Collagen peptides for skin and joints: what the studies show

A scoop of powder in your coffee. Tasteless, supposedly. The tub says something about skin, hair, nails and joints, next to a picture of a woman who looks like she has never had a bad night.

And somewhere between the third and the fourth month you ask yourself: am I actually noticing anything? Or do I only want to notice something, because the stuff cost 40 euros?

I like that question. It is more honest than most advertising promises.

And the answer is more interesting than either camp makes it sound. Collagen peptides are neither magic nor expensive nonsense. They are a research field with a fascinating mechanism, a mediocre body of evidence and a large funding shadow. All three belong on the table.

The thinking error that almost every collagen ad uses

Imagine you want to build a brick house and a delivery arrives. Only the delivery gets ground to dust at the property line. What reaches you is sand and powder.

That is roughly how your digestive tract treats proteins. It is not a doorman passing parcels through. It is a dismantling plant.

Collagen is a very large molecule. Three long chains, wound around each other like a rope. Stomach acid and enzymes from your pancreas take this rope apart into ever smaller pieces. What passes through the gut wall are single amino acids and very short fragments of two or three building blocks. An intact collagen molecule is not among them.

That is why the sentence “eat collagen so collagen appears” is both so seductive and so imprecise. It only holds if you read it very generously: yes, the building blocks arrive. No, they do not travel with an address label towards your face.

Hydrolysed collagen changes little about this basic principle. It simply saves your digestion some work. The collagen has already been pre-cut with enzymes, which is why the powder dissolves in cold water while gelatine sets into a jelly. Collagen hydrolysate, collagen peptides and hydrolysed collagen mean the same thing.

And now you know why the interesting question is not whether collagen arrives, but what part of it arrives.

The fragments that really do make it into the blood

This is where it gets interesting. Because the dismantling plant does not work completely.

Collagen contains a peculiarity that hardly any other dietary protein has: hydroxyproline. It is a chemically modified form of the amino acid proline. And fragments containing hydroxyproline appear to be more resistant to digestive enzymes than others.

The best known of these fragments is called prolyl-hydroxyproline, Pro-Hyp for short. It consists of exactly two amino acids and is measurable in the blood after eating.

Why is this more than a laboratory curiosity? Because the research has built a hypothesis from it that I find rather elegant.

The idea goes like this: these fragments are not building material, they are a message. If collagen fragments suddenly appear in your blood, your body might read that as a sign that connective tissue is being broken down somewhere. And respond with increased new formation.

Put differently: it is not the brick that gets delivered, it is the fax to the building firm.

By now a triple building block is being discussed as well, Gly-Pro-Hyp. A Korean research group compared different collagen preparations in the blood of volunteers and found that one particular hydrolysate mainly raised this tripeptide. In cell culture, Gly-Pro-Hyp protected human skin fibroblasts from switching down certain matrix genes under oxidative stress.

Here too I stay careful. This work comes from the research centre of a cosmetics company, together with a peptide manufacturer. That does not devalue the data. It does change how I weigh it.

And now you know why I speak of an interesting hypothesis here and not of an established principle of action.

The skin studies: mostly positive, methodically thin

Now to the numbers most people buy this powder for.

And here I have to say two things at once that seem to contradict each other. First: the meta-analyses mostly point in a positive direction. Second: the studies behind them are mostly small, short and inconsistent.

This is exactly where the field got restless. A group of Brazilian dermatologists wrote a letter to the same journal in which one of the meta-analyses had appeared. Their objection concerned the heterogeneity of the included studies. In plain language: investigations were added up that differ considerably in preparation, dose, duration, measurement method and target group.

That is a fair objection. If you add up apples, pears and quinces, you get a large number, but no fruit.

One of the methodically cleaner single studies came from Kiel and is still among the most cited.

And now you know why I find the decimal places less interesting here than the question of how the studies were built.

The funding shadow you have to read along with

May I ask you an uncomfortable question? Who actually pays for the studies on a product that is sold exclusively as a food supplement?

The answer is rarely the ministry of health.

With collagen peptides it is striking how often the same names appear: research institutes owned by manufacturers, peptide departments of gelatine companies, research centres of cosmetics groups. That is not a scandal, it is normality in this market segment. But it belongs in the picture.

What I do with that in practice: with every collagen study I look at the section on conflicts of interest. I check whether the control group received a sensible placebo. And I ask whether the endpoints were defined in advance or only became interesting afterwards.

And now you know why I will not give you a percentage to the decimal point on this topic.

Joints, bones, muscles: the noisier chapter

Maybe you know this: the knee speaks up on the stairs. Not dramatically. Just persistently. And at some point you reach for whatever is on the shelf.

With joints the evidence gets harder than with skin. Not because there are no studies, but because they diverge more strongly.

An Iranian research group reached a clearly more reserved verdict in a systematic review of 19 papers on osteoarthritis and rheumatoid arthritis. It described that several of the included studies reported low effectiveness compared with usual treatments and in part side effects, and that the quality of most studies was weak. Their conclusion: a final assessment is not possible at present.

A third, more recent systematic review screened 4,246 papers and included 36 randomised studies on bones, muscles and joints. Its picture is nuanced: for joint complaints the studies mostly reported favourable results. For bone density the data were not sufficient for a clear statement. For muscle the results were inconsistent, with positive effects mainly when training was added. Here too, one author works for a collagen manufacturer.

Two single studies from Freiburg are often cited in this context and deserve a fair presentation.

And now you know why with joint complaints I talk about load, mobility and drivers of inflammation first, and only then about powder.

Vitamin C: the cofactor without which the structure does not hold

Now comes the part where the biochemistry is unambiguous for once. That is rare enough to be enjoyed.

