OJAI, CA—For the past 20 years, the perception among headache investigators has been that, in clinical trials, placebo response rates have been increasing, said Randall Weeks, PhD, Director of the New England Institute of Behavioral Medicine in Stamford, Connecticut. Furthermore, much variability in placebo response rates has been seen between different types of studies and patient populations. For example, response to placebo tends to be higher in acute headache treatment trials than in headache prophylaxis trials, and adolescent patients tend to have higher placebo response rates than adult patients.
So what contributes to these varied rates? At the Headache Cooperative of the Pacific’s 2009 Winter Colloquium, Dr. Weeks reviewed the neurobiology of the placebo response, as well as the many factors that affect the placebo response in clinical research and practice.
The Neurobiology of the Placebo Response
According to Dr. Weeks, there is evidence of a valid biological basis for the placebo response. Several studies have shown a connection between the placebo response and the endogenous opioid system, which is implicated in the regulation of stress responses and pain. Placebo-induced analgesia appears to be caused by b-endorphin binding to µ-opioid receptors, which blocks pain signals, he explained. In addition, treatment with naloxone, an opioid antagonist, has been shown to suppress the analgesic effect of placebo.
Dr. Weeks added that PET scans have shown activation in specific parts of the brain, such as those involved in emotional and pain processes, in people receiving placebo.
Is Something Better Than Nothing?
Dr. Weeks noted that physicians sometimes use placebo or placebo-like treatments when they have run out of evidence-based treatments, with the attitude that “doing something is better than doing nothing.” According to one study, however, this practice is not beneficial, he stated. Researchers analyzed data from more than 100 clinical trials and found no justification for the use of placebo. Compared with no treatment, placebo had no significant effect on binary outcomes, regardless of whether these outcomes were subjective or objective.
In a recent study, however, investigators found that of 231 physicians who responded to a 16-question Web-based survey, 96% believed that placebos could have therapeutic effects. Forty-five percent said that they had prescribed “placebo medications” in their regular clinical practice at some point in the past year. Only 12% felt that use of placebo medications in routine medical care should be prohibited. In another study, approximately 50% of surveyed internists and rheumatologists reported practicing placebo medicine on a regular basis. About 62% believed this practice to be ethical, and 68% described placebo treatment to patients as a potentially beneficial medicine or treatment not typically used for their condition. The most commonly used placebo-like treatments were vitamins and OTC analgesics.
The Placebo Response/Effect in Clinical Practice
Factors thought to contribute to the placebo effect in clinical practice include branding and cost of the medication. One study conducted in the United Kingdom showed that patients responded better to branded active treatment (versus generic active treatment) and branded placebo (versus generic placebo). In another study, half of patients were told that they would be receiving a great new drug that cost $2.50 per pill; the other half were told that it cost $0.10 per pill. Approximately 86% of patients who received the $2.50 pill reported pain relief, compared with 61% of those who received the $0.10 pill.
According to Dr. Weeks, some data in the literature indicate that color, shape, size, and taste of medications contribute to the placebo effect. For example, blue pills are considered calming, while red pills are considered activating; bigger pills are thought to be better; and capsules are preferable to tablets.
Cognitive and behavioral factors—such as acquiescence, expectancy, conditioning, suggestion, and motivation—also influence the placebo response/effect, said Dr. Weeks. The most commonly discussed theoretical positions involve expectancy and conditioning, he noted. “They are not diametrically opposed. Expectancy is the personal belief that something is going to happen. Conditioning involves paired associations, historically…. If conditioning theory runs counter to your expectations, expectations would probably take precedence regarding response.”
The Placebo Response/Effect in Clinical Research
In clinical research, variables thought to contribute to the placebo response and effect include type of treatment (the more invasive, the higher the placebo response and effect), treatment setting, and provider attitudes.
Standardization of transmission of information is also an important factor. “In a great many multisite/multicenter studies, this is extremely variable and may present a significant problem in terms of creating different placebo responses rates at these different sites,” said Dr. Weeks.
Another factor that influences the placebo response in clinical research is variability of symptoms. “What do we know about migraine? Some people get migraines that last 12 hours. Some last only four hours. Some people get one [migraine] a month. [The] same person may get six the next month,” he posited. The key is “understanding that there is variability with respect to what we’re studying.” Placebo responses can be affected by symptom variability and regression toward the mean symptomatically. Additional variables affecting clinical research include open versus hidden paradigms, single versus double-blind study designs, and number and types of study sites.