The "placebo effect" doesn't work for all. Logically, there is no reason to expect it to work -- you may just as well argue that if you are *looking* for an effect to occur, then you will be more apt to notice it *not* occurring. This will particularly be the case with pessimistic people and those
who tend to play "Devil's advocate."
I doubt that I could be tricked into thinking a placebo works; I am more likely to think that it won't work, and may even notice a *worsening* of my symptoms rather than an improvement.
Apparently, for some patients there is a "placebo effect", and for others, there is an oppositely directed "devil's advocate effect".
Before a real drug trial or drug regimen, a preliminary exam should be run, wherein a patient is given a placebo, and the amount of improvement/ disimprovement he reports should be mathematically converted into his "Placebability Rating" factor.
If the patient turns out to be a Devil's Advocate, this is written on his medical chart for all future doctors to see.
Future doctors could then, when seeing this patient, prescribe a pill, all the while saying "This pill almost never works; but I suppose you can try it if you insist - in my opinion it's a waste of time..."
If the pill is a placebo, it will then have a positive effect even upon the devil's advocate (due to the doctor's sly delievery); and moreover, if the pill is real, its effect will be even more beneficial than its initial estimated value.
Drug companies could certainly make use of placebability/devil's advocacy measurements, for they will then be able to more skillfully administer the doses during experimental trials, so as to arrive at more favorable outcomes. And you know what? If this idea is fully adopted by the national medical board, the reported outcomes will be true.