Treatments and gadgets. 12 minute read.
Do joint supplements work? Glucosamine and chondroitin, and where collagen sits
By the PocketPhysio editorial team. Published 2026-10-01, updated 2026-10-02.
What the 2026 guideline says about glucosamine and chondroitin
The American College of Rheumatology guideline summary approved on 8 September 2026 puts both of them in one sentence: "In people with knee, hip, and hand OA, we strongly recommend against glucosamine, chondroitin, and combination preparations of glucosamine and chondroitin." OA there is osteoarthritis, the wear-related joint problem, and the sentence covers all three of the joints people most often buy these products for.
Two labels travel with that sentence, and both belong in any honest report of it. The recommendation is graded strong. The certainty of evidence is graded low for glucosamine alone and moderate for the combination. Those are separate judgments.
Strength describes the recommendation. In GRADE, the system those two labels come from, strong means the panel expects nearly everyone who sees the same evidence to make the same choice. The weighing behind it covers burden and cost as well as benefit and harm. Certainty describes the research: low means the true effect could turn out to be substantially different from the estimate, so further studies could still change it. The summary document prints the labels without their definitions, so those come from the grading system rather than from the guideline text.
The two can sit together without contradicting each other, because a panel can be firm about the decision while the trials behind it stay shaky. A conditional recommendation is the one that says the opposite, that reasonable people with the same facts would choose differently.
The document also marks this one as a change in direction since 2019. Three of its recommendations carry that mark and only one, a conditional recommendation about GLP-1 medicines for people who have knee osteoarthritis and obesity, is marked as new, so this is a reversal rather than a new idea. That matters because the previous guideline did leave a door open: the National Center for Complementary and Integrative Health notes that the 2019 version from the same college "conditionally recommends chondroitin for patients with hand osteoarthritis", while strongly recommending against glucosamine alone or with chondroitin for the knee. The 2026 document closes that door.
In the UK the position has been settled for longer. NICE guideline NG226, published on 19 October 2022, puts glucosamine in a do-not-offer recommendation, 1.4.6: "Do not offer glucosamine or strong opioids to people to manage osteoarthritis", and recommendation 1.4.7 tells clinicians that if someone asks about it, to explain that "there is no strong evidence of benefit for glucosamine". NICE does not publish its recommendations as a strength and certainty pair, so there is no grading to quote alongside that; it is an instruction not to offer it. NICE names glucosamine only, and the NHS fills in the other half without hedging: "GPs no longer prescribe chondroitin and glucosamine on the NHS because there's no strong evidence that they are effective as a treatment."
What the trials behind that actually found
Not nothing, which is why the subject refuses to die. The National Center for Complementary and Integrative Health describes a 2018 combined analysis of 29 studies in people with knee osteoarthritis, 6,120 participants in total, in which "global pain was significantly reduced by glucosamine or chondroitin taken separately but not by the combination of the two". Its own summary of the field is that "it's still uncertain whether glucosamine and chondroitin are helpful for knee osteoarthritis symptoms".
For the hand there is one trial worth knowing about: 162 participants, six months, in which hand pain fell and hand function improved more with chondroitin than with placebo, as the same agency reports. That single study is the reason chondroitin had a conditional recommendation for hand osteoarthritis in 2019.
For the hip the answer is flatter. The American Academy of Orthopaedic Surgeons concluded in 2017 that moderate strength evidence does not support glucosamine sulfate for hip osteoarthritis, and that glucosamine was no better than placebo for pain, joint function or joint structure measured as joint space narrowing, based on one high quality trial in 222 people.
One body still reads the knee evidence more warmly. The 2021 orthopedic surgeons' guideline includes glucosamine among dietary supplements that may help pain and function in mild to moderate knee osteoarthritis, "although it cautions that the evidence is inconsistent", in the words of the agency summarizing it. Two bodies, one literature, opposite conclusions. The newer rheumatology document is the one that graded its certainty openly.
