Search This Blog

25 October 2010

When is a Placebo Not Really a Placebo? Maybe More Often Than You Think

By Katherine Hobson

When you read in a study that a drug performed better or worse than a placebo, you probably don’t give much thought to whether the ingredients of the placebo had anything to do with the results.

Beatrice Golomb, an associate professor of medicine at the University of California, San Diego, School of Medicine, was researching cholesterol when she came across some older drug studies that actually said what was in the dummy pills — corn oil and olive oil, which of course we know now may have their own cholesterol-fighting properties. Study researchers noticed that the mortality rate in the control group was unexpectedly low, and that the drug being studied showed no improvement over the placebo, but they didn’t connect the dots and hypothesize that the placebo might actually have been helping.

“We hear the word ‘placebo’ and think ‘inert,’” even though there no known substances that are completely physiologically inert, Golomb tells the Health Blog. Even “filler” ingredients that pass through the body without being absorbed can have effects, say, by also preventing other things from being absorbed, she says.

Substances in placebos can also bias findings in favor of the drug being studied, she says, pointing to a study of a drug aimed at helping cancer patients keep weight on. The placebo included lactose, and cancer patients are prone to lactose intolerance, leading to increased GI symptoms in the control group. That may have made the drug look better than it was, she says.

There are no regulations on what goes into placebos, and very few studies actually report the composition of dummy pills or capsules: just 8.2% of the two years’ worth of studies published in four major medical journals that Golomb and colleagues combed through. The results of their research appear in Annals of Internal Medicine. For all placebos, including injections and other treatments, the disclosure rate was 26.7%.

“There’s a lot of attention paid to making them look and taste similar, but there’s very little attention paid to the actual ingredients,” she says. The placebo and nocebo effects (the latter is when a study subject reports ill effects from a supposedly inert dummy pill) get chalked up to patient suggestibility rather than possible physiological effects, she says.

Given that there’s no perfectly inert placebo, Golomb and her co-authors suggest that disclosure is key and call for major journals to lead the way by instituting a reporting requirement. “We can just accept that among the range of things that are never perfect in a clinical trial, placebos are seldom perfect,” she says.

Clarification: Golomb is an associate professor of medicine in the division of general internal medicine at UCSD School of Medicine. A previous version of this post referred imprecisely to her title.

No comments: