When Is A Placebo Not A Placebo?

By Neuroskeptic | December 28, 2010 10:50 am

Irving Kirsch, best known for that 2008 meta-analysis allegedly showing that “Prozac doesn’t work”, has hit the headlines again.

This time it’s a paper claiming that something does work. Actually Kirsch is only a minor author on the paper by Kaptchuck et al: Placebos without Deception.

In essence, they asked whether a placebo treatment – a dummy pill with no active ingredients – works even if you know that it’s a placebo. Conventional wisdom would say no, because the placebo effect is driven by the patient’s belief in the effectiveness of the pill.

Kaptchuck et al took 80 patients with Irritable Bowel Syndrome (IBS) and recruited them into a trial of “a novel mind-body management study of IBS”. Half of the patients got no treatment at all. The other half got inert cellulose capsules, after having been told, truthfully, that the pills contained no active drugs but also having been told to expect improvement in a 15 minute briefing session on the grounds that

placebo pills, something like sugar pills, have been shown in rigorous clinical testing to produce significant mind-body self-healing processes.

Guess what? The placebo group did better than the no treatment group, or at least they reported that they did (all the outcomes were subjective). The article has been much blogged about, and you should read those posts for a more detailed and in some cases skeptical examination, but really, this is entirely unsurprising and doesn’t challenge the conventional wisdom about placebos.

The folks in this trial believed in the possibility that the pills would make them feel better. They just wouldn’t have agreed to take part otherwise. And when those people got the treatment that they expected to work, they felt better. That’s just the plain old placebo effect. We already know that the placebo effect is very strong in IBS, a disease which is, at least in many cases, psychosomatic.

So the only really new result here is that there are people out there who’ll believe that they’ll experience improvement from sugar pills, if you give them a 15 minute briefing about the “mind-body self-healing” properties of those pills. That’s an interesting addition to the record of human quirkiness, but it doesn’t really tell us anything new about placebos.

ResearchBlogging.orgKaptchuk, T., Friedlander, E., Kelley, J., Sanchez, M., Kokkotou, E., Singer, J., Kowalczykowski, M., Miller, F., Kirsch, I., & Lembo, A. (2010). Placebos without Deception: A Randomized Controlled Trial in Irritable Bowel Syndrome PLoS ONE, 5 (12) DOI: 10.1371/journal.pone.0015591

CATEGORIZED UNDER: papers, placebo, woo
  • http://www.blogger.com/profile/00926204336056169207 Disgruntled PhD

    True, it doesnt really add to the knowledge of people researching in the field. That being said, if it encourages doctors to prescribe sugar pills openly then we might see some change (or at least some extremely interesting case reports).

  • spike

    but isn't it all a pure 'constructive' expectation effect?
    I bet they would have got no effects if they had not told them this has proven to help in some patients. Actually to test this they could tell different populations (same number) that this has proven to work in x% of the patients and then compare the size of the effect as a function of x.
    I bet you'd get a monotonically increasing function there!

  • http://gehirniskraemerei.wordpress.com/ gehirniskraemerei

    I don't see why this study should encourage doctors to prescribe open placebos in general. Even in this study they used some form of deception by telling patients “3) a positive attitude helps but is not necessary” – i doubt that. E.g. Patients with depression or simply many other patients not interested in voluntary taking part in a “a novel mind-body management study” advertised in a newspaper are going to benefit that much from open placebos as they very well might lack the positive attitude the patients in that paper seemed to have. Plus most doctors won't take their time for a 30min educational 1on1 promotion of pills.

  • Anonymous

    I heard a report on this yesterday and it sounded to me like the people in the study were responding to the extra attention and information they received from the medical staff, not the placebo itself.

  • http://www.blogger.com/profile/08461338194309128443 Paul

    Did he actually say “Prozac doesn't work”?

  • Anonymous

    He said Prozac and other antidepressants don't really work better than placebo — a finding which has been replicated now a zillion times.

  • http://www.blogger.com/profile/06647064768789308157 Neuroskeptic

    I'm not sure if he did but this is how it was widely reported, hence why I said “allegedly” but fair to say I don't think he made any attempt to dispel that impression.