Where two expert bodies openly disagree
The disagreement worth explaining is about a prescription product, not about the tub in a pharmacy aisle. The European Society for Clinical and Economic Aspects of Osteoporosis, Osteoarthritis and Musculoskeletal Diseases, ESCEO, issued a statement in 2019 that "strongly recommends prescription crystalline glucosamine sulfate for knee osteoarthritis but discourages the use of other glucosamine formulations", in the words of the National Center for Complementary and Integrative Health. A 2021 paper written by a working group set up between members of the international ESCEO task force and a group of Southeast Asian experts, with ESCEO figures among its authors, spells that algorithm out: "ESCEO specifically recommends the use of pCGS and long-acting chondroitin sulfate products", where pCGS is prescription crystalline glucosamine sulfate. The same paper says the other versions, including glucosamine hydrochloride, "have consistently been shown to be ineffective in the treatment of knee OA" (Yeap and colleagues, 2021).
Then it explains the split: "ESCEO strongly supports the use of pCGS and chondroitin sulfate, whereas OARSI does not support their use. The main reason for this discrepancy is that the OARSI guidelines, which were mainly prepared in a US-centric perspective, do not recognize the concept of 'pharmaceutical-grade' or 'prescription-grade' SYSADOAs, such compounds being unavailable on the US soil." That is as clear a statement of a guideline disagreement as you will find in print, from the side that disagrees. The other side of it is not one body on its own: as the National Center for Complementary and Integrative Health records, the 2019 guideline from Osteoarthritis Research Society International also strongly recommends against glucosamine for knee osteoarthritis for lack of efficacy, and against chondroitin on the grounds that the evidence is of low quality.
Read it with the disclosures in view. Several of that paper's authors declared advisory board work and conference support from pharmaceutical companies, which is disclosed in the paper and is normal in this field. The trial evidence that makes the prescription preparation look better is itself under a cloud: summarizing a 2014 analysis of 25 glucosamine trials in 3,458 people with knee osteoarthritis, the National Center for Complementary and Integrative Health reports that those trials did give more favorable results, then adds that "most of the studies of the prescription drug formulation had a high risk of bias because of weaknesses in their study design, were published more than 20 years ago, and were funded by the pharmaceutical company". And if you live in the USA, the UK, Canada, Australia or New Zealand, the preparation the European algorithm is arguing for is not the one you would be buying, so the argument does not transfer to the product in your hand.
Collagen is a different question, and a more open one
Look first at what the guideline did say about the other things people swallow, because that sets the scale. Vitamin D is a conditional recommendation against for all three of those joints, at low certainty of evidence. Fish oil is a conditional recommendation against for the same joints, at moderate certainty. Conditional means the panel expects informed people to differ, which is a milder verdict than the strong one glucosamine got. Collagen appears nowhere in that document, in either direction, so there is no grading of it to report at all.
Collagen has not been through what glucosamine has been through, and pretending it has would be as wrong as selling it. A 2023 meta-analysis found four randomized controlled trials, 507 people with knee osteoarthritis graded 1 to 3 on X-ray or 1 to 3 by functional class, run in Ecuador, Taiwan, China and the Czech Republic. Three of those trials, 375 people, went into the pain analysis: pain was lower with collagen peptide than with the comparison capsule, a standardized mean difference of minus 0.58 (95% confidence interval minus 0.98 to minus 0.18, p = 0.004), with 68% heterogeneity between the trials, which the authors graded moderate certainty of evidence. All four trials, 507 people, went into the side effect analysis, where the difference did not reach significance, an odds ratio of 1.66 (0.99 to 2.78, p = 0.05), graded very low certainty (Lin and colleagues, 2023). That interval does not rule out more side effects with collagen, so read it as not shown rather than as shown to be equal.
