What Can Dogs Safely Chew? An Honest Look at What the Evidence Actually Says

Search “what can my dog chew on” and you will drown in confident answers. Give them a bully stick. Never give rawhide. Antlers are great. Antlers are terrible. Frozen carrots for teething. The thumbnail test. It is stated with total certainty, and almost none of it comes with a source.

We went looking for the sources. We read the veterinary dental guidelines, the behavior literature, the foreign-body case series, and the actual biomechanics of a dog’s tooth. What we found is more useful than another confident list, and a little uncomfortable: the experts are clear about what to avoid, nearly silent about what is safe, and a lot of the advice you have been given traces back to nobody at all.

Here is the short version, and then the reasoning behind every piece of it, including where the evidence runs out.

The Short Version

• There is no single safe chew, because “safe” is not one question. A chew can crack a tooth, slowly wear one down, get swallowed and lodge on the way down, or drive a splinter into soft tissue. A product can pass one of those tests and fail another, which is why no material is the universal answer.
• For teeth, it should give. If it will not dent, flex, or yield, it can break a tooth. That rules out antlers, bones, hard nylon, hooves, and horns. It rules out nothing else.
• Softness is not a clearance. A gritty tennis ball gives, sails through the thumbnail test, and still sands teeth down over years. That risk arrives through play, not through anything you would have evaluated as a chew.
• Size it to your dog, and understand that the swallow is the danger. The real emergency is a piece lodged in the esophagus on the way down, not a nibble. Take a chew away once it wears down small enough to gulp.
• Anything edible is a supervised activity, not a crate activity. We do not recommend rawhide, and we explain why below.
• Be stricter with puppies, not more lenient. Baby teeth are thin and fracture easily, and a broken one can mark the adult tooth forming underneath it.
• No veterinary authority publishes a list of safe chew materials. We went looking. That vacuum is precisely why the internet is so confident about this, and it is the most useful thing we can tell you.
• Chewing is a normal need, not a calming ritual. No study in dogs shows it moves a stress marker. Provide for it anyway, because the need is real.

First, Why Dogs Chew (and Why “Just Stop Them” Is the Wrong Answer)

Chewing is not a bad habit to be trained away. It is a normal part of being a dog. In one kennel study, beagles given suspended toys and chews spent about a quarter of their observed daytime using them, and were still using them after two months without losing interest, though individual dogs varied enormously and a couple barely touched them. Chewing is something dogs are built to want to do, and the reasonable expectation, though it is proposed in the literature rather than demonstrated, is that a dog denied an appropriate outlet may redirect onto something you like less, such as your baseboards.

So the goal is not to stop your dog chewing. It is to give the drive something appropriate to land on.

One honest correction while we are here, because you will see it everywhere: the claim that chewing “relieves stress” or “calms” a dog is not established. Here is the actual state of it, which is more interesting than either the confident version or a flat denial.

No study in dogs has shown that chewing moves a physiological stress marker. The one kennel study that measured stress hormones in saliva alongside chews found nothing worth reporting and abandoned the rest of the analysis. There is one study pointing the other way, and it deserves a fair hearing: kennel dogs given a long-lasting chew during a brief period of isolation were rated by observers as calmer and more positive than dogs given other enrichment items. That is real, and we are not going to pretend it does not exist. But it was funded by a pet food manufacturer, it measured how the dogs looked to human raters rather than anything in their bloodstream, and, critically, it compared a chew against other toys rather than against nothing at all. Its own authors say so. So it can tell you a chew held a dog’s attention better than a treat dispenser. It cannot tell you chewing calms a dog.

The sweeping “chewing lowers stress” story you see repeated online is not built on that study anyway. It is extrapolated from human gum-chewing and rodent research. Chewing is a normal behavioral need, and that is reason enough to provide for it. You do not need the calming claim, and we are not going to sell it to you.

The Uncomfortable Truth: Nobody Official Will Tell You What Is Safe

Here is the thing that reframes this whole topic once you notice it.

