Listen to this story.
Magic pill
When a dog participating in a clinical trial of a supplement for cognitive health in geriatric pet dogs became more social and slept better, the researchers were almost certain the dog was responding to the experimental treatment.
They didn't know for sure because the trial, done last year, was randomized, placebo-controlled and double-blinded. In other words, the dogs in the study were given either the supplement or a placebo — a faux medical treatment — and neither the dogs' owners nor the researchers knew which was receiving which.
By the end of the six-month experiment, the dog seemed dramatically better. But it turned out that this dog, along with others in the study who showed signs of recovery, was on the placebo.
The dogs' apparent recuperation, it was later deduced, may not have been a recuperation after all. Rather, the researchers potentially were witnessing a psychological phenomenon called the caregiver placebo effect.
"He had his owner and our whole team fully convinced that he was on the active treatment," said Kate Simon, the study's lead researcher and a veterinary and doctoral student at North Carolina State University (NCSU) College of Veterinary Medicine. "His at-home cognitive behavioral symptoms that had been changing with age really seemed to be improving.”
The caregiver placebo effect occurs when a patient receiving a placebo appears to be improving in the judgment of the person caring for the patient. In human medicine, it usually happens when the sick individual, such as an infant or older adult, can't verbally express how they feel. So the assessment must come from the observations or examinations of others, like parents or doctors. Hopeful the treatment will work, the caregiver sees signs of recovery even when the patient isn't actually receiving treatment.
In veterinary medicine, the phenomenon happens when a pet owner or even a veterinarian perceives healing in an animal receiving an otherwise ineffective treatment. If unaccounted for, the effect can make an experimental treatment in a clinical trial appear more successful at producing the desired result than it truly is. Even approved drugs could, in a given case, appear more effective than they really are, considering patient individuality.
The potential for the caregiver placebo effect to emerge made it all the more challenging for Simon's team to decipher their results and the supplement's true efficacy. But the experiment was far from fruitless: It inspired the group to look into the caregiver placebo effect itself.
"A huge part of our cognitive assessments are owner questionnaires and surveys, which we use all the time in veterinary medicine because, of course, our patients can't speak to us," Simon said. "But there's a lot of criticism that they can be really susceptible to bias, and the results of our study got our gears turning."
During the initial trial, owners were asked to routinely bring their dogs to the clinic for a behavioral evaluation by a researcher and complete at-home questionnaires. Now, to determine whether the caregiver placebo effect was, in fact, responsible for the dogs' reported behavioral changes and to what degree, the researchers are comparing the results with those of a new group of dogs that are not receiving an intervention (not even a placebo) and being assessed in the same way. The team is watching to see if the new batch of dogs displays any signs of natural cognitive recovery.
The ultimate goal, Simons said, is to determine which evaluation methods prove most robust against any caregiver placebo effect and learn how to better analyze their results. The effect may still be evident in the new control group — caregivers can subconsciously treat and perceive their pets differently simply because they are aware of the trial. If that's the case, the team hypothesizes that the effect will be more apparent in the at-home assessments than those done by researchers at the clinic. There, Simon explained, the environment is more controlled and her team on guard for bias.
But the fresh study still has limitations, Simon acknowledged, pointing out that animals can act differently outside of their normal environment. Moreover, evaluating subjective aspects of animals' health, like cognition and pain, is inherently challenging.
Tricky to tackle
People have been studying the caregiver placebo effect in veterinary medicine for nearly 20 years. "The pain field is where it has been really looked at most," said Dr. Margaret Gruen, an associate professor of behavioral medicine at NCSU. "But it can show up in really any clinical trial, particularly those that are measuring a construct — something that we can't directly measure, such as anxiety" or, as in the study conducted by Simon's team, age-related cognitive decline.
Gruen has published research over the past decade on managing pain in cats with degenerative joint disease in which a caregiver placebo effect was identified. And in 2017, she co-authored a literature review of five other placebo-controlled experiments on cats with the same ailment that all saw the effect. As more veterinary researchers encounter and study the phenomenon, the key, she said, will be to figure out how they can reliably avoid and account for it.
