Collagen Peptides: What the Randomised Trials Show for Skin, Joints and Bone

First, this is a different category from the rest of this site

Most compounds in our encyclopedia are injected, unapproved, and supported by a handful of human studies or none. Collagen peptides are none of those things. They are eaten, sold openly in supermarkets as a dietary supplement, and tested in dozens of randomised placebo-controlled trials. Searching "peptides" and landing on collagen is the single commonest confusion we see, so it is worth stating plainly: a tub of collagen powder and a vial of BPC-157 have almost nothing in common beyond the word peptide.

"Hydrolysed" is the other word worth decoding. Whole collagen is a large, tough protein your gut cannot absorb intact. Hydrolysis breaks it with enzymes into short fragments, typically a few thousand daltons, which can cross the intestinal wall. Some of those fragments, notably the tripeptide glycine-proline-hydroxyproline, do appear in blood after a dose. "Collagen peptides", "hydrolysed collagen" and "collagen hydrolysate" describe the same thing; "marine", "bovine" and "porcine" describe only the animal it came from.

The mechanism usually offered in marketing — you eat collagen, so you make collagen — is not how protein digestion works, and the trials do not test it. What the fragments may do is act as signals rather than bricks. That distinction matters when you read a claim, because an effect on skin hydration does not require, or demonstrate, that any eaten collagen became skin.

Skin: three meta-analyses, two outcomes that hold up

This is where the evidence is strongest and where almost all the marketing points. Three systematic reviews with meta-analyses have pooled the randomised trials, and they broadly agree.

The largest by trial count pooled 26 randomised controlled trials in 1,721 people and found significant improvements in skin hydration and in elasticity against placebo. Its authors also identified several biases in the trials they included and called for larger studies.¹ An earlier analysis of 19 double-blind randomised trials in 1,125 participants, 95% of them women aged 20 to 70, found favourable results for hydration, elasticity and wrinkles, and concluded that 90 days of intake reduces signs of skin ageing.² A third, of 14 trials in 967 participants, again found moisture and elasticity improved over a 12-week regimen.³

Two cautions belong beside those results, and they come from the literature itself rather than from us. First, a letter published in the same journal as the 19-trial analysis criticised it directly for heterogeneity: for pooling studies that differed too much in product, dose, duration and measurement to be combined into one estimate.⁴ That is a specific, published, on-the-record objection, and anyone citing the headline figure should know it exists.

Second, the outcomes are instrument readings. Hydration is measured by corneometry, elasticity by cutometry — devices pressed against the forearm or face. An earlier dermatology review of 11 randomised trials in 805 patients, covering collagen hydrolysate at 2.5 to 10 g a day for 8 to 24 weeks, noted the same pattern and the absence of regulation on quality, absorption and efficacy.⁵ No trial we found demonstrates a change visible to the person in the mirror, as opposed to a change a cutometer can detect. That is not the same as no effect; it is a smaller claim than the advertising makes.

Joints: a large short-term effect, in a field full of them

The most useful single paper here is not about collagen at all. A systematic review and meta-analysis of 20 dietary supplements across 69 randomised trials in hand, hip and knee osteoarthritis put collagen hydrolysate among seven supplements showing large effect sizes, above 0.80, and clinically important pain reduction in the short term.⁶ Good news, with three qualifications in the same paper: those effects were short-term only, no supplement showed clinically important pain reduction at long term, and the quality of evidence across the field ranged from very low to high.

A trial often cited for people without arthritis tested 5 g a day of bioactive collagen peptides for 12 weeks in 139 athletic adults with activity-related knee pain, and reported a statistically significant improvement in pain intensity against placebo.⁷ That is a real randomised result in a defined population: young active people with knee discomfort, not patients with diagnosed osteoarthritis, and not a structural change in the joint.

What none of this establishes is that collagen rebuilds cartilage. Pain and function are what the trials measured. Structural outcomes, where they have been measured across this class of supplement, have been the weakest part of the evidence.

Bone and muscle: smaller literatures, more specific claims

For bone, the key trial randomised 131 postmenopausal women with age-related reduction in bone mineral density to 5 g a day of specific collagen peptides or placebo for 12 months. Bone mineral density increased in the collagen group at the spine and femoral neck, and the bone formation marker P1NP rose in that group while the resorption marker CTX-1 rose in the control group.⁸ An open-label follow-up of 31 of those women after four years reported continued increases.⁹ Note what the follow-up is: open-label, uncontrolled, and a quarter of the original sample. It is supportive, not confirmatory.