The major veterinary dental authorities are willing to tell you what to avoid. The 2019 American Animal Hospital Association dental guidelines state that the task force “believes that there is not a strong rationale for offering hard treats (antlers, synthetic, or natural bones) that could damage the structural integrity of the tooth.” That is a clear “don’t.”

Note what that warning is actually about. It is not about clean teeth. A broken tooth is an injury, painful and often needing extraction or a root canal to fix, which is why the dental bodies are the ones on record here. Chew safety is a question about injuries that happens to fall in their jurisdiction.

But go looking for the matching list of what you should give, from an equivalent authority, and it mostly is not there. The American Veterinary Dental College publishes no position statement on chew objects. It does endorse chew materials in general terms for controlling plaque at home, with nothing about hardness, size, or material, which is the whole problem in one line: the recommendation exists and the specification does not. AAHA’s behavior guideline, meanwhile, does not discuss chewing as a need to provide for at all.

We should be straight about the limits of that search. We read the guidelines and position statements people actually cite, and none of them names a safe chew material. We have not read every document every veterinary body has published. So treat this as “the sources everyone points to do not say it” rather than proof that nothing anywhere says it.

That vacuum is exactly why the internet is full of confident, unsourced chew advice. Nature abhors it, so pet blogs fill it. Once you know the authorities are quiet on the constructive side, you can stop trusting anyone who sounds too sure.

On Hardness, One Principle Holds Up: It Should Give

Strip away the folklore and one real, physical principle remains.

A dog’s large upper chewing tooth, the one that does the heavy work, fractures under enough force. A veterinary study measured it: that tooth broke at an average of about 1,281 newtons of force in the lab. Dogs’ own bite forces can reach into that range at the extreme upper end. The researchers themselves concluded that objects which do not “give” meaningfully below that point can raise the risk of fracturing that tooth.

An honest caveat, because we promised sources not scare tactics: dogs do not chew at their maximum force most of the time. The same paper notes that voluntary bite forces recorded in dogs span roughly 13 to 1,394 newtons, so that fracture load sits at the very top of the range rather than in the middle of it. And those are different dogs than the teeth that were broken, so it is a comparison between studies, not a measurement of one animal. It is a lab test on teeth potted in acrylic, not a live prediction. So the risk is real but it is not “every hard chew snaps a tooth.”

The practical version is simple: a chew should have some give to it. If it does not flex, dent, or yield at all, it is in the category that can break a tooth. That is the honest core of what the popular “thumbnail test” was fumbling toward.

One thing to be clear about, because plenty of articles blur it: this rule answers the tooth question and only the tooth question. “It gives” is not a general safety clearance. Rawhide gives, and it is the thing we are about to tell you not to use. A soft hollow rubber toy gives, and it has its own documented failure mode. Give is the test for whether something can crack a tooth. The other ways a chew hurts a dog need their own tests, which is most of the rest of this article.

Speaking of which: that famous rule, that a chew is too hard if you cannot dent it with your thumbnail or it would hurt to tap your kneecap? We tried to find where it comes from. It appears nowhere in the AAHA guidelines and nowhere in the Veterinary Oral Health Council’s materials, the two bodies it usually gets attributed to. It is a reasonable rough idea, but it is folk wisdom wearing a lab coat. Use the principle, not the party trick.

The Risk That “It Should Give” Misses Completely

There is one more way a chew damages a tooth, and it slips past every rule we have given you so far. It gets its own heading because it is the exception that proves the hardness rule is not the whole story.

Teeth do not only break. They also wear down. Veterinary dentistry splits that into attrition, where teeth grind against each other, and abrasion, which is a tooth wearing against something foreign. Abrasion is the one that matters here, and the clinical description of which dogs get it should make a Golden owner sit up: it comes from very aggressive chewing, and it is more common in large breeds.

Now the part that breaks our own principle. The textbook example is a tennis ball. A tennis ball gives. It dents, it flexes, it sails through the thumbnail test. And chronic chewing on toys like tennis balls wears down the canine and premolar teeth anyway, because the fuzzy cover collects grit and the ball effectively becomes fine sandpaper. Softness protects a tooth from cracking. It does nothing to stop a tooth being sanded away.

