HealFern

Best supplements for joint pain, ranked by the evidence

This page has affiliate links. A purchase through them can pay HealFern a commission at no extra cost to you, and it never changes what we report. How HealFern makes money.

Short answer

Curcumin and Boswellia have the most consistent evidence for knee arthritis pain, though the benefit is small and many trials are low quality. Glucosamine and chondroitin did no better than placebo in the 1,583-person GAIT trial. Fish oil may help rheumatoid arthritis alongside medicine. Vitamin D did not help knees, and thunder god vine is risky. Exercise still comes first.

HealFern card on a deep plum background with faint capsule outlines, reading Joint health, Joint supplements ranked, and a line saying seven ingredients are graded by their trials.
The picks at a glance
#PickBest forScore
01Curcumin (turmeric extract)Knee osteoarthritis pain, short trialModerate
02BoswelliaKnee osteoarthritis, thinner dataModerate
03Glucosamine and chondroitinHand osteoarthritis (chondroitin)Limited
04Fish oilRheumatoid arthritis, alongside medicineLimited
05Undenatured type II collagen (UC-II)Knee osteoarthritis, after other optionsLimited
06Vitamin DCorrecting a low level, not knee painWeak
07Thunder god vineNo one, safety risksAvoid

We read the NCCIH pages on glucosamine, turmeric and thunder god vine, the 2019 American College of Rheumatology guideline, the 1,583-person GAIT trial, and the pooled analyses of curcumin and Boswellia. Then we graded each ingredient type by the size and quality of its human trials.[1][2][3][6][7]

Curcumin and Boswellia show the most consistent small benefit for knee arthritis pain, glucosamine and chondroitin are a coin toss, and no supplement replaces exercise, weight loss or standard pain medicine.[8][9][1][6]

Joint supplements at a glance
What they are Plant extracts (turmeric, Boswellia), cartilage building blocks (glucosamine, chondroitin, collagen) and fish oil, sold in the US as dietary supplements.[1][2]
Used for Mostly knee and hand osteoarthritis pain. Fish oil has also been studied in rheumatoid arthritis.[1][12]
Amounts studied Glucosamine 1,500 mg and chondroitin 1,200 mg a day; UC-II collagen 40 mg a day; fish oil 0.45 g to 4.5 g of omega-3 a day.[7][13][11]
Best evidence Curcumin and Boswellia, from pooled small trials of mixed quality.[8][9]
Safety Mostly mild stomach upset. Enhanced-absorption curcumin has been linked to liver damage, and glucosamine may raise blood sugar.[2][1]
Who should avoid Pregnant people (turmeric, thunder god vine), people on warfarin without their doctor’s approval, and anyone taking thunder god vine at all.[2][3][1][12]
Guideline view The ACR and Arthritis Foundation strongly recommend against glucosamine and chondroitin for knee osteoarthritis.[1]
HealFern infographic ranking joint supplements by evidence. Three stats: 1,583 people took part in the GAIT trial, where glucosamine and chondroitin did not beat placebo; 60.1% of the GAIT placebo group improved anyway; 37 of 48 outcomes in curcumin reviews were rated extremely low quality evidence. Evidence grades: curcumin Moderate, small drops in knee pain; Boswellia Moderate, 7 trials with 545 people, lower pain and stiffness after at least 4 weeks; glucosamine and chondroitin Limited, GAIT 1,583 people for 24 weeks, not significantly better than placebo for knees; fish oil Limited, fewer anti-inflammatory drugs in rheumatoid arthritis and 4.5 g a day no better than 0.45 g for knee arthritis; vitamin D Weak, 413 people on 50,000 IU monthly for 2 years saw no change in knee pain; thunder god vine Avoid, kidney and liver damage, low white cell counts and rare deaths reported. Takeaways: curcumin or Boswellia is the most reasonable try for a fixed few weeks; glucosamine and chondroitin did not beat placebo for knees; exercise and weight loss have stronger evidence than any capsule. Talk to a doctor first if you take warfarin or diabetes medicine, are pregnant, have liver disease, or have joint pain with fever or weight loss.
Joint supplements graded by evidence. Sources: NCCIH, NIH Office of Dietary Supplements, the GAIT trial (NEJM 2006), Chen 2026, Yu 2020, Hill 2016 and Jin 2016 (JAMA).

