In episode #89 we had a vigorous discussion about a resent paper published by Dr. Beatrice Golomb (et all) entitled “What’s in Placebos: Who Knows? Analysis of Randomized, Controlled Trials” in the Annals of Internal Medicine published by the American College of Physicians.
It stated that “there isn’t anything actually known to be physiologically inert. On top of that, there are no regulations about what goes into placebos, and what is in them is often determined by the makers of the drug being studied, who have a vested interest in the outcome. And there has been no expectation that placebos’ composition be disclosed. At least then readers of the study might make up their own mind about whether the ingredients in the placebo might affect the interpretation of the study” .
What this means is that when a study is published the contents of the placebo are not listed. This is important because placebo’s have effects on the study in two ways. There is the well-studied ‘psychological’ effect of placebo which is independent of the composition of placebo; there is also physical effects based on the composition of the placebo. Now independent scientist can assess the psychological effects but if they don’t know the composition of the placebo they are unable to assess the physical effects and thus ultimately cannot determine the validity of the comparison between drug and placebo.
Let me give you an example; my company (let’s pretend) wants to release a drug to treat type 2 diabetes. We did two trials; in group A my drug was tested against placebo A (composed of Peanut oil, high in B3) and in group B it was against placebo B (composed of Corn Oil, low in B3). The results showed a significant improvement compared to placebo A and no improvement against placebo B. If I am honest I will investigate the reason for the difference which due to the fact that B3 aggravated diabetes thus it is explained by the fact placebo A is really a nocebo (were symptoms are worsened by the administration of an inert, sham, or dummy (simulator) treatment) making the drug appear more effective than it really is.
However, it is (being overly cynical) more likely I will release results from group A as though it was an unmitigated success. Now, if my study included the composition of the placebo, independent scientist could point out that the placebo may not have been a neutral factor and cast honest doubt on my reported results. However, under current regulations, I don’t have to publish the contents of my placebo; thus there is NO way for independent scientist to know that my placebo contained B3 which may have made my results less reliable.
Now, not all examples have to be nefarious. One study testing a cholesterol drug in the 70s used Olive Oil as a placebo and showed little effect; later it was learned that olive oil has a natural ability to reduce cholesterol so the effectiveness of the drug was underrepresented in the study. The point being is that 1) scientist only can’t know the biological effects of placebo’s if they don’t know their composition, 2) that there currently no requirement (either legal or cultural) to include in a studies result the composition of the placebo and 3) this leave the processes of testing open to, at best, inaccurate results or, at worst, dishonest reports. So, the end run is there should be both profession and regulatory rules regarding at least the publication of the composition of placebos because note every fake pill is the same.