Then there is the small print. All four trials were rated at high risk of bias on the Cochrane tool, the authors say the pooled sample was small, and their conclusion asks for "well-designed randomized controlled trials" to confirm the result. A moderate certainty grade on top of four trials the same reviewers rated at high risk of bias is a tension inside the paper, and the risk of bias is the half to lean on. In one of the four trials the comparison group took glucosamine sulfate rather than an inert capsule, so the word placebo does not cover everything that was compared. Four trials is where glucosamine was decades ago.
Funding runs through this literature. The single trial most often quoted for collagen randomized 191 people with knee osteoarthritis, graded 2 or 3 on X-ray, at 13 centers in southern India, to an undenatured type II collagen supplement, to glucosamine plus chondroitin, or to placebo for 180 days, with 63 people in the collagen group, 66 in the glucosamine and chondroitin group and 62 on placebo. Total WOMAC score, the summed knee index used in this trial, runs from 0 to 2,400, and it fell by 551 with collagen, 454 with glucosamine and chondroitin and 414 with placebo at day 180, giving p values of 0.002 against placebo and 0.04 against the other supplement. Those three figures rest on the 164 people who finished: 54 in the collagen group, 57 in the glucosamine and chondroitin group, 53 on placebo. The study was sponsored by the company that makes the ingredient, two of its three authors were employees of that company and the third had consulted for it, all of which the paper discloses (Lugo and colleagues, 2016).
What is left is an open question with industry fingerprints on it, no guideline position either way, and a signal small enough that a few better trials could erase it. That is a different statement from "recommended against", and the two should not be blurred into each other.
Nothing here has been shown to rebuild cartilage
This is the claim to watch for, because it sells, and because what the X-ray studies found is not it. The plainest summary available comes from the National Center for Complementary and Integrative Health: "Whether glucosamine and chondroitin actually have an effect on joint structure is uncertain. Studies have had inconsistent results."
Inconsistent is the right word, and the detail matters here, because a page that says the structural trials all came back empty would be wrong. In the two-year US trial that measured the gap between the bones on X-ray in 572 participants, "changes in joint space width did not differ between the placebo group and any of the other groups". In a two-year Australian trial in 605 participants, the group taking glucosamine and chondroitin together did show a reduction in joint space narrowing, while neither supplement on its own did. Two further two-year trials of chondroitin alone, one in 622 people and one in 300, found joint space improvements against placebo, which the Australian and US trials did not. For the hip, the one high quality trial in 222 people found glucosamine no better than placebo for joint structure.
Here is what the labels blur. A gap on an X-ray is an indirect measure taken from a flat image of a joint, and a smaller loss of it is not cartilage growing back. No trial on this page looked at cartilage itself and found more of it. In the collagen trial, the markers of cartilage turnover in blood and joint fluid showed no significant difference between the groups, and the X-ray was taken at the beginning only, so that trial could not have detected a structural change even if one had happened.
So the fair position is narrower than "nothing changed" and far narrower than what gets printed on a label: the joint space findings contradict each other, nothing here has been shown to rebuild cartilage, and the guideline that weighed all of it still recommends strongly against glucosamine and chondroitin. Pain relief and cartilage repair are different claims with different evidence.
Safety, and the conversation to have first
These products are not dangerous for most people, and that is not the same as being free of consequences. The National Center for Complementary and Integrative Health states that "no major safety problems have been identified in large studies of glucosamine and chondroitin for osteoarthritis", then adds that glucosamine "may cause increases in blood glucose (sugar) levels in some people", and that both "have been associated with an increased risk of bleeding in people who are taking the anticoagulant warfarin". It also says little is known about their safety in pregnancy or while breastfeeding.
That makes the pharmacist the right first stop, not the internet. Take the list of everything you already take, including anything herbal, and ask about interactions before you add something. If you have diabetes, ask specifically about blood glucose. If you take warfarin, that bleeding signal is a reason to speak to the clinician who monitors your blood tests before you start anything, not after. Never stop or change a prescribed medicine on your own because a supplement is supposed to replace it.
