Curcumin, the active compound group in turmeric, has reduced pain and WOMAC scores in knee osteoarthritis across multiple meta-analyses, with some finding it comparable to anti-inflammatory drugs and no increase in side effects. The certainty of that evidence is low: a 2025 critical review rated every existing systematic review of curcumin for knee osteoarthritis as critically low quality. Turmeric powder in food is roughly 2 to 5% curcuminoids, which is why the spice and the extract are not the same intervention.
Turmeric is one of several supplements sold for joints, and how it compares with the rest is set out in PhysioHub’s assessment of what supplement evidence supports.
Ranked by how often each one is the reason turmeric does nothing for someone taking it.
| Factor | The common error | What to do instead |
|---|---|---|
| 1. Curcuminoid content | Reading the turmeric figure on the front | Read the curcuminoid figure on the back |
| 2. Absorption | Plain curcumin powder, which is poorly absorbed | A formulation designed for bioavailability, or piperine |
| 3. Duration | Judging it after two weeks | 8 to 12 weeks, then stop and compare |
| 4. The condition | Assuming all arthritis responds the same | Most evidence is in knee osteoarthritis |
| 5. Interactions | Assuming a spice cannot interact with medication | Check with a pharmacist if on anticoagulants |
The first two account for most of the disappointment. Someone takes a product whose label emphasises 1000mg of turmeric root, which contains perhaps 30mg of curcuminoids in a form the gut barely absorbs, and concludes turmeric does not work for them. What they tested was not what the trials tested.
This is a documentary review of published trial evidence rather than a recommendation for you. Five factors were selected because each is a variable that plausibly determines whether an individual sees any effect, and they are ranked by how commonly each goes wrong.
The sources read were a 2025 critical review of systematic reviews and meta-analyses of curcumin for knee osteoarthritis, which assessed the quality of the reviews themselves; a systematic review and network meta-analysis of turmeric products for knee osteoarthritis; and a systematic review and meta-analysis of curcuminoids alone for pain and function in knee osteoarthritis. Evidence checked in August 2026.
The honest position, stated plainly: the individual trials point consistently in a positive direction, and the methodological quality of the reviews summarising them is poor enough that the size of the real effect is genuinely uncertain. That is an unusual combination and it is the most important thing on this page.
No prices or specific products are named, because retail costs could not be verified for this review and no brand comparison was carried out.
Find the curcuminoid figure. It is usually on the back, often as a percentage of a standardised extract. That number is the active dose, and the large number on the front is usually the weight of turmeric root.
Look for a named delivery system or piperine. Plain curcumin is poorly absorbed. Products either add black pepper extract or use a formulation designed to improve absorption.
Do not compare doses across formulations. A bioavailability-enhanced product at a lower stated dose may deliver more than a plain extract at a higher one, which makes direct dose comparison between brands meaningless.
Check what else is in it. Combination joint products frequently include a small amount of turmeric alongside several other ingredients, none at a tested dose.
Remember that food is not the intervention. Turmeric in cooking is a good ingredient and delivers a small fraction of any trial dose. Nothing here is an argument against using the spice.
What it is
Turmeric is the root. Curcuminoids are the group of compounds within it that the trials tested, of which curcumin is the main one. Turmeric powder is only around 2 to 5% curcuminoids by weight.
Why it matters most
The gap between the two is roughly twentyfold to fiftyfold. A product stating 500mg of turmeric root is a completely different proposition from one stating 500mg of standardised curcuminoid extract, and both can be labelled as turmeric supplements.
What the evidence supports
Trials use standardised curcuminoid extracts rather than turmeric powder, which is why the trial results do not transfer to eating more curry. Doses in trials vary widely, and the important comparison is curcuminoid content rather than the weight of the source material.
Limits and cautions
Because effective doses differ so much by formulation, no single number can be given as the right dose. What can be said is that a product not stating its curcuminoid content is not giving you enough information to judge it.
Why it ranks first
It is the single most common reason someone takes turmeric for three months and notices nothing.
What it is
Curcumin is poorly absorbed from the gut, rapidly metabolised and quickly cleared, so a substantial proportion of what is swallowed never reaches the bloodstream in an active form.
