Where a collagen pack in the EU carries a health claim about collagen, the claim belongs to vitamin C. The EU list of permitted claims names collagen only inside claims for vitamin C, such as normal collagen formation for bones, skin and cartilage. When a collagen maker applied for a joint claim, the EU food safety authority found in 2011 that cause and effect was not established. So, does collagen powder work? Osteoarthritis pain eased by a small to moderate effect, fat-free mass rose only alongside strength training (against placebos mostly without protein), and skin showed no measurable effect without industry funding. The table below sorts five targets.

Key takeaways

1. Does collagen powder work for skin? Not measurably in trials without industry funding or in high-quality trials.

2. Joints: osteoarthritis pain eased a little; an older review found no clinically important long-term effect.

3. Muscle: an incomplete protein that stimulated muscle building less than whey; extra fat-free mass came only with training, against placebos mostly without protein.

4. Bone: one trial, a small density change, no fracture data.

5. The pack: the permitted health claim is about vitamin C.

Does collagen work? What reaches the blood, and the evidence by target

Collagen is digested. Small fragments containing hydroxyproline reach the blood; the absorption studies do not show that they rebuild skin or cartilage. Per target, the table shows some easing of osteoarthritis pain, more fat-free mass only with training, and no measurable skin effect in trials without industry funding.

Collagen in trials, by target
In trials, collagen had small to moderate effects on osteoarthritis pain and, with training, on fat-free mass against placebos mostly without protein; skin trials without industry funding showed no measurable effect
TargetProduct, amountWho was studiedWhat changedCertainty
Skinhydrolysate, 2.5 to 10 g a day23 trials, 1,474 peopleAll trials: better hydration, elasticity, wrinkles; trials without industry funding and high-quality trials: no measurable effectNote: effect seen only in industry-funded trialslow
Joints (osteoarthritis)hydrolysate (grams) or undenatured type II (40 mg)25 trials, 2,856 people with osteoarthritisPain eased by a small to moderate effect size (0.35); no clinically important long-term effect shownNote: hydrolysate and type II pooledmoderate
Tendons15 or 30 g with vitamin C before training, in 1 trial of 8 men8 trials, 257 people aged 18 to 52, almost all menThicker tendons in 3 of 4 trials that measured it; no extra strengthNote: small trials, almost no women, nobody over 52low
Muscle15 g a day with training in 2 trials; 20 g twice a day without training8 training trials, 418 healthy adults (fat-free mass); a trial of 208 adults over 65 (collagen without training)With training: more fat-free mass (moderate, 0.48); without training: no measurable difference from a carbohydrate drinkNote: placebos mostly without protein; maker funding disclosed in 2025 for one trialmoderate
Bone5 g a day for 12 months1 trial, 131 women after menopause with low bone densitySpine T-score +0.1 with collagen, −0.03 with placebo; fractures not studiedNote: product from a collagen makerlow to moderate
Each row in the unit of its own trials; the rows share no common scale. Effect sizes: about 0.2 small, 0.5 moderate, 0.8 large.

Collagens, a family of 28 types, are the most abundant proteins in mammals; the “30 percent of body protein” seen online is not in that review. After 9.4 to 23 grams of hydrolysate, fragments such as prolyl-hydroxyproline (two linked amino acids) peaked in the blood at 1 to 2 hours and were down by half at 4 hours. In 4 volunteers, blood levels rose with the dose.

Terms: an effect size (standardized mean difference) makes trials with different scales comparable; about 0.2 is small, 0.5 moderate, 0.8 large. Fat-free mass is body weight minus fat, mostly muscle and water.

Collagen for skin: no measurable effect in trials without industry funding

For skin, who paid changes the answer. A 2025 meta-analysis of 23 trials (1,474 participants) found better hydration, elasticity and wrinkle scores across all trials. Yet the trials without industry funding showed no measurable effect, and neither did the high-quality trials.