For muscle, the design matters more than the supplement. Fifty-three older men with sarcopenia did a 12-week guided resistance training programme and took either 15 g a day of collagen peptides or a silica placebo; the collagen group gained more fat-free mass and strength.¹⁰ Everyone trained. The question the trial answers is whether collagen adds to resistance training, not whether it does anything without it.

A 2025 study went after the mechanism directly, using infused labelled phenylalanine to measure whether 30 g of hydrolysed collagen builds muscle connective protein better than the same amino acids given free, after resistance exercise in 45 young adults.¹¹ That is the right kind of experiment for the "collagen is special" claim, and the reason to read it rather than a summary of it.

Who paid for these trials

This is the part most coverage leaves out, and it is not a reason to dismiss the evidence. It is a reason to read it carefully.

A large share of the collagen trial literature is funded by, supplied by, or conducted with collagen manufacturers, and the recurring phrase "specific collagen peptides" or "bioactive collagen peptides" in trial titles is a clue: these are branded, proprietary preparations, tested by the companies that make them. Industry funding is routine in nutrition research and does not make a result false. It does mean three things. Trials that found nothing are less likely to have been published. The product tested is a specific preparation, not collagen in general. And the dose, duration and outcome measure were chosen by a party with an interest in the answer.

The practical consequence for a reader is that the evidence supports a narrower claim than it appears to. "Collagen peptides improved skin elasticity" is supported. "This tub of collagen will improve your skin elasticity" requires that the tub contains the same preparation at the same dose, which is rarely checkable from a label.

The doses actually studied

Trials use specific amounts, and they are not the amounts on every label.

• Skin: 2.5 to 10 g a day, for 8 to 24 weeks, across the trials in the dermatology review.⁵ One frequently cited trial compared 2.5 g with 5 g in 69 women aged 35 to 55 over 8 weeks.¹²
• Knee discomfort: 5 g a day for 12 weeks in athletic adults.⁷
• Bone mineral density: 5 g a day for 12 months in postmenopausal women.⁸
• Muscle with resistance training: 15 g a day for 12 weeks in older men with sarcopenia.¹⁰

Two things follow. The effective dose for one outcome is not the effective dose for another, and the muscle dose is three times the skin dose. And every one of these results came with a fixed duration: the shortest skin trials run eight weeks, the bone trial a year. Nothing here supports judging a product after a fortnight.

What a seller is allowed to say

In the United States collagen peptides are a dietary supplement, which sets the rules for the marketing around them. A seller may make a structure/function claim — a statement about how an ingredient affects the structure or function of the body. FDA does not pre-approve those claims. The manufacturer must hold substantiation that the claim is truthful and not misleading, must notify FDA of the claim's text within 30 days of marketing, and must print a disclaimer stating that FDA has not evaluated the claim and that the product is not intended to diagnose, treat, cure or prevent any disease.¹³

That is why labels say "supports skin elasticity" rather than "treats wrinkles". The first is a structure/function claim; the second is a drug claim, and only an approved drug may make it. Reading that distinction on a label tells you which legal category you are in, and therefore what evidence stands behind the sentence.

We checked the European position directly and could not read it: EFSA's site refused automated access on October 1, 2026. Rather than repeat a widely circulated claim about what EFSA has authorised, we state only what we verified ourselves, which is the US position above.

What a fair summary looks like

Collagen peptides have better human evidence than almost any compound on this site. Three meta-analyses agree on hydration and elasticity, and the bone and knee trials are real randomised results. They are also small, measured by instruments, concentrated in trials run by the industry that sells the product, and criticised in print for being pooled too readily.

So the fair summary is narrow and dull: at 2.5 to 10 g a day for at least two to three months, hydrolysed collagen produces measurable improvements in skin hydration and elasticity against placebo, with weaker and less consistent evidence for wrinkles, short-term benefit for joint pain in a field where many supplements show short-term benefit, and one well-conducted year-long bone trial that deserves replication by someone with no commercial interest in the answer.

If what brought you here was a search for peptides, our collagen peptides entry keeps the evidence and status in one place, and the rest of the encyclopedia covers the injectable compounds that share the word and nothing else.

Frequently asked questions

Do collagen peptides actually work?

For two specific outcomes, the randomised evidence says yes: skin hydration and skin elasticity improved against placebo across three meta-analyses covering 19, 26 and 14 trials. Those are instrument measurements rather than visible change, the effects are modest, and much of the literature is industry-funded.

How much collagen should I take?

The trials used 2.5 to 10 g a day for skin over 8 to 24 weeks, 5 g for knee discomfort and for bone density, and 15 g alongside resistance training for muscle. The dose that worked for one outcome is not the dose that worked for another, and nothing was tested for less than eight weeks.

