THE EFFECTS OF PLACEBOS ON PARTICULAR DISEASES. One indication of progress in learning about placebos would be a showing that placebos are helpful with particular diseases. In this article in the Wall Street Journal (July 18), Sumathi Reddy says that: “Studies have shown that administering placebos reduces pain and symptoms in patients with irritable bowel syndrome and migraines, even when patients know they are taking a placebo.” She quotes Ted J. Kaptchuk, director of the Program in Placebo Studies and Therapeutic Encounter at Beth Israel Deaconess Medical Center in Boston and a professor at Harvard Medical School: “The pathways that we know the placebo effects use are the pathways many significant drugs use.â€
Prof. Kaptchuk says that a placebo can ease symptoms and pain even when patients know they are getting a placebo and cites a randomized controlled trial of 80 patients with irritable bowel syndrome and a study involving 66 migraine patients. Of course, the numbers in each sample are small, and there are the difficulties of obtaining a random and representative sample that Professor Kramer raised in the book I posted on yesterday.
When I referred in yesterday’s post to placebos as “the most studied drugs” I had in mind the paradox that more data has been collected about placebos than any other drug because all randomized controlled experiments have collected data on placebos. Focusing on the placebo results of those experiments is a step forward in establishing whether or not placebos are drugs that can be used in a predictable fashion or are only random unexplained residuals.
Do the research reports typically identify the composition and weight of the placebo?
I would expect that the placebo varies from study to study.
It would be good practice to describe each placebo used very specifically.
Why can’t statistical techniques be used to estimate a placebo “function” dependent on relevant variables? Yes, knowledge about what influences a placebo effect could help identify ways to develop a placebo-type effect without its being attached to a treatment or drug, but bounds might also be estimated for use in analysis of a treatment when a control group has not been set up. And shouldn’t the presence of a placebo effect be considered as a benefit of a treatment or drug?