How we graded: Moderate means several pooled trials with a small, consistent effect. Limited means some trials, but small, inconsistent or low quality. Weak means pilot data only, or good trials that found no benefit. Avoid means little evidence for joint pain plus a real safety problem. We rank ingredient types, never brands, and this page has no affiliate links. Our full method is on the how we research page.

The main studies behind the rankings
Study Who and how many Dose and length Result
GAIT trial, NEJM 2006[7] 1,583 adults with knee osteoarthritis Glucosamine 1,500 mg, chondroitin 1,200 mg, or both, for 24 weeks Not significantly better than placebo overall. The drug celecoxib was.
Chen review of reviews, 2026[8] 7 systematic reviews of curcumin for knee arthritis Varied doses and products Possible benefit, but 37 of 48 outcomes rated extremely low quality evidence
Yu meta-analysis, 2020[9] 7 trials, 545 people with osteoarthritis Boswellia extracts, at least 4 weeks Lower pain and stiffness scores, better function
Hill trial, 2016[11] 202 adults with knee osteoarthritis 4.5 g vs 0.45 g omega-3 daily for 2 years High dose no better. Low dose did better on pain at 2 years.
Lugo trial, 2016[13] 191 adults with knee osteoarthritis UC-II 40 mg daily for 180 days Better symptom scores than placebo. Two of three authors worked for a supplement company.
Jin trial, JAMA 2016[14] 413 adults with knee osteoarthritis and low vitamin D 50,000 IU vitamin D3 monthly for 2 years No difference in knee pain or cartilage loss

1. Curcumin (turmeric extract): knee arthritis pain

Curcumin, the main active compound in turmeric, has the most positive pooled research for knee osteoarthritis. NCCIH says several meta-analyses found improvements in knee pain, stiffness and mobility, and calls the early evidence positive while saying higher-quality studies are still needed.[2] Grade: Moderate.

The reviews behind that finding are weak, though. A 2026 critical appraisal of 7 systematic reviews agreed curcumin showed potential benefit, then rated the reviews themselves as extremely low in method quality. Of 48 outcomes, 37 rested on extremely low quality evidence, and the reviews overlapped heavily.[8]

A 2025 network meta-analysis of 39 trials also found curcumin helped some outcomes, with a possible edge for joint function.[10]

Products vary a lot. Many add piperine from black pepper to raise absorption, so two bottles labeled “turmeric” can deliver very different amounts.[2] That matters for safety too. Plain turmeric is likely safe in usual amounts for 2 or 3 months, but enhanced-absorption curcumin has been linked to liver damage in some people.[2]

Who it suits: adults with knee osteoarthritis who want a time-limited trial, have no liver disease and are not pregnant.[2]

2. Boswellia: knee arthritis, if you accept thinner data

Boswellia serrata, a tree resin extract, showed consistent pain relief in small trials. A 2020 meta-analysis of 7 trials with 545 people found lower pain and stiffness scores and better function than control groups, and suggested taking it for at least 4 weeks before judging.[9] Grade: Moderate, with less certainty than curcumin.

The 2025 network meta-analysis compared seven supplements across 39 randomized trials and 4,599 patients. Boswellia had the highest chance of ranking first for pain and stiffness, and no supplement in the analysis caused more side effects than placebo.[10]

The total number of people studied is small, and the authors asked for large, well-designed trials with standard doses before anyone treats these results as settled.[10] Boswellia also does not appear among the treatments the 2019 ACR guideline recommends.[6]

Extracts differ from brand to brand, and the trials used different products. If you try it, pick one standardized extract and stay with it for the whole trial window.

Who it suits: people with knee osteoarthritis who have tried exercise and standard pain relief and want to test one more option with a doctor’s knowledge.

3. Glucosamine and chondroitin: hand arthritis, or a fair trial for knees

Glucosamine and chondroitin are the most studied joint supplements and among the most disappointing. NCCIH says it is still uncertain whether they help knee osteoarthritis symptoms, and the ACR and Arthritis Foundation strongly recommend against them for the knee.[1] Grade: Limited.

