I recently participated in a free webinar discussing neutering of dogs and cats (always a hot-button topic!). I made a screen recording of my segment of the presentation, which I am posting here.
There were some technical glitches, including some popup notifications that I have only been able to hide imperfectly. I was also asked at the last minute to shorten the presentation by 10 minutes, so it’s a bit rushed and doesn’t flow perfectly, and it ends abruptly since I handed over quite quickly to the next speaker. Nevertheless, I hope there will be some useful concepts and information in there for you!
With the dozens of posts covering hundreds of scientific publications and my own experience with acupuncture trainingand practice, I have surely exhausted the subject. Right? Well, in my own mind I have since it is clear to me that acupuncture is founded on a combination of folk mythology and unreliable science, and it almost certainly does little to nothing useful for our pets.
That said, it is still accepted as beneficial by plenty of vets and pet owners, largely on the basis of unreliable anecdotal evidence and an enormous, though ultimately unconvincing, scientific literature. Therefore, I feel obliged to continue my occasional, and almost certainly futile, efforts to protect pets and their owners from wasting time and money, and failing to effectively treat serious medical problems, by employing acupuncture.
A recent study of acupuncture as a treatment for pain in dogs after ovariohysterectomy (spay surgery) reinforces the case against this practice-
The investigators compared pain scores in dogs after spay surgery between two groups of dogs: all getting the standard treatment (an opioid called butorphanol, a sedative called medetomidine, general anesthesia, and an NSAID called meloxicam), some randomly assigned to “electroacupuncture” (electrical stimulation through needs at purported acupuncture points, though these don’t really exist) and “fake” electroacupuncture (needles at the same locations not actually penetrating the skin). The pain assessment was don’t by vets who were, in theory, unable to tell which group got the real and which the fake acupuncture (the needles were passed through a piece of opaque foam so the skin underneath couldn’t be seen).
The result was no difference in pain scores between the groups at any point. Some dogs did show evidence of increased discomfort shortly after surgery, and this was likely due to the use of butorphanol, which is a weak and short-acting opioid and no longer considered adequate for use in a surgical procedure like this.
Of course, negative findings like this never convince believers in acupuncture, and the results didn’t disturb the investigators at all. They simply pointed out positive findings in other studies and came up with a list of excuses- the wrong length of time for treatment, treatment during general anesthesia, possible interference from drugs used, etc. Every possible explanation except for the most plausible- acupuncture is nonsense and doesn’t reduce pain except through placebo effects!
Do you trust your vet? How much? It turns out your answer may well be a good predictor of how likely your dog or cat is to be appropriately vaccinated.
“a large minority of dog owners consider vaccines administered to dogs to be unsafe (37%), ineffective (22%), and/or unnecessary (30%).
A slight majority of dog owners (53%) endorse at least one of these three positions.”
48% of dog owners opposed mandatory rabies vaccination and agreed with the statement, “The decision to vaccinate dogs that are kept as pets should be left up to individual pet owners.”
These views were less likely in those with a college education, and more common in those who also held misinformed views about vaccination for children.
A new study has looked at this issue again, in both cat and dog owners.
The findings support and amplify messages from previous work, showing that while most pet owners trust their vet and rely on them as the main source of information, all too many are still less trusting and more influenced by less reliable sources of information. The basic findings were these-
Overall, 62.9% of dog and 61.2% of cat owners were classified as trusting their veterinarians.
The longer people know their vets, the more confidence they have in them.
Vets were the primary source of information about vaccines for dog and cat owners, followed by the internet (Figure 1)
Owners with less trust in vets were more likely to turn to other sources, especially the internet, for information.
People who trust their vets are more likely to have their pets properly vaccinated than those who don’t.
Those who rely more on the internet for information are less likely to have their pets properly vaccinated.
Other variables, such as political affiliation and education level, were inconsistently associated with reliance on vets as a primary information source.
This is not a revolutionary study, but one that supports the broad understanding we already have about the role of personal relationships built over time in establishing trust and support clients in making science-based decisions about their pets’ health. It also highlights the unsurprising deleterious role of the internet as a source of information about vaccination (and many other topics!).
In my presentations on mistrust and misinformation, I have emphasized that trust in vets and science remain high overall, though it is eroding among some segments of society. We still have the confidence of many pet owners, and we can still help them protect their pets from the mistakes encouraged by sources of anti-scientific information. But we cannot blindly rely on that confidence, and we must continually nurture the relationships with our clients that allow us to help them resist the rising forces of mistrust and misinformation.
Integrative Medicine- Mixing Apple Pie and Cow Pie A recent issue of the journal Today’s Veterinary Practice contained a couple of articles form some familiar folks who represent the problematic notion of “integrative medicine.” This is a poorly defined term which typically is a way of signaling support for including alternative therapies with little scientific validation alongside legitimate science-based medicine. It creates an aura of open-mindedness and signals that proponent are not extremists but reasonable and willing to use any tool that helps their patients without prejudice. Science-based and alternative therapies are simply tools in a toolbox, and whatever tools suits the job is the one that should be chosen.
This, of course, ignores the fact that we have good reason to believe science-based “tools” actually work, while the safety and effectiveness of alternative methods is often unproven, and sometimes clearly disproven. All therapies should have to meet the same standard of evidence, that of scientific testing, and integrative medicine is simple a way of exempting some treatments from proving their worth objectively.
Mixing treatments that don’t work with those that do isn’t going to improve care. Or, as infectious disease specialist Mark Crislip has put it, “If you mix cow pie with apple pie, it does not make the cow pie taste better; it makes the apple pie worse.”
Nevertheless, the integrative ideology is popular because it feeds into the middle-ground fallacy, the belief that when there are two opposing sides to an issue, the truth is always a compromise somewhere in the middle. In reality, sometimes one side is actually right and the other side is wrong, so eschewing both exclusively scientific and alternative medicine in favor of an integrative mix of the two isn’t necessarily the best path.
Fresh Foods vs Conventional Diets The first article that drew my attention was concerning the possible benefits of “fresh” pet foods. The first author was Dr. Donna Raditic, a nutritionist I have written about a number of times. She was involved in running an integrative medicine fellowship program, and she has been a consistent proponent of mixing alternative therapies, such as herbal medicine, acupuncture, Bach flowers, and homeopathy, with conventional treatments.
The focus of this article is essentially that fresh foods are safer and healthier than “highly processed” conventional diets, due mostly to the concern about potentially carcinogenic chemicals created by “heat processing” (“cooking” to most of us). She advocates using the NOVA classification system, or something like it, to rate pet foods and avoiding those that are highly processed.
I have written in detail about the issue of processing, and the limitations of the concept. While the idea that fresh diets might have some health benefits is a reasonable hypothesis, it is currently supported mostly by associations in observational studies in humans or by lab animal studies that poorly reflect the reality of what people and pets actually eat. Nutritional composition and quality may well be more important for health outcomes, and processing may be a confounder rather than a problem in itself. There is also significant doubt about the value of the NOVA classification system in human nutrition, and about whether using it as a primary guide to the healthiest diet is actually beneficial.
In keeping with the tone of “integrative medicine” thinking, the article contains reasonable hypotheses and raises issues worth considering. However, it presents the benefits of the NOVA system and the risks of processed foods with far more confidence and certainty than is justified by the existing evidence, and it assumes the superiority of the alternatives without any evidence at all. This reveals the underlying bias of the authors, who have clearly already made up their minds regardless of the limitations of the evidence. Assuming the truth of the proposition you are arguing is a type of logical fallacy known as begging the question, and it is prominently displayed in this article.
