The Emperor's New Drugs

Riki's picture


Controversial news has broken loose in the mental health community: antidepressants are hardly better than placebos!

So I have decided to read the recently published The Emperor’s New Drugs by Irving Kirsch for my book commentary. Kirsch certainly elucidated some troubling meta-analyses (albeit in a dry, boring way), but by the end of the book, I wished I had just read the preface and the epilogue instead of the whole book. The book constantly reiterated that antidepressants are hardly more effective than placebos.  I also found it frustrating that nearly a third of the book to explain what placebos are and how the placebo effect works. In spite of these feelings, there are a few interesting points I would like to discuss.

            Clinical studies of patients with varying degrees of depression show that as long as the patients are unaware of their treatment group, there is no relationship between symptom improvement and antidepressant dose. Pharmaceutical companies instruct physicians to increase the dosage of the antidepressants if the patient does not respond favorably to the drugs. Whether or not physicians currently know about the lack of dose-related therapeutic effects is uncertain. Kirsch points out that by increasing the dose, doctors are really just increasing the placebo effect. I’m not so sure about this. If my depressive symptoms showed no improvement after a few weeks of an antidepressant and my doctor decided to up the dose, I don’t know if I would necessarily feel better due to knowledge of an increased dose. If this happened several times, it is likely the side effects would intensify. Often, the side effects of these drugs are unbearable and can be a contributing factor to depression. Moreover, if I knew my doctor was increasing the dose and I still didn’t feel much better as a result, I would probably become more discouraged about my recovery. Feeling depressed about one’s condition is certainly an element of depression. Increasing the dose may have theoretical placebo benefit, but from personal experience and conversations with friends, if the patient’s symptoms don’t improve after the first dose of antidepressants, it would probably be more helpful to include some sort of adjunctive therapy.

            What I found most interesting was the notion that depression is not due to a chemical imbalance in the brain, a point Kirsch drew out in this book. We’ve all seen the ads for various antidepressants on TV, which grossly over-simplify the “cause” of depression to a chemical imbalance. Even I was fooled into thinking this. Perhaps my doctors too, because they give me the chemical imbalance line every time I am diagnosed with depression – “Depression is thought to be caused by an imbalance of certain chemicals in your brain. This drug increases the chemical levels in your brain, which will make you feel better.” Replace the word ‘chemical’ with ‘serotonin’, ‘norepinephrine’ or ‘dopamine’ and you have the standard explanation of depression. Kirsch reveals that when researchers lower serotonin, norepinephrine, or dopamine levels in the brains of healthy patients (read: not depressed), the patients’ moods are not affected. Moreover, the various types of antidepressants affect different neurotransmitters, yet patients respond equally when given one or the other. If a patient does indeed suffer from a lack of serotonin, then why would a drug that increases norepinephrine levels improve their symptoms just as much as a drug that increases serotonin? In fact, researchers only buy into the chemical theory of depression because depressed patients respond to these medications. I don’t understand why researchers only experimentally altered the levels of those three neurotransmitters and also why they only did so with “healthy” patients. Maybe the chemical theory of depression is not completely invalid after all. When drug companies develop new/improved antidepressants, they try to specify the drug’s action to a specific receptor subtype binding site to minimize side effects. If a drug company succeeded in completely reducing all side effects of an antidepressant, this means patients would not be able to break blind in randomized controlled trials. I am curious to know the difference in placebo and drug efficacies if the patients truly did not know which treatment group they had been placed in.

            Meta-analyses show that antidepressants are no more effective than placebos in mild to moderate depression. In treating severe depression, though, antidepressants are slightly more effective than placebos. Clinical trials comparing psychotherapy to antidepressants demonstrate that while psychotherapy is no better or worse than antidepressants, it is effective for all degrees of depression, not just severe depression. Comparing the long-term effects of antidepressants to psychotherapy reveals that patients treated with antidepressants are more likely to relapse than those treated with psychotherapy. This makes sense because while both treatments result in new neural networks, psychotherapy, especially cognitive behavioral therapy, can equip the patient with tools to deal with future stress. Of course antidepressants change the brain, but they aren’t actively teaching the patient new thinking patterns. What surprised me is that a combination of antidepressants and psychotherapy is not necessarily the most effective treatment for depression, according to Kirsch. While a blend of antidepressants and psychotherapy is more effective than antidepressants alone, it is not more effective than psychotherapy alone. It seems that psychotherapy is the most effective treatment for depression. If this is the case, why do so many people take antidepressants? Perhaps people simply don’t have time to devote an hour a week to therapy; taking a pill once a day is not very time consuming. But by covering up depression with a quick fix, a person is not uncovering the root of their depression. No wonder so many people relapse when only treated with antidepressants – they aren’t treating the problem at all, only the symptoms.

            Kirsch is quick to discuss the possibility of psychotherapy being a placebo as well since psychotherapy is roughly as effective as antidepressants, and as Kirsch is wont to restate, antidepressants are only slightly more effective than placebos, possibly due to an additive placebo effect. He argues that because behind a placebo lies the meaning of treatment to a patient, and because psychotherapy examines the meaning behind events in the patient’s life, then psychotherapy must be a placebo. I don’t think the effects of psychotherapy can be completely contributed to a placebo effect. I agree in the sense that based on societal views, if a person attends therapy for depression, they are likely to feel better simply by knowing they are receiving treatment proven to be effective, just like antidepressants. However, I do not think this can account for all of the effects of psychotherapy. There must be some difference between psychotherapy and antidepressants/placebos if the effects of psychotherapy last longer. In cognitive behavioral therapy, the patient is actively working on observing behaviors and thoughts and then changing them based on discussions with the therapist. How could the results be a placebo effect if the patient is causing the change directly?

            Kirsch ends with a chapter devoted to alternative treatments for depression. One of his suggestions is exercise, which he says is most effective for moderate to severe depression. However, if a person is severely depressed, how could they possibly gain the energy let alone motivation to get out of bed and go to the gym? Of course exercise will help with depression, but it is unrealistic to expect a patient in the throes of severe depression to have the ability to exercise. Kirsch also recommends self-help books. I can see this being helpful for mild depression, but in the case of moderate to severe depression, I think the patient needs to have the support of an outside source (i.e. a therapist) to help untangle self-rationalizations and thinking patterns. It is difficult to do so alone. Kirsch’s least helpful suggestion for treating depression is to somehow decrease social inequality. This is a ridiculously unrealistic expectation.

            The medical community and no doubt pharmaceutical community are upset over the publication of studies proving antidepressants to be marginally more effective than placebos. They are concerned that patients will lose trust in the health care system, and antidepressants may lose efficacy. I suspect this actually translates to lost profits rather than the well-being of depressed patients. Revealing the nature of antidepressants to the public may initially damage treatment efficacy, but perhaps depressed patients would benefit in the long run with the knowledge of cognitive behavioral therapy as the most effective treatment for depression.




Kirsch, Irving. The Emperor's New Drugs. Basic Books, 2010. Print



Post new comment

The content of this field is kept private and will not be shown publicly.