The Collagen Paradox: Good Science, Inconvenient Paperwork

Why a wall of clinical data still can't earn collagen a single approved word in the UK — and why that's not quite the scandal it sounds.*

---

Walk into any pharmacy and you'll find a shelf of collagen promising firmer skin, stronger nails and the broad reversal of time. Read those same products' UK websites closely and you'll notice something strange: legally, almost none of them are allowed to promise any of it. Not because collagen got caught lying — but because of a quiet, gloriously pedantic decision made over a decade ago in a building in Parma.

This is the story of how an ingredient with a real research record became officially unmentionable.

First, a word on Parma

Fair question: of all the places to headquarter Europe's food-safety brain, why a mid-sized city in northern Italy?

The authority was set up in 2002, in the wake of the mad-cow panic, to rebuild public trust in food. It needed a permanent home, and two cities went to war over it — Parma and Helsinki. The contest got memorably petty. Italy's then–prime minister championed Parma's bid and, at an EU summit, dismissed Finland's chances on the grounds that the Finns "don't even know what prosciutto is." Finland's largest newspaper hit back with a full-page advert reading "Prosciutto is ham" — and, underneath, "Now 1.2 million Finns know this."

Parma won. Which means the body that decides whether your collagen may breathe a single word about your skin is headquartered in the city that gave the world Parma ham and Parmesan — a place defined entirely by food, now in charge of telling everyone what they can't say about it. Keep that image handy. It helps.

What the studies actually found

Collagen hasn't been hiding in the shadows of pseudoscience. It has been put through proper human trials — randomised, placebo-controlled, the works. Across a number of them, participants taking marine collagen peptides showed measurable improvements in skin elasticity and reductions in the depth of wrinkles. There's a substantial body of work on joints, too, particularly in athletes and in osteoarthritis. Whatever you make of the supplement aisle's wilder promises, the underlying data was not fringe.

So if the research exists, why can't anyone say so?

The list nobody mentions

Because in the UK — and in the EU before Brexit — you cannot make a health claim about a food or a supplement unless that exact claim already sits on an official register. It's a permitted list. A scientific body assesses the evidence, regulators authorise the wording, and only then may a brand put it in front of you. No entry, no claim. And here is the part that catches everyone out: if a claim isn't on the list, no amount of evidence outside it counts. You could hold the finest clinical trial ever conducted. Off-list, it's legally invisible.

Collagen, for all its studies, never made it onto the list. The question is why.

The moment it went sideways

The pivotal case landed in 2013. A German ingredients firm, Gelita, applied to authorise a skin claim for its collagen peptide, VeriSol, and submitted human trial data to back it. This was the proper route — a formal application, real science, the front door rather than a marketing loophole.

Here's where it gets almost comic.

The regulator's panel didn't call the studies rubbish. It accepted that the product was properly characterised. It didn't even dispute that the elasticity results were real. Instead, it explained — with the unblinking seriousness of people who have never knowingly rounded a number — that "skin looking better" and "skin working better" are, for regulatory purposes, two entirely different things. And only the second one qualifies as a health claim.

Wrinkle reduction and improved elasticity, the panel ruled, are matters of appearance, not skin function. The single trial that did test an actual function of the skin — its water-barrier — showed no effect on it. And so the conclusion was delivered: a cause-and-effect relationship had not been established. Claim refused.

Read that back slowly. The science showed an effect. The regulator agreed it showed an effect. The effect was then filed under the wrong heading and shown the door.

And the joints? Same wall, different brick

The skin saga wasn't the first time this company knocked. Two years earlier, in 2011, the very same firm applied to authorise a "maintenance of joint health" claim for collagen hydrolysate — and here the story takes an interesting turn.

This time, the regulator didn't quibble over categories at all. It accepted, plainly, that maintaining your joints is a genuine, claim-worthy physiological benefit. No appearance-versus-function hair-splitting. The door was open.

The claim failed on the evidence instead. The single headline trial put forward — 147 active student athletes, fifteen measures of joint discomfort — didn't reach statistical significance against placebo once the numbers were adjusted, and the panel ruled the dossier insufficient. Cause and effect: not established.

But notice the date. That verdict was reached on one company's file, in 2011, and it has sat frozen ever since. The science didn't freeze with it. In the years since, collagen and joints has become one of the busier corners of the field: a 2024 systematic review pooling eleven randomised trials and some 870 people with knee osteoarthritis reported improvements in pain and function that were both statistically significant and clinically meaningful. That isn't proof to EFSA's exacting standard — it was never re-submitted as a fresh dossier, and the authors themselves press for larger, independent trials — but it is a very long way from "does nothing."

Which gets us to the single most important distinction in this whole story: the register says no authorised claim. It does not say no effect. Those are different sentences in different languages, and almost everyone — brands and sceptics alike — reads the first as if it were the second. Collagen has now been turned away twice, by two different doormen, for two different reasons — wrong category for skin, one thin dossier for joints. What it has never received is a verdict that it doesn't work. It simply hasn't been waved onto an official list. The gap between those two things is exactly where this article lives.

Falling between two stools

This isn't malice; it's architecture. The UK runs two separate regimes that don't quite meet in the middle. On one side, cosmetics — where claims about appearance live. On the other, food and supplement law — which governs anything you swallow, and which only recognises claims tied to a genuine physiological function.

A collagen drink falls into the gap. Because you ingest it, it's judged under food law. But the benefit people actually care about — how their skin looks — is, in the eyes of food law, a cosmetic matter that doesn't qualify. The thing collagen was shown to do is precisely the thing the relevant rulebook isn't built to credit. Every formal application to authorise a skin-maintenance claim through this route has met the same fate, mostly for measuring the "wrong" outcomes.

It's less a verdict that collagen does nothing, and more a bureaucratic category error with a decade of staying power.

And hyaluronic acid? The same gap — and a myth worth retiring

Hyaluronic acid sits in exactly the same position: no authorised health claim in the UK or EU, for skin, joints or anything else you swallow. By the letter of the register, another silent ingredient.

But "silent" and "inert" are different words. Collagen and hyaluronic acid aren't two unrelated trend ingredients sharing a tub by coincidence — they're the two principal components of the same extracellular matrix. Collagen is the structural scaffold; hyaluronic acid is the molecule that binds water into it. A structure built on one without the other would be telling half the story.

There's also a stubborn old myth worth retiring: that HA does nothing unless a clinician injects it. Injected HA — into an arthritic knee, as a licensed medical device — is real medicine and an entirely separate regulatory world. But the assumption that the oral route is useless has been overtaken by the research. Ingested hyaluronic acid appears to be broken into smaller fragments, absorbed, and distributed to tissue; randomised trials and a recent pooled analysis report measurable improvements in skin hydration and elasticity, and separate, higher-dose trials report eased knee-osteoarthritis symptoms. Promising — not yet authorised, held back by the same evidentiary bar collagen keeps meeting. Once again: the science is moving faster than the filing cabinet.

Then Brexit pressed pause

When the UK left the EU, the GB register simply inherited the EU's list of authorised claims and froze it in place. So even in the unlikely event that European regulators warmed to collagen tomorrow, Britain wouldn't automatically follow. The filing cabinet was copied, locked, and the key put somewhere sensible.

So what can you actually trust?

Here's the part the dry people in Parma will happily sign off on — because it cleared their gate fair and square.

Vitamin C carries an authorised claim: it contributes to normal collagen formation for the normal function of skin, cartilage and bone. That's not marketing varnish; it's on the register, assessed and approved, which is more than the headline ingredient in most collagen products can say. Note the word cartilage sitting in there — that's the one legitimate, on-the-register way to gesture at joints, and it rides on the vitamin C, never on the collagen. It also contributes to protecting cells from oxidative stress and to reducing tiredness and fatigue. Several B-vitamins earn their place too — niacin and biotin, for instance, contribute to the maintenance of normal skin.

None of it is glamorous. All of it is proven. There's a quiet kind of confidence in only saying the things you can stand behind.

The honest version is the interesting one

The temptation, in this category, is to pretend the register doesn't exist and promise miracles anyway. The opposite temptation is to throw up your hands and say nothing. Neither is the truth.

The truth is more interesting than either: the science on collagen is genuinely promising, the regulator is genuinely strict, and the gap between the two is exactly where the marketing usually lies — in both senses of the word. We'd rather just show you where the line is, tell you what's actually on the register, and let you decide what to do with the rest.

Structure, stated honestly. It was never going to be a miracle. It was always going to be the truth, told well.

---

Sources & further reading

• 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 (Article 13(5), Regulation (EC) No 1924/2006). EFSA Journal 2013;11(6):3257. doi:10.2903/j.efsa.2013.3257
• EFSA NDA Panel (2011). Scientific Opinion on the substantiation of a health claim related to collagen hydrolysate and maintenance of joints (Article 13(5), Regulation (EC) No 1924/2006). EFSA Journal 2011;9(7):2291. doi:10.2903/j.efsa.2011.2291
• EFSA NDA Panel (2012). Guidance on the scientific requirements for health claims related to bone, joints, skin and oral health. EFSA Journal 2012;10(5):2702.
• Regulation (EC) No 1924/2006 on nutrition and health claims made on foods (retained in GB law).
• GB Nutrition and Health Claims Register — authorised claims for vitamin C and the B-vitamins. To date, no health claims are authorised for collagen, collagen peptides or hyaluronic acid.

On the evolving evidence base (not regulatory-grade, included for context):

• Simental-Mendía M, et al. (2025). Effect of collagen supplementation on knee osteoarthritis: an updated systematic review and meta-analysis of randomised controlled trials. Clin Exp Rheumatol 43(1):126–134.
• de Miranda RB, et al. (2021). Effects of hydrolyzed collagen supplementation on skin aging: a systematic review and meta-analysis. International Journal of Dermatology.
• Oral sodium hyaluronate, randomised double-blind placebo-controlled trial (2025), Scientific Reports — skin hydration and barrier function.
• Reviews of oral hyaluronic acid for knee osteoarthritis symptom relief (various RCTs, ~80–240 mg/day).

This article discusses the history of collagen research and food-claims regulation. It is not a health claim for any product, and nothing here is intended as medical or dietary advice.