Randomizing and double-blinding a placebo-controlled clinical trial is a critical part of dodging the caregiver placebo effect, according to Dr. Mark Rishniw, director of research at the Veterinary Information Network, an online community for the profession and parent of the VIN News Service. It's the best attempt researchers can make to prevent observer bias from skewing their outcomes. But, Rishniw stressed, they must also anticipate seeing the effect in both control and active groups and analyze their results accordingly.
"Doing the randomized control trial doesn't remove the caregiver placebo effect — it simply allows you to account for it," said Rishniw, who is also an adjunct professor at Cornell University College of Veterinary Medicine.
Generally, if the treatment being tested is actually effective, researchers will theoretically see improvements in the active group that are notably greater than those in the placebo, Rishniw explained. The treatment may be ineffective or less effective if both groups exhibit similar degrees of response. Of course, study results aren't always particularly consistent. That's why, Rishniw believes, the caregiver placebo effect is exceptionally tricky to bypass.
In the trial done last year by Simon's team, which had a wide variety of results, it's possible that dogs in the active treatment group could have benefited from the experimental supplement, in addition to any effects from the caregiver placebo effect.
Comparing the original placebo group with a new control group of dogs that aren't receiving an intervention might help the researchers know more about the magnitude of the caregiver placebo effect in their original study, but even observational groups aren't impervious to bias, Rishniw said.
"With something like cognitive decline, it's usually the primary caregiver who can determine if a patient is improving or not, because there's really not a lot of ways an independent observer can detect changes," Rishniw said. "But even the pet owners in an observational group might [subconsciously] treat their dogs differently just because they know they are in a trial. It's what makes these sorts of studies so difficult to conduct."
A potential related complication is that patients could get a real health benefit from a placebo due to the social activity its administration entails, notes Dr. Dawn Boothe, a clinical pharmacology consultant at VIN. "The simple act of giving medicines — interacting, that pat on the head when it is done — is going to make the animal feel better," Boothe said. "So what the caregiver perceives is actually real, but it is not the treatment [that's helping] as much as it is the giving of the treatment."
There are other ways, though, in which researchers can combat the caregiver placebo effect. In one of Gruen's double-blind studies examining pain relief in cats with degenerative joint disease, her team was able to successfully circumvent it by adding a three-week "washout period" following the trial's regular treatment period. During the washout, participants in both the active and placebo groups received only the placebo. When one group was taking the pain medication and the other was taking the placebo, both were reportedly improving. But after the cats on the active treatment started getting the placebo, their pain symptoms relapsed, suggesting the painkiller was effective after all.
Other types of experiments would require different strategies to circumvent the caregiver placebo effect. For example, if testing the efficacy of a physical procedure such as a physiotherapy treatment, Rishniw said, a person separate from the evaluators would need to perform all procedures behind closed doors or conduct a faux maneuver on placebo-receiving participants that, for the evaluators and pet owners, is indistinguishable from the real one.
Delivering some home truths
A psychological tendency to see what you want or expect to see happens beyond veterinary clinical settings, as well. The caregiver placebo effect can also show up when pet owners try to treat patients themselves, according to Dr. Brennen McKenzie, a small animal veterinarian in Los Altos, California. The phenomenon, he said, often surfaces when owners apply products or practices that haven't been scientifically validated, such as homeopathic remedies.
"I think that it's just built into the human brain," McKenzie said, who also authors the SkeptVet blog. "We care for our pets so deeply and want to see them get better. And sometimes, even when a therapy has no real scientific evidence behind it, we think we see them improving."
Even for home remedies that are essentially harmless, it's best to try to curb the caregiver placebo effect, McKenzie added. If a pet owner believes that their animal is recovering with a treatment lacking scientific validity, that may delay the animal's access to a treatment that is clinically proven to be effective. So McKenzie does his best to encourage pet owners to opt for science.
"People like to sometimes believe that just because our dogs and cats can't believe things about their therapy, that there's no placebo effect," he said. "It just doesn't occur to them that their perceptions of what's going on with [their animal] might not be what's most accurate. ... So, I ask them to trust me, as compassionately as I can. That's my duty to the pet.”