In 2004, the U.S. Food and Drug Administration (FDA) issued a black box warning about antidepressants, stating that their use by adolescents increases the risk of suicidal ideation, suicidal feelings, and suicidal behavior. There is a lot of direct debate about this FDA warning: some people in medical organizations are concerned that this warning may have more negative effects than something beneficial.
Because then, when depressed patients exhibit clinical symptoms, it will prevent depressed patients from seeking help and doctors from prescribing antidepressants. Now, a decade later, there is a wealth of epidemiological data that addresses these important concerns.
What is the impact of the FDA’s warning on the rate of measurement and treatment of depression? In retrospect, it appears that the FDA had little choice at the time but to issue a black box warning. The authorities analyzed a series of meta-analysis studies, including 372 randomized controlled studies of antidepressants, with about 100,000 subjects.
The results showed that none of the participants in the studies had fatal suicidal intent, though. However, the incidence of suicidal ideation and suicidal behavior was 4% for participants taking antidepressants, compared to 2% for those taking placebo. Subsequent age-stratified analysis of the study demonstrated a significant increase in this risk only among children and adolescents under the age of 18.
But in adults over 24 years of age, and in older adults over 65 years of age and even older, there was no evidence of a wind increase in the risk of suicide. And, antidepressants have a significant protective effect against the development of suicidal ideation and suicidal behavior.
This meta-analysis stirred up a lot of controversy, and different methodological issues were discussed. In particular, some experts questioned the validity of the studies included in the meta-analysis, with their suicide assessment methodology, and this study was not designed to assess suicide.
Of course the FDA took this into account: they needed to balance the small risks of pit depressant treatment with the proven benefits, and in 2007 they came up with a modified black box warning, specifically that depression itself was associated with an increased risk of suicide. Did this modified warning accomplish its mission – to inform clinicians of the risks but not to compromise the rational treatment of depression?
Since the black box warning was introduced, practice has shown that there are dynamics in depression treatment rates that worry us, and the answer to the diagnosis of suicidal intent in new patients with depression is that this may not be the case.
In a large sample cohort study – which included 1.1 million adolescents, 1.4 million adults, and 5 million older adults – researchers looked at claims data for self-help medicine (from 2000 to 2010). The source was the American Mental Health Research Network’s 11 health plans. The study noted that two years after the FDA issued its warning, antidepressant use decreased significantly: 31 percent, 24.3 percent and 14.5 percent for adolescents, young adults and adults, respectively.
Although there was a small increase in antidepressant use among adolescents after 2008, the rate of use was still lower than the 2004 data. Since the FDA warning was issued, antidepressant use appears to be lower in all age groups than would have been expected prior to the warning.
A similar trend was reported in earlier studies, which showed a nearly 20% decrease in the use of selective 5-hydroxytryptamine inhibitors among medications prescribed for adolescents between 2003 and 2005. During the same period, instead of a decrease in antidepressants, treatment for depression in adults increased from 20% to 30%. Prior to the black box warning, primary care physicians were describing an increase in the proportion of antidepressant prescriptions (an average increase of 4.97%).
However, immediately after the warning was issued, the percentage of prescriptions decreased (by an average of 4.61%). Thus, although the FDA announcement focused on children and adolescents, it also appears to affect community treatment of depression in adults – a population not included in the announcement, for whom strong evidence is lacking as to whether antidepressants pose a suicide risk.
Moreover, with the issuance of the black box warning, not only did the rate of prescription of antidepressants decline, but so did the rate of diagnosis of newly depressed patients. For example, a study of a national sample showed a significant reduction in the rate of diagnosis of newly depressed patients in primary care after the warnings were issued.
There was a 44% decrease among children, a 37% decrease among young adults, and a 29% decrease among all adults. This downward trend is a conclusion drawn from a case study rather than a change in the epidemiology of depression in practice.
And, the decline in antidepressant prescription rates has not been accompanied by an increase in other depression treatments, such as psychotherapy and other antipsychotics, as a complement to antidepressants. Of course, these observational data only show that the enactment of these trends and black box warnings occurred concomitantly; they do not allow for a causal relationship between the warnings and the diagnosis and treatment of depression.
It is also important to realize that these observational studies have methodological limitations. For example, predictions of trends in antidepressant use or depression diagnosis rates are not based on historical trends or are subject to error.
However, the decrease in depression diagnoses and the decrease in antidepressant prescription rates may affect the attitudes of patients and psychiatrists, influencing how they approach the controversy, media coverage, and FDA warnings. Patients may be reluctant to speak up about the symptoms that can lead to a depression diagnosis, and subsequent antidepressant treatment, and psychiatrists may be hesitant to prescribe these medications. But there are many other factors that could explain this trend.
For example, the limitations imposed by managing mental health care patients, or antidepressant prescription management programs, or the fallibility of mental health care, affect the entire U.S. population. With the FDA announcement, in addition to the decrease in depression diagnoses and prescription medications, there were other trends that had an impact: there was also a significant increase in statistics for antipsychotic intoxication, which was used as an indirect measure in one study to measure suicidal intent.
In the second year of the warning, poisoning rates increased by 21.7% among adolescents (10-17 years) and 33.7% among young adults (18-29 years). The poisoning phenomenon so far represents a possible suicidal intent.
It is possible that the downward trend in antidepressant use has led to this increase in suicidal intent. However, one thing is uncertain, is any psychiatric drug poisoning intentional or accidental? On the contrary, the rate of intact suicides among people aged 10-34 years increased gradually between 1999 and 2010.
There was no abrupt change before or after the FDA warning, so there is little evidence that this announcement led to a change in the true suicide rate. However, considering the full dynamics of these trends, it is very important to provide primary medical care for everyone, and in this service a significant percentage of depressed patients can be identified and treated. We know that untreated depression causes much more suicide – morbidity and mortality – than antidepressants do.
We need better training for physicians to help them understand that although they cannot ignore small risks, they can safely manage these small risks by monitoring their patients, especially with medications for children and adolescents. what should the FDA do when they see these observable data? Consider that the authorities’ modification of the black box warning in 2007 did not suffice to completely stop the aggressive cold effect that seems to exist in depression treatment.
It is possible that the warning was presented in a way that was clearer than any clarification might have contained – I believe that further modifications are unlikely, but that would have been useful. So I would say that the FDA should consider removing the warning altogether.
At the very least, there should be a discussion in the medical community about this possibility. I believe that we cannot ignore the importance of these epidemiological data, or the real possibility that the FDA warning has unintentionally influenced depressed patients and physicians, preventing them from finding treatment and preventing physicians from prescribing antidepressants.