Hydrolysed collagen: wrinkles, joints… does it really work?
In short: hydrolysed collagen is well absorbed, but that does not prove it rejuvenates skin or repairs cartilage. On skin, the effect is modest (slightly better hydration and elasticity) and disappears in independent and better-designed studies; on wrinkles, the evidence is lacking. For knee osteoarthritis, pain drops a little, but no major rheumatology society recommends it. The product is safe for most people, except those with a fish allergy or kidney stones. If you want to try it, run a three-month test on a limited budget and judge for yourself.
Updated 14 October 2026.
What exactly is collagen?
Collagen is the most abundant protein in the body. It acts as a kind of “scaffolding”: it gives strength to skin, bones, tendons and cartilage. There are several types:
- type I, found in skin, bones and tendons;
- type II, found in joint cartilage;
- type III, found alongside type I in skin and blood vessels.
Collagen sold as a supplement comes from animals: pig skin, cowhide, fish skin or scales (“marine collagen”), chicken feet or cartilage. There is no such thing as plant collagen: “vegan” products that mention it usually contain vitamin C and amino acids, not collagen.
Hydrolysed, peptides, daltons: three words to understand
- Hydrolysed means “cut up”. Raw collagen is a very long molecule. Manufacturers cut it into small pieces using enzymes.
- These pieces are called peptides: short chains of amino acids, the building blocks of proteins.
- Their size is measured in daltons (Da), the unit of molecular mass. Depending on the manufacturer and the process, the peptides in commercial products are often in the range of 2,000 to 5,000 Da: very small fragments.
- Gelatine, by contrast, is simply cooked collagen (described as “denatured”): it is what makes jelly desserts set.
It is well absorbed… but then what?
After a dose of hydrolysed collagen, small fragments show up in the blood, mainly a “dipeptide” (two linked amino acids) called Pro-Hyp (prolyl-hydroxyproline). A Japanese study in volunteers showed this in 2005 (Iwai et al., 2005), and a 2024 study confirmed it (Virgilio et al., 2024). The latter was funded by Rousselot, a major collagen producer based in Ghent.
So absorption is proven. What is not proven is the next step: that these fragments go on to “instruct” the skin or cartilage to make more collagen in humans. That is the hypothesis clinical trials are trying to test. The body digests most proteins into amino acids and small peptides such as Pro-Hyp, then reuses them wherever it needs them: eating collagen is not the same as “topping up” your skin with collagen.
What you can expect: the verdict table
| Claim | What the studies show | Level of evidence |
|---|---|---|
| Skin hydration and elasticity | Modest effect when all studies are pooled; no effect in independent or high-quality studies | Limited, highly dependent on funding |
| Fewer wrinkles | Inconsistent results, not significant in the most recent meta-analysis | No solid evidence |
| Knee osteoarthritis pain | Pain reduction slightly above the noticeable threshold, but studies are very heterogeneous and often of low quality | Limited |
| Bone density after menopause | One positive 12-month trial, funded by a manufacturer, with no measurement of fractures | Very limited |
| Muscle in older people | One positive trial with strength training, funded by a manufacturer, disputed by other researchers | Very limited |
| Nails, hair | Small studies, often without a placebo | Very weak |
| Tendons, sports recovery | Changes in blood markers, no clear clinical benefit | Very weak |
A randomised controlled trial randomly assigns who receives the product and who receives a placebo (a dummy product that looks identical). A meta-analysis statistically combines several trials: it is only as good as the trials it includes.
Skin: wrinkles, hydration, elasticity
This is the most heavily marketed promise, and the one where recent studies have changed the picture the most.
The study that toppled the myth (2025)
In 2025, two researchers, S.-K. Myung and Y. Park, published a meta-analysis in The American Journal of Medicine of 23 randomised trials totalling 1,474 participants, followed for an average of 12 weeks (Myung & Park, 2025). When everything is pooled, collagen appears to improve hydration, elasticity and wrinkles. But the authors sorted the studies:
- by funding: industry-funded studies show an effect on every outcome; studies without industry funding show none;
- by quality: high-quality trials show no significant effect, whether on hydration, elasticity or wrinkles.
