You wouldn't lie to me, would you?

Dec 24 2010 Published by under Medicine

"Fake Pills Can Work, Even If Patients Know It"

"Irritable Bowel Syndrome: Placebo Works Even if Patients Know"

"Knowingly taking a placebo helps, study finds"

"Sugar Pills Work Even When People Know They Are Fake"

That's just a sample of the headlines accompanying a new study on placebos published in PLoS ONE.  The study attempts to nail down some questions about the placebo effect, especially regarding how to harness it ethically, without deception.  The placebo effect, a medical improvement not due to a biologically active treatment, is a difficult concept.  Some of it is stochastic or artifactual, that is, any time you study human beings, it is possible to measure statistical "improvement" that may not have real clinical meaning.  The other part of placebo is the response to being cared for independent of other biological interventions.  Journalist Steve Silberman (with whom I have disagreed about placebo in the past, and I'm still not sure who's right) quotes anthropologist Dan Moerman (emphasis mine):

"I was really surprised at how well the non-placebo group did,” Moerman says in email. “Note I don’t call them a ‘no treatment group’ because they, and everyone else, received exemplary treatment here: they were listened to, examined, encouraged, supported. They were able to talk with, and be taken seriously by, people who understood their issues, things they probably had serious difficulty discussing with their own families. I think it likely that the effectiveness of the placebos above and beyond all the other treatment would have been diminished without the whole system of compassionate care.”

So the non-artifactual part of placebo isn't really "no treatment", it is less tangible treatment.  In the current study, this distinction becomes important.  There are really two types of placebo: the kind used as a "dummy" treatment in placebo-controlled trials, and the placebo effect observed when we treat people well.  In attempting to harness placebo as an actual treatment, we run into a number of problems.  The first is teasing out what we mean by "placebo", the second the ethics of deceiving patients.  The current study by Kaptchuck, et al attempts to tease out the second question in order to answer the first.

They chose to study irritable bowel syndrome (IBS), an common, unpleasant disorder of unknown cause.  It is often exacerbated by stress and anxiety, and the natural course of the syndrome and of its attacks is to come and go spontaneously.  This fact itself makes studying interventions difficult, as it is harder to decide whether a treatment worked, or the patient simply got better as they would have.  Since IBS is often made worse with stress, it is also possible that simply comforting a patient may improve the symptoms.  Can this comfort be labelled "placebo" simply because it is non-pharmacologic?

All of this points to the difficulty of defining what a "placebo treatment" actually might be.  Giving comfort and care is active treatment.   The current study did not compare "placebo" to no treatment as they claim, but an active intervention, albeit one that was non-pharmacologic:

The provider clearly explained that the placebo pill was an inactive (i.e., “inert”) substance like a sugar pill that contained no medication and then explained in an approximately fifteen minute a priori script the following “four discussion points:” 1) the placebo effect is powerful, 2) the body can automatically respond to taking placebo pills like Pavlov's dogs who salivated when they heard a bell, 3) a positive attitude helps but is not necessary, and 4) taking the pills faithfully is critical.

Both groups were treated  "in the context of a warm supportive patient-practitioner relationship," but the subjects were essentially told that the pills would work, negating any real claim that this is a "placebo without deception."

This study simply shows what we already know: treating patients well helps them.  All the rest is commentary; don't bother to go and learn it.


Kaptchuk, 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

Two other excellent discussions online:

Ed Yong at NERS

Orac at Respectful Insolence

10 responses so far

  • Mara says:

    "All the rest is commentary; don’t bother to go and learn it."

    Paraphrasing Hillel for the win! (It's my favorite parable. I even told a version of it at my sister's wedding.)

    I'm fascinated by the placebo effect and ethics related to it, so I hope you continue to write about it.

  • Christina Pikas says:

    aw bummer. But thanks for explaining the study.

  • Richard says:

    The study showed that giving comfort and care is not quite as good as giving comfort and care + a sugar pill [and you don't have to lie about what the pill is]. This is a little more informative than 'treating patients will help them', no?

    • PalMD says:

      I guess the whole point is there's really do difference other than different levels of showing care for people, and that improvement in symptoms of self-limiting illness can be somewhat dependent on showing ongoing care for people. It's not about "harnessing" anything else.

  • The Blind Watchmaker says:

    I have recently thought that a placebo based treatment, whether overt or covert, is analogous to the "MaGuffin" in movies. This was coined by Alfred Hitchcock and means,
    "a plot element that catches the viewers' attention or drives the plot of a work of fiction". It does not really matter what the MaGuffin actually is. It could be "the stolen plans", or "the Heart of the Ocean" diamond. The point of a MaGuffin is that all of the characters think that it is important and all of their efforts are focused on it.

    Perhaps we all are looking for some kind of MaGuffin to focus our attention on. If I am ill, I can focus my anxieties on the illness and feel worse, or I can focus them on a medical MaGuffin of sorts, a placebo. The actual form of this nostrum is immaterial. It may be a sugar pill, a drop of a solution diluted to zero concentration, needles placed in my skin, a supplement, or an actual pharmaceutical product. My previous beliefs may make this misdirection stronger, but knowledge of its actual composition may be relatively immaterial. It acts as a focus for my anxieties. It may also represent a physical manifestation of a caring person. If the caregiver is compassionate and gives you a remedy, that compassion may be represented in my mind in the remedy.

    In the end, we feel better by misdirecting our bad feelings and displacing them with the good feelings that we get from caring individuals.

    As an evidence based doctor, I feel morally obligated to explain to patients that 20 to 40% of the subjective benefits of my treatments may be due to the placebo effect. I always feel a bit cold when I do this. At least now I can feel a little less like a kill-joy by explaining to my patients the placebo effect. It is the compassion with which it is given that is important.

  • The Blind Watchmaker says:

    This brings to mind magician and skeptic, Jamie Ian Swiss, who bills himself as "The Honest Liar".

    He will..."Convince you beyond a doubt that he has supernatural powers. But all the while, speaking from in-depth firsthand knowledge, he will tell you that he is NOT doing these things -- he is merely creating a terrifyingly convincing illusion."

  • The Blind Watchmaker says:

    On the other hand, leave it to Orac to deconstruct the study and point out the selection bias created by recruiting patients with the phrase..."placebo pills, something like sugar pills, have been shown in rigorous clinical testing to produce significant mind-body self-healing processes." This self-selects for a population predisposed to woo-type thinking. There was also selection bias on the part of the authors by putting less women in the control group and putting more constipation-predominant IBS patients in the "treatment" group.

    Oh well.

  • I'm not intentionally lying. I thought the splint you suggested was mostly placebo. And it is starting to feel a little better even without the splint, which totally conforms my bias. QED!

  • ceekay says:

    Watchmaker: the authors did not intentionally put more women than men in one group or more constipation patients in one group...What happened was -- the randomization failed to distribute these groups randomly....

    Failure of randomization, especially when assessed against numerous variables, happens all the time in small studies and is completely unsurprising.... --what this means is that the study needs to be repeated with a larger sample size

  • The Blind Watchmaker says:


    Yah, and with recruitment language that does not imply that the sugar pills are perhaps more than just sugar pills (not pills that have a 'mind-body' effect).