I am wondering if an experiment - if run long enough - would automatically rule out the placebo effect.

Or do we know that the placebo effect can successfully last, say, years? Or does it tend to only last a set period of time and then fizzle out (such as a few weeks at most)?

Is it possible to say whether or not something is likely not a placebo effect, provided the researchers are unable to introduce a placebo to try to control for it? If your only options are "do the treatment" and "not do the treatment" is there some point where one would feel statistically confident that it is not the placebo effect? By this, I am trying to give a thought experiment of "suppose you have a drug where you cannot replicate something to have the same shape and texture, so all you have is 'consume' or 'not consume' as options."

Because if the placebo effect reliably will last forever in test subjects, then the answer is no, we can never be statistically confident. However, maybe it tends to diminish over time, and after some period of time (like 3 months?) we begin to be able to say "this is likely not the placebo effect from the real thing" despite being unable to administer a placebo.

Are there any studies on this?


1 Answer 1


I think that depends on the condition or symptom that the placebo is supposed to ameliorate. For instance, any pain that is diminished by a placebo pill is likely to return, but the same is true after taking a medication for the pain because it has limited duration. So clearly, the fact that that the placebo effect, after a single administration of placebo, has a time-limited effect isn't very useful if the active treatment also has a time-limited effect.

The situation might be different with, for instance, treatment with antidepressants or placebo for depression. Here, the treatment is typically given every day for a period and depressive symptoms are measured. In this case, the usual pattern is not that the placebo group have lower scores on the depression symptoms scales initially and then having rising scores again after a while, so this is contrary to your hypothesis.

However, a problem in placebo research is the effect of regression toward the mean. An expression of this effect is that in a medical setting, people tend to seek medical care when their symptoms are at their worst. Because many disorders and diseases have fluctuating symptomatology over time, this means that many of the patients that seek medical attention (and thus might be enrolled in a treatment study) are at their worst and will improve regardless of treatment. So in many treatment studies, all patients whether they get active treatment, placebo or no treatment at all actually improve. This poses a problem in regards to the placebo effect because much of the placebo effect is probably due to regression toward the mean and no "true" placebo effect.

So no, I wouldn't say that there are any hard evidence that I'm aware of that shows that the placebo effect diminishes over time any more than any active treatment effects diminish over time.

I think your question would benefit from a concrete example, as it's difficult to know what kind of scenario you have in mind.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.