Do collagen supplements work? What the research shows

Is collagen just hype?

Not entirely, but the promises run well ahead of the science. Collagen supplements have been tested in dozens of randomised controlled trials, and several meta-analyses pool their results. Some find measurable, modest changes. Others, which look harder at who paid for the studies and how well they were run, find little or nothing.

What is certain is the regulatory position. The European Food Safety Authority (EFSA) has assessed claims for collagen hydrolysate – for example on joints – and did not accept them. No health claim for collagen is authorised in the UK or the EU, so no product can legally promise that collagen improves your skin, joints, hair or nails.

Is collagen good for skin?

Skin is the most studied area. Most trials give adults – mostly women, often aged between their 30s and 60s – a daily dose of hydrolysed collagen, usually 2.5–10 g, for 8–12 weeks and measure skin hydration, elasticity and wrinkle depth with instruments.

• A 2021 meta-analysis of 19 trials (1,125 participants, 95% women) found improvements in hydration and elasticity compared with placebo.
• A 2023 meta-analysis of 26 trials (1,721 participants) also found better hydration and elasticity, but flagged several sources of bias in the studies.
• A 2025 meta-analysis of 23 trials split them by funding and quality. Industry-funded and low-quality studies showed effects; studies without industry funding and high-quality studies showed no effect on hydration, elasticity or wrinkles.

Put together: there may be a small effect on how well-hydrated and springy the surface of the skin is, but it is far from proven, and there is little to suggest collagen changes deep wrinkles. Sun protection and not smoking have much stronger evidence for skin ageing.

Do collagen supplements work on joints?

The joint research falls into two groups, and the trials in both are relatively small.

• Osteoarthritis: a 2019 meta-analysis of placebo-controlled trials found lower overall symptom and stiffness scores, but no significant difference in the pain or physical function sub-scores of the same questionnaire.
• Athletes with activity-related joint pain: a 24-week US study in student athletes (10 g a day) and a 12-week German study (5 g a day) reported less activity-related joint pain than with a placebo. The differences were modest, and in the US study only 97 of the 147 people who started could be included in the analysis.

Does collagen help bones, hair or nails?

Bones: one German trial gave 131 post-menopausal women with low bone density 5 g of collagen peptides a day for a year and reported a small rise in bone mineral density compared with placebo. It has not been widely repeated, and 102 of the 131 women completed it. Osteoporosis is a medical condition – speak to your GP about it.

Hair and nails: the evidence is thin. There are a few small studies of brittle nails, usually without a proper placebo group, and almost nothing reliable on hair. Many “hair, skin and nails” products also contain biotin, zinc or other nutrients, which makes it impossible to tell what any effect comes from.

Why is the evidence on collagen so mixed?

Researchers who are sceptical of collagen supplements point to the same handful of problems:

• Small studies – many trials have fewer than 100 participants, and some only a few dozen.
• Industry funding – a large share of trials are paid for by the companies that sell the collagen being tested, and those trials are more likely to report positive results.
• Different doses and products – studies use anything from 2.5 to 15 g a day, different sources and sometimes added vitamins, so results are hard to compare.
• Short follow-up – most run for 8–12 weeks; long-term effects are largely unknown.
• A plausibility gap – collagen is digested into amino acids and small peptides. It does not reach the skin or joints intact, and the idea that certain peptides act as signals in tissues is still being worked out.

Are there collagen supplements that actually work?

No collagen product has been shown to work better than another in independent research, and none is allowed to claim an effect on skin, joints, hair or nails. Some studies test branded peptides, but the results of one trial cannot be applied to every product with a similar name.

What you can check is what is in the tub or pouch: the collagen source (bovine, marine, porcine), how many grams of collagen a daily serving gives, and what else has been added. Collagen is also a protein, so it counts towards your daily protein intake – see How much collagen per day?.

Does collagen powder work better than capsules or drinks?

There is no good evidence that the format matters. What differs is the dose: studies have used 2.5–15 g a day, which would mean a lot of capsules, whereas a powder or a drink can give that amount in one serving. Liquid “collagen shots” often contain much less collagen than the studies used, plus sugar or sweeteners.

How do I judge a collagen study?

• Was it randomised and placebo-controlled, and were participants and researchers blinded?
• How many people took part, and for how long?
• Who funded it, and did the authors work for the manufacturer?
• Was the result measured with instruments, or only how people said they felt?
• Was the effect big enough to notice in daily life, or only statistically significant?
• Has an independent group repeated it?

What are the signs that collagen is working?

There is no reliable sign you can see or feel. In studies, changes in skin hydration or elasticity were picked up by measuring instruments, not always by the participants themselves, and usually after 8–12 weeks. If you notice a difference, it may be real, but changes in season, skincare, sleep and diet can all have similar effects – which is exactly why trials need a placebo group.