The authors may, of course, turn out to be right about some of their claims, and if the data showing this appear, I will happily shift my position on fresh diets from neutral to strongly in favor. For now, however, as veterinary professionals and pet owners, we should not obsess about the “processing” involved in the making of the pet foods we use. While inclusion of some fresh or whole foods may well turn out to have some health benefits, it is far more important that making sure our pets are eating complete and balanced diets that meet their nutritional needs, regardless of the form they take.
An Integrative Approach to Aging Regular readers will know that for the last several years I have been deeply involved in canine aging science, working at a company developing drugs to extend healthy lifespan and as part of several scientific collaborations in this area. So I always pay attention to articles and books about aging and longevity.
Unfortunately, there is at least as much hyperbole and outright nonsense in this field as there is legitimate science, and lifespan extension is claimed freely by lots of folks as a benefit for unscientific or unproven treatments that they have long been advocating. I have illustrated this in a number of book reviews, most clear for the book Forever Dog, which is a superficial reframing of many myths and cliches from alternative medicine into an “anti-aging” approach.
Dr. Gary Richter is another of the integrative medicine crowd I have discussed previously, and he has a book on a “holistic” approach to longevity for dogs that I have read but not yet reviewed. Suffice it to say the book contains the usual mixture of reasonable lifestyle advice (maintain a healthy body weight, exercise regularly), mainstream veterinary advice (see a vet regularly, get at least some vaccinations), unproven or exaggerated claims (for unconventional diets and supplements), and outright nonsense (Chinese and other herbal medicine, chiropractic, etc.). Mixing the good, the bad, and the ridiculous may sound enlightened and “open-minded,” but it’s really just cow pie and apple pie again.
In this article, Dr. Richter reviews the hallmarks of aging, a common system used to categorize factors contributing to aging in a way that facilitates study and that might someday be useful as targets for therapies to slow aging. However, as usual there is a lack of any distinction between scientifically validated tools, plausible but unproven hypotheses, and pet theories with no substance or evidence whatsoever. Almost everything is presented equally positively, which makes for an inspiring, but ultimately misleading read.
The real danger of the integrative approach is evident in the fence-straddling Dr. Richter does about vaccination and parasite prevention. He does recommend these as “protect[ing] pets’ immune systems,” but then he almost immediately adds, “inappropriate or overuse of vaccines, antibiotics, and other medications, as well as early spay and neuter surgeries, may have equally impactful negative effects on longevity.” This broad statement is supported by a citation to a paper on the potential risks and benefits of neutering different breeds at different ages, which isn’t really relevant to the doubt he is casting on vaccines or “other medications.”
The reality is that there is zero evidence that recommended vaccination practices constitute “overuse” or have any negative effects on health and longevity. Antibiotic overuse is absolutely a problem in terms of antimicrobial resistance and some effects on the microbiome, but it is not at all evidence that it plays a role in reducing lifespan. And while there are complexities to the issue, the bulk of the evidence in multiple species suggests that neutering increases lifespan. This caveat is fundamentally misleading and creates unnecessary anxiety about beneficial veterinary treatments.
I am all for generating excitement about the potential for new preventive approaches to extending lifespan and healthspan. I think new therapies, from lifestyle changes to drugs and more we haven’t even considered yet, will be available and practical not too far in the future. However, we do no favors to our patients or the longevity field by stoking inappropriate anxiety or unjustified enthusiasm about current practices or those still under investigation. Dr. Richter’s book reads very much like this article, and while it is not as egregiously anti-science as some have reviewed (such as Forever Dog and Nutrigenomics), its strengths are undermined by the credulous and unhelpful “integrative medicine” perspective.
A snippet from a free webinar I co-present with Dr. Lippman and Dr. Tarantino from the Senior and Geriatric Dog Society. You can take the full course here- https://loyal.dog/course-5B-BM
What’s the Fuss About? In May of 2023, the FDA approved a new treatment for osteoarthritis (OA) pain in dogs. This was Librela (bedinvetmab). A monoclonal antibody that targets nerve growth factor (NGF), Librela was a completely new category of treatment for OA pain in dogs. This was greeted with significant excitement because, although there are many treatments used for this condition (e.g. 1, 2), all of them have their pros and cons, and it is not always possible to adequately maintain comfort and function in every dog with existing tools.
To begin with, we should consider the context. Osteoarthritis is a common and serious condition that causes significant pain and disability in aging dogs and can be a reason for euthanasia due to diminished quality of life. As mentioned, there are many therapies available for OA. Some are well-understood and there is strong scientific evidence to support their use, such as non-steroidal anti-inflammatory drugs (NSAIDs. See 3, 4, 5). The value of others is less clear (e.g. glucosamine, gabapentin, acupuncture). Other factors, such as obesity and activity level, also play a role in the management of OA.
No one treatment alone is going to be a perfect solution for all patients, and the best approach will vary with circumstances., The most effective and best studied pain relievers are NSAIDs. While these drugs do have potential risks, they are well understood and the vast majority of dogs with OA can benefit from them safely. However, there seems to be great anxiety among pet owners and even vets about NSAIDS, and an unfortunate reluctance to use them appropriately. This can lead to inadequate treatment of pain in individual dogs, or to the use of dubious or ineffective treatments. This reluctance is also part of why Librela was greeted with such enthusiasm when it was introduced. There was a hope, encouraged by the manufacturer, that this would be a safer alternative first-line treatment than NSAIDs.
Where We Started So was such hope justified by the scientific evidence when Librela was first introduced? Well, sort of.
The initial data used to gain FDA approval did show evidence of benefit. About 45% to 48% of treated dogs were judged as responding well, which was more than the improvement seen in the placebo groups (the placebo effect, contrary to what a lot of people believe, is common in veterinary patients and clinical research studies).
In a study done in Europe, only 17% of the placebo group showed an adequate response; considerably less than the 45% response in the Librela group. In a US study, however, the placebo response was about 36%, which is not nearly as different from the 48% response seen with Librela. Overall, the drug appeared to be effective based on the mechanism of action (biologic plausibility), the preclinical research in dogs, the response to similar drugs in humans, and the clinical studies done for Librela approval in several countries (Librela has been approved and used in many European countries since 2021)
The initial approval also included evaluations of safety. Studie in laboratory Beagles lasting form 2 weeks to 6 months were done at doses ranging from the usual clinical dose up to 10 times this amount, and no serious adverse effects were reported. Clinical dogs also reported few serious adverse effects compared with the placebo groups.
This is not, of course, a perfect or complete assessment of the possible risks of Librela. However, there is an inevitable compromise in the development of new medicines between the strength of the information produced to show safety and efficacy and the time, cost, and risk to study subjects needed to obtain that information.
A “perfect” study for determining safety would be to use the drug in thousands of dogs with arthritis, with all the complex mix of ages, breeds, medical conditions, and other treatments seen in the real world, for many years. However, this is impossible to achieve in most cases, due to logistics and cost, and requiring this would result in virtually no medicines ever being approved for use in veterinary patients.
The majority of drugs we have available in veterinary medicine are used those approved for humans and used “off-label” in our patients (see this post for a more detailed discussion of the drug approval process in veterinary medicine). We typically have a lot less information about the safety and effectiveness of these drugs than we originally had about Librela. So while the initial data were imperfect, it would be wrong to say that the approval was unusual or inappropriate.
Where We Are Now? So what has changed? Since the US approval, there have been a lot of anecdotal reports to the FDA about bad things happening to dogs getting Librela. These have ranged from minor to severe, and they have included a tremendous variety of events. Some of the most concerning are neurologic changes and severe physical disability leading to euthanasia. This has raised the concern that some of these events may have been caused by the Librela.