  • http://www.blogger.com/profile/06647064768789308157 Neuroskeptic

    And there's the fact that working no better than placebo is not the same as not working, strictly speaking, but w.r.t drugs it's generally understood that if it only works as a placebo, the drug itself doesn't work though the “placebo effect” might.

  • http://www.blogger.com/profile/08461338194309128443 Paul


    I would argue differently. Indeed, after listening to him on various R4 programs and IOP debates, I remember him saying that Prozac DID work – the problem as he saw it was that placebos worked as well.

  • http://www.blogger.com/profile/08461338194309128443 Paul
  • http://pharmtastic.tumblr.com/ Krishna

    I don't think that it is the study that is causing false hope in people that sugar pills can cure their ailments if they truly hope for it. It is the media & blogger statements such as “Prozac doesnt work” and “all placebos work even without deception” thats is going to make ppl think that. Why doesnt anyone mention that all the criticism towards this paper is already pointed out by the authors themselves in their paper – http://www.plosone.org/article/info:doi/10.1371/journal.pone.0015591
    please read the full paper before coming to a conclusion that this is s 'stupid study'.

  • Anonymous

    it's generally understood that if it only works as a placebo, the drug itself doesn't work though the “placebo effect” might.

    “Scientists” caught red handed in magical thinking, if there is NO statistically significant difference pray tell what is the evidence for this?

  • http://www.blogger.com/profile/06647064768789308157 Neuroskeptic

    Anonymous: Eh? All I was saying was, that if a given drug works no better than a placebo, it's fair to say that that drug “doesn't work”.

    Even though, technically, it might work in the sense that it makes you feel better, because of the placebo effect.

    I.e. when talking about drugs, “works” implicitly means “better than placebo”. I think most people would agree and that's how I use the word. It's just quicker to say “drug X doesn't work” even if it might be more complete to say “drug X doesn't work any better than a placebo”.

  • Anonymous

    Thanks for your take in this – a bit more sober than mainstream media outlets. Like one other commenter I wondered about the effects of extra attention – of being part of a study etc.

    Do drug trials take into account the effects of knowingly and voluntarily being in a drug trial?

    BTW in reports of this story I saw sugar being described as inert, and though my chemistry is pretty rusty I recall it being a powerful oxidising agent (it has been used as the oxidiser in rocket fuel for instance), and pure sugar can have quite noticeable acute affects on body chemistry even in small amounts I gather. I presume the experimenters meant inert with respect to IBS over the time period of the study? But did they say how they established this?

  • veri

    I have a sweet tooth and read motivational books, helps when I'm sad. This should be renamed the Willy Wonker effect.

  • veri

    Placebo lab conditions should be same no? This looks to be tampered with. If they asked them to hold a magic wand and use these conditions.. is that a placebo effect? I'd say no.

  • veri

    Anon, study looks to be studying deception the sciency way, how much sheeples trust experts. They should replicate the study with a man climbing out of the dumpster telling them it's gooood for them.. I bet they'll all be rushing to the loo.

  • Roger Bigod

    Is there a placebo rebound effect? If people find out they've been given a placebo and responded, is there a loss of self-esteem or an increase of distrust? Or if they relapse while still believing in a placebo, are the effects more severe?

  • http://www.blogger.com/profile/06473801377995007024 N.M.Levesque

    Given that the people were to swallow a pill, and the condition has to do with the digestive system–the effects of concentrating on the affected area may have been in play here.

  • Anonymous

    Isn't it remarkable how powerful placebo effects are? We are in an age of biological and neurochemical reductionism. And yet belief is as powerful as chemical agents. Interesting to say the least. And idiots proclaim that the mind-body dualism has been conquered by neuroscience. I think not.

  • veri

    – rolls eyes –

    babe, be nice. anyway i'm going shopping. laterz.

  • http://www.blogger.com/profile/06647064768789308157 Neuroskeptic

    Re: the sugar pills, it wasn't actually sugar, I misread the paper. The pills were described to the patients as being “like sugar pills”, but they were actually gelatine caps filled with avicel, which is basically cellulose (which is not digestible).

  • http://www.blogger.com/profile/06832177812057826894 pj

    Bollocks, comment lost in the aether.

    Shorter PJ:

    Hah, compare this study to Kirsch et al 2008 on antidepressants. In this study they find that there is no 'clinically significant' effect of placebo* (note that they say 50 points is needed and the difference between the two arms is less than this) and this is massively exciting and means that placebos are brilliant whereas in the 2008 study they failed to find a clinically significant effect of antidepressants and therefore we should stop using them.

