Collagen Amounts in Studies: Reference Table by Type, Source and Outcome

A reference table of the collagen amounts used in published studies, by product type, length, group studied and outcome. Study amounts only, never advice.

Content last updated October 23, 2026. This page has not been medically reviewed.

What this table is, and what it is not

This page lists the collagen amounts that researchers used in a handful of published trials, next to who was studied, for how long and what was measured. It is a reading aid, not a dosing guide. Nothing here tells you how much to take. No health body in the sources we read gives an official daily amount (RDA, AI or upper limit) for collagen supplements, so there is no official figure to compare these study amounts against. A clinician or pharmacist can help with any personal decision.

We chose these trials to show how different the amounts, products and outcomes are, not to rank them or to say which "works". For what the evidence says by use, see Collagen for Skin, Joints and Bone. For what the product types are, see Collagen Types and Sources. For how studies and labels compare, see How Much Collagen Do Studies and Labels Use?.

Study amounts by product type, group and outcome

Read each row on its own. A result in one product type does not carry over to another.

Study (year)	Product type, as described	Amount in the study	Length	Who was studied	Compared with	What was measured
Proksch et al. (2014)	Collagen hydrolysate ("specific collagen peptides")	2.5 g or 5.0 g once daily [1]	8 weeks [1]	69 women aged 35 to 55 [1]	Placebo [1]	Skin elasticity, an instrument reading. Both amounts were statistically better than placebo [1]
Clark et al. (2008)	Collagen hydrolysate in a liquid drink	10 g in 25 mL [2]	24 weeks [2]	147 randomized athletes with activity-related joint pain; 97 could be analyzed [2]	Placebo drink [2]	Pain, mobility and inflammation on a visual