Their conclusion: there is “currently no clinical evidence” to justify collagen for preventing or treating skin ageing. The industry disputed how some studies were classified and the way different products were lumped together (17 of the 23 trials came from Asia, often at around 3 g a day). These criticisms are fair to discuss, but they do not make an effect appear in the independent studies.
What other analyses say
- 2023, 26 trials (Pu et al., 2023): improvement in hydration and elasticity, with no analysis of wrinkles and no breakdown by funding.
- 2026, 35 trials, 2,534 participants (Batool et al., 2026): slightly better hydration; elasticity only improves after removing two very influential studies; no effect on wrinkles. The authors note clear publication bias, meaning that negative studies tend never to be published, which inflates the average effect.
- 2026, “umbrella” review (a synthesis of 16 meta-analyses and 113 trials): a “modest” improvement in elasticity and hydration, less consistent results on wrinkles (Ravindran et al., 2026).
- October 2026: David Robert Grimes and dermatologist Anjali Mahto reanalysed two meta-analyses from 2021 and 2023. One of them presented a non-significant result as significant. Once corrected, there is no longer any link between dose and effect. Their verdict: oral collagen “should not be promoted or recommended” in dermatology until good-quality independent studies are available (Grimes & Mahto, 2026).
In plain terms: if you are hoping to erase your wrinkles, the best studies say no. A slight gain in hydration is possible, nothing more. The proven measures for skin remain daily sun protection, stopping smoking and, on medical advice, retinoid creams (see our article bakuchiol or retinol (in French)).
Joints and osteoarthritis
Hydrolysed collagen
Osteoarthritis is a disease of the whole joint, in which the cartilage covering the bones gradually wears down; it most often affects the knee, hip or hands. Studies measure pain on a visual analogue scale (VAS) from 0 to 100 mm, and disability with the WOMAC, a questionnaire on pain, stiffness and the ability to move.
- A 2025 meta-analysis (11 trials, 870 participants) found pain reduced by 13.6 mm out of 100 more than with placebo (Simental-Mendía et al., 2025). That is slightly above the 9 mm threshold the authors consider noticeable to the patient. But the studies contradict each other a great deal, and the total WOMAC score does not reach the minimal difference a patient really feels.
- Another analysis from 2023 (4 trials, 507 patients) points the same way, with trials all judged at high risk of bias (Lin et al., 2023).
Undenatured type II collagen (UC-II)
This is a different product: minimally processed chicken cartilage, taken at a very small dose (40 mg a day, of which only 1.2 mg is active collagen). It is not supposed to “rebuild” cartilage, but to work through oral tolerance: the gut would “train” the immune system to attack cartilage less. The reference trial (191 patients, 6 months) is positive, but it was funded by the manufacturer, and two of its three authors were employees (Lugo et al., 2016). A 2025 review considers the product safe, but calls for larger trials before routine use (Gupta & Maffulli, 2025).
What rheumatologists say
None of the major professional societies consulted recommends collagen for osteoarthritis. Collagen is not among the treatments recommended by the European (ESCEO, 2019) and American (ACR, 2019) societies; the Royal Australian College of General Practitioners (RACGP) says it can recommend it “neither for nor against”. What works best, and for the long term, remains suitable physical activity, weight loss if you are overweight and, if needed, painkillers on medical advice.
Bones, muscles, tendons, nails and hair
- Bones: a German trial gave 5 g of peptides a day for a year to 131 postmenopausal women with low bone density. Their bone mineral density (the amount of mineral in the bone, measured by a special X-ray called DXA, an indicator of fracture risk) rose slightly compared with placebo (König et al., 2018). But the trial was partly funded by the manufacturer GELITA, the groups were not identical at the start, and fractures were not measured. Calcium, vitamin D (see our article vitamin D) and physical activity remain the foundation.
- Muscles: 53 older men with sarcopenia (age-related muscle loss) who were doing strength training gained more muscle with 15 g of collagen than with a placebo (Zdzieblik et al., 2015). Sports nutrition specialists described these results as “exceptional”, 2.7 to 5.6 times larger than those of studies on standard proteins (Phillips et al., 2016). Collagen is also an incomplete protein: it lacks tryptophan, an essential amino acid. For muscle, a complete protein (dairy, eggs, pulses) makes more sense. See also our article on sarcopenia (in French).
- Tendons and sport: 15 g of vitamin C–enriched gelatine before exercise doubles a blood marker of collagen synthesis, in 8 men (Shaw et al., 2017). A marker is not a cure: no clear benefit for recovery or tendons has been shown.
- Nails and hair: small studies, often without a placebo or with products mixed with other ingredients. No conclusions can be drawn.
What the law forbids brands to promise you
This is the most telling point, and the least known.
- No health claim is authorised for collagen in the European Union. A health claim is a statement such as “contributes to…”; it is only allowed if it appears on the official list (Regulation (EU) 432/2012), after assessment by EFSA, the European Food Safety Authority.
- EFSA rejected two applications submitted by a manufacturer: collagen and joints in 2011, collagen and skin elasticity in 2013. In both cases, it found that a cause-and-effect relationship had not been established (EFSA, 2011, 2013).
- The trick: vitamin C, on the other hand, is allowed the claim “contributes to normal collagen formation for the normal function of skin, bones, cartilage…”. That is why so many collagen products contain a little vitamin C: the statement legally refers to the vitamin, not to the collagen.
- In Belgium, a supplement is notified to the FPS Public Health, not authorised. The FPS says so itself: the notification number “is not an approval or authorisation of the product”, and effectiveness is not assessed (FPS Public Health, 2026).
- In September 2024, Test Achats filed a complaint with the FASFC and the FPS Economy against several brands over prohibited claims (Test Achats, 2024).
Risks to be aware of
Collagen is well tolerated by most people: in skin trials, few or no adverse effects were reported. A few situations call for caution:
- Fish allergy: 21% of people allergic to fish also react to fish collagen, and this collagen remains allergenic after cooking (Kalic et al., 2020). Severe reactions have been described. If you are allergic to fish, avoid marine collagen unless your allergist advises otherwise.
- Red meat allergy (alpha-gal syndrome): reactions to bovine gelatine have been described.
- Kidney stones: gelatine increases oxalate in the urine at doses from 5 to 10 g, the usual supplement dose (Knight et al., 2006). Oxalate is the main component of most stones, even though no study directly shows that collagen causes them. If you have had stones before, or if you have kidney disease, ask your doctor.
- Heavy metals: an Italian study found no exceedance of European limits in fish and jellyfish collagens (Cammilleri et al., 2025). An American report not published in a journal flags them in some products bought online. Choose a brand that publishes its test results.
- Pregnancy and breastfeeding: little data; ask your doctor or midwife.
Adverse effects of a supplement can be reported to the FPS Public Health (nutrivigilance).
In practice: if you still want to try it
Nothing stops you from trying, as long as you know what you are buying.
- Set yourself a measurable goal: a photo of your skin in the same light, or your knee pain scored from 0 to 10 every week.
- Dose: most trials use 2.5 to 10 g of hydrolysed collagen a day, as a powder to dissolve. Beyond that, nothing shows that more collagen gives more effect. Be wary of capsules: you often need four or more to reach 2.5 g.
- Duration: at least 8 to 12 weeks, the length of most trials. If there is no clear change after three months, stop.
- Price: from €15 to more than €200 a month depending on the brand. The most expensive products are not the best studied.
- Read the label: source (bovine, porcine, marine), actual amount of collagen per dose, no promises such as “erases wrinkles” (which are prohibited).
- Bone broth is no substitute for a powder: its collagen content varies greatly from one recipe to another and is generally much lower than that of a supplement dose (Alcock et al., 2019).
Going further: why these studies contradict each other
If the results vary so much, it is no accident. Several flaws keep coming up in collagen studies:
- Funding: the overwhelming majority of trials are funded by manufacturers or run by researchers linked to them. This is not proof of fraud, but it is a well-known driver of more favourable results, and the 2025 meta-analysis shows it clearly.
- Fragile blinding: in a “double-blind” trial, neither the participant nor the researcher knows who is getting the real product. But collagen has a recognisable taste and texture: a participant may guess what they are taking.
- Products that are all different: source, peptide size, dose (from 0.3 to 10 g), other ingredients (hyaluronic acid, vitamins). Pooling these studies is like comparing apples and pears.
- Short, small studies: 8 to 12 weeks for most, a few dozen participants.
