Sunday, February 13, 2011

Homeopathy: The Make-Believe Medicine

While I was watching TV last weekend, I saw an odd commercial.

Oscillococcinum. That’s a long word that sounds a lot like the technical name that you might associate with many other medicines: acetaminophen, pseudoephedrine, fexofenadine, and so on. You would logically think that it is a new drug to treat flu symptoms.

What is oscillococcinum (oss-sillo-kok-sin-um), exactly? According to the website for this particular product, it is "anas barbariae hepatis et cordis extractum 200CK” with added sucrose and lactose (that is, sugar).

Nothing like a dead language to obscure what the ingredient really is: “Muscovy (wild) duck liver and heart extract.” Notice that nowhere on the product’s website do they actually explain this.

White Muscovy duck (Cairina moschata) with wings outstretched
(Photo by Steven H. Keys, from Wikimedia Commons)

Now this sounds weird on its own—who would drink duck liver and heart extract to ease flu symptoms?—but ultimately not too different from other natural/herbal-type remedies, right? Well, not exactly. Do you see that “200C” at the end of the active ingredient? This indicates that the original extract has been diluted in 200 steps, each step consisting of a 1:100 dilution with high agitation. Just how dilute is the final solution?

1:100200 = 1:10400 = 1:10,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000.

Now that’s a lot of zeros. How does one even imagine something this dilute? Imagine taking all of the water on earth (1,385,000,000 cubic kilometers/332,500,000 cubic miles, assuming they corrected for density) and placing it in a sphere out in space. If you put a single protein molecule (e.g. albumin and assuming that protein volume = (1.21 * molecular weight) Å3) into that volume, you would only have a ~22C dilution from pure albumin. If you had enough water to fill the sun, you would still only have a ~26C dilution. In fact, in order to reach the equivalent of 200C with a single molecule of albumin, you would need a volume of water equivalent to 10294 times the volume of the observable universe!!!

Amounts of water required to make homeopathic-scale solutions containing a single molecule of the protein albumin. 10C: 13 M&Ms; 23C: All of Earth's Water; 26C: The Sun (Sol); 44C: The Milky Way; 53C: observable universe; 200C: 10^294 observable universes.

Now take a small drinkable teaspoon-size dose from that whole volume, and you’ve got your homeopathic preparation. What this really means is that oscillococcinum isn’t really anything more than sugar water that at one time was in contact other water, that was at one time in contact with other water (repeat many times), that was at one time in contact with duck liver and heart extract.

Welcome to the world of homeopathic medicine.

The word “homeopathic” is often conflated colloquially with all alternative/herbal/naturalistic medicine, though it actually refers to the specific type of preparation where the original active ingredients are extremely diluted, usually to the point where there aren’t any molecules of the original substance remaining in any given dose of the preparation.

To anyone even passively familiar with the way the world works, and especially to the known laws of chemistry and physics, this sounds ludicrous—how could anything like this be thought to work? Notice, too, that none of this is explained on the product’s website.

Homeopathy is governed by two three main laws:

  • Law of similars – A substance that in large doses causes certain symptoms in healthy people can, in small doses, treat those symptoms in disease states (“like cures like”).
  • Law of infinitesimals – The more dilute a substance is, the more powerful it is.
  • Law of succession – Intense shaking between dilutions (beyond simply mixing thoroughly) makes the preparation more potent.

To understand these seemingly crazy laws, we need to understand their genesis. Homeopathy was defined and developed at the end of the 18th century by German physician Samuel Hahnemann. At this time, it was largely unknown what caused disease and ailments—the germ theory of disease would not be developed until Louis Pasteur and Robert Koch’s work over 50 years later. Vitalism, the belief that living matter was imbued with an animating spirit that separated it from other matter, was still widely accepted. Evidence for atomic theory—the finite divisibility of matter—would just begin to accumulate with John Dalton’s work in the early 1800s.

