December 12, 2016

expert reaction to literature review on efficacy of homeopathy in livestock

A literature review published in Veterinary Record looks at the evidence for the efficacy of homeopathy in livestock. They report that there is insufficient evidence for homeopathy as a way to prevent or treat infectious diseases in livestock.

Prof Peter Lees, Emeritus Professor of Veterinary Pharmacology, Royal Veterinary College and Prof Pierre-Louis Toutain, European Veterinary Specialist in Pharmacology and Toxicology, European Board of Veterinary Specialisation, said:

“Overall, there are some positive aspects of this review – in terms of its validity as an assessment of the literature on homeopathic trials conducted between 1981 and 2014 in cattle, pigs and poultry.    We suggest that several of the findings and conclusions should be.  A fundamental conclusion is that, for a range of reasons, the 52 trials reviewed provide no convincing evidence for the use of homeopathic products in these three species.

“It is the most comprehensive review on homeopathy in farm animals yet published and takes us beyond recent reviews of Mathie et al. on use of homeopathic products in animals. Of 48 publications identified, 33 were published in homeopathic journals. The purposes of applying homeopathic remedies were prevention of disease (30) treatment of disease (18) and metaphylaxis (4).”

“It addresses the important question of publication bias. For example, in contrast to homeopathic journals, in journals with a broader focus on veterinary medicine, 12 out of 18 trials found the homeopathic treatment was ineffective (Odds Ratio 0.27, 95 per cent Confidence Interval 0.05 to 1.57).”

“It addresses the issue of lack of reproducibility and prognostic validity in trials of homeopathic products. Conclusions based on single clinical trials (either positive or negative) cannot be conclusive; reproducibility is the classical mertonian principle of science. Not one of the studies reviewed was repeated under comparable conditions.”

“It addresses the issue of lack of prognostic validity. The authors conclude that replacement or reduced usage of antibiotics in favour of homeopathic products is not justified.”

“When homeopathic products are used with intention of prevention of disease (prophylaxis) or treating in contact animals exhibiting signs of disease (metaphylaxis) there is a fundamental contradiction in that there is lacking a complete homeopathic clinical picture and also the matching remedy picture. We can extend this comment to any mass medication  – namely it is against the very principles of homeopathy, which  require individual animal examination and treatment.”

“The review authors emphasise that animal welfare considerations demand that treatments applied should always be the most effective to avoid unnecessary suffering.  Justifying homeopathic products on grounds of concerns with conventional/pharmacological-based treatments is not justified.  Examples of concerns that can be properly managed are those of drug residues (manageable by adopting an appropriate withdrawal time) and antimicrobial resistance (managed by appropriate stewardship).”

Dr Martin Whitehead, Vet at Chipping Norton Veterinary Hospital, said:

“This review is not fully a systematic review in the Evidence Based Medicine (EBM)sense that the experimental design and conduct of the eligible trials were fully analysed in terms of their quality, with those trials having weaker designs/conduct (so allowing bias or other non-specific effects to influence the results) being eliminated.  Thus, this study specifically includes some trials that were unblinded and/or unrandomised and/or did not involve a control group and/or involved a control group that was not placebo treated.

“Furthermore, 33 of 52 (63%) of studies did not define criteria for exclusion of animals from trial participation or for exclusion of animals during the trial. Numerous other features of the eligible trials could also allow bias and other non-specific effects to influence the results.  Therefore, there is substantial potential for effect of bias and other non-specific effects, and so the positive findings cannot be taken as good-quality evidence that homeopathy is effective.

“However, as part of the study, attempts are made to evaluate the effect of trial design and conduct on whether the results were positive or negative, and this includes a not-very-well defined systematic review.  This part of the study does show that better designed trials were less likely to show a positive effect of homeopathy.

“Some of the trials compared the homoeopathically-treated group to placebo-treated or untreated control groups, whereas others compared the homeopathically-treated group to a conventionally-treated group.  Thus, there is certain ‘apples and oranges’ factor to the review!  Some of the 28 positive trials will be ‘false positive’ due to bias.  Some of the trials demonstrating no benefit will have been in comparison to conventional treatment, and may possibly have demonstrated benefit relative to placebo-treated control group.