You might reasonably object that a tennis ball is not a chew, it is a toy. That is exactly right, and it is precisely why this risk gets missed. Nobody hands a dog a tennis ball as a chew object and evaluates it as one. The damage comes from the dog who carries a ball everywhere and works it between the back teeth for years. It arrives through play, which is not a category anybody thinks to apply chew rules to.

Wear is not automatically an emergency. But it can outrun the tooth’s ability to lay down protective dentine underneath, and once the pulp is exposed you are in the same place a fracture puts you: pain, infection, and a choice between a root canal and extraction. That is why a visibly worn tooth is worth a set of dental X-rays rather than a shrug, because a tooth can be dying without looking dramatic.

For a sense of how common it is: one recent study at a veterinary dental service compared working dogs with ordinary companion dogs and found wear on at least one tooth in about 90 percent of the working dogs and a third of the non-working ones. Two caveats we will not paper over. Those were dogs already referred to a dental service, so that is not a prevalence figure for dogs in general. And the teeth most often worn in that series were the incisors, not the canines and premolars that the tennis-ball pattern hits. Still, a third of the ordinary dogs is not a rarity.

The practical version is easy. Rotate the ball out of circulation rather than leaving one down permanently as furniture, and if your dog is a devoted ball carrier, ask your vet to take a proper look at the canines and premolars at the next visit.

Bones, Honestly

Bones are where the conventional wisdom is loudest, so it is worth taking apart carefully. The short version: the folk rules are half right, for reasons people usually get wrong.

“Cooked bones splinter.” There is genuine science under this. When bone dries out, it gets dramatically more brittle. In materials testing, drying bone cut its fracture toughness by about a third and its ability to absorb energy before shattering by around 85 percent, because the water and soft protein that make bone flexible are gone. Cooking drives off moisture and denatures that protein, so the mechanism behind “cooked bone shatters into sharp pieces” is real. The honest caveat: that testing was done on bone as an engineering material, not on a roast chicken bone in a dog, and for obvious reasons no one is going to feed dogs cooked bones just to measure the difference. So the mechanism is documented and the caution is well founded; the specific “cooked kitchen bone hurts dogs more than raw” claim is a very reasonable inference rather than something a study has directly proven.

“Chicken and turkey bones are the dangerous ones.” This is the part that surprised us. Look at the actual clinical records of dogs treated for bone injuries and nobody has tested that question. The case series sort bones by size and where they get stuck, and report outcomes on those terms. None of them compares chicken against beef, or cooked against raw, to see whether it changes what happens to the dog. Poultry bone splintering is mechanically plausible, but the claim that it is specifically the bird bones landing dogs at the emergency vet is not something the clinical data has actually tested.

What the data does show is more useful than the folklore. In one study of dogs treated for bone obstructions, the bones that lodged dangerously in the esophagus were about three times larger relative to the dog’s body than the ones that passed safely to the stomach. And of the bones that did reach the stomach, most were simply left alone to dissolve, with no complications reported. The esophagus is where a bone becomes an emergency. A piece that reaches the stomach has better odds, but read that carefully: those were referral cases where a clinician made the call to leave the bone alone and monitor it. That is a record of what vets decided, not a clearance for you to decide the same thing at home. If your dog swallows a bone, the call to your vet is the move. Whether it needs an X-ray is their judgment, not yours or ours.

So the real rule is not “chicken bad, beef fine.” It is: size relative to your dog is the thing the clinical data actually points at, and a bone getting stuck on the way down is the danger it points at. A chunk small enough to gulp and big enough to jam is the worst case, regardless of which animal it came from. Worth saying plainly: that study established size as an important factor, not as the only one. Shape, how the bone breaks up, and how hard a given dog goes at it all plausibly matter too. Nobody has sorted them out.

And to be clear about where we land, since we have just spent several paragraphs explaining how bone injuries work: understanding a risk is not the same as endorsing the thing that causes it. Bones do not give, so they carry the fracture risk from two sections back, and they are the most common object found lodged in dogs’ esophaguses. We do not recommend them. The size-and-location detail still matters, because plenty of dogs get hold of a bone whether anybody planned it or not, and knowing that a piece stuck on the way down is the real emergency is what tells you when to stop watching and start driving.

