Did the FDA Black Warning Cause Suicides? Reply to Soumerai and Lu

I’m not a medical doctor. I don’t even play one on TV. So why should anybody listen to me on the subject of antidepressants?

That’s a very good question, and here’s my answer: because I have listened to women and men – good, kind, caring, creative people who took these drugs, or whose loved ones took these drugs – in good faith, as prescribed – and suffered horrifying toxic effects, including suicide and homicide, as a result.

I have also listened to men who have served as expert witnesses in lawsuits against the drug companies and got to see the secret files that the drugmakers hoped never would see the light of day.

I know what these companies are capable of.

So that’s why it was disheartening for me to see in STAT a recent article by Stephen Soumerai and Christine Lu resurrecting (yet again) the zombie myth that the FDA black box warning on youth suicides led to an increase in youth suicides. That idea was easily refuted by data when it was first proposed, and it is just as easily refuted now.

A bit of background is in order here. In October of 2004, faced with mounting evidence, the FDA issued a black box warning linking antidepressants to youth suicides. In May of 2007, this warning was extended to young adults up to twenty-five years of age. No explanation was offered as to why this effect magically turns itself off after the twenty-fifth birthday, but never mind that for now.

In September of that same year, statistician Robert Gibbons published a paper in the American Journal of Psychiatry titled “Early Evidence on the Effects of Regulators’ Suicidality Warnings on SSRI Prescriptions and Suicide in Children and Adolescents.” Dr. Gibbons and his co-authors looked at the rates of both antidepressant prescriptions and suicides in young people, both before and the black box warning of October 2004. They found that antidepressant prescriptions for children and adolescents decreased by twenty-two percent, while the youth suicide rate rose fourteen percent.

The authors concluded “If the intent of the pediatric black box warning was to save lives, the warning failed, and in fact it may have had the opposite effect: more children and adolescents have committed suicide since it was introduced.”

Predictably, the media had a field day with this one. The Washington Post informed readers, “Youth Suicides Increased as Antidepressant Use Fell.” The New York Post went for the jugular: “Warning May Be Killing Kids,” while the Chicago Tribune weighed in with the more restrained “As Youth Suicides Increase, FDA’s Label Rule Criticized.” Dr. Gibbons told the Tribune the FDA black box warning had a “horrible and unintended effect.”

There was just one problem: Gibbons and his colleagues looked at antidepressant prescriptions for the years 2004 and 2005, but they only looked at suicide data for the year 2004. Examination of Figure 1 of the AJP paper shows that antidepressant prescriptions barely budged in 2004, and almost all the decrease in prescriptions occurred in 2005. What happened to suicide rates that year? The paper doesn’t say.

But it’s easy enough to find out. According to the CDC, in 2005, the first full year after the FDA black box warning, the rate of youth suicides dropped. It dropped again in 2006, and again in 2007. That is exactly what we would expect to find if the FDA black box warning had its intended effect, either by means of decreasing prescriptions for antidepressants, or increased monitoring of pediatric patients for suicidality, or both.

In a 2018 article in STAT, Dr. Soumerai and a co-author reiterated Gibbons’s conclusions, only to have their arguments completely debunked by author Robert Whitaker in his Mad in America blog (as discussed in Chapter Seven my book, Prescription for Sorrow). Now Soumerai, along with his colleague Dr. Lu, are trying again.

The two experts begin with this jab at Health and Human Services Secretary Robert F. Kennedy Jr.:

Kennedy has in particular made skepticism toward antidepressants for adolescents a centerpiece of his public health agenda, even claiming that antidepressants may be harder to quit than heroin …

In fact, a 2018 survey of antidepressants users on an internet message board showed that withdrawal can cause a dizzying variety of often truly horrifying effects:

• Neurological: dizziness, ringing in the ears, burning sensations, sensitivity to light
• Psychological: suicidality, anger, insomnia, obsessive thoughts, poor concentration and memory, depersonalization, paranoia, terrifying dreams
• Gastrointestinal: constipation, diarrhea, acid reflux
• Cardiovascular: palpitations, chest pain, racing heart, skipped beats, high blood pressure
• Musculoskeletal: Muscle weakness, aches and pains
• Psychosexual/genitourinary: difficulty urinating, erectile dysfunction, “numb penis”
• Other: recurring infections, bad skin, hives

Some of these complaints are literally indescribable in standard medical terminology: “vision lagging behind eye movements,” “head like cotton balls stuffed in,” “brain sloshing.”