“Doctoral theses are not peer reviewed.  Homeopathic and other alternative medicine journal publications tend to receive poor-quality peer review (from a scientific viewpoint), thereby tending to give a bias towards publication of ‘false-positive’ studies.  This data is clear evidence of that specific form of publication bias.

“Studies using even the most basic methods, e.g., control groups and blinding, to reduce bias showed less efficacy, suggesting some or all of the positive outcomes are due to bias.

“A key missing factor in this analysis is whether the outcomes reported were specified as ‘primary outcomes’ at the stage of trial design.  There is a huge range of outcomes looked at, and it is possible that many of those reported were not initially specified, but where just factors found to be significantly different at the end of the trial when the data was analysed.  In terms of statistics, such post-hoc findings are not valid evidence of a genuine effect (but they can be viewed as useful for generating hypotheses for further study).

“As noted above, studies with ‘low’ risk of bias may not have used some of the other design features required to fully exclude bias.  That is why many of these studies were excluded in the systematic review of RCTs of veterinary homeopathy conducted by Mathie & Clausen (2014).

“On three of the four occasions that odds ratios are presented in Doehring & Sundrum (2016) they appear to be presented both positively and negatively, as if those are two different things to compare, when in fact a positive and a negative odds ratio are the same information.  This observation raises doubts about whether this paper was reviewed (for Veterinary Record) by an epidemiologist or statistician – who would have spotted this error!”

Danny Chambers, Locum Veterinary Surgeon, said:

“It is important that the livestock industry continues to take measures to reduce the amount and inappropriate use of antibiotics in farmed animals, but this clearly cannot be achieved by substituting antibiotics with homeopathy.  In cases where homeopathy appears to have been effective, simply giving no pharmaceutical treatment would have achieved the same result.

“This review demonstrates that the overall evidence is negative for homeopathy but not totally conclusive.  It is therefore vital that trials showing a positive effect are replicated under more rigorous standards to demonstrate a genuine effect.   However, not a single positive trial of homeopathy in livestock has been replicated.  Hence Doehring & Sundrum’s conclusion that “replacing or reducing antibiotics with homeopathy currently cannot be recommended.” ”

 Prof Tim Morris, Professor of Laboratory Animal Welfare and Science at the School of Veterinary Medicine and Science, University of Nottingham, said:

“This study asks important questions by systematically evaluating the scientific literature as to the evidence that homeopathy can have a role in assuring animal welfare as an effective veterinary treatment and in reducing antibiotic use to address resistance.

“Objective assessment of use of homeopathy is clouded by entrenched positions, and the mixed findings of this systemic review will no doubt be used to reinforce claims on both side of this debate. However, the most useful conclusion of this review is to highlight absence of any repeated studies under comparable conditions.

“If advocates of homeopathy wish to move from case by case claims to claim its general efficacy under a range of conditions, and that it has a general role in assuring animal welfare, a number of well designed and repeated studies in farm animals which show such general applicability and satisfactory outcomes are required.”

* ‘Efficacy of homeopathy in livestock according to peer-reviewed publications from 1981 to 2014’ by C. Doehring et al. will be published in Veterinary Record  at 23:30 UK time on Monday 12th December, which is also when the embargo will lift. 

Declared interests

Prof Peter Lees:

Self-employed consultant on pharmacological/toxicological aspects of conventional veterinary products.

Grants from e.g. Defra, BBSRC and some with pharmaceutical company support.

Served on RCVS  Council and its Disciplinary Committee from 2000 to 2014.

Memberships of several professional bodies e.g. European Association for Veterinary Pharmacology and Toxicology.

Served for 20 years on Veterinary Products Committee (1978- 1998) and Scientific Committee of Home of Rest for Horses (1990- 2000).

Prof Pierre-Louis Toutain: No conflicts of interest.

Dr Martin Whitehead: I am a founder member of the Campaign for Rational Veterinary Medicine:  https://www.vetsurgeon.org/microsites/private/rational-medicine/

Danny Chambers: No conflicts of interest

Prof Tim Morris: No declared interests

None others received.