“Just feed them raw, then.” This is the escape hatch a lot of people reach for, and it runs into a different wall. The American Veterinary Medical Association discourages feeding any raw or undercooked animal protein, not mainly over choking, but because healthy-looking dogs fed raw can shed Salmonella, Campylobacter, Listeria, and other pathogens in their stool and infect the household, with young children, elderly, pregnant, and immunocompromised people most at risk. The WSAVA’s global nutrition committee lands in the same place, and separately lists ingested bones as a risk for dental fractures and gastrointestinal obstruction, without carving out an exception for raw ones. Raw feeding is its own large topic with passionate advocates, and we are not going to litigate all of it here. But “raw makes bones safe” trades a mechanical risk for an infection one, and that tradeoff is rarely mentioned by the people recommending it.

What About Sticks?

It is the most natural thing in the world. You throw a stick, the dog brings it back, and sooner or later the dog settles in to gnaw on it. Almost nobody thinks of a stick as dangerous. The veterinary literature disagrees more than you would expect.

A stick or splinter driven into the mouth or throat, what vets call an oropharyngeal penetrating injury, is a well-recognized problem in referral practice, and wood is the usual cause. In a review of 50 referred dogs, a piece of wood was the apparent cause in 72 percent of them, and it was medium to large breeds it happened to. It usually happens during ordinary play: a dog runs onto a planted or bouncing stick, or bites down and a splinter buries itself in the soft tissue at the back of the mouth. The immediate signs are pain, blood in the saliva, gagging, or trouble swallowing.

The worse version is the one you do not see coming. In that same series, 82 percent of the dogs did not arrive as an emergency at all. They came in later, after a fragment of wood had migrated and festered into an abscess or a draining wound. That delay is the whole story: clinical signs resolved in every dog that presented acutely, but in only 62 percent of the chronic ones. A more recent series of acute injuries treated with endoscopy reported no major long-term complications in 95 percent of the dogs followed up. So the injury itself is usually fixable when it is caught. The one that goes unnoticed for weeks is the one that does lasting damage.

None of that means a stick will hurt your dog, and we are not going to tell you to pry every stick out of a happy dog’s mouth. A quick carry-and-drop during fetch is low risk. What raises the odds is a dog running full speed onto a stick, or settling in to chew one down. So: fetching a stick, fine. Letting the stick become the chew toy, not a great idea. Hand them something built for the job instead.

Rawhide: Why We Do Not Recommend It

Our answer on rawhide is simple, so here it is up front: we do not recommend it. The reason is not a vague bad feeling about the product. It is where rawhide turns up when things go wrong.

In a 2025 series of dogs treated at a single hospital for upper gastrointestinal obstruction, dried chew snacks, the category rawhide sits in, were the second most common thing found lodged in the esophagus, behind bones. The revealing part is the other half of that finding: the same category accounted for none of the objects found in the stomach. Bones and dried chews are not simply things dogs swallow. They are things that get stuck on the way down, in the one location where a foreign body is genuinely dangerous.

One limit on that figure, since we are being careful with everything else: a case series tells you what was found inside dogs that already had an obstruction. It does not tell you the odds that any particular rawhide causes one, because nobody counted the dogs who chewed rawhide uneventfully. Rawhide is popular, and popular things show up often in case counts. What the study does support is the comparison inside it, which is the part we are leaning on: when dried chews turn up, they turn up in the esophagus rather than the stomach.

That is a swallowing problem, not a chewing problem, and supervision only partly covers it, because the failure happens in a few seconds.

There is one counterargument worth answering, because it confuses people. Some rawhide chews carry the Veterinary Oral Health Council’s seal of acceptance, which sounds like a safety clearance. It is not one. By the VOHC’s own description, its mission is to review products against standards “for effective plaque and tartar control,” and the seal is awarded when a product demonstrates dental efficacy. That is an effectiveness mark, not a safety certification, and it should not be read as evidence that a product cannot choke a dog or block an esophagus.

