Collagen for skin and joints: does it work?
Collagen is everywhere right now. So here's the honest version — including the parts the ads leave out.
What the evidence shows
Skin. Pooled trials are genuinely mixed. Some meta-analyses report improvements in skin hydration and elasticity after 8–12 weeks. But a large 2025 review found that the benefit largely disappeared in independent, high-quality studies — the positive results clustered in industry-funded and lower-quality trials. Its blunt conclusion was that there's no solid clinical evidence that collagen prevents or treats skin ageing.
Joints. For knee osteoarthritis, meta-analyses find a modest reduction in pain versus placebo — real, but based on trials rated at high risk of bias, and usually shorter than six months.
Safety is consistently good across trials.
The honest caveats
- Funding bias heavily shapes the skin literature.
- High variability in doses, sources and measures.
- Trials measure symptoms and instrument readings — not regrown collagen or cartilage.
What the doses looked like
Where effects were seen: roughly 2.5–10 g/day of hydrolysed collagen peptides for skin (over 8–12 weeks), and around 10 g/day for joints (over 3–6 months).
Important for a supplement brand
There are no EU-authorised health claims for collagen — for skin or joints. EFSA has specifically rejected them. So while the category is popular and competitors lean on "beauty from within" language, strictly speaking those aren't permitted health claims. We'll always keep product claims clean and put the real evidence here, in the journal.
In short
Collagen is safe and might help skin and joints for some people — but the strongest, independent evidence is underwhelming, and no health claim is approved. Promising, not proven.
Sources
- The American Journal of Medicine (2025) — systematic review/meta-analysis, collagen and skin.
- Osteoarthritis meta-analyses on collagen peptides and knee OA.
- EFSA — rejected collagen skin and joint claims.