|McMan's Depression and Bipolar Web|
|Home Articles Links News Newsletter Books Forum Community Search Donate|
Your Depression and Bipolar Disorder Source
Knowledge is Necessity
In the future, antidepressants will carry suicide warnings. Is the FDA justified in its action?
"A success rate of three out of 15 studies is clearly, a concern."
Main articles page.
More Treatment Articles
FDA Antidepressant Suicide Warning
On Sept 14, 2004, an FDA panel voted 18 to 5 to require manufacturers of all antidepressants to add black box warnings to their product labeling. A month later, the FDA adopted the panel's recommendations. The warning reads in part:
"Antidepressants increased the risk of suicidal thinking and behavior (suicidality) in short-term studies in children and adolescents with Major Depressive Disorder (MDD) and other psychiatric disorders. Anyone considering the use of [Drug Name] or any other antidepressant in a child or adolescent must balance this risk with the clinical need. Patients who are started on therapy should be observed closely for clinical worsening, suicidality, or unusual changes in behavior."
The warning specifically links antidepressant use to suicidal behavior in four percent of kids on these drugs compared to two percent for kids on placebos.
A "black box" warning is the most serious warning placed on the labeling of a prescription medication, but does not ban physicians from prescribing the drug to children and adolescents.
The FDA action comes out of a series of enquiries that began in May 2003 with the investigation into one pediatric trial involving Paxil. That investigation was subsequently expanded to include all pediatric trials for the newer antidepressants.
Issue One: From the time SSRIs were first introduced, a handful of experts have raised safety concerns that, for certain individuals, the new antidepressants may cause mental restlessness and suicidality. The FDA considered the issue in 1990, with a public hearing, but decided the evidence was insufficient to warrant any labeling changes.
Related to Issue One: The potential for antidepressants to induce mania, agitation, and similar behavior, and whether this in turn puts patients at increased risk for suicide. Although this issue was not addressed by the FDA, the new labeling (but not in the black box) advises, "there is concern that such symptoms may represent precursors to emerging suicidality."
Issue Two: Are antidepressants effective in kids? In other words, is the benefit worth the risk? Only Prozac has passed two pediatric trials for depression, and is the only newer antidepressant approved by the FDA and its UK counterpart to treat depression in kids.
Antidepressants - The Problem
Major depression affects approximately two percent of children, and four to eight percent of adolescents. By the time a youngster reaches age 18, there is a 20 percent prevalence rate for major depression. The rate of dysthymia is slightly lower and is often under-recognized. Seventy percent of youth with dysthymia break out into major depression. Compared to adults, adolescents have more behavioral problems. Co-occurring disorders may be present in up to 90 percent of youth with major depression, including anxiety, disruptive behavior, and substance use.
Six to 10 percent of kids with major depression have a protracted course, 40 to 60 percent relapse after remission, and 20 to 60 percent have recurring episodes after recovery.
According to a 1993 nine-year study by Kovacs et al, children with major depression had a 74 percent rate of suicidal thinking and a 28 percent rate of suicide attempts. The percentages were about the same for those with dysthymia.
Antidepressants - The Background
An estimated 157 million prescriptions for newer antidepressants were dispensed to patients of all ages in the US in 2002, approximately 10.8 million to kids aged one through 17, and about 8.1 million to adolescents.
Following reports that a small percentage of kids in Paxil trials exhibited signs of suicidal behavior, the FDA in July requested the manufacturers of eight other antidepressants to look at data from their pediatric drug trials for evidence of suicidality. In August, Wyeth made a labeling change to Effexor and sent a letter to doctors recommending against its pediatric use. In October, the FDA sent a public health advisory, urging caution for all antidepressants for pediatric use. In December, citing lack of evidence of efficacy in clinical trial data and safety concerns, the FDA's UK counterpart, the MHRA, contraindicated all new generation antidepressants for pediatric use except for Prozac
The FDA acted on the strength of a report by its lead investigator into the drug trials, Andrew Mosholder MD. The report suggested evidence of an increased suicide risk in kids based on a review of 25 pediatric trials, 16 related to depression. The report also found that only three in 15 pediatric depression trials had succeeded, thus putting the issue of benefits vs risk on the table. At the time, the FDA did not regard these findings as conclusive, and deferred making a recommendation pending a second opinion from an independent team at Columbia University.
