The Scoop Equine Nutrition Podcast: Ep 56 The Placebo Effect

Annie Emery of Revolution Equine Nutrition takes a look at research behind the placebo effect- what is it? Why should we care? What does the science say? 

The Scoop Equine Nutrition Podcast is self-funded and isn’t here to sell you any feeds or supplements.

Find Annie:

https://revolutionequinenutrition.org

https://revolutionanimalnutrition.com

Annie is a certified canine, feline and equine nutrition advisor loc

Transcript

Copy Transcript Copied

0:01

Welcome to The Scoop Equine Nutrition podcast. Thank you so much for joining me. My name is Annie Emery. I own and operate Revolution Equine Nutrition, so I'm an independent equine nutrition professional, and this is the podcast that is self-funded, not sponsored by any company. I'm not gonna sell you a supplement, don't worry. But that said, I did wanna make you aware of some education opportunities coming up, because I'm doing a lot of cool things, and the more you support, the more I'll be able to do in the future. I am kicking off some fall travel. if you live near Minneapolis-St. Paul or Western Wisconsin, and you want me to come see you, I'd be happy to do in-person consultations to meet with your horse and you. And then I can also do some teaching times with your barn, with your riding club, tack store, et cetera. Contact me if you would like help with that. And then I am continuing to see clients virtually through the United States and Canada for nutrition programs diet balancing, medical diets, et cetera, as well as general webinars and teachings. There's a variety of offerings, and I would love to connect with you. And I also own and operate Revolution Animal Nutrition. I do dog and cat diets as well. I do all the different diets, raw food, kibble, fresh, et cetera. Today we're gonna talk about a topic that I thought was pretty interesting, and I think you'll find it interesting as well. What is the placebo effect? hang in there with me, kids. It's actually really cool. The question is, okay, so we have a horse, dog, cat, a child. The animals and children don't read the label. They don't know the advertising. They don't know what's gonna happen here. And they don't read the ads. Today's topic is gonna be a brief discussion about the expectation, observation, natural fluctuations, learning, and human-animal interactions can influence what we believe we're seeing So before we look at the animal studies and the research, we need to separate here two terms that are frequently used as though they mean the same thing. we have a placebo response and a placebo effect. A placebo response is any change measured while someone or an animal is receiving a placebo, so that's not the active thing being studied. Because we can't just give everyone the same thing, we need to know against a baseline. If an animal improves during the placebo phase of the study, that improvement is part of the placebo response. That does not automatically mean that the placebo caused the improvement. the condition fluctuates. I know we see that in animals all the time. Something else in the animal's environment may have changed. The caregiver team may be different. They might give more or less exercise, daily care, weather, ground situations outside, mud, et cetera. And the person measuring the outcome may interpret things differently as well. We all have our own flavor or our own style, our own personalities, and that feeds into what we're seeing, and the measurements itself may vary. And there's also something that's a science word called regression towards the mean. Now why do we care about this?" Because I want you to be the smartest kids in the class. This is important when we evaluate supplements and herbs and, treatments. I cannot do medical work, so today we're gonna be talking about supplements. Although in my study here, I'm gonna be bringing in more medical things 'cause that's simply where we have the research, right? I'm using medical terms to illustrate principles that we can, backfill and apply to supplements, herbs, whatever we give our horse that's not a drug, right? And there's getting to be a little bit more of a gray area about, supplements are being used for medical conditions. what is regression towards the mean? The treatment, the supplement, the big push is starting when typically there's a problem. We don't typically start things before we have a problem. We start them when a problem is noticeable. The horse is stiff, the dog is uncomfortable, the cat had stopped jumping into the bed. There's something that the animal is doing that is saying, "Hey, there's a problem here." So then we take action, potentially a supplement. the, quote, "unusually bad period" is when we typically make a change, which is totally normal. But the extreme observations are usually followed by observations that kind of become close to that animal's normal, whether or not we intervene. If the animal moves better a- from an unusually bad period, we might say that the product did the thing. The product made the cat jump on the bed easier or the dog move easier. But not so fast We have to be s- clear and follow it over time. A true placebo effect is more specific. Is the change caused by the context surrounding this treatment, this inactive treatment? 