Crisis of Confidence: Antidepressant Risk Versus Benefit
Andrew A. Nierenberg, MD (Chair)
Department of Psychiatry, Harvard Medical School and Massachusetts General Hospital, Boston
Andrew C. Leon, PhD
Departments of Psychiatry and Public Health, Weill Cornell Medical College, New York, New York
Lawrence H. Price, MD
Department of Psychiatry and Human Behavior, Brown University and Butler Hospital, Providence, Rhode Island
Richard C. Shelton, MD
Department of Psychiatry, Vanderbilt University Medical Center, Nashville, Tennessee
Madhukar H. Trivedi, MD
Mood Disorders Research Program and Clinic, University of Texas Southwestern Medical Center, Dallas
Two recent events in the field of psychiatry have captured the attention of the lay media: (1) the results of research examining the efficacy of antidepressants for patients with mild-to-moderate depression and (2) the FDA requirement that antidepressants carry a black box warning regarding suicidal thinking and behavior in patients younger than 25 years. The media’s interpretations of these events have affected public perception of the safety and utility of antidepressants and raised challenging questions for clinicians who treat patients with these medications. Andrew A. Nierenberg, MD, assembled a panel of nationally recognized experts in psychiatry, consisting of Andrew C. Leon, PhD; Lawrence H. Price, MD; Richard C. Shelton, MD; and Madhukar H. Trivedi, MD, to discuss these issues.
Interpreting the Data From Meta-Analyses
Dr Nierenberg began the discussion by referencing 3 meta-analyses1–3 that have been interpreted in the lay press in a manner that emphasizes the risks of antidepressants while minimizing the benefits. The challenge for clinicians, researchers, and educators is to communicate to the public the true import of the findings. Part of interpreting the data from these meta-analyses is to consider the strengths and limitations of randomized controlled trials and to examine how clinically applicable the results of the trials are.
Much of the recent media attention has focused on a meta-analysis by Dr Shelton and colleagues (Fournier et al),1 which was conducted in response to earlier meta-analyses by Kirsch et al2 and Khan et al3 (AV 1). Kirsch et al2 suggested that aggregated data from all industry-sponsored trials submitted to the FDA found a minimal difference in efficacy between antidepressants and placebo for those with severe depression (initial HDRS scores > 28) and no drug-placebo difference for patients with moderate depression. Khan et al3 reported that baseline severity had no clinically significant effect on drug-placebo differences. Dr Shelton commented, however, that many industry-related clinical trials use exclusionary criteria, which can homogenize the severity of depression in samples, and placebo wash-out periods, which eliminate placebo responders prior to randomization. Dr Shelton et al focused on placebo-controlled trials that included patients with a broad range of depression severity.
Dr Shelton explained that he and his colleagues found a relationship between the baseline severity of illness and the degree of difference between the responses to drug versus placebo. Among patients with baseline HDRS scores of ≥ 25, the drug-placebo efficacy difference was clinically significant.1
Dr Price noted that the results of this meta-analysis1 reinforced what most clinicians think about how antidepressants work and how they should be used, ie, antidepressants are effective in those who are seriously depressed, but the media focused on the finding that antidepressants have less efficacy in milder depression and interpreted the study as more evidence that antidepressants are not effective medications. Dr Shelton agreed, adding that he noticed 2 recurring themes from the lay press regarding psychiatry: (1) that psychiatric disorders (especially depression) are not taken seriously, and (2) that patients should be treated with psychotherapy, not medications.
Dr Trivedi emphasized that meta-analyses are helpful in examining specific aspects of a question, while more definitive answers require prospective studies. As meta-analyses purposely exclude certain data in the literature to answer very focused questions, they should not then be used to produce interpretations about evidence that was not evaluated. Dr Shelton added that meta-analyses are not, in many ways, very informative for clinical practice, but rather illuminate what needs to be addressed in future studies.
Antidepressants and Suicidality
Dr Nierenberg then focused on the controversy surrounding evidence that suggests that suicidal ideation and behavior is more likely in young patients taking antidepressants than in those receiving placebo (AV 2).4 In a recent US House Committee on Veterans’ Affairs hearing5 regarding antidepressants and suicide in military personnel, Dr Leon testified on the results of meta-analytic analyses from the FDA pediatric and adult antidepressant trials.6,7 His recommendation was to continue to use antidepressants to treat veterans and active military personnel with depression, while closely monitoring for suicidality, because depression itself confers a risk for suicide.
The panel agreed that the issue can be complicated by the use of the term suicidality, which can denote a range of behaviors, including suicidal ideation, intent, or actions. Additionally, the black box warning in package inserts states that “patients of all ages who are started on antidepressant therapy should be monitored for clinical worsening,” even though the RCT data reviewed by the FDA only showed a risk for adolescents and young adults.
The panel agreed that patients with mild depression should receive an evidence-based psychotherapy, rather than being immediately treated with an antidepressant. Although psychotherapy is perceived as being more expensive than pharmacotherapy, the cost balances out over the long-term. Regarding efficacy, a recent meta-analysis8 suggested that an overestimation of effect size exists for psychotherapy due to publication bias, but the meta-analysis did not stratify results by depressive severity. However, the REVAMP trial9 compared 2 different forms of psychotherapy plus medication with medication alone for the treatment of chronic or recurrent depression (including mild severity), and the 3 arms of the trial produced similar outcomes (AV 3). These results suggest that measurement-based pharmacotherapy produces a robust effect that is difficult to improve by adding adjunctive psychotherapy.
