Psychiatrists, along with other physicians, try to practice evidence-based medicine– to use the treatments supported by the best scientific research. The most convincing studies are double- blind and placebo-controlled. A placebo is a “fake” treatment, such as a pill without active ingredients. Scientists use placebos because people often initially improve based on the just the expectation that they are getting help.

When I was a teenage babysitter, I had a special “treatment” for children’s “boo-boo’s.” I would touch the hurt child with a “magic leaf” plucked from a tree. I explained that the “magic” would take away all the pain. Almost every child touched by the leaf would immediately stop crying and say that the pain was gone. The “magic leaf “ was a placebo- it worked because the child thought it would work. Similarly, patients who gets a psychiatric treatment might improve because they think that they should be improving. (Usually, placebo responses don’t last more than a few weeks.)

Psychiatrists want to make sure that our patients are improving because the treatment actually works. Therefore, when possible, we compare new treatments against placebos. In a double- blind, placebo controlled study, neither the patients (and their families) nor the scientists rating the treatments know who has the active treatment and who has the placebo. Scientists’knowledge of which patients are getting the real treatment could influence how well they rate the treatment as working.

For example, say I’m doing a study on the effectiveness of pop music on my pets’ health. I’m convinced that pop music is bad for animal, and therefore only classical music (my favorite) should be played in our house. To prove my hypothesis, I keep my dog in my daughter’s bedroom while she listens to pop music for an hour, and my cat in my son’s quiet room at the other end of the house. At the end of the day, I examine the pets. I’m convinced that my dog seems a little upset. Don’t his whiskers look a little droopy? Didn’t he hesitate before batting chasing the ball that I offered him? My cat, on the other hand, looks very content to me. Her whiskers seem very perky while she enthusiastically bats at the yarn. Based on my findings, I rate the dog an “8” and the cat a “5” on my Moody Mammal Scale.

Can I use my research to justify that pop music shouldn’t be played in the house because it depresses pets? No, that study wouldn’t fly with scientists. I was a biased observer- I saw what I wanted to see. Similarly, if researchers know which is the active treatment and which is the placebo, they may unintentionally falsely perceive that patients getting the real treatment are the most improved.

Sometimes, decisions are based on open-label studies. These are studies that are not blinded- the patients and researchers know that they are getting the treatment. These are often done prior to the double-blind, placebo controlled studies, since they are easier to do and can help determine if the further research is warranted.

For example, a new antidepressant medication might have already been found to be safe and helpful for adults, but has not yet been evaluated for teenagers. Teenagers admitted to an inpatient psychiatry ward may participate in a study in which half are given the new medication and half are given an older antidepressant that has already been shown to be effective. (Of course, the risks and benefits would have been explained to the patients and their families in advance, and it would be their choice to be in the study. They would in no way be penalized if they declined to be in the study. Patients in these studies are closely monitored with medical assessments.)

It can be harder to draw conclusions from open-label studies, however, since we don’t know if any response is due to placebo. Sometimes, medications that seem effective in open-label studies turn out to be no more effective than placebo when the double-blind, placebo controlled studies are done. All we can say from open-label studies is that a treatment seems promising.

Why would doctors want to do these types of studies, and why would anyone agree to participate in them when there are already treatments proven to work? No treatment has been found that is 100% effective or free of side effects. A treatment that works for one person might not work for another. Researchers are always looking for better treatments. A patient who has tried other treatments with limited success might want to try something new.