Skin trials, split by who paid
Across 23 skin trials collagen improved hydration, elasticity and wrinkles; in trials without industry funding the effect was not measurable
All 23 trials (1,474 people)
  1. Hydration: improved
  2. Elasticity: improved
  3. Wrinkles: improved
Trials without industry funding and high-quality trials
  1. Hydration: no measurable effect in either group
  2. Elasticity: no measurable effect in either group
  3. Wrinkles: no measurable effect in either group
Earlier meta-analysis of 26 trials, all funding sources, against placebo: effect size 0.63 for hydration (18 trials) and 0.72 for elasticity (19 trials), both moderate.

Without that split, an earlier meta-analysis of 26 placebo-controlled trials (1,721 participants, 2 to 12 weeks) pooled hydration from 18 of them and elasticity from 19: moderate effect sizes of 0.63 and 0.72. A 2026 review of 25 trials (13 to 236 participants each) rated most of them at high risk of bias.

The popular “1 percent a year from age 25” does not trace to its usual source. The 1975 measurement study found less skin collagen at older ages, and less in women than men at every age, but its summary gives no yearly rate. In the lab, skin cells from people over 80 made about a third less type I precursor than cells from people aged 18 to 29 (56 against 82 nanograms per milliliter); the study gives no figure for the 40s or 50s.

Collagen for joints: some relief in osteoarthritis, from two different products

In osteoarthritis, collagen products eased pain by a small to moderate effect size without a measurable rise in side effects; two products sit behind this, hydrolysate in grams and undenatured type II collagen in milligrams. In an older review of 20 osteoarthritis supplements, none held a clinically important effect at long-term follow-up, on very low-quality evidence. For healthy, active joints, the EU food safety authority found no established benefit.

In a 2024 meta-analysis, pain data from 25 trials with 2,856 patients showed an effect size of 0.35 against control (small to moderate, moderate certainty); function improved by 0.31 (small, high certainty). A 2025 knee-only update (11 trials, 870 participants) also pointed to symptom relief, with widely varying results.

Which collagen helps joints depends on the product. Undenatured type II is intact cartilage collagen; in a trial with 191 people with knee osteoarthritis, 40 milligrams a day for 180 days beat placebo on the overall symptom score and also beat glucosamine with chondroitin, side effects not clearly different. The EU authority's healthy-joint evidence was one trial in 147 student athletes, where none of 15 joint measures differed significantly from placebo after correction for multiple comparisons.

Tendons and ligaments: collagen taken before training

The tendon trials paired collagen with training. In 8 trained men with a mean age of about 49, a heavy leg session alone did not measurably raise a collagen-building blood marker; 15 grams of collagen with vitamin C beforehand raised it, and 30 grams more so. In 3 of 4 trials that measured tendon size, tendons grew thicker with collagen than with placebo, but collagen added no measurable strength.

In 8 healthy men, 15 grams of vitamin C-enriched gelatin doubled the marker when taken an hour before 6 minutes of rope skipping. The tendon review's 257 people were 246 men and 11 women, none over 52, and a thicker tendon is not the same as fewer injuries.

Collagen vs whey: why collagen builds little muscle on its own

Collagen lacks tryptophan, an essential amino acid, so it is an incomplete protein. In older women and men, whey raised the muscle-building rate and collagen did not raise it measurably. In a one-year trial with 208 adults over 65, collagen without training (20 grams twice a day) changed no muscle measure measurably against a carbohydrate drink. With strength training, trials found more fat-free mass, mostly against placebos with no protein.

Whey against collagen, women aged about 69
In 22 older women, whey raised the muscle-building rate over 6 days; collagen did not raise it measurably
Bars: change in muscle protein synthesis, percent per day, from zero.
Whey, rested leg0.063 %Whey, trained leg0.173 %Collagen, rested legnot significant0.011 %Collagen, trained legnot significant0.020 %
Whey, rested leg0.063 %Whey, trained leg0.173 %Collagen, rested legnot significant0.011 %Collagen, trained legnot significant0.020 %
30 g twice a day for 6 days, 11 women per group. In 31 men aged about 72 who already ate the recommended protein amount, an extra 50 g a day of whey or pea protein raised the rate; the same amount of collagen did not raise it measurably.
In 22 older women, whey raised the muscle-building rate over 6 days; collagen did not raise it measurably
GroupChange, percent per day
Whey, rested leg0.063 %
Whey, trained leg0.173 %
Collagen, rested leg (not significant)0.011 %
Collagen, trained leg (not significant)0.020 %

In 31 men aged about 72 already at the recommended protein intake, 50 extra grams of whey or pea a day raised the rate; collagen did not measurably.