'Cause remember, the placebo is not supposed to do anything. Ideally, we would love to see research having a placebo group and a no treatment group because the placebo itself could be doing something. We need to separate is the receiving treatment from a placebo something that would occur anyways? Is the effect of apparently receiving a treatment enough to cause a change? 'Cause you see what I'm saying here? There's doing nothing, there's giving a placebo, and there's giving a treatment, and again, I'm using the word treatment to mean supplement. this distinction becomes important when someone says the animal receiving the placebo improved, so the placebo worked. We can't really make that claim scientifically. We can say the improvement occurred during placebo treatment. Determining why it occurred is the harder question, right? there's a study here that I wanted to talk about that looks at osteoarthritis in dogs. this is from twenty twelve. We have a veterinary study looking at dogs with arthritis. We have a paper called Caregiver Placebo Effects for Dogs with Lameness. this is coming out of the Journal of American Veterinary Medical Association. this is a prospective, randomized, double-blind, placebo-controlled, multi-center clinical trial. this is a real research study. The researchers examined fifty-eight dogs with lameness caused by osteoarthritis, joint pain. The dogs were in the placebo arm of a larger FDA-approved clinical study. The researchers were interested in the relationship between subjective assessment and objective measurements. the subjective assessments are coming from the owner and the veterinarians. The objective measurements come from force platform gait analysis. the force platform measures the force a dog produces when they're stepping on a plate. In a dog with an uncomfortable limb, a sore leg, the researchers can measure how much weight or force the dog puts on that limb. this is giving a numerical measurement of the limb use rather than relying on what the dog appears. "Oh, the dog is moving better." Okay let's look at the numbers. The owners evaluated their dog's lameness, and the veterinarians also evaluated the dogs at the walk and the trot. They checked for pain during palpation and the affected joint, they looked at the posture and did other parts of the standard examination for arthritis. The researchers defined a caregiver placebo effect as a favorable assessment that was not supported by the objective force platform results. in other words, the caregivers believe the dogs have improved even though the forces shown on the plate were unchanged or even got worse. over this forty-two day study, the average reaction forces from the placebo dogs remained unchanged. When the dogs were considered individually, five out of the fifty-eight dogs worsened by at least five percent. Seven improved by at least five percent, and forty-six were remain unchanged. objectively, the large majority of the dogs did not show meaningful change in limb loading. Okay. That's not good or bad, that's just the way it was. what's interesting is in about 40% of the cases, the caregivers thought the dog got better, and in about 45% of the time, the veterinarians thought the dogs got better and the effect became stronger over time, which is fascinating to me. as the study continued, the caregivers thought the dogs got better and better. The perceived improvement was there, even though the dogs did not show any difference in their plate forces. Their gaits did not show that improvement. this is important because we often assume that when we see something over time, it makes it more reliable. This would say, "Oh, I've, I saw my dog, my horse, my cat walk for several weeks now. They look better. I'm sure I'm seeing a difference." But the time here is showing the opposite effect. The placebo perception grew where the force plate analysis in this study did not back that up. We follow the instructions, and we want the animal to feel better. We are stressed by something. We want to see improvement. Everyone does. And the more we wait, the more meaning we might attach, like we're in it. Now, this is also interesting because we love our vets, and you guys know I'm a huge supporter of the veterinary community. Nutrition people are not veterinarians, but veterinarians are also affected along with others. this doesn't mean the veterinarians are bad or they're not doing their job. It just means we have certain clinical judgments and assessments. this study does not mean that observations are worthless. And we need to also explain that a force plate captures a particular part of the dog's movement during a brief test. This isn't gospel truth. Now, as owners, we see our pets in their environments. If a dog is not jumping on the bed one week and jumps on the bed another week, that's real. we just need to remember that one study is not the be and end all of everything. It's just the information we have available as a starting point. again, if you believe a supplement is working, I always honor that, and I'm not going to say it's not