23 September 2009
Wired has a fascinating article about the placebo effect and how the pharmaceutical companies deal with it. Not only is there evidence that the placebo effect is growing (some drugs approved in the 80s and 90s would struggle to pass the FDA now), but it turns out there may be significant geographic differences in the strength of the effect:
Assumption number one was that if a trial were managed correctly, a medication would perform as well or badly in a Phoenix hospital as in a Bangalore clinic. Potter discovered, however, that geographic location alone could determine whether a drug bested placebo or crossed the futility boundary. By the late '90s, for example, the classic antianxiety drug diazepam (also known as Valium) was still beating placebo in France and Belgium. But when the drug was tested in the US, it was likely to fail. Conversely, Prozac performed better in America than it did in western Europe and South Africa. It was an unsettling prospect: FDA approval could hinge on where the company chose to conduct a trial.
I'm not sure how you separate out the geographic confounding of the drug response versus the geographic confounding of the placebo response when looking at differences between the two, but it is interesting nonetheless.
UPDATE: I just wanted to clarify why I thought this article was interesting so that folks do not think that I believe all the analysis contained in the article. The "effect" of the placebo treatment is clearly nonsensical as effects always need to about comparisons. What is identified from a clinical trial is the difference between the placebo response and the treatment response. My interpretation of the article (which is different than the author's interpretation) is that there is a lot of variation in that difference, both over time and over geography within the same drug. Since I have not read the academic articles that inform the article, I'm not sure if this variation is about what we would expect or not giving sampling variation, but the possibility of a systematic relationship is intriguing.
As Kevin notes in the comments below, there are some that are criticizing the article. It took a bit of searching (not that simple!), but I found a good response:
The author of the response simply claims that variation in the placebo response is simply sampling variance.
Posted by Matt Blackwell at September 23, 2009 10:04 AM