Do Joint Supplements Work for Dogs? The Evidence, Graded

The joint supplement aisle is built for exactly one customer: somebody whose dog has started to slow down, who has just been told the words hip dysplasia or arthritis, and who wants to do something about it today.

That is a real thing to want, and I have been that customer. So this article is not going to tell you that supplements are a scam and leave it there. It is going to tell you what the evidence actually shows, which is more specific and more useful than that.

The short version: the best-selling joint supplement in the pet aisle does not work, and the one that does is cheaper, less marketed, and sitting three shelves away.

This is general information, not veterinary advice. A dog who is stiff, lame, or slowing down needs a veterinarian who can put hands on the joint — not a supplement chosen from an article. Nothing here is a dose, and that omission is deliberate.

Jump To Section

Table of Contents

The short answer

Glucosamine and chondroitin do not appear to work for canine arthritis pain. The largest analysis of the evidence found a “very marked non-effect,” and its authors recommended in print that these products should no longer be recommended for pain management.

Omega-3 fatty acids do appear to work — they came out of the same analysis with the strongest effect of anything tested.

And the single most effective thing has nothing to do with a bottle. Keeping a dog lean delayed the onset of hip arthritis by six years in a controlled lifetime study. It costs nothing, which is exactly why nobody advertises it.

Do glucosamine supplements work for dogs?

The evidence says no. A 2022 systematic review and meta-analysis covering 72 trials found a “very marked non-effect” for chondroitin-glucosamine products, with 88.9% of trials showing no efficacy and an effect size of -1.39. The authors recommended these products should no longer be recommended for pain management in dogs and cats. A 2023 randomized trial using force-plate measurement reached the same conclusion.

What the biggest analysis of the evidence found

In 2022 a research group at the Université de Montréal published a systematic review and meta-analysis of nutraceuticals and enriched diets in canine and feline osteoarthritis in International Journal of Molecular Sciences. It is open access, so you can read it yourself.

They screened 1,578 publications, plus 20 more from their own files, and ended up with 57 articles containing 72 separate trials across nine categories of product. The study was pre-registered on PROSPERO, which matters — it means the analysis plan was published before the results were known.

Here is what came out.

What was tested	Effect size	What that means
Omega-3 supplements	d = 1.19	The strongest result of anything in the analysis
Omega-3-enriched diets	d = 0.58	Medium to large. “Evident clinical analgesic efficacy”
Cannabidiol	Positive, smaller	High-quality studies, good evidence, lesser degree
Collagen	Weak	Something there, not much
Chondroitin-glucosamine	d = -1.39	“A very marked non-effect.” 88.9% non-effect, 0% effect

The authors did not hedge their conclusion. From the abstract, verbatim:

“Our analyses showed a weak efficacy of collagen and a very marked non-effect of chondroitin-glucosamine nutraceuticals, which leads us to recommend that the latter products should no longer be recommended for pain management in canine and feline osteoarthritis.”

That is a research group telling veterinarians to stop recommending the best-selling product in the category.

Grading it honestly, including the part that cuts the other way

This site’s rule is that the same scrutiny applies to research it agrees with. So:

In its favor: pre-registered, open access, no proprietary funding — the paper states there was “no proprietary interest or funding directly provided for this project or to any of the authors,” with indirect support from Canadian public research grants.

Against: the authors disclose having conducted work in canine osteoarthritis for a long list of animal health companies, including Boehringer-Ingelheim, Elanco, Merck, Nestlé Purina and Royal Canin. That is normal in a small specialist field — there are not many labs doing this work — but it is worth knowing, and they published it themselves rather than being caught at it.

A meta-analysis is also only as good as the trials inside it, and many of the trials in this field are small.

Then somebody tested them head to head

A meta-analysis pools other people’s work. A 2023 trial went and measured it directly, and it is the more persuasive study of the two because of how it measured.