Deciding whether this is actually the case or not, however, is quite a challenge. The majority of these adverse events have happened in older dogs, many of whom have medical conditions other than OA and are getting drugs or other medical treatments besides Librela. This is a population that is at high risk for all kinds of medical problems, including those we are hearing about. Determining what, if any, role Librela has in these issues isn’t as simple as just assuming any problem a dog getting Librela has is caused by the drug.
One thing the FDA has done to try and sort this out is something called disproportionality analysis. Essentially, they do statistical analysis to see if the reported problems are more common dogs getting Librela than in dogs getting other drugs, and they attempt to compare similar dogs based on age and other factors.
This is not a clear indicator that Librela is the cause, but it can raise the level of suspicion if there are a lot more instances of a particular problem than are seen in similar dogs getting other drugs. The FDA did find reason for concern in this analysis for several of the types of problems reported, inclduing some of the most serious.
The agency also looked at whether the reported problems were more common in dogs also using some other medical treatment along with Librela. Sometimes, this was the case, and they found that about 25% of the cases evaluated were also taking the medicine gabapentin. Again, this doesn’t show that there is truly a risk to combining these drugs, but it suggests this is a question that needs to be looked into.
Not surprisingly, the manufacturer, Zoetis, has tried to reassure vets and dog owners about the safety of Librela. In an email to vets, the company pointed out that despite the apparently large number of reports of adverse events, they are actually quite uncommon relative to the number of doses of the drug they have sold (the company cannot, of course, know how many doses have actually been given to how many dogs, but the sales figures are a rough estimate of this since vets don’t tend to keep a lot of unused drugs in stock). The estimated adverse event rate is less than 10 per 10,000 dogs treated by this estimate, though there is significant uncertainty behind this number.
It is also important to remember that people who have negative experiences with any drug are a lot more likely to tell others about it than those who have positive experiences, which is one of the reasons anecdotes are not a very reliable guide to the true safety of any treatment. In online discussion groups of veterinarians, there are often many anecdotes indicating vets have used the drug frequently without seeing any problems, and there are stories of dogs saved from euthanasia when they responded to Librela after failing other OA treatments. Dueling anecdotes often leaves us with more questions than answers.
We can try to make some sense of the limited data we have by looking at the way Librela works and experiences in other species. The drug reduces the amount of nerve growth factor (NGF). This is an important chemical with many roles in the body. Most critically, it is involved in development of the nervous system, which is why drugs like this should not be used in growing or pregnant animals. However, NGF is also part of the pathway that creates pain in OA, which is why reducing it can help control pain. It may also be involved in the maintenance and repair of nerve tissue and the bone and cartilage of the joints, so adverse effects regarding nerve function or bone and cartilage health might plausibly be related to the drug. However, there is not yet sufficient evidence for us to know if and when this is the case and which patients might be at greatest risk.
There has also been a lot of concern focused on the fact that similar drugs for reducing NGF have failed to be approved or have been restricted by the FDA due to concerns about adverse effects (e.g. 6, 7). The most serious problem was the development of Rapidly Progressive Osteoarthritis (RPOA), which can involve the dramatic and devastating collapse of joints. This has mostly been seen in humans taking both NSAIDs and an anti-NGF drug at the same time. It is not clear if RPOA happens in dogs or not.
While RPOA was not reported in the initial clinical studies, these involved fairly small numbers of dogs and the subjects were not allowed to take NSAIDs during the trial. Some vets have reported joint damage consistent with RPOA in dogs taking Librela, but again these haphazard case reports are not yet sufficient to know when, if, and in which dogs this problem might occur.
As always, we have to bear in mind that humans and dogs share a lot of aspects of our biology, but there are also important differences that affect how we handle drugs. Humans can chew sugarless gum with xylitol safety while this chemical easily kills dogs. Dogs can take the antibiotic enrofloxacin routinely while humans get unpleasant hallucinations from it. The evidence in humans is worth paying attention to, and it can be a clue to what might be happening in dogs, but it is not sufficient proof in itself to make a confident judgement.
What Do I Do? The frustrating reality is that right now we don’t have a firm, final answer to how risky Librela is for each patient. The best we can do is integrate what we know with the level of uncertainty and the needs of each individual patient and make a choice now, knowing that our decision might be different in the future when there is more, better information.
Though I am involved in canine aging research, and I have even worked a bit on osteoarthritis, in my clinical practice I mostly see emergency and surgical patients, so I haven’t used much Librela myself. Many of my colleagues in the practice have used it, and anecdotally they have seen mostly positive results and few of the concerning adverse effects discussed here.
My own approach, then, is based not so much on anecdote or personal experience but on my analysis of the evidence and uncertainty at this moment. My judgement will certainly change as new information comes in, which is, after all, the whole point of evidence based medicine. For now, here’s what I tell my clients and colleagues-
Like all drugs, Librela has both risks and benefits
There is reasonable but limited evidence for beneficial effects on OA pain, though the response rates in the published studies are less than 50%, and this may include 15%-30% placebo effects. I think it is appropriate to use for OA pain with the caveats below.
While the risks appear quite low based on studies for approval, there are a lot of reports of more serious issues now that it is on the market. We don’t know for sure if these are actual problems with the drug, but they might be.
Anyone who says Librela is absolutely killing and disabling dogs in droves is going well beyond the data. Anyone who says Librela is a clearly safe, first-line OA drug with minimal side-effects is also going beyond the data. The uncertainty about the real risks and the factors that determine which dogs can safely use it and which can’t is enormous at this time. Because of this,
Librela should not be the first thing we try for OA treatment.
We know a lot more about the safety and benefits of some other treatments. NSAIDs are the best first-line drug treatment, and the fears about them are excessive, irrational, and quite likely part of the reason we may have jumped too early and enthusiastically to using Librela. Some dogs truly cannot use NSAIDs due to other conditions or medications, but I routinely see dogs who could benefit from these drugs not using them because of unreasonable anxiety on the part of vets and dog owners.
There also other therapies for OA with decent supporting evidence, including weight loss and a few supplements (e.g. fish oils, green-lipped muscle). Rehabilitation (aka physical therapy) is a plausible and promising option, though we have little data as yet about what works for what cases.
There are also lots of implausible, unproven, or useless therapies out there, from CBD to glucosamine, acupuncture to cold laser, tramadol to gabapentin. While most of these are likely to have minimal risks, they don’t have strong evidence that they work either. If we use the current concern over Librela to justify greater use of such treatments, we will very likely be leaving a lot of dogs with inadequate pain control. It is also worth remembering the basic priciple of biology that there are no benefits without risks. Or, as I like to call it, McKenzie’s Law:
In dogs that can’t use NSAIDs or other treatments with good supporting evidence, or dogs who aren’t sufficiently comfortable on these, Librela can be effective and even life-saving. I think it is worth trying in these cases, with appropriate discussion of the potential risks and the uncertainty around these, and with proper monitoring.
Librela should probably not be given along with NSAIDs.
We don’t yet know if dogs will have the same issue of rapid cartilage destruction seen in humans. Some cases that look like this condition have been reported, but it’s too soon to know if/when/how often/to whom this happens. Most of the dogs who could safely use NSAIDs probably won’t need Librela anyway, but this limitation seems prudent right now.
Some specialists recommend x-rays of joints before and regularly during use to watch for signs of cartilage destruction. This might increase safety, but it can be expensive and may need repeated sedation, so not always practical.
Librela probably shouldn’t be used in dogs with neurological symptoms, such as weakness or incoordination (aka ataxia). Partly this is because it won’t help if they don’t have pain or OA anyway, and a lot of the time hind limb weakness in of large-breed older dogs is due to neuromuscular disease rather than OA. However, this might also be a group at higher risk of adverse neurological effects based on both the reported adverse effects and the mechanism of action of the drug.