- Effect size: researchers often express it as an SMD (“standardised mean difference”), a unitless measure; around 0.2 the effect is small, around 0.8 it is large. They also give a confidence interval, the plausible range of the effect: if it includes zero, the effect is not demonstrated.
The day several large, long, well-blinded independent trials show an effect, this verdict will change. For now, that is not the case.
Test yourself
Sources
Alcock, R. D., Shaw, G. C., & Burke, L. M. (2019). Bone broth unlikely to provide reliable concentrations of collagen precursors compared with supplemental sources of collagen used in collagen research. International Journal of Sport Nutrition and Exercise Metabolism, 29(3), 265–272. https://doi.org/10.1123/ijsnem.2018-0139
Batool, A., Abbas, M. S., Farooq, R., Akbar, M. A., Ahmed, A., Kanwal, W., Ali, H., Ahmad, A., Junaid, M., Hanif, M. S., Nunes Junior, E. B. D., & Aminpoor, H. (2026). Oral collagen peptides and skin rejuvenation: A systematic review and an updated meta-analysis of randomized controlled trials. Journal of Cosmetic Dermatology, 25(7), e71041. https://doi.org/10.1111/jocd.71041
Cammilleri, G., Tortorici, M., Pantano, L., Galluzzo, F. G., Pulvirenti, A., Buscemi, M. D., Lo Dico, G. M., Macaluso, A., Calabrese, V., Jacob, U. M., & Ferrantelli, V. (2025). Toxic metals and metalloids in collagen supplements of fish and jellyfish origin: Risk assessment for daily intake. Open Medicine. https://doi.org/10.1515/med-2025-1141
EFSA NDA Panel. (2011). Scientific opinion on the substantiation of a health claim related to collagen hydrolysate and maintenance of joints. EFSA Journal, 9(7), 2291. https://doi.org/10.2903/j.efsa.2011.2291
EFSA NDA Panel. (2013). 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. EFSA Journal, 11(6), 3257. https://doi.org/10.2903/j.efsa.2013.3257
European Commission. (2012). Règlement (UE) n° 432/2012 de la Commission du 16 mai 2012 établissant une liste des allégations de santé autorisées portant sur les denrées alimentaires. Journal officiel de l'Union européenne, L 136, 1–40. https://eur-lex.europa.eu/eli/reg/2012/432/oj
FPS Public Health, Food Chain Safety and Environment. (2026, July 2). Commercialisation des compléments alimentaires et aliments enrichis. https://www.health.belgium.be/fr/professionnels/entreprises/alimentation/aliments-specifiques/complements-alimentaires-aliments-enrichis/commercialisation-complements-alimentaires-aliments-enrichis
Grimes, D. R., & Mahto, A. (2026). Hydrolyzed collagen for skin aging meta-analyses are unreliable. Journal of Cosmetic Dermatology. https://doi.org/10.1111/jocd.71208
Gupta, A., & Maffulli, N. (2025). Undenatured type II collagen for knee osteoarthritis. Annals of Medicine, 57(1). https://doi.org/10.1080/07853890.2025.2493306
Iwai, K., Hasegawa, T., Taguchi, Y., Morimatsu, F., Sato, K., Nakamura, Y., Higashi, A., Kido, Y., Nakabo, Y., & Ohtsuki, K. (2005). Identification of food-derived collagen peptides in human blood after oral ingestion of gelatin hydrolysates. Journal of Agricultural and Food Chemistry, 53(16), 6531–6536. https://doi.org/10.1021/jf050206p
Kalic, T., Kamath, S. D., Ruethers, T., Taki, A. C., Nugraha, R., Le, T. T. K., Humeniuk, P., Williamson, N. A., Hira, D., Rolland, J. M., O'Hehir, R. E., Dai, D., Campbell, D. E., Breiteneder, H., & Lopata, A. L. (2020). Collagen—An important fish allergen for improved diagnosis. The Journal of Allergy and Clinical Immunology: In Practice, 8(9), 3084–3092. https://doi.org/10.1016/j.jaip.2020.04.063
Knight, J., Jiang, J., Assimos, D. G., & Holmes, R. P. (2006). Hydroxyproline ingestion and urinary oxalate and glycolate excretion. Kidney International, 70(11), 1929–1934. https://doi.org/10.1038/sj.ki.5001906
