Hydrolyzed Collagen: What the Science Says

Hydrolyzed collagen — sold as collagen peptides or collagen hydrolysate — is the one joint supplement with genuinely emerging evidence. A 2023 meta-analysis in the Journal of Orthopaedic Surgery and Research pooled 4 randomised controlled trials involving 507 patients with knee osteoarthritis and found a statistically significant reduction in pain scores compared with placebo ¹. The effect size was modest, not dramatic. Effective doses in those trials were around 10 g per day, taken for at least 3 months ¹⁴. This is a supplement measured in months, not days — and the effect is on pain and function, not on reversing structural joint damage.

What is hydrolyzed collagen?

Collagen is the most abundant protein in the body, forming the structural scaffold of cartilage, tendons, ligaments, bone and skin. In its raw form it is a large triple-helix molecule that the digestive system cannot absorb efficiently. Hydrolyzed collagen — also labelled as collagen peptides or collagen hydrolysate — is collagen that has been broken down with enzymes or heat into short peptide chains of 2,000–10,000 daltons. These smaller fragments are absorbed more readily from the gut and, crucially, have been detected in circulating blood and in cartilage tissue within hours of ingestion ⁵¹¹.

Once absorbed, these peptides appear to act as signalling molecules: they stimulate chondrocytes (cartilage cells) and fibroblasts to increase their own collagen and proteoglycan production, essentially reminding connective tissue to rebuild itself ⁵¹¹. This mechanism is biologically plausible — it is not just wishful thinking — but translating a plausible mechanism into a meaningful clinical effect requires evidence from trials in real patients.

What does the 2023 meta-analysis actually show?

The clearest summary of the trial evidence comes from a meta-analysis published in the Journal of Orthopaedic Surgery and Research in September 2023 ¹. The authors searched PubMed, Scopus, EMBASE, Web of Science, Cochrane and ClinicalTrials.gov for randomised controlled trials of collagen peptides in osteoarthritis, finding 4 eligible RCTs with a combined 507 patients, all with knee OA. The pooled analysis showed statistically significant improvements in pain (measured on visual analogue or WOMAC pain scales) in the collagen group versus placebo.

The honest caveats: the authors rated all four included trials as having a high risk of bias, the sample sizes were modest, follow-up durations ranged from 13 to 24 weeks, and there was meaningful heterogeneity between studies. A 2024 trial-sequential meta-analysis in Osteoarthritis and Cartilage reached broadly similar conclusions — real signal, modest magnitude, limitations in the underlying evidence base ³. A larger updated systematic review also published in 2024 (11 RCTs, 870 participants) found significant improvements in both pain and function, with collagen supplementation favoured, but noted the need for larger, more standardised trials ¹².

The take-away: the evidence is promising, not conclusive. Hydrolyzed collagen is in a different category from glucosamine — which major guidelines now recommend against — precisely because there is a plausible mechanism, accumulating trial data with a consistent direction of effect, and no serious safety concerns at 10 g/day. But it is not a proven therapy in the way paracetamol or physiotherapy are proven.

Tendon adaptation: what the RCTs show

Cartilage is not the only connective tissue studied. A 2017 randomised crossover trial by Shaw et al. at UC Davis gave participants either 15 g of vitamin C-enriched gelatin (a form of collagen) or a placebo 1 hour before a brief jumping protocol, and measured circulating markers of collagen synthesis ⁶. Gelatin plus vitamin C produced a doubling of procollagen I synthesis markers compared to placebo. This is a mechanistic study, not a pain trial, but it established the biological principle: collagen precursors, taken with vitamin C before exercise, significantly increase collagen synthesis — relevant to tendons and ligaments as well as cartilage ⁶.

A 2019 RCT by Praet et al. published in Nutrients took this further: 18 patients with chronic mid-portion Achilles tendinopathy received either 2.5 g of specific collagen peptides per day or placebo, combined with daily eccentric calf exercises, for 6 months in a cross-over design ⁸. VISA-A scores (a validated measure of Achilles tendon function and pain) improved significantly more in the collagen group: +12.6 points versus +5.3 points in placebo after 3 months. A 2021 systematic review by Khatri et al. across 15 RCTs concluded that collagen peptide supplementation in conjunction with exercise was beneficial for managing degenerative bone and joint disorders and supported connective tissue recovery ⁵. The key phrase: in conjunction with exercise. Collagen supplements are not a substitute for loading the tendon — they appear to amplify the adaptive response to that loading.

