Asking For Evidence: Does Homeopathy Work?

The controversial medical practice of homeopathy has made the news again recently, as the UK’s National Health Service is debating whether to “blacklist” it as a treatment.  Supporters of the practice have pointed to evidence that homeopathic treatments have performed “better than placebo” in some studies, while detractors point to its lack of plausible mechanisms for actually improving a patient’s health.  So what are the facts?

What is homeopathy?

Homeopathic treatments are based on the “like for like” principle, a concept which has its roots in the Ancient Greek tradition of medicine, as outlined by Hippocrates (he of the Hippocratic oath).  The idea resurfaced in the 19th Century as a German physician, Samuel Hahnemann, attempted like for like treatment of malaria using small doses of quinine from tree bark.  By eliciting a malaria-like fever in himself, though much weaker and less damaging, he believed he had discovered a means of treating the condition.  This is superficially similar to how vaccines inoculate us against diseases, and indeed Hahnemann was inspired by the contemporary vaccinations against cowpox.  However, they don’t act against disease in the same way.

This is particularly evident in the relative dose of a vaccine compared to a homeopathic treatment.  Vaccines do contain a small amount of disease or pathogen (though in a weakened form or dead) , and when the body receives the vaccine it can develop antibodies to resist it.  Homeopathic treatments are famous for their extremely low doses.  In fact, most homeopathic substances are diluted so much that not even a single molecule of the substance is expected to remain, i.e. the homeopathic treatment does not actually contain anything other than water (or in the case of pills, sugar).

Homeopathy is billed as a “holistic” treatment.  Homeopathic supplements are accompanied by lengthy interviews with a homeopathic specialist, taking detailed patient histories and discussing all aspects of the patient’s life.

How can homeopathy work if there’s no active ingredient?

Great question.  Homeopathy practitioners point to “the memory of water” as the means by which that ultra-dilute solution is not merely a glass of water.  Essentially, the idea is that water molecules can “remember” the influence of other molecules that are no longer there, and it is this changed water that allows homeopathy to work.

There is a great deal of quantum woo out there attempting to tie this into modern particle physics, but I should be clear – as a physicist, I see no mechanism by which water retains the memory of dissolved substances after they are diluted out of the solution, and I’ve yet to find a physicist who thinks otherwise.  Personally, I’m rather disturbed by the idea that recycled drinking water retains a memory of the waste that once floated in it.

Evidence For Homeopathy

The British Homeopathic Association points to several studies of randomised controlled trials (RCTs) of homeopathic medicine as is appropriate for clinical trials of any medicine, which are double-blind (neither the participant nor the researcher know what group they’ve been assigned to) and use placebos (people in one group are given all the same aspects of the treatment, including pills; while they might contain water, they’ve not been prepared using the homeopathic dilution process).  They report that around:

“41% supported a balance of positive evidence”,

which they compare to their own meta-analysis, or ‘review of reviews’ of conventional medicine studies, which claims for standard medicine,

“44% of the interventions studied were likely to be beneficial (positive)”.

Note the difference in language between each statement.  They link to a PDF summarising evidence both for homeopathy’s effectiveness and for evidence of “memory of water”.  As far as the evidence for homeopathy, most hinges on systematic reviews of homeopathic treatment which deliver either positive or inconclusive outcomes on treatments for a handful of ailments (e.g. hay fever, influenza, childhood diarrhoea).

Evidence Against Homeopathy

There are a large number of systematic reviews of homeopathic treatments which show no positive outcomes from its use as a treatment (for example, this study published in the Lancet).  Also, a systematic review of systematic reviews of homeopathy shows that many pro-homeopathy studies in the review above are not as positive as labelled after critical analysis, and that the effect sizes reported in these studies were over-estimated, most damningly because they suffered from publication bias, i.e. they chose to ignore studies that did not support their desired conclusions.  It’s important to note this review comes from a former homeopathy supporter.  Much of the subsequent “debate” in homeopathy appears to be torturing of statistics by those in favour, in order to discredit those against (see for example this amusing vignette).

The papers referring to “memory of water” are yet to pose a meaningful cause for what are admittedly measurable electrical effects in water.  We should also note that the papers make some fundamental errors in chemistry (for example, that lithium is larger than sodium, which it isn’t).  They also tend to be working at much higher concentrations than the ultra-dilute solutions that homeopathy uses.  In short, they are yet to show convincing evidence of “memory”, and there is significant criticism of their methods for measuring changes to the rearrangement of molecules in water (see for example here and here and here).

This highlights the fallibility of peer-review, especially when a new raft of predatory open-access journals will publish anything for a profit.  This is why it’s important not only to look at peer-reviewed articles, but also the journals they’re published in, to measure how rigorous the peer-review might have been.

In these circumstances, one might be tempted to point to anecdotal evidence of effective homeopathic treatment, but that’s not good enough for researchers to be convinced.  It’s also quite possible that the real power of homeopathic remedy is a placebo effect of a slightly different kind.  The “holistic” approach of allowing patients to discuss in detail with a “specialist” for hours about a myriad of problems (compared to increasingly brief GP interviews) may have extra benefits beyond the standard placebo effect of taking pills.

The Bottom Line

The NHS has concluded:

There is no good-quality evidence that homeopathy is effective as a treatment for any health condition.

This conclusion was supported by the House of Commons Science and Technology Committee in 2010, as well as by the US Department of Health, and most recently by the Australian Government’s National Health and Medical Research Council. Personally, I agree, and tend to support the view of this website.

Ernst, E. (2002). A systematic review of systematic reviews of homeopathy. British Journal of Clinical Pharmacology. DOI: 10.1046/j.1365-2125.2002.01699.x

Shang, A. et al. (2005). Are the clinical effects of homoeopathy placebo effects? Comparative study of placebo-controlled trials of homoeopathy and allopathy. The Lancet.