You do not have to take our word for the gap, either, because the VOHC addresses it directly. Its own guidance tells owners that obstructions from swallowed dental chews can be reduced by giving the correctly sized product for the dog’s weight and by limiting chews to times when someone is there to watch. That is the organization behind the seal telling you the seal does not handle this risk, and handing the job to you. So a rawhide chew can carry the seal and still be a piece your dog gulps too large.

Our recommendation is to skip it. If you are going to use it regardless, then at least stack the odds: strictly supervised, and taken away the moment it shrinks to a size your dog could swallow whole. That is harm reduction on a product we would not have chosen, not a quiet endorsement. It is never a leave-in-the-crate item.

The Question You Actually Have: What Can I Leave in the Crate?

This is the practical fork nobody addresses cleanly. “Give your dog a chew” means one thing when you are sitting next to them and something very different when they are crated and you are at work.

We have to be honest here: there is no veterinary authority or safety standard that defines a “safe to leave unsupervised” category of chew. That distinction, the reassuring idea that this toy is fine to leave and that one is not, is essentially industry convention and manufacturer guidance, not something a governing body has established. The VOHC’s own advice on dental chew treats is to limit them to times when someone is available to watch the dog chewing, which, taken literally, is stricter than how most people actually use an edible chew.

So rather than pretend there is an official rulebook, here is the honest framework, sorted by risk:

• Safer to leave, with judgment: a sturdy rubber toy, sized correctly and too big to swallow, ideally stuffed with food to occupy the dog. This is the common recommendation, and it is a reasonable one, but know that it is convention, not a guarantee. One documented failure mode: hollow rubber toys have caused tongue entrapment when a dog compresses the toy and the resulting vacuum traps the tongue. It is rare, and the reported fix is to break the seal by piercing the toy, which is the logic behind hollow toys built with a hole at both ends. That second hole is not decorative; it lets the pressure equalize, which should make suction less likely to build. If you use a hollow rubber toy, that vented design matters, and so does checking it for cracks and wear.
• Supervised only: anything edible and consumable. Dental chews, any long-lasting edible chew. These are the ones that get gulped, splinter, or shrink to a swallowable nub, and those are failures somebody needs to be there to catch. This is not a theoretical category: in a review of 31 dogs treated for esophageal obstruction caused by a dental chew treat, the esophageal damage was moderate or severe in 26 of the 30 that could be assessed, and about a quarter of the dogs did not survive. The honest caveat is that 84 percent of those were small breeds, so it does not transfer cleanly to a large dog. A chew that has worn down small enough to swallow whole should be taken away.
• Not at all: hard items that do not give (antlers, bones, hard nylon, hooves, horns), because of the fracture risk, and stringy or fabric items left unattended, because swallowed strands can bunch the intestine.

If you take one thing from this section: the crate is exactly where the “supervised only” chews should never end up.

Puppies Are a Different Problem

Everything above is written for an adult dog. Puppies need the same principles applied harder, and the reason is not the one you have been told.

The popular story is that puppies chew to soothe sore, teething gums. That is shakier than it sounds, and it is not why we would be careful. The real reason is that baby canine teeth are long, thin and fragile, puppies fracture them easily, and a broken baby tooth is not a throwaway problem: inflammation at its root can interfere with the permanent tooth developing directly underneath it, leaving thin enamel or a malformed crown on an adult tooth that has not even come in yet.

So the direction of travel is the opposite of what most people assume. Hard chews are a worse idea at four months than at four years. And the trigger for acting on a baby tooth that has not fallen out is not a date on the calendar, it is the permanent tooth appearing next to it.

Teething itself, the teething rings, the vegetable chews and the retained-baby-tooth window are a big enough subject that we are giving them their own article rather than bolting them onto this one.

So What Do We Actually Recommend?

After all of that, here is the short, honest version.

Start by dropping the idea that there is a single safe chew, because “safe” is not one question. A chew can crack a tooth. It can slowly wear one down. It can be swallowed and lodge on the way down. It can drive a splinter into soft tissue. And any of those goes worse when nobody is in the room. A product can pass one of those tests and fail another, which is exactly why no material is the universal answer and why anything marketed as one deserves a raised eyebrow.