The vitamin C connection — the only legally claimable link

EU law (Regulation EC No 1924/2006) prohibits health claims on food supplements unless they have been specifically authorised by the European Food Safety Authority (EFSA). Under this framework, collagen itself carries no authorised joint claim — so any product advertising "supports your joints" based purely on its collagen content is on shaky legal ground.

Vitamin C is different. EFSA has authorised the claim that vitamin C contributes to normal collagen formation for the normal function of cartilage (ID 131) and bones (ID 130) ⁷. This is because vitamin C is an obligatory cofactor for prolyl hydroxylase and lysyl hydroxylase — the enzymes that stabilise and cross-link collagen molecules into functional tissue. Without adequate vitamin C, collagen fibres are structurally fragile (this is why severe deficiency causes scurvy — connective tissue literally breaks down).

The practical implication: the Shaw 2017 trial showed that adding 48 mg of vitamin C to gelatin before exercise more than doubled collagen synthesis markers ⁶. Taking collagen without vitamin C is probably less effective than taking it with a vitamin C source — whether that is a glass of orange juice, kiwi fruit or a low-dose supplement. Most people with a varied diet are not deficient in vitamin C, but it is a simple, evidence-grounded pairing.

Practical guidance: dose, form, timing

The following reflects what the trial literature describes — it is attributed information, not a prescription, and you should discuss any supplement with your doctor or pharmacist, particularly if you take medications or have a medical condition.

Dose: trials showing joint pain benefit used 10 g per day of hydrolyzed collagen (collagen peptides) ¹⁴. The tendon-focused Praet trial used only 2.5 g of a specific branded peptide formulation ⁸ — doses vary by formulation and endpoint.
Duration: minimum 3 months before expecting any detectable effect; some trials ran to 6 months ¹⁴. Discontinue if no benefit is apparent by month 4.
Timing: the Shaw 2017 evidence suggests taking collagen (or gelatin) 30–60 minutes before exercise to maximise delivery to active tissues ⁶.
Vitamin C pairing: take with a food or drink containing vitamin C at the same time ⁶⁷.
Form: hydrolyzed collagen powder dissolved in liquid is the form used in most trials. Products labelled "collagen peptides" or "marine collagen hydrolysate" are generally the same category. Undenatured type II collagen (UC-II) is a different product with a different mechanism and dose (~40 mg) ³.
Safety: no serious adverse effects have been reported at 10 g/day in trials to date ¹⁴. It is a food protein. Vegans should note that most collagen supplements are animal-derived (bovine or marine).

Who should NOT expect miracles

This section matters as much as the evidence for benefit. Hydrolyzed collagen is not appropriate as a primary treatment, and some expectations are simply not supported by the science.

Advanced osteoarthritis with severe structural damage: the trials involved patients with mild-to-moderate knee OA. If joint space is severely narrowed or bone-on-bone contact is present, a supplement that supports cartilage matrix is unlikely to produce meaningful functional change. NICE NG226 and the 2019 ACR guidelines are unambiguous: exercise-based rehabilitation and, in severe cases, orthopaedic review, are the priority ⁹¹⁰.
Anyone expecting fast results: participants in the RCTs experienced gradual, modest improvement over weeks and months — not rapid or dramatic relief ¹⁴. If you need meaningful pain control now, collagen is not the right primary tool.
Using collagen instead of treatment or exercise: the tendon and joint data consistently shows better results when collagen is combined with exercise, not used instead of it ⁵⁸. Replacing physiotherapy, osteopathy or a structured exercise programme with a supplement is not supported by the evidence.
Inflammatory arthritis (rheumatoid arthritis, psoriatic arthritis): the evidence base is for osteoarthritis. Inflammatory arthropathies are driven by immune mechanisms that collagen peptides do not address. These require specialist medical management.
Expecting it to rebuild cartilage visible on MRI: no supplement trial to date has demonstrated radiographically measurable cartilage regeneration in humans. The measured outcomes are pain and function scores — clinically relevant, but not structural reversal.

