I will dutifully report on yet another professional’s opinion about the research literature on antidepressants. This time the “antidepressant is just a placebo effect” argument comes from a psychologist.
Irving Kirsch, a professor of psychology at the University of Hull in the U.K., says that antidepressants are nothing more than fancy and expensive placebos. He, of course, does not say this in a vacuum. No, of course not. He’s saying this in promoting his new book, The Emperor’s New Drugs (which, you know, is a “funny” play on the phrase “the emperor’s new clothes”).
Read on for a quick deconstruction of his argument (his argument as presented in an interview online, anyways).
The crux of the issue for me comes down to a question similar to this one: What about the hundreds of research studies published in the past 4 decades showing that antidepressants are better than placebo, or do have a positive impact on treatment? How do you explain all of that prior data without seeming like some kind of conspiracy theorist?
Kirsch has an answer.
Nevertheless, many studies suggest that antidepressant drugs do have an edge over placebos. How do you account for that?
When you do a clinical trial, you tell people that they might get a placebo. When researchers give placebos, what they are trying to control for is the expectancy of improvement, which can produce a sense of hope. You also tell them that the active drug causes side effects and what those side effects are. If I were a patient in one of these trials, I’d be wondering, well, what am I getting? And if I’ve started noticing side effects, and especially the side effects that had been described to me, I would no longer be “blind.” I would think, “Oh, my mouth is dry, that’s great—that means I got the active drug.” That would further increase my expectation that the drug was going to help. In the few studies where that has been assessed, about 80 percent of patients do figure out what group they are in. So it’s actually the side effects, the undesirable chemical effects of these drugs, that cause subjects on antidepressants to do a little better than those on the placebo.
So his answer is simple — people know they are on the active treatment. All these decades of experimental design research — the kind of stuff we predicate the very foundation of all scientific knowledge when it comes to a drug’s effect in humans — is flawed. Fundamentally flawed. “Blind” placebo studies aren’t really blind to anyone.
There’s some truth to his argument. There is research to suggest that blinded randomized placebo control studies aren’t as randomized or blinded as we might think. I’ll give him that.
But that means that virtually all research that has relied on this kind of study design is flawed and useless. Not just for antidepressants, but for virtually any psychiatric medication (and many others as well).
This line of reasoning also seems to call into other types of treatment as well. How can you show the same effect isn’t happening in all those blinded control studies of cognitive-behavioral therapy? One could argue it’s even worse for those types of treatment, as all too often, the “control” group didn’t receive any kind of placebo — they were placed on a wait-list for treatment. Of course some sort of human interaction with another human being is going to come out on top.
Even if your control group was “education” or “social talk,” I think most people are cognizant enough that they’re not engaging in psychotherapy.
Is this the new house of cards we have built? By pulling apart one set of studies, isn’t it possible to use the same logic and reasoning to pull apart virtually any scientific study done on the subjective human nature of “feeling better” or “improvement” on symptom checklists?
Read the full interview: Psychologist Says Antidepressants Are Just Fancy Placebos