THE QUESTION POSED in the title of this column may strike many readers as odd. How can medications that have proved helpful in reducing depression also cause suicide? After all, suicide is a tragic complication of some cases of depression. Yet research and clinical observations over the past 40 years have raised concerns that these drugs produce suicidal thoughts, suicide attempts and possibly even suicide in a small subset of depressed patients. What are the risks?
In 2006 psychiatrist Tarek A. Hammad and his associates at the U.S. Food and Drug Administration published a meta-analysis (quantitative review) of studies involving a large number of children and adolescents taking antidepressants for depression, anxiety disorders and attention-deficit hyperactivity disorder (ADHD). Their results demonstrated that subjects on antidepressant medications had twice the risk (4 versus 2 percent) of suicidality (suicidal thoughts and attempts) as compared with those on placebo. No completed suicides occurred during any of the studies reviewed. A 2007 meta-analysis by psychologist Jeffrey Bridge of Ohio State University and colleagues at several institutions included additional studies and confirmed these results, although the percentages for suicidality were slightly lower.
These findings point to the drugs as the cause of the increased suicidality rather than depression.
On supporting science journalism
If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
The FDA Acts
The initial results of Hammad's meta-analysis were available to the FDA well before the article's publication. In the wake of these findings, the FDA placed a “black box warning” on all antidepressants in 2004. This warning applied to the entire class of antidepressants and stated that these drugs can increase the risk of suicidality in children and adolescents who have major depressive disorder or other psychiatric disorders. The black-box warning is the strongest one that the FDA can issue (and is so named because of the black border that usually surrounds the text of the warning). In 2007 the agency extended this warning to include people up to the age of 24, noting that the data did not show this increased risk in adults older than 24.
The risk for suicidality usually occurs within the first days or weeks of starting the medication. According to a 2004 FDA Public Health Advisory, public health officials are concerned that patients showing certain symptoms early in treatment or during a change in medication dosage may be at heightened risk for worsening depression or suicidality. This added risk is more likely if the symptoms are severe, abrupt in onset and not part of the depressive symptoms for which the patient initially sought treatment. These symptoms include anxiety, agitation, panic attacks, insomnia, irritability, hostility, impulsivity, severe restlessness, hypomania and mania. Fortunately, only a small percentage of people who show these symptoms are at risk for suicidality. Nevertheless, if these symptoms start to occur, they should be reported to the prescribing physician.
Advice about Antidepressants
Numerous studies in patients of all age ranges have found that antidepressants are helpful in about two thirds of depression cases. For children and adolescents, the data show that Prozac (fluoxetine) is fairly effective but that other antidepressants are not any more effective than a placebo pill. For adults, many different antidepressants have proved helpful. One major problem with antidepressants is the high rate of relapse (approximately 40 percent) that occurs after they are discontinued. Many researchers have found that the combination of an antidepressant and psychotherapy (especially cognitive-behavioral therapy) leads to greater symptom reduction and less suicidality and suicide than either treatment affords alone.
At a practical level, the risk of suicidality can often be adequately addressed by careful monitoring of patients, especially early in treatment. Unfortunately, most primary care doctors and psychiatrists do not do this; they usually arrange a follow-up meeting several weeks or more after the drug is prescribed. This may be the result of the doctors’ lack of awareness of the risk of suicidality or the limits placed on the number of sessions by some insurance companies. Some even tell patients that such side effects are possible and that they will wear off in a week or two. This warning may discourage patients from calling doctors if the symptoms do occur. Treatment with both antidepressants and psychotherapy carries the built-in safeguard that the psychotherapist can monitor side effects and inform the patient and prescribing physician if any problems do occur.
Although there have been no research findings demonstrating completed suicide resulting from antidepressant usage in children and adolescents, a number of case reports suggest that there might be cause for concern. Case reports, however, are only suggestive and do not constitute hard scientific evidence. In addition, most studies that have found suicidality have been relatively short-term with small samples as compared with those needed to address this question; long-term risks have not yet been carefully evaluated. Most studies of people older than 24 have not reported increased suicidality with antidepressants. Nevertheless, we should be cautious about accepting this conclusion because we do not know the mechanism that explains why antidepressants can trigger suicidality in younger people.
In addition, these data may be subject to serious sources of bias, as we will discuss. To be on the safe side, physicians should also closely monitor adults who are taking antidepressants. If people who are taking antidepressants wish to stop, it is important that they consult their doctor before doing so and that they gradually taper off the drug rather than going “cold turkey.” Stopping antidepressants abruptly can trigger an array of distressing symptoms, including dizziness, nausea, headache, fatigue, anxiety, irritability and sadness, to name just a few.