valuable. We just need to kinda take everything in context. And the force plate itself is gonna be measuring different aspects than you as an owner is measuring. But when we have a subjective and objective measurement that is so far apart, we might just, have to take a look at that. We cannot assume that we're seeing something if it's like really not there. Now, switching gears here, we have another paper from twenty ten. We have a paper called "The Placebo Effect in Canine Epilepsy Trials" in the Journal of Veterinary Internal Medicine So this is a paper understanding the difference between a placebo response and a proven placebo effect. what does this even mean? we have a meta-analysis of three prospective placebo-controlled trials, including dogs with epilepsy. The researchers compiled a number of seizures each dog experienced per week and compared the seizure frequency during the baseline and placebo periods. we have thirty-four dogs in the overall analysis, and twenty-eight received placebos during the study So of those 28 dogs, 22 of the 28, which is 79%, had fewer seizures during the placebo administration than they had during the baseline. Eight dogs, or 29%, experienced a reduction of at least 50%, which is not nothing. And any other study would say, "Hey, that's classifying that dog as a responder to the treatment." Problem was, there really wasn't a treatment. Across the three individual trials, the average reduction in seizure frequency during placebo, was 26%, 29% and 46%. two of those changes were statistically significant, while one was not. Why do we care about this? the numbers are impressive, but why did an inactive capsule, quote, "biologically treat epilepsy" in nearly 80% of the dogs? We don't know. seizure disorders fluctuate. That's my point. Dogs may enter a clinical trial after a severe cluster of seizures and then move back towards the regression of the mean. They go back to their baseline. this is regression towards the mean. Now, enrolling a study may also change the dynamics, right? Like caregivers might have to be on a protocol. They're a little more accountable for following the process. They might keep better records, they might have different routines and be less likely to miss a dose. they're paying more attention. And seizure control, the seizure counts also depend on the caregiver observation and recording. again, a seizure is more concrete than saying an animal seems better, but the data is still passing through a human observer. the researcher's point was not that inert treatment should be used for epilepsy, but the point is that open label study without a placebo comparison could make an ineffective treatment appear effective because the placebo, which shouldn't have done something, quote, "did something". If 29% of placebo treated dogs can meet a common definition of a, quote, "treatment response," simply reporting that 29% of dogs improved after receiving a new product tells us very little by itself. Oh, very interesting Now let's switch gears to cats. In twenty seventeen, we have a study called The Caregiver Placebo Effect in Analgesic Clinical Trials with Cats with Naturally Occurring Degenerative Joint Disease Associated Pain. Okay, so it's about cats. It's about pain relief in joint pain. They looked at scientific literature and six studies looking at placebo-controlled of analgesic treatments. So cats with degenerative joint disease with the pain And they have a lot of data from five studies involving ninety-six cats. That's a fair amount of cats. these five studies included a supplemental diet, which is very interesting, a nutraceutical, a study evaluating e- feline pain, a questionnaire, low-dose NSAIDs, and an anti-nerve growth factor antibody. the cats received physical, orthopedic, and neurologic examinations. The joints were identified through how the cats responded to palpation and manipulation, and they took X-rays to look at the diagnosis of degenerative joint disease. Okay, so what did they find? Now, they looked at two different ways of measuring improvements, which I think is key. we have the client-specific outcome measure. the owners said "Hey, my cat likes to jump on the bed." So they could kinda say for that cat, that was the metric. Other cat might be, okay, this cat is going up a certain set of stairs. This other cat's going to the litter box, being more comfortable playing, or getting into a chair. there were different things based on the cat. This is very reasonable because cats, all are very different, and what's meaningful to one might not be meaningful to another. some of the studies used three owner-selected activities, with each activity scored from zero to four. success required total improvement of at least two of these things. this is about a sixteen percent chance. Okay, so we're saying about a twenty to twenty-four percent reduction in pain and improvement from baseline. Another study used activities. then their metrics were they wanna see improvement in at least four of the points, that represented a twenty percent change across the scale or about thirty percent decrease in pain and impairment. Numbers are important. That's why I'm harping on this. This