Does Collagen Work? Skin, Joints and the Gym

A 2026 review pooled 113 randomised trials of collagen supplements with 7,983 participants. Collagen was linked to better skin elasticity and hydration and to less osteoarthritis pain and stiffness. In a handful of small trials, it showed no clear benefit for strength recovery or muscle soreness after exercise.

This guide sets out what the review and five other meta-analyses found for skin, joints, muscle and recovery, how strong that evidence is, and what the trials used.

What collagen supplements are

Collagen is a structural protein found in skin, tendons, cartilage and bone. The 2026 review notes it makes up about 25% to 30% of the protein in the human body, and that some research suggests the body's own production falls by about 1% a year from early adulthood. Many of the trials below used hydrolysed collagen, where the protein is broken into smaller pieces called peptides.

What the 2026 umbrella review found

An umbrella review pools earlier systematic reviews. This one, led by Roshan Ravindran with researchers including a team at Anglia Ruskin University in Cambridge, screened 573 papers and included 16 systematic reviews with meta-analyses, covering 113 randomised trials and 7,983 participants. The authors reported no funding and no conflicts of interest.

Two quality ratings matter here. GRADE rates how sure we can be about each pooled result, from high to very low certainty. AMSTAR 2 rates how well each review was carried out. Of the 16 reviews, 1 was rated high quality, 4 low and 11 critically low, so a result can grade high certainty while the reviews behind it were poorly run.

Results are given as standardised effect sizes, which let trials that used different measures be pooled. A larger number means a bigger difference from placebo. The chart shows selected outcomes where collagen came out ahead.

For skin, collagen improved elasticity (20 trials, 1,217 people) and hydration (19 trials, 954 people), both rated high certainty, but not skin roughness. These are device measurements of how skin stretches and holds water, not ratings of wrinkles or appearance, and the authors wrote that collagen may be best positioned not as an anti-wrinkle quick fix. The plausible range for elasticity was wide (0.44 to 1.59), and results varied with the year trials were published.

In people with osteoarthritis, collagen reduced self-reported pain (25 trials, 2,687 people) and stiffness on the WOMAC questionnaire, a standard osteoarthritis score, both rated high certainty. Two other WOMAC scores, pain and functional limitation, did not reach statistical significance. Trials that ran longer tended to report better scores on some measures, a finding the authors said needs confirming in larger head-to-head trials.

For muscle, collagen was linked to a moderate gain in fat-free mass (moderate certainty) and a modest gain in maximal strength (high certainty, 11 trials). Tendon shape improved at low certainty, but tendon mechanical properties did not change significantly.

Where collagen showed no clear benefit

The review found no significant effect on strength recovery 24 or 48 hours after exercise, or on muscle soreness immediately after exercise or at 24 and 48 hours. Each of these rested on 4 or 5 small trials and was rated low certainty, so this is a lack of evidence of benefit rather than proof of none. The authors concluded that collagen's effect looks chronic and structural rather than a quick recovery aid.

Results elsewhere were mixed. Fat mass fell (moderate certainty), but body fat percentage did not change significantly, and fasting blood glucose, blood lipids and blood pressure showed no significant effect. For oral health, gum thickness fell slightly (low certainty), while other measures showed no significant change.

The authors also listed limits: certainty was often moderate to low, sample sizes were often small, and most trials were short, so effects over years are not known.

Skin: what the other meta-analyses show

A 2023 meta-analysis in Nutrients, led by Szu-Yu Pu at Taipei Medical University, pooled 26 randomised trials with 1,721 people. Hydrolysed collagen improved skin hydration and elasticity against placebo. The effect on hydration varied with the collagen source and how long people took it, and the authors flagged several biases in the trials.

A 2023 meta-analysis in Cureus, led by Dian Andriani Ratna Dewi, pooled 14 trials with 967 participants and reported better skin moisture and elasticity than placebo after a 12-week course. A 2025 meta-analysis led by Gabriela Danessa pooled 10 trials with 646 participants. Doses ran from 1 to 10 g a day, the most common was 4 g and the median was 3.5 g. The authors noted that the findings apply mainly to women and that risk of bias was generally unclear.

These skin analyses draw on many of the same trials as each other and as the 2026 review, so they are not independent confirmation.

Knees: osteoarthritis pain

A 2025 meta-analysis in Clinical and Experimental Rheumatology, led by Mario Simental-Mendia, pooled 11 randomised trials with 870 people with knee osteoarthritis. Collagen improved pain and function scores against placebo, and the authors judged the gains clinically meaningful, though results varied widely between trials. In the 2026 review, by contrast, the WOMAC functional limitation score did not reach statistical significance.

