Andy writes:

The third notable front page story in yesterday's NY Times (in addition to the Guiliani and skull stories) was the brilliant study concluding that a common knee surgery, costing taxpayers billions, confers a benefit no greater than the placebo effect. It is perhaps the only time that someone has compared the effect of a surgery against the placebo effect.

Not sure how the materialists here will respond to that one. To many of them, the placebo effect itself is not really accepted. It lacks material explanation.

There are really 2 placebo effects -- how people who are deceived into thinking that they are getting effective treatment seem to improve, and how people with no treatment also seem to improve.

The latter can often be explained by the fact that people tend to seek treatment when conditions are worst. Last winter I got antibiotics for a sore throat. After taking the pills, my throat is much better. But I'll never know if the pills did anything or not.

Whether there is much a placebo effect in the first sense is debatable. A recent study says no. See this discussion.

The current Newsweek reports that prozac and other SSRI-antidepressants perform only slightly better than placebos.

I have a theory about prozac -- that a slightly different placebo effect is at work here. Prozac has very mild side effects. It was considered a breakthrough (in part) because earlier drugs had more serious side effects. It may be that those mild side effects are crucial to getting results because they remind the patients that they are getting medicine for their condition. The placebos don't have any side effects at all.

They should really test prozac by comparing it against a placebo that has mild side effects similar to prozac. Until they do that test, I will be skeptical about whether prozac is effective. The very slight advantage that prozac shows in tests might be entirely explainable by the failure of the placebos in the tests to have side effects.

Update: USA Today says, "An estimated 5 million to 6 million U.S. children are currently taking at least one psychiatric drug. ... About 1.5 million children currently take Prozac-type drugs". Seems like a lot to me. Most of these kids are on ritalin, which is controversial for other reasons.

Update: Here is the study. Haven't read it yet, but why does it take a Freedom Of Information Act request to get this data out of the FDA? The data should be published as it is submitted to the FDA.