The "belief' of the importance or relevance of a placebo effect always bothered me in the same way as the power of prayer.
I've never worked on clinical trials statistics so I was never privy to data from double-blind randomized controlled trials (RCT). Someone in the DC's Improbable Science did a great article on the subject.
The statistical problem associated with the incorrect assignment of an 'effect' to a placebo or prayers (my addition) is the missing group of people who didn't take the medicine nor the placebo. That group will revert to mean (get better by themselves - cure their cancers - with less probability, etc).
That group provides a baseline which should be subtracted from the two other group results prior to doing any regression. Without that subtraction, placebo(prayer) would be assigned the 'cured people' results in their group.
This incorrect interpretation plus a healthy dose of vanity is behind the all popular belief that people can effect their own health (to a larger degree that they really can), that they can cure other people (I am a Healer - a mutant or someone in close relationship with Divinity)...etc.
I might as well be honest and confess that this Pet Peeve is the result from the unnecessary early demise of my dear uncle Eradio. He was intelligent, most of the time. Once he argue with me that NASA was working on creating artificial atmospheres in space to allow for rocket jets to have something to push against. I replied with Newton's Third Law... He didn't have a retort..instead he notice that I had a Sardonic smile...:)
When facing stomach cancer, he decided to avoid a simple surgery and instead relied on a Healer...:) Needless to say, he died. I assign the existence of Healers to Utter Vanity (i am endowed with Superpowers by God) or Financial Gains ( a sucker is born every second) or Both.
Placebo effects are weak: regression to the mean is the main reason ineffective treatments appear to work
December 11th, 2015
|“Statistical regression to the mean predicts that patients selected for abnormalcy will, on the average, tend to improve. We argue that most improvements attributed to the placebo effect are actually instances of statistical regression.”
“Thus, we urge caution in interpreting patient improvements as causal effects of our actions and should avoid the conceit of assuming that our personal presence has strong healing powers.”
McDonald et al., (1983)