This increase in available evidence triggered the recent controversy about the safety and the efficacy of antidepressants, in particular SSRIs fluoxetine, citalopram, escitalopram, paroxetine and sertraline and SNRIs mirtazapine, nefazodone and venlafaxine.

Following many months of reviewing the safety data, regulatory agencies in the United Kingdom, the United States and Canada have issued warnings about the increased risk of suicidal behaviour in children and adolescents using these medications.

Yet, untreated depression is also problematic. In the recent Treatment for Adolescents With Depression Study TADS , conducted at 13 centres in the United States, specific psychotherapy cognitive behavioural therapy was not found to be more effective than placebo.

Therefore, drug therapy may continue to play an important role in the treatment of major depression in children and adolescents.

Understanding the efficacy and safety profiles of antidepressants is therefore important. Yet, for clinicians, interpreting the data is often difficult given the competing opinions on both efficacy and safety.

In this article, we will review the evidence from published and unpublished RCTs on the efficacy of antidepressants in treating major depressive disorder in children and adolescents and on the emergence of suicidality suicidal thinking and behaviour associated with antidepressants.

In addition, we will discuss the methodologic differences between the RCTs reviewed and how these differences may have contributed to the variation in outcomes reported.

This discussion should make it easier for clinicians to evaluate the available data. We obtained data for review from published RCT reports in peer-reviewed journals and meeting abstracts and from unpublished RCTs in the public domain available from the Web sites of the FDA, the UK Medicines and Healthcare products Regulatory Agency [MHRA] and GlaxoSmithKline.

Clinical trials in this review included large RCTs of antidepressants involving subjects less than 18 years old with depression that were identified through the extensive safety review conducted by the FDA in The FDA reviewed 6 published reports 3 of fluoxetine 13 — 15 and 1 each of paroxetine, 16 sertraline 17 and citalopram 18 and 10 unpublished studies 2 each of paroxetine, 19 , 20 venlafaxine, 21 nefazodone 22 and mirtazapine [unpublished data on file with Organon Inc.

Table 1 provides the primary outcomes and response rates CGI-I scores reported in each of the 16 SSRI and SNRI trials. The results of the SSRI trials were mixed except for those of the 3 fluoxetine trials.

Fluoxetine is the only drug approved for the treatment of depression in children and adolescents. The most recent study TADS compared fluoxetine alone, cognitive behavioural therapy alone, fluoxetine plus cognitive behavioural therapy, and placebo alone in subjects aged 12—17 years over 12 weeks.

Despite widespread scrutiny of the outcomes of many of the antidepressant trials, the results of all 3 of the fluoxetine trials were considered positive by the MHRA and the FDA. The other SSRI trials reported mixed results.

In the published trial of paroxetine, 16 many of the secondary outcomes, including CGI-I scores, were positive, but no significant difference between patient groups was found in the primary outcome. In this study, the Hamilton Depression Rating Scale HAMD , an adult rating scale, was used as the primary outcome measure.

The findings led researchers to determine that adult measures are not appropriate for pediatric populations. This study was the only RCT to use another drug imipramine as a comparator.

Imipramine was not found to be significantly better than placebo in any of the outcomes. Two identical studies were conducted in 53 centres in the United States, India, Canada, Costa Rica and Mexico to compare sertraline and placebo for 10 weeks.

Neither of the individual studies was positive for sertraline; however, the drug—placebo difference did reach significance for the combined data. Although the trials were not powered to detect differences by age group, there was some suggestion that sertraline may have been more effective among the adolescents than among the children in the study.

There were 2 placebo-controlled studies of citalopram. Many studies use a CDRS-R score of 28 or less as a response criterion; thus, the low response rates in this study are consistent with the remission rates reported in other studies.

The second citalopram study showed no evidence of efficacy. A recent placebo-controlled trial of escitalopram 25 showed no difference between the treatment group and the placebo group in the primary outcome measure of CGI-I scores. In summary, there appears to be some evidence of efficacy for paroxetine, sertraline and citalopram, but fluoxetine was found to be efficacious in more than 1 RCT.

Two RCTs were conducted of each of the SNRIs nefazodone, venlafaxine and mirtazapine for the treatment of pediatric depression. One of the nefazodone trials demonstrated efficacy on some of the outcome measures but not on the primary outcome measure change in CDRS-R score.

reported negative results owing to high response rates in the placebo groups. When the results of the venlafaxine trials were pooled, there was some evidence of benefit among adolescents. The studies used a variety of outcome measures and response criteria. In general, the RCTs involved a standard assessment of symptoms by a clinician during an interview with the patient and his or her parents.

