The role of placebo (from the Latin "I shall please") in the treatment of disease dates back more than 2500 years, with a multi-cultural history that includes its use in China, Babylonia, and ancient Greece. Placebo is defined as inert substances or procedures with no recognized specific physiologic action to treat a specific disorder. Thus, the practitioner may administer inert medication or perform sham procedures. If a measurable improvement is observed, the patient may be said to have a placebo response. If the improvement is attributable to the placebo, then the patient may be said to have exhibited a placebo effect. Response may be measured both by objective criteria such as changes in blood pressure and heart rate, and subjective measures.
Previous research on the potential efficacy of placebo has primarily focused on pain relief. In these studies, patients have been shown to release opioid-like substances that appear to exert an analgesic effect. Overall, the pain-relief response falls in the range of 30-40%. Interestingly, the more aggressive and invasive the intervention, the greater the response tends to be, with patients receiving sham surgery displaying a response of up to 70%. However, as placebo treatments are repeated, analgesic effect seems to dissipate. In trials, placebo response often appears greatest at 3-4 months, but diminishes over a year or so.
Some predictors of response have been identified. Placebo analgesia is greater in anxious patients and severe pain is more likely to be relieved by placebo. Patient expectation and practitioner attitude appear to have a significant influence on response, as well.
While we know that placebo can result in improvement in some patients, many important questions remain. Can true placebo effects (within the placebo response) be identified? While an effect may be observable, it’s important to consider the natural history of the disease. This is particularly true in the case of inflammatory arthritis, which is characterized by flares and remissions. Moreover, the perceived effect may be a reflection of the statistical phenomenon, regression to the mean.
Can we predict in advance who will respond? As noted, patient attitude and that of the practitioner may have significant influence. But to what extent will they alter outcome? According to Galen, "He cures most successfully in whom the people have the most confidence."
Does placebo have a role in clinical practice? And are some of the effects of the "tried and true" but untested interventions that we use, actually placebo? A recent study published in the New England Journal of Medicine provides a case in point. Mosely et al2 studied 180 patients with osteoarthritis. Patients were divided into three groups. One received arthroscopic debridement, the second received arthroscopic lavage, and the third had sham surgery. While all groups initially improved, they slowly returned to their previous state over the course of two years.
Patients seeking non-traditional interventions for illness present an interesting opportunity for study of these questions about placebo. Acupuncture constitutes one such therapy. Although complementary and alternative medicine is not widely taught in medical schools in the United States, as many as 42% of Americans have explored this avenue for relief of injury or illness. Interestingly, these same patients are reluctant to report this information to their physicians; one study showed only 38.5% would do so.
In the Houston Arthritis and Acupuncture Treatment Study (HAATS), researchers from Baylor College of Medicine, the American College of Acupuncture and Oriental Medicine, and Texas A&M (Suarez-Almazor et al), are studying the role of patient expectations, and influence of the practitioner on the efficacy of acupuncture in the treatment of osteoarthritis. Acupuncture is defined for the purposes of this study as "Insertion of dry needles into the body at specially chosen sites for the treatment and prevention of symptoms." In the HAATS study, the procedure is actually electroacupuncture; insertion of the needles is accompanied by low-voltage electrical stimulation.
Response to acupuncture is thought to involve the release of endorphins resulting from both the placebo effect and neuroelectrical stimulation. Opioid peptides released in the brain may inhibit naloxone. Moreover, there may be additional neurochemical changes in the brain and brain stem that are responsible for pain relief.
While acupuncture is used for a wide variety of conditions including musculoskeletal problems, general pain, headaches, allergy and respiratory disease, available data is most conclusive in demonstrating its efficacy in post-surgical pain and for nausea. There is some evidence to suggest limited efficacy in osteoarthritis. In the HAATS study, patients with osteoarthritis of the knee are randomized to one of three groups: 1) a group put on a waiting list for acupuncture (to serve as a natural control and to observe disease progression); 2) A group receiving electroacupuncture, based on traditional Chinese acupuncture modalities; and 3) a group receiving non-traditional acupuncture, using a slightly different technique.
The investigators will assess the effect of the practitioners’ communication patterns on the response to treatment by audiotaping the treatment visits with the acupuncturists. The investigators hypothesize, that through placebo effects, patients receiving high expectations about treatment from their acupuncturists, will have better responses than those receiving more neutral verbal cues. In the analysis, phrases like "There is no guarantee about the results of treatment." are considered neutral and statements such as "I am confident that this will work for you." are rated as delivering high expectations of efficacy.
While both acupuncture procedures involve insertion of needles and administration of electrical charges, the needles used, the points selected for treatment and the voltages and cycles of electrical stimulation differ. Practitioners and research assistants were trained to deliver information specific to the two groups. Those in the neutral expectation group are instructed to use language that focuses on the technical aspects of treatment, and qualifiers in describing potential benefits.
To test the efficacy of the process, 8 staff members underwent acupuncture. The investigators videotaped interactions between practitioners and "patients" in order to give feedback to practitioners and assistants. Of the 8 individuals who participated in this exercise, only one guessed correctly that they he had undergone non-traditional treatment.
Two hundred patients have enrolled in the HAATS study, with a planned enrollment goal of 580 participants. Interactions between practitioners and patients continue to be audiotaped to ensure quality control.
Acupuncture treatments are administered two days a week, over the course of 6 weeks. Results are measured primarily using the WOMAC™ Index for Knee and Hip Osteoarthritis, as well as through a questionnaire that measures frequency, intensity, and feelings of pain. All patients who receive non-traditional acupuncture are also eligible to receive traditional acupuncture at the end of the treatment period.
 The Placebo Effect. An Interdisciplinary Exploration. Herrington A. Ed. Harvard University Press, Cambridge, MA, 1997.
 Moseley JB, O’Malley K, Peterson, NJ, Menke TJ, Brody BA, Kuykendall DH, Hollingsworth JC, Ashton CM, Wray NP. A controlled trial of arthroscopic surgery for osteoarthritis of the knee. N Engl J Med. 2002 Jul 11;347(2):81-8.
 Eisenberg DM, Davis RB, Ettner SL, Appel S, Wilkey S, Van Rompay M, Kessler RC. Trends in alternative medicine use in the United States, 1990-1997: results of a follow-up national survey. JAMA. 1998 Nov 11; 280(18):1569-75.
 Hochberg MC, Berman BM. Traditional Chinese Acupuncture is Effective as Adjunctive Therapy in Patient with Osteoarthritis of the Knee. Data presented at the ACR/ARHP Annual Scientific Meeting, October 20, 2004.
Summary prepared by Nancy Novick