Cinchona Flowers
Cinchona, a source of the malaria drug quinine. (From USGS)

The idea behind the law of similars was that the disease symptoms were due to an imbalance in a person’s vital force, and that by treating a sick person with small doses of something else that would normally also cause those symptoms, the short-term effect would displace the longer-term disease and allow the vital force to right itself. Perhaps ironically, Hahnemann developed homeopathy after finding that cinchona bark—a source of quinine and actual treatment for malaria—caused symptoms in him similar to malaria’s symptoms. (Of course, now we know that quinine directly kills Plasmodium falciparum and it has nothing to do with causing similar symptoms.) It also helped that homeopathy was relatively safe compared to the standard medical treatments of the day, including bloodletting, which often harmed more people than they cured.

To be honest, I have no idea why the law of infinitesimals or succession would exist—why would only the supposedly positive effects of a substance be ‘potentized’ and not the negative effects, and why would shaking help? Though without the knowledge of the finite divisibility of matter, it isn’t much of a surprise why it was thought you could actually have such small doses.

Modern Homeopathy

Today, we have atomic/molecular theory and know that there are a finite number of particles of any given substance. Vitalism has long ago been discarded as false. And we know now that diseases have underlying causes (e.g. bacteria, viruses, genetics, nutrition, inflammation, etc.) and that treating the symptoms may make people feel better, but it doesn’t fix the problem.

So why does homeopathy still exist? In my opinion, it is a mix of wishful thinking and a mistaken notion that anything natural is good and anything developed in a lab is bad. The term “wishful thinking” may sound somewhat harsh, but the capacity of the human mind for self-delusion is quite high, and ignoring this capacity can be dangerous. This capacity is the very reason why scientists like myself insist on removing any possible bias from a study. In my own experiments in which I have to count cells under a microscope, I scramble the images beforehand so that I don’t know which sample is which, lest my subconscious desire for a certain result influence my decisions.

Yet those who are strong believers in homeopathy discount this system. For instance, Larry Malerba who blogs on Huffington Post, wrote this in response to critics on a post of his:

“Anecdotal­” to me has become a derogatory term used in polarized arguments to silence differing medical points of view. The same can be said when we casually dismiss patient’s experience­s as “subjectiv­e.” To the contrary, anecdotal to me is a specific example that illustrate­s a potentiall­y important point. To me, firsthand clinical experience is a more practical form of informatio­n than the statistica­l analyses and abstractio­ns of research studies, which are several steps removed from the patient sitting in my consulting room.

He bases his belief on his clinical experience, where he really has no idea whether the things he has treated would have gotten better on their own, as most ailments do. A fellow homeopathy advocate, Dana Ullman, has espoused similar views. Research studies, the best-controlled of which show very little, if any, effect of homeopathic remedies over placebo, have failed to uphold their beliefs and so they now reject them.

That doesn’t mean there aren’t those trying to find some basis for homeopathy to work. Although not directly a supporter of homeopathy, Dr. Luc Montagnier, a virologst who discovered HIV and was awarded the Nobel prize for it, has done some odd recent research purporting to show radio signals from DNA at very high, though not homeopathically-high (only ~9C) dilutions, and that DNA ‘teleports’ to other containers based on this. Others suggest that the high agitation of water in between dilutions allows water to strongly retain some sort of molecular memory. Others still suggest that there are “nanocrystals” of the starting material which doesn’t really make sense for a number of reasons. And of course, none of this indicates why such things would have the effects they are claimed to, even if they were true.

I’ll spend a separate post going more into more detail about these proposals and the research that has been done on them. In many ways this resembles the transformation that creationism undertook when it assumed the pseudoscientific drapings of Intelligent Design, and supporters of homeopathy seem to be quite willing to latch on to anything that provides the slightest veneer of support.

Back to Oscillococcinium and Other Homeopathic Remedies

Before I close out this post, there’s a bit more to say on specific homeopathic remedies. When homeopathy was being formulated, Hahnemann and his followers would undertake “provings” for the various remedies, by treating themselves and recording the symptoms that result from taking the remedy—a crude version of a modern clinical trial. Notably, they started using actual doses, but later tried with more dilute mixtures.

