
Roasted Red Pepper Soup
This roasted red pepper soup combines tomatoes, parsnips and basil for a warming, vegetable-rich lunch that’s ideal for colder ...
0 item(s)
Free delivery on tests
Key takeaways
Collagen is one of the most popular supplements in the UK, sold for smoother skin, easier joints and a healthier gut. But does collagen actually work? In this guide, I look at what collagen is, what happens when you swallow it, and what the research shows for each claim. I cover the studies that support collagen and the ones that question it, so you can decide for yourself.
In this guide: What is collagen? · Does it work? · Skin · Joints, bones and muscle · Gut · Safety · How to take it · FAQ
Collagen is the most common protein in your body [1]. It gives structure and strength to your skin, tendons, ligaments, bones and blood vessels [1]. It is also part of the connective tissue in your gut wall.
Your body makes its own collagen. To do this, it needs amino acids such as glycine and proline, plus vitamin C [11]. Production slowly falls as you get older. With age, you have fewer of the skin cells that make collagen. The enzymes that finish building it also slow down [12].
Scientists have found at least 28 types of collagen. Type I makes up more than 90% of the collagen in your body [1]. The three types you will see on supplement labels are:
| Type | Where it is found | Where you will see it |
|---|---|---|
| Type I | Skin, bone, tendons, ligaments | Most hydrolysed collagen powders |
| Type II | Cartilage in your joints | Joint supplements, often from chicken |
| Type III | Skin, blood vessels and organs | Often sold alongside type I |
Some joint products contain “undenatured” type II collagen. This form has not been broken down, so it is a different product from collagen peptides.
Most collagen supplements are made from the skin, bones or scales of cows, pigs, chickens or fish. They are usually “hydrolysed”. This means the protein has been broken into smaller pieces, called peptides, so it dissolves easily in drinks. Gelatine is a less broken-down form of collagen, which is why it sets as it cools.
| Source | Made from | Things to know |
|---|---|---|
| Bovine | Cow hides and bones | Not suitable if you avoid beef |
| Porcine | Pig skin and bones | Not suitable for halal or kosher diets |
| Marine | Fish skin and scales | Avoid if you have a fish allergy; suits pescatarians |
| Chicken | Chicken cartilage | Often used for type II joint products |
No good trials have compared one source against another. Choose based on your diet, any allergies and whether the brand tests its products.
I made a video on this story. You can watch it below, then read on for the wider evidence.
Collagen supplements show some benefits in research, but the evidence is weaker than the marketing suggests. The best human data are for joints, bones and muscle, and even those trials are small and mostly industry-funded. The skin evidence is hotly disputed. For gut health, hair and nails, good human trials are still missing. What I want to add early on is that the vasy majority of people, including me, that supplement collagen notice visible improvement to their hair, skin, and nails (and at the right dose their bloating too).
| Claim | What the research shows | Strength of evidence |
|---|---|---|
| Skin hydration and elasticity | Positive in pooled trials; disputed by independent re-analyses | Contested |
| Wrinkles | Not robust in the newest reviews | Weak |
| Joint pain and stiffness | Modest benefit in placebo-controlled trials | Moderate to low |
| Bone density | One 12-month trial of one product | Low |
| Muscle (with strength training) | One 12-week trial | Low |
| Gut lining | Lab studies; one small study with no placebo | Very low |
| Nails | One small study with no placebo | Very low |
| Hair | No good trials of collagen alone | None |
Mostly, no. You digest collagen like any other protein, into single amino acids and small peptides. Whole collagen does not travel through your blood to rebuild your skin or gut lining.
That said, “just amino acids” is not the full story. In one study, volunteers took a drink of hydrolysed gelatine after a 12-hour fast. Small collagen fragments, mainly one called prolyl-hydroxyproline (Pro-Hyp), rose in their blood. Levels peaked one to two hours later and halved by four hours [2]. Some researchers think these fragments may act as signals to the cells that make collagen. This idea comes mainly from lab work. Good human evidence that collagen fragments reach the skin is still lacking [4].
Think of it like a brick wall. Eating collagen does not deliver a wall. It delivers bricks, plus a few small signals. Your body then decides where to use them.
Collagen is also an incomplete protein. It contains no tryptophan, one of the essential amino acids [10]. It is fine as an extra, but it should not replace other protein foods such as fish, eggs, dairy, pulses or meat.
The honest answer is that the skin evidence is contested. Several reviews find small gains in skin hydration and elasticity. Others find these gains disappear when you look only at the best-run or independent trials, or when you correct errors in the data.
