In a previous article, I shared the results of a survey that we conducted following the conclusion of a recent multisite, double-blinded trial. We found that 82% of the owners had enrolled their pet to improve its health and/or well-being. Encouraged by that positive statistic, we then wondered how each subject owner evaluated their pet’s reaction to the study treatment. Was it purely subjective judgment, or did they employ some sort of metric? And could there be a “placebo effect?”
The word “placebo” comes from the Latin word meaning “I shall please.” Whether they remember their Latin or not, some medical practitioners embody that meaning when they administer placebos solely to placate their patients. The oft-observed “improvements” in patients’ conditions following the prescription of a sugar pill gave rise to the term “placebo effect.”
In clinical research, a placebo has a more scientific purpose. Clinical trials often employ a placebo as a comparison benchmark, a “control” or “negative control,” to assess the results of an experimental therapeutic intervention. In some trials, there may be a “positive” control, a compound already approved for treating the indication being studied.
This comparative study design is thought to render an unbiased evaluation of the relative effectiveness of the new treatment. However, the true effectiveness of the experimental compound can be called into question when study subjects on a negative control exhibit a greater degree of improvement in the indication being treated than manifested by those given the test article.
The placebo effect in human clinical trial subjects has been addressed in myriad articles in the professional literature, but the topic is less frequently covered in the veterinary press. (Perhaps the language barrier between animals and human makes the collection of reliable data more challenging?) According to Franklin D McMillan (“The Placebo Effect in Animals,” Journal of the American Veterinary Medical Association, 1999), a placebo effect is recognized as that which is favorable, non-specific, psychological, or psycho-physiological, but has no specific treatment for the condition being studied. Are animals prone to the same reaction as humans?
Some ascribe the placebo effects observed in veterinary studies to the animal’s caregiver, based on the theory of expectancy. In other words, there seems to be a body of evidence that suggests that what the owner of a study subject expects from treatment with the test article will, in many cases, be “observed” in the pet and reported by the owner.
In a post-study survey, nine of the 27 owners whose animal had received a placebo responded that the study treatment had been effective. Assuming no test article administration errors, it would appear that 33% experienced the placebo effect.
According to Dr Alan H Roberts in “The Power of Nonspecific Effects in Healing” (Clinical Psychology Review, 1993), a placebo response rate of 35% is commonly found in human trials. So, is the placebo effect contagious between owners and pets? Or is something else going on?
The table below summarizes the owner-reported results of a recently-completed randomized, double-blinded, placebo-controlled study.
|Treatment Assignment||Effective||Not Effective||Total|
In this study, test article effectiveness was evaluated by each animal’s owner and not substantiated by laboratory test results or medical examinations. Therefore, it is possible that some of the owners of placebo subjects expected the treatment to work and, consequently, “observed” effectiveness.
Another explanation could be that some owners were more attentive to their pets because of the recording requirement of the study, and that increased attention generated a “healthier” response. Perhaps some owners were less stressed once their pet became enrolled in the study, and that calmer demeanor manifested itself in fewer observed symptoms, real or perceived.
On the one hand, we can be happy that these nine animals apparently enjoyed relief from the symptoms that prompted their owners to enroll them in the trial. On the other hand, these results might suggest that a more robust study design is required for future trials to minimize the possibility that owners of study subjects “catch” the placebo effect.