Oscillococcinium, as we said, is from duck liver and heart extract, and is supposed to treat flu symptoms. We just discussed the concept of the law of similars, so why would this have been used to treat flu? Well, it seems that in 1925 a French physician thought he had discovered the cause of the Spanish flu in a bacterium, and for some reason thought duck liver and heart was a good source for this bacterium. Like cures like, right?

Too bad that we’ve since found out that the flu is a viral infection and has nothing to do with bacteria.

Since homeopathy’s methods lack rigor and its reasoning borders on magical thinking, it isn’t much of a shock that oscillococcinium would stick around. It also isn’t much of a surprise that other extensions could easily make their way in that in any normal world would seem insane—for instance, making sugar pill forms of homeopathic remedies claiming that the sugar somehow takes up the essence of the water version, or collecting rainwater from a thunderstorm to make a love potion (please read this short write-up, its highly entertaining). Even if we assume for the moment that homeopathy has something to it, that would not automatically validate anything and everything claimed to be a homeopathic remedy.

To illustrate the point that homeopathic remedies are nothing more than sugar and water, the 1023 campaign (1023 is the order of magnitude of Avogadro’s number, the number of discrete particles of a substance in 1 mole) is trying to raise awareness of this fact by holding regular mass ‘overdosing’ on homeopathic remedies and showing that there are no effects or side-effects. But this isn’t all just for fun and games. In the majority of cases, people promoting and using homeopathic treatments isn’t a problem because most ailments tend to resolve on their own. But there are actually those who advocate foregoing medical treatment for homeopathic treatment with AIDS and other diseases, which often means the death of a patient that could have lived.

Hormesis and Allergy Tolerance

Finally, I would be remiss if I did not mention a somewhat related concept called hormesis. Hormesis is the idea that certain poisons may actually provide a benefit at very low (but real) doses. While this is also controversial as far as some of the specific claims that are being made, there does seem to be a basis for the general principle.

For instance, in some cases severe allergies, like peanut allergies, have been gradually reduced by administering very small doses, allowing the body to build up a tolerance to it. If nothing else, it is certainly plausible that small doses of poisons—many of which we would have evolutionarily-speaking had periodic small exposures to—provide a protective effect, perhaps by priming protective stress response pathways. An example of this sort of behavior can be found in the bacterium E. coli, which when exposed to one stress often induces many other types of stress-protection pathways.

Whatever the facts are, the effects of low doses will be as dependent on the particulars of the biological pathways they interact with as are the high doses. It could very well be that hormesis isn’t going to be true of most or even all poisonous substances. But unlike homeopathy, it doesn’t require a complete reconfiguration of the known laws of physics and chemistry for it to work.

Credits for images used in my figure: M&M’s, Earth, the Sun, Milky Way, observable universe, serum albumin
Much of the information used was obtained from Wikipedia and its referenced sources unless otherwise noted.
Update 2/22/11: Added information on the law of succession, which was not explicitly enumerated previously.


  1. Nice post.
    See more on Dana Ullman here


  2. Thanks all. And eLucida, good link. I'll be referring to RationalWiki in the future I'm sure.

  3. British Journal of Clinical Pharmacology (British Pharmacological Society) (1989)

    Cochrane (2003)

  4. Dr. Nancy Malik:
    The first study is not as convincing as you think it is.

    First, there are several oddities. For some reason the treatment was only apparently effective among 12-29 year olds and mild to moderate cases, something which homeopathic theory would not predict.

    Second, the "time trend" curve shows that the potential error between most of the data points for both treatment and placebo overlap, and the curve for "recoveries related to active drug" has 95% confidence intervals that at almost every point include 0. For lay people, a 95% confidence interval means that we are 95% sure that the true value lies within the interval, and in this case, '0' could easily be the true value.

    As for the meta-analysis (your second link), finding a slight decrease in duration is far from a ringing endorsement. If this homeopathic treatment worked, it should be far more successful than that.