Seventy-five pet dogs with radiographically confirmed hip osteoarthritis were randomly assigned to five groups: glucosamine and chondroitin sulfate, two marine fatty acid compounds, carprofen — a prescription veterinary anti-inflammatory — and a placebo of sunflower oil. Double-blinded, block-randomized, six weeks.

The measurement was a force plate, recording peak vertical force as the dog trotted across it. That is the dog’s own weight going through the leg, in newtons. It does not know what group it is in, it is not influenced by hope, and it does not improve because an owner spent money.

Group	Change in peak vertical force at 4 weeks
Marine fatty acid compound (PCSO-524)	+3.90
Marine fatty acid compound (EAB-277)	+4.17
Carprofen (the prescription drug)	+3.08
Placebo (sunflower oil)	+0.08
Glucosamine and chondroitin	-0.05

Read the bottom two rows together. The glucosamine group did not merely underperform. It landed on the wrong side of a spoonful of sunflower oil.

Meanwhile the two marine compounds performed as well as the prescription anti-inflammatory, and that held at six weeks.

The finding in that trial that nobody quotes

There is a second result in the same study, and leaving it out would be dishonest.

Alongside the force plate, the researchers recorded a subjective orthopedic assessment score — the kind of judgment an owner or a clinician makes by watching the dog. That score did not change significantly in any group. Not in the marine oil groups, and not in the carprofen group either.

Six weeks of a real prescription drug, and the watching-the-dog measure could not see it.

Two things follow. First, real improvements can be too small to notice by eye over six weeks, which is worth remembering before you conclude that something is working. Second, and more usefully: if a subjective score cannot reliably detect an actual anti-inflammatory drug, it certainly cannot validate your supplement. “He seems a bit brighter” is not evidence. It is the thing every supplement in this category is sold on.

And now the incentive, because it runs both ways here

It would be easy to finish that section with “so buy the marine oil instead.” Here is why the article does not.

That trial was funded by a company that sells one of the compounds it found effective. The funding statement was also wrong when the paper was first published: a corrigendum issued in August 2023 records that the stated funder was incorrect and names the actual one. A senior author also discloses having been a paid consultant for a related company, while stating he received no payment connected to this study.

None of that makes the force plate lie. Randomization, blinding and an objective outcome are exactly the defenses that make industry-funded work believable, and they were all in place.

But it does mean the two findings sit on different footings. “Glucosamine did not beat sunflower oil” comes from a group with no stake in that answer. “This marine extract worked as well as carprofen” comes from a trial paid for by someone who profits if it did. Both can be true. They are not equally disinterested, and you should hold them differently.

This is the same test this site applies everywhere: ask who benefits if you believe it. It is the reason how I review things exists as its own page.

Why nobody checks whether the bottle contains what the label says

Underneath all of this sits a regulatory gap most owners have no idea exists.

There is no legal requirement in the United States that a pet supplement contain what its label claims.

The Dietary Supplement Health and Education Act — the 1994 law that created the whole regulatory category for human supplements — does not apply to animals. Congress did not include them. So a pet supplement has no legal category of its own. The FDA treats these products either as animal food, or as unapproved animal drugs if they claim to treat a disease. And because they are not labeled and sold as animal feeds, they largely fall outside state feed regulation too.

The result is a product that is not quite a food and not quite a drug, and is therefore properly policed as neither.

There is one voluntary check. The NASC Quality Seal, from the National Animal Supplement Council, is awarded to member companies that submit to auditing and adverse-event reporting. It is meaningfully better than nothing, and I look for it.

But be clear about what it is. NASC is an industry association funded by the companies it certifies. It is the industry policing itself, not a regulator. And membership is voluntary and costs money — which means the companies least willing to be audited are precisely the ones with no reason to join. A missing seal tells you more than a present one.

If you want the fuller version of how thin the checking is on things your dog eats, what the dog food recalls actually tell you covers the same gap on the food side, and it is not reassuring.

What actually works, in order

Here is the honest ranking, strongest evidence first. Only one item on it comes in a bottle.