I recommend regular recheck visits to look for any problems in dogs taking Librela.
I would typically recommend and exam 1 month after first dose and then ever three months for at least a few visits, and x-rays of any joints that seem to be getting worse. This is a pretty arbitrary schedule, and other approaches are also reasonable, but this level of monitoring also isn’t a bad idea for dogs with chronic musculoskeletal issues anyway, especially if we are trialing multiple treatments.
I also recommend the use of objective tools for measuring pain and disability, such as a validated pain scale or mobility tool. Placebo effects influence owners and vets alike, and these can make ineffective treatments look like they work.
Bottom Line
Librela is likely mildly to moderately effective for OA pain, though we do not yet know which dogs will respond best and for how long.
Librela has risks as well as benefits, but we don’t really know how common these are. We also cannot predict very accurately which dogs are most likely to have adverse effects.
Librela should not be used in dogs with neurologic disease or taking NSAIDs
Librela should not be the first choice for most dogs with OA (weight management and NSAIDS are the best first-line for many dogs, and there are other options with some supporting evidence if these are not possible or sufficient).
Dogs on Librela should be closely monitored on a regular basis
Things will change. As more information is produced, we will have a better sense of how well Librela works, what the risks are, and which dogs are more or less likely to experience both benefits and risks. My hope is that we will be cautious but not hysterical in our use of this drug while more research is done. We don’t want to cause harm, but we also don’t want to abandon a potentially useful treatment without sufficient cause.
What Is It? Though I haven’t devoted a single post to the subject, I have mentioned Reiki many times and explained why it is one of the clearest examples of pseudoscientific nonsense out there. Believing Reiki has therapeutic benefits requires abandoning science and believing in magical “energy” that cannot be measured or evaluated in any way except the subject perceptions of practitioners with special abilities. In other words, magic.
Reiki is a variety of “energy medicine,” and there are many other forms of this, which differ in the theoretical explanations for their actions and in the degree of connection to particular religious traditions, including various styles of Christian faith healing, healing touch, therapeutic touch, Qigong, Tellington Touch, and others.
Reiki was invented (or discovered, depending on your point of view) by a Japanese monk named Mikao Usui in the early 20th century. Supposedly, he had been looking for a form of spiritual healing and had a vision during a long period of solitary meditation and fasting that led to his development of Reiki. Subsequent Reiki Masters have modified and develop the technique as well as spreading it to other countries.1–3
Does It Work? Like all vitalist treatments, the underlying principles of Reiki cannot be scientifically validated and so must be taken on faith. The notion that nonphysical forces which cannot be seen or scientifically studied can nevertheless be intuitively sensed and manipulated with significant impact on physical diseases is scientifically implausible.
Despite the inherent difficulty in studying a therapy based on undetectable forces, many clinical trials have been conducted on Reiki, and many more on other similar types of energy medicine. Most are methodologically flawed and subject to placebo effects and other forms of bias and error. No convincing research evidence has emerged that Reiki, therapeutic touch, or other similar practices have any objective or significant impact on physical disease in humans or in veterinary species.4–9 Several studies have suggested that despite claims to the contrary, practitioners cannot reliably sense the presence of a patient or their “energy” when effectively blinded, which begs the question of how they can be channeling this mysterious force intentionally if they cannot detect it.10
Some studies do suggest, however, that the attention and human contact associated with Reiki and similar therapies is comforting to many human patients and can lead to improvement in subjective symptoms or the patients’ experience of their illness and their medical care. There is reason to believe that many domestic animals, including dogs and cats, may also experience comfort and positive physiologic responses to human contact, so Reiki may induce these responses regardless of the existence of any putative spiritual forces.11–18 This may have some mild clinical benefits, but it is unlikely to significantly affect the outcome of serious illness.
This study does a decent job of setting up a blinded, randomized trial using standard, validated measures for assessing pain. They also compared the practice to no treatment and to a placebo. Sounds good, but there is one pretty obvious choice that undermines this methodology– the placebo intervention.
Since Reiki involves touching, or passing hands closely over, the skin of the patient, one of the potential confounders is human contact. It is well-demonstrated that gentle contact and interaction with humans has calming effects on domesticated animals, lowering heart rate, blood pressure, stress levels, and potential behavioral signs of pain.
In this study, rather than having “real” Reiki as the test treatment and the same type of interaction without the channeling of mystical energy, the investigators used a rather bizarre placebo- “physical contact from gloves connected to 50-centimeter wooden rods applied by a non-Reiki therapist individual for the same length of time and method of application as the Reiki group.”
In other words, instead of standing close and gentle touching or almost touching the dogs, the investigators touched them with a rubber glove attached to a 1 ½ foot-long stick. Hmmmmm, I wonder if this awkward and unfamiliar behavior might have been less comforting, maybe even stress inducing, compared with normal petting, magical energy or no?
In any case, the results were not impressive. Comparing pain levels within each treatment at a series of time intervals after surgery, the only statistically significant differences in the first measure (Glasgow canine pain scale short form, CMPS-SF) were that pain scores were higher than before surgery at 2, 4, 8, and 12 hours for the placebo and no-treatment groups, but only at 2, 4, and 8 hours for the Reiki group (Fig. 1). For the other measure (visual analog pain scale, VAS), the non-treatment group had higher pain levels at 2, 4, 8, 12, and 24 hours, the placebo group showed higher levels at 2, 4, 8, and 12 hours, and the Reiki group had higher scores at 2, 4, and 8 hours after surgery (Fig. 2).
Figure 1. CMPS-SF scores
Comparing between the groups, the Reiki group had lower scores on the first scale (CMPS-SF) than the placebo group (but not the non-treatment group) at 2 hours and lower scores than both at 4 hours, with no other significant differences at 8, 12, and 24 hours (Fig. 1). On the second scale (VAS), the Reiki group had lower pain scores than the other two groups at 4 and 8 hours, but not at 2, 12, or 24 hours Fig .2).
Figure 2. VAS scores.
So, the apparent differences were only seen at a couple of time points and were not the same between the two measurement scales. Much more importantly, most of the scores were below the level usually considered high enough to warrant treatment (6/24 on the CMPS-SF and 35/100 on the VAS). So, dogs with mild pain already well-treated with conventional pain meds sometimes showed lower scores after Reiki, most mostly not. Not exactly a ringing validation.
At most, this might suggest that gentle petting has a mild comforting effect on dogs after surgery compared with no contact or with fake petting using a glove on a stick. There is no need to invoke undetectable magical energy fields. And there is certainly no reason to suggest that Reiki has a meaningful effect on pain with such small and inconsistent differences in dogs with already minimal discomfort.
(Interestingly, the placebo group actually showed higher scores than the no-treatment group at several of the early time points, especially on the VAS scale, and the Reiki group seemed to have lower scores than the placebo group but not lower than the non-treatment group at one point. Could this be because the weird fake petting was actually worse than no contact at all? The study wasn’t designs or the data analyzed to test this, but it’s an intriguing alternative possibility.)
Individual dogs were given additional pain medication if they had any score on either pain scale above the threshold for treatment (the investigators used 30/100 for the VAS score, while 35 is mentioned in the article they sit for this instrument. There is no consensus on the “right” cutoff for this tool.) About half the dogs in the placebo and no-treatment groups received this extra medication, and none of the dogs in the Reiki group did. That would seem to be a pretty dramatic difference!
However, this is not consistent with the much smaller and less consistent differences between the groups in pain scale results. Such a dramatic finding is suspicious for a failure to adequately blind the people making the assessment, though of course there is no way to evaluate this.
