Short answer: Both beat placebo for knee osteoarthritis pain and function in meta-analyses, but the trials are mostly small and low quality.1 Turmeric extracts did about as well as NSAID pain relievers in some trials.2 A network analysis favored boswellia for function and modified curcumin for pain.3 NIH warns highly absorbable curcumin may harm the liver.4
| Turmeric / curcumin | Boswellia | |
|---|---|---|
| Versus placebo | Less knee pain and better function across 11 placebo trials2 | Less pain and stiffness, better function across 7 trials5 |
| Versus NSAIDs | Similar effect, with fewer stomach side effects1 | No trials against approved NSAIDs as of 20181 |
| Network analysis (20 trials) | Modified formulations lowered pain scores3 | Modified extracts improved pain, stiffness and function scores3 |
| Mixed results | Less benefit in people with higher BMI2 | Overall effect not significant in one 2024 meta-analysis; significant only in subgroups6 |
| Safety | Highly absorbable curcumin formulations may harm the liver4 | Side effects similar to placebo in pooled trials1 |
A 2021 meta-analysis of 16 trials in 1,810 adults with knee osteoarthritis, up to 16 weeks long, found turmeric extracts reduced pain and improved function versus placebo and had similar effects to NSAIDs, with fewer side effects than NSAIDs. Results varied a lot, and benefit was smaller with higher BMI.2 NIH says we do not know enough to conclude turmeric or curcumin helps any health condition, products vary in what they contain, and highly absorbable curcumin may harm the liver.4
A 2020 meta-analysis of 7 trials in 545 people found boswellia reduced pain and stiffness and improved function.5 A 2024 meta-analysis of 13 trials found no significant overall effect because the trials differed so much, though subgroup analysis favored boswellia over placebo.6
Looking at both together, a 2018 meta-analysis of 11 trials found both better than placebo, with low study quality and mostly fewer than 100 people per trial.1 A 2026 network analysis of 20 trials found modified boswellia extracts improved function and modified curcumin lowered pain.3
Neither clearly wins. Both show modest benefit in small trials of mixed quality, and the authors of the 2018 review said the evidence is not good enough for practice recommendations.1 This is general information, not personal advice. More detail: the Curcumin and Boswellia (Boswellia serrata) guides.