In a 12-week trial completed by 53 men with sarcopenia (mean age 72), everyone lifted three times a week; the group on 15 grams of collagen peptides gained 4.2 against 2.9 kilograms of fat-free mass and 16.5 against 7.3 newton-meters of thigh strength, about 1.3 kilograms and 9 newton-meters more. The placebo held no protein, and the men's diets supplied only about 0.91 grams of protein per kilogram. A 2025 correction disclosed maker payments and a patent application. In 77 premenopausal women aged 18 to 50 who finished, training plus 15 grams beat training plus placebo on fat-free mass and grip strength. Pooled over 8 training trials (418 healthy adults, placebos mostly without protein), the fat-free mass gain was moderate (0.48, moderate certainty); over 11 trials, maximal strength rose by a small amount (0.19, low certainty).

As protein, collagen counts only in part: as an example, a 70-kilogram (154-pound) person over 65 aiming for the advised 1.0 to 1.2 grams per kilogram needs 70 to 84 grams a day; a 15-gram scoop covers 18 to 21 percent of that, and the modeled 36 percent ceiling is about 25 to 30 grams. The protein calculator for your daily target does the sum; sarcopenia tests and treatment covers muscle loss.

The fat-free mass gains in these trials all came with training; strength training after 40 shows how to build it.

Collagen and bone density: one trial after menopause

For bone density, our search found one trial: 131 women after menopause with low bone density (mean age 64) took 5 grams of collagen peptides or a placebo daily for 12 months. Spine and hip density rose slightly more than with placebo; fractures were not studied.

A T-score counts standard deviations from young-adult bone density, so the gap between the group means works out to about a tenth of a unit (0.13 at the spine, 0.10 at the hip); 29 of the 131 women (22 percent) dropped out. The product came from a collagen maker, and our search found no independent repeat. Creatine for women over 50 covers another supplement studied alongside training.

How much, how long and side effects

The skin, muscle-training and bone trials gave 2.5 to 15 grams of hydrolysate a day; the knee trial gave 40 milligrams of undenatured type II collagen. Skin was measured after 2 to 24 weeks, the knee after 180 days, bone after 12 months. Side effects were few. Morning against evening intake was not compared in the trials we found.

How long the trials ran
Skin trials ran 2 to 24 weeks, the knee trial 180 days and the bone trial 12 months before results were measured
Bars: weeks from the start of the trial, from zero; lighter part: the spread across trials.
Skin, 26 trials (pooled)2 to 12 weeksSkin, hydrolysate trials8 to 24 weeksSkin, effects read after about 90 days (one review)about 90 daysMuscle with training12 weeksKnee osteoarthritis, 40 mg type II180 daysBone density12 months
Skin, 26 trials (pooled)2 to 12 weeksSkin, hydrolysate trials8 to 24 weeksSkin, effects read after about 90 days (onereview)about 90 daysMuscle with training12 weeksKnee osteoarthritis, 40 mg type II180 daysBone density12 months
Skin trials ran 2 to 24 weeks, the knee trial 180 days and the bone trial 12 months before results were measured
TrialsDuration
Skin, 26 trials (pooled)2 to 12 weeks
Skin, hydrolysate trials8 to 24 weeks
Skin, effects read after about 90 days (one review)about 90 days
Muscle with training12 weeks
Knee osteoarthritis, 40 mg type II180 days
Bone density12 months

A review of 11 dermatology trials (805 patients) reports 2.5 to 10 grams of hydrolysate over 8 to 24 weeks, or 3 grams of collagen tripeptide over 4 to 12 weeks, with no adverse events reported. Another meta-analysis (19 trials, 1,125 participants aged 20 to 70, 95 percent women) read effects after about 90 days, an observation, not a tested threshold. The signs to judge it by are what trials measured: skin hydration, joint pain, fat-free mass.

