James Bowery <[email protected]> wrote:
> The null hypothesis upon which to base the control experiment: Even > though a placebo effect may be present and in fact much larger . . . > This is drifting off topic, so let me rename the header. I do not think the placebo effect exists. I read several papers years ago about this. I do not have time to go into the details but anyway, it was a mistake many decades old. Sloppy statistics and untested assumptions led people to think there is a placebo effect -- That is, an effect in which a prognosis improves because the patients think they are being treated when they are actually taking by fake medicine (something with no efficacy). One hypothesis is that people respond well because they think the doctor cares for them or is concerned about their well-being. This was tested in recent decades by dividing patients into two groups. One group is given a placebo and treated with kindness. The other is dismissed and sent home with nothing. The two groups recover at the same rate, with the same percent reporting success. The placebo and the treatment have no effect on the outcome. It is a fact that people often get better on their own. Nature cures many diseases. This fact clouded the issue and made doctors think that a placebo was curing people almost as well as some drugs. Or if not that well, it was curing them in significant numbers. They were comparing a drug that was supposedly effective against a control group of people who got a fake drug. What they should have done instead of this -- or in addition to this -- would be to compare the drug against a control group of people who get nothing. No fake drug, no sympathy. They would see that many of them also get better. - Jed