So screen for each risk separately:

1. Provide for the chewing, do not fight it. Chewing is normal, and most dogs do a lot of it. Without an appropriate outlet, some dogs will redirect onto your baseboards. Give the drive somewhere to land.
2. For teeth: it should give. Pliable rubber, not rock-hard. If it will not dent, flex, or yield, it can break a tooth. That screens out the fracture risks, which is to say antlers, bones, hard nylon, hooves, and horns. It screens out nothing else, and notably it does not screen for wear: a gritty tennis ball gives and still sands teeth down over years.
3. For swallowing: size it to your dog and mind the gulp. Too big to swallow whole. Take it away when it wears down small. The version of a swallowed piece that turns into a real emergency is the one that lodges in the esophagus on the way down.
4. Supervise the consumables. Anything your dog eats down is a supervised activity, not a crate activity. We would skip rawhide altogether; it turns up lodged in the esophagus too often for the risk to be worth it.
5. Do not pick a chew for dental reasons. Some products genuinely do reduce plaque and tartar, and the Veterinary Oral Health Council’s accepted list is where to find those. Treat that as a bonus on a chew you chose for the reasons above, not as the reason to hand one over, and do not treat it as a replacement for brushing. Chewing is a much bigger subject than clean teeth, which is why most of this article is about injuries rather than tartar.

Notice what is not on this list: a brand, a magic material, or a promise that any of this de-stresses your dog. The evidence does not support those, so we will not pretend it does.

Quick Answers

Is the thumbnail test real? Not as an official rule. It traces to no veterinary authority. The idea behind it, that a chew should have some give, is sound as far as your dog’s teeth are concerned. The specific test is folklore, and it says nothing about whether a chew can be swallowed or splinter.

Are antlers safe? The one veterinary position on them, from AAHA, is a non-endorsement: no strong rationale for offering a treat that hard given the tooth-fracture risk. There is no study proving antlers are safe, and the whole category of things-that-do-not-give carries fracture risk. We do not recommend them.

What about bully sticks? A bully stick is a dried animal-protein chew, which puts it in the same category as the other dried chews here: soft enough that it is not a tooth-fracture worry, edible and consumable so it is supervised-only and never a crate item, and it needs taking away once it wears down to a gulpable stub. We could not find any evidence specific to bully sticks, for or against, so we are not going to invent a verdict. Treat it as what it is: a consumable chew that gets smaller while your dog works on it.

Frozen carrots for teething? We found no evidence either way. Freezing anything makes it harder, and hard is the wrong direction for a mouth full of thin baby teeth. A carrot is also food, so it belongs in the supervised column. We cover teething items separately.

Are tennis balls bad for my dog’s teeth? They can be, and not because they are hard. The fuzzy cover picks up grit and acts like fine sandpaper, and chronic chewing on tennis balls is a recognized cause of worn canine and premolar teeth, particularly in big, enthusiastic chewers. A ball for fetch is fine. A ball that lives on the floor and gets ground on for years is the problem. Rotate them out, and have your vet check the back teeth of a committed ball carrier.

Does chewing calm my dog down? Not established. No dog study has shown chewing moves a physiological stress marker. One manufacturer-funded study found kennel dogs looked calmer to human observers with a chew than with other enrichment items, which is worth knowing, but it had no “nothing at all” comparison, so it cannot settle the question. Chewing is a normal need worth providing for either way. The sweeping calming claim online is borrowed from human and rodent research.

Are chicken bones more dangerous than beef bones? The mechanism for splintering is plausible, but nobody has tested whether species changes what happens to the dog. What the data does point at is bone size relative to your dog, and a bone getting stuck in the esophagus. Size is an important factor rather than the only one.

Do cooked bones really splinter? The physics support it: drying and cooking make bone markedly more brittle. That has not been formally tested in dogs, but the mechanism is real and the caution is reasonable.