1. Keep the dog lean. Nothing else is close.

In a lifetime study of 48 Labrador Retrievers, littermates were paired and one of each pair was fed 25% less from eight weeks of age until death.

The lean dogs lived a median 1.8 years longer. And the age at which hip osteoarthritis appeared moved from six years old to twelve.

Six years of joint disease, prevented by a measuring cup. Disclose the funding, because the rule applies here too: that study was funded by Nestlé Purina. It is also a randomized controlled lifetime trial with littermate pairs, which is about as strong as veterinary nutrition evidence gets.

Body condition is scored by hand, not by the number on the scale. How much to feed and how to score body condition is the practical version, and there is a calculator on it.

2. Veterinary care, properly staged

Arthritis is a diagnosis, not a shopping category. International consensus guidelines for canine osteoarthritis published in 2023 set out staged management — weight programs, exercise and physical therapy, pain medication where it is warranted — matched to how advanced the disease is.

Funding disclosed: those guidelines were funded by Elanco, an animal health company that sells arthritis medication, and the funder organized the meetings and reviewed drafts. That is a real conflict on a document that recommends treatment. It is still the profession’s current consensus position, and it is still worth reading with that in mind.

The practical point stands regardless: a dog with painful arthritis has options that are not sold in an aisle, and a supplement bought instead of a veterinary appointment is the expensive way to lose six months.

3. Omega-3 fatty acids

This is the one supplement with evidence behind it — the strongest effect in the 2022 meta-analysis, and consistent with the marine oil results in the 2023 trial.

This article will not give you a dose, and you should be suspicious of any that does. There is no dose I could verify from a primary source that I would be willing to put in front of somebody’s dog. Fish oil is not harmless at arbitrary amounts, the right quantity depends on the dog’s weight and health, and it can interact with other medication.

Ask your veterinarian for the amount, and ask for it in milligrams of combined EPA and DHA — not in capsules, not in teaspoons of oil, and not in “one scoop.” Those units tell you nothing, and a product that only speaks in scoops is not telling you what is in it.

4. Feeding an older dog properly, which is the opposite of what most people were told

The old advice was to cut protein for senior dogs. It is outdated, and the National Academies’ own guidance says so:

“Older dogs appear to require somewhat more protein to maintain their protein reserves, perhaps as much as 50% more.”

And in the same document, from the National Research Council:

“Because of decreased physical activity and slowed metabolism, older dogs need 20% fewer total calories than do middle-aged adult dogs.”

Fewer calories and more protein means a higher protein density, not a lower one. An older dog losing muscle is losing the thing that holds the joint together. What working dogs actually need from their food goes through this in full.

The three-month script, and why it works on people who love their dogs

There is a sales pattern in this category that deserves naming, because it is what turns a product that does nothing into a recurring charge.

“Give it three months. Some dogs take six.”

Think about what that sentence does. It moves the test date past the point where you can remember what the dog was like when you started. It runs past most return windows. And it converts an absence of results into evidence that you have not waited long enough.

Nobody takes a child to the doctor, accepts a prescription that may show no benefit for six to twelve months, and simply keeps administering it without ever asking again.

The test is not how long they told you to wait. It is who benefits from the wait.

Note what the actual trials did: six weeks, with a measurement at the start and a measurement at the end. When somebody genuinely wants to know whether a thing works, they measure it early.

How to test one properly, if you are going to try it anyway

You might reasonably decide to try something. Here is how to find out whether it did anything, rather than whether you wanted it to.

Write down the date and what you are giving, including the milligrams of the active ingredient, not the number of chews.
Record something countable before you start. Stairs without stopping. Whether he gets up on the first try. How far into the walk the limp appears. A video of him rising from lying down.
Change one thing at a time. A supplement started the same week you cut his food and added a walk has told you nothing about the supplement.
Set the review date yourself, before you buy, and make it six to eight weeks — the interval the actual research used.
Cancel the subscription before that date, not after. You can always start again. It is much harder to stop.
Take the video again on review day and compare it with the first one. Memory is not evidence, and in this category memory is the product.