There is a second cost that nobody prints on a label. Months spent waiting for a capsule to work are months not spent on the treatment that has the strong recommendation behind it.
What the same guidelines put first
Exercise, and weight management for people who need it. The 2026 rheumatology summary strongly recommends exercise for knee and hip osteoarthritis, at moderate certainty of evidence, and strongly recommends weight loss, also at moderate certainty, for people with knee or hip osteoarthritis who meet criteria for overweight or obesity. Those are stronger recommendations resting on better evidence than anything else on this page. NICE agrees on the order: recommendation 1.3.1 of NG226 says to offer therapeutic exercise tailored to their needs to everyone with osteoarthritis, and 1.2.2 asks clinicians to explain that the core treatments are therapeutic exercise and weight management where that applies. The NHS says the same in plainer words: "Exercise is one of the most important treatments for people with osteoarthritis, whatever your age or level of fitness."
Our programs are the staged version of that work. Start with knee osteoarthritis or hip osteoarthritis depending on the joint, hand osteoarthritis for finger and thumb joints that are sore and stiff, and rheumatoid arthritis if you have inflammatory joint disease, where your rheumatology team's plan always comes first. If your joints started aching around the menopause, joint pain in menopause explains what else changes at that point.
Each exercise page carries its own dose. Many knee and hip programs start at 2 to 3 sets of 10 to 15 repetitions, most days of the week, and build load slowly over 6 to 12 weeks; your physio will adjust this to your joint, your other conditions and what you want to get back to.
When to see a physio (physical therapist)
Before you spend anything, if the joint has not been assessed. A supplement cannot tell you which structure hurts, and the commonest reason a knee or hip program fails is that it was never matched to the problem in the first place.
Book sooner if a joint has been painful for more than a few weeks, if it is swelling after activity, if you are avoiding stairs or hills, or if pain has started waking you at night. Bring what you have already tried, including any supplement, how long you took it and what changed. The warning list below comes first, at the speed each line sets.
See a doctor promptly if
- Same day: a joint is hot, red and swollen, or you have joint pain with a high temperature or feel hot, cold or shivery. Get medical advice the same day, and go to an emergency department if you feel very unwell. This can be a sign of infection in the joint, which needs treatment quickly. Gout can look much the same, and a doctor needs to tell the two apart. If you have a very high or very low temperature, shivering you cannot control, you are breathing very fast or finding it hard to breathe, you are confused or your speech is slurred, your skin, lips or tongue look blue, gray, pale or blotchy (on brown or black skin this can be easier to see on the palms of the hands or soles of the feet), or you have a rash that does not fade when you press it, call emergency services, as this can be sepsis.
- Within a few days: morning stiffness in your hands that lasts longer than 30 minutes, soft, puffy swelling over the big knuckles at the base of the fingers or over the wrists, or painful, swollen joints in your feet as well. This is not an emergency, but see your doctor within a few days, because it can be a sign of an inflammatory arthritis such as rheumatoid arthritis rather than osteoarthritis, and early treatment can stop it getting worse.
- Within a few days: the pain is there at night or at rest and keeps getting worse. See your doctor within a few days. If you have had cancer, now or in the past, or are losing weight without trying, see your doctor sooner, within a day or two, and mention it. This applies even if the pain is not getting worse. If you are being treated for cancer now, contact your cancer team the same day.
Common questions
Does glucosamine work for knee pain?