But the damage had been done. As more information came out, a wider public learned for the first time how drug companies routinely suppress negative trials, manipulate data, and only report positive results. Parents were up in arms, trial lawyers started licking their chops, the US Congress held hearings, and the New York Attorney General would soon launch an investigation. For the first time in recent memory, the drug companies were on the defensive. Any resistance they put up, in public anyway, was token at best..
In February 2004, the FDA's combined Psychopharmalogic Drugs Committee and Pediatric Advisory Committee convened a public hearing, and as a result of that meeting, the FDA announced its intention to strengthen the warning labels on antidepressants. By the time the gavel came down on its meeting in September, the only real question was one of execution and degree. John Walkup MD, a child and adolescent psychiatrist from Johns Hopkins, adroitly framed the issue by urging the panel that any warning should clarify rather than magnify. “Magnification of the warning may actually do more harm than good to the kids who actually need to come to care,” he cautioned.
But Mark Miller, whose 13-year-old son committed suicide while on Zoloft, countered, “I would rather be scared to death by a label on a medication than be changed forever by the death in our family seven years ago.”
The Suicide/Self-harm Issue
Dr Mosholder's report analyzed suicidality data from 25 drug trials (16 related to depression) involving 4,000 patients. Of these, 109 events were considered "possibly suicide-related," rated by the drug companies that had sponsored the trials as such. Forty-seven patients were hospitalized, 66 committed self-harm. There were no completed suicides. All the patients concerned fully recovered. Self-harm ranged from hanging and overdose to cutting and slapping. The MHRA undertook a similar review, but relied on summary data from the drug companies while the FDA had a closer look at individual trial data. The MHRA, on the other hand, released data on mania and related effects in some trials while the FDA did not. A brief summary:
An instance of a self-inflicted slapping that had been labeled as suicidal suggested that Dr Mosholder's findings could fall apart under closer scrutiny, but just the opposite happened. The Columbia University team that had a second look at the data did expunge the self-slapping episode and 11 other events from the suicidal category, but they also found 26 new cases that drug company investigators had considered benign. Kelly Posner PhD, who led the effort, told the FDA panel at its September hearing of a child who impulsively swallowed 11 pills, which had been originally coded as a medication error and an attempted hanging that was attributed to a personality disorder rather than a suicidal event.
An FDA team then crunched the numbers, with its leader Tarek Hammad MD, PhD testifying that in September that “out of 100 patients treated we might expect two to three patients to have some increase in suicidality due to short-term treatment ... that is beyond the risk that occurs with the disease being treated.”
How Serious Is The Antidepressant Risk, Really?
At both the February and September hearings, Andrew Vickery, the Houston lawyer who won a $6.4 million wrongful death suit against GSK, the makers of Paxil, noted that the drug trials were not designed to capture suicidal behavior, implying the numbers would turn out much higher if they were. He reminded the panel that Eli Lilly had promised the FDA back in 1990 to conduct a “rechallenge protocol,” which they never did, and which the FDA never made them do.
Two witnesses, including psychiatric scourge Peter Breggin MD, referred to a 2001 retrospective review of hospital admissions by Preda et al that found that 8.1 percent of patients admitted to a psychiatric unit over 14 months were the result of antidepressant-induced mania or psychosis. Dr Breggin also cited instances of unauthorized use of tranquilizers in adult trials.
In an article in the April 10, 2004 British Medical Journal, Jurendi et al criticized the authors of the published studies for exaggerating the drug’s benefits or downplaying their harm or both. For example, one trial of 93 kids on Paxil produced 11 serious adverse events in a short space of time compared with two of 87 in the placebo group. Despite this, together with the fact that seven of the patients were admitted to the hospital, the authors of that study concluded that the drug "was generally well tolerated in this adolescent population, and most adverse effects were not serious."
Since doctors increasingly rely on online journal abstracts, which report the misleading claims but not the contradictory data, the potential for mischief is high, Jurendi et al pointed out.
But John Mann MD of Columbia University, representing the American Foundation of Suicide Prevention, presented convincing population data at the September hearing showing a telling association between antidepressant use and lower suicide rates in adults and children worldwide. He told the panel that since 1987, after a steady three-decade rise, suicide rates in the US began dropping. Women, who were prescribed twice the SSRIs as men, experienced proportionately lower drops in suicide rates. Areas in the US with the highest SSRI prescription rates had the biggest decline in suicides. According to Dr Mann, for every 10 percent increase in prescription rates, the US suicide rate declined three percent.