The largest US trial, GAIT, gave 1,583 people glucosamine, chondroitin, both, the drug celecoxib, or placebo for 24 weeks. On placebo, 60.1% improved. Glucosamine added 3.9 percentage points, chondroitin 5.3 and the combination 6.5, none of them statistically significant. Celecoxib added 10.0 points, which was.[7]

A 2018 analysis of 29 studies with 6,120 people found each one alone reduced pain but the combination did not.[1] Prescription glucosamine sold in Europe tended to do better in trials, but most of those studies were old, at high risk of bias and funded by the manufacturer.[1]

Hands are the exception. In one 6-month trial of 162 people, chondroitin improved hand pain and function, and the ACR guideline conditionally recommends chondroitin for hand osteoarthritis.[1][6]

Who it suits: people with hand osteoarthritis, or anyone who wants a 3 to 6 month trial and will stop if nothing changes.

4. Fish oil: rheumatoid arthritis, alongside your medicine

Fish oil has more support in rheumatoid arthritis than in osteoarthritis. In rheumatoid arthritis trials, omega-3 supplements helped people use fewer anti-inflammatory drugs, but their effect on painful or swollen joints and morning stiffness was inconsistent.[12] Grade: Limited.

For knee osteoarthritis, the best trial we found compared two doses and had no placebo. Hill and colleagues gave 202 people either 4.5 g or 0.45 g of omega-3 a day for 2 years. The high dose gave no extra benefit, and the low-dose group actually reported more pain relief at 2 years.[11] Without a placebo group, that trial cannot show either dose beats nothing.

Safety depends on dose. FDA has concluded that supplements giving up to 5 g a day of EPA and DHA are safe when used as directed. Two large trials found 4 g a day for several years slightly raised the risk of atrial fibrillation in people with or at high risk of heart disease. Fish oil may also raise clotting times in people on warfarin, though most research found 3 to 6 g a day did not change anticoagulation much.[12]

Who it suits: people with rheumatoid arthritis who keep taking their prescribed drugs and clear fish oil with their rheumatologist.

5. Undenatured type II collagen: a small bet with company ties

Undenatured type II collagen (UC-II) comes from chicken sternum cartilage and is taken in a tiny 40 mg daily dose. Its main trial looked good, but the evidence is thin and the trial had industry authors.[13] Grade: Limited.

In the Lugo trial, 191 adults with knee osteoarthritis took UC-II, glucosamine with chondroitin, or placebo for 180 days. The UC-II group improved more on the WOMAC symptom score than both the placebo group and the glucosamine group, and side effects did not differ between groups.[13]

Two of the three authors listed InterHealth Nutraceuticals, a supplement company, as their employer.[13] It is still a single trial, and an independent replication would be needed to raise the grade. The 2025 network meta-analysis did find collagen helped in some outcomes.[10]

If you came here comparing collagen powders for skin, our collagen for wrinkles guide covers that research.

Who it suits: people who have tried the higher-ranked options and want a low-dose capsule, knowing the evidence rests mostly on one company-linked trial.

Which joint supplements should you skip?

Two popular choices did poorly. Vitamin D did not ease knee pain in a two-year trial, even in people who started with low levels, and thunder god vine carries risks of kidney and liver damage, with rare deaths reported.[14][3]

6. Vitamin D: Weak

In the Jin trial, 413 adults with knee osteoarthritis and low vitamin D took 50,000 IU of vitamin D3 or placebo once a month for 2 years. Blood levels rose, but knee pain and cartilage volume did not differ from placebo. Adverse events were reported by 56 people on vitamin D and 37 on placebo.[14] The good reason to take vitamin D is a low level found by your doctor. Our vitamin D2 vs D3 guide explains which form raises levels more.

7. Thunder god vine: Avoid

Thunder god vine is a Chinese herb promoted for rheumatoid arthritis. NCCIH says it might help some symptoms, but its side effects include lowered white blood cell counts, kidney damage and liver problems, rare deaths have been reported, and it is considered unsafe in pregnancy because of possible birth defects.[3] NCCIH also says it should not replace standard rheumatoid arthritis care, which can delay joint damage.[3]

Why do joint supplements show such small effects?