isn't just guessing. So the researchers did not simply ask whether they thought the cat was better. They used premeditated, predetermined numerical criteria to define success, which I think is very important. That's why I'm spending so much time on it And the second measurement was physical activity. The cats actually wore accelerometers, which is simply activity monitors like an activity tracker. activity was recorded in one-minute intervals, and the cats wore the same monitor through the whole study. the researchers looked at each cat's average daily activity during baseline and treatment. And for them this a success, the cat needed at least a ten percent increase in average activity to be classified as success. smaller increase, no change, or decrease was called a failure. this allowed the research team to look at what the owners believed they were seeing with what the activities monitors actually recorded. I think this is really cool. Okay, so here's what they found. during the placebo treatment, which the owners didn't know what they were giving the cat if it was a treatment or a placebo 80, about 87% of the cases showed at least one point improvement on questionnaire. And when the researchers applied the stricter predetermined definitions of success, between 54 and 74% of the placebo-treated cats were still classified as successes by their caregivers. the objective activity results were more variable. They found that depending on the study, between 10 and 63% of the cats met the definition of activity success. And here's the kicker, only 36% the cats classified as success by their owners were also classified as success by the activity monitors. And at the same time, 19% of the cats whose owners did not classify them as a success actually did meet the activity monitor definition of improvement. it goes both ways. Isn't that interesting? bias does not only happen in one direction. They missed the activity. The caregivers can miss measurable improvement, which again, it's not trying to be negative, it's just trying to understand the situation so we make good decisions. the researchers tested whether caregiver-reported success was associated with cat's age, weight, baseline activity, X-ray disease scores, orthopedic pain, or the type of study. They did not find significant effects from their variables. That part was extremely small with only eleven owners. we could not conclude and should not conclude that personality never matters. However, it does suggest that the cl- caregiver placebo effect cannot be neatly dismissed as something that happens only to unusually optimistic people. thought that was a very important point to consider. The researchers also discussed how a study design can influence results. some of the studies include a two-week baseline period. This gave the owners time to use the assessment tool and help reduce the effect of suddenly paying unusually close attention to the cat simply because the study had begun. the authors noted that a owner's first assessment might make the cat look worse than later assessments. the owner knew and knows that the cat must meet certain criteria to qualify to further study and maybe you want it to happen, and they may focus intently on every difficulty during the initial evaluation. And then we have a nutraceutical study that tried another strategy. It included an initial placebo period, but the owner did not know that everyone was receiving a placebo. There was no quote, treatment. the theory was here that the initial excitement or the expectation surrounding this treatment and this nutraceutical might stabilize before the active treatment was introduced. So interesting, a substantial placebo response still occurred. some of the tested treatments simply might have not been very effective. A larger placebo response makes it harder for genuinely effective treatment to separate itself statistically from a placebo. But it can also expose a treatment that does not add much beyond the surrounding context. another interesting thing is the activity monitor results included another possibility. It's called placebo by proxy. And remember, slum placebo-treated cats improved not only in their owner's reports but also in their activity monitors. One explanation is that the owners correctly observed real changes in their cats caused by a natural fluctuation. Another explanation is that the owner's behavior changed because the owner believed the cat was receiving treatment. Perhaps the owner paid more attention to the cat, interacted more, provided more opportunities for movement, or interacted with the cat in a more positive way. That could legitimately produce a real increase in the cat's activity, even though the placebo itself was not causing that effect. And the researchers were very careful here, as they should be. They did not claim that they had a proven placebo by proxy effect. They said it was a possibility that deserved additional study. Again, we wanna make sure we're not leaping in logic here. The paper says what this paper says. We don't take one thing and translate it into a million other things. cats do vary in their activity, and we have to keep that in context. Now let's