This past Sunday, CBS’ 60 Minutes aired a segment on antidepressants being no better than placebo. Harvard researcher and author Irving Kirsch says antidepressants do not work because of the ingredients inside them or any effect they have on serotonin, but because we believe we’ll get better when we take them. We could just as easily be taking a sugar pill, without knowing it, and we’d get the same effect, he argues.
If you saw the story, you might be feeling very confused right now. Even Lesley Stahl, who reported the story and whose husband takes an antidepressant, said she was confused. Kirsch and others interviewed were very confident that SSRIs don’t have the impact that pharmaceutical companies would have us believe.
I take one. It seems like it works for me. Yet does it really? Am I just fooled into believing it works?
John Grohol, editor-in-chief of Psych Central, responded, “What wasn’t mentioned in the 60 Minutes piece, because it was opinion journalism forwarding a specific viewpoint, is that Kirsch’s research is selective. He hasn’t looked at every antidepressant study ever done (now numbering in the thousands). He only looked at the clinical trials required to gain U.S. Food and Drug Administration approval for 6 antidepressant drugs (there are over a dozen on the market).”
I reached out for input from Dr. Marlene Freeman of the Massachusetts General Hospital Center for Women’s Mental Health for input as it applies to women with postpartum depression — well, everyone really, but my focus is women with postpartum depression — and I want you to see her response. It really helped me to understand the placebo effect better and why people like Hirsch make the claims they make:
The gold standard in the research of medications is the randomized, placebo-controlled trial. This means that patients enter a research study, and receive either the treatment being studied (such as an antidepressant) or a “fake” or placebo pill (one that has no active ingredients). The reason that studies are placebo-controlled is that the act of participating in a study itself is huge, and patient expectations also can influence treatment outcomes. [so it’s better if patients don’t know what they are getting]
In a typical, real-world setting (not a study), a patient decides to pursue or not pursue treatment, and if she selects treatment she picks herself she carries with her expectations about that treatment. In a clinical trial of an antidepressant for depression, patients enter the study and are thoroughly evaluated. They usually have blood work and hours of communication with study staff from the start of the study and then regularly (for example, every two weeks throughout the trial). Studies usually take place in clinical settings with doctors and nurses, and patients are carefully monitored for symptoms and side effects. Basically, participating in a clinical trial results in a lot of attention for the patient, often more than she would receive in routine care at her doctor’s office, as visits tend to be frequent with a lot of detailed evaluations.
Therefore, it is not surprising that many patients will feel better while they participate in a clinical trial, regardless of whether they receive an active medication or placebo. The goal of an antidepressant trial is to demonstrate that the antidepressant is better than placebo in the particular context of a clinical trial. Placebo response rates in depression studies are historically quite high, between 30 – 50%, meaning that factors other than the fake pill treatment helped patients improve. To show superiority over the placebo, the active antidepressant needs to show statistically more benefit than the placebo group, which can have substantial improvement. A study design needs to factor in the placebo response. Smaller studies are usually less likely to show a benefit of either treatment group compared to another.
It is not unusual for an antidepressant to fail to show statistically substantial benefit over placebo, particularly in smaller studies. In a “negative study,” one in which the antidepressant did not “beat” placebo, it doesn’t necessarily mean the antidepressant is ineffective. The study may just not have been one in which the efficacy of the antidepressant showed up beyond the general benefits of study participation and expectation that both the antidepressant and placebo groups experienced.
The placebo effect raises challenges for studying antidepressants. Huge trials are generally required, which are very expensive. The number of “negative trials” with high placebo response rates also show that antidepressants are only part of the experience patients receive when pursuing treatment for depression. Medication treatment should include thorough evaluation, frequent follow up, support and attention.
Receiving a placebo in an antidepressant study is a far cry from suffering from depression without getting help. Because depression is a serious illness, to purposefully leave a patient without a full evaluation and careful attention and monitoring would be unethical.
I have participated in a research study. It’s true that you get tons of attention and support. Everyone is watching you and talking to you about your problems and how you’re feeling and you feel like you have a team of people behind you. It’s nothing like what one might expect from regular visits with your primary care physician or another doctor. It makes perfect sense to me that people participating in a trial, especially if they have mild or moderate depression, would get massive benefit just from the participation alone, regardless of whether they’re getting the active or inactive medication.