Copyright © New Therapist


More Chinese Placebo Puzzles

By John Söderlund


Some time ago, I appear unwittingly to have provoked the ire of a handful of readers of a daily newspaper for which I write the occasional column. The disgruntled disliked my statistical approach to understanding the value of placebos when compared with psychiatric medication.

Well, I'm back at the scene of the crime with a new psychopharmacological Chinese puzzle inspired by the writings of Ellio Frattaroli in his recent book Healing the Soul in the Age of the Brain (see New Therapist 22 for an interview with Frattarolli)

On August 27, United Press International reported on a study of the effectiveness of Zoloft, one of the most commonly prescribed anti-depressants around the world.

The study examined the effectiveness of Zoloft compared with that of the placebo (or sugar pill) in a sample of paediatric patients diagnosed with major depressive disorder.

The drug, the study found, was effective for use in children and adolescents, supported by the finding that 69 percent of those receiving it improved significantly in terms of the symptoms of depression that they demonstrated after treatment. This compared to significant improvements in 59 percent of those who were administered a placebo - a drug with no active ingredients.

In my earlier musings on this matter, I noted - to the chagrin of some - the unavoidable fact that those who received the real drug also received the placebo, since the placebo (from the latin "to please" and thought to have an effect because of the attention that comes with the administration of the sugar pill) is impossible to strip out of such trials.

In short, when one receives the drug - whether it be the real thing or the sugar pill one also receives the effect of wanting to please the giver of the drug.

In previous articles, this fact was the basis on which I suggested that the placebo effect, considered on its own, could reasonably be said to be more effective than the active drug, since 59 percent of the effect of the active drug, in this case, could reasonably be attributed to the placebo effect, all things being equal.

But here's the new Chinese drug puzzle. If the net difference between giving your child a sugar pill and a Zoloft is an increase of a mere10 percentage points (the difference between 59 and 69 percent) in the likelihood that they will recover, surely it is reasonable to consider the compensatory value of avoiding the other (mostly undesirable) things that appear to come with the real drug?, the web site of manufacturer Pfizer, lists the following possible side effects:

Upset stomach, trouble sleeping, diarrhea, dry mouth, sexual side effects, feeling sleepy or tired, tremor, indigestion, sweating, feeling agitated, and having less appetite.

Children who take Zoloft, it adds, may also have other side effects such as excessive movement or twitching, fever, not "feeling well," trouble concentrating, not thinking normally, nosebleeds, weight loss, easy bruising, manic or excited behaviour, or rapid mood swings.

What it doesn't address in its list of side effects, however, is suicidality. Why suicidality, you might ask. Isn't that the very thing it should be preventing?

That question would take us back a few years. To 1990, in fact, when the two prominent Harvard researchers and psychiatrists, Drs. Martin Teicher and Jonathan Cole published an article entitled "Emergence of Intense Suicidal Preoccupation During Fluoxetine Treatment", in the American Journal of Psychiatry. The article reported that " persistent, obsessive, and violent suicidal thoughts had emerged in a small minority (3.5 percent) of patients treated with fluoxetine (the active ingredient in a range of this class of antidepressants, including prozac)." This translated, the researchers, concluded, into an estimated incidence of 1.3 to 7.5 percent with 95 percent confidence limits.

You can guess the response of the makers of such drugs. And the tit-for-tat raged for another ten years, including brief, polite inquiries from the American Food and Drug Administration (FDA) about continued reports of suicidality following treatment with Sertraline (the chemical name for Zoloft's active ingredient, particularly within the pediatric/adolescent population.

But all the quiet diplomacy over the potential damage to a multibillion dollar industry was shattered in a couple of more recent court cases.

On June 6, 2001, a jury found that Paxil (a drug of the same class as Zoloft, called Aropax in South Africa) caused a man in Wyoming to kill his wife, daughter, granddaughter and himself. Smithkline, the manufacturer of Paxil, was ordered to pay the family of Donald Schell damages of $8,000,000.

A year earlier, the Supreme Court of New South Wales ruled that a man's ingestion of Zoloft had caused him to murder his wife.

Which brings us to the more recent past and closer to the answer to the question of suicidality, antidepressants and the placebo effect. On June 19 this year, the FDA said it was reviewing reports of a possible increased risk of suicidal thinking and suicide attempts in children and adolescents under the age of 18 treated with the drug Paxil for major depressive disorder (MDD). It recommended that Paxil not be used to treat children and adolescents for MDD, given that there was no evidence that Paxil was effective for this purpose and that the drug was not approved for use in children and adolescents.

Following its recent review of the same data, the UK Department of Health issued a Press Release on June 10 this year, stating that paroxetine (the active ingredient of Paxil) must not be used to treat children and teenagers under the age of 18 years for depressive illness because UK authorities have concluded that there is an increase in the rate of self harm and potentially suicidal behaviour in this population when given paroxetine.

All of this goes a long way to explaining where we began: The recent interest in testing the effectiveness of the selective serotonin re-uptake inhibitors (SSRI's) among adolescents.

And while the 69 percent hit-rate with Zoloft may be higher than the 59 percent yielded by the severely maligned placebo, I can't help suspecting that, given the history of the whole saga, some of those most at odds with my past hypothesising might themselves be a little hesitant about dishing out Zoloft over the sugar pill when their children next get a bout of the blues.

John Soderlund is a Counselling Psychologist and media consultant in Pietermaritzburg, South Africa, and publishing editor of New Therapist Magazine.



Return to New Therapist home page