discussion about this possibility.”47  This statement aims to highlight the failed attempt at the revised BBW to warn about the inherent risk of untreated depression and its consequences on rates of morbidity and mortality. The viewpoint presented in Dr Friedman's article is based on many of the published studies discussed in this review article. He also does not discount the risk associated with antidepressant use, urging primary care physicians (who, he notes, treat a “substantial proportion” of depressed patients) to be aware of the small risk of suicidality with antidepressant treatment and outlining that they can safely manage it by closely monitoring patients.

Conversely, Dr Marc B. Stone, MD, senior medical reviewer for the Division of Psychiatry Products of the FDA and primary author of the large meta-analysis that was the basis for the expanded/revised warning, argues that the BBW had no effect on suicide rates.48  The following are a few highlights from his commentary. Dr Stone first speaks to the article by Gibbons et al,35  published at the peak of the antidepressant and suicidality controversy, that claimed that the reduction in issuance of SSRI prescriptions in the United States and the Netherlands resulted in an increased suicide rate in children and adolescents. He points out that the authors, the reviewers, and editors did not notice that the antidepressant prescription issuance rate decreased in 2005, and suicide rates increased in 2004; this is now a known and admitted error. He adds that other research has made similar claims, but the trends could be random fluctuations, arbitrary before and after periods are used, and data from inferior sources were commonly used to identify these trends. Also explained is the thought that yearly fluctuations in antidepressant prescription rates are not likely to directly influence suicide rates because most of any possible reduction in suicide rates would most likely occur when antidepressants are first introduced into a population. Of note, the issuance of antidepressants decreased from 2004 to 2005 in every age subgroup except those 55 years and older. Dr Stone elucidates the possibility that this may have been due to the fact that from 2004 to 2006 drug companies' advertising expenditures decreased by 35% because this was a time that a lot of brand name medications were going generic. Specifically, generic approvals (varied formulations) were granted for 6 different antidepressants during that time: bupropion, paroxetine, mirtazapine, citalopram, sertraline, and venlafaxine. The advertising that was happening was targeting older individuals who would soon be eligible for the new Medicare Part D prescription drug coverage benefit. Finally discussed is the study published by Lu et al39  that compared the number of antidepressant prescriptions with the number of psychotropic drug poisonings in a large number of patients in the US Mental Health Research Network. His main critique centers on the fact that better suicide data are available from the CDC, which prospectively collects suicide data from a statistically valid sample of the entire US population. In children and adolescents ages 10 to 17 years, the superior CDC data indicate that attempted suicide rates peaked in 2003 or 2004, and then decreased thereafter. Completed suicides followed a similar pattern, as did intentional nonfatal poisonings. Concluding remarks that were made highlight that continued pursuit of a causal link will “…provide additional examples that achieve little beyond reinforcing concerns about something that did not happen…” because of the fact that there are only so many data available that can be analyzed in so many different ways.

The controversy surrounding antidepressants and suicidality has been debated during the last 10 years and will likely continue as more original research, meta-analyses, and population-level studies are published. More studies on antidepressants conducted to prospectively identify suicidality in patients with MDD and other varying diagnoses need to be completed to truly identify the incidence of suicidal behavior and ideation when initiating antidepressants. There is a risk of suicidality, albeit seemingly low, in children, adolescents, and young adults when initiating antidepressants; there is also risk in not treating depressed patients with antidepressants for whom they are indicated. Conversely, antidepressants prescribed to people 65 years and older have shown decreased suicidality compared with placebo, and antidepressants prescribed to patients ages 25 to 64 years have not shown an increase in suicidality compared with placebo. Clinicians must be cognizant of this and monitor high-risk patients per the FDA recommended guidelines.

1
Teicher
MH
,
Glod
C
,
Cole
JO.
Emergence of intense suicidal preoccupation during fluoxetine treatment
.
Am J Psychiatry
.
1990
;
147
(
2
):
207
-
10
.
2
Mayer-Gross
W
,
Slater
E
,
Roth
M.
Clinical psychiatry. 2nd ed
.
London
:
Cassell;
1960
.
3
US Food and Drug Administration Center for Drug Evaluation and Research
. (
2004
).
Memorandum: background comments for February 2, 2004 meeting of Psychopharmacological Drugs Advisory Committee (PDAC) and Pediatric Subcommittee of the Anti-Infective Drugs Advisory Committee (Peds AC)
[Internet]
.
Silver Spring
:
US Food and Drug Administration
;
2004
[cited
2014
.
4
GlaxoSmithKline
.
