In the case of something like acupuncture (as an example), it is very hard to set a placebo group or control group, since this therapy is not like taking drugs. Lets say if the needle is not in the right place (as a sort of placebo treatment), it will still have some other external effect.
So is it appropriate to say that a control / placebo group really exists in this kind of intervention?
If you are testing a medication, the right control is a similar pill without the active chemical.
If you are asking whether it matters exactly where acupuncture needles are placed, you can compare "accurate" placement with placement in the "wrong places".
If you are asking whether skin penetration matters, you can compare actual acupuncture with sham acupuncture where the needles retract so don't penetrate.
A summary of such work:
http://www.dcscience.net/2013/05/30/acupuncture-is-a-theatrical-placebo-the-end-of-a-myth/
