Bruce McCullough writes:
The Sept 2009 issue of Wired had a big article on the increase in the placebo effect, and why it’s been getting bigger.
Kaiser Fung has a synopsis.
As if you don’t have enough to do, I thought you might be interested in blogging on this.
I thought Kaiser’s discussion was good, especially this point:
Effect on treatment group = Effect of the drug + effect of belief in being treated
Effect on placebo group = Effect of belief in being treated
Thus, the difference between the two groups = effect of the drug, since the effect of belief in being treated affects both groups of patients.
Thus, as Kaiser puts it, if the treatment isn’t doing better than placebo, it doesn’t say that the placebo effect is big (let alone “too big”) but that the treatment isn’t showing any additional effect. It’s “treatment + placebo” vs. placebo, not treatment vs. placebo.
That said, I’d prefer for Kaiser to make it clear that the additivity he’s assuming is just that–an assumption. Like Kaiser, I don’t know much about pharma in particular, but like Kaiser, I feel that the assumption of additivity is a reasonable starting point. I just think it would be clearer to frame this as a battle of assumptions (much as in Rubin’s discussion of Lord’s Paradox).
I also agree with Kaiser that the scientific questions about placebos are interesting. As in much medical research, it’s frustrating how the ground seems to keep shifting and how little seems to be known. Or, to put it another way, a lot is known–lots of studies have been done–but nothing seems to be known with much certainty. There are few pillars of knowledge to hold on to, even in a field such as placebos that has been studied for so many decades.
Also, as Kaiser points out, the waters can be muddied by the huge financial conflicts of interests involved in medical research.
I would assume that, in most cases, the effectiveness of the treatments would tend asymptotic or at least bounded (for example, pain relief never exceeds 100%). Doesn't this argue against additivity?
@Mark: On the margin (with a 100% effective drug and a dichotomous measure) then yes, you could have cases where the additivity assumption isn't met. But it is a pretty good working hypothesis in most ranges of drug efficacy.
An assumption that seems to be implicit in Kaiser's discussion of additivity of effects is: the variances of the outcome stay the same even as the effect sizes get larger. If we assume that the variances increase with effect size, then the larger the placebo effect, the harder it is to detect a true treatment effect.
I think that the above equations are not quite right. I suggest:
placebo effect = effect of symptoms ameliorating naturally + effect of belief in being treated
That is, for many problems, people tend to get better even if they get no treatment and they know that they are getting no treatment. This effect contributes to what is usually called the placebo effect, even tho the patient is not being tricked.