Kaleb, I hope you’ll agree, is a handsome beast. In his youth, he cut an athletic figure, and was quite the wanderer. Indeed, without his lust for independent travel, he’d never have come into our lives. Having roamed once too often from owners who showed little interest in taking him back, he ended up in a rescue shelter in Ithaca, New York, and was adopted in 2005 by my girlfriend, Nadia.
Now in his twilight years, Kaleb doesn’t get around so well. He’s part German Shepherd, and is afflicted by the breed’s curse: hip dysplasia and osteoarthritis. In plain English: his hip joints are loose, which leads to cartilage damage and inflammation. It’s painful to watch him struggling to his feet, and while he still loves to go for a walk, his back legs start to give out before too long. But at least he’s getting the best possible veterinary care, we told ourselves.
At least we did, until an email hit my inbox some weeks back, sent to a discussion list of science writers. It referred to a study indicating that two food supplements—glucosamine and chondroitin—do little to help cats with disintegrating joints. I recognized the names as ingredients in Kaleb’s breakfast: Our dog-feeding ritual involves taking a chew containing these nutrients, then adding a dollop of peanut butter containing a couple of pills of a painkiller called tramadol.
Chondroitin is an important component of cartilage, and glucosamine a potential building block for its repair, so it makes sense that they might help his aching joints. But I’ll take hard facts over intuition any day, so I forwarded the email to Nadia and started to look at what studies on dogs with osteoarthritis have to say.
What I found was eye-opening. There’s scant evidence that either the supplements or the painkillers are doing much to ease Kaleb’s suffering. There is a treatment that clearly could do some good: a non-steroidal anti-inflammatory drug, or NSAID. But we’d rejected that after discussion with his vet a year or so ago because of fears—possibly overblown—that it might damage his kidneys.
If you have a pet, this should be a cautionary tale. Americans spent $14.2 billion on veterinary care for their pets in 2013—and that doesn’t include proprietary health diets and food supplements. Put another way, pet owners pay about $850 annually in veterinary expenses per dog, and about $575 per cat. Factor in the emotional energy we invest in keeping our companion animals healthy, and you’d hope for high confidence in the end results. But as I’ve learned, much of veterinary medicine is based on shaky scientific foundations: The drugs prescribed for your dog or cat may work no better than those we’ve been giving to Kaleb.
Before you get angry, realize that mostly this isn’t your vet’s fault. The biggest problem is that their medicine cabinets are relatively bare. Like it or not, most of what we know about whether drugs work and are safe comes from clinical trials conducted by pharmaceutical companies to win marketing approval. Even though the sums we spend on our pets’ health may seem lavish, they’re a fraction of the budgets involved in human medicine, making it hard for companies to justify the costs of developing new veterinary drugs. That’s why the U.S. Food and Drug Administration’s database of approved human drugs contains more then 6,500 entries, and the list for dogs fewer than 650.
If it’s on the approved list for animals, you can be reasonably confident that a drug does what it says on the label. Most worming tablets, for example, have been well tested for their ability to clear parasites from dogs or cats. But things can get murky with many commonly prescribed drugs, including antibiotics and painkillers, which have not specifically been approved for use in animals and where practice is based on extrapolations from human medicine—which may or may not be relevant to creatures with subtly different physiology, prone to different diseases.
Still, vets could make better use of the available scientific knowledge. Today, most doctors treating human patients accept the principles of evidence-based medicine, where best practice is based on data from multiple scientific studies. But many vets are reluctant to jump on that bandwagon, arguing that there’s not enough data on animals to justify this approach. “A lot of vets think that it will undermine client confidence,” says Brennen McKenzie, president of the Evidence-Based Veterinary Medicine Association and a vet at the Adobe Animal Hospital in Los Altos, California.
Think about what he’s saying: Some vets are reluctant to delve into what science has to say, out of fear that they’ll have to admit that they don’t know for sure how to make our pets well. A comment added to one of McKenzie’s blog posts, from a vet who had learned that glucosamine does little for osteoarthritis, underlines the point. “I can tell you it was hard for me to stop selling the stuff,” the vet wrote. “I was making money, the clients thought it was working … and I did not want to fess up and tell them they had bought something from me that was a waste of money.”
My own journey of discovery about Kaleb’s treatment began at a website called BestBETs for Vets, where the Centre for Evidence-based Veterinary Medicine at the University of Nottingham helps vets to ask the right questions—and shows them how to find answers in the scientific literature. Its examples include one relevant to Kaleb, considering the effectiveness of glucosamine and chrondroitin versus an NSAID called carprofen in treating dogs with osteoarthritis. The bottom line: “Carprofen is superior to glucosamine/chrondroitin supplements in reducing the clinical signs.”
