In a recent Undercover Economist column in the Financial Times (The Ig Nobel prizes: in praise of ridiculous research), Tim Harford mentioned that there has been a study showing that cheap placebos are less medically effective than expensive placebos.

This sounds slightly ridiculous, but, then, the placebo effect is fairly strange anyway.

So how strong is the real evidence? Do expensive placebos have stronger effects than cheap ones?

  • So the Placebo effect might just work on itself. It must be related to that pheomenon that people view more expensive things as better, even if its just a cheep thing with a large price tag.
    – Ryan
    Commented Oct 17, 2016 at 15:56
  • 2
    Here is a link to the original study. My statistics background is not strong enough to judge how strong the evidence is.
    – Tim
    Commented Oct 19, 2016 at 5:59
  • 7
    It doesn't sound ridiculous at all, if you interpret the placebo effect as based on the patient's belief in the efficacy of the drug. People will generally assume a more expensive product as better (using price as a heuristic measure of quality); add a scalable placebo effect (the stronger the belief, the stronger the effect) and you will get a stronger placebo effect if you say the drug was more expensive.
    – IMSoP
    Commented Oct 20, 2016 at 13:28
  • @IMSoP There is a problem with that theory. Recent studies have shown that a placebo can be effective even if the patient knows that what they're taking is a placebo. So yes, the placebo effect is weird. nhs.uk/news/2015/07July/Pages/…
    – GordonM
    Commented Nov 16, 2016 at 16:11
  • Yes, patients report that they feel better in certain ways (less pain, reduced nausea, etc), but that doesn't mean placebos have any real "effects". See this related question
    – BradC
    Commented Nov 21, 2016 at 17:56

2 Answers 2


There isn't strong evidence that any placebos have objective clinical effects.

Placebo interventions for all clinical conditions

"We did not find that placebo interventions have important clinical effects in general". They only found some small effects on patient-reported outcomes, not objective outcomes.

  • 6
    An interesting finding, but not, I expect, the consensus view. Placebos are widely used as comparing treatment to nothing is thought to bias the result to the treatment. whether placebos have been studied independently is a different issue, though relevant.
    – matt_black
    Commented Oct 17, 2016 at 7:48
  • 4
    I suspect the key word in that conclusion is important. Since their overall result was "The overall pooled effect of placebo was a relative risk of 0.93 (95% confidence interval (CI) 0.88 to 0.99)" where "a value of less than 1 indicates a beneficial effect of placebo", it seems they in fact did find a statistically significant result
    – Henry
    Commented Oct 17, 2016 at 8:02
  • 2
    @matt_black Placebos control for measurement and reporting artifacts in addition to any possible clinical effect from "any treatment". It seems very plausible to me that placebos have no real effects themselves.
    – Paul
    Commented Nov 16, 2016 at 18:26
  • @matt_black I think that is the consensus medical view about placebos, even though they are widely misunderstood by the general public. Placebos aren't some "mind over matter" trick that somehow magically improve objectively measurable bodily symptoms, they are simply an acknowledgement that our perception of our own non-specific symptoms (pain/nausea/etc) is filtered through our messy mind. When you perform tests with objectively measurable outcomes, you find placebos don't do much.
    – BradC
    Commented Nov 21, 2016 at 19:53
  • @BradC Maybe, but truly objective things are a lot more scarce than you might think. Placebos also affect the doctor's perception of the patient's symptoms. Otherwise medical trials would be a lot easier.
    – matt_black
    Commented Nov 22, 2016 at 0:07

This is the article where the research results were published in. The double blind study was quite small, but the results were still significant. In other trials analogous results were found; if people have a larger expectation that a treatment will work, then the placebo effect will be stronger. E.g. in this study the effects of oxygen was compared to placebo oxygen in the treatment of high-altitude headache. It was found that the effect of the placebo oxygen was stronger, involving more biochemical pathways, if the patient had previously been given real oxygen. See also this article.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .