Placebo drift - why are placebos getting stronger?

Arthur Barsky, the director of psychiatric research at Brigham and Women's Hospital in Boston, says one big reason is conditioning.

"You've learned over the course of your life that going to the doctor, being examined, having him write out a prescription, waiting in line at the pharmacy, taking a pill, (these things) are generally followed by some benefit," Barsky says. "So you've kinda learned a pattern of reacting that seems to be pretty powerful."

Therefore the mere fact of participating in these activities, Barsky says, often makes you feel better.

Now this is incredibly important when it comes to drug trials — studies that try to figure out whether or not a drug is working. Researchers testing a new medication feel that because of the placebo effect, they can't just give their drug to patients then ask them how they feel, because patients are liable to report the drug is helping when it's just the act of taking medicine that's making them feel better.

Barsky says this is the real question: "Is your new medication more effective than a sugar pill?"

And so most of the time when there's a drug study in this country, the drug is compared to a placebo. But, says Barsky, this whole system is premised on some big assumptions about how placebos work.

"We have always assumed that it was a pretty constant effect. That the same person would have the same response to a placebo at different points in time. That similar illness would respond in a similar way to placebo. It was a constant, predictable, stable, and very important phenomenon that you saw in any medical interaction."

But it turns out the placebo effect might not be as stable as we've assumed. Barsky recently published a study that looked at a bunch of antidepressant trials that had taken place between 1980 and 2005, and he found that in 2005 patients in these trials responded to placebos way more than patients did back in the 1980s.

"The placebo response was about twice as powerful than it was in the 1980s," Barsky says. "That's a pretty significant difference."

In other words, placebos seemed to be twice as powerful as they were 30 years ago.

No one, including Barsky, really knows why the placebo effect appears to be changing. But Ted Captchuk, another Harvard professor who studies the placebo effect, says that placebo "drift" as it's now known, appears to be real. He says it's shown up in more than just antidepressant trials. And one possible explanation, according to Captchuk, is that there's been a change in our expectations.

For example, Chaptchuk points out that by 2005 our belief in the power of antidepressant drugs was very strong, and that might account for the shift. "There's a lot of confidence, and that changes both providers impression of what happened, and presumably the patient's experience of what could happen."

But also, says Kaptchuk, it could be that because drug companies mostly pay for drug experiments, doctors who do the research have a subtle incentive to say the drugs are working. And since doctors don't know who's taking a real pill and who's not, the fact that they see benefits in all patients would also inflate the placebo effect.

Then there's another possible explanation.

Researchers, especially in pharmaceutical trials, get paid for every patient they recruit. But often, Kaptchuk says, it's hard to find people, so doctors will sometimes admit patients to trials who simply aren't that depressed. And typically, he says, people who aren't that depressed are much more susceptible to the placebo effect.

"I don't think there's out-and-out fraud," says Kaptchuk. "I think that you're under pressure to recruit. It's really hard to recruit people. And you know, (when) it's borderline, (you) put them in. And those people on the borderline at the end, they are better in the placebo group."

Whatever the cause, placebo drift is something that has the potential to cause real mischief in medical trials.

"If the placebo response — that baseline — is shifting all the time, then it really confuses the issue of whether the drug is effective or not," Barsky points out.

More at NPR