I’ve written about placebo effects in animals several times. These appear in both clinical research studies and in real-world medical practice in many different ways. However, one of the most significant phenomena that fools us into thinking useless therapies are working in our pets is the caregiver placebo effect. This is when people, including owners, vets, and researchers, see what they hope and expect to see rather than what is really happening to the animals in their care. Since our pets can’t speak for themselves, we interpret their actions in deciding whether they feel better or worse, and we fool ourselves easily and often.
The classic illustration of the caregiver placebo effect in dogs comes for a study of arthritis medication. When given a placebo, a fake treatment that does nothing at all, about half of owners, and nearly as many vets, saw significant improvement in subjective measures of pain and function. Objective measures, such as the amount of weight a dog puts on an arthritic leg, did not show improvement with placebo because such measures are harder to fool than our unaided observations.
This phenomenon is known from studies of caregivers for human patients as well as for dogs, and last year another paper appeared showing us that the exact same phenomenon affects owners and vets when evaluating pain in cats.
Gruen, M. E., Dorman, D. C., & Lascelles, B. D. X. (2017). Caregiver Placebo Effect in Analgesic Clinical Trials for Painful Cats with Naturally-Occurring Degenerative Joint Disease. The Veterinary Record, 180(19), 473. http://doi.org/10.1136/vr.104168
This review looked at six published clinical trials of treatments for arthritis in cats. They assessed the apparent improvement reported in cats getting a placebo on both subjective and objective measures. What they found was that 50-70% of cats getting placebo (meaning fake, useless treatment) were classified as treatment successes on subjective client measures. This means that a majority of clients truly felt their cats were better just because they mistakenly believed they were given an effective treatment. However, only 10-50% of objective measures showed an improvement with placebo. (the reasons cats may have improved even while getting no treatment at all are other aspects of placebo effects, which I have discussed elsewhere.)
This finding reinforces, yet again, that we cannot trust anecdotes or our own observations to determine if treatments work or not. We need placebo controlled, blinded studies with real, objective measures of effect or we risk making ourselves feel better without actually helping our patients and pets.