How products address it
The oldest approach is adding piperine from black pepper, which slows metabolism of curcumin and substantially increases how much reaches the circulation. Newer approaches include phospholipid complexes, micelle formulations and particle-size reduction, all sold under proprietary names.
What the evidence supports
A network meta-analysis of turmeric products compared conventional curcuminoid preparations against bioavailability-enhanced ones and found all preparations significantly reduced WOMAC pain, with different effect sizes between categories. Formulation is a real variable rather than marketing alone.
Limits and cautions
Better absorption raises the interaction risk as well as the effect, which is directly relevant for anyone on anticoagulants. Piperine also affects the metabolism of other drugs, so it is not a neutral addition.
Why it ranks second
It determines how much of the correct compound actually reaches the joint, and it is the variable most obscured by proprietary branding.
What it is
The duration over which any effect builds, and the period after which a fair judgement can be made.
What the trials did
Most trials of curcumin in knee osteoarthritis run between six weeks and four months, with meaningful differences from placebo typically emerging over several weeks rather than days. There is little evidence for an immediate effect.
What to actually do
Give it eight to twelve weeks at a known curcuminoid dose in an absorbable formulation, then stop for four weeks and compare. Stopping is the part people skip, and it is the only informative half of the experiment, since joint pain fluctuates over exactly this timescale.
Limits and cautions
Long-term safety data beyond a few months is limited, so indefinite use is not well studied. That is an absence of evidence rather than evidence of harm.
Why it ranks third
It costs nothing to get right, and without it there is no way to distinguish the supplement from the natural course of the condition.
What it is
The question of whether the evidence applies to your particular joint problem.
Where the evidence sits
Almost all of it is in knee osteoarthritis. That is where the trials were done and where meta-analyses report reductions in pain, WOMAC total, pain, function and stiffness subscales.
Where it is thinner
Rheumatoid arthritis has some supporting work, and it is a much smaller body of evidence than for knee osteoarthritis, and omega-3 has better support in inflammatory arthritis. Other sites of osteoarthritis, including hands and hips, have little direct evidence. Tendinopathy has essentially none, despite curcumin frequently being recommended for it.
What the evidence supports
A meta-analysis of curcuminoids alone found significant improvements against placebo across VAS pain and WOMAC total, pain, function and stiffness scores in knee osteoarthritis, with curcuminoids not inferior to NSAIDs on pain and function outcomes.
Limits and cautions
Not inferior to NSAIDs in these trials is a statement about the trials rather than a licence to stop prescribed medication. Discuss any change with whoever prescribed it.
Why it ranks fourth
It determines whether the evidence applies to you at all, and most people taking it for knees are in the group it was tested in.
What it is
The situations where curcumin at supplement doses is a genuine medical consideration rather than a harmless kitchen ingredient.
Anticoagulants and antiplatelets
Curcumin can affect platelet function, so anyone taking warfarin, a direct oral anticoagulant, clopidogrel or regular aspirin should speak to a pharmacist before starting. Piperine-enhanced products increase this concern by increasing absorption.
Gallbladder and reflux
Curcumin stimulates gallbladder contraction, so it is best avoided by anyone with gallstones or bile duct obstruction. It commonly worsens reflux and indigestion, which is the most frequently reported side effect.
Other situations
It may lower blood sugar, which matters alongside diabetes medication. Iron absorption may be reduced with high doses. Safety in pregnancy at supplement doses is not established, and culinary amounts are considered fine.
What the evidence supports
Meta-analyses consistently report that curcuminoids do not significantly increase adverse events compared with placebo or with NSAIDs, which is a favourable overall safety picture. The interactions above are specific situations rather than general risks.
Why it ranks fifth
It applies to a minority of people, and for those it applies to it is the factor that matters more than any of the four above.
| Form | Curcuminoid content | Absorption | Matches trial conditions |
|---|---|---|---|
| Turmeric as a cooking spice | Around 2 to 5% of the powder | Poor | No |
| Turmeric root capsules | Low, often not stated | Poor | No |
| Standardised curcuminoid extract | Typically 95% of the extract | Poor without an enhancer | Partly |
| Curcuminoid extract with piperine | Typically 95% | Substantially better | Yes |
| Bioavailability-enhanced formulation | Varies, check the label | Designed for it | Yes |
The last column is the practical test. Two of these five are what the trials studied, and the other three are what most people buy.