For collagen to become stable, it is not enough to string the amino acids together. Certain proline and lysine building blocks have to be chemically modified, they get a hydroxyl group attached. Only this hydroxylation allows the three chains to twist into a stable triple helix.

The enzymes that carry out this step are called prolyl and lysyl hydroxylases. And they need vitamin C as a cofactor.

Picture a workshop where the screws get driven in. Vitamin C is not the screw. It is the cordless screwdriver. Without it the parts just lie around.

What happens when this step fails over a longer period is historically well documented: scurvy. Bleeding gums, poor wound healing, fragile vessels, joint pain. All consequences of connective tissue that does not get its twist lock.

There is even a small but much noticed study that tested exactly this combination.

And now you know why, with collagen, I ask about peppers, berries, cabbage and citrus fruit first.

Why homeostasis applies here too: glycine, glutathione and the whole network

There is one point in this topic that interests me as a physician more than any wrinkle measurement. And it appears in no advertisement.

Collagen is an extremely glycine rich protein. About every third amino acid in the chain is glycine. That is not a coincidence, it is a design principle: glycine is the smallest amino acid there is, and only because of that do the three chains fit so tightly into each other.

But glycine is not only building material. It is an active substance in the body in its own right.

• Glycine is an inhibitory neurotransmitter in the spinal cord and the brainstem.
• Glycine is one of the three building blocks of glutathione, the body's most important antioxidant.
• Glycine is involved in the formation of haem, the red pigment in blood, and in creatine synthesis.
• Glycine plays a role in the conjugation of bile acids and in detoxification steps in the liver.

So a person taking collagen powder automatically takes in a relevant amount of glycine. Whether that is good or irrelevant depends on where they stood before.

And here comes the point I make in every article of this guide, because it stands above everything else: your body regulates nutrients in loops. Whoever turns one screw moves others with it.

With collagen that is not an abstract warning, it is concrete. Collagen is an incomplete protein. It contains no tryptophan and little of several essential amino acids. Anyone who counts large amounts of collagen powder into their protein balance shifts their amino acid pattern. With an already tight protein supply that can backfire.

And now you know why I would rather talk about the whole network here than about a single powder.

Covering needs or therapy: why the dose shifts everything

There is a distinction I make in every article of this guide, because it clears up almost all misunderstandings.

On one side stands covering your needs. Those are the amounts you buy online or in the drugstore. They fill gaps. No therapeutic effect is to be expected from them, and that is not their job.

On the other side stands the orthomolecular or therapeutic high dose. Considerably higher amounts, used deliberately, limited in time, under medical supervision and with laboratory monitoring. Only here can a nutrient act like a medicine at all.

The classic example of this is not collagen, it is vitamin D. For covering needs, small over the counter amounts are enough. Which amount fits you depends on your measured level and belongs in a medical conversation, not in a guide article. In the so-called Coimbra protocol for multiple sclerosis, by contrast, doses many times higher than that are used, with close monitoring of calcium, parathyroid hormone and kidney values and with strict calcium restriction. I deliberately give no concrete figures here.

And now the point that often gets lost with collagen: for collagen peptides this second range simply does not exist.

The amounts in the studies lie between about 2.5 g and 15 g per day. Those are food amounts, not medicinal doses. A serving of collagen powder supplies no more protein than a larger piece of meat. So anyone hoping for a drug-like effect is looking in the wrong place.

That is exactly why I consider collagen a good example of honest supplementation: it is a food component that can fill a gap. No more, but also no less.

And that gap does indeed exist more often than it used to. Not because supplements are fashionable, but because our eating habits have shifted.

• We mostly eat muscle meat. Skin, cartilage, tendons, rind and bones cooked along with the meat have disappeared from many kitchens.
• That removes the part which traditionally supplied glycine and proline. The amino acid pattern can shift towards methionine as a result.
• Highly processed foods supply energy but little micronutrient density. Precisely the micronutrients that collagen formation needs fall short.
• Chronic stress, smoking, UV exposure and environmental burdens can raise the consumption of protective substances such as vitamin C, zinc and glutathione precursors.

Honesty also requires this: for the thesis of nutrient loss in soils and foods over decades there are comparative data, but their significance is methodologically disputed. Old and new analytical methods are hard to compare, and part of the effect is explained by higher yielding varieties, in which the nutrients are spread over more mass. I consider this a real but often overdrawn argument.

And now you know why I name real food from good sources first and the powder second.

Origin, allergies and where I get careful

Collagen powder is an animal product. Always. Vegan collagen does not exist, even if some products are named that way. What is sold there are building blocks and cofactors meant to support your body's own formation. That is a different approach and should also be named differently.

On the subject of allergy there really are data, and they matter more than most guides suggest.

On contaminants I have to be transparent: I did not find solid comparative data specifically on heavy metal content in collagen powders in the literature. So I phrase this as a precautionary principle and not as a study result.

The thought behind it is simple. Collagen is obtained from skin, bones and connective tissue, meaning from tissues in which some environmental burdens can accumulate. With marine sources there is the added fact that they come from side streams of fish processing. A review on the industrial use of fish cartilage describes exactly this raw material chain and discusses safety and sustainability questions. For me the practical consequence is this: I look for a statement of origin, for batch analyses on heavy metals and for a manufacturer who hands over both on request.

And now you know why I consider the question of origin more important here than the question of the brand.

Three levers you can put in place today

No protocol, no weekly plans, no product names. Just three things that lie in your hands and that stay sensible even if you decide against the powder.

And now you know why with collagen I neither advise against it nor push it, but ask about the aim.

Frequently asked questions about collagen peptides

How this topic connects with the others

Collagen never stands alone. If you have the feeling that one building block is not enough, these paths lead further.