Treatment of Adolescents With Major Depression: Title and subTitle BreakContributions of a Major Trial

The publication of the results of the Treatment for Adolescents With Depression Study¹ (TADS) in this issue of JAMA is certainly timely in view of recent and widely publicized controversies regarding treating adolescents with depression.^{2 - 4} Beyond that, these results shed new light on some older controversies about psychiatric treatments. As is usually the case with good research, the study provides important answers but also raises several important questions.

In the TADS study, a randomized controlled trial funded by the National Institute of Mental Health and conducted at 13 academic and community centers in the United States, 429 patients aged 12 to 17 years (mean age, 15 years), who had a primary diagnosis of major depressive disorder, provided written informed consent (along with consent from at least one of their parents) to participate in the study. Participants on average had moderate to severe symptom levels based on scores on the Children's Depression Rating Scale-Revised⁵ and 27% had at least minimal suicidal ideation at baseline. The average depressive episode duration was 72 weeks, adding to the evidence that these mood disturbances were not temporary fluctuations. Potential participants were excluded if they had been hospitalized for dangerousness within the preceding 3 months, had made a suicide attempt causing medical attention within 6 months, had clear suicidal intent or an active plan for suicide, or if they had suicidal ideation with a disorganized family believed by the investigators to be unable to guarantee adequate safety monitoring. Such exclusions are reasonable and necessary for research participation, but unfortunately limit the controlled research evidence base available to clinicians treating depressed patients with the highest potential danger to themselves.

Participants were randomly assigned to 1 of 4 treatments for 12 weeks: fluoxetine with clinical management (6 physician visits lasting 20-30 minutes to monitor clinical status and medication effects), pill placebo with clinical management, cognitive-behavioral therapy (CBT), or the combination of fluoxetine with CBT. Fluoxetine and equivalent placebo doses could be adjusted from a starting dose of 10 mg/d to 40 mg/d; the mean highest dose received by participants was about 30 mg/d in the fluoxetine alone, fluoxetine with CBT, and placebo groups. The CBT was planned for 15 sessions over 12 weeks and was designed to restructure the negative thought patterns typical of depression and to increase active, positively reinforcing behaviors. In an important consideration for treatment of adolescents, parent and family sessions were integrated with individual CBT. The fluoxetine alone and placebo treatment groups were double-blind, but participants and their clinicians in the fluoxetine with CBT group knew they were receiving active medication. Those receiving CBT, either alone or combined with fluoxetine, knew they were receiving an active psychotherapy because there was no placebo therapy control group receiving equivalent amounts of time and contact. Outcomes were assessed by independent evaluators who were blind to treatment assignments.

After 12 weeks of treatment, total scores on the Children's Depression Rating Scale-Revised as a measure of depression symptoms based on interviews with adolescents and their parents indicated that the combination of fluoxetine with CBT was significantly superior to placebo, fluoxetine alone, and CBT alone. Fluoxetine alone was significantly better than CBT alone, and also better than placebo, although the latter was true only based on the week 12 adjusted mean scores and not on an analysis of the changes over time across the 12 weeks. Based on a global rating of being much improved or very much improved, response rates were 71% in the fluoxetine with CBT group, 61% in the fluoxetine alone group, 43% in the CBT alone group, and 35% in the placebo group.

In view of concerns about possible increases in suicidal behavior in children receiving selective serotonin reuptake inhibitors (SSRIs),² including recent warnings from the US Food and Drug Administration³ and the British Medicines and Healthcare Products Regulatory Agency,⁴ the safety data provided by TADS are of particular interest. Suicidal ideation (measured by a self-report questionnaire⁶ ) declined in all of the treatment groups. About 6% of the patients experienced a suicide-related event defined as either worsening suicidal ideation or making a suicide attempt, with no statistically significant difference among the 4 treatment groups. Seven patients made a suicide attempt and there were no completed suicides. A broader measure of harm-related adverse events that included any self-harm or harm to another person or property indicated an increased risk (odds ratio, 2.19; 95% confidence interval, 1.03-4.62) for patients receiving fluoxetine compared with those who were not, with the suggestion of a protective effect for CBT because the odds ratio was higher for fluoxetine alone compared with fluoxetine with CBT.

What answers does the TADS report provide at this point regarding treatment of adolescents with major depression? Perhaps most important is the finding that treatment of carefully evaluated adolescents with moderate to severe major depression can be effective within 12 weeks, even for patients with the psychiatric comorbidities commonly encountered in clinical practice. About half of the participants had comorbid disorders such as dysthymia (long-term mild depression), anxiety disorders, disruptive behavior disorder, and attention-deficit/hyperactivity disorder. Patients with bipolar disorder, severe conduct disorder, current substance abuse or dependence, or evidence of a psychotic disorder were appropriately excluded because they would need different treatments not provided in the study.

Another answer that is intuitively appealing, and is certainly in line with controlled studies of adult patients with major depression,^{7 - 9} is suggested by the finding that combination treatment using fluoxetine with CBT had the best outcomes, whether measured by symptom ratings or the global response rate of 71%. This conclusion must be qualified by the design limitation of open treatment with fluoxetine in the combined treatment group, meaning that positive expectational affects from receiving an active medication could confound comparisons with the CBT alone group and the double-blind fluoxetine alone and placebo groups. However, one could also argue that open treatment with both medication and psychotherapy essentially imitates the conditions of combined treatment in nonresearch clinical practice.

