Whenever I give a public lecture about homeopathy, I explain what it is, briefly go in to its history, explain what its assumptions are, and what the evidence tells us about its efficacy and safety. When I am finished, there usually is a discussion with the audience. This is the part I like best; in fact, it is the main reason why I made the effort to do the lecture in the first place.

The questions vary, of course, but you can bet your last shirt that someone asks: “We know it works for animals; animals cannot experience a placebo-response, and therefore your claim that homeopathy relies on nothing but the placebo-effect must be wrong!” At this stage I often despair a little, I must admit. Not because the question is too daft, but because I did address it during my lecture. Thus I feel that I have failed to get the right message across – I despair with my obviously poor skills of giving an informative lecture!

Yet I need to answer the above question, of course. So I reiterate that the perceived effectiveness of homeopathy relies not just on the placebo-effect but also on phenomena such as regression towards the mean, natural history of the condition etc. I also usually mention that it is erroneous to assume that animals cannot benefit from placebo-effects; they can be conditioned, and pets can react to the expectations of their owners.

Finally, I need to mention the veterinary clinical evidence which – just like in the case of human patients – fails to show that homeopathic remedies are better than placebos for treating animals. Until recently, this was not an easy task because no systematic review of randomised placebo-controlled trials (RCTs) of veterinary homeopathy was available. Now, I am happy to announce, this situation has changed.

Using Cochrane methods, a brand-new review aimed to assess risk of bias and to quantify the effect size of homeopathic interventions compared with placebo for each eligible peer-reviewed trial. Judgement in 7 assessment domains enabled a trial’s risk of bias to be designated as low, unclear or high. A trial was judged to comprise reliable evidence, if its risk of bias was low or was unclear in specified domains. A trial was considered to be free of vested interest, if it was not funded by a homeopathic pharmacy.

The 18 RCTs found by the researchers were disparate in nature, representing 4 species and 11 different medical conditions. Reliable evidence, free from vested interest, was identified in only two trials:

  1. homeopathic Coli had a prophylactic effect on porcine diarrhoea (odds ratio 3.89, 95 per cent confidence interval [CI], 1.19 to 12.68, P=0.02);
  2. individualised homeopathic treatment did not have a more beneficial effect on bovine mastitis than placebo intervention (standardised mean difference -0.31, 95 per cent CI, -0.97 to 0.34, P=0.35).

The authors conclusions are clear: Mixed findings from the only two placebo-controlled RCTs that had suitably reliable evidence precluded generalisable conclusions about the efficacy of any particular homeopathic medicine or the impact of individualised homeopathic intervention on any given medical condition in animals.

My task when lecturing about homeopathy has thus become a great deal easier. But homeopathy-fans are not best pleased with this new article, I guess. They will try to claim that it was a biased piece of research conducted, most likely, by notorious anti-homeopaths who cannot be trusted. So who are the authors of this new publication?

They are RT Mathie from the British Homeopathic Association and J Clausen from one of Germany’s most pro-homeopathic institution, the ‘Karl und Veronica Carstens-Stiftung’.


One alternative therapy that I have so far almost entirely neglected is Ayurveda. It is said to be one of the fastest growing system within this sector. Ayurvedic healing includes herbs, nutrition, panchakarma cleansing, acupressure massage, Yoga, Sanskrit, and Jyotish (Vedic astrology). The website of the ‘Choppra Center’ explains: Recognizing that human beings are part of nature, Ayurveda describes three fundamental energies that govern our inner and outer environments: movement, transformation, and structure. Known in Sanskrit as Vata (Wind), Pitta (Fire), and Kapha (Earth), these primary forces are responsible for the characteristics of our mind and body. Each of us has a unique proportion of these three forces that shapes our nature. If Vata is dominant in our system, we tend to be thin, light, enthusiastic, energetic, and changeable. If Pitta predominates in our nature, we tend to be intense, intelligent, and goal-oriented and we have a strong appetite for life. When Kapha prevails, we tend to be easy-going, methodical, and nurturing. Although each of us has all three forces, most people have one or two elements that predominate.

However, the evidence for its effectiveness is not overwhelming. In 2007, we published a systematic review of Ayurvedic treatments for rheumatoid arthritis (RA). Seven studies met our inclusion criteria. Trials tested either Ayurvedic medicine against placebo or other Ayurvedic medicines. Of 3 placebo-controlled RCTs, one high-quality trial did not show benefit of the active treatment against placebo, while another incompletely reported study indicated beneficial effects of an Ayurvedic medicine. A further incompletely reported study showed no significant difference. The remaining 4 trials were difficult to interpret because they tested an Ayurvedic medicine against other Ayurvedic medicines whose effects were not proven. We concluded that there is a paucity of RCTs of Ayurvedic medicines for RA. The existing RCTs fail to show convincingly that such treatments are effective therapeutic options for RA.

Because of this paucity of reliable evidence, any new assessments are welcome.

The aim of this article was to review and meta-analyze the effectiveness and safety of different Ayurvedic interventions in patients with osteoarthritis (OA). 138 electronic databases were searched through August 2013. Randomized controlled trials, randomized crossover studies, cluster-randomized trials, and non-randomized controlled clinical trials were eligible. Adults with pre-diagnosed OA were included as participants.

Interventions were included as Ayurvedic, if they were explicitly labeled as such. The main outcome measures were pain, physical function, and global improvement. Risk of bias was assessed using the Cochrane risk of bias tool.

19 randomized and 14 non-randomized controlled trials on 12 different drugs and 3 non-pharmaceutical interventions with a total of 2,952 patients were included. For the compound preparation, Rumalaya, large and apparently unbiased effects beyond placebo were found for pain (standardized mean difference [SMD] -3.73; 95 % confidence interval [CI] -4.97, -2.50; P < 0.01) and global improvement (risk ratio 12.20; 95 % CI 5.83, 25.54; P < 0.01).