If such a difference is real, it would mean Reiki is a pretty powerful pain treatment. Given the implausibility of the premise behind the treatment, and the failure to find such dramatic effects in many other studies in other species, an error in the study seems a lot more likely. Such an unusual and dramatic finding would need to be replicated by independent researchers before it could be taken more seriously.
Almost none of the objective physiologic measurements different between the groups (Table 1), reinforcing the concern that bias may have affected the findings of more subjective pain measurements. This is actually a bit of a surprise since human contact has been previously shown to influence heart rate and blood pressure. Such inconsistencies weaken the argument that there is a clear and real pattern of effect here.
Table 1. Physiologic measurements.
Is It Safe? No direct harm has been seen in studies of Reiki. The major risk of Reiki and other forms of energy healing is the delay or rejection of appropriate diagnosis and effective treatment that may occur if people mistakenly believe these therapies offer real hope of improving their pets’ condition.
Bottom Line The idea that a magical energy force detectable only by specially trained people (who have been shown, in controlled studies, not to be able to detect this force after all) can have meaningful health benefits is highly implausible. Much of well-established scientific understanding would need to be totally wrong for this idea to be true. The evidence is generally poor quality, and consistent benefits have not been reliably demonstrated. Some non-specific effects from gentle, pleasant physical contact may be present, but we and our veterinary patients can enjoy that without the mystical nonsense and the false claims of important objective health effects.
Selected References
1. Singh S, Ernst E (Edzard). Trick or Treatment??: Alternative Medicine on Trial. London: Corgi; 2009. https://www.worldcat.org/title/trick-or-treatment-alternative-medicine-on-trial/oclc/788887954&referer=brief_results. Accessed November 10, 2018.
2. Carroll RT. The Skeptic’s Dictionary?: A Collection of Strange Beliefs, Amusing Deceptions, and Dangerous Delusions. Wiley; 2003.
4. Robinson J, Biley FC, Dolk H. Therapeutic touch for anxiety disorders. Cochrane Database Syst Rev. 2007;(3):CD006240. doi:10.1002/14651858.CD006240.pub2
5. O’Mathúna DP. Therapeutic touch for healing acute wounds. In: O’Mathúna DP, ed. Cochrane Database of Systematic Reviews. Chichester, UK: John Wiley & Sons, Ltd; 2016:CD002766. doi:10.1002/14651858.CD002766.pub5
6. vanderVaart S, Gijsen VMGJ, de Wildt SN, Koren G. A Systematic Review of the Therapeutic Effects of Reiki. J Altern Complement Med. 2009;15(11):1157-1169. doi:10.1089/acm.2009.0036
7. Lee MS, Pittler MH, Ernst E. Effects of Reiki in clinical practice: a systematic review of randomised clinical trials. Int J Clin Pract. 2008;62(6):947-954. doi:10.1111/j.1742-1241.2008.01729.x
8. Ferraz GAR, Rodrigues MRK, Lima SAM, et al. Is Reiki or prayer effective in relieving pain during hospitalization for cesarean? A systematic review and meta-analysis of randomized controlled trials. Sao Paulo Med J. 2017;135(2):123-132. doi:10.1590/1516-3180.2016.0267031116
9. Joyce J, Herbison GP. Reiki for depression and anxiety. Cochrane Database Syst Rev. 2015;(4):CD006833. doi:10.1002/14651858.CD006833.pub2
10. Rosa L, Rosa E, Sarner L, Barrett S. A Close Look at Therapeutic Touch. JAMA. 1998;279(13):1005. doi:10.1001/jama.279.13.1005
11. Shahin M. The effects of positive human contact by tactile stimulation on dairy cows with different personalities. Appl Anim Behav Sci. 2018;204:23-28. doi:10.1016/J.APPLANIM.2018.04.004
12. Beetz A, Uvnäs-Moberg K, Julius H, Kotrschal K. Psychosocial and psychophysiological effects of human-animal interactions: the possible role of oxytocin. Front Psychol. 2012;3:234. doi:10.3389/fpsyg.2012.00234
13. Vormbrock JK, Grossberg JM. Cardiovascular effects of human-pet dog interactions. J Behav Med. 1988;11(5):509-517. http://www.ncbi.nlm.nih.gov/pubmed/3236382. Accessed November 17, 2018.
14. Dudley ES, Schiml PA, Hennessy MB. Effects of repeated petting sessions on leukocyte counts, intestinal parasite prevalence, and plasma cortisol concentration of dogs housed in a county animal shelter. J Am Vet Med Assoc. 2015;247(11):1289-1298. doi:10.2460/javma.247.11.1289
15. Mariti C, Carlone B, Protti M, Diverio S, Gazzano A. Effects of petting before a brief separation from the owner on dog behavior and physiology: A pilot study. J Vet Behav. 2018;27:41-46. doi:10.1016/j.jveb.2018.07.003
16. Shiverdecker MD, Schiml PA, Hennessy MB. Human interaction moderates plasma cortisol and behavioral responses of dogs to shelter housing. Physiol Behav. 2013;109:75-79. doi:10.1016/j.physbeh.2012.12.002
17. HAMA H, YOGO M, MATSUYAMA Y. Effects of stroking horses on both humans’ and horses’ heart rate responses. Jpn Psychol Res. 1996;38(2):66-73. doi:10.1111/j.1468-5884.1996.tb00009.x
18. Lynch JJ, Fregin GF, Mackie JB, Monroe RR. Heart rate changes in the horse to human contact. Psychophysiology. 1974;11(4):472-478. http://www.ncbi.nlm.nih.gov/pubmed/4852234. Accessed November 17, 2018.
I recently gave a lecture at a continuing education conference looking at how the regulation of veterinary treatments works, and how vets can use this process to help them make evidence-based treatment recommendations. Understanding what is required to claim a treatment is “FDA approved,” and what it means when a product is not, is also useful for pet owners considering their options.
Making Evidence-based Treatment Choices The core of evidence-based medicine is integrating the best current scientific evidence with our own clinical expertise and the needs of the patient and the client when making diagnostic and treatment recommendations.1 Accurately understanding the potential risks and benefits of a treatment is necessary for us to guide clients in making choices about their pets’ care.
However, it is impossible for each of us to be an expert on all the treatments available to choose from, and the time needed to keep current with the complex and expansive body of research evidence can be prohibitive. Clinicians frequently use shortcuts or proxies to help us understand how much confidence we can have in a given therapy and how much scientific uncertainty there is about it.
One of these proxies is the degree of regulatory evaluation and oversight applied to a product. There are different general categories of medicines available for vets and pet owners to choose from, and understanding the differences between these categories, including the evidence available supporting the safety and efficacy of each, can be a useful aid in evaluating our treatment choices.
A relatively new category of regulatory approval is Expanded Conditional Approval (XCA).2 In the last few years, some useful new drugs have become available to veterinarians through this pathway, such as Laverdia and Panoquel. Some not-so-new drugs have also been approved under this process, such as Vetmedin and potassium bromide. But what is conditional approval? How is it different from “regular” FDA approval?
This session will review the broad general categories of treatments available (in terms of regulatory oversight), with special focus on XCA and how vets can explain the significance of these categories to clients.
What Are Our Options? In general, vets can choose among medicines in several broad categories:
Prescription drugs fully approved under the supervision of the Food and Drug Administration’s Center for Veterinary Medicine (the FDA CVM)*
Prescription drugs conditionally approved under a formal FDA process
Drugs approved for use in humans or other species and used “off label”† in dogs
Dietary supplements or other non-prescription treatments not tested or approved under a specific regulatory process
*Some veterinary products are regulated by other agencies, such as certain parasite preventatives (EPA) and vaccines (USDA), but FDA is responsible for approving most prescription drugs.
