FactoWiki

What supplements are best for joint pain and stiffness?

Last updated: August 2026 · Reviewed by the FactoWiki Editorial Team for clarity and source accuracy

The three with the most consistent human evidence are collagen peptides (or UC-II), Boswellia serrata, and curcumin in a bioavailability-enhanced form. Glucosamine and chondroitin are the most popular but the most equivocal. None of them matches an anti-inflammatory drug, and all take 8–12 weeks.

Key takeaways

  • UC-II collagen at 40 mg has better trial data than the far more common 10 g collagen peptide dose.
  • Boswellia is the fastest-acting of the group, sometimes showing effects within 5–7 days.
  • Plain curcumin is barely absorbed — the form on the label matters more than the milligrams.

The ones with real trial support

Undenatured type II collagen (UC-II) at 40 mg daily has outperformed glucosamine-chondroitin in head-to-head knee osteoarthritis trials, and works by an immune-tolerance mechanism rather than by supplying building blocks. Boswellia serrata standardised to AKBA inhibits the 5-LOX inflammatory pathway and is unusual in showing effects within a week or two. Curcumin has performed comparably to ibuprofen in some knee osteoarthritis trials — but only in enhanced-absorption forms such as phytosome, piperine-combined or nanoparticle preparations.

The popular but uncertain ones

Glucosamine and chondroitin dominate the shelf and have the most mixed literature of anything here. The large NIH-funded GAIT trial found no overall benefit versus placebo, with a possible signal in the moderate-to-severe pain subgroup. Sulphate forms are better studied than hydrochloride. If you try them, three months is a fair trial; if nothing has changed by then, nothing will.

What actually matters more

For most people with joint pain, load management, strengthening the muscles around the joint and body weight do more than any capsule. A 5 kg weight loss removes roughly 20 kg of force from the knee with each step. Supplements are a reasonable addition to that, and a poor substitute for it.

Key ingredients to understand

The ingredients most relevant to this question, with what the evidence actually supports:

Frequently asked questions

How much improvement should I expect?

Modest. Trials typically show pain score reductions in the range of 20–30%, which is meaningful for some people and unnoticeable for others.

Can I take several at once?

Yes, and combinations are common. The trade-off is that if something works you will not know which component did it, and cost rises quickly.

When should I see a doctor instead?

Joint swelling, redness, heat, locking, giving way, or pain that wakes you at night all warrant assessment rather than a supplement.

Related on FactoWiki

This article is general information, not medical advice. FactoWiki may earn a commission from links on product review pages (never on comparisons). Always check with a qualified healthcare professional about your own situation.