The main guideline bodies say no, and the newest one says so in the strongest terms it has: the American College of Rheumatology osteoarthritis guideline summary approved in September 2026 states, "In people with knee, hip, and hand OA, we strongly recommend against glucosamine, chondroitin, and combination preparations of glucosamine and chondroitin." It grades that strong, with low certainty of evidence for glucosamine alone and moderate certainty for the combination, so the panel's confidence sits in the advice rather than in the trials behind it. The US National Center for Complementary and Integrative Health, reviewing the same field, writes that "it's still uncertain whether glucosamine and chondroitin are helpful for knee osteoarthritis symptoms", and describes a 2018 combined analysis of 29 studies in 6,120 participants in which "global pain was significantly reduced by glucosamine or chondroitin taken separately but not by the combination of the two". The UK recommendation is older and shorter: NICE guideline NG226 tells clinicians "Do not offer glucosamine or strong opioids to people to manage osteoarthritis", and, if someone asks about it, to explain that "there is no strong evidence of benefit for glucosamine"; NICE does not attach a strength and certainty pair to its recommendations, so that one is a plain instruction not to offer it. One European society disagrees about a prescription-only preparation that is not what is sold on a shop shelf, and if you are already taking a supplement and it seems to help you, discuss that with your doctor or pharmacist rather than stopping anything else to make room for it.
Does collagen help joint pain?
Possibly, modestly, on weak evidence, and no guideline has graded it. A 2023 meta-analysis found four randomized controlled trials in 507 people with knee osteoarthritis, and pooled three of them, 375 people, for pain: pain was lower with collagen peptide than with the comparison capsule, a standardized mean difference of minus 0.58 (95% confidence interval minus 0.98 to minus 0.18, p = 0.004), with substantial disagreement between the trials, which the authors graded as moderate certainty of evidence by the GRADE system. The same authors report that all four trials were at high risk of bias and that the total sample was small, and they call for well-designed trials to confirm the finding (Lin and colleagues, 2023); moderate certainty resting on four trials the same reviewers rated at high risk of bias is a tension inside that paper, and the risk of bias is the half to lean on. The 2026 American College of Rheumatology guideline summary does not mention collagen anywhere, in either direction, so no panel has graded it yet, and industry funding runs through this literature: the best known single trial of an undenatured type II collagen supplement was sponsored by the ingredient manufacturer, with two of its three authors employed by that company (Lugo and colleagues, 2016). Nothing in these trials shows collagen rebuilding cartilage, because the pooled analysis measured pain and side effects only, so treat it as an open question rather than a proven treatment, and never as a replacement for the exercise and weight management the guidelines recommend strongly.
Is it safe to take glucosamine or chondroitin with other medicines?
This is a question for a pharmacist, because two interactions are documented and both matter. The US National Center for Complementary and Integrative Health states that "no major safety problems have been identified in large studies of glucosamine and chondroitin for osteoarthritis", and then adds that "glucosamine may cause increases in blood glucose (sugar) levels in some people, and glucosamine and chondroitin have been associated with an increased risk of bleeding in people who are taking the anticoagulant warfarin". The same page says little is known about using them during pregnancy or while breastfeeding. Take a list of everything you already take, including anything you bought yourself, to a pharmacist before you start, especially if you are on a blood thinner or a diabetes medicine. Never stop or change a prescribed medicine on your own.
Can joint supplements rebuild cartilage?
No supplement has been shown to do that, and the X-ray studies that looked at joint structure contradict each other rather than supporting it; the National Center for Complementary and Integrative Health's summary is that "whether glucosamine and chondroitin actually have an effect on joint structure is uncertain" and that studies have had inconsistent results. The two-year US trial that measured joint space on X-ray in 572 participants found that "changes in joint space width did not differ between the placebo group and any of the other groups", while a two-year Australian trial in 605 participants did find less joint space narrowing in the group taking glucosamine and chondroitin together, and two further two-year trials of chondroitin alone, one in 622 people and one in 300, found joint space improvements. A gap on an X-ray is an indirect measure, and a smaller loss of it is not cartilage growing back: no trial here looked at cartilage itself and found more of it. The 2017 American Academy of Orthopaedic Surgeons guideline on hip osteoarthritis concluded that moderate strength evidence does not support glucosamine sulfate there, and that glucosamine was no better than placebo for pain, joint function or joint structure. In the collagen trial most often quoted, the markers of cartilage turnover in blood and joint fluid showed no significant differences between groups, and the X-ray was taken at the start only (Lugo and colleagues, 2016), so cartilage repair is the claim to treat as a warning sign about the seller.