Two things shrink the measured benefit. Osteoarthritis damages the whole joint, not just cartilage, so a capsule aimed at one tissue has limited reach. And arthritis trials have a strong placebo response, so a supplement has to beat the sizable relief people get from a dummy pill.[17][18]

Osteoarthritis is more than worn cartilage

NIAMS says osteoarthritis does not come from simple wear and tear. Changes in the joint tissue start a gradual breakdown that can reach the cartilage, tendons and ligaments, the joint lining, the bone and, in the knee, the meniscus.[17]

Pain then feeds itself. People with sore joints move less, the muscles around the joint weaken, and weaker muscles put more stress on the joint.[17] That cycle helps explain why exercise is in the guideline and capsules are not.[6]

Most supplement trials measure pain and function scores, not repair. The one trial in this list that measured cartilage directly, the Jin vitamin D trial, found no difference in cartilage volume after 2 years.[14]

The placebo response is large in arthritis trials

A 2008 meta-analysis pooled 198 randomized trials with 16,364 people in placebo groups. Placebo relieved osteoarthritis pain with an effect size of 0.51, while untreated control groups barely changed, at 0.03. Placebo also improved stiffness and function.[18]

The placebo response was bigger when starting pain was higher and when the dummy treatment came by injection or needle.[18] GAIT shows the same thing: 60.1% of people on placebo improved.[7]

So feeling better after you start a supplement does not prove the supplement did it. Score your pain for a week before you start, keep the rest of your routine steady, and compare at the end of your trial window.

How do you choose a joint supplement and read the label?

Pick the ingredient and amount used in the trials, not the boldest claim on the bottle. Store products can differ from the ones researchers tested, the FDA does not review supplements for safety or effectiveness before they go on sale, and words like “standardized” do not guarantee quality.[15]

Match the trial dose and length

The table lists what the main trials used. If a product gives far less, or does not state the amount per serving, you cannot tell whether it matches the research.

What the trials used, and what to check on the label
Ingredient Amount studied Trial length Label check
Glucosamine 1,500 mg a day[7] 24 weeks Amount per daily serving, not per capsule
Chondroitin 1,200 mg a day[7] 24 weeks Amount per daily serving; ask first if you take warfarin[1]
Curcumin Varied across trials[8] Varied Whether it is an enhanced-absorption formula or adds piperine[2]
Boswellia Varied extracts[9] At least 4 weeks[9] One named extract, kept for the whole trial
Fish oil 0.45 g omega-3 a day did as well as 4.5 g[11] 2 years Grams of EPA and DHA per serving, not grams of oil[12]
UC-II collagen 40 mg a day[13] 180 days The words undenatured type II collagen, the form tested

What a quality seal does and does not mean

Seals from groups such as the U.S. Pharmacopeia (USP) and NSF International mean the product was made under good manufacturing practices, contains what the label lists, and has no harmful levels of contaminants such as lead. NIA suggests looking for the USP verified mark.[16]

A seal says nothing about whether the ingredient eases joint pain. Supplement labels also cannot claim to cure or prevent a disease. Claims about how a product affects the body must carry a disclaimer saying the FDA has not evaluated them.[15]

Skip big multi-ingredient joint formulas if you want a clear answer. NIA advises against buying supplements with ingredients you do not need.[16] One ingredient at a time also lets you judge what helped.

Are joint supplements safe, and who should avoid them?

Most joint supplements caused only mild problems in large trials. In GAIT, side effects were mild, infrequent and evenly spread across groups, and NCCIH reports no major safety problems for glucosamine and chondroitin in large studies.[7][1] The risks sit with specific groups and specific products.

See a doctor soon if you have a fever without flu symptoms, have lost 10 pounds or more without trying, have joint pain that lasts more than several days, or have severe, unexplained joint pain and swelling.[5] Gout causes sudden painful joints too; our gout diet guide covers it.

Who should not take joint supplements at all?