The efficacy of fluoxetine alone (61% response rate), while less than that of combined treatment, was similar to the results of previous trials in children and adolescents, and comports with fluoxetine having received approval from the US Food and Drug Administration for labeling and marketing for treatment of depression in children and adolescents—the only SSRI that currently has such approval. Although the TADS results certainly do not provide a definitive answer to the recent concerns about SSRI treatment and suicidality among adolescents,^{10 - 11} it is of interest and clinical importance that suicidal ideation decreased in all 4 treatment groups, with the suggestion of a particularly beneficial effect for CBT.

What questions are raised by the TADS results? One important question is whether fluoxetine is unique among the SSRIs in having beneficial effects for major depression in children and adolescents. Fluoxetine's long half life does give it an unusual pharmacokinetic profile among the SSRIs, but other than the expected lower incidence of withdrawal affects this has not been observed to confer any within-class efficacy advantage in treating major depression among adults.¹² Answers to this question are clouded by the nonpublication or omission of important data from controlled trials evaluating SSRIs,^{13 - 15} certainly reinforcing the need for a trial registry to ensure the availability of the full results of controlled trials regardless of their sponsorship or outcome.^{16 - 17} Based on the currently available evidence, it appears that clinicians should use fluoxetine as the first choice drug if they decide that antidepressant drug treatment is indicated for an adolescent with major depression, unless there is good individual patient evidence to the contrary such as a history of a poor response to fluoxetine or a good response to a different antidepressant.

Another open question is whether CBT alone is effective in treating adolescents with moderate to severe major depression. As noted by the authors, the 43% response rate for CBT alone by 12 weeks was lower than reported in some previous studies of that form of psychotherapy, possibly due to the higher level of severity and chronicity (mean episode duration of 72 weeks in the participants assigned to CBT alone) of patients in the TADS study or to lower levels of positive expectancy for CBT in this randomized trial. Of interest in this regard is a recently published trial reporting effectiveness of interpersonal psychotherapy, which is an individual psychotherapy that is focused on interpersonal problems associated with depression, for depressed adolescents.¹⁸ The use of different measures does not allow direct comparisons, but it appears that the patients in the interpersonal psychotherapy trial, who were treated in school-based health clinics and not all of whom met criteria for major depression, had less severe depressive disorders than the patients in the TADS trial.

As previously noted, the TADS results raise a mixture of questions and answers regarding the complicated issue of whether antidepressant drug treatment prescribed to treat depression might increase the risk for depression's most dangerous outcome, suicidal behavior. The reduction of suicidal ideation over time is reassuring, but the increased risk for harm-related adverse events observed in the patients receiving fluoxetine suggests that the increase in activation, irritability, or disinhibition sometimes associated with SSRI treatment^{12 ,19} warrants careful monitoring in clinical practice. However, the TADS results leave this as a question rather than a firm answer because those specific symptoms were rare among the patients receiving fluoxetine. Future results from the ongoing stages of the TADS trial should provide important data regarding the long-term efficacy and safety of fluoxetine and CBT treatment for depressed adolescents.

The findings of the TADS study regarding SSRI treatment of depression and suicide risk support the conclusion reached by Boland and Keller^{12 (p856)} in a current psychopharmacology textbook: "Although some concerns about potentiating suicidal behavior may remain, these should be balanced over the clear risk of suicide in patients with untreated depression." The TADS results do not provide clinicians with direct evidence about treatment for depressed adolescents deemed to be at high suicide risk. The same is true for virtually all controlled trials because of the ethical problems of enrolling such patients in a randomized trial. Ironically, the evidence base for treatment effectiveness is the most meager for the most worrisome patients. Preventing suicide is obviously an important goal in treating depression. Prudent clinical practice should include assessment and monitoring of patients for suicidal risk, particularly during periods of psychosocial stress and at times of increased activation or irritability whether or not associated with medication.

The TADS findings provide some new perspectives on several other issues in psychiatric treatment and research. One controversial issue²⁰ has been the use of a placebo control group in randomized trials of treatment for a disorder in which a proven therapy is available (as is the case regarding fluoxetine for major depression in adolescents), particularly for potential participants, such as minors, who require proxy consent.²¹ The TADS study illustrates that with fully informed consent and careful monitoring to guard against undetected clinical decline, inclusion of a placebo group in randomized trials can be acceptable and important to document the efficacy and safety of specific treatments. The TADS study thus provides an excellent example of the wisdom of the 2002 clarification of the World Medical Association Declaration of Helsinki noting that a placebo-controlled trial may be ethically acceptable even if proven therapy is available under several circumstances, including "Where for compelling and scientifically sound methodological reasons its use is necessary to determine the efficacy or safety of a prophylactic, diagnostic, or therapeutic method."²²

Another area of lingering controversy is the role of psychotherapy in the treatment of psychiatric disorders, including major depression. It is gratifying that the TADS study provides a good illustration of how this field has progressed from the highly polarized beliefs of the past to assessment of carefully controlled data.²³ As in other areas of health care, good empirical data always trump beliefs and ideology.

Probably the most important message from TADS is that carefully assessed, empirically validated treatments are available for adolescents with major depression. A corollary of that message is that depressive illness is a major public health problem with substantial morbidity and mortality for adolescents as well as for adults.²⁴ The results from this major new trial¹ demonstrate that although treatment of a depressive illness is often successful and gratifying for patients and clinicians, such success typically requires more than a brief visit for prescription of medication. Rather, it requires careful assessment and monitoring in the context of an ongoing patient-physician relationship. Furthermore, the current evidence suggests that the likelihood of a good outcome is enhanced by the combination of appropriate and carefully monitored drug treatment with an empirically validated psychotherapy.