A smaller 2023 meta-analysis led by Chun-Ru Lin pooled 4 trials with 507 people and also found less pain with collagen peptides than placebo. All 4 trials carried a high risk of bias. These trials studied people with diagnosed osteoarthritis, not other kinds of knee pain.

Keep this in context. The NHS lists exercise as one of the most important treatments for osteoarthritis, and lists keeping to a healthy weight among the main lifestyle measures. On osteoarthritis supplements in general, naming glucosamine and chondroitin, it notes that NICE, the National Institute for Health and Care Excellence, recommends they should not routinely be offered on the NHS.

How to use collagen if you choose to

The points below are our suggestions, built on the trials above and NHS guidance.

• •Give it time. The Dewi analysis pointed to a 12-week course for skin. Judge it after about 3 months, not 2 weeks.
• •Know what the dose evidence covers. Skin trials in the Danessa analysis used 1 to 10 g a day, most often 4 g. The sources here give no clear dose for joints, so follow the product label.
• •Keep exercising. For osteoarthritis, the NHS recommends a mix of muscle-strengthening and general fitness exercise, built up gradually with a plan from your GP or a physiotherapist. Collagen, if you use it, sits on top of that.
• •Eat a balanced diet with plenty of vitamin C. The NHS says vitamin C helps maintain healthy skin, bones and cartilage, and that adults aged 19 to 64 should be able to get the 40 mg they need each day from foods such as oranges, peppers, strawberries and broccoli. This is general nutrition advice, not a way to make a supplement work better.
• •Check the source animal. Our advice: products come from fish or cattle, among others, so check the label if you have a fish allergy or dietary or religious needs.
• •Ask a pharmacist or GP before starting if you are pregnant, take regular medicines or have a medical condition. This is our advice, as the trials were short and long-term safety is not known.

Common mistakes

• •Buying collagen mainly for recovery. Small, low-certainty trials so far show no clear benefit for soreness or strength recovery.
• •Stopping after a couple of weeks. The trials ran for weeks to months.
• •Using it in place of exercise for osteoarthritis. The NHS lists exercise as one of the most important treatments.

Frequently asked questions

The bottom line

Across pooled trials, collagen supplements were linked to better measured skin elasticity and hydration and to less osteoarthritis pain and stiffness. In a few small trials, they showed no clear benefit for recovery or soreness after exercise, and most of the reviews behind these results were rated low in quality. If you try collagen, give it about 3 months, keep exercising and eating well, and check with a pharmacist or GP if you are pregnant or take regular medicines.

Related reading

• •The best exercise for knee osteoarthritis
• •Turmeric and curcumin for joint pain
• •How much protein per meal
• •Do multivitamins work

Sources

• •Ravindran R, Pizzol D, et al. Collagen Supplementation for Skin and Musculoskeletal Health: An Umbrella Review of Meta-Analyses on Elasticity, Hydration, and Structural Outcomes. Aesthetic Surgery Journal Open Forum 2026. pubmed.ncbi.nlm.nih.gov/41809116
• •Pu SY, Huang YL, et al. Effects of Oral Collagen for Skin Anti-Aging: A Systematic Review and Meta-Analysis. Nutrients 2023. pubmed.ncbi.nlm.nih.gov/37432180
• •Dewi DAR, Arimuko A, et al. Exploring the Impact of Hydrolyzed Collagen Oral Supplementation on Skin Rejuvenation: A Systematic Review and Meta-Analysis. Cureus 2023. pubmed.ncbi.nlm.nih.gov/38192916
• •Danessa G, Notario D, Regina R. Effects of collagen-based supplements on skin's hydration and elasticity: A systematic review and meta-analysis. Indian Journal of Dermatology, Venereology and Leprology 2025. pubmed.ncbi.nlm.nih.gov/40826844
• •Simental-Mendia M, Ortega-Mata D, et al. Effect of collagen supplementation on knee osteoarthritis: an updated systematic review and meta-analysis of randomised controlled trials. Clinical and Experimental Rheumatology 2025. pubmed.ncbi.nlm.nih.gov/39212129
• •Lin CR, Tsai SHL, et al. Analgesic efficacy of collagen peptide in knee osteoarthritis: a meta-analysis of randomized controlled trials. Journal of Orthopaedic Surgery and Research 2023. pubmed.ncbi.nlm.nih.gov/37717022
• •NHS. Osteoarthritis: treatment and support. nhs.uk/conditions/osteoarthritis/treatment
• •NHS. Vitamin C. nhs.uk/conditions/vitamins-and-minerals/vitamin-c