Placebos Don't Cure Anything
Hall, Harriet, Skeptic (Altadena, CA)
In Newberg and Waldman's response to Steve Salerno (Forum, Vol. 15 No. 2), there was a glaring error. They said "placebos can cure, on the average, 30% of the majority of physical and emotional diseases" This is so false it is laughable. Placebos can't "cure" any disease. The very definition of placebo is that it has no active ingredient capable of curing of even mitigating disease. Placebos can elicit a response where patients report less pain and other subjective symptoms. Placebos can be used in double blind research to help determine the effectiveness of test treatments. Placebos do not produce any objective effects on cancer, rheumatoid arthritis, infections, or any other disease. Placebos have not been demonstrated to alter the course of any disease, much less "cure" it.
The 30% figure probably comes from Beecher's 1955 paper (1) where he reviewed studies that compared ah active treatment to a placebo, and found that on average 35% of patients in the placebo arm reported improvement. Most of the improvement was likely due to the natural course of disease: people do sometimes get better with no treatment at all. …