These findings were confirmed by another study released that same year by the All-Party Parliamentary Group for Prescribed Drug Dependence. On a scale of one to ten, the average self-reported severity of withdrawal symptoms was nine. Thirty percent of respondents reported being out of work indefinitely because of antidepressant withdrawal symptoms. Many reported these symptoms persisted for years and years after stopping antidepressants, and some found themselves unable to kick the drugs and gave up entirely.

Quitting heroin sounds like a walk in the park by comparison.

The two clinicians go on to state:

Kennedy has not only loudly and repeatedly questioned the value of antidepressants for adolescents, but he has also speciously linked them to violent behavior — specifically to a mass shooting — with, again, absolutely no evidence.

In fact, there is a mountain of evidence linking antidepressants to violence and suicide. This evidence began coming in as soon as these drugs were released on to the market, and continues to this day – clinical case histories, healthy volunteer studies, data mining of after-market violence case reports, and the drugmakers’ own trials.

Psychiatrist David Healy and his colleagues analyzed RCT data for five antidepressants and found that on a per-patient basis, compared to placebo, these drugs were associated with a doubling of the rate of suicidal acts and a whopping nearly seven-fold increase in the rate of completed suicides. If, for some perverse reason, the drugmakers wanted to promote their wares as suicide enhancers, these data would be considered overwhelming evidence that these nostrums are effective at doing just that.

Or as Dr. Healy put it, graphically and succinctly, “When it comes to dead bodies in current psychotropic trials, there are a greater number of them in the active treatment groups than in the placebo groups. This is quite different from what happens in penicillin trials or trials of drugs that really work.”

Since then, testimony in Dolin v GlaxoSmithKline revealed that GSK’s own trials of Paxil showed a seven-fold increase in suicide attempts per patient in patients given the active drug, as opposed to placebo. They also had twenty-two completed suicides in treatment arm, as opposed to zero in those given placebo. These astonishing findings become even more astonishing when you realize that actively suicidal patients are excluded from drug trials.

Absolutely no evidence?

The two authors go on to favor us with these words of wisdom:

Kennedy’s antidepressant rhetoric is not only based on bad science, it fuels distrust in mental health treatments at a time when adolescent depression, anxiety, and suicide rates are at record highs.

Do you know what else is at a record high? Antidepressant prescriptions for adolescents. Maybe this “distrust in mental health treatments” is justified.

Then we get to the meat of the piece:

We have seen before how fear-driven messaging about antidepressants can cause catastrophic harm. In our Health Affairs systematic review, we analyzed the totality of rigorous evidence on the FDA’s existing black box warnings for youth antidepressants. We found that well-intended but mismanaged warnings to doctors, patients, and parents about possible negative impacts have cost several thousand lives.

And yet, the authors’ own data preposterously undermine their conclusions. An examination of Figure 2 in their 2020 paper shows clearly that youth suicides showed a precipitous drop in the first quarter of 2008, but they have gone up steadily since then. So have prescriptions for antidepressants.

The problem with the sort of interrupted time-series analysis employed by the two experts is that the null hypothesis is overly specific. It assumes that any change in a trend must be due to whatever variable to which they have decided to attribute the change, rather than any of a myriad of other factors that may be operating. If anything, I’d say these data demonstrate that Dr. Gibbons’s 2007 paper supposedly debunking the link between antidepressants and youth suicides is the culprit here, causing an increase both in antidepressant prescriptions and youth suicides.

Soumerai and Lu’s paper is part of a disturbing trend: a full-court press to make people believe that drugs don’t kill people – warnings about drug harms kill people.

The two clinicians go on to bemoan the fact that non-pharmaceutical treatment (i.e. psychotherapy) for depressed youth has declined in the wake of the black box warning. They do not adduce a shred of evidence that the black box warning caused this decline, but never mind that for now. If Drs. Soumerai and Lu wish to discuss how distressed youths might more easily access meaningful help with problems with living, without anyone having to pretend they suffer from a drug-treatable brain disease, I for one would be happy to have that conversation. But somehow I doubt they would be interested.