Some people are better off not starting: anyone thinking about thunder god vine, people who are pregnant or nursing, people with surgery coming up, and people whose joint pain has no diagnosis yet. Older adults on several medicines should start only after a doctor or pharmacist checks for interactions.[3][15][16]

If you are pregnant or breastfeeding, know that many supplements have never been tested in pregnant women, nursing mothers or children.[15] Turmeric supplements may be unsafe in pregnancy, and thunder god vine is considered unsafe because of possible birth defects.[2][3]

If you have surgery coming up, be aware that some supplements may raise bleeding risk or change how you respond to anesthesia, so tell your surgical team about everything you take, as early as you can.[15] Glucosamine, chondroitin and fish oil already carry bleeding cautions for warfarin users.[1][12]

For older adults on many medicines, NIA warns that a supplement can make a prescribed medicine weaker or stronger, and suggests keeping a written list of every medicine and supplement to bring to each visit.[16] Osteoarthritis is more common with age, so this group overlaps heavily with people shopping for joint pills.[17]

If nobody has diagnosed your joint pain yet, start there. Rheumatoid arthritis, gout and osteoarthritis need different care, and standard rheumatoid arthritis care can delay joint damage.[3][4] See a doctor soon for fever or severe, unexplained joint pain and swelling.[5]

People with liver disease have a specific reason to pass on curcumin. The enhanced-absorption kind has been linked to liver damage.[2] NCCIH also notes that sudden liver injury from supplements, while rare, is being reported more often.[15]

Does exercise work better than a supplement for knee pain?

Yes, and the evidence behind it is much stronger. A Cochrane review of 44 trials with 3,537 people found land-based exercise cut knee osteoarthritis pain by about 12 points on a 100-point scale, rated high-quality evidence. No supplement in this ranking has evidence of that quality.[19][8]

What the exercise trials found

In the 2015 Cochrane review, exercise also improved physical function by about 10 points and quality of life by about 4 points on 100-point scales, measured right after the program ended.[19]

Part of the benefit lasted. Twelve studies with 1,468 people checked again 2 to 6 months later and found pain still about 6 points lower. People dropped out of exercise groups no more often than control groups, 14% versus 15%.[19]

Side effects were limited to more knee or low back pain in 8 studies, and no study reported a serious adverse event. Cochrane updated the review in 2024; we read the 2015 version.[19]

How the two compare

The numbers come from different scales, so they cannot be lined up point for point. The quality gap is plain, though. Most curcumin review outcomes rested on extremely low quality evidence, while the exercise result for pain was rated high quality.[8][19]

The ACR guideline strongly recommends exercise and, for people with knee or hip arthritis who are overweight, weight loss.[6] Being overweight is also a listed risk factor for getting osteoarthritis in the first place.[17] If you add a supplement, keep the exercise plan going.

What do doctors recommend first for joint pain?

For osteoarthritis, doctors start with exercise, weight loss when needed, and braces or shoe inserts. Over-the-counter pain relievers and creams come next, then joint injections or prescription drugs, and surgery when nothing else works.[4] Supplements sit outside that list.

The 2019 ACR and Arthritis Foundation guideline strongly recommends exercise, weight loss for people with knee or hip arthritis who are overweight, tai chi, self-management programs, canes, knee braces, topical and oral NSAIDs, and steroid injections for the knee. It conditionally recommends acupuncture, heat and cold, topical capsaicin, acetaminophen, duloxetine and tramadol.[6] Osteoarthritis has no cure, but these steps ease pain and help people move.[4]

Checklist before you buy a joint supplement

More guides like this one are in our joint health and best of sections.

So which joint supplement is worth trying?

If you want to try one supplement for knee osteoarthritis, curcumin or Boswellia is the most reasonable pick, for a fixed trial of a few weeks, with your doctor aware. The benefit in trials was small and the studies were often low quality, so stop if you feel no change.[8][9]

Our judgment call: glucosamine and chondroitin are not worth the money for most knees, since the biggest US trial found no real gain over placebo. Chondroitin for hand arthritis is the exception worth discussing. Spend the effort on exercise first.[7][1][6]

Questions people also ask

  1. What is the best supplement for joint pain?

    Curcumin and Boswellia have the most consistent trial evidence for knee osteoarthritis pain. The benefit is small, many studies are low quality, and neither replaces exercise, weight loss or standard pain relievers. Try one at a time with your doctor aware.

