Placebos work, even when patients are in the know, study finds | Hacker News Reader

Placebos work, even when patients are in the know, study finds

articles.latimes.com

17 pointsAlex391715 comments

I just spent an hour yesterday explaining to a coworker that correlation != causation, data and statistics can easily deceive, and that most published study findings are likely incorrect. It was a little soul-crushing, therefore, to run across this headline and simultaneously hear an uncritical summary of this study on NPR this morning...

Said coworker told me, after seeing that correlation != causation, that she was never taught that in school. I replied that likely she had but that the concept is easy to forget and almost never reinforced in daily life. Thinking about it today, it seems like nearly all marketing efforts try to convince the intended audience that correlation == causation.

We're used to scientific studies being misrepresented, but additionally the study itself seems to have some problems:

http://scienceblogs.com/insolence/2010/12/more_dubious_state...

The author of that blog post is a moron. He couldn't find a single flaw in the methodology that the researchers don't bring up themselves. All he's doing he cut and pasting directly from the journal article and pretending to be intelligent, when in fact he's adding absolutely nothing to the conversation.

I would like to see this repeated with a medical condition that can be diagnosed without blindly trusting the word of the patient.

It should also be verified that the patient fully understands the meaning of "placebo".

This is perfectly logical, assuming the patients were aware of the placebo effect. I wonder what the results would be if some patients were given real medication while being told it was placebo.

No, it's surprising. I was under the impression that a placebo worked only if you didn't know it was a placebo. If you feed me a placebo while telling me it's a placebo then I would be under the impression that it would be non-effective, thus we wouldn't expect to get any placebo effect since I wasn't expecting it to work.

Having read this headline, though, I now have the opposite expectation: that placebos work even when you know they're placebos. This, for me, is excellent news, since I can now start taking placebos every time I feel sick and thus get better due to the placebo effect due to the fact that I now expect placebos to work and hence will now get a significant placebo effect from placebos!

The best part is that this works even if this study is bullshit.

I can now start taking placebos every time I feel sick and thus get better due to the placebo effect due to the fact that I now expect placebos to work and hence will now get a significant placebo effect from placebos!

I wonder if this is related to the Hawthorne effect:

http://en.wikipedia.org/wiki/Hawthorne_effect

Maybe not. From the article:

>> "My personal hypothesis is this would not happen without a positive doctor-patient relationship," Kaptchuk said.

>> Others agreed.

>> "What seems to be the active ingredient is the warm, personal relationship," said Dr. Howard Brody of the University of Texas Medical Branch in Galveston.

Those quotes excellently work with the theory of health care & placebos in http://hanson.gmu.edu/showcare.pdf

I don't know about you, but I have an excellent doctor-patient relationship with myself and I fully expect that if I prescribe myself placebos it would work just fine.

Your second paragraph describes the situation I expected at least some of the test subjects to be in.

Firstly, placebo studies don't require deception - patients are generally told that they might be getting the placebo, or they might be getting the real thing; they don't know which one, but uncertainty is different to deception. In a double blind study) the people interacting with the patient don't know who gets which either - so they aren't deceiving the patient, they genuinely don't know.

Secondly, in this case, what if the placebo is not genuinely inert? The 'placebo' was cellulose (dietary fibre) in a gelatine capsule, and apparently fibre is the standard treatment for irritable bowel syndrome.

Thirdly, showing that open label placebos might be better than controls is not enough to justify using an open label placebo in a clinical; instead, they would need to have compared an open to a closed label placebo. If a closed label placebo works better than an open label one, then a drug which is not efficacious could be accepted by a clinical trial because it was an unintentional closed label placebo and did better than the intentional open label placebo. In addition, closed label placebos are not just for the patient - they also play a role in double blind clinical trials of keeping the doctors dealing with the patient / assessing results from knowing.

There's also the possibility that the placebo effect in this case was due to some attempt to reduce cognitive dissonance: "why would they have me take this pill if it's actually useless?"

I suspect the Hawthorne Effect is the most likely explanation, though.

And here's the study: http://www.plosone.org/article/info:doi/10.1371/journal.pone...

This post is a placebo. You are now cured.

Reply on news.ycombinator.com