Elusive Placebos

Alas, I have treated the Placebo effect too uncritically in my health econ class.  I’ll do better next year.  From Wikipedia:

The original 1955 article of Beecher "The Powerful Placebo" claimed a 35% placebo effect in 15 studies. The original article was in 1997 re-analysed and "no evidence was found of any placebo effect in any of the studies" used by Beecher. … The claimed "effects" were produced by spontaneous improvement, fluctuation of symptoms, regression to the mean, additional treatment, conditional switching of placebo treatment, scaling bias, irrelevant response variables, answers of politeness, experimental subordination, conditioned answers, neurotic or psychotic misjudgment, psychosomatic phenomena, misquotation, etc. …

Hróbjartsson and Gøtzsche published a study in 2001 and a follow-up study in 2004 questioning the nature of the placebo effect. … They performed two meta-analyses involving 156 clinical trials in which an experimental drug or treatment protocol was compared to a placebo group and an untreated group, and … found that in studies with a binary outcome, meaning patients were classified as improved or not improved, the placebo group had no statistically significant improvement over the no-treatment group. Similarly, there was no significant placebo effect in studies in which objective outcomes (such as blood pressure) were measured by an independent observer. The placebo effect could only be documented in studies in which the outcomes (improvement or failure to improve) were reported by the subjects themselves.

HT to Brandon Reinhart.

GD Star Rating
loading...
Tagged as:
Trackback URL:
  • http://liveatthewitchtrials.blogspot.com/ davidc

    I understand the placebo effect to be based on the belief that the mind can effect health. Are there similar effects you would expect to see if expectations really effect health?

    Would you expect people to survive up to important days? http://researchnews.osu.edu/archive/holideath.htm

    Should death curses work on believers if “live curses” (placebos) do not? http://www.rationalistinternational.net/article/2008/20080310/en_1.html

  • zanzibar

    Placebos tend to work for outcomes like pain, mood, etc. But a placebo won’t increase your CD4 count…

  • http://www.extropica.com Brandon Reinhart

    “I understand the placebo effect to be based on the belief that the mind can effect health. Are there similar effects you would expect to see if expectations really effect health?”

    The research seems to indicate that there is actually no placebo effect at all. It isn’t based on belief. The “placebo effect” is the same as no effect. Many sources will claim the “healing power of the placebo effect” but any improvement in the placebo group is apparently entirely attributable to other factors, such as flaws or biases in the study. One particular bias appears to be that in many of these studies the subjects are asked how they feel and usually answer in the positive giving a subjective result that isn’t really that useful. Many answer out of politeness that they are feeling better, others express a reduction in pain, but no physical indication this would be the case. Etc.

    The H&G Paper: http://content.nejm.org/cgi/content/full/344/21/1594

    I believe there has been at least one study where the placebo group was TOLD they were receiving a placebo and then compared to an untreated group and a treated group. The placebo group members still expressed the belief they were improving despite being told they were being given a placebo and despite the fact that their health status didn’t improve.

    Here it is:

    http://www.leecrandallparkmd.net/researchpages/placebo1.html

    Here’s another:

    http://jama.ama-assn.org/cgi/content/full/287/14/1840

    That one indicates that responses to the placebo effect have actually increased over the past few decades.

    I found a bunch of these sources from an article on bad science blog the other day, so it might make sense to find a broader set of references.

  • Silas

    The claimed “effects” were produced by … psychosomatic phenomena,

    Um … that *is* the placebo effect.

    zanzibar: I wish placebos did that for me. But somehow, in ten years, the doctors have never been able to work up a placebo that did something about my pain, although I’m sure it’s not for lack of trying.

  • Psychohistorian

    Subjective well being is highly relevant in many cases. Pain, for example, is purely subjective, so if I think I’m in less pain, I basically am in less pain. The same applies to discomfort from non-seriousness illnesses. If I think my cough is better, it is now more bearable and you’ve improved my subjective well being.

    Also, wikipedia is often a strong forum for dissenters. Dissenters are rather more likely to attack conventional wisdom than people who agree with conventional wisdom are to defend it, given the relative satisfaction each gets out of doing so. Not saying they’re wrong, just saying one should hardly view it as definitive.

    And, forgive me if I have this wrong, but isn’t one of RH’s major meta-points that you should defer to the consensus of experts in other fields (when prediction markets are unavailable)? He’s not a doctor, so it seems he should defer to the consensus of doctors on medical issues, which as far as I know supports the importance of the placebo effect. I may be confusing that philosophy with someone else, though.

  • Unnamed

    I can’t tell – is the claim here that placebos are less effective remedies than previously thought (and completely ineffective for many conditions), or is it that they have no medical benefit whatsoever and only influence what people say about their health? The latter conclusion seems unwarranted, especially given the physiological research that is starting to look at the effects that placebos have on the brain (especially in the processing of pain). I’m not an expert on this, but it looks like these studies are finding effects on lower-level brain processes (like sub-cortical mu-opioid receptors and P2 ERPs) that are inconsistent with a mere reporting bias.

  • Jor

    @Brandon
    One particular bias appears to be that in many of these studies the subjects are asked how they feel and usually answer in the positive giving a subjective result that isn’t really that useful

    A lot of outcomes in medicine are inherently subjective.Pain is _SUBJECTIVE_. If anything, the fmri studies on pain seem to indicate there probably is a role for psycho-somatic (placebo) control of pain. There is no physical measurement available. Depression is subjective, there is no know physical measurement available. Chest Pain (Angina) to some extent is _subjective_. Many interventions in medicine are aimed at improving these subjective measures as these things really can effect quality of life. You need a control arm. You need a control arm that doesn’t have built in expectations to the questions they are going to answer.