So in 2002, Dr. Kallmes convinced the National Institute of Arthritis and Musculoskeletal and Skin Diseases, or NIAMS, part of the National Institutes of Health, to give him money to compare vertebroplasty to a placebo.

Researchers rarely test surgical procedures, even minimally-invasive ones, against a placebo because unlike drug studies, where it’s possible to hand out harmless sugar pills, placebo surgeries involve breaking the skin, a move that can be ethically dicey because it opens the door to infections and other kinds of dangerous complications.

Dr. Kallmes also didn’t want to leave patients in pain if they got the dummy injection, so he told all study participants that they could get the other intervention if they had not experienced adequate relief after 30 days. 

He eventually enrolled 131 patients who were randomly assigned to receive either the bone cement or a dummy injection of short-acting painkillers.

Study investigators took great pains to make patients who got the dummy injection believe they were getting the real thing, including opening a container of the bone cement so the distinctive smell, something like nail polish remover, would waft through the air.

And each month, as he submitted his data, he said he half expected someone would call and tell him to cancel the study, something that happens when results show overwhelming benefit, or danger, early on.

But the call never came.

In the final analysis, patients in both the placebo group and the treatment group reported that on a scale from 1 to 10, their pain dropped by an average of about three points after 30 days, whether they got the bone cement or not.

The Australian study, which was very similar, had almost the same results.

 “The placebo effect, I don’t think, can be underestimated,” said Dr. Buchbinder.

“Especially for an invasive procedure, people have a higher expectation that they will benefit. They put themselves on the line if they have to have an invasive procedure, so they really have to buy into it and I think that really heightens the placebo response.”

Dr. Buchbinder said that after the study, she received a visit from a grateful patient who had been flown to the hospital by air ambulance because the fracture in his back was so painful. 

He was enrolled in the study, and after the procedure, his pain was so much better than he was able to drive home.

Curious, Dr. Buchbinder looked up his number to see which treatment he’d gotten. 

He had gotten the placebo.

Whether his own belief was enough to blunt his pain, or whether the fracture may have healed on its own, as many do, is anyone’s guess, Dr. Buchbinder said.

 “Unless you do rigorous studies, no matter how amazing it looks in these instances, I don’t think you actually can know unless you’ve done the studies,” she said.