Two earlier reviews pooled the trials and found a benefit. A 2021 review of 19 trials with 1,125 people found that hydrolysed collagen improved skin hydration, elasticity and wrinkles compared with placebo [13]. A 2023 review of 26 trials with 1,721 people also found better hydration and elasticity [12].
A 2025 review looked at 23 trials with 1,474 people [3]. Pooled together, collagen still looked helpful. But trials without industry funding showed no effect on hydration, elasticity or wrinkles. Higher-quality trials showed no effect either. The authors concluded there was no clinical evidence to support collagen for skin ageing [3].
Collagen industry groups pushed back on the 2025 review. They said some trials were wrongly labelled as independent, some data were entered wrongly, and pure collagen was mixed with multi-ingredient blends [14]. These points come from groups with a commercial interest, but they are worth knowing.
In 2026, a biostatistician and a UK dermatologist re-checked the 2021 and 2023 reviews, the ones the industry relies on [4]. They found errors in both, including mixed-up statistics. In one review, a result that was not significant was reported as significant. Once the data were corrected, there was no link between the collagen dose and the size of the skin benefit [4].
They raised other problems too [4]:
They concluded that collagen should not yet be promoted as a proven anti-ageing treatment [4].
Yes. Two 2026 reviews still found a benefit for hydration and elasticity, but not for wrinkles.
These reviews pool many of the same trials that the critics questioned. Pooling more of the same trials cannot fix problems inside them. The umbrella review was also published before the 2026 re-analysis.
| Review | What it found | Main limitation |
|---|---|---|
| de Miranda et al., 2021 [13] | Better hydration, elasticity and wrinkles | Data errors found in 2026 re-analysis |
| Pu et al., 2023 [12] | Better hydration and elasticity | Data errors found in 2026 re-analysis |
| Myung & Park, 2025 [3] | No benefit in independent or high-quality trials | Industry says it mislabelled some trials |
| Ravindran et al., 2026 [15] | Better elasticity and hydration; no change in roughness | Pools earlier reviews; published before the re-analysis |
| Batool et al., 2026 [16] | Better hydration, elasticity and barrier; wrinkles not robust | Built on the same short trials |
| Grimes & Mahto, 2026 [4] | No dose–response once errors were fixed | A re-analysis, not new trial data |
Collagen may give a small boost to skin hydration and elasticity for some people. But we cannot say that with confidence yet, and the case for fewer wrinkles is weak. Sun protection, not smoking and a protein-rich diet remain the foundations for healthy skin as you age.
This is where collagen has its most consistent results. Several placebo-controlled trials show modest benefits. But the trials are small, and most were funded or supplied by collagen makers.
Collagen may modestly ease osteoarthritis symptoms. A 2019 review pooled placebo-controlled trials in people with osteoarthritis [5]. Collagen improved overall symptom scores, stiffness and pain on a simple pain scale. However, the separate pain and physical function scores within the main symptom questionnaire did not differ from placebo. The review also mixed hydrolysed collagen with undenatured type II collagen [5]. The 2026 umbrella review also found less pain and stiffness, with bigger effects from longer use [15].
One trial suggests it may. In a 24-week study, 147 college athletes with activity-related joint pain took 10 g of collagen hydrolysate or a placebo each day [17]. The collagen group reported less joint pain at rest, when walking, when standing and when lifting. The difference was clearer in people with knee pain. The trial was sponsored by a collagen maker, which also employed one of the authors [17].
One trial suggests it may. In a 12-month study, 131 postmenopausal women with lower bone density took 5 g of specific collagen peptides or a placebo daily [6]. Bone density in the spine and hip rose in the collagen group compared with placebo. Bone formation markers also moved in a helpful direction [6]. This is one trial of one branded product, and it needs repeating.
Collagen may add a little to strength training, but it is not clear it beats other protein. In a 12-week trial, 53 older men with age-related muscle loss did strength training three times a week [7]. Some also took 15 g of collagen peptides daily. They gained more lean mass and leg strength than those on placebo, and lost more fat [7]. The catch is that the placebo was silica, which contains no protein. So we cannot tell if collagen did better than any other protein would have. The 2026 umbrella review found modest gains in muscle mass, but no meaningful benefit for recovery or soreness after exercise [15].
| Area | Strongest evidence | Typical dose studied | Confidence |
|---|---|---|---|
| Osteoarthritis symptoms | Reviews of placebo-controlled trials [5] | Varied by trial | Moderate to low |
| Activity-related joint pain | One 24-week trial [17] | 10 g a day | Low |
| Bone density | One 12-month trial [6] | 5 g a day | Low |
| Muscle with strength training. | One 12-week trial [7] | 15 g a day | Low |
| Exercise recovery | Umbrella review [15] | Varied by trial | No meaningful benefit |
No supplement replaces the basics. Strength training, enough total protein, vitamin D and calcium remain the foundations for joints, bones and muscle.