Seroxat® Article 31: consolidated response document. Question 2: risk of possibly suicide-related events, self-harm and hostility with paroxetine data from all available sources
[Internet]
.
Brentford, London
:
2004
[cited 2014 Oct 7]. Available from: http://www.gsk.com/media/388729/question_02.pdf
.
5
Carrey
N
,
Virani Pharm A. Suicidal ideation reports from pediatric trials for paroxetine and venlafaxine
.
Can Child Adolesc Psychiatr Rev
.
2003
;
12
(
4
):
101
-
2
.
6
Hammad
TA.
Review and evaluation of clinical data
[Internet]
.
Silver Spring
:
US Food and Drug Administration
;
2004
[updated 2004; cited
2014
.
7
US Food and Drug Administration Center for Drug Evaluation and Research
.
Historical information on antidepressant use in children, adolescents, and adults [Internet]
.
Silver Spring
:
US Food and Drug Administration;
2004
.
8
Hammad
TA
,
Laughren
T
,
Racoosin
J.
Suicidality in pediatric patients treated with antidepressant drugs
.
Arch Gen Psychiatry
.
2006
;
63
(
3
):
332
-
9
. DOI: . PubMed PMID: 16520440.
9
US Food and Drug Administration Center for Drug Evaluation and Research
.
Suicidality in children and adolescents being treated with antidepressant medications: FDA public health advisory [Internet]
.
Silver Spring
:
US Food and Drug Administration;
2004
.
10
US Food and Drug Administration Center for Drug Evaluation and Research
.
Summary minutes of the CDER Psychopharmacologic Drugs Advisory Committee and the FDA Pediatric Advisory Committee Meeting: September 13-14, 2004 [Internet]
.
Silver Spring
:
US Food and Drug Administration;
2004
.
11
US Food and Drug Administration Center for Drug Evaluation and Research
.
Memorandum: overview for December 13 meeting of Psychopharmacologic Drugs Advisory Committee (PDAC) [Internet]
.
Silver Spring
:
US Food and Drug Administration;
2006
.
12
Stone
M
,
Laughren
T
,
Jones
ML
,
Levenson
M
,
Holland
PC
,
Hughes
A
,
et al
.
Risk of suicidality in clinical trials of antidepressants in adults: analysis of proprietary data submitted to US Food and Drug Administration
.
BMJ
.
2009
;
339
(
2
):
b2880
. DOI: .
13
US Food and Drug Administration Center for Drug Evaluation and Research
.
Summary minutes of the CDER Psychopharmacologic Drugs Advisory Committee meeting: December 13, 2006 [Internet]
.
Silver Spring
:
US Food and Drug Administration;
2006
.
14
Adegbite-Adeniyi
C
,
Gron
B
,
Rowles
BM
,
Demeter
CA
,
Findling
RL.
An update on antidepressant use and suicidality in pediatric depression
.
Expert Opin Pharmacother
.
2012
;
13
(
15
):
2119
-
30
. DOI: . PubMed PMID: 22984934.
15
Libby
AM
,
Orton
HD
,
Valuck
RJ.
Persisting Decline in Depression Treatment After FDA Warnings
.
Arch Gen Psychiatry
.
2009
;
66
(
6
):
633
. DOI: .
16
Libby
AM
,
Brent
DA
,
Morrato
EH
,
Orton
HD
,
Allen
R
,
Valuck
RJ.
Decline in treatment of pediatric depression after FDA advisory on risk of suicidality with SSRIs
.
Am J Psychiatry
.
2007
;
164
(
6
):
884
-
91
. DOI: . PubMed PMID: 17541047.
17
Zito
JM
,
Safer
DJ
,
dosReis
S
,
Gardner
JF
,
Soeken
K
,
Boles
M
,
et al
.
Rising prevalence of antidepressants among US youths
.
Pediatrics
.
2002
;
109
(
5
):
721
-
7
. DOI: .
18
Skaer
TL
,
Sclar
DA
,
Robison
LM.
Trends in prescriptions for antidepressant pharmacotherapy among US children and adolescents diagnosed with depression, 1990 through 2001: an assessment of accordance with treatment recommendations from the American Academy of Child and Adolescent Psychiatry
.
Clin Ther
.
2009
;
31
(
Pt 1
):
1478
-
87
. DOI: . PubMed PMID: 19698905.
19
Rushton
JL
,
Whitmire
JT.
Pediatric stimulant and selective serotonin reuptake inhibitor prescription trends
.
Arch Pediatr Adolesc Med
.