A study published in the Oct 2003 Archives of General Psychiatry supports Dr Mann's analysis. Researchers from Columbia University reviewed 588 case files of kids aged 10 to 19 and found that a one percent increase in antidepressant use was associated with a decrease of 0.23 suicides per 100,000 adolescents per year.
This population data came as cold comfort to the parents who testified at both hearings. Terri Williams told the panel that shortly after starting his initial dose of Prozac, her son began having strange dreams, began behaving aggressively, and started breaking things. In her own words: "On December the 5th, 2000, I discovered Jacob's body hanging from the rafter in our attic. He had hung himself with his own belt. A letter was placed on the ladder leading up to our attic thanking us for giving him 14 years of a happy life ..."
From the transcripts of both hearings, it was evident that the parents had made a very strong impression on the panel. For one, some of these kids had been prescribed the drugs for other conditions such as OCD, ruling out depression as the cause of their suicidal behavior. A clear pattern that emerged was that these meds had been prescribed by primary care physicians who failed to advise parents on any possible adverse effects and who did not provide adequate follow-up care. Moreover, these kids had started acting suicidal very soon after taking their initial dose.
The FDA panel clearly required a statistical rationale for any decision it would make, but once the parents had spoken the final outcome was never in doubt.
The Efficacy Issue
The MHRA provided the only detailed breakdown of the pediatric trials in question. To summarize:
It Gets Worse
Testifying at both FDA hearings, Thomas Laughren MD of the FDA noted that a 20 percent success rate in pediatric clinical trials (three out of 15) “is clearly a concern.” Nevertheless, he observed that these results could be interpreted more positively if the data from two separate trials were pooled and another one were evaluated based on secondary endpoints. Dr Laughren pointed out that adult trials are also subject to high failure rates (50 percent). Thus, lack of success “is not the same thing as saying that we have proof that the drugs have no benefit.”
These remarks were seized on by Tom Woodward, a parent who lost his daughter while she was on Zoloft. In his testimony, he referred to Dr Laughren's logic as “Orwellian,” saying, “this gibberish is an insult to the American public and would be laughable if the consequences weren't so terribly tragic.”
At the February hearing, David Kirsch PhD of the University of Connecticut and co-author of two papers that found the placebo effect accounted for 75 to 80 percent of an antidepressant’s result was not nearly so charitable as Dr Laughren. Dr Kirsch reported on a new study of his that found that eight of 12 published pediatric trials failed to find any significant benefit of the medication over the placebo. The four trials that reported significant differences did so on clinician-rated measures rather than patient-related measures. Combining the data from the trials yielded a placebo response that accounted for 87 percent of the antidepressant’s response. Or, to put it another way, "the drug effect is only 13 percent of the drug response."
David Antonuccio PhD of the University of Nevada and Dr Kirsch’s collaborator in his latest study testified that industry studies typically contain design flaws that favor the study drug, such as excluding placebo responders and relying on ratings from clinicians who have a vested interest in the outcome (and who know which patients are on the drug, thanks to the side effects).
In their BMJ article, Jurendi et al analyzed the quality of six of seven published pediatric studies (one was considered irrelevant) of the newer antidepressants underwritten by the pharmaceutical companies. In just two trials did the antidepressant show a clear advantage over the placebo. Of 42 reported measures in all the studies, say the authors, "only 14 showed a statistical advantage for an antidepressant." Claims for effectiveness were based entirely on ratings by doctors. Of 10 measures relying on patient or parent reports, none "showed significant clinical advantage for an antidepressant." No study presented data on rates of attempted self-harm, emergency treatment, or school attendance. Drop-out rates ranged from 17 to 32 percent for those on antidepressants and 17 to 46 percent for placebo-treated patients, too high, say the authors to produce reliable results, notwithstanding statistical fictions such as "last observation carried forward."
In an aside, the authors noted that eight of nine unpublished studies failed.
