hi perry,

sorry for being a little late to the game.

i am intrigued by your questioning why the placebo effect isn't just considered an integral part of an effective therapy... my question is, if it were, would that eliminate the need to determine whether there is any non-placebo therapeutic benefit from any given therapy?

wouldn't we want to know if benefit were entirely attributable to placebo effect? if we would, there would still be a need for placebo-controlled trials, wouldn't there? if we wouldn't, then would we be effectively deeming the placebo effect an actual therapeutic benefit? we would have to be, i would think... which would then lead to trials testing different methods of producing placebo effect against one another... just thinking out loud here - this was not my original question.

my original question relates to the following statement:

"That is so because random assignment dilutes positive effect of patients’ expectation that they will improve from the on-going uncertainty about whether they are on the real thing, and conversely it elevates the placebo group’s expectation that they may be on the real thing."

if there are both positive and negative placebo effects, then, assuming blinding is effective, the former could not affect just the placebo group and the latter just the group receiving the real thing - the potential would exist for each participant, regardless of which group, to be experiencing positive, or negative, or both effects.

thus, if blinding is effective, wouldn't one have to assume that each of these potential effects would occur, on average, with the same frequency and to the same degree in both the placebo and the real thing groups?

thanks for your help
rosie