The one NHS cut we should support

Homoeopathic medicines do nothing that a placebo does not do. This is because they contain no active ingredient…like a placebo. But the NHS spends our money on them.

The “rationale” behind homoeopathy is that “like cures like”. This idea had been around at least since the time of Hippocrates (about 400 BC) but was formulated as the basis of a “natural” system of medicine by Samuel Hahnemann, a doctor in Germany in the 1780s. He rightly objected to medical practices of the time, such as bloodletting, which did more harm than good, and soon gave up his practice, fearing being “a murderer or malefactor towards…[his] fellow human beings.”

Hahnemann continued to ponder medicine’s mistakes. He observed the actions of several drugs, such as quinine, on himself and got the impression that they caused symptoms similar to those of the diseases they cured. He developed the notion, to him a law, that whatever causes symptoms of a disease in a healthy person can cure that disease in a sick one.

This is a type of magical thinking (“sympathetic magic”) with not a shred of evidence to support it and would have actually been quite dangerous if applied without modification. Sensibly, he decided to dilute these harmful drugs but developed the even more magical notion that the drugs would be more effective if diluted in a special way. This “potentisation” involved dilution and succussion (vigorous shaking to you and me) by which, somehow, the water maintained a “memory” of the active ingredient. There was unlikely to be any molecule of the drug present: the commonly-used dilution of 30C (30 times a 1 in 100 dilution) contains 1 part in 10*60 of the original drug. This is equivalent to one molecule in a volume of water 20 times the size of the Sun!

But perhaps homoeopathy does work, despite its apparent implausibility. What’s the evidence? This is a job for systematic reviewers of randomised controlled trials (RCTs). An RCT matches a treatment against a placebo (that is, nothing, but supplied in an identical form) in such a way that participants do not know which they are getting and, crucially, neither do the health care staff. Other people analyse the results and use statistics to see whether the treatment is better than nothing to a high degree of certainty, meaning that the result is very unlikely to have been caused by chance. A systematic review1 is an unbiased synthesis of all RCTs of a particular treatment. Such reviews are fundamental to evidence-based medicine. What do such reviews tell us about homoeopathy?

The Cochrane Collaboration2 lists nine systematic reviews of homoeopathic treatments for various conditions, all but two of which showed no evidence of a positive effect. One review showed evidence of benefit in irritable bowel syndrome from just two rather inadequate trials. Another showed benefit in treating adverse effects of cancer therapy: it examined eight trials of which four showed no benefit and four showed benefit (only two of these were good quality, with little risk of bias). This was out of a total of 33 RCTs analysed in the nine reviews – rather underwhelming!

You get a different picture from the website of the Faculty of Homeopathy (FoH).3 It refers to 42 systematic reviews of RCTs, of which only 15 (36%) are said to indicate benefits. Six of these reviews are of homeopathic treatments in general. Since two of these analyse the same data, there are really only five reviews of this kind. A closer look at one of these five (Cucherat et al., 2000)4 gives a more nuanced picture: the authors say that the 16 RCTs analysed indicated a highly significant benefit of homoeopathy but, excluding RCTs of poor quality, the benefit disappeared. They conclude that there is weak evidence of benefit, with better designed RCTs more likely to be negative. The other five systematic reviews of this type5 listed on the FoH website give similarly lukewarm conclusions – and these are the ones they are proud of!

The nine other reviews are of homoeopathy for single conditions, some of which offer some evidence of vague positive effects (see footnote for details).5 It admits that the other 27 studies of the 42 initially referred to were inconclusive or negative.

The British Homeopathic Association proudly cites the most recent systematic review (one of the six mentioned previously),6 partly carried out by its own researchers, which shows a benefit of homoeopathic treatments. But this review includes 32 RCTs with either a high (20) or uncertain (12) risk of bias (three of these were said to have minor uncertainties and their data was classed as “reliable” by the study authors). Only 22 could be analysed, with an overall finding of a benefit from homoeopathic treatments. However, the authors warn that caution is needed in interpreting this finding because of the low or unclear quality of the evidence. This caution was wholly lacking when prominent homoeopath (and Physician to HM The Queen) Dr Peter Fisher was interviewed on BBC Radio 4’s Today Programme on 13 November. He triumphantly cited this study but failed to mention the low quality of the trials.7

So even the evidence quoted by homoeopathic organisations gives only weak support to their claims, while there is a wealth of evidence showing that homoeopathic treatments are indistinguishable from placebos. But does homoeopathy do any harm? In earlier times, bloodletting made patients weaker and can hardly have improved their chances. It is frequently quoted (by homoeopaths) that cholera patients in the Royal Homoeopathic Hospital in the 1840s survived better than those in the nearby Middlesex Hospital. This is because the latter were “treated” with bloodletting and opium, not because the former could cure cholera. When the alternative is a harmful treatment, simply looking after the patient and giving them pills or potions with no active ingredient is better. Nowadays, there are many minor illnesses which conventional medicine cannot treat, such as colds and ‘flu. Patients tend to get better without treatment so homoeopathic remedies will do no harm.

