One of the occasional arguments used in support of “alternative” approaches to human medicine is the observation that since “alternative” medicine is used (with anecdotal success) in animals, and animals don’t know anything about the treatment that they’re getting, then they must work a priori. Of course, the fallacy of such an observation is pretty obvious to anyone with a logical/skeptical frame of mind, because it assumes that the therapies do work (even though there’s little evidence of that).
Clearly, however, some people perceive that the therapies work, including veterinarians – there are veterinary acupuncturists, chiropractors, homeopaths, etc., etc. Since there’s very little scientific support for the idea that the therapies actually have any clinically significant effect on biological processes, including the processes that result in disease, questions arise as to whether there are other effects of “alternative” treatments on animals. Specifically, people may wonder whether or not animals can benefit from placebo effects.
Fully discussing placebo effects is a topic for another blog (or several other blogs). Regardless, whether or not placebo effects exist in human medicine, there is little evidence that they exist in animals. (1) In general, for a placebo response to occur, it would seen require that the patient being treated recognize that there is an intentional effort to treat. Animals would appear to lack the ability to comprehend such intentions (other than they may not like a particular intervention). As such, animals would not be able to participate in placebo-generating experiences. So, for example, one couldn’t rationally suggest to a dog that a particular therapy might help it get better, or that it was beneficial because it was “natural;” one presumably wouldn’t wax eloquent to a horse that a particular therapy might give it a window of hope for recovery. They just wouldn’t understand.
Still, there are many explanations for how a placebo-like effect might be explained in animals. Take conditioning. Conditioning theory proposes that bodily changes result following exposure to a stimulus that previously produced that change. This is perhaps the most intuitively acceptable explanation for any placebo effects in animals. Indeed, animal studies support such a model for placebo effects, starting with the first descriptions on salivating dogs by Pavlov. (2) Both human and animal studies support the idea that conditioning forms some basis for placebo responses. (3) Since conditioning requires learning, it would be expected that repeated visits to a practitioner (of any particular persuasion) might increase the strength of the association between a learned stimuli and response in animals, good or bad. There are numerous examples of dogs shaking in fear when being taken into a veterinary clinic; on the other hand, a dog that enjoyed being handled in a soothing environment might appear to receive some relief from a chronic condition; as it learned to associate its visits with the comforting handling, conditioning effects could occur. This could certainly serve as reasonable explanations for purported placebo effects in animals. Nevertheless, the hypothesis that a healing or therapeutic effect can be dependably provoked as a result of conditioning cannot be supported at this time by any evidence.
Expectancy theory proposes that bodily changes may occur to the extent that the person receiving the therapy expects them to. There is considerable overlap between expectancy and conditioning, because learning is one of the major ways that expectancies are formed. To the extent that therapies are expected to provide relief from disease, or at least provide the client and/or veterinarian with a feeling of control over the disease process, they may alleviate adverse mental states (in the humans). Certainly, in humans, therapies that help restore patient control may evoke therapeutic effects, at least short term, but studies that investigate the expectancy model in animals have so far not been performed. Still, if animals were able to form an association between treatment-related signals (the attention and handling received, the way that the owner behaves towards the animal when it is receiving treatment) and the relief of its distress, expectancies of treatment effects might develop (on the part of both animal and owner).
There is a good body of research that demonstrates that human contact has measurable effects on animals. For example, petting by humans reduces heart rates in dogs (4) and horses (5) and causes major vascular changes in dogs. (6) Gentle handling increases productivity in dairy heifers (7) and increases reproductive efficiency in sows. (8) Thus, it is plausible that human-animal contact might play an important role in the observed responses to therapeutic interventions. To take an “alternative” example, it has been shown that a single acupuncture treatment is as effective as petting a horse, when it comes to relief of signs of chronic airway disease; that is, there’s no demonstrable effect of acupuncture beyond simple handling. (9) On the other hand, handling may also be stressful to animals, so responses to handling may not necessarily be beneficial. Still, there’s no question that human contact can invoke responses from animals and animals may behave quite differently when they are not being observed; those shouldn’t be confused with placebo-effects, however.
Can Therapies Induce Placebo Effects in Owners?
The reported intensity of subjective symptoms such as pain, fatigue, and depressed mood in an animal may vary over time for all sorts of reasons, not all of which have to do with actual changes in symptom severity. Further complicating such analyses are treatment effects that might exist on the part of both the animal owner, as well as the veterinarian with a personal investment in an “alternative” approach.
Client expectations can be very powerful motivators. Having participated in a therapeutic transaction, clients generally expect to see some results. Optimistic owners may be more likely to diligently pursue treatments. Even failing obvious results, normal reciprocal responses often result in clients reporting improvement, at least initially, even when no improvement has occurred. At the very least, veterinarians can help clients understand what problems are occurring in the animal – such comfort and reassurance may make a problem easier for the client to deal with. That’s a good thing, mostly, unless the veterinarian steers the client into areas that are unsupported by evidence.
Good veterinary care should include a healthy dose of understanding and compassion, and veterinarians should be interested in proven effective care. However, there’s no evidence whatsoever that animals can benefit from, or even experience, placebo effects. Indeed, when doctors claim effectiveness for a treatment beyond the evidence in the belief that they are doing the patient a favor by inducing a “placebo effect” to the animal’s supposed benefit, they are abusing three trusted roles: expert, authority figure, and comforter. Animals deserve better.
- McMillan, FD. The Placebo Effect in Animals. J Am Vet Med Assoc 1999; 215(7): 992-9.
- Pavlov, IP. Conditioned Reflexes. London: Oxford Press, 1927. 23–78.
- Voudouris, NJ, Peck, CL, Coleman, G. Conditioned placebo responses. J Pers Soc Psychol 1985; 48: 47-53.
- Gantt, WH, Newton, JO, Royer, FL, et al. Effect of person. Conditional Reflex 1966; 1: 18-35.
- Lynch, B. Heart rate changes in the horse to human contact. Psychophysiology 1974; 11: 472-478.
- Newton, JF, Ehrlich, WW. Coronary blood flow in dogs: effect of person. Conditional Reflex 1966; 1: 81.
- Gross, WB. The benefits of tender loving care. Int J Stud Anim Prob 1980; 1: 147-149.
- Heinsworth, PH, Brand, A, Willems, PJ. The behavioral response of sows to the presence of human beings and their productivity. Livestock Prod Sci 1981; 8: 67-74.
- Wilson, DV, Berney, CE, Peroni, DL, et al. The effects of a single acupuncture treatment in horses with severe recurrent airway obstruction. Equine Vet J 2004; 36(6): 489-94.