†”off label” is a common term vets use to describe what the FDA more often calls “extra-label” drug use. I explain the details of what this means a bit later.
There is also a category of fully approved drugs that can be labeled for use without the supervision of a veterinarian.13These over-the-counter (OTC) drugs have the same approval requirements as prescription products, but the manufacturer has demonstrated to the FDA that the drug can be safely and effectively used without supervision by a veterinarian, relying only on the directions on the label, so a prescription from a vet is not required. (This is slightly different from the process for approving OTC drugs for human use, which does not always require a regular drug approval process.)14
In general, the strength of the evidence identifying both the benefits and the risks of these treatments in the target species is greater the closer to the top of the list they are. For example, it takes a lot more research and data in dogs to achieve FDA approval for a prescription medication to be used in dogs than is required for a human drug used off-label or for a dietary supplement.
The FDA has legal authority to set the conditions for the use of any “drug,” which the agency defines as anything “intended for use in the diagnosis, cure, mitigation, treatment, or prevention of disease” and anything “(other than food) intended to affect the structure or any function of the body.”3 The agency regulates drugs for both humans and other animals, but the guidelines are often different depending on the species a product will be used in.
The categories of available treatments reflect the need for a compromise between the best possible evidence and the availability of treatments. Ideally, every medicine would go through the rigorous testing required for a fully approved prescription drug, or even more than that! However, this is a lengthy and expensive process, and a strict requirement to have this level of evidence for everything would drastically limit the treatment options available for vets and dog owners. Therefore, the FDA allows vets to use products other than approved prescription drugs. The agency also sets specific guidelines for each category to minimize the potential risk associated with the greater uncertainty in categories that require less extensive pre-approval testing.
This is helpful to veterinarians because it gives us more treatment options, but it also requires us to understand, and explain clearly to clients, the differences between these categories and what this means for how confident we can be in the safety and effectiveness of different types of treatments.
Fully Approved Prescription Drugs The highest bar for scientific evidence is set by full FDA drug approval. This is a complex process that requires demonstrating to the agency that a drug proposed for use in dogs is safe, for the dogs taking it as well as for the humans around them, and that it is effective, meaning it consistently does what it is expected to do. Full approval requires showing that a drug can be consistently made to a high standard of quality. The FDA also has specific requirements for labeling the drug, for disclosing the evidence produced to support approval, and for what a company can and cannot claim about the drug once it is approved.
The major components of FDA approval for animal drugs are called the major technical sections. These include:3
Target Animal Safety (TAS)
Effectiveness
Human Food Safety
Chemistry, Manufacturing, and Controls
Environmental Impact
Of greatest interest to clinicians are the TAS and effectiveness sections because these relate directly to our assessment of the potential risks and benefits of a drug for our patients. The TAS typically involves testing the drug at both the intended label dose and higher doses in healthy animals of the species for which the drug is intended. This helps establish the potential side effects of a drug and gives a sense of the “safety margin” or possible consequences of exposure to doses higher than the label dose or of using the drug in animals that are not healthy or may be especially sensitive in some way.
Effectiveness is often determined in a field study, which is usually a randomized, blinded, controlled clinical trial of the drug in individuals of the target species with the condition the drug is meant to treat. Such trials are conducted with oversight by the FDA, including review of the methodology before the study is run and review of the data after it is completed. This oversight helps reduce the influence of bias on the results.
All research studies involve bias, processes which systematically influence the design, conduct, or analysis of a trial in ways that alter the outcome. It is well-known, for example, that studies funded by industry are more likely to produce results favorable to the study sponsors than independently funded research.4 This is sometimes misunderstood as meaning that companies or researchers deliberately “fudge” the data to get the results they want, but that is rarely how bias works.
Most of the time, bias in research takes the form of small, unconscious effects on the numerous decisions that must be made at every step of designing and conducting a study and analyzing the data. Such processes are just as problematic when they involve ideology (e.g. homeopaths rarely conduct studies that show homeopathy doesn’t work), career advancement, or factors other than financial interests.5
Such biases occur in every study, and the classic research methods of randomization, blinding, and placebo controls are a few of many scientific methods that help minimize the influence bias has on the outcomes of research studies. Independent regulatory oversight, as is part of the FDA drug approval process, is another useful tool in minimizing the influence of bias on study results, and it allows us to have greater confidence in the findings of such clinical trials.
The FDA also requires companies marketing prescription drugs to monitor the experience that patients and vets have using the drug and submit regular reports after the drug comes on the market. This further strengthens our ability to identify and fix any problems once a product is in regular clinical use.
Of course, no scientific or regulatory process is ever perfect. Biology is complicated, so the effects of drugs are also complicated and never completely predictable. However, the standard of FDA approval is rigorous, and an approved prescription drug has been tested extensively and has more scientific evidence to support safety and efficacy than other categories of treatments available to us.
Expanded Conditional Approval- XCA Established in 2018, the expanded conditional approval process is a relatively new regulatory category which recognizes that for conditions that are serious or life-threatening and for which there is no existing effective treatments, a quicker process that makes new drugs available for vets to use sooner is beneficial.1,6 The clinical studies needed to demonstrate effectiveness under the full approval process can be lengthy and complex for some kinds of drugs, and while these studies are necessary, patients may go for years without treatment while waiting for such research to be completed.
XCA is available for drugs intended for use in the seven designated “major species” (horses, dogs, cats, cattle, pigs, turkeys, and chickens) if they meet two core criteria:
They are intended to address a serious or life-threatening condition, but demonstrating effectiveness would require complex or particularly difficult studies
OR
They address an unmet animal or human health need, but demonstrating effectiveness would require complex or particularly difficult studies
XCA requires most of the same evidence needed for full approval, including all of the major technical sections except for the usual effectiveness package. Therefore, confidence regarding safety and manufacturing quality can be just as high as for fully approved products. Evidence is still needed to establish effectiveness, but the specific standard is different.
Under XCA, rather than the “substantial evidence” for effectiveness standard, which typically requires a clinical trial in the target species, expanded conditional approval requires a “reasonable expectation of effectiveness” (RXE). This standard can potentially be met with studies that involve fewer dogs and shorter time periods than the clinical trials needed for full approval or with evidence already existing in the scientific literature from studies done previously. This shortens the time necessary to achieve approval which should ideally make it possible to bring needed products into use faster.
The XCA process is not intended to permanently replace the standard approval requirements. Unlike full approval, conditional approval is limited to one year. It is possible to renew XCA annually for up to five years so long as the company marketing the drug is making progress in generating the evidence required to meet the substantial evidence standard. Ultimately, the drug must be given full approval once this evidence is provided and accepted by the agency, or it must be withdrawn from the market.
A number of novel drugs have been approved under the XCA process, such as Laverdia (verdinexor) for treatment of lymphoma in dogs7 and Panoquel (fuzapladib) for management of canine pancreatitis symptoms.8 However, some drugs that have already been in widespread use under a different category, such as off-label use of products approved for humans or other species, have also been given XCA for specific uses in dogs.
Vetmedin (pimobendan) was fully approved for treatment of congestive heart failure (CHF) due to mitral valve disease (MMVD) or dilated cardiomyopathy in dogs in 2007 (and the generic version was approved for this indication in 2024).9,10 However, Vetmedin was approved conditionally under XCA in 2022 for use in ACVIM Stage B2 MMVD cases to delay the onset of congestive heart failure.11 The drug is the same, but the indication is different, and it is approved for CHF and pre-clinical MMVD under two different regulatory processes.