Earlier studies involving adults with depression used adult rating scales. Later studies involving children and adolescents used child-specific scales e. In addition, the clinical assessment of global improvement i. The CGI is the only outcome that was consistent across all of the RCTs reviewed.

Regulatory agencies such as the FDA require 2 positive trials before it will declare a medication effective. There are methodologic issues that likely had an impact on the results of the 10 studies.

More specific details about the methodology of these studies can be found in a review of the study designs used by the FDA Advisory Committee to evaluate the studies in September www.

We will review some of the major methodologic issues here. Many of the trials that failed to show evidence of efficacy did so because the placebo response rate was so high. Several factors may have contributed to these high response rates.

It is unclear in these cases whether the trial results were negative because the medication was not effective unlikely, since the response rate was also high in the treatment groups or whether there were methodologic issues leading to high response rates for all of the subjects.

Because the trials did not use another drug as an active comparator, it is impossible to determine the cause. The first difference between the trials was the number of sites involved. Increasing the number of sites in a trial increases the variability in the conduct of the study, which may influence outcomes.

Similarly, the number of subjects per site is relevant. With more subjects per site, investigators become more familiar with the study protocol and intervention strategies. Another difference is the experience of the site staff and investigators in a assessing and treating depression in children and adolescents and b using rating scales for outcome measures.

As suggested by Wagner and colleagues in the sertraline study, 17 the possible reasons for the high placebo rate in that study included the experience of the investigators, the large number of participating sites, the low number of subjects at each site and the fact that the sites were in different countries.

Another area in which the trials differed was patient selection. Although most of the studies recruited only subjects who met Diagnostic and Statistical Manual of Mental Disorder criteria for major depressive disorder of at least moderate severity, other inclusion and exclusion criteria varied, including age, severity of illness, comorbid disorders, and ancillary pharmacologic and psychotherapeutic interventions.

For example, some of the trials recruited only adolescents, whereas others included children and adolescents. Some of the studies that reported negative findings showed evidence of efficacy among the adolescents but not among the children. In a study of cognitive behavioural therapy for depression in adolescents, Brent and colleagues reported higher response rates among subjects who learned of the study through advertisements than among those who were clinical referrals.

Another factor that may have affected outcomes is study design. Duration of assessment, use of placebo run-in periods, duration of acute treatment and dosing all affect results.

For example, Rintelmann and colleagues reported that an extended evaluation period 2 weeks led to improvement in some subjects during the course of the evaluation. Two of the fluoxetine studies used placebo run-in periods, and the third the TADS used an extended evaluation period without a placebo run-in period; all 3 reported low placebo response rates.

Finally, special attention must be paid to the selection of the primary outcome measure, since this is the focus of regulatory agencies. However, in reviewing the studies, a decision about whether a study's findings are positive or negative cannot always be based strictly on the primary outcome measure.

Future trials should consider having standard outcome measures, which would allow for easier comparison and interpretation of results across studies. Suicidal behaviour has been reported in children and adolescents prescribed antidepressants, both in case reports and in clinical trials.

One difficulty in interpreting suicidal behaviour in depression studies is that suicide attempts and suicidal ideation are common symptoms of depression. If a suicide attempt is made during the course of treatment, it is difficult to identify the cause or causes of the event.

It could be a lack of improvement or worsening of depressive symptoms, increased activation increased energy either from improvement in the mood disorder or due to the medication , or it could be directly linked to the medication. The use of a placebo control group helps to answer some but not all of the questions.

The FDA conducted an independent examination of the adverse event data from all clinical trials of SSRIs and SNRIs, both for depression and for all indications. Table 2 presents the results of the FDA analysis of suicide-related events attempts or ideation in pediatric RCTs of antidepressant therapy for major depressive disorder and for all indications.

For all indications, the relative risk of suicide-related events was significantly increased among subjects given medication relative risk [RR] 1.

The same was true in the trials of therapy for major depressive disorder RR 1. Because of the relative rarity of these events 97 among more than children and adolescents included in the trials , the difference was only significant when data from all of the trials were pooled.

Except for venlafaxine, individual medications were not statistically more likely than other medications to lead to suicidal behaviour. No suicides occurred in these trials. Contrary to the results of the clinical trials, data from observational studies involving both adults and youths suggest an inverse relation between treatment with antidepressants and suicidality.

During the late s, the incidence of suicide among youths in the United States decreased as the number of antidepressant prescriptions increased. A more recent epidemiologic study showed an association between increased prescription of antidepressants and decreased incidence of completed suicides among youths in the United States after adjustment for such factors as physician availability.