Meanwhile, in one Paxil study and two Prozac studies the primary outcome measures were changed after the fact. Despite a review of the Prozac studies by the US Center for Drug Evaluation and Research (part of the FDA) that determined that "the sponsor did not win ... based on the protocol specified endpoint," and that the evidence based on the pre-specified endpoint was not convincing, the FDA approved Prozac for pediatric depression.
In an April 24, 2004 Lancet article, Whittington et al examined 11 published and unpublished pediatric trials involving five newer antidepressants, concluding that only Prozac suggested a favorable risk-benefit profile. Published data from one Paxil trial produced mixed results, but when pooled with data from two unpublished studies the risks outweighed the benefits. Pooled data from two published Zoloft studies showed similar mixed results plus a new study outcome (brought to light by a UK public authority ) that did not support the drug.
The authors noted that published trials form the basis of treatment guidelines, but that suppressing unpublished data "can ultimately lead to recommendations for treatments that are ineffective, cause harm, or both." The authors of the BMJ article made a similar point, concluding that "accurate trial reports are a foundation of good medical care. It is vital that authors, reviewers, and editors ensure that published interpretations of data are more reasonable and balanced than is the case in the industry dominated literature on childhood antidepressants."
But What About Real-World Experience?
The fact that the pharmaceutical industry could not produce convincing evidence that antidepressants worked in treating kids for depression amounted to the equivalent of a crisis of faith in the psychiatric community. In order to continue to justify treating kids with these drugs, psychiatrists would have to forsake the very same science that they had so convincingly employed in the past to build their patients' trust.
"Pay no attention to that man behind the curtain," they would now have to tell their patients in so many words.
At the September hearing, Lawrence Diller MD, who has authored two books critical of prescribing psychiatric meds to kids, put it this way: “For years we were told to practice evidence-based medicine, and now, when there is no evidence for SSRI effectiveness and yet higher risk of suicidality, the leaders say, 'wait, not so fast.' I say, where's the beef?”
Perhaps the beef lies in the population data that shows an association between lower suicide rates in both adults and kids and the introduction of SSRIs. True, spotting an association is not the same as finding a causal link, but any clinician who disregarded these findings could arguably be found guilty of malpractice.
In the meantime, the best that psychiatrists can do for now is go by their clinical experience. This formed the basis of the testimony by two leading spokespersons:
Larry Greenhill MD, a child and adolescent psychiatrist representing the American Academy of Child and Adolescent Psychiatry, testified that many child psychiatrists have found antidepressants useful for treating their patients, but did emphasize the need for kids to be “carefully diagnosed,” and “in a well-monitored treatment program.”
David Fassler MD, a child and adolescent psychiatrist representing the American Psychiatric Association, testified that SSRIs “can be extremely helpful and even life-saving for some children and adolescents,” though he noted they should not be regarded as a total solution for a complex illness. He also reported that although between 30 to 40 percent of kids may not respond to an initial medication, many will ultimately respond to a different medication.
A number of patients' groups echoed these sentiments, but it was a parent testifying on her own behalf who was the most convincing. Suzanne Vogel-Scibilia MD, psychiatrist, patient, NAMI board member, mother of five, two with mental illness, told the panel: "I shudder to think of their plight if these medications were not available. One of my sons has had suicide attempts and violent incidents with knives. He has also run out of our house - in a fit of terror - in subzero weather only to be found freezing and hypothermic by our local police department in the next town. These incidents all occurred while his illness was not adequately treated with an antidepressant medication."
Meanwhile, scientific inquiry is looking at how antidepressants work with talking therapy. At the September hearing, John March MD of Duke University and director of the ongoing NIMH Treatment for Adolescents with Depression Study (TADS) told the panel of a recently published TADS study showing that Prozac combined with cognitive behavioral therapy (CBT) worked better (71 percent response) for the 437 youths in the study compared to Prozac alone (60.6 percent response) and CBT alone (43.2 percent). Only 34.8 percent of the placebo group responded.
But treatment that optimally combines meds with talking therapy may only be a pipe dream for most kids. Ken Duckworth MD, a child and adolescent psychiatrist representing NAMI (which supports treating kids with medications) cautioned the panel: “If you think cognitive behavior therapy and thoughtfully applied medications, with good monitoring, is going to happen on a routine basis, there is not much evidence to suggest that. The system is not set up for that.”