Conflicts of Interest?
The controversy regarding antidepressants and suicidality has both scientific and economic dimensions. In 2004 sales of commonly prescribed classes of antidepressants (including selective serotonin reuptake inhibitors, or SSRIs) in the U.S. alone reached $10.9 billion. Clearly, pharmaceutical companies have a large stake in these drugs. David Healy, a psychiatrist and former secretary of the British Association for Psychopharmacology, has argued that some drugmakers have often published biased results of studies on these medications and hidden or disguised their risks for more than 40 years. Others have vigorously challenged his conclusions.
Pharmaceutical makers have funded most drug trials to date, creating a potential conflict of interest for investigators and leading to the increased likelihood of an “experimenter expectancy effect.” This effect occurs when researchers expect certain results—in this case, that antidepressants are safe and effective—and unintentionally influence the study design or analyses to find these results. Although double-blind procedures (in which neither the patient nor the research staff knows whether the patient is receiving an active medication or a dummy pill) can provide some protection against this bias, often controversy exists with respect to how well these procedures are implemented in medication studies.
It is also relevant that a major study funded by the National Institute of Mental Health found higher suicidality rates than those obtained in studies funded by drug companies. Moreover, many drug companies have provided ghostwriters to write up drug trials for publication, potentially biasing the document by slanting its descriptions toward favorable drug effects. Finally, there have been reported instances (for example, the recent scandal regarding the drug Vioxx) in which pharmaceutical companies have hidden negative evidence. Although this type of nondisclosure has not been clearly demonstrated for antidepressants, it remains a disturbing possibility.
In response to these concerns, the International Committee of Medical Journal Editors (ICMJE) has recently required that authors submitting papers must disclose all financial and personal relationships that might bias their work, state whether potential conflicts exist, identify individuals who provided writing assistance, and disclose the funding sources for this assistance. Perhaps most important, the ICMJE requires authors to register all clinical trials they have conducted with a not-for-profit organization, whether or not the studies have been published. In this way, all studies will be publicly available at no charge, not just those that have obtained positive results. In addition, the FDA requires drug companies to submit the results of all studies that have been conducted, including efficacy and safety data, in their application for FDA approval.
So, do antidepressants cause suicide? The available data suggest that they probably do not, but potential biases in the research render this conclusion tentative. Nevertheless, some antidepressants do appear to cause at least a slight increase in suicidal thoughts and attempts in children, adolescents and young adults.
Suicide is the third leading cause of death in adolescents (10 to 14 years of age) in the U.S. Many investigators have found that increased antidepressant use is associated with decreased rates of completed suicides. If black-box warnings lead physicians to decrease their administration of antidepressants, the rate of completed suicides may increase. Results of a recent study in the Netherlands suggest that this may be the case: the suicide rate in children and adolescents has increased by more than 40 percent since 2003, when prescription rates of antidepressants began to decrease.
A better alternative to prescribing antidepressants to fewer children and adolescents would be to continue to prescribe them but with close and careful monitoring by the physician. Otherwise, we may be faced with a tragic irony: efforts to protect children from suicidal thoughts and attempts through lower antidepressant prescription rates may lead to greater harm through increased suicide rates because of inadequately treated depression.
Many questions remain. Will future studies conducted independently of the influences of drug companies show different risks of suicidality and suicide in young people? Will we continue to find a lack of antidepressant-induced suicidality and suicide in adults? What will longer-term studies reveal about the effects of antidepressants on suicide and suicidality? We hope that recent efforts to reduce sources of bias in research will provide us with more accurate answers to these important questions.
Fortunately, there is a safe alternative to antidepressants. Many, but not all, studies have shown that short-term psychotherapies (12 to 16 sessions), with or without medication, are at least as effective as medication for depression and anxiety, maintain patient improvement better in the long run, and carry little or no associated risk of suicide. Yet because psychotherapists do not have advertising budgets remotely approaching those of drug companies, many mental health consumers are unaware of these important findings.
Given the present state of knowledge, we support the use of antidepressants, particularly fluoxetine, as one treatment for depression and anxiety in children, adolescents and young adults, as long as the treatment is closely monitored. Fluoxetine and many other antidepressants have proved effective in adult populations, and we support their use with those older than 24 as well. The data show, however, that a combination of psychotherapy, particularly cognitive-behavioral therapy, and antidepressant medication is the most effective, enduring and safe treatment.