A new report calls for standardizing placebos.
A different placebo effect. I'm Bob Hirshon and this is Science Update.
In most drug trials, the drug is compared to a placebo, a fake pill that's supposedly inert. But medical professor Beatrice Golomb, of the University of California at San Diego, worries that placebos may sometimes do more than we think.
We don't know how often placebo composition has affected the outcome of trials, because we have no idea what's been in the placebo most of the time.
She and her colleagues found that the vast majority of studies don't say what's in their placebos. The few that do reveal potential conflicts. For example, an old cholesterol drug study used olive oil and corn oil as placebos—oils that were later found to improve cholesterol levels themselves. Standardizing placebos is a tricky idea, but Golomb says more disclosure would at least help researchers figure out which placebos work best—or rather, least—in different situations. I'm Bob Hirshon, for AAAS, the science society.
Making Sense of the Research
The use of a placebo (Latin for "I will please") is an essential component of many medical trials. Although it's commonly used as a synonym for a fake pill, there are other kinds of placebo treatments, including simulated surgeries. Placebos are supposed to resemble the real medicine or treatment as much as possible, except that they lack the drug, device, or key technique that's being tested.
In a typical medical trial, patients who receive the actual, experimental treatment are compared to those who receive a placebo. Since the patients aren't told which one they're getting, this comparison is meant to separate any benefits from simply believing you're being treated from the benefits of the treatment itself. It's common for at least some patients to improve after taking just the placebo—that's what scientists call the "placebo effect." And it's generally assumed that the placebo effect stems from a psychological response to getting treated, even when the ultimate effect is on physical health.
This study doesn't completely negate that idea, and it doesn't suggest that placebos are useless as experimental tools. Without a doubt, modern medicine would be far less advanced without its long history of placebo-controlled experiments. What Golomb's paper does suggest, however, is that the assumption that a placebo does absolutely nothing isn't valid, and in some cases, the placebo might even confuse the results of a trial.
As Golomb puts it, there's really nothing you can put in your body that's completely neutral—not plant fiber, not sugar, not vegetable oil, or any number of things traditionally used as placebos. It may be that a given substance doesn't significantly affect the medical outcome you're looking at: for example, olive oil may have no real impact on a bacterial infection. But if it does, as in the experiments that used olive oil as a placebo for a cholesterol drug, then it may distort the results of the trial. For instance, since olive oil is now known to improve cholesterol levels overall, using it as a placebo probably made the drug look less effective than it was. That's because the drug may not have been that much better than the olive oil placebo—but because both the drug and the placebo "worked," not because neither of them worked.
Another example Golomb cites was an early HIV drug trial. Patients were given either an experimental anti-viral drug or a placebo made from lactose, or milk sugar. As scientists now know, patients with HIV/AIDS tend to develop lactose intolerance. As a result, many patients in the placebo group had to drop out of the study, because they were getting sick.
You might think that researchers wouldn't make those mistakes today, and they probably wouldn't. But the trials were done at a time when the potential effects of those placebos weren't as well known. Golomb notes that many of the placebos we're using today may have effects that we haven't discovered yet. The trouble is, medical researchers usually don't report the contents of their placebos. They're not required to, and in a review of recent studies, Golomb found that the vast majority don't volunteer the information. She says that publishing the contents of placebos across the board would at least give other researchers the opportunity to challenge them, or to re-examine them later. She also suggests that research teams should take more care to vet their placebos before they start a trial.
Now try and answer these questions:
- What does the term "placebo effect" usually mean?
- How are the potential effects of placebos that Golomb addresses different from the placebo effect?
- If no placebo is truly neutral, how might scientists use them more effectively?
- What are the benefits of reporting the contents of a placebo in a medical study?
You may want to check out the November 5, 2010, Science Update Podcast to hear further information about this Science Update and the other programs for that week. This podcast's topics include: a new report calls for better standards for placebos used in clinical trials, the genetics of centenarians, Parkinson's disease research reveals cluttered brains, and the importance of adequate training for marathon runners.