Challenges in Psychiatry
Challenges in psychiatry include addressing the belief that antidepressants are ineffective and the FDA warning that antidepressants may increase the risk of suicidality in young patients. Additionally, the panel discussed the widespread conviction among certain critics of psychiatry and some members of the public that a conspiracy exists between the pharmaceutical industry and psychiatrists to profit from taking advantage of the public. This conviction then causes any data that support the efficacy and safety of antidepressants to be viewed with suspicion. Misunderstanding of the safeguards, meticulous oversight, and careful monitoring of clinical trials adds to the public’s perception of bias in these studies. Intentional misinformation about the efficacy and safety of antidepressants and the continuing stigma that mental illness carries add to the concern that people may be deterred from seeking treatment for depression.
To address the crisis of confidence in psychiatric medications, practitioners need to educate the public about mental illness and its treatments. Also, researchers should study the patient populations typically seen in clinical practices, rather than the often less severely ill patients seen in clinical trials. New treatment approaches concentrating on novel mechanisms of action are another path to regaining the trust of the public.
Formulating a clearer definition of depression is another piece of the puzzle. Although psychiatrists may see the more severely ill patients, primary care practitioners see many patients who present with a range of symptom severity. Primary care physicians who prescribe antidepressants for milder forms of depression or in response to patients reporting that they “feel down” may be confusing the public about what a depressive disorder really is. Clarifying the pathophysiology of the disparate conditions now grouped together as the single entity of depression will help to advance treatment.
The panel agreed that the data regarding the efficacy of antidepressants are complex, making it easy to misinterpret meta-analyses. Additionally, the issue of suicidality is quite complicated but the risk is not great enough to abandon the use of antidepressants, although patients should be monitored carefully. The concept of depression may have become too broad, and patients who are more mildly depressed may benefit from receiving evidence-based psychotherapy before antidepressants.
The consensus of the panel was that, although depression is a long-term, chronic disease and additional research and novel treatments are needed to improve outcomes for patients, measurement-based pharmacotherapy is an effective tool for helping many patients achieve remission and recovery. Clear communication with the public and the media, as well as with other physicians, about the safety and efficacy of antidepressants will reduce societal skepticism and encourage those who need treatment to seek it, and thus help more people who suffer from depression to recover.
For Clinical Use
- When possible, educate the public about mental illness and its treatments
- Expect greater antidepressant efficacy in patients with more severe depression than in patients with mild or moderate depression
- Closely monitor all patients who have begun antidepressant treatment for suicidal ideation or behavior
- Recommend evidence-based psychotherapy to patients with mild or moderate depression
ADM = antidepressant medications, BSP = brief supportive therapy, CBASP = cognitive behavioral analysis system of psychotherapy, FDA = US Food and Drug Administration, HDRS = Hamilton Depression Rating Scale, NICE = National Institute for Clinical Excellence, RCT = randomized controlled trial, REVAMP = Evaluating the Value of Augmenting Medication with Psychotherapy
Take the online posttest.
- Fournier JC, DeRubeis RJ, Hollon SD, et al. Antidepressant drug effects and depression severity: a patient-level meta-analysis. JAMA. 2010;303(1):47–53.
- Kirsch I, Deacon BJ, Huedo-Medina TB, et al. Initial severity and antidepressant benefits: a meta-analysis of data submitted to the Food and Drug Administration. PLoS Med. 2008;5(2):e45.
- Khan A, Leventhal RM, Khan SR, et al. Severity of depression and response to antidepressants and placebo: an analysis of the Food and Drug Administration database. J Clin Psychopharmacol. 2002;22(1):40–45.
- Leon AC. The revised black box warning for antidepressants sets a public health experiment in motion. J Clin Psychiatry. 2007;68(7):1139–1141.
- House Committee on Veterans' Affairs. Exploring the relationship between medication and veteran suicide. http://democrats.veterans.house.gov/hearings/hearing.aspx?NewsID=525. February 24, 2010. Accessed September 9, 2010.
- Hammad TA, Laughren T, Racoosin J. Suicidality in pediatric patients treated with antidepressant drugs. Arch Gen Psychiatry. 2006;63(3):332–339.
- FDA News. FDA proposes new warnings about suicidal thinking, behavior in young adults who take antidepressant medications [news release]. Washington DC: US Food and Drug Administrtion; May 2, 2007. http://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/2007/ucm108905.htm. Accessed December 9, 2010.
- Cuijpers P, Smit F, Bohlmeijer E, et al. Efficacy of cognitive-behavioural therapy and other psychological treatments for adult depression: meta-analytic study of publication bias. Br J Psychiatry. 2010;196(3):173–178.
- Kocsis JH, Gelenberg AJ, Rothbaum BO, et al. Cognitive behavioral analysis system of psychotherapy and brief supportive psychotherapy for augmentation of antidepressant nonresponse in chronic depression: the REVAMP Trial. Arch Gen Psychiatry. 2009;66(11):1178–1188.