Is rawhide bad if it has the VOHC seal? The seal means the product met a standard for plaque or tartar control. It is not a safety certification, and the VOHC itself tells owners to size dental chews correctly and supervise them because of obstruction risk. Rawhide’s main risk is a softened or partly chewed piece being swallowed and lodging in the esophagus, which the seal does not speak to. Given the alternatives, we do not recommend it.

Can I leave a chew in the crate? A correctly sized, sturdy stuffed rubber toy is the usual answer, and a reasonable one, though “safe unsupervised” is convention rather than an official standard. Anything edible or consumable should be supervised, never left in the crate.

What about puppies teething? Be stricter about hardness with a puppy than with an adult, not more relaxed. Baby canine teeth are long, thin and easily fractured, and infection at the root of a broken one can mark the permanent tooth developing underneath. Teething rings and the retained-baby-tooth window get their own article.

Sources

• American Animal Hospital Association. 2019 AAHA Dental Care Guidelines for Dogs and Cats. Hard-treat position naming antlers and bones, and the referral to the VOHC list. Stated as a Task Force belief and carries no citation, so it is expert opinion rather than a graded evidence finding.
• Soltero-Rivera M, Elliott MI, Hast MW, et al. Fracture limits of maxillary fourth premolar teeth in domestic dogs under applied forces. Front Vet Sci 2019;5:339. PMID 30761310. Mean fracture load about 1,281 N. Ex vivo cadaveric teeth under static point load, not live chewing.
• Veterinary Oral Health Council. Accepted Products for Dogs and VOHC home. Mission is to review products against standards “for effective plaque and tartar control”; the seal is awarded on demonstrated dental efficacy. VOHC reviews sponsor-submitted trial data rather than testing products itself. Its own owner guidance on reducing obstructions from swallowed dental chews: give the correctly sized product for the dog’s body weight, and limit chews to times when the owner can observe the dog.
• Quinn R, Masters S, Starling M, White PJ, Mills K, Raubenheimer D, McGreevy P. Functional significance and welfare implications of chewing in dogs (Canis familiaris). Front Vet Sci 2025;12. PMID 40206258. Narrative review from a single group; its own conclusions are hedged and it states the amount of chewing needed remains unknown.
• Hubrecht RC. A comparison of social and environmental enrichment methods for laboratory housed dogs90123-7). Appl Anim Behav Sci 1993;37(4):345-361. 432 hours of observation on 48 beagles; dogs spent 24 percent of observed time using the toys after two months. That is a share of observed daytime, not of a full day, and individual dogs varied widely.
• Flint HE, Atkinson M, Lush J, Hunt ABG, King T. Long-Lasting Chews Elicit Positive Emotional States in Dogs during Short Periods of Social Isolation. Animals (Basel) 2023;13(4):552. PMID 36830339. Funded by Mars Petcare UK. Outcomes are observer ratings and qualitative behavioural assessment, with no cortisol, heart rate or heart rate variability collected. The authors state: “there was no control intervention used in the current study, so it is unknown how these emotional responses would compare to social isolation events with no enrichment present.”
• Barash NR, Lashnits E, Kern ZT, Tolbert MK, Lunn KF. Outcomes of esophageal and gastric bone foreign bodies in dogs. J Vet Intern Med 2022;36(2):500-507. PMID 35156732. Esophageal bones had a median cross-sectional area of 98.21 mm2/kg versus 28.6 mm2/kg for gastric bones (P < .001); most bones reaching the stomach were left in place without complications.
• Griffiths LG, Tiruneh R, Sullivan M, Reid SW. Oropharyngeal penetrating injuries in 50 dogs: a retrospective study. Vet Surg 2000;29(5):383-388. PMID 10999451. Wood was the apparent cause in 72 percent; 82 percent presented chronically; signs resolved in all acute cases but only 62 percent of chronic ones.
