Evidence Files

Collagen for Skin: The Benefit Vanishes When Someone Else Pays for the Trial

A 2025 meta-analysis of 23 randomised trials found that collagen supplements improve skin hydration, elasticity and wrinkles. Then the authors split the trials by who paid for them. In the ones without pharmaceutical funding, the effect disappeared — and it disappeared again in the high-quality trials.

Published: 27 September 2026

Verdict

TRICK (weak evidence)

We searched PubMed and Cochrane Library for oral collagen and skin. This is a crowded field: 126 records tagged as randomised trials for collagen peptides and skin, and 36 systematic reviews or meta-analyses. Cochrane has not reviewed oral collagen for skin ageing at all. Most of the reviews come out positive, but the one that asked the obvious question found the benefit confined to industry-funded and lower-quality trials. A 2026 review of 25 trials rated the majority at high risk of bias.

Why Europe keeps drinking it

Collagen sits at the centre of what the industry calls “beauty from within”: powders stirred into morning coffee, ampoules, gummies, sachets sold next to the face creams rather than the vitamins. The pitch is intuitive to the point of being hard to question. Skin loses collagen with age; wrinkles follow; so put collagen back in. No other supplement has a story that makes quite so much sense on first hearing.

The EU regulatory picture is revealing. Under Regulation (EU) No 432/2012, the list of permitted health claims on food, collagen has no authorised claim of its own. What is authorised is a claim for a different ingredient: “Vitamin C contributes to normal collagen formation for the normal function of skin.” Look at the back of a collagen box sold in Europe and you will usually find vitamin C in the formula — which lets the label talk about collagen and skin without attributing anything to the collagen itself.

So the marketing and the regulation are having two different conversations. The front of the pack promises firmer, plumper skin from the protein you are drinking. The legally defensible sentence on the back is about a vitamin that costs a fraction of the price.

What the science actually says

The collagen evidence is not empty, and it is not uniformly negative. It is a textbook case of an effect that depends on which trials you let into the room.

Pooled, it looks like it works. Several meta-analyses report benefits. One pooled 26 trials with 1,721 participants and found hydrolysed collagen improved skin hydration versus placebo, while noting several biases in the included trials. An earlier one of 19 double-blind trials in 1,125 participants — 95% of them women — reported better hydration, elasticity and wrinkles
Split by funding, it stops working. The 2025 meta-analysis in The American Journal of Medicine (23 RCTs, 1,474 participants) was the first to break the results down by who paid. Trials funded by pharmaceutical companies showed significant effects. Trials without that funding showed no effect on hydration, elasticity or wrinkles. Restricting to high-quality studies gave the same result: no significant effect in any category. The authors’ conclusion: there is currently no clinical evidence to support collagen supplements for preventing or treating skin ageing
The trials themselves are the weak link. A 2026 systematic review of 25 RCTs counted favourable results in most of them — hydration improved in 10 of 15 trials, wrinkles in 9 of 10 — but rated the majority at high risk of bias, with sample sizes as small as 13 participants. Its own verdict is “preliminary evidence”
What an industry trial looks like. A typical example: a 2016 double-blind trial found that a collagen hydrolysate rich in two specific peptides beat both placebo and a lower-peptide version on facial moisture, elasticity and wrinkles. Two of its three authors worked in the peptide division of Nitta Gelatin, a collagen manufacturer. That does not make the result false. It does make it exactly the kind of result that vanishes when the independent trials are analysed separately
“It just gets digested” is too simple. The standard sceptic’s line — collagen is broken down into amino acids like any protein — is not quite right. After people drank gelatin hydrolysates, collagen-derived peptides, mainly Pro-Hyp, appeared in their blood, peaking one to two hours later. Something does get through. But showing a peptide in plasma is not the same as showing visibly better skin, and that second step is where the independent evidence runs out

Marketing claim vs. evidence

What the box says

"Reduces wrinkles." "Restores elasticity." "Replenishes the collagen your skin loses with age." The implication is that the collagen you drink goes back where it came from, and that the effect is established.

What the research shows

The benefit appears in industry-funded and lower-quality trials, not in independent or high-quality ones. Most trials are at high risk of bias. Peptides do reach the blood, but no independent trial has shown they rebuild skin. And in the EU, collagen has no authorised skin claim at all — the one on the label belongs to vitamin C.

Frequently asked questions

+

Do collagen supplements actually work for skin?

The honest answer is that it depends on which trials you trust. Pooled together, trials show modest improvements in hydration and elasticity. But when a 2025 meta-analysis looked only at trials without pharmaceutical funding, or only at high-quality trials, the effect disappeared.

+

Does collagen reduce wrinkles?

Some trials say yes — a 2026 review found improvements in wrinkle depth or volume in 9 of 10 trials that measured them. The same review rated most of those trials at high risk of bias, and the independent-funding analysis found no effect on wrinkles.

+

Does collagen just get digested in the stomach?

Mostly, but not entirely. After people drank collagen hydrolysates, small collagen peptides — mainly one called Pro-Hyp — were detected in their blood within one to two hours. What remains unproven is that these peptides produce a visible difference in skin.

+

Is collagen a waste of money?

On current independent evidence, the case for paying for it is weak. The trials that found benefits were largely industry-funded or lower quality. If you are buying it for skin, you are paying for a result that has not held up when manufacturers were out of the picture.

+

Why does my collagen supplement contain vitamin C?

Partly because of regulation. In the EU, collagen has no authorised health claim, while vitamin C does: it “contributes to normal collagen formation for the normal function of skin.” Adding vitamin C allows a label to mention collagen and skin in the same legally approved sentence.

+

Is taking collagen safe?

The trials suggest so. A 2026 systematic review found hydrolysed collagen generally well tolerated, with a low incidence of adverse events across 25 randomised trials. The question is not whether it is harmful but whether it is worth paying for.

Our Conclusion

Collagen is not homeopathy. There is a plausible mechanism, peptides that measurably reach the bloodstream, and dozens of trials. But the trials that find a benefit are, disproportionately, the ones paid for by the people selling it, and the benefit does not survive when the analysis is limited to independent or high-quality studies. That is what weak evidence looks like: not nothing, but nothing that holds up without the manufacturer in the room. If you are considering it, the question to ask is not “does collagen work?” but “who ran the trial on the label?” It is probably harmless, and it is certainly not free. For comparison, the best-known independent trial of skin ageing — 903 adults, funded by the Australian government, not a manufacturer — found 24% less skin ageing over four and a half years with daily sunscreen. The supplement tested alongside it, β-carotene, had no overall effect.

Sources

Every claim in the text is linked to a primary source. Links go to the publisher's site, to EUR-Lex, or to PubMed, the US National Library of Medicine database.

Share:

This article is for informational purposes only and does not constitute medical advice. Trick or Treatment analyses the presence of clinical studies in open scientific databases — PubMed and Cochrane Library. The absence of studies in these databases does not automatically mean a drug is ineffective, but it does mean its effectiveness has not been confirmed by evidence-based medicine standards. Any treatment decisions should be made together with your doctor.