The gut claims for collagen run well ahead of the science. The idea is reasonable, but good human trials have not been done yet.
The theory is simple. Collagen is rich in glycine and proline, amino acids your body uses to maintain connective tissue, including in the gut. That is why collagen and bone broth are often promoted to “seal” a leaky gut.
Check the label before you buy. Many flavoured collagen powders contain sweeteners such as sorbitol or added fibres such as inulin. These are fermentable and can trigger bloating, especially if you have IBS or follow a low-FODMAP diet. NICE advises people with IBS and diarrhoea to avoid sorbitol [19]. If your gut is sensitive, choose an unflavoured product with no added fibre.
The evidence for nails is limited, and for hair it is largely missing.
In one study, 25 people with brittle nails took 2.5 g of a specific collagen peptide daily for 24 weeks [9]. Nail growth rose by 12% and broken nails fell by 42% [9]. However, there was no placebo group, it tested one manufacturer’s product, and two authors worked for a collagen research company. I could not find good-quality human trials of collagen alone for hair.
Some of the benefit people notice may come from extra protein, new habits or expectation, rather than from collagen itself. Many people tell me they notice better skin, hair or nails on collagen. That experience is real, and there are several possible reasons for it:
From clinical practice
“In clinic, I wonder whether supporting the gut lining might partly explain some of the skin changes people describe on collagen. It is an idea worth testing, but it has not been tested in trials yet.” — Alex Manos, MSc, Registered Nutritional Therapist
This is a clinical observation, not research evidence.
Collagen is generally well tolerated, and reviews of the skin trials describe hydrolysed collagen as safe [12]. Some people notice mild fullness, bloating or a lingering taste.
Take extra care in these situations:
The trials in this article used 2.5–20 g a day, and most collagen trials lasted 8–12 weeks [4]. If you decide to try collagen, give it a fair test:
Yes. Slow-cooked meat on the bone, skin-on fish and bone broth all contain collagen, along with other nutrients. Bone broth is not a reliable dose, though. One analysis found that bone broth had fewer collagen building blocks than research supplements. The amount also varied a lot from batch to batch [20]. Vitamin C from fruit and vegetables helps your body make its own collagen [11].
If you are reaching for collagen because your gut does not feel right, it may help to find out what is going on first. Collagen will not tell you why you are bloated or uncomfortable.
Our Ultimate Gut Plan is a stool test that looks at your gut bacteria, digestion and gut lining markers. If bloating is your main symptom, especially soon after eating, our SIBO Test uses a breath test you can do at home. Neither test measures collagen. They can help you and a practitioner understand what may be driving your symptoms. Then you can choose your next steps with more confidence.
They show modest benefits in some trials, mainly for joint symptoms, bone density, and muscle when combined with strength training [5][6][7]. The skin evidence is disputed [3][4]. Gut, hair and nail claims have very little human evidence behind them.
Type I is the most common and is found in skin, bone and tendons. Type II is found in joint cartilage. Type III is found in skin and blood vessels. Type I makes up more than 90% of the collagen in your body [1].
Most trials lasted 8 to 12 weeks [4]. Longer use has been linked to bigger effects on skin and joints [15]. Give any product at least 8–12 weeks before you judge it.
There is no good head-to-head evidence that one source works better. Choose based on your diet, any allergies, and whether the brand tests its products.
Trials have used 2.5 g to 20 g a day, depending on the goal. Higher doses have not clearly given bigger skin benefits [4].
It may help a little. Reviews of placebo-controlled trials found less pain and stiffness in osteoarthritis [5][15]. One trial in athletes found less activity-related joint pain [17]. Most trials were funded by collagen makers.
Lab studies suggest collagen peptides may protect gut cells [18]. But there are no good human trials showing collagen improves gut barrier function.
Not reliably. Bone broth contains collagen and gelatine. But one analysis found it had fewer collagen building blocks than research supplements, and the amount varied a lot [20].
Some people notice mild fullness or bloating. Added sweeteners such as sorbitol, or fibres such as inulin, in flavoured products are common culprits [19].
No. Collagen has no tryptophan, so it is an incomplete protein [10]. Keep eating a mix of other protein foods.
See your GP promptly if you notice any of the following:
These need checking, and no supplement should delay that. If you have kidney disease or take regular medication, speak to your GP before starting collagen.
Disclaimer: This article is for education only. It is not intended to diagnose, treat, cure or prevent any disease. Speak to your GP before starting a supplement if you have a medical condition or take regular medication.