To Box or Not to Box
Panel deliberations took place on the second day of the September hearing. Its members were uniformly impressed with the rigor of the Columbia team analysis and agreed by a vote of 25 to one (with one abstention) that the combined industry clinical trials and TADS study, imperfect as they were, showed evidence of suicidality. Although the need for some kind of warning was universally acknowledged, there was some hesitation over putting it in a black box. A black box is the FDA's strongest warning, prominently displayed at the top of the drug's product labeling. Its presence on a product severely crimps direct-to-consumer advertising.
Matthew Rudorfer MD of the NIMH voiced his objection, saying, “I believe that while we are concerned about a two to three percent increase of risk of suicidality, I think the underlying illness carries a 15 percent risk of suicide if left untreated, and I fear that the black box would impede access to treatments.”
The panel's chair, Wayne Goodman MD of the University of Florida, agreed in part with Dr Rudorfer by acknowledging that a black box would “make prescribing more difficult” and “there will be an alarm from parents and the child.” But he added: “I think that is worth that complication, because it will raise the threshold to prescribing and force an engagement of a discussion, not only about the risks, but the potential benefits and alternatives to medication.”
Robert Nelson MD, PhD of the University of Pennsylvania had this to add: “I think in this day and age, a lot of the information we get about drugs we pull from palm-based databases. What comes up first is a black box warning. If it is not there, you don't find it, you don't see it, you will go right to dose, you will miss it entirely. So, I think that is the only way to get it out to people.”
By the time the issue came to a vote, members were beginning to leave or had just left (read into that what you may). So it was that the panel rushed through its 18 to five decision in favor of recommending that the FDA require manufacturers to issue black box warnings on their product labeling. The panel also gave hurried approval to recommending that separate informational MedGuides be distributed by pharmacists with every prescription, and adjourned without addressing sundry loose ends. One month later, the FDA adopted the panel's recommendations.
One Final Matter
Nearly all those who testified, as well as the FDA panel, emphasized a crying need for quality studies, ones that specifically test for suicidality and are long term. But with black box warnings a reality, what parent is about to enroll their kid in such a trial?
Background materials, testimony from the FDA public hearings, FDA announcements, and labeling changes can be found here.
For free online issues of McMan's Depression and Bipolar Weekly, email me and put "Sample" in the heading and your email address in the body.
Updated Nov 3, 2004
Dena (May 2, 2004): While it is extremely important to evaluate
all information regarding the relation between antidepressants and
suicide, I thought it only caused a mass hysteria of sorts when the FDA
put out a vague warning that was-- as expected--exacerbated by the media.
I understand that they need to cover their #ss, but to throw such a
warning out there without explaining it in layman's terms can actually
cause more harm than good, as people who really do need those meds might
decide to or be told by loved ones to stop taking them. The warning, when
explained thoroughly on the other hand, can provide much insight for
people who have dealt with treatment-resistant depression.
Jacqueline (Nov 20, 2004): We need laws that disallow members of the FDA to make ANY profit off of medications being sold. Therefore, they can NOT be members of the manufacturing company's. The same goes for the rest of the government making money from Oil company's. It just can't be allowed because it creates distrust, and there is good reason for it. People are money-hungry and they're putting money before people. People need to take a stand. I was a victim. I was placed on medications when I wasn't depressed. I hadn't had a perfect life, and I could have used some good therapy and more hobbies to keep me out of trouble and boost my self esteem. BUT THAT's IT. Instead, I was placed on over 10 medications in a ONE year period. I self-mutilated for the first time. I became suicidal for the first time. I wasn't even 18 years old.
Myra (Dec 19, 2004): When i am not on medications, i am overwhelmed with thoughts of planning suicide., when i am on them (in very high doses) i am not in the planning mode == just wishful thinking
MJM 6/16/05: Re articles on antidepressants and suicide -it was not until after my brother's death by suicide , that I started looking into the side effects and suicide--I was horrified and shocked to read all this information, that none of us in the family knew about. My brother was diagnosed with depression and put on several drugs and on high doses--I now feel, he did not even realize what was happening to him--he went for help--this was a person who wanted to live and be with his family, not kill himself. I feel for all the families who went through this--no one should.
Post your opinion here.
Order my book on Amazon
Your online source for issues that matter to you.
For free samples, email me and put "Sample" in the heading and your email address in the body.
Find out more.
Shop for depression and bipolar books online here.