We all know that so-called alternative medicine (SCAM) is currently a popular option for treating all sorts of human diseases. Therefore, it can hardly come as a surprise that a similar situation applies to animals. Consumers who are fond of SCAM are also likely to take their pets to SCAM practitioners when they feel that they are ill.
But SCAM for animals can be a rather tricky issue. Animals cannot consent to the treatments they are given. This can render the use of SCAM for animals ethically problematic: Where does the treatment end and the animal abuse begin? Which SCAM works for animals and which doesn’t? Which are safe and which harmful?
A recent systematic review assessed the evidence for the clinical efficacy of 24 SCAMs used in cats, dogs, and horses. The authors performed literature searches on Web of Science Core Collection, CABI, and PubMed. Relevant articles were assessed for scientific quality, and information was extracted on study characteristics, species, type of treatment, indication, and treatment effects.
Of 982 unique publications screened, 42 were eligible for inclusion, representing 9 different SCAMs:
- aromatherapy,
- gold therapy,
- homeopathy,
- leeches (hirudotherapy),
- mesotherapy,
- mud therapy,
- neural therapy,
- sound (music) therapy,
- vibration therapy.
For 15 other predefined SCAMs, no study was identified. The risk of bias was assessed as high in seventeen studies, moderate to high in ten, moderate in ten, low to moderate in four, and low in just one single study. In those studies where the risk of bias was low to moderate, there was considerable heterogeneity in reported treatment effects.
The authors concluded that
the present systematic review has revealed significant gaps in scientific knowledge regarding the effects of a number of “miscellaneous” SCAM methods used in cats, dogs, and horses. For the majority of the therapies, no relevant scientific articles were retrieved. For nine therapies, some research documentation was available. However, due to small sample sizes, a lack of control groups, and other methodological limitations, few articles with a low risk of bias were identified. Where beneficial results were reported, they were not replicated in other independent studies. Many of the articles were in the lower levels of the evidence pyramid, emphasising the need for more high-quality research using precise methodologies to evaluate the potential therapeutic effects of these therapies. Of the publications that met the inclusion criteria, the majority did not have any scientific documentation of sufficient quality to draw any conclusion regarding their effect. Several of our observations may be translated into lessons on how to improve the scientific support for SCAM therapies. Crucial efforts include (a) a focus on the evaluation of therapies with an explanatory model for a mechanism of action accepted by the scientific community at large, (b) the use of appropriate control animals and treatments, preferably in randomized controlled trials, (c) high-quality observational studies with emphasis on control for confounding factors, (d) sufficient statistical power; to achieve this, large-scale multicenter trials may be needed, (e) blinded evaluations, and (f) replication studies of therapies that have shown promising results in single studies.
Of particular interest was, I think, what the authors revealed in relation to homeopathy. The included studies, with moderate risk of bias, such as homeopathic hypotensive treatment in dogs with early, stage two heart failure and the study on cats with hyperthyroidism, showed no differences between treated and non-treated animals. An RCT with osteoarthritic dogs showed a difference in three of the six variables (veterinary-assessed mobility, two force plate variables, an owner-assessed chronic pain index, and pain and movement visually analogous scales).
Such findings are disappointing and should make SCAM-loving pet owners think again. Crucially, they are supported by another systematic review of 18 RCTs of homeopathy, representing 4 species (including two dog studies) and 11 different indications.
The authors excluded generalized conclusions about the effect of certain homeopathic remedies or the effect of individualised homeopathy on a given medical condition in animals. A meta-analysis of 9 trials with a high risk of bias, and two studies with a lower risk of bias, concluded that there is very limited evidence that clinical intervention in animals using homeopathic remedies can be distinguished from similar placebo interventions.
In essence, these data confirm what I have been pointing out repeatedly, for instance on my blog: SCAM for animals has one important feature in common with SCAM for humans – it is not evidence-based.