For a dog whose working life is ending rather than beginning, the calculation is different again — retiring a service dog is about that, and what mobility work physically demands matters if your dog braces or counterbalances, because those tasks put load through the joints you are now worried about.

What this means if your dog works

A service dog’s joints are not an abstract concern. They are the reason the team keeps functioning.

Which is why the ranking above matters more here than it does for a pet. The three interventions with real evidence — keeping the dog lean, veterinary treatment, and omega-3 — are all available to a handler on a fixed income. The expensive one is the one that does not work.

And if a dog’s joints are becoming the limiting factor in what he can do, that is a conversation with a veterinarian about the work itself, not a shopping decision. Load-bearing tasks on a painful joint are not something a supplement fixes.

Frequently asked questions

What joint supplement actually works for dogs?

On current evidence, omega-3 fatty acids. In a 2022 meta-analysis of 72 trials, omega-3 supplements produced the largest effect of any category tested, and omega-3-enriched diets showed clear analgesic efficacy. A 2023 randomized trial using force-plate measurement found marine fatty acid compounds performed comparably to a prescription anti-inflammatory over six weeks. Ask your veterinarian for the amount in milligrams of combined EPA and DHA.

Is glucosamine bad for dogs?

The evidence is about a lack of benefit rather than harm. Glucosamine and chondroitin are generally regarded as well tolerated in dogs, and the meta-analysis finding was that they do not relieve arthritis pain, not that they cause injury. The cost is money and time — money spent on something that does not work, and months during which a painful dog is not getting treatment that would help.

How long should a joint supplement take to work?

The published trials measured at four and six weeks and found real effects in that window. If a seller tells you to wait three months, or six, before judging, notice that this pushes your assessment past the point where you can accurately remember the starting condition, and usually past any refund window. Set your own review date before you buy, and record something countable at the start.

Are pet supplements regulated in the United States?

Much less than most owners assume. The Dietary Supplement Health and Education Act, which governs human supplements, does not apply to animals. The FDA treats pet supplements as animal food, or as unapproved animal drugs if they carry disease claims, and because they are not sold as feeds they largely escape state feed regulation as well. There is no general legal requirement that the bottle contains what the label says.

What does the NASC Quality Seal mean?

It means the company is a member of the National Animal Supplement Council and has submitted to periodic auditing and adverse-event reporting. It is a genuine check and better than nothing. It is also an industry association funded by its member companies rather than a regulator, and membership is voluntary and costs money — so its absence is more informative than its presence.

Do older dogs need less protein?

No, and this is one of the most persistent pieces of outdated advice in dog nutrition. National Research Council guidance states that older dogs may require as much as 50% more protein to maintain their protein reserves, while needing about 20% fewer calories. That means a higher protein density in the food, not a lower one. Restricting protein in a healthy senior dog risks the muscle that supports the joints.

Should I give my service dog a joint supplement?

That is a question for the veterinarian who examines your dog, not for an article. What the evidence supports is keeping the dog lean, getting arthritis properly diagnosed and staged rather than self-treated, and asking about omega-3. If your dog performs mobility, bracing or other load-bearing tasks and has started to slow down, the more urgent question is whether the work itself needs to change.

Why do so many joint supplements have good reviews?

Because owner assessment is unreliable at exactly this task. In the 2023 trial, the subjective assessment score did not detect a significant change in any group — including the dogs receiving a real prescription anti-inflammatory that the force plate showed was working. If watching the dog cannot reliably detect a functioning drug, it cannot validate a supplement either. Reviews measure hope and expense, not joint function.

I’m a service dog handler of twelve years, not a veterinarian, and nothing here is veterinary or medical advice. Every effect size and quotation above is taken from the papers linked, which are open access where possible so you can check them. No product is named or recommended anywhere in this article, and nothing on this page earns me anything. If your dog is in pain, the next step is a veterinary appointment, not a purchase.

Sources: