Printed from https://bottledhope.co/supplements/collagen/. Last reviewed 2026-09-26. Check the site for the latest version.
Collagen
The body's most abundant structural protein, made into supplements from cattle, pig, fish or poultry. It is studied for skin, joints, muscle and bone.
Explained simply · Last reviewed
What it is
Collagen is the most abundant structural protein in the body. A 2021 review of collagen and exercise says it makes up about a third of the protein in humans, and about 65 to 80% of the dry weight of tendons. A 2023 review of skin trials says collagen makes up 80% of the dry weight of human skin, and that skin makes less of it with age. It names collagen supplements among the ways people try to keep their skin looking young.
More about Collagen
A 2021 review of collagen and exercise says it makes up about a third of the protein in humans, and about 65 to 80% of the dry weight of tendons. A 2023 review of skin trials says collagen makes up 80% of the dry weight of human skin, and that skin makes less of it with age. It names collagen supplements among the ways people try to keep their skin looking young.
Supplements are made from cattle, pig, marine and poultry sources, the 2021 review says, and a 2023 review of knee trials names cattle hides, fish scales and chicken skin. The 2021 review explains that collagen is broken down with enzymes into short fragments, called peptides, the main form in supplements. A 2024 review of collagen with exercise training says gelatine is also made from collagen, and that it can be broken down further into the smaller collagen peptides. Unlike gelatine, the 2021 review says, the peptides do not set into a gel, and they dissolve in cold water.
The 2024 review says collagen peptides are generally considered low in essential amino acids (the ones the body cannot make for itself), and are often classed as a low-quality protein. The 2021 review says research on collagen peptides has mainly focused on bone and joint health. The claims below also cover skin and muscle.
Also known as: hydrolysed collagen, collagen peptides, collagen hydrolysate, marine collagen
At a glance
Safety: known risks and 1 group who should avoid it or take extra care. See Safety ↓Claims made about Collagen, and how strong the evidence is for each:
- Improves skin hydration and elasticityEvidence rating: Limited
- Eases knee osteoarthritis painEvidence rating: Moderate
- Eases activity-related joint pain in people without arthritisEvidence rating: Limited
- Builds muscle and strength alongside exercise trainingEvidence rating: Limited
- Increases bone mineral densityEvidence rating: Limited
Each rating follows our published rules and the sources cited for it. We don’t add our own medical judgment. How we rate
Safety
Read this before the ratings below.
Risks
- Allergy. The European Food Safety Authority (EFSA) states that mammalian collagen can cause severe allergic reactions.
- EFSA describes fish collagen as a sensitizer whose ability to trigger allergic reactions is uncertain. In two challenge studies in 33 fish-allergic people, the researchers observed no allergic reactions of clinical relevance to fish gelatine, although very few of those tested had antibodies specific to collagen. EFSA reports that fish and mammalian collagens do not cross-react, and that some fish muscle tissue is likely to remain on the fish skin used to make collagen.
- In the knee osteoarthritis trials pooled by a 2023 meta-analysis, adverse events were not significantly more common with collagen than in the comparison groups, but the reviewers rated how sure they are of that finding (certainty) as very low. They list gastrointestinal complaints as the most common events reported, and note that not every trial specified its adverse events.
- A 2025 systematic review of 36 trials found few adverse events reported, but states it is not clear whether participants had none or whether the trials did not evaluate or report them.
Who should avoid it or take extra care
- People with a fish allergy: EFSA describes fish collagen as a sensitizer whose ability to trigger allergic reactions is uncertain, and states that some muscle tissue is likely to adhere to the fish skin used for collagen preparation.
Safety sources: Source [1]Source [2]Source [3]
What the evidence says
Each claim is rated on its own evidence. Ratings follow the scale on our methodology page.
Improves skin hydration and elasticity Evidence rating: Limited
People take it hoping for firmer, better-hydrated skin as they age.
A 2023 review of 26 trials found better skin hydration and elasticity with collagen; EFSA said a benefit to skin function was not established for one product.
A 2023 meta-analysis pooled 26 randomized trials in 1,721 people. It found that hydrolysed collagen improved skin hydration and elasticity compared with placebo. The review was funded by a hospital, and its authors declare no conflicts.
Its authors identified several sources of bias in the included trials. They note that some trials had fewer than 40 participants, and state that large randomized trials are needed to confirm the findings. The review does not rate its certainty, and it did not analyze results by who funded the trials.
EFSA assessed an application from the company Gelita AG for one porcine collagen product. In 2013 it concluded that a cause-and-effect relationship had not been established between that product and a change in skin elasticity leading to an improvement in skin function. Of the two trials submitted, one measured wrinkles, which EFSA did not treat as a measure of skin function. The other found no difference in skin water loss or hydration.
Neither source cited here reports a result for wrinkles. EFSA's conclusion covers one product and skin function only. Because the two sources point in different directions, this page takes the lower reading.
Sources: Source [1]Source [2]
Eases knee osteoarthritis pain Evidence rating: Moderate
People with knee osteoarthritis take it hoping it will ease the pain in their knees.
A 2023 review pooling three trials found less knee pain with collagen and was moderately sure of it, though it judged all the trials it included at high risk of bias.
A 2023 meta-analysis of hydrolysed collagen in knee osteoarthritis included four trials with 507 participants, and pooled three of them, with 375 participants, for pain. It found less pain with collagen than in the comparison groups (standardized mean difference −0.58, a measure that puts results from different scales on one scale). The reviewers rated that pain result moderate certainty. They rated their finding of no significant difference in adverse events very low. The review had no funding and declares no conflicts.
Their GRADE table, a standard system for rating certainty, marks the pain result down for serious risk of bias and records its inconsistency as moderate. The reviewers judged all included trials at high risk of bias. One of the three trials in the pain analysis compared collagen with glucosamine sulfate rather than an inactive placebo. The authors write that the certainty of their results is limited by potential bias, small samples and inconsistency, and call for well-designed trials.
A 2025 network meta-analysis compared seven supplements for knee osteoarthritis, with no external funding. It found collagen improved pain on a visual scale and physical function compared with placebo, but not the WOMAC pain score (a standard osteoarthritis questionnaire). It pooled hydrolysed collagen with undenatured type II collagen, and did not rate certainty.
A 2025 systematic review of type I collagen hydrolysate found that the joint studies reported benefits. It states that most trials in its review were commercially funded, which raises concerns about potential conflicts of interest. One of its authors is employed by a collagen producer, which also paid its submission fees.
This rating follows the one source that rates certainty. The network meta-analysis's result on the WOMAC pain score is a partial disagreement.
Sources: Source [1]Source [2]Source [3]
Eases activity-related joint pain in people without arthritis Evidence rating: Limited
Active people take it hoping it will ease the joint aches that come with sport and exercise.
EFSA said a joint benefit in active people was not established; a 2021 review of 15 mostly commercially funded studies says all five on joint pain or joint injury reported benefits.
EFSA assessed an application from the company Gelita AG for collagen hydrolysate and joint health in physically active people. In 2011 it concluded that a cause-and-effect relationship had not been established. The one controlled trial it could use was in 147 student athletes with activity-related joint discomfort. Once the analysis was adjusted for multiple comparisons, that trial found no significant difference from placebo on any of its 15 endpoints. EFSA also declined to apply results from people with osteoarthritis to people without it.
A 2021 systematic review of collagen peptides with exercise includes that same trial. It reported that all five studies it found on joint pain or joint injury showed benefits. It concluded that collagen has the potential to reduce joint pain and improve joint function, especially alongside rehabilitation exercise. One of those five studies used undenatured type II collagen rather than hydrolysed collagen.
That review states that 13 of its 15 included studies were commercially funded. One of its authors has received research funding from a collagen manufacturer. The review states that the company did not have any involvement in the conception or writing of this review.
Neither source rates certainty. Because the sources disagree, this page takes the lower reading.
Sources: Source [1]Source [2]
Builds muscle and strength alongside exercise training Evidence rating: Limited
People who train take it hoping it will help them build muscle and get stronger.
A 2024 review of 19 trials found larger gains with collagen than placebo alongside training: in fat-free mass at moderate certainty, and in muscle size and strength at low certainty.
A 2024 meta-analysis pooled 19 trials in 768 healthy adults taking collagen peptides alongside exercise training. It found gains over placebo in fat-free mass (8 trials, moderate certainty). It also found gains in measures of muscle size (5 trials, low certainty) and in maximal strength (11 trials, low certainty, standardized mean difference 0.19).
Fat-free mass is everything in the body except fat, so it is not a measure of muscle alone. This page records the claim as low, the lowest certainty among the muscle and strength outcomes. The review rated its tendon outcomes very low.
The reviewers note that in the one trial comparing collagen with whey protein matched for leucine, whey increased muscle mass more. The review had no external funding and declares no conflicts. It does not report who funded the trials it included.
Sources: Source [1]
Increases bone mineral density Evidence rating: Limited
People take it hoping it will help keep their bones dense as they age.
A 2025 review pooling five widely varying trials, several also giving calcium and vitamin D, found higher bone density; another found the bone studies too limited for firm conclusions.
A 2025 meta-analysis reported that collagen peptides increased bone mineral density at the spine and femoral neck. It pooled five trials whose results varied substantially (I² 80.1%, a measure of how much the trial results differ). Several of those trials gave collagen together with calcium and vitamin D. The review applied GRADE to each included study rather than to each outcome, so it gives no certainty rating for bone density itself. It was publicly funded and declares no conflicts.
A 2025 systematic review limited to type I collagen hydrolysate found three reports on bone. Only one of them measured bone density, in a trial whose groups differed at baseline. The review concluded that the bone studies had limitations that prevent definitive conclusions. It states that most trials in its review were commercially funded, which raises concerns about potential conflicts of interest. One of its authors is employed by a collagen producer.
Neither review reports fracture results from trials; the meta-analysis calls for trials with fractures as the main outcome. Because the sources disagree, this page takes the lower reading.
Sources: Source [1]Source [2]
Which form the evidence used
Supplements are sold in several forms. This is what the sources on this page say about them — which form the research used, and any comparison those sources themselves made. Where they made none, this page does not rank the forms.
Hydrolysed collagen (collagen peptides), What the sources supportDetails
The cited reviews tested this form: the knee osteoarthritis meta-analysis included only hydrolysed collagen, the exercise meta-analysis only collagen peptides (no gelatine), and the skin meta-analysis hydrolysed collagen. Of the 26 trials in the 2023 skin meta-analysis, most used fish, a few cattle, pig or chicken, and nine gave no source; its per-source counts add up to 27. For hydration, its one chicken trial showed an effect of −0.03 (95% CI −0.40 to 0.34), an interval that includes no effect. The reviewers found no significant difference between sources for elasticity, and report no such test for hydration. This covers only that review's skin measures, which this page rates limited. No cited source compared sources for joints, muscle or bone. Source [1]Source [2]Source [3]
Undenatured type II collagen (for example UC-II), Details
A different product: it is not hydrolysed. The 2023 knee osteoarthritis meta-analysis did not include it. The 2025 knee network meta-analysis pooled it with hydrolysed collagen under a single "collagen" heading, and one of the joint-pain trials in the 2021 exercise review used it. No cited source reports its results separately, so the ratings below cannot be applied to it on its own. Source [1]Source [2]Source [3]
Sources
- Effects of Oral Collagen for Skin Anti-Aging: A Systematic Review and Meta-AnalysisNutrients, 2023 · Meta-analysis · Primary source
- Scientific Opinion on the substantiation of a health claim related to VeriSol®P and a change in skin elasticity leading to an improvement in skin function pursuant to Article 13(5) of Regulation (EC) No 1924/2006EFSA Panel on Dietetic Products, Nutrition and Allergies (EFSA Journal), 2013 · Health authority summary · Primary source
- Analgesic efficacy of collagen peptide in knee osteoarthritis: a meta-analysis of randomized controlled trialsJournal of Orthopaedic Surgery and Research, 2023 · Meta-analysis · Primary source · Certainty: very low
- Comparative Effectiveness of Nutritional Supplements in the Treatment of Knee Osteoarthritis: A Network Meta-AnalysisNutrients, 2025 · Meta-analysis · Primary source
- The Effects of Type I Collagen Hydrolysate Supplementation on Bones, Muscles, and Joints: A Systematic ReviewOrthopedic Reviews, 2025 · Systematic review · Primary source
- Scientific Opinion on the substantiation of a health claim related to collagen hydrolysate and maintenance of joints pursuant to Article 13(5) of Regulation (EC) No 1924/2006EFSA Panel on Dietetic Products, Nutrition and Allergies (EFSA Journal), 2011 · Health authority summary · Primary source
- The effects of collagen peptide supplementation on body composition, collagen synthesis, and recovery from joint injury and exercise: a systematic reviewAmino Acids, 2021 · Systematic review · Primary source
- Impact of Collagen Peptide Supplementation in Combination with Long-Term Physical Training on Strength, Musculotendinous Remodeling, Functional Recovery, and Body Composition in Healthy Adults: A Systematic Review with Meta-analysisSports Medicine, 2024 · Meta-analysis · Primary source · Certainty: very low
- Efficacy of collagen peptide supplementation on bone and muscle health: a meta-analysisFrontiers in Nutrition, 2025 · Meta-analysis · Primary source
- Scientific Opinion on the evaluation of allergenic foods and food ingredients for labelling purposesEFSA Panel on Dietetic Products, Nutrition and Allergies (EFSA Journal), 2014 · Health authority summary · Primary source
Page history
Show all 4 changes to this page
- — Rewritten to the content style guideline; ratings and sources unchanged.
- — Source audit. A 2025 skin meta-analysis (The American Journal of Medicine) was removed because its full text could not be read, and with it everything the page took from it: its funding and trial-quality subgroup results, its conclusion, and the page summary's sentence about it. The skin claim now rests on the 2023 meta-analysis and the 2013 EFSA opinion; its rating stays limited. Its heading no longer mentions wrinkles, because neither remaining source reports a result for them.
- — Published. All five claims, the safety section and every source passed the site's source and wording reviews. The unread Myung 2025 full text and letters, and two unread graded osteoarthritis reviews, stay logged in docs/08-open-decisions.md.
- — First draft. Skin, active-people joint pain, muscle and bone rated limited; knee osteoarthritis pain rated moderate on the one GRADE-rated review. Hair and nails left off: no systematic review was found.