There are some differences in the rules for use of drugs that are fully approved versus those approved under XCA. The most significant is that vets cannot use a conditionally approved drug off-label, meaning for any purpose or in any way other than what is dictated by the label directions. For example, if a drug is only conditionally approved to treat staphylococcal skin infections in adult dogs, it would not be legal to use the drug to treat a urinary tract infection, a puppy, or a cat.
This restriction applies even if there is reason to believe the drug might be effective for these other uses, such as published research literature or anecdotal reports of such uses. This is a major difference from fully approved drugs, which sometimes can be used off label. Fortunately, it is easy for vets to recognize a conditionally approved drug because these will have the designation “CA” as part of the name (CA1 for the first conditionally approved use, CA2 for the second, and so on).
The XCA process allows needed treatments to be available to patients more quickly but also ensures a high standard of evidence. Vets using conditionally approved drugs can be confident that the evidence for safety is as robust as for any other drug, and the evidence for effectiveness is strong, though not as extensive as for a fully approved drug. Understanding this, vets can counsel their clients about the differences between full and conditional approval so owners can make informed choices about the risks and benefits of these treatments.
Off-label Use of Drugs Approved in Humans or Other Species Many more drugs are approved and available for use in humans than in dogs, obviously because the biomedical industry for humans is much larger and better funded than veterinary medicine. The FDA recognizes that many potentially beneficial drugs are available in people, or other species, that will never be put through the approval process for use in dogs due to limitations of time, money, or other factors. Since 1994, the agency has allowed vets to use these drugs under specific conditions, and this is known as “extra-label” use (more commonly called “off-label”).12
Because the testing required for approval of a drug used off label was not specifically meant for the species or indication a clinician intends to treat, the approval data are less reliable indicators of how safe and effective the drug will be for this new use. As we all know, extrapolation across species is a necessary evil in veterinary medicine, but there are plenty of cases in which a drug that is safe and effective in one context may be neither in another. Tylenol is safe enough to be designated an over-the-counter (OTC) medication in humans, but it must be used with great caution in dogs and never in cats!
In recognition that the standard of evidence supporting off-label use of approved drugs is lower than for drugs fully or conditionally approved for the target species and indication, the FDA has set some rules for extra-label drug use.
Veterinarians must, of course, have an established veterinarian-client-patient relationship (VCPR) in order to prescribe a drug off-label, just as they should for any medical treatment. The FDA provides a set of criteria for the existence of a valid VCPR:
you have recently seen and are personally acquainted with the keeping and care of the animal by virtue of examination or, for food animals, a visit to the place where the animals are housed
you have assumed responsibility for making medical judgments about the health of an animal, and the owner has agreed to follow your instructions
you have sufficient knowledge of the animal to form at least a general or preliminary diagnosis of the medical condition
you are readily available for follow-up in case of adverse reactions or treatment failure.
In addition to establishing a valid VCPR, vets can only use an approved drug off label when an animal’s health is threatened or where the animal may suffer or die without treatment.
Off-label use of such treatments for cosmetic or performance enhancement purposes would not be allowed. Because the confidence we have in the safety and effectiveness of drugs used off-label is lower, the risks are considered greater, and thus there must be a meaningful need to justify taking those risks.
Vets also cannot use a drug off-label when there is an equivalent drug already approved for dogs, unless they determine that the approved drug is not appropriate or effective for a specific patient. Exactly what is necessary to determine this is not spelled out, so there is some uncertainty and room for discretion here. There are also detailed requirements for record-keeping, labeling, and compounding of drugs used off label.
Extra-label use of drugs approved for humans is one of the most common types of medication use in veterinary medicine. We understand that the evidence for safety and effectiveness in our patients is not as strong as it is for approved animal drugs, but these drugs are an indispensable clinical tool. As long as we, and our clients, understand the limits of the evidence, we can balance the potential risks and benefits, and the limits of the evidence, in a rational way and monitor patients appropriately.
Supplements Dietary supplements are the category with the least oversight and the greatest uncertainty about safety and efficacy. In veterinary medicine, this is not even an official category. As the FDA explains, “there is no ‘dietary supplement’ regulatory classification for animal food substances and products.”15 The law that regulates dietary supplements for humans, the Dietary Supplement Health and Education Act of 1994 (DSHEA) has been ruled not to apply to veterinary species. Products marketed for these species “are considered either ‘foods’ or ‘new animal drugs’ depending on the intended use.”15
Essentially, anything that is intended to cure, treat, or prevent disease is considered a drug and should, in principle, go through the process for full or conditional approval before being marketed for use in companion animals. Some health claims are allowed for therapeutic diets, if they can be supported with evidence, but these are limited. Despite complex legal language and some degree of uncertainty, the bottom line is that food is food and medicine is medicine. If something is meant to be used as medicine, then it is a drug and not food. Foods don’t need to be approved by the agency before being marketed, but drugs do.
In practice, the FDA does not aggressively enforce this policy, and it typically does not insist on removal of supplements marketed for pets even if they have not been through the required drug approval process. In the past, the agency had an articulated policy indicating it would not object to marketing of pet supplements so long as they did not claim to treat or prevent disease or make other inaccurate claims and so long as there was no specific evidence suggesting they were harmful.16 That policy was withdrawn in early 2020, so exactly how the agency views regulating these products at this point isn’t entirely clear.17
The importance of this for veterinarians is that we must be aware that supplements for pets aren’t required to prove they are safe and effective before, or after, being marketed. Most of these products have never been tested in clinical trials, and the few studies that are done are typically set up and run by the company selling the product without any independent supervision. These studies can be interesting, but not surprisingly they often come up with results favorable to the company, and without the independent FDA monitoring, it is harder to be confident that these results are objective and unbiased.
There is also no official regulatory monitoring of quality for most dietary supplements. One study found, for example, that the labels for many probiotic supplements were inaccurate—often the products didn’t have the species or amount of probiotic organism claimed on the label.18 Similar results have been found for tests of other supplements.16
Even if a specific supplement might have therapeutic effects, your patients won’t benefit from these if the product you use doesn’t actually contain the ingredients it’s supposed to have! There are voluntary industry groups that make some effort to establish guidelines for supplement quality, but these are not mandatory nor independent of the manufacturers.
There may be research evidence relevant to determining the safety and efficacy of specific dietary supplements intended for companion animal species. However, the research is often sparse and rife with potential bias and error, and independently evaluating it for every product is impractical for most veterinarians. The absence of a robust regulatory process for pre-market approval of such products leaves the veterinarian with less guidance and less confidence in these treatments than we can have in those that are subject to more regulatory oversight.
Bottom Line Science is imperfect, and medicine is an imperfect science. No medical treatment, whether a prescription drug or a dietary supplement, is ever completely effective or completely safe for every patient. There is always some level of risk in doing anything in medicine, and this has to be balanced against the potential benefits of the treatment in the context of the degree of evidentiary uncertainty.
As veterinarians, we can only decide what is best for our patients on the basis of the best evidence we have, integrated with our clinical expertise and the needs and goals of our clients. Uncertainty is inevitable, but the more evidence we have, the less uncertainty there is and the easier it is to make confident recommendations. Understanding the different levels of regulatory oversight for various categories of treatments can help us assess the strength of the scientific evidence and the degree of confidence we can have in the medicines we recommend.
References
Cockroft, P. Holmes, M. (2003). Handbook of Evidence-Based Veterinary Medicine. Oxford: Blackwell.