15 grams a day, the upper end of those trials, came with few side effects and sits below the modeled 36 percent share of daily protein. The case against collagen is about benefit: no measurable skin effect without industry money, no measurable muscle gain on its own, and joint relief in osteoarthritis that was not shown to last.

More health topics to explore

References

  1. European Commission — Commission Regulation (EU) No 432/2012 establishing a list of permitted... (2012)
  2. EFSA Panel on Dietetic Products, Nutrition and Allergies (NDA) — Scientific Opinion on the substantiation of a health claim related to... (2011)
  3. Liang CW et al. — Efficacy and safety of collagen derivatives for osteoarthritis: A trial... (2024)
  4. Bischof K et al. — Impact of Collagen Peptide Supplementation in Combination with Long-Term... (2024)
  5. Myung SK et al. — Effects of Collagen Supplements on Skin Aging: A Systematic Review and... (2025)
  6. Ricard-Blum S — The collagen family (2011)
  7. Iwai K et al. — Identification of food-derived collagen peptides in human blood after oral... (2005)
  8. Shigemura Y et al. — Dose-dependent changes in the levels of free and peptide forms of... (2014)
  9. Pu SY et al. — Effects of Oral Collagen for Skin Anti-Aging: A Systematic Review and... (2023)
  10. Bassila C et al. — Efficacy and safety of hydrolyzed collagen supplementation on skin health... (2026)
  11. Shuster S et al. — The influence of age and sex on skin thickness, skin collagen and density (1975)
  12. Varani J et al. — Decreased collagen production in chronologically aged skin: roles of... (2006)
  13. Liu X et al. — Dietary supplements for treating osteoarthritis: a systematic review and... (2018)
  14. Simental-Mendía M et al. — Effect of collagen supplementation on knee osteoarthritis: an updated... (2025)
  15. Lugo JP et al. — Efficacy and tolerability of an undenatured type II collagen supplement in... (2016)
  16. Nulty CD et al. — Hydrolyzed collagen supplementation prior to resistance exercise augments... (2024)
  17. Buchalski A et al. — Collagen Supplementation on Tendon-Related Structural and Performance... (2026)
  18. Shaw G et al. — Vitamin C-enriched gelatin supplementation before intermittent activity... (2017)
  19. Paul C et al. — Significant Amounts of Functional Collagen Peptides Can Be Incorporated in... (2019)
  20. Oikawa SY et al. — Whey protein but not collagen peptides stimulate acute and longer-term... (2020)
  21. Mertz KH et al. — The effect of daily protein supplementation, with or without resistance... (2021)
  22. McKendry J et al. — The effects of whey, pea, and collagen protein supplementation beyond the... (2024)
  23. Zdzieblik D et al. — Collagen peptide supplementation in combination with resistance training... (2015)
  24. Zdzieblik D et al. — Collagen peptide supplementation in combination with resistance training... (2025)
  25. Jendricke P et al. — Specific Collagen Peptides in Combination with Resistance Training Improve... (2019)
  26. Bauer J et al. — Evidence-based recommendations for optimal dietary protein intake in older... (2013)
  27. König D et al. — Specific Collagen Peptides Improve Bone Mineral Density and Bone Markers in... (2018)
  28. Choi FD et al. — Oral Collagen Supplementation: A Systematic Review of Dermatological... (2019)
  29. de Miranda RB et al. — Effects of hydrolyzed collagen supplementation on skin aging: a systematic... (2021)

About this article · Written by the huuman Team. Our content is based on peer-reviewed research and clinical guidelines. We follow editorial standards grounded in scientific evidence.

This article is for educational purposes only and does not constitute medical advice. Health and training decisions should be discussed with qualified professionals.

November 12, 2026
November 17, 2026