SMART NEWS

Is That Pill a Placebo? This Program Can Tell, Even If You Can’t

A new algorithm could make it faster and less expensive to develop new painkillers

Correspondent

February 19, 2015

Chances are, you’d be hard put to tell the difference between a placebo and a painkiller. But a computer can—if it gets a good look at your brain. New research has yielded a computer program that can help determine whether you’ve taken an active painkiller or a harmless dummy pill.

The placebo effect has long stumped researchers eager to develop new ways to relieve pain. When people take experimental drugs, they often experience a profound pain response that makes it seem like the drug is working—until wider trials show that the drug itself wasn’t relieving pain. These false starts, coupled with a lengthy clinical trial process, can add to the expense and inefficiency of developing new drugs.

That could be changing, reports Francie Diep for Popular Science. Eager for a better way to research potential new painkillers, a team of scientists assembled brain imaging data from previous studies and conducted fMRI scans on more than 130 patients. In the past, they admit, it’s been difficult to figure out just how brain activity indicates drug efficacy. So they enlisted a partner in their research—a computer algorithm designed to learn as it analyzes more and more data.

After taking scans of patients who had taken an active painkiller and those who had taken a placebo, the researchers fed the data to their algorithm, using it to “train” the program to spot the effects of placebos versus painkillers. Then they put their program to the test, feeding it an unrelated dataset from a trial of a painkiller that was known to relieve pain. The algorithm worked about 70 percent of the time.

Even better, says Diep, the algorithm “never identified a placebo as a working drug.” That’s great news for researchers looking to make the clinical trial process quicker and more accurate. Eugene Duff, who helped head the study, told the Pharmaceutical Journal that the new technique could make studies less expensive, too. It could “potentially enable us to do smaller studies in the human population before devoting all these resources to more promising trials,” he noted. And with new research indicating that a painkiller with fewer side effects could be just around the corner, perhaps the study will speed up the process even further.