U.S. Department of Health and Human Services Food and Drug Administration Center for Veterinary Medicine (CVM). Eligibility Criteria for Expanded Conditional Approval of New Animal Drugs Guidance for Industry. 2021. Accessed on July 18, 2024. Available at: https://www.fda.gov/media/130706/download
U.S. Department of Health and Human Services Food and Drug Administration Center for Veterinary Medicine (CVM). From an Idea to the Marketplace: The Journey of an Animal Drug through the Approval Process. Accessed on July 18,. 2024. Available at: https://www.fda.gov/animal-veterinary/animal-health-literacy/idea-marketplace-journey-animal-drug-through-approval-process#approved
Resnik DB, Elliott KC. Taking financial relationships into account when assessing research. Account Res. 2013; 20(3): 184-205.
Simmons JP, Nelson LD, Simonsohn U. False-positive psychology: undisclosed flexibility in data collection and analysis allows presenting anything as significant. Psychol Sci. 2011; 22(11): 1359-1366.
U.S. Department of Health and Human Services Food and Drug Administration Center for Veterinary Medicine (CVM). Oral Dosage Form New Animal Drugs; Pimobendan. Fed Reg. 2007; 72(95):27733.
U.S. Department of Health and Human Services Food and Drug Administration Center for Veterinary Medicine (CVM). Frequently Asked Questions about Animal Drugs: What are the differences between an over-the-counter animal drug, a prescription animal drug, and a veterinary feed directive animal drug? Accessed July 18, 2024. Available at: https://www.fda.gov/animal-veterinary/safety-health/frequently-asked-questions-about-animal-drugs#difference
Screening One of the topics I have written about frequently here is screening—testing apparently healthy individuals to look for disease that hasn’t yet cause clinical symptoms. Screening is popular in human and veterinary medicine because of the widespread belief that the earlier we detect disease, the more effectively we can treat it, and that a proactive approach is better for patients than waiting for diseases to show up with symptoms. Sometimes this is true. And sometimes it isn’t.
Screening is a complicated subject because it is often hard to know in general whether or not we are really benefitting patients, and it is nearly impossible to know for sure whether screening has helped or harmed any specific individual.
There are some situations in which it is pretty obvious that screening would be of no benefit-
When the test is wrong and the patient doesn’t actually have the disease it says they have (a false positive result or misdiagnosis)
When the disease is there, but it is not progressive and will never cause any symptoms or illness (overdiagnosis)
When we don’t have any effective treatments, so knowing that a disease exists before symptoms appear just gives us more time to worry but no way to help the patient
The problem is that we often don’t know if a patient is misdiagnosed or overdiagnosed. Sometimes, an initial test is positive and follow-up testing shows they didn’t have the disease after all. But we don’t always get to do this testing, and we may treat patients, or even euthanize them, without discovering that our first test was wrong. And once we find a disease and treat it, we have no way to know if it was ever going to make the patient sick, so we usually assume we’ve helped the patient even when we haven’t.
We can often determine how likely misdiagnosis and overdiagnosis are in general by looking at statistics for groups of patients. (here’s a paper I wrote with more details about how this works) However, even this is sometimes difficult in veterinary medicine, where we don’t always have the information we need. This figure shows some of the important measures of how useful a test might be.
We can usually determine the sensitivity (how many of patients who have the disease correctly test positive for it) and specificity (how many patients without the disease correctly test negative for it) of a test by applying it to patients we already know have, or don’t have, the disease we are testing for. This is usually done when new tests are developed.
However, these values are not as useful as they seem in the real world. What we really want to know is the predictive value of a test. This is how many of the patients who test positive or negative actually have, or don’t have, the disease. This depends on how common the disease is in the group of patients we test.
If, for example, only 1% of dogs have a certain cancer and I test 100 dogs with a test that is 95% sensitive and 95% specific (pretty good numbers), only 16 of the dogs who test positive will actually have the disease. That means that 84% of the positive tests are wrong! You can imagine the potential harm we could do if we mistakenly told 84% of our patients they had cancer when they didn’t!
When a disease is rare, then negative tests are pretty likely to be true (99.95% in the example I just gave), but positive tests aren’t. Conversely, with common diseases, positive tests are a lot more likely to be right. Unfortunately, we don’t always know how common a disease is in the general population, so it can be hard to know how much to trust our test results.
So despite the fact that a lot of vets recommend very frequent and comprehensive testing, especially in older animals, we have to be careful about both when we do this and also how we use the results. An example I was asked about recently is the cancer screening test NuQ
NuQ I have written about another cancer screening test, OncoK9, several times (1,2). This test was announced to much fanfare, and I was initially quite skeptical about its value. The company, to its credit, did pursue ongoing testing to improve the evidence base, and I did come to believe the test might be useful as an aid in diagnosis of suspected cancer, though I still felt we never had adequate evidence to recommend routine screening of health animals, as the manufacturer suggested. However, the company producing this test went out of business, and it is no longer available.
NuQ is another purported cancer screening test which is getting more attention recently, so I wanted to take a look at the evidence behind it. A paper published in 2022 reported the basic data, including the sensitivity and specificity of this test overall and for a few specific cancers. In general, the manufacturer reports a sensitivity of about 50% overall, and a specificity of 97%. The sensitivity is a bit higher for some cancers (e.g. 81% for hemangiosarcoma), and lower for others (e.g. 19% for mast cell tumors). This, of course, represents the numbers generated when testing dogs who are already know to have or not have cancer, and the values might well be different in another population based on age, breed, and other factors.
To figure out the predictive value of the test, we need to know the prevalence of cancer in the population we are testing. We seldom have this data, but a couple of papers have given us a general range of estimates we can use. A study earlier this year found cancer in about 3%-6% of apparently healthy dogs tested, and this is sufficiently in line with previous estimates as to be a reasonable ballpark.
Using these numbers, the overall positive predictive value of NuQ for cancer ranges from 34%-51.5% and the negative predictive value ranges from 96.8%-98.4%. Another way of thinking about this, is that of the apparently health dogs who test positive, as many as 75% of them might not actually have cancer (even with the higher 6% prevalence, 49% of them will falsely test positive for cancer).
If we look at the best-case performance of the test, such as with hemangiosarcoma, assuming. Prevalence of 6% and a sensitivity of 81.8%, the positive predictive value is still only 63.5%. That means we will tell about 64% of the owners whose dogs have a positive test that they have a cancer we can do little to treat, and we will give this grim diagnosis falsely to about 37% of these owners.
Of course, presumably vets will follow up on these results with imaging or other tests rather than treating or euthanizing patients on this basis alone. But this still means a significant cost for owners and, potentially, invasive tests that can cause pain or injury, when the chances of finding cancer are low. And for owners who cannot afford such testing, or treatment for cancer, we have given them terrible news they can do nothing about even if it’s true. Determining whether or not this is a good idea for a given patient is something that requires some careful thought and extensive, clear discussions with owners.
It is also important to remember that for most cancers we have no idea if earlier detection leads to longer, healthier lives for patients or not. It might, but this hasn’t always been true for human cancer patients, and it is quite possible we could harm more patients than we help by treating cancers that would never have caused illness or death. And there are some cancers for which we have no meaningfully effective treatments at all, so diagnosing these earlier is clearly not going to improve quality or length of life in these patients.
In general, I think vets and pet owners are entirely too optimistic about the value of screening given both how little evidence we have to show it helps in veterinary patients and the numerous examples of how it can cause harm in human medicine. For NuQ in particular, I find it difficult to picture many situations in which the likely benefits of running the test outweigh the uncertainties or potential harms.
Perhaps in a population at high risk of a treatable form of cancer, on age and breed or family history, we might get a useful early warning, but even then it isn’t yet clear that earlier diagnosis and treatment will make a meaningful difference. Certainly, routine screening of all healthy senior dogs does not seem justified based on the information we have now.