Why do some European doctors still recommend glucosamine?
Because one society reads one specific preparation differently, and it is a prescription product in parts of Europe rather than a supplement off a shelf: a 2021 paper co-authored by members of the ESCEO working group, the European osteoarthritis and osteoporosis society behind that algorithm, states that "ESCEO specifically recommends the use of pCGS and long-acting chondroitin sulfate products", where pCGS means prescription crystalline glucosamine sulfate, and that other formulations "have consistently been shown to be ineffective in the treatment of knee OA". The same paper explains the split plainly: "ESCEO strongly supports the use of pCGS and chondroitin sulfate, whereas OARSI does not support their use. The main reason for this discrepancy is that the OARSI guidelines... do not recognize the concept of 'pharmaceutical-grade' or 'prescription-grade' SYSADOAs, such compounds being unavailable on the US soil" (Yeap and colleagues, 2021), and several of its authors declared advisory board work and conference support from pharmaceutical companies. The US National Center for Complementary and Integrative Health, summarizing a 2014 analysis of 25 glucosamine trials in 3,458 people with knee osteoarthritis, confirms that the trials of the prescription formulation gave more favorable results, and then says why that is hard to read: "Most of the studies of the prescription drug formulation had a high risk of bias because of weaknesses in their study design, were published more than 20 years ago, and were funded by the pharmaceutical company." The practical point for a reader in the USA, the UK, Canada, Australia or New Zealand is that the product the European algorithm is talking about is probably not the one in front of you, and the American and international guideline positions recommend against the ones that are.
References
- American College of Rheumatology. 2026 update of the American College of Rheumatology recommendations for the management of osteoarthritis of the knee, hip, and hand. Guideline summary, approved by the ACR Board of Directors on 8 September 2026. https://assets.contentstack.io/v3/assets/bltee37abb6b278ab2c/bltb3d12c34020da842/oa-guideline-summary-2026.pdf
- National Institute for Health and Care Excellence. Osteoarthritis in over 16s: diagnosis and management. NICE guideline NG226. Published 19 October 2022. Read 1 October 2026. https://www.nice.org.uk/guidance/ng226/chapter/Recommendations
- National Center for Complementary and Integrative Health. Glucosamine and chondroitin for osteoarthritis: what you need to know. Last updated October 2023. Read 1 October 2026. https://www.nccih.nih.gov/health/glucosamine-and-chondroitin-for-osteoarthritis-what-you-need-to-know
- NHS. Osteoarthritis: treatment. Page last reviewed 20 March 2023. Read 1 October 2026. https://www.nhs.uk/conditions/osteoarthritis/treatment/
- Lin CR, Tsai SHL, Huang KY, et al. Analgesic efficacy of collagen peptide in knee osteoarthritis: a meta-analysis of randomized controlled trials. Journal of Orthopaedic Surgery and Research. 2023. doi:10.1186/s13018-023-04182-w
- Lugo JP, Saiyed ZM, Lane NE. Efficacy and tolerability of an undenatured type II collagen supplement in modulating knee osteoarthritis symptoms: a multicenter randomized, double-blind, placebo-controlled study. Nutrition Journal. 2016;15:14. doi:10.1186/s12937-016-0130-8
- Yeap SS, Tanavalee A, Perez EC, et al. 2019 revised algorithm for the management of knee osteoarthritis: the Southeast Asian viewpoint. Aging Clinical and Experimental Research. 2021. doi:10.1007/s40520-021-01834-x
Written and checked by the PocketPhysio editorial team. Last updated 2026-10-02.
This page is general education, not a diagnosis or a personal treatment plan. See a physiotherapist or doctor for advice about your own situation.