Placebos Might Be Potent Enough to Affect Results: Study
By Lynne Peeples
NEW YORK (Reuters Health) Oct 18 - Some placebo pills may be potent enough to taint research results, hints a new review of more than 150 clinical trials.
The authors aren't just talking about a "placebo effect" on the mind, but rather physiological effects of the pills' constituents on the body. What's more, they found that less than one out of every 10 studies published in four top medical journals actually divulged what ingredients were used in placebo pills.
"We've been trained to associate placebos with being inert," lead researcher Dr. Beatrice Golomb of the University of California, San Diego, told Reuters Health. "But there isn't evidence that anything is truly physiologically inert."
"This really does question the primary foundation on which medical care is based," she added.
In looking at older studies of heart disease, Dr. Golomb noticed that placebos often consisted of things like olive or corn oil, which are now known to lower cholesterol levels.
Some earlier clinical trials of cancer and HIV treatments, she found, used placebo pills composed of lactose sugar and found relatively few gastrointestinal problems in the experimental group - but AIDS and cancer patients can be at an increased risk for lactose intolerance.
And these were just the rare studies that included the recipe. Further, she noted that the company producing the experimental drug often supplies the study placebo as well.
"This led me to wonder," said Dr. Golomb. "What rules are there about what goes into placebos?"
She contacted the U.S. Food and Drug Administration and learned that, in fact, there were no rules.
So Dr. Golomb and her colleagues decided to dig deeper. They reviewed 167 placebo-controlled trials published in 2008 and 2009 in English-language general and internal medicine journals with high impact factors and found that placebo components were very rarely described.
Only about 8% of the trials shared the contents of the pills. Studies were a little more open regarding placebo injections and other treatments with about one in every four divulging the information, report the researchers in the Annals of Internal Medicine.
"We can only hope that this hasn't seriously systematically affected medical treatment," Dr. Golomb said.
But she and her team suggest that it is a very real possibility, with potentially serious consequences. "An ineffective treatment might appear effective, or an effective treatment might appear ineffective in trials," senior researcher Dr. Jeremy Howick of the University of Oxford, in England, told Reuters Health in an email. "This is obviously harmful for public health since it could lead to use and payment for ineffective treatments, or failure to recognize effective ones."
"Placebos are best viewed as treatments in their own right," he added.
Dr. Matthew Falagas, director of the Alfa Institute of Biomedical Sciences in Athens, Greece, who was not involved in the study, also noted that ongoing research has shown how various substances used as placebos can have both expected and unexpected effects.
"From this point of view, it is useful to know the exact constituents of a placebo pill," he told Reuters Health in an email.
"The real hope is that this will lead the major medical journals to demand disclosure so we can do a better job of interpreting results as they come in," said Dr. Golomb.
She suggested that such openness could also ultimately lead to improvements in the composition of future placebos.
"When people talk about the placebo effect, they always presume it to be some kind of psychological affective suggestibility," she added. "But we need to keep in mind the possibility that it could be an actual effect of the drug."
Ann Intern Med. Posted October 19, 2010. Abstract