Valuck and colleagues conducted a retrospective cohort study to examine the link between antidepressant treatment and suicide attempts among patients aged 12—18 with newly diagnosed depression. The results indicated that antidepressant therapy did not increase the risk of suicide attempts and that such therapy taken for more than days reduced the likelihood of suicide attempts compared with treatment taken for less than 55 days.

Leon and colleagues examined serum toxicology results in cases of suicide involving youths less than 18 years of age in New York City between and Only 4 of the 58 youths tested positive for antidepressants imipramine 2, fluoxetine 2.

Similar results have been found in postmortem studies of completed suicides by adults. Therefore, there is emerging evidence from observational studies that contradicts the findings of the FDA's review. It is difficult to reconcile the epidemiologic data showing decreased numbers of completed suicides with the data from the clinical trials showing increased suicidal ideation and behaviour among youths prescribed antidepressants.

Increases in suicidal ideation and behaviour are relatively common in depression, whereas completed suicides are far less common. It is possible that slight increases in ideation do not, in fact, result in an increase in actual completed suicides. The reliability of the data on suicidal behaviours in these trials was greatly improved when the suicide-related events were reviewed by the FDA.

The original data, which were based on classifications by the drug manufacturers, clearly had some issues with inadequate labelling of the events. However, even the FDA review was based on site-level documentation.

Although some trials had relatively elaborate descriptions of the events, others provided only minimal information. In addition, any prior suicidal behaviours of the subjects were not reported. Because suicidal behaviours are generally highly associated with suicide-related events, 39 it is crucial for studies to report both the adverse event as well as the known history of similar behaviours of these subjects.

From our review, only one trial collected and reported this information. Suicidal ideation accounted for the majority of the events, and there were no completed suicides among the youths included in the trials.

Given the methodologic differences between the trials and the range of drug and placebo response rates reported, clinicians need to be judicious in their interpretation of data from pediatric antidepressant trials.

The current recommendations for antidepressant therapy in children and adolescents are being debated, and many question are being raised about the strength of the evidence supporting them.

In this evolving field, outcome measures have varied widely across studies. Finding the most effective outcome measures remains a challenge. The results of most trials are not definitive all positive or all negative. Clinicians, therefore, must evaluate the overall trial. Evidence from SSRI trials involving adults indicates that the drugs are equally efficacious.

It is unlikely that these drugs will behave any differently in youths. Does that mean that there is different efficacy among placebos? We would suggest that the differences are likely the result of methodologic issues between studies rather than a true difference between the medications.

Finally, the most controversial issue involves the rates of suicide-related events. It is important to distinguish between suicidal ideation and completed suicides, or even suicide attempts.

As has been known for years, children and adolescents with depression requiring drug therapy need to be closely monitored, particularly during the early stages of treatment or after dose escalations.

Clinicians need to be informed about the nature of the events reported in the clinical trials in order to adequately describe the benefits and risks of antidepressants to patients and their families.

Further details regarding the FDA's review and the data presented here can be found on the FDA's Web site www. Clinicians are faced with the difficult task of treating depression in this population.

Data from clinical trials have suggested a possible association between worsening or new suicidal behaviour in youths prescribed antidepressants.

However, no predictive factors have been determined. Therefore, clinicians need to be vigilant about the emergence of these adverse events.

Other important points for clinicians who care for children and adolescents with major depression include the following:. Advocacy agencies e. They should ask about suicidal behaviours at all subsequent visits. Drug therapy should be started at a low dose equivalent of 5—10 mg of fluoxetine and the dose increased every 2 weeks until the maximum dose is reached if no significant adverse events emerge.

Families should be asked to monitor patients closely for worsening depression, worsening or new suicidal ideation or behaviours, and other behavioural side effects. To aid in this process, families may wish to use tools available through such organizations as the National Alliance on Mental Illness www.

org , Families for Depression Awareness www. org and the Canadian Mental Health Association www. Contributors: All of the authors contributed to the development and writing of the manuscript and approved the final version to be published.

Competing interests: None declared for Taryn Mayes. Amy Cheung has received speaker fees from Eli Lilly and Company. Graham Emslie has received research grants from Eli Lilly, Organon Inc.

and Forest Laboratories Inc. and Wyeth-Ayerst; and is with the Speaker Bureau of McNeil-PPC, Inc. NOTE: We only request your email address so that the person you are recommending the page to knows that you wanted them to see it, and that it is not junk mail.

We do not capture any email address. Skip to main content. Amy H. Cheung , Graham J. Emslie and Taryn L. Abstract The dramatic increase over the past 10 years in the amount of available clinical research on the use of antidepressants to treat major depression in children and adolescents has substantially improved our knowledge of the safety and efficacy of these medications in the pediatric population.

