Placebo effects in pain management can exert very real and powerful influences either for good or harm. The question is, how can positive powers of placebo be harnessed to improve patient outcomes? The answer depends on the clinical skills of healthcare providers.
An article in USA Today last month [November 13, 2009], entitled “Placebo effect behind many 'natural' cures,” asserts that “People looking for natural cures will be happy to know there is one. Two words explain how it works: ‘I believe.’" The article goes on to note that the placebo effect — the ability of a fake pill or presumably ineffective treatment to make people feel better just because they think that it will — “looms large in alternative medicine, which has many therapies and herbal remedies based on beliefs versus science.” While acknowledging that this statement has pejorative implications, the author quotes Robert Ader — a psychologist and placebo researcher at the University of Rochester in New York — as saying, “placebos can have real and beneficial effects. Much of the results of certain alternative procedures are largely placebo effects."
The nature of placebos and their effects are actually quite complex and still poorly understood, but a closer look at placebo powers suggests interesting possibilities for better pain management research and practice. In a classic paper, more than 50 years ago Harvard researcher Henry Beecher  discussed placebos as having a double meaning: (1) a “pure placebo” intended or expected to have no therapeutic effect (eg, a “dummy drug” used in research), or (2) something known or intended to act through psychological rather than physiological mechanisms. Whether a “pure placebo”— a completely inert agent or a therapy exerting absolutely no mental or physical response — actually exists in medical science is debatable.
After examining the limited clinical research of the time, Beecher found that placebos were consistently effective in more than 30% of patients treated for pain conditions, providing clinically significant reductions in pain of 50% or greater. A further important observation was that placebos were most effective when the pain state was greatest. Beecher also makes distinctions between two phases of pain suffering: (a) the original pain sensation, and then (b) the reaction to it, or its processing by the brain. As might be expected, placebos exert their most profound effects during the second phase, which is psychological and subjective in nature, by altering the reaction to or perception of pain. In certain circumstances, placebos also can incur toxic effects (adverse reactions) that are both objective (signs) and subjective (symptoms).
Today, there is an emphasis on randomized, double-blind, placebo-controlled trials as the “gold standard” for clinical research. This approach allows researchers to filter out the influence of confounding factors, such as psychological placebo effects, that could make it difficult to know if the active treatment in question is truly worthwhile or not. However, placebos in pain medicine research often exert surprising benefits. In previous UPDATES blogposts we reported on a number of pain therapy trials that demonstrated robust placebo effects: for example, sham procedures — during acupuncture, vertebroplasty, and surgery for migraine — a placebo analgesic cream, and a dummy pill for abdominal pain were all noted to produce considerably beneficial outcomes. In some cases, whether or not the placebo interventions were actually physiologically inert must be questioned and, in other cases, the role of patients’ positive expectations, beliefs, or hopes must be taken into account [Tilburt et al. 2008].
Brain imaging studies show that positive expectations or beliefs ("I know these pills will help") can incur neurobiological changes (largely in limbic regions of the brain) and affect levels of chemical messengers that modulate pain [Tracy 2008]. Conversely, anticipation of pain or harm can influence brain activity in opposite ways to produce adverse reactions. For example, a current literature review found that the specific adverse event profiles associated with antimigraine medications (either triptans, NSAIDS, or anticonvulsants) were matched by negative effects reported in placebo groups. In other words, research subjects knowing of possible adverse events associated with a particular active drug that they might be taking as part of a clinical trial actually developed those effects to a certain extent even when administered a placebo — subjects anticipated certain harms and they happened, which is called a “nocebo” effect [Amanzio et al. 2009].
While the potential for nocebo effects might advocate for not informing patients of possible adverse reactions associated with a therapy, there are ethical and medicolegal concerns associated with such practice. Another questionable practice is the prescribing of agents expected to have little or no pharmacologic benefit for the particular condition in hopes of inducing a positive placebo effect. A fairly recent survey of 679 internists and rheumatologists in the U.S. found that up to 58% prescribed placebo treatments on a regular basis [Tilburt et al. 2008]. Most prescribers (62%) believed the practice was ethically permissible; however, the most frequently prescribed “placebos” included over-the-counter analgesics (41%) and vitamins (38%), and to a lesser extent sedatives (13%). Actually, however, these were not truly inert agents, and nearly 7 out of 10 practitioners described the agents to patients as potentially beneficial treatments not typically used for their condition; curiously, roughly 5% of respondents said that they honestly described the agents to patients as placebos.
Clinical Implications: In their survey report, Tilburt and colleagues  reasoned that most physicians probably felt that doing something was better than doing nothing; for example, nearly 6 out of 10 respondents said they would recommend a sugar pill for patients with chronic pain if it had been shown to be more effective than no treatment at all. Unfortunately, the survey did not inquire about outcomes: Did the “placebo treatments” actually help the patients?
Intentionally recommending or prescribing otherwise ineffective treatments solely to invoke positive placebo effects seems to involve an element of deception that is contrary to the principle of informed patient consent. Yet, there are several principles of placebos that might be helpful to practitioners:
- As Beecher  suggested long ago, for any analgesic or other pain therapy, the total benefit may be equal to its “active” effect plus its placebo effect. For example, a clinical trial may show a 50% response rate in the placebo group and a 70% response in the treated group; the 20% difference is the “active” effect of the treatment itself. Could the success rate be improved by enhancing the placebo effect?
- The reactive or suffering component of pain is affected by neurobiological processes that may be altered by a patient’s expectations, beliefs, or hopes. In such cases, the placebo power of any therapy might be bolstered by the healthcare provider’s own positive attitudes and beliefs communicated to the patient. If the practitioner reinforces a positive and optimistic mindset in the patient regarding the effectiveness of a therapy it lends added potency.
- The more severe the pain or suffering, the greater latitude there is for invoking favorable placebo effects, at least to some extent.
- While discussing potential adverse effects associated with any pain therapy could be a prudent component of informed patient consent, the possibility of invoking negative expectations and nocebo effects must be considered. The ways in which possibly negative aspects of treatment are presented by the healthcare provider can make a critical difference in the outcomes.
> Amanzio M, Corazzini LL, Vase L, Benedetti F. A systematic review of adverse events in placebo groups of anti-migraine clinical trials. Pain. 2009(Dec);146(3):261-269 [abstract here].
> Beecher HK. The powerful placebo. JAMA. 1955;159(17):1602-1606.
> Tilburt JC, Emanuel EJ, Kaptchuk TJ, Curlin FA, Miller FG. Prescribing “placebo treatments”: results of national survey of US internists and rheumatologists. BMJ. 2008(Oct 23);337:a1938(online) [see abstract].
> Tracey I. Imaging pain. Br J Anaesth. 2008;101(1):32-39 [article PDF here].