  2. Does glucosamine actually work for knee pain?

    Probably not for most people. In the 1,583-person GAIT trial, glucosamine was not significantly better than placebo, and the ACR and Arthritis Foundation strongly recommend against it for knee osteoarthritis. Chondroitin for hand arthritis has better support.

  3. Is turmeric safe to take every day for arthritis?

    NCCIH says plain turmeric is likely safe in usual amounts for 2 or 3 months. Enhanced-absorption curcumin products have been linked to liver damage, and turmeric supplements may be unsafe in pregnancy. Stop and call a doctor if you notice dark urine or yellow skin.

  4. Does fish oil help joint pain?

    In rheumatoid arthritis trials, fish oil helped people use fewer anti-inflammatory drugs, but its effect on joint pain and swelling was inconsistent. For knee osteoarthritis, a 2-year trial found a high dose gave no extra benefit over a low dose.

  5. Does vitamin D help arthritis pain?

    Not in the best trial we found. In 413 adults with knee osteoarthritis and low vitamin D, 50,000 IU monthly for 2 years raised blood levels but did not change knee pain or cartilage loss compared with placebo.

  6. How long should I try a joint supplement?

    Boswellia trials suggest at least 4 weeks, and GAIT ran 24 weeks. Write down your pain score before you start, set a fixed trial window, and stop if nothing changes. Tell your doctor, especially if you take warfarin or diabetes medicine.

Sources

  1. Glucosamine and Chondroitin for Osteoarthritis: What You Need To Know, NCCIH, NIH
  2. Turmeric: Usefulness and Safety, NCCIH, NIH
  3. Thunder God Vine: Usefulness and Safety, NCCIH, NIH
  4. Osteoarthritis, MedlinePlus, National Library of Medicine
  5. Joint pain, MedlinePlus Medical Encyclopedia
  6. 2019 American College of Rheumatology/Arthritis Foundation Guideline for the Management of Osteoarthritis of the Hand, Hip, and Knee, Kolasinski et al., Arthritis Care & Research, 2020
  7. Glucosamine, chondroitin sulfate, and the two in combination for painful knee osteoarthritis (GAIT), Clegg et al., New England Journal of Medicine, 2006
  8. A critical review of systematic reviews and meta-analyses of curcumin for knee osteoarthritis, Chen et al., Frontiers in Pharmacology, 2026
  9. Effectiveness of Boswellia and Boswellia extract for osteoarthritis patients: a systematic review and meta-analysis, Yu et al., BMC Complementary Medicine and Therapies, 2020
  10. Comparative Effectiveness of Nutritional Supplements in the Treatment of Knee Osteoarthritis: A Network Meta-Analysis, Zhang et al., Nutrients, 2025
  11. Fish oil in knee osteoarthritis: a randomised clinical trial of low dose versus high dose, Hill et al., Annals of the Rheumatic Diseases, 2016
  12. Omega-3 Fatty Acids: Health Professional Fact Sheet, NIH Office of Dietary Supplements
  13. Efficacy and tolerability of an undenatured type II collagen supplement in modulating knee osteoarthritis symptoms, Lugo et al., Nutrition Journal, 2016
  14. Effect of Vitamin D Supplementation on Tibial Cartilage Volume and Knee Pain Among Patients With Symptomatic Knee Osteoarthritis, Jin et al., JAMA, 2016
  15. Using Dietary Supplements Wisely, NCCIH, NIH
  16. Dietary Supplements for Older Adults, National Institute on Aging, NIH
  17. Osteoarthritis: Symptoms, Causes and Risk Factors, NIAMS, NIH
  18. The placebo effect and its determinants in osteoarthritis: meta-analysis of randomised controlled trials, Zhang et al., Annals of the Rheumatic Diseases, 2008
  19. Exercise for osteoarthritis of the knee, Fransen et al., Cochrane Database of Systematic Reviews, 2015

This article is general information, not medical advice. These statements have not been evaluated by the FDA, and supplements are not intended to diagnose, treat, cure or prevent any disease. Talk to your doctor or pharmacist before you start a supplement. How we research.

+

More on Joint Health