    * The amount of work they put in to get expectation effects here is astounding and borders on outright lying to the patient.

  • Roger Bigod

    The placebo effect: change you can believe in.

  • veri

    lol. makes you wonder how deceptive the marketing field is. The potato: change you can believe in. GET A LIFE: change you can believe in, the new fragrance from Armani. The more obscure, the sexier the campaign, as if some girly guy guyish girl is breathing pelvic thrusts down your spine. seems out of place during recession.

  • http://www.blogger.com/profile/08149271199649905788 mercurialmind

    Apprently the people with IBS have more faith than the ones with depression.

  • http://www.blogger.com/profile/06647064768789308157 Neuroskeptic

    It would have been nice if they'd had a third condition: placebo pills but with none of the 15 minute briefing.

    Just give them the pills and say “these are placebos. They don't contain any medicines. Please take 4 per day, every day.”

  • Roger Bigod

    According to feely-touchy hemisphere theory, there's a person over in the non-dominant hemisphere who doesn't understand language but pays attention to demeanor and facial expressions. So it sees an authority figure in white coat, picks up emotional twinges from words like “help”, “feel better”, “relief” and decides to respond.

    The obvious control would be a surly, unkempt male of an ethnic minority.

  • http://www.blogger.com/profile/06647064768789308157 Neuroskeptic

    On that note, maybe you could do a version of the Milgram Experiment designed to maximize the placebo effect?

    “You will get better. The experiment requires it. You have no other choice.”

  • Anonymous

    Why isn't anyone doing research on the mechanism(s) of HOW the placebo effect works?

    If we are all just bags of neurons and chemicals, then shouldn't we be able to reduce down HOW mental processes can alleviate pain, the perception of pain, and even in some cases, (Rogaine) promote hair growth?

    Funny how the very people who defend the materialist/scientism paradigm won't look into the mechanics of how mind affects body.

  • Roger Bigod

    “So far, every patient responded perfectly, so if you don't, you'll be the deviant who spoils the study.”

  • Anonymous

    Materialists loathe the placebo effect! It smacks them in the face and threatens to shake their belief that matter trumps mind, every time and every place. But it doesn't. A kind of “inconvenient truth” for the materialists. Another inconvenient phenomenon taken from psychiatry is conversion disorder. No matter how you play it, psychology (and psychogenic) conflict rules supreme in explaining these pseudo-neurological ills. It's funny: My experience with neurologists is that they are all keen on psychological explanations for things like “non-epileptic” seizures and unexplained paralysis. But our biological psychiatrists have an extreme meltdown when faced with these puzzles, and usually go off frothing at the mouth sputtering about “biochemical imbalances” and other such nonsense! Dream on.

  • http://www.blogger.com/profile/06647064768789308157 Neuroskeptic

    Actually there's been some work on the biology of the placebo effect. Fabrizio Benedetti from Italy has done some nice studies using fMRI and has also shown that the opiate antagonist naloxone can block placebo-induced analgesia.

    However his work has focussed on placebo effects in pain, which may be a special case, because we know that there are specific neural pathways (right down to the spinal cord) involved in “gating” pain. Placebo effects in other conditions may be different.

  • Anonymous

    we're this far from nuking all of you….

    the X-MAS vacuum cleaner for the atheists….

    shermer, randi, myers, pz, dawkins, harris



    why does everyone always want to PUNCH you, shermer?


    take your meds, you little fckers…


    now we are going to bury you…

    And the lesson from all of this? DOUBLE!

    What do you want, you little ****ers?

    more of these idiots






  • Roger Bigod

    “I don't know if anyone mentioned it when you enrolled in the study, but just FYI, if you don't have a response, we'll have to report you to Homeland Security so they can enter you in their list of Atypicals. It shouldn't matter unless you need to fly commercial or get a driver's license.”



No brain. No gain.

About Neuroskeptic

Neuroskeptic is a British neuroscientist who takes a skeptical look at his own field, and beyond. His blog offers a look at the latest developments in neuroscience, psychiatry and psychology through a critical lens.


See More

@Neuro_Skeptic on Twitter


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Collapse bottom bar