talk about children. In twenty thirteen, there's a placebo-by-proxy study in parents' belief on therapy for children. This is in the Journal of Behavioral Medicine, so their study in twenty thirteen recruited fifty-eight children between the ages of two and five whose parents reported frequent temper tantrums. the children were given pharmacologically inactive flower essences. this is a product that was presented as something that was intended to reduce tantrums, but it's essentially an inactive scent So the study collected measurements rather than asking the parents one question before treatment and one question after. the tantrum frequency, severity, and parental mood were measured in five different occasions during an eight-day pretreatment period and another in five occasions during the ten days after the treatment began. And remember, this is an inactive flower essence in this study. Again, that's not saying all flowers are inactive. It's just this is for the study. during the treat, quote, "treatment period," which remember wasn't a treatment, the reported frequency of tantrums and severity of the tantrums continued to decline. this was statistically significant. It was a real thing that they noticed. And there was a between the parents' mood and reported tantrums. Better mood of the parents resulted in fewer and less severe tantrums. Now, we need to be clear that this particular study did not involve an objective observer watching the children's behavior. It did not have an, quote, "untreated control group." And, there was only the group with this inactive flower essence. There wasn't a group that didn't have the essence. We just have to understand context. It doesn't mean it's bad or good. And the authors are appropriately cautious. I have always found that scientists don't make leaps in logic. They could not determine whether the children's behavior objectively changed, whether the parents perceived the same behavior differently, or whether believing the treatment had begun changed the way the parents interacted with the children. It could have been a combination of all three. So a parent who believes that help has arrived may become calmer and more patient. The parent may respond differently when the tantrum begins. So the child may then respond to the parent's changed behavior. So in that case, the child's behavior changes. It really does change, even though it wasn't the substance itself driving it through a pharmacological effect. So this is the heart of placebo by proxy. expectation can affect the caregiver. The caregiver can affect the individual receiving care, whether that's a child, dog, cat, or horse. Imagine you're giving your horse a calming supplement because you believe it will help. Perhaps you approach the horse with less tension. Your breathing changes. Your hands become quieter. You give the horse more time to process. You feel like you've got another thing in your toolbox that, that's supporting and you're less likely to jump the gun. So the horse might respond to your response. So something real is definitely happening there. This is positive, and this is fantastic. So it's not the supplement responsible, but the results are still great. It's just the results are from the environment shifting, not the substance. And again, let's talk about a dog with, a painful leg. If you believe the new treatment is helping, you might be less hesitant about inviting the dog to move. You might take a few more walks, provide more and the dog kind of loosens up their joints and maybe they move better. So the influence the arthritis, and the dog is getting more confident, you're getting more confident, and one thing that I found really interesting in my study is animals can develop conditioned psychological responses. This doesn't mean that they have human-style consciousness for treatments associated. There's one more layer to this. Animals can develop what call conditioned psychological responses. So this isn't human-style consciousness. So there's a classic study from nineteen seventy-five The researchers paired saccharin flavored water with an immunosuppressive drug that also produce illness and taste aversion. So there's a sweet taste, and we have a condition with rats. The rats were given this material, and then they were injected with red blood cells to stimulate an immune response. Then six days later, the researchers measured antibody titers. How much did the immune system kick off? The non-conditioned animals produced high antibody titers. Conditioned animals that were not re-exposed to the sweet drink also produced high titers, and animals receiving Cy with the antigen did not produce an antibody, which was expected because the drug, the Cy drug, suppressed the immune response. That said, the conditioned rats were re-exposed to the sweet taste around the time of the antigen injection, and they were also significantly immunosuppressed, even though the taste itself did not have the Cy in it. And there's an important control here, which makes this a very powerful study. They also conditioned a taste aversion using lithium chloride, which could make the animal sick but was not itself an immunosuppressive