Are collagen peptides the same as the peptides sold for healing or weight loss?

No. Collagen peptides are eaten, sold as a dietary supplement and tested in dozens of randomised trials. Compounds like BPC-157 or retatrutide are injected, unapproved as research chemicals or still in trials, and supported by far less human evidence. They share a word, not a category.

Does it matter whether collagen is marine, bovine or porcine?

The source describes the animal, not the effect. One meta-analysis found that effects on hydration varied by source and duration, while effects on elasticity did not differ significantly between sources. No source has been shown superior for a specific outcome.

Does eating collagen become collagen in my skin?

That is not how protein digestion works, and the trials do not test it. Hydrolysed collagen is broken into short fragments, some of which appear in blood. Any effect is more plausibly a signal than a building material, and an improvement in a skin measurement does not show that eaten collagen became skin.

Can a company say collagen treats wrinkles?

Not in the United States. As a dietary supplement it may carry a structure/function claim such as 'supports skin elasticity', must hold substantiation, must notify FDA within 30 days, and must print that FDA has not evaluated the claim and that the product is not intended to diagnose, treat, cure or prevent any disease.

Related on Grey Peptides

• Peptides and women's health: approved, studied, thin evidence
• Collagen peptides
• GHK-Cu
• Copper peptides for skin
• Peptide encyclopedia

Sources

1. Pu, S.-Y. et al. (2023). Effects of Oral Collagen for Skin Anti-Aging: A Systematic Review and Meta-Analysis. Nutrients, 15(9), 2080. PMID: 37432180
2. de Miranda, R. B. et al. (2021). Effects of hydrolyzed collagen supplementation on skin aging: a systematic review and meta-analysis. International Journal of Dermatology, 60(12), 1449-1461. PMID: 33742704
3. Dewi, D. A. R. et al. (2023). Exploring the Impact of Hydrolyzed Collagen Oral Supplementation on Skin Rejuvenation: A Systematic Review and Meta-Analysis. Cureus, 15(12), e50231. PMID: 38192916
4. Peres, G. et al. (2022). Concerning the heterogeneity of the studies included in meta-analyses. Comment on: 'Effects of hydrolyzed collagen supplementation on skin aging'. International Journal of Dermatology, 61(3), e99-e101. PMID: 34196407
5. Choi, F. D. et al. (2019). Oral Collagen Supplementation: A Systematic Review of Dermatological Applications. Journal of Drugs in Dermatology, 18(1), 9-16. PMID: 30681787
6. Liu, X. et al. (2018). Dietary supplements for treating osteoarthritis: a systematic review and meta-analysis. British Journal of Sports Medicine, 52(3), 167-175. PMID: 29018060
7. Zdzieblik, D. et al. (2017). Improvement of activity-related knee joint discomfort following supplementation of specific collagen peptides. Applied Physiology, Nutrition, and Metabolism, 42(6), 588-595. PMID: 28177710
8. König, D. et al. (2018). Specific Collagen Peptides Improve Bone Mineral Density and Bone Markers in Postmenopausal Women - A Randomized Controlled Study. Nutrients, 10(1), 97. PMID: 29337906
9. Zdzieblik, D. et al. (2021). Specific Bioactive Collagen Peptides in Osteopenia and Osteoporosis: Long-Term Observation in Postmenopausal Women. Journal of Bone Metabolism, 28(3), 207-213. PMID: 34520654
10. Zdzieblik, D. et al. (2015). Collagen peptide supplementation in combination with resistance training improves body composition and increases muscle strength in elderly sarcopenic men. British Journal of Nutrition, 114(8), 1237-1245. PMID: 26353786
11. Aussieker, T. et al. (2025). The Effects of Ingesting a Single Bolus of Hydrolyzed Collagen versus Free Amino Acids on Muscle Connective Protein Synthesis Rates. Medicine and Science in Sports and Exercise. PMID: 40523226
12. Proksch, E. et al. (2014). Oral supplementation of specific collagen peptides has beneficial effects on human skin physiology: a double-blind, placebo-controlled study. Skin Pharmacology and Physiology, 27(1), 47-55. PMID: 23949208
13. US Food and Drug Administration. Structure/Function Claims. Read October 1, 2026. FDA

This article is for education. It is not medical advice and it is not a recommendation to take or avoid any supplement. Collagen peptides are a food ingredient sold as a dietary supplement; if you are being treated for osteoporosis, arthritis or any other condition, the treatment your clinician prescribed is the one with evidence behind it for that condition.