Not All Placebos Are Created Equal

Variation Shown in Placebo Effect, Sham Care Process Influences Outcome

Photo by Graham Ramsay In collaboration with researchers throughout the Harvard Medical community, Ted Kaptchuk is exploring the placebo effect using a variety of approaches in diverse experimental environments.

The debate about the existence of a placebo effect has heated up over the past year as more and more lab experiments are detecting immediate physiological responses to placebos. A new study takes placebo investigations out of the lab and into a clinical trial, demonstrating a discernible placebo effect over time, according to an article in the Feb. 1 British Medical Journal.

While researchers usually use placebos in clinical trials to test the relative effectiveness of a new treatment, this trial pitted one placebo against another. “It’s upside down research,” said Ted Kaptchuk, HMS assistant professor of medicine at the Osher Institute and Beth Israel Deaconess Medical Center. “Our aim was to study the effect of the rituals in which the medical interventions were embedded rather than the efficacy of any particular treatment.”

Placebo vs. Placebo
The study of 270 individuals with chronic arm pain had two phases. In the first phase, 135 patients were given sham acupuncture and another 135 were given a placebo pill for two weeks. During this period, investigators found no strong evidence for an enhanced effect with placebo devices compared with placebo pills.

In the second phase of the study, the same patients were randomized again, with half the patients entering a trial involving a sham acupuncture device versus real acupuncture and the other half entering a trial of placebo pills versus real pain pills. The acupuncture trial lasted four more weeks (the length believed needed to see improvement), and the pill trial ran for six more weeks (the time needed to have the real drug in the bloodstream).

Patients receiving sham acupuncture in this second phase reported a more significant decrease in pain and symptom severity than those receiving placebo pills for the duration of the trials. The results demonstrate that a placebo effect may vary by type of placebo used and indicate that since two different effects were apparent, the placebo effect itself exists.

“These findings suggest that the medical ritual of a device may deliver an enhanced placebo effect beyond that of a placebo pill. There are many conditions in which ritual is irrelevant when compared with drugs, such as in treatment of a bacterial infection,” said Kaptchuk, “but the other extreme may also be true. In some cases, the ritual may be the critical component.”

The enhanced placebo effect illustrated in this study applied only to subjective reports from patients about their perception of pain and the severity of their condition. More objective measures of grip strength showed no difference in improvement between the two placebos.

The results also provide evidence that what doctors tell patients about side effects may directly influence their experience of them. Prior to participating in the study, doctors provided informed consent forms alerting the patients to the side effects they might experience: temporary soreness for acupuncture and fatigue and dry mouth for the pills. Of those receiving placebos, 25 percent of subjects undergoing sham acupuncture and 31 percent taking placebo pills reported experiencing the very side effects suggested to them—even when nothing was administered to cause these effects.

Impact Analysis
This study is the first rigorous look at the effects of placebos in a clinical environment since results in a May 2001 New England Journal of Medicine article showed no significant difference between placebos and no treatment in a review of previous clinical trials. Unlike this earlier meta-analysis, which relied on trials not designed to answer questions about placebos, Kaptchuk and colleagues designed this study specifically to investigate placebos.

The study takes the first step away from examining the placebo effect as a generalized phenomenon to investigating how it varies in specific clinical environments. Kaptchuk and his colleagues in several HMS-affiliated hospitals have initiated other National Institutes of Health–funded studies that will explore the placebo phenomenon in clinical trials for different illnesses and in laboratory experiments that focus on underlying neurobiological, biochemical, genetic, and psychological mechanisms.

Though the study results add evidence for the existence of a placebo effect in a clinical environment, Kaptchuk does not recommend the use of placebos with patients or deception in the doctor–patient encounter. The aim is to understand how the ritual of healing affects health outcomes. “While the science of medicine has a long history, the scientific investigation of the ritual of healing is still in its infancy,” said Kaptchuk. “My colleagues and I are trying to bring rigorous research to the questions of placebo and the clinical encounter. We hope to facilitate a more balanced appreciation of both the science and art of medicine.”

—Elizabeth Dougherty

top