There was also some evidence that effects of the herbal compound preparation Shunti-Guduchi are comparable to those of glucosamine for pain (SMD 0.08; 95 % CI -0.20, 0.36; P = 0.56) and function (SMD 0.15; 95 % CI -0.12, 0.36; P = 0.41).

Based on single trials, positive effects were found for the compound preparations RA-11, Reosto, and Siriraj Wattana. For Boswellia serrata, Lepidium Sativum, a Boswellia serrata containing multicomponent formulation and the compounds Nirgundi Taila, Panchatikta Ghrita Guggulu, and Rhumayog, and for non-pharmacological interventions like Ayurvedic massage, steam therapy, and enema, no evidence for significant effects against potential methodological bias was found.

No severe adverse events were observed in any of the trials.

The authors concluded that the drugs Rumalaya and Shunti-Guduchi seem to be safe and effective drugs for treatment of OA-patients, based on these data. However, several limitations relate to clinical research on Ayurveda. Well-planned, well-conducted and well-published trials are warranted to improve the evidence for Ayurvedic interventions.

I am, of course, pleased that other too have noticed the paucity of good evidence and recommend more and better research into this area. There are, however, several things that worry me about this systematic review:

  • How can there be a total absence of adverse effects? Even placebos would generate some.
  • The conclusion that Rumalaya and Shunti-Guduchi are safe does not seem justified on the basis of just a few trials.
  • My own review found quite encouraging effects for Boswellia serrate.
  • 138 electronic databases? I did not even know that so many existed!
  • I am also concerned by the way the treatments found to be ‘safe and effective’ are being promoted on the internet:

Rumalaya is a phytopharmaceutical formulation that relieves joint and bone ache associated with various orthopedic ailments. Its natural ingredients possess potent anti-inflammatory properties that alleviate pain. As an immunomodulator, Rumalaya modulates both the humoral and cell-mediated immune response to aches and pain. The medicine has strong anti-arthritic properties that work to combat arthritis.


  • Rheumatic arthritis
  • Rheumatoid arthritis
  • Cervical and lumbar spondylosis
  • Osteoarthritis
  • Arthralgia
  • Gout
  • Frozen shoulder
  • Traumatic inflammatory conditions like fibrositis, bursitis, synovitis, capsulitis, tenosynovitis, myositis and sciatica.

I fail to see good evidence to support most of these claims.

Lastly, I find that the authors fail to warn the public in sufficiently strong terms of some of the drawbacks of Ayurvedic medicines. Many of them seem not to be safe. One of several problems is that they have been shown to be often contaminated/adulterated with toxic substances such as heavy metals.

My conclusion about the value of Ayurvedic medicines is therefore not so optimistic: EFFICACY IS USUALLY MORE THAN DOUBTFUL, WHILE RISKS ARE WELL-DOCUMENTED.

When we consult a health care provider because we feel unwell, the first step invariably is to arrive at a diagnosis. Sometimes this is fairly obvious but often it is not. Typically the health care provider will do a few tests to aid the process. This may involve doing some physical examinations, or taking a blood sample, or ordering some high tech investigation, like a scan, for instance.

All these tests have to fulfil certain criteria to be valid. The most basic of these is that repeated tests should produce roughly the same result. If not, the test is not reproducible, i.e. it is not better than pure guess-work.

Alternative practitioners often use diagnostic tests that are unknown in conventional medicine. But this fact does not mean that these test do not need to be as valid as any other test used in medicine. If the alternative tests are not reproducible, the diagnosis of the alternative practitioner is likely to be pure invention. And if the diagnosis is just a figment of imagination, the treatment aimed at curing it will be nonsense as well.

So, how reliable are those tests used by alternative practitioners? Amazingly, this fundamental question has attracted very little research. I have repeatedly investigated this area and found that, generally speaking, alternative diagnostic techniques are bogus. And because there is so little research, every new trial of alternative diagnostic methods is important.

A brand-new study assessed the inter-rater reliability of Ayurvedic pulse (nadi), tongue (jivha), and body constitution (prakriti) assessments. Fifteen registered Ayurvedic practitioners with 3-15 years of experience independently examined 20 healthy subjects. Subjects completed self-assessment questionnaires and software analyses for prakriti assessment. Weighted kappa statistics for all 105 pairs of practitioners were computed for the pulse, tongue, and prakriti data sets.

These pairwise kappas ranged from poor to slight, slight to fair, and fair to moderate for pulse, tongue, and prakriti assessments respectively. The average pairwise kappas for pulse, tongue, and prakriti were 0.07, 0.17, and 0.28, respectively. For each data set and pair of practitioners, the null hypothesis of random rating was rejected for just 12 pairs of practitioners for prakriti, one pair of practitioner for pulse examination, and no pairs of practitioner for tongue assessment.

The authors of this investigation conclude that  the results demonstrate a low level of reliability for all types of assessment made by doctors.

This is worrying, I think. It is comparable to a situation where you go to see your GP and he measures your blood pressure, or weight, or cholesterol, or any other parameter with a test that produces a different result each time someone tries to repeat it. Your blood pressure could be 160/90 when measured with this fictionally unreliable test and 110/ 60 when repeated two minutes later by the practice nurse, for instance. Any treatment based on such random numbers would be idiotic…and so, it seems, is any Ayurvedic treatment based on the pulse (nadi), tongue (jivha), and body constitution (prakriti).

Recent Comments

Note that comments can now be edited for up to five minutes after they are first submitted.

Click here for a comprehensive list of recent comments.