Methods We obtained data for review from published RCT reports in peer-reviewed journals and meeting abstracts and from unpublished RCTs in the public domain available from the Web sites of the FDA, the UK Medicines and Healthcare products Regulatory Agency [MHRA] and GlaxoSmithKline. Efficacy Selective serotonin reuptake inhibitors Table 1 provides the primary outcomes and response rates CGI-I scores reported in each of the 16 SSRI and SNRI trials.

View this table: View inline View popup Download powerpoint. Table 1. Serotonin—norepinephrine reuptake inhibitors Two RCTs were conducted of each of the SNRIs nefazodone, venlafaxine and mirtazapine for the treatment of pediatric depression. Methodologic issues regarding efficacy outcomes The studies used a variety of outcome measures and response criteria.

Safety Suicidality Suicidal behaviour has been reported in children and adolescents prescribed antidepressants, both in case reports and in clinical trials. Table 2. Methodologic issues regarding safety outcomes The reliability of the data on suicidal behaviours in these trials was greatly improved when the suicide-related events were reviewed by the FDA.

Interpretation Given the methodologic differences between the trials and the range of drug and placebo response rates reported, clinicians need to be judicious in their interpretation of data from pediatric antidepressant trials.

Footnotes This article has been peer reviewed. Childhood and adolescent depression: a review. Harv Rev Psychiatry ; 3 : OpenUrl PubMed. Fleming JE, Offord DR, Boyle MH. Prevalence of childhood and adolescent depression in the community. Treatment and ongoing management.

Cipriani A, Furukawa TA, Salanti G, et al. Comparative efficacy and acceptability of 21 antidepressant drugs for the acute treatment of adults with major depressive disorder: a systematic review and network meta-analysis. Brent DA, Gibbons RD, Wilkinson P, et al.

Antidepressants in paediatric depression. BJPsych Bull. Locher C, Koechlin H, Zion SR, et al. Efficacy and safety of selective serotonin reuptake inhibitors, serotonin-norepinephrine reuptake inhibitors, and placebo for common psychiatric disorders among children and adolescents. JAMA Psychiatry.

Brent D, Emslie G, Clarke G, et al. Switching to another SSRI or to venlafaxine with or without cognitive behavioral therapy for adolescents with SSRI-resistant depression: the TORDIA randomized controlled trial [published correction appears in JAMA.

American Academy of Child and Adolescent Psychiatry. Severe shortage of child and adolescent psychiatrists illustrated in AACAP workforce maps. Accessed December 28, In Reply: We agree with Drs. Anvari, Carroll, and Klein in recommending a nuanced approach to depression treatment, taking into account patient comorbidities; drug-drug interactions; adverse effect profiles; prior responses or preference for a particular antidepressant; and the availability of school, peer, and community resources.

However, treating children and adolescents with complex psychiatric or medical conditions was beyond the scope of our article. Evidence supports the use of fluoxetine and escitalopram as first-line agents for unipolar depression in children and adolescents without complex medical or psychiatric histories.

We believe in basing recommendations on the best evidence available, and the evidence for sertraline and venlafaxine is not as compelling as that for fluoxetine and escitalopram. Evidence showing that sertraline is more effective than placebo is based on two trials that were reported in one publication as if they were one trial.

The analysis broke randomization protocols and ignored potential between-trial variability heterogeneity , which could have inflated the treatment effect.

We agree that some primary care physicians may feel comfortable prescribing antidepressants that are not approved for use in children and adolescents. However, we believe physicians should consult with or refer patients to mental health specialists whenever they are uncomfortable with this.

We acknowledge the shortage of child and adolescent mental health specialists and that referral may not be easy e. an in-person visit 4. But at a minimum, phone consultation with a mental health specialist should occur whenever the primary care physician lacks the comfort and expertise needed to appropriately treat patients when first-line therapies have not been successful.

We agree that safety plans for at-risk children and adolescents should include limiting access to traditional weapons, as well as household items that could be used to inflict harm e. Wagner KD, Ambrosini P, Rynn M, et al. Efficacy of sertraline in the treatment of children and adolescents with major depressive disorder: two randomized controlled trials.

Efficacy and safety of selective serotonin reuptake inhibitors, serotonin-norepinephrine reuptake inhibitors, and placebo for common psychiatric disorders among children and adolescents: a systematic review and meta-analysis.

Totten AM, Hansen RN, Wagner J, et al. Telehealth for acute and chronic care consultations. Comparative effectiveness review no.

Accessed May 18, Email letter submissions to afplet aafp. Letters should be fewer than words and limited to six references, one table or figure, and three authors. Letters submitted for publication in AFP must not be submitted to any other publication.

Letters may be edited to meet style and space requirements.