2001
;
155
(
5
):
560
. DOI: .
20
Olfson
M
,
Marcus
SC
,
Weissman
MM
,
Jensen
PS.
National trends in the use of psychotropic medications by children
.
J Am Acad Child Adolesc Psychiatry
.
2002
;
41
(
5
):
514
-
21
. DOI: . PubMed PMID: 12014783.
21
Ma
J
,
Lee
KV
,
Stafford
RS.
Depression treatment during outpatient visits by U.S. children and adolescents
.
J Adolesc Health
.
2005
;
37
(
6
):
434
-
42
. DOI: . PubMed PMID: 16310120.
22
Gibbons
RD
,
Hur
K
,
Bhaumik
DK
,
Mann
JJ.
The relationship between antidepressant prescription rates and rate of early adolescent suicide
.
Am J Psychiatry
.
2006
;
163
(
11
):
1898
-
904
. DOI: . PubMed PMID: 17074941.
23
Delate
T
,
Gelenberg
AJ
,
Simmons
VA
,
Motheral
BR.
Trends in the use of antidepressants in a national sample of commercially insured pediatric patients, 1998 to 2002
.
Psychiatr Serv
.
2004
;
55
(
4
):
387
-
91
.
PubMed PMID: 15067149
.
24
Pamer
CA
,
Hammad
TA
,
Wu
YT
,
Kaplan
S
,
Rochester
G
,
Governale
L
,
et al
.
Changes in US antidepressant and antipsychotic prescription patterns during a period of FDA actions
.
Pharmacoepidemiol Drug Saf
.
2010
;
19
(
2
):
158
-
74
. DOI: . PubMed PMID: 20049836.
25
Busch
SH
,
Frank
RG
,
Martin
A
,
Barry
CL.
Characterizing declines in pediatric antidepressant use after new risk disclosures
.
Med Care Res Rev
.
2011
;
68
(
1
):
96
-
111
. DOI: . PubMed PMID: 20675349.
26
Lineberry
TW
,
Bostwick
JM
,
Beebe
TJ
,
Decker
PA.
Impact of the FDA black box warning on physician antidepressant prescribing and practice patterns: opening Pandora's suicide box
.
Mayo Clin Proc
.
2007
;
82
(
4
):
518
-
20
. DOI: . PubMed PMID: 17418083.
27
Valluri
S
,
Zito
JM
,
Safer
DJ
,
Zuckerman
IH
,
Mullins
CD
,
Korelitz
JJ.
Impact of the 2004 Food and Drug Administration pediatric suicidality warning on antidepressant and psychotherapy treatment for new-onset depression
.
Med Care
.
2010
;
48
(
11
):
947
-
54
. DOI: .
28
Kurian
BT
,
Ray
WA
,
Arbogast
PG
,
Fuchs
DC
,
Dudley
JA
,
Cooper
WO.
Effect of regulatory warnings on antidepressant prescribing for children and adolescents
.
Arch Pediatr Adolesc Med
.
2007
;
161
(
7
):
690
. DOI: .
29
Olfson
M
,
Marcus
SC
,
Druss
BG.
Effects of Food and Drug Administration warnings on antidepressant use in a national sample
.
Arch Gen Psychiatry
.
2008
;
65
(
1
):
94
. DOI: .
30
Bhatia
SK
,
Rezac
AJ
,
Vitiello
B
,
Sitorius
MA
,
Buehler
BA
,
Kratochvil
CJ.
Antidepressant prescribing practices for the treatment of children and adolescents
.
J Child Adolesc Psychopharmacol
.
2008
;
18
(
1
):
70
-
80
. DOI: . PubMed PMID: 18294090.
31
Singh
T
,
Prakash
A
,
Rais
T
,
Kumari
N.
Decreased use of antidepressants in youth after US Food and Drug Administration black box warning
.
Psychiatry (Edgemont)
.
2009
;
6
:
30
-
4
.
32
Morrato
EH
,
Libby
AM
,
Orton
HD
,
Degruy
FV
,
Brent
DA
,
Allen
R
,
et al
.
Frequency of provider contact after FDA advisory on risk of pediatric suicidality with SSRIs
.
Am J Psychiatry
.
2008
;
165
(
1
):
42
-
50
. DOI: . PubMed PMID: 17986680.
33
Cheung
A
,
Sacks
D
,
Dewa
CS
,
Pong
J
,
Levitt
A.
Pediatric prescribing practices and the FDA black-box warning on antidepressants
.
J Dev Behav Pediatr
.
2008
;
29
(
3
):
213
-
5
. DOI: . PubMed PMID: 18550990.
