There are cheaper means of achieving the placebo effect, if that's what people desire

Expensive placebos work better. http://blogs.sciencemag.org/pipeline/archives/2015/01/29/exp...

Cheaper than water? =]

Cheaper than a $20 bottle of water masquerading as a real treatment.

A placebo effect always helps by definition.

To show something only has a placebo effect is the equivalent of showing it does not work.

In some contexts, you may be legitimately concerned only with the absolute level of "works" and not the level relative to a placebo baseline.

The (understandable, traditional) reason for measuring relative to placebo is that, "well, you don't need the drug if it's only as good as what a placebo gives -- just use the cheaper placebo! You should only bother with some special drug if it's better than what mere belief-in-effectiveness yields."

But that assumes that people are equally credulous of every placebo; if a person is more credulous of homeopathy, and credulity is hard to manufacture, then it's not quite wrong to say that it "works" under a relevant definition.

(I know this is kind of an angels-on-the-head-of-a-pen technicality, but eventually we'll want to actually understand why belief-in-effectiveness can matter at all, so we can exploit it, and reasoning like yours tends to give the idea that such an effort, by definition, can't work.)

Placebo effect is not some magic, like faith healing. It's statistical quirk. So placebo effect works whether or not you believe in it :-D

it's not a "statistical quirk". The placebo effect is reproducible and appears to be something that biology "does". how this works is an open area of research but not one that is really studied in detail.

As far as I know, a placebo is all about symptoms and how a patient perceives them. When you take care of a patient and give him/her a pill, statistically the patient will report to have recovered quickly or suffered less. There might a body reaction where some hormone is released which blocks pain (http://www.brainfacts.org/sensing-thinking-behaving/mood/art...), but no direct action as a cure.

Ok, I'll be more specific: the reason why drug trials are placebo-controlled is that human body is already equipped with the most miraculous drug of all: our own immune system. So, to accurately measure effectiveness of the drug we have to benchmark it against effectiveness of immune system. Placebo effect is just our own body fighting disease, no mystical forces involved.

Are you sure? I read into it a while ago and it seemed people were converging on statistical quirk as the accepted answer.

AFAIK, the "real" placebo effect is a very small and specific thing that goes on when pain alleviation or inflammation are experimental variables, because of the body's μ-opioid system.

The rest of the placebo effect, outside of that domain, is being increasingly found to be the result of under-powered experiments.

We can choose the argue about the scope of the effect (I think it goes beyond just pain and inflammation, and don't think it's limited to mu-opioids), but generally, yes, pain and inflammation are the two areas where the placebo effect has been seen to occur in a well-powered, well-controlled study.