It is well established that placebo response to treatments for pain (eg migraine treatments) can last a considerable time and can make it difficult to separate true effect and placebo effect. However one would expect the placebo effect to wear off with time
I have tried to find research work on how long one might reasonably expect a placebo response to last. Or to put it another way, how quickly can it decay?
It would be marvellous if a placebo response to treatment for migraines were to last 5 or 10 years, or even longer, for the placebo would then verge on being a cure in itself!
Of course a well constructed RCT is important in illustrating the difference between a placebo and "true" response, but RCTs are expensive things to organise and one has to be very clear about what you are looking for. Earlier in the evaluation process of a new treatment that appears to be producing an excellent result for longer than a year, it would be good to have some idea of when you can begin to discount a placebo response in favour of a true response to the treatment, be it 3,6,9 or 12 months
Does anyone have any idea where I can get more information on the decay rates of a placebo response in migraine treatments?