34
Ali
MM
,
Teich
J
,
Woodward
A
,
Han
B.
The implications of the Affordable Care Act for behavioral health services utilization. Adm Policy Ment Health
.
Published online
:
November 9
,
2014
.
35
Gibbons
RD
,
Brown
CH
,
Hur
K
,
Marcus
SM
,
Bhaumik
DK
,
Erkens
JA
,
et al
.
Early evidence on the effects of regulators' suicidality warnings on SSRI prescriptions and suicide in children and adolescents
.
Am J Psychiatry
.
2007
;
164
(
9
):
1356
-
63
. DOI: . PubMed PMID: 17728420.
36
Olfson
M
,
Shaffer
D
,
Marcus
SC
,
Greenberg
T.
Relationship between antidepressant medication treatment and suicide in adolescents
.
Arch Gen Psychiatry
.
2003
;
60
(
10
):
978
. DOI: .
37
Centers for Disease Control and Prevention
.
National suicide statistics at a glance: trends in suicide rates among both sexes, by age group, United States, 1991–2009 [Internet]
.
Atlanta
:
Centers for Disease Control and Prevention [cited
2014
.
38
Wolitzky-Taylor
KB
,
Ruggiero
KJ
,
McCart
MR
,
Smith
DW
,
Hanson
RF
,
Resnick
HS
,
et al
.
Has adolescent suicidality decreased in the United States? Data from two national samples of adolescents interviewed in 1995 and 2005
.
J Clin Child Adolesc Psychol
.
2010
;
39
(
1
):
64
-
76
. DOI: . PubMed PMID: 20390799.
39
Lu
CY
,
Zhang
F
,
Lakoma
MD
,
Madden
JM
,
Rusinak
D
,
Penfold
RB
,
et al
.
Changes in antidepressant use by young people and suicidal behavior after FDA warnings and media coverage: quasi-experimental study
.
BMJ
.
2014
;
348
:
g3596
. DOI: . PubMed PMID: 24942789.
40
Carpenter
DJ
,
Fong
R
,
Kraus
JE
,
Davies
JT
,
Moore
C
,
Thase
ME.
Meta-analysis of efficacy and treatment-emergent suicidality in adults by psychiatric indication and age subgroup following initiation of paroxetine therapy
.
J Clin Psychiatry
.
2011
;
72
(
11
):
1503
-
14
. DOI: .
41
Hetrick
SE
,
McKenzie
JE
,
Cox
GR
,
Simmons
MB
,
Merry
SN.
Newer generation antidepressants for depressive disorders in children and adolescents
.
Cochrane Database Syst Rev
.
2012
;
11:CD004851
. DOI: . PubMed PMID: 23152227.
42
Julious
SA.
Efficacy and suicidal risk for antidepressants in paediatric and adolescent patients
.
Stat Methods Med Res
.
2013
;
22
(
2
):
190
-
218
. DOI: .
43
Gibbons
RD
,
Brown
CH
,
Hur
K
,
Davis
JM
,
Mann
JJ.
Suicidal thoughts and behavior with antidepressant treatment
.
Arch Gen Psychiatry
.
2012
;
69
(
6
):
580
. DOI: .
44
Bridge
JA
,
Iyengar
S
,
Salary
CB
,
Barbe
RP
,
Birmaher
B
,
Pincus
HA
,
et al
.
Clinical response and risk for reported suicidal ideation and suicide attempts in pediatric antidepressant treatment: a meta-analysis of randomized controlled trials
.
JAMA
.
2007
;
297
(
15
):
1683
-
96
. DOI: . PubMed PMID: 17440145.
45
Leon
AC.
The revised warning for antidepressants and suicidality: unveiling the black box of statistical analyses
.
Am J Psychiatry
.
2007
;
164
(
12
):
1786
-
9
. DOI: . PubMed PMID: 18056231.
46
US Food and Drug Administration Center for Drug Evaluation and Research
.
Guidance for industry: suicidal ideation and behavior: prospective assessment of occurrence in clinical trials [Internet]
.
Silver Spring
:
Food and Drug Administration;
2012
.
47
Friedman
RA.
Antidepressants' black-box warning--10 years later
.
N Engl J Med
.
2014
;
371
(
18
):
1666
-
8
. DOI: . PubMed PMID: 25354101.
48
Stone
MB.
The FDA warning on antidepressants and suicidality--why the controversy?
N Engl J Med
.
2014
;
371
(
18
):
1668
-
71
. DOI: . PubMed PMID: 25354102.