• Kilduff-Taylor A, Baines SJ. Endoscopic treatment of acute oropharyngeal stick injuries in dogs: 46 cases (2010-2020). J Small Anim Pract 2023;64(10):635-641. PMID 37340777. 38 of 40 dogs with long-term follow-up (95 percent) had no major long-term complications.
• Yan J, Daga A, Kumar R, Mecholsky JJ. Fracture toughness and work of fracture of hydrated, dehydrated, and ashed bovine bone. J Biomech 2008;41(9):1929-1936. PMID 18502430. Drying cut fracture toughness by about 31 percent and work of fracture by about 85 percent. This is materials testing on bone as an engineering material, not a study of cooked bones fed to dogs.
• Niemiec BA. Disorders of Dental Hard Tissues in Dogs. Today’s Veterinary Practice, May/June 2014. Abrasion is “a tooth wearing against something foreign”; attrition is “caused by tooth-on-tooth contact.” “Chronic chewing on toys, such as tennis balls, wears down the canines and premolars,” and abrasion “in general, is due to very aggressive chewing, which is more common in large breed dogs.” All worn teeth should be radiographed to confirm vitality. Clinical guidance from a boarded veterinary dentist, not a research study. The grit-as-sandpaper mechanism is the standard clinical explanation rather than a measured finding.
• Han S, Reiter AM, Stefanovski D, Castejón-González AC. Traumatic dentoalveolar injury, tooth wear, and periodontal disease in working and non-working dogs (2018-2022). Front Vet Sci 2026. 102 dogs (51 working, 51 non-working). “Tooth wear was identified at least in one tooth in 46 (90.2%) WD and 17 (33.3%) NWD. The mean number of teeth affected was 12.7 ± 11.5 in WD and 1.9 ± 4.9 in NWD.” Incisors were most commonly affected, premolars second. Limit: a veterinary dental service population, so not a general prevalence estimate.
• Leib MS, Sartor LL. Esophageal foreign body obstruction caused by a dental chew treat in 31 dogs (2000-2006). J Am Vet Med Assoc 2008;232(7):1021-1025. PMID 18380620. Esophageal lesions moderate or severe in 86.7 percent (26/30); strictures in 6 of 25 survivors (24 percent); overall mortality 25.8 percent. Important limit: 83.9 percent (26/31) were small dogs, so the figures do not transfer directly to large breeds. Questionnaire data from 19 hospitals; the treat’s composition is not specified in the report.
• Charlier C. Deciduous Canine Tooth Removal in the Dog. Today’s Veterinary Practice, January/February 2022. “Active, chewing puppies easily fracture their long, thin, fragile deciduous canine teeth,” and periapical inflammation from a fractured deciduous tooth “may interfere with development of the permanent tooth and lead to focal enamel hypoplasia, hypomineralization, or crown malformation of the developing permanent tooth.” Clinical guidance from a veterinary dentist, not a research study.
• American Veterinary Medical Association. Raw or Undercooked Animal-Source Protein in Cat and Dog Diets.
• WSAVA Global Nutrition Committee. Statement on the Risks of Raw Meat-Based Diets. Lists ingested bones as a risk for dental fractures and gastrointestinal obstruction, and finds no properly documented evidence of health benefits.
• Rubio A, Van Goethem B, Verhaert L. Tongue entrapment by chew toys in two dogs. J Small Anim Pract 2010;51(10):558-560. PMID 21029100. “Compression of a chew toy during regular play activity can create a vacuum effect eventually causing entrapment of oral tissues.” Two case reports.
• Laiket L, Sutthiprapa W, Khattiya S, Temwichitr P, Rattanasrisomporn J, Thengchaisri N. Prevalence and characteristics of foreign body-induced upper gastrointestinal obstruction in dogs. J Adv Vet Anim Res 2025. PMC12506697. Esophageal foreign bodies were 61 percent bone and 23 percent dried chew snacks; dried chew snacks accounted for none of the gastric cases. Single-centre series.
• US Food and Drug Administration. No Bones About It: Reasons Not to Give Your Dog Bones. Draws no distinction by species or by cooked versus raw, and rebuts the idea that large bones are safer.
• Merck Veterinary Manual. Dental Development of Dogs (permanent teeth appear around 4 to 5 months, complete by 7 months) and Developmental Abnormalities of the Mouth and Dentition in Small Animals (a deciduous tooth should be extracted once its permanent successor has begun eruption).