But there are serious diseases, such as malaria, yellow fever, dysentery, typhoid,8 and even cancer,9 as well as vaccinations, where some overseas homoeopaths claim efficacy for their treatments. Despite celebrity endorsements, these claims lack evidence and it would be extremely dangerous to reject conventional medicine for homoeopathy in these cases. It is reassuring to learn that the FoH “denounces the use of homeopathy as an alternative in the management of the Ebola outbreak,” though they see a place for it in conjunction with conventional medicine. However, you can buy homoeopathic medicines on the internet for such conditions as: diabetes (reduces your need for insulin); tumours (stops them returning after surgery); heart conditions (including valve disease, irregular heartbeat and heart attack); kidney disease (protein in the urine); anti-viral drops (prevents swine flu; epilepsy; brain haemorrhage; and dwarfism!10

Paradoxically, official homoeopathy in the UK (BHA, FoH, Dr Peter Fisher) advises following conventional medical advice on malaria and vaccination – there is no hint of “like cures like.” But some practitioners offer their own homoeopathic “vaccines.” You can get these for your pets in Britain. In North America, Health Canada has licensed homoeopathic “vaccines” for influenza, poliomyelitis, measles and whooping cough though it states that these are not alternatives to vaccination (one wonders what they are, then). However, the US Food and Drug Agency comes down quite hard on unfounded claims for treatments. The BHA proposes working with conventional medicine for treating prostate and breast cancers but claims that its methods can prolong life in the latter “by months or even years.”

Homoeopathy gets very positive ratings from its patients (71% reported positive health changes in a survey at the Bristol Homeopathic Hospital) and here the words of another alternative medicine practitioner are relevant: “The fact that patients swear by us does not mean we are actually helping them. Satisfaction is not the same thing as effectiveness.”11 Homoeopathy costs the NHS in Britain a bit over £5 million per year. It’s only a small proportion of the total budget but…£5 million is £5 million and it could be used for something worthwhile.

Notes and references:


2 The Cochrane Collaboration is the most respected organisation carrying out such reviews: its findings are used to inform health service policies world-wide. Here is a summary of its nine homoeopathy reviews: Asthma, Dementia, ADHD, Menopausal symptoms in breast cancer patients, Induction of labour, Molluscum infection (skin virus), Influenza prevention or treatment: little or no evidence of effectiveness; Irritable Bowel Syndrome: possible benefit shown in two small old low quality trials; Adverse effects of cancer treatments: four of eight studies showed benefit (two poor quality).


4 Cucherat et al. Evidence of clinical efficacy of homeopathy – A meta-analysis of clinical trials. Eur J Clin Pharmacol 2000; 56: 27–33.

5 All from FoH website (see reference 3).

Kleijnen (1991) 105 RCTs: “the evidence… is positive but not sufficient to draw definitive conclusions because most trials are of low methodological quality.”

Linde (1997) 89 RCTs: “clinical effects … [not] completely due to placebo. However, …insufficient evidence…that homeopathy is clearly efficacious for any single clinical condition.”

Linde (1999) Same data as 1997 but excluding low quality RCTs. “…studies with better methodological quality tended to yield less positive results.”

Linde and Melchart (1998) Individualised treatments (the “gold standard” for homoeopathy): 19 RCTs: “overall …superior to placebo” but better quality trials have less positive results. “The evidence… is, therefore, not fully convincing.” [my emphasis]

Mathie (2014) See discussion in text.

Barnes (1997) Recovery from gut surgery: “apparently positive overall result” but drawbacks in studies “preclude a firm conclusion.”

Taylor (2000) Hayfever: 4 RCTs: air flow through the nose is improved. Curiously, the participants don’t feel different from those taking placebo. Review queried since placebo response of 3% is surprisingly low.

Jacobs (2003) Childhood diarrhoea: 3 RCTs: duration reduced from 4.1 to 3.3 days. Homoeopathy recommended in addition to conventional treatment.

Linde (2003) Data from 6 RCTs already analysed in Linde (1997) for three painful conditions. Was this an example of ‘cherry-picking’? This small number of different studies gave a positive result, though two were not significant (i.e. inconclusive).

Several of the other 5 studies have been criticised for poor methods. See Bewley BMJ 2011; 343 doi: “If evidence of efficacy as poor as this was offered for registration of a new drug, it would not succeed.”

6 Mathie et al. Randomised placebo-controlled trials of individualised homeopathic treatment: systematic review and meta-analysis. Systematic Reviews 2014, 3:142.

7 To be fair, there are plenty of poor quality studies of medical drugs funded by drug companies which tend to show positive results where better designed studies don’t. (See Bad Pharma by Ben Goldacre).


9 Frenkel M et al.(2010). Cytotoxic effects of ultra-diluted remedies on breast cancer cells. Int J Oncol. 2010 Feb;36(2):395-403. This “study” took place in a dish, not in humans. It contained not one statistic on the significance of its findings so the best we can say is that we don’t know if homoeopathic treatment has any effect on cancer cells in culture, still less in actual people!


11 Chiropractor Preston Long in

You might also like: Xmas present suggestion! Trained as a homoeopathic doctor, Ernst realised that it was ineffective and has worked hard to ensure that groundless claims for complementary and alternative medicine are challenged. He was shamefully made to resign from his department at Exeter University after an unfounded accusation by the secretary of homoeopathy supporting Prince Charles. (

Homeopathic A&E (That MItchell and Webb Look):


Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s