agent. That conditioned taste aversion did not reduce the antibody titers. So this control helps show that immune changes were not simply caused by remembering that a particular taste had made the previous animal feel sick. The conditioned response reflected the specific physiological effect that had been paired with the taste. So why do we care about this? This doesn't mean that giving an animal a supplement created a conditioned immune response. What it does show is it demonstrates the animal nervous system and physiological systems can learn associations between treatment cures and biological responses. Conscious belief is not required. So what about horses? This is a horse podcast after all. The direct caregiver placebo literature is much stronger in dogs and cats than horses, I have not found a horse study equivalent to the dog paper, and that directly talks about the owner improvements with objective measures such as the force plate analysis and identifies the disagreement as a caregiver placebo effect. What is my take-home message here? Let's just say we're using a joint supplement or an allergy supplement When we say, "Oh, I want my horse to feel better," okay, let's translate that into concrete steps Say, "Okay, I have a horse with arthritis, first of all, I would tell you kinda when you would expect to see a timeline, when we would expect to see changes, 'cause every supplement program we would expect different things at different times. Knowing what window you would look at is very important to me, and I educate on that in my consultations. And then we can take that into, okay, when we say that the supplement has, quote, "kicked in," I wanna say that my horse takes longer steps during the warm-up, or my horse isn't refusing a jump, or my horse goes up a certain hill a certain way. Create a baseline before starting. Record behavior look at measurements. Use numbers if you can, because I think that helps. Make notes, have a calendar, use video. Use the same surface, preferably the same time of day, the same camera position. Take before and afters. I'm a big believer in taking pictures of things. People do that with me, and they're like, "Oh my gosh, Annie, I forgot how my horse looked before our nutrition program, and now the horse looks amazing." But you start to adjust your eye a little bit, right? And if you're doing a dog evaluation, record the dog getting out of bed, jumping, et cetera. For cats, again, you can look at how they jump, how they use the litter box, how they play with their toy. And then if you have a paddock camera for your horses, how are they moving? Where are they going? Are they traveling more? So again, if you wanna play with it, this is just ways that we can break down studies into real world measurements, 'cause I believe in empowering you guys to make good decisions. Now, here's another key. Don't try to change 50 things at once. Keep things simple. Use subjective and objective measurements when possible. If you have something that would be changed with veterinary work, bring your veterinarian in, perform lameness evaluations, blood work, whatever think they wanna do. Now, again, bias is not dishonest. A bias doesn't mean someone is stupid. There's a reason that scientists use randomization, placebos, blinding, and outcome measures. So everyone is capable of bias, including the research teams. Big picture question, can animals experience a placebo effect? Yes. Don't jump to too many conclusions. And the next time you begin with something new, pause as you consider things. Look at the big picture. I do believe owners, like I fundamentally believe if you're seeing something, I believe you, but at the same time, we have to have a full conversation about this. My job is to help people pick supplements. Literally, that's my job we pick feeds, we pick supplements, and if it's a probiotic, we expect a result in this amount of time, generally. If it's a hoof supplement, there's that amount of time. If it's topline, it's that. If it's joint, it's that. My job is to guide the conversation around when do you expect generally to see something And keep your veterinarian in the loop. Supplements are not treatments, and as nutrition professionals, we cannot diagnose or treat medical conditions. It's just creating supportive environments for horses, and we have certain nutritional diets for problems like muscle myopathies, gastric ulcers, et cetera. Those are specific nutrition programs, but will not replace veterinary work. For example, if you have a horse with Lyme disease, put them on an antibiotic, don't treat them with herbs. If you decide to treat them with herbs, have that work with your veterinarian. This kind of whole animals thinking is what I do best. If you need help with this, I can support you. I still take out more supplements than I add by a lot. But, it is important to understand things. So again, I own and operate Revolution Equine Nutrition, until next time, stay curious, keep asking good questions, be the smartest kids in the class, and remember that noticing a change is the beginning of the investigation, not necessarily the end. We'll see you next time on The Scoop Equine Nutrition Podcast