König, D., Oesser, S., Scharla, S., Zdzieblik, D., & Gollhofer, A. (2018). Specific collagen peptides improve bone mineral density and bone markers in postmenopausal women—A randomized controlled study. Nutrients, 10(1), 97. https://doi.org/10.3390/nu10010097
Lin, C.-R., Tsai, S. H. L., Huang, K.-Y., Tsai, P.-A., Chou, H., & Chang, S.-H. (2023). Analgesic efficacy of collagen peptide in knee osteoarthritis: A meta-analysis of randomized controlled trials. Journal of Orthopaedic Surgery and Research, 18, 694. https://doi.org/10.1186/s13018-023-04182-w
Lugo, J. P., Saiyed, Z. M., & Lane, N. E. (2016). Efficacy and tolerability of an undenatured type II collagen supplement in modulating knee osteoarthritis symptoms: A multicenter randomized, double-blind, placebo-controlled study. Nutrition Journal, 15, 14. https://doi.org/10.1186/s12937-016-0130-8
Myung, S.-K., & Park, Y. (2025). Effects of collagen supplements on skin aging: A systematic review and meta-analysis of randomized controlled trials. The American Journal of Medicine, 138(9), 1264–1277. https://doi.org/10.1016/j.amjmed.2025.04.034
Phillips, S. M., Tipton, K. D., van Loon, L. J. C., Verdijk, L. B., Paddon-Jones, D., & Close, G. L. (2016). Exceptional body composition changes attributed to collagen peptide supplementation and resistance training in older sarcopenic men. British Journal of Nutrition. https://doi.org/10.1017/S000711451600221X
Pu, S.-Y., Huang, Y.-L., Pu, C.-M., Kang, Y.-N., Hoang, K. D., Chen, K.-H., & Chen, C. (2023). Effects of oral collagen for skin anti-aging: A systematic review and meta-analysis. Nutrients, 15(9), 2080. https://doi.org/10.3390/nu15092080
Ravindran, R., Pizzol, D., López-Gil, J. F., Rahmati, M., Boyer, L., Fond, G., Butler, L., Stellato, A., Gawronska, J., Barnett, Y., et al. (2026). Collagen supplementation for skin and musculoskeletal health: An umbrella review of meta-analyses on elasticity, hydration, and structural outcomes. Aesthetic Surgery Journal Open Forum, 8. https://doi.org/10.1093/asjof/ojag018
Shaw, G., Lee-Barthel, A., Ross, M., Wang, B. H., & Baar, K. (2017). Vitamin C–enriched gelatin supplementation before intermittent activity augments collagen synthesis. The American Journal of Clinical Nutrition, 105(1), 136–143. https://doi.org/10.3945/ajcn.116.138594
Simental-Mendía, M., Ortega-Mata, D., Acosta-Olivo, C. A., Simental-Mendía, L. E., Peña-Martínez, V. M., & Vilchez-Cavazos, F. (2025). Effect of collagen supplementation on knee osteoarthritis: An updated systematic review and meta-analysis of randomised controlled trials. Clinical and Experimental Rheumatology, 43(1), 126–134. https://doi.org/10.55563/clinexprheumatol/kflfr5
Test Achats. (2024, September 16). Dangers et intérêts des compléments alimentaires au collagène. https://www.test-achats.be/sante/alimentation-et-nutrition/complements-alimentaires/news/dangers-et-interets-des-complements-alimentaires-au-collagene
Virgilio, N., Schön, C., Mödinger, Y., van der Steen, B., Vleminckx, S., van Holthoon, F. L., Kleinnijenhuis, A. J., Silva, C. I. F., & Prawitt, J. (2024). Absorption of bioactive peptides following collagen hydrolysate intake: A randomized, double-blind crossover study in healthy individuals. Frontiers in Nutrition, 11, 1416643. https://doi.org/10.3389/fnut.2024.1416643
Zdzieblik, D., Oesser, S., Baumstark, M. W., Gollhofer, A., & König, D. (2015). Collagen peptide supplementation in combination with resistance training improves body composition and increases muscle strength in elderly sarcopenic men: A randomised controlled trial. British Journal of Nutrition, 114(8), 1237–1245. https://doi.org/10.1017/S0007114515002810