If you have knee osteoarthritis and want to try it, choose a product that states its curcuminoid content and includes piperine or a named absorption technology, and give it eight to twelve weeks.
If you have taken turmeric for months with no effect, check the label before concluding it does not work for you. Turmeric root capsules with no stated curcuminoid content are not what was tested.
If you take an anticoagulant or antiplatelet, speak to a pharmacist first. This is the interaction that matters most and it is easy to check.
If you have gallstones, avoid supplement doses, since curcumin stimulates gallbladder contraction.
If you have rheumatoid arthritis, omega-3 has better evidence for inflammatory arthritis than curcumin does, and neither replaces disease-modifying medication.
If you just want to use more turmeric in cooking, that is a fine thing to do for its own sake and it is a different question from the one this page answers.
The question that comes up, almost always phrased the same way, is whether turmeric is worth bothering with. It is a good question and the honest answer has two halves that people rarely hear together. The trials are consistently positive, more so than for most things on a supplement shelf. And a critical review published in 2025 looked at the systematic reviews summarising those trials and rated all of them as critically low quality, which means the effect could be meaningful or could be considerably smaller than the headlines suggest.
What that translates to in a conversation is: it is reasonable to try, at a proper dose, for a defined period, while doing the things with larger effects. The specific failure worth heading off is someone taking a turmeric root capsule for two years, having never read the back of the label, and having never started strengthening the knee it was bought for. That is a lot of effort and expense directed at the smallest available lever.
Where a knee has been painful long enough that supplements are being considered, the musculoskeletal pain and injuries service is where the larger lever is.
Does turmeric help joint inflammation?
Multiple meta-analyses report that curcumin reduces pain, WOMAC total, pain, function and stiffness scores in knee osteoarthritis compared with placebo, with some finding it comparable to anti-inflammatory drugs. A 2025 critical review rated every systematic review on the topic as critically low quality, so the direction is positive and the true effect size is uncertain.
How much turmeric do I need for arthritis?
No single figure applies, because effective doses differ substantially between formulations. What matters is the curcuminoid content stated on the label rather than the weight of turmeric root, and whether the product includes piperine or a bioavailability-enhancing technology. Turmeric powder is only around 2 to 5% curcuminoids.
Why do turmeric supplements contain black pepper?
Curcumin is poorly absorbed and rapidly cleared. Piperine, the active compound in black pepper, slows its metabolism and substantially increases how much reaches the bloodstream. It also affects the metabolism of other drugs, which is why interactions matter more with piperine-containing products.
Is turmeric as good as ibuprofen for joint pain?
Some meta-analyses have found curcuminoids not inferior to NSAIDs for pain and function outcomes in knee osteoarthritis, with no increase in adverse events. Those are findings within trials of limited quality, and they are not a reason to stop a prescribed medication without discussing it with whoever prescribed it.
Who should not take turmeric supplements?
Anyone on warfarin, a direct oral anticoagulant, clopidogrel or regular aspirin should check with a pharmacist first, since curcumin can affect platelet function. It should be avoided with gallstones or bile duct obstruction, used cautiously with reflux, and supplement doses are not established as safe in pregnancy.
How long does turmeric take to work for joint pain?
Trials generally run between six weeks and four months, with differences from placebo emerging over weeks rather than days. Eight to twelve weeks is a fair trial, followed by stopping for four weeks to see whether anything changes, since joint pain fluctuates naturally over that timescale.
Is cooking with turmeric as good as taking a supplement?
No, not for the effects described in the trials. Turmeric powder is only around 2 to 5% curcuminoids and is poorly absorbed, so culinary use delivers a small fraction of any trial dose. It remains a perfectly good ingredient, and that is a separate question from the supplement evidence.
PhysioHub – Empowerment through Evidence-Based Education.
Every body is different. Book a thorough assessment and get advice that is specific to you and your goals.
Book an Appointment