This item in the latest RSS News caught my attention since I have written about the placebo effect (here, here).
Stephen [Senn] argued that the placebo effect is really an example of [regression to the mean]. Selecting only patients for a clinical trial on a treatment of diastolic blood pressure (DBP) that have a DBP above a certain cut-off, for example, means that as a group the patients are bound to improve since their starting point is at the extreme. Excluding patients with lower DBP misses those in that group that would actually get worse (ending up with 'high' DBP).
Senn is the author of one of my favorite non-technical introductions to statistics (Dicing with Death), and I think he's pointing out something very intriguing. He is, in effect, arguing that the so-called placebo effect does not exist; it is a statistical artefact.
Is this a crazy idea? Or is this brilliant? What do you think?
Man, the placebo effect has been looked at so many ways by so many people, I'd be amazed if truly no one had suggested something like this before. But you never know...
Posted by: jme | 04/10/2010 at 07:49 PM
jme: you're right... a Google search yields many links to discussions of regression to the mean with respect to the placebo effect. It is considered one of many factors explaining the effect, though.
Posted by: Kaiser | 04/11/2010 at 12:44 AM
If it doesn't exist, why is there a difference between a placebo group and a control group (the group that doesn't get any treatment at all)? What about all these studies showing the difference between red and blue placebo pills?, "cheap" and "expensive" pills?, one pill vs 3 pills a day?... The placebo effect certainly exists.
The way patients are selected for trial does have an influence on the trial, as is pointed out in the quote, but this does not disprove the placebo effect.
Posted by: Cris | 04/11/2010 at 06:26 AM
No. The placebo effect cannot be explained away by regression to the mean. If it does then how do you explain nocebo effects? How do you explain psychogenic illness? It has been shown, for example, that people who get headaches when they think they’ve eaten MSG are just experiencing a placebo (nocebo) effect. How would that be regression to the mean?
Posted by: Charles Lambdin | 04/26/2010 at 03:25 PM
Actually, there are now two completely divergent schools of thought on the placebo effect. I don't know where "Kaiser" got the idea that there is a difference between the placebo group and the control group, because when researchers finally looked at this issue by examining EVERY trial that had ever been done with a third group that go nothing, Hrobjartssen, et al, found that there is no significant clinical effect associated with placebos.
Prior to that, Kienle and Kienle examined all the original work that led to the definition of the placebo effect, and concluded there was no placebo effect there. That is, placebo effect defined as an effect from the belief in treatment. They found a lot of different influences such as potential researcher bias, patient bias, patient reporting bias, all jumbled up into this one definition, that were all then attributed to belief in treatment. They looked at placebo work since and found a lot of "sloppy methodological thinking."
How do you explain the nocebo effect? Same as the placebo effect. Things like researcher bias, patient bias (noticing or reporting more negative than positive things if you are asked to report them (by being led to expect them) regardless of whether there is clinical change), etc.
There is a junk drawer of things that can cause changes that aren't due to the treatment, including just a natural tendency to improve, but in the evaluation, they all get attributed to a belief in treatment.
It's very important not to overattribute changes that aren't due to belief in treatment to belief in treatment, because it causes doctors to avoid using empirical observations and problem solving. This has, I believe, had a devastating effect on our ability to solve tough medical problems over time. How many times have doctors seen patients improve or experience unusual remissions or cures -- improvements that could have led to solving problems and helping other people -- but they didn't because doctors attributed what they saw to an irreproducible product of their patients' belief in treatment?
Posted by: A.J. | 01/12/2011 at 10:58 PM