So are Kaleb’s supplements doing anything at all? He started taking glucosamine and chrondroitin in 2007, advised by vets in Ithaca who hoped that they might help stave off joint damage. But now that he’s already arthritic, there’s little evidence that they will help, according to a recent systematic review of available studies. (In this case, human medicine provided a good guide to the likely effects in dogs: A huge clinical trial concluded in 2006 that the supplements don’t reduce arthritic knee pain.) If we want to work on Kaleb’s diet, the same review suggested, we might try formulas rich in fish oils, which have promising results in placebo-controlled trials. (And if you’re wondering why placebo controls are needed in veterinary studies, read the note to the right.)
So much for glucosamine and chondroitin. Now I needed to find out about tramadol, the painkiller that we add to Kaleb’s breakfast. I turned to Steve Budsberg of the University of Georgia, who specializes in canine osteoarthritis. “That’s too bad,” he responded, when I told him that Kaleb was taking the drug. “I think it just gets the dogs high.”
Digging into the scientific literature, I learned why Budsberg is skeptical. Tramadol is an opioid—essentially a synthetic version of morphine—and its painkilling effects in people depend largely on its conversion in the body to a substance called M1. But dogs don’t seem to convert tramadol to M1 as well as humans. I found just one controlled trial comparing carpofren and tramadol to treat dogs with osteoarthritis. The drugs were given for only a couple of weeks, and the main conclusions were that placebo effects are large, and that findings vary depending on how you measure a dog’s symptoms.
Why is tramadol widely prescribed to dogs with Kaleb’s condition, when the best evidence indicates that NSAIDs like carpofren are the most effective option? Fear of liver and kidney damage, two known dangers of NSAIDs, seems to be the main reason. But Budsberg believes this concern is overplayed, and worries that the vogue for tramadol has achieved little apart from reassuring vets and dog owners that they aren’t risking side effects. “They’re treating themselves,” he says.
As you can imagine, Nadia and I aren’t feeling so good about ourselves right now. Each of us has a PhD in biology, and yet we’d failed to ask all of the right questions about Kaleb’s treatment. We plan to get some fresh tests to see how stable his kidney function is, and talk to our current vet in San Francisco about whether it’s time to try carprofen. (Warning: asking more questions may mean spending more money.)
I’m pleased that we’re seeking better answers for Kaleb, but the big question remains: Why are vets recommending treatments that probably don’t work? One explanation is what psychologists call confirmation bias: Once we get an idea into our heads, we tend to pay attention to information that supports it, and dismiss facts that don’t. Vets aren’t immune, so it’s easy to see how initial positive experiences with a drug could color their judgment. They’re especially likely to be fooled into overestimating a drug’s impact on a condition like arthritis, which can wax and wane of its own accord.
If you’re a cat lover, you may be wondering if things are any better in feline medicine. Sadly not. There are even fewer studies on cats, which suffer from a number of mysterious conditions that are hard to treat. Particularly distressing is feline gingivostomatitis, a severe inflammation of the gums that occurs when the immune system overreacts to plaque in the mouth. “It’s a terrible thing,” says Karen Langeman, who runs the Porte Veterinary Hospital in Campbell, California, and sees one or two cases each month. Eating becomes very painful, and some cats have to have all their teeth pulled.
What triggers this exaggerated immune response is unclear, although it’s most common in cats with viral or bacterial infections. And without a good understanding of the cause, vets can do little but try to ease the symptoms with corticosteroids and painkillers.
Mysteries like these could be solved by more research, but how can we get vets to pay attention to the studies that have been done? It would help if professional bodies took a strong evidence-based stand. Sadly, the American Veterinary Medical Association flunked a test of its commitment to scientific principles in January, when its governing body voted down a resolution rejecting homeopathy as an “ineffective practice.” The association’s Australian and British counterparts already discourage homeopathy because of a lack of evidence for therapeutic effects—not to mention the absence of a good explanation of how the extremely dilute solutions used in homeopathic remedies might work. Yet the AVMA’s leadership feared the resolution was divisive, and argued that evaluating specific therapies isn’t its job.
You and I can also make a difference, by pressing vets to consider the evidence that does exist. I’m not suggesting repeating my exercise of digging into the research literature; that’s heavy-going, even for someone who makes his living writing about science. But we can keep our vets on their toes by asking better questions. “Very few of my clients come to me wanting to know what my rationale is for doing what I’m doing,” McKenzie says.
So ask your vet why they think the drugs your animal is being given will work. We’re going to have to confront our own psychological biases, here: research shows that people prefer confident advice, sometimes even when we know those giving it have been wrong before. And good answers to these questions will inevitably be hedged with caveats about the small number of studies that have been done, and their limitations. If all you get from your vet is a bland assurance that they’ve been doing this for years, and see great results, get them to talk you through the scientific evidence. If they can’t do so, that should be a warning sign: It might be time to look for another vet.
Our companion animals do great things for us, improving not just our psychological well-being but also our physical health through knock-on effects like reduced blood pressure. The least we can do in return is to challenge vets to base their decisions on the best available science.
Kaleb, buddy, we owe you one.