Last year, Harvard psychologist Irving Kirsch made a major splash on CBS' "60 Minutes" by proclaiming that most antidepressant drugs were little better than placebos.
Placebos are the so-called sugar pills given to patients in clinical trials of new antidepressant drugs to test whether the drugs do better than the fake pills. "The difference between the effect of an antidepressant and the effect of a placebo is minimal for most people," Mr. Kirsch told CBS' Lesley Stahl. "It's not the chemical ingredients of the drugs that are making them better; it's largely the placebo effect."
Those statements are intensely interesting to Bret Rutherford, a psychiatrist at Columbia University. It's not that he wants to argue with Mr. Kirsch about the value of antidepressants. Instead, he wants to elevate the reputation of placebos, and take their effects seriously.
"When we say 30 percent of people are responding to placebo" in these drug trials, "the important point is, we don't know what is causing that," he said.
Dr. Rutherford has a hunch about what is driving the placebo effect, and he is conducting a study now to see whether he is right.
In his view, it has two primary elements.
The first is "expectancy, which is a patient's belief in whether or how much a treatment is going to help him or her." The other is what he calls "therapeutic contact" -- the benefit of seeing doctors, nurses and other health care staff on a regular basis.
To test how much expectancy might make a difference in outcomes, Dr. Rutherford is doing an experiment now in which patients with depression are being put in one of two groups. The first group's members are told they will definitely get an antidepressant, and then begin taking the medication. The second group's participants are told they have a 50-50 chance of getting the real drug or a placebo.
While the study isn't finished, he already can say that the group that knew it was getting a real antidepressant is doing better than the other two groups, including the people who are getting the real drug but were told they only had a 50-50 chance of receiving it. That suggests that there is a real impact when patients strongly expect to improve.
He also is doing brain imaging on all patients before they start the trial and will then follow up afterward to see how much the parts of the brain involved in depression may change among the different groups.
There already are studies in other diseases that have shown the placebo effect creates real alterations in brain activity.
For instance, researchers in Italy have shown that when people think they are getting a pain medication, they subjectively experience less pain and their brains produce more natural pain-fighting chemicals. When they are given a substance that blocks those brain chemicals, the placebo effect goes away.
There also have been studies showing that the brains of people with Parkinson's disease produce the neurotransmitter dopamine and they experience fewer movement tremors after getting a sham therapy.
The intensity of the doctor-patient interaction in clinical trials may also play a big role in the placebo effect, Dr. Rutherford said.
When a patient with depression enters a clinical trial, even if he is getting a placebo, "it entails going somewhere every week when you've been isolated and lethargic. You meet with a bunch of young and enthusiastic people who care about you; you get your blood pressure taken and lab tests done and people in white coats mill about and give you medical sounding stuff. A warm, empathic person sits down and meets with you -- it's a massive amount of care."
In fact, it's far more care than a typical person with depression gets, because most of them get their drugs from family practice doctors who often have limited time to spend with them and even less time for follow-up visits.
"So when an antidepressant fails to beat placebo in mild depression," he said, "it does not mean that getting an antidepressant is no better than nothing at all. It just means that for mild depression, these kinds of therapeutic interventions are powerful."
David Kupfer, a depression researcher at the University of Pittsburgh, agrees.
When Pitt led the way decades ago in testing early antidepressants vs. placebos at several centers, the placebo effect at Pitt was as high as the drug treatment effect at the other centers, Dr. Kupfer said, and he believes that was the direct result of a philosophy here that psychiatrists ought to be available to the patients and their families around the clock.
Drug companies often have disliked the placebo effect, Dr. Rutherford said, because they see it as an enemy of getting their new drug approved.
But if scientists can figure out how the placebo effect works its magic and what actual brain changes it produces, he said, they should be able to use that information to improve patients' lives.