Homeopathy for Low Back Pain? This clinical trial made me laugh out loud

If you think that scanning through dozens of new scientific articles every week is a dry and often somewhat tedious exercise, you are probably correct. But every now and then, this task is turned into prime entertainment by some pseudoscientists trying to pretend to be scientists. Take, for instance, the latest homeopathy study by Indian researchers with no less than 9 seemingly impressive affiliations:

• ¹Department of Organon of Medicine and Homoeopathic Philosophy, National Institute of Homoeopathy, Ministry of AYUSH, Govt. of India, Salt Lake, Kolkata, West Bengal, India.
• ²Department of Organon of Medicine and Homoeopathic Philosophy, National Institute of Homoeopathy, Ministry of AYUSH, Govt. of India, Block GE, Sector III, Salt Lake, Kolkata, West Bengal, India.
• ³Department of Homoeopathy, State Homoeopathic Dispensary, Karaila, Pratapgarh, Uttar Pradesh, India.
• ⁴Department of Homoeopathy, State Homoeopathic Dispensary, Tulsipur, Shrawasti, Uttar Pradesh, India.
• ⁵Department of Materia Medica, National Institute of Homoeopathy, Ministry of AYUSH, Govt. of India, Salt Lake, Kolkata, West Bengal, India.
• ⁶State Homoeopathic Dispensary, Mangalbari Rural Hospital, Matiali Block, Jalpaiguri, West Bengal, under Department of Health & Family Welfare, Govt. of West Bengal, India.
• ⁷Department of Repertory, The Calcutta Homoeopathic Medical College and Hospital, Govt. of West Bengal, Kolkata, West Bengal, India.
• ⁸Department of Homoeopathy, East Bishnupur State Homoeopathic Dispensary, Chandi Daulatabad Block Primary Health Centre, Village and Post Office: Gouripur (South), Police Station Bishnupur, West Bengal, under Department of Health & Family Welfare, Govt. of West Bengal, India.
• ⁹Department of Repertory, D. N. De Homoeopathic Medical College and Hospital, Govt. of West Bengal, Tangra, Kolkata, West Bengal, India.

Now that I have whetted your appetite, here is their study:

Lumbar spondylosis (LS) is a degenerative disorder of the lumbar spine. Despite substantial research efforts, no gold-standard treatment for LS has been identified. The efficacy of individualized homeopathic medicines (IHMs) in lumbar spondylosis (LS) is unknown. In this double-blind, randomized, placebo-controlled trial, the efficacy of IHMs was compared with identical-looking placebos in the treatment of low back pain associated with LS. It was conducted at the National Institute of Homoeopathy, West Bengal, India.

Patients were randomized to receive IHMs or placebos; standardized concomitant care was administered in both groups. The Oswestry low back pain and disability questionnaire (ODQ) was used as the primary outcome measure; the Roland-Morris questionnaire (RMQ) and the short form of the McGill pain questionnaire (SF-MPQ) served as secondary outcome measures. They were measured at baseline and every month for 3 months. Intention-to-treat analyses (ITT) were used to detect any inter-group differences using two-way repeated measures analysis of variance models overall and by unpaired t-tests at different time points.

Enrolment was stopped prematurely because of time restrictions; 55 patients had been randomized (verum: 28; control: 27); 49 could be analyzed by ITT (verum: 26; control: 23).

The results are as follows:

• Inter-group differences in ODQ (F _{1, 47} = 0.001, p = 0.977), RMQ (F _{1, 47} = 0.190, p = 0.665) and SF-MPQ total score (F _{1, 47} = 3.183, p = 0.081) at 3 months were not statistically significant.
• SF-MPQ total score after 2 months (p = 0.030) revealed an inter-group statistical significance, favoring IHMs against placebos.
• Some of the SF-MPQ sub-scales at different time points were also statistically significant: e.g., the SF-MPQ average pain score after 2 months (p = 0.002) and 3 months (p = 0.007).
• Rhus Toxicodendron, Sulphur, and Pulsatilla nigricans were the most frequently indicated medicines.

The authors concluded that owing to failure in detecting a statistically significant effect for the primary outcome and in recruiting a sufficient number of participants, our trial remained inconclusive.

Now that I (and hopefully you too) have recovered from laughing out loud, let me point out why this paper had me in stitches:

• The trial was aborted not because of a “time limit” but because of slow recruitment, I presume. The question is why were not more patients volunteering? Low back pain with LS is extremely common. Could it be that patients know only too well that homeopathy does not help with low back pain?
• If a trial gets aborted because of very low patient numbers, it is probably best not to publish it or at least not to evaluate its results at all.
• If the researchers insist on publishing it, their paper should focus on the reason why it did not succeed so that others can learn from their experience by avoiding their mistakes.
• However, once the researchers do run statistical tests, they should be honest and conclude clearly that, because the primary outcome measure showed no inter-group difference, the study failed to demonstrate that the treatment is effective.
• The trial did not “remain inconclusive”; it was squarely negative.
• The editor of the journal HOMEOPATHY should know better than to publish such nonsense.

A final thought: is it perhaps the ultimate proof of homeopathy’s ‘like cures like’ assumption to use sound science (i.e. an RCT), submit it to the homeopathic process of endless dilutions and succussions, and – BINGO – generate utter nonsense?