Last week it was reported that police in Bangladesh had made a major bust at a factory that was producing counterfeit homeopathic drugs. The counterfeiters were attaching the labels of other drug producers to the remedies they were producing in their own factory. Dhaka's Daily Star reported the bust with the rather ironic headline "Fake Medicine Factory Busted".
Of course, even homeopathic remedies need to be guaranteed safe if they are sold in stores, and counterfeiters are not bound by the same safety controls as other more reputable sources. There are also 'intellectual property' issues concerning the use of other company's labels and trademarks. So I am not here to tell you that this drug bust was unnecessary or ridiculous. In fact I want to challenge The Star's implicit suggestion that homeopathic remedies are by their nature counterfeit therapies.
I should acknowledge right away that homeopathic medicines don't do anything beneficial in any kind of pharmacological or somatic sense. In homeopathic theory, dilution is not considered to reduce the effectiveness of a treatment. This is why you can sell a preparation of homeopathic chamomile even when the chamomile is so dilute that there is not even a single molecule of chamomile left in the preparation. In other words, homepathic medicines work (when they work) through the placebo effect.
So how is it that some trials have found that these medicines work better than placebo? The answer, put simply, is that some placebos are better than others. The distilled water and sugar pills that are used as placebos in clinical trials make particularly bad placebos. To explain why, it helps to look at another kind of 'fake' medicine.
In 2008, Kirsch et. al. published a paper which showed, fairly convincingly in my view, that the entire multi-billion-dollar industry in antidepressants is involved in the sale of placebos. There are two reasons this discovery has gone largely unheralded in the ethics literature and the popular media. First, Kirsch titled the paper "Initial Severity and Antidepressant Benefits: A Meta-Analysis of Data Submitted to the Food and Drug Administration", which doesn't exactly roll off the tongue. And second, antidepressants — despite being placebos—work extremely well. 60% of people see a significant reduction in depression when they take a course of these drugs, significantly better than the number who improve after taking an inert sugar pill placebo.
How can it be that antidepressants are simultaneously placebo-based and significantly better than placebo? I won't rehearse Kirsch's findings here, but part of the answer lies in the fact that antidepressants — unlike sugar pills — make really excellent placebos. The placebo response depends on an unconscious association between drug-taking and medical benefits, so the best placebos are those that best mimic active therapies, warts and all. If a placebo gives you insomnia, nausea or headache (as prozac can) or if it tastes bad (as a not-too-dilute infusion of chamomile might) then you will associate it strongly with active medications you have taken in the past, like pseudoephedrine, which causes insomnia and tastes bad.
The problem with homeopathic remedies is not that they are no better than placebos. Placebos can be wonderful and powerful drugs, and when they are used to treat the felt symptoms of a disease (such as pain or depression) rather than the underlying biological processes behind it, there is nothing fake or counterfeit about them. The problem with most homeopathic remedies is that they are often just too weak and bland to make decent placebos.
As well as making sure that the factories manufacturing them are producing safe, non-toxic remedies, we should make sure that their homeopathic products taste bad and produce mild, reversible, unpleasant side-effects. A slightly sickening, slightly revolting placebo would not be a counterfeit medicine at all.