Skin versus joints: same product, different evidence base

Many collagen products are marketed for skin, and the skin evidence is actually more consistent than the joint evidence — a 2023 systematic review and meta-analysis covering 967 participants found significant improvements in skin moisture and elasticity with hydrolyzed collagen supplementation ³¹¹. The mechanism for skin (stimulating dermal fibroblasts) partially overlaps with the mechanism for cartilage (stimulating chondrocytes), which is why the same product category has been studied for both.

However, do not assume interchangeability. The doses, formulations, peptide molecular weights, and outcome measures differ between skin and joint studies. A product optimised for skin hydration may not be the same formulation used in the Achilles tendinopathy RCT. If you are taking collagen specifically for a joint or tendon problem, look for products and doses consistent with the joint literature.

Exercise first — the non-negotiable foundation

No honest discussion of joint health can skip this: NICE guideline NG226 (2022) and the 2019 ACR/Arthritis Foundation guidelines both place therapeutic exercise at the centre of osteoarthritis management ⁹¹⁰. Exercise improves pain, function and quality of life through mechanisms — strengthening periarticular muscles, improving proprioception, reducing inflammatory mediators — that no supplement replicates. A patient who does structured exercise and takes collagen is on firmer ground than a patient who takes collagen instead of exercise.

As an osteopath, the question I hear most often is not "does this supplement work?" but "why do my joints keep hurting?" The answer almost always involves movement patterns, load distribution, muscle function and tissue quality — things that hands-on assessment and guided exercise address directly. Collagen may play a supporting role in that picture. It is not the main event.

Already taking supplements for pain or joints? Bring the bottles to your first visit — we'll review them together, no extra cost and nothing to sell you. → Book / WhatsApp

FAQ

Does hydrolyzed collagen actually work for joint pain?

The honest answer is: modestly, and only over time. A 2023 meta-analysis published in the Journal of Orthopaedic Surgery and Research pooled 4 randomised controlled trials involving 507 patients with knee osteoarthritis and found statistically significant reductions in pain scores favouring collagen peptides over placebo ¹. The effect was real but small — meaningfully different from a dramatic cure. The authors noted all included trials had a high risk of bias, so these results should be interpreted with caution.

How long does it take and how much should I take?

The trials showing benefit typically used 10 g per day and ran for at least 3 months, with some extending to 6 months ¹⁴. This is not a supplement you will notice in a week. If you have not seen any change after 3–4 months of consistent daily use, the evidence does not support continuing. Discuss dosage with your doctor or pharmacist — these are attributed figures from the research literature, not a prescription from this clinic.

Which type of collagen is best — type I, type II, hydrolysate, or peptides?

Most of the joint-pain evidence uses hydrolyzed collagen (also called collagen peptides or collagen hydrolysate), which is broken down into shorter chains for easier absorption ¹⁴. Undenatured type II collagen (UC-II, ~40 mg/day) works via a completely different immune-tolerance mechanism and has its own trial base ³. Both are different from the gelatin used in some tissue-synthesis research ⁶. There is no head-to-head trial large enough to declare a winner; the evidence base for each is still emerging.

Why does vitamin C matter with collagen?

Vitamin C is an essential cofactor for the enzymes that cross-link collagen fibres into stable tissue. This is not marketing language — it is basic biochemistry confirmed by the European Food Safety Authority (EFSA), which has authorised the claim that vitamin C contributes to normal collagen formation for the normal function of cartilage (Reg. EC 1924/2006, ID 131) ⁷. A 2017 RCT by Shaw et al. in the American Journal of Clinical Nutrition showed that vitamin C-enriched gelatin taken before intermittent exercise significantly increased circulating collagen synthesis markers ⁶. If you take collagen supplements, pairing them with a vitamin-C source (food or supplement) is scientifically reasonable.

Does the evidence for skin and the evidence for joints come from the same research?

No — they are largely separate bodies of evidence. The skin literature tends to use lower doses (2.5–10 g/day) and measures outcomes like skin elasticity and hydration ³. The joint/OA literature uses higher doses (10 g/day) and measures pain and function scores ¹. Both suggest real but modest effects. Do not assume that a collagen product marketed for skin will behave identically in your cartilage or tendons, or vice versa. The mechanisms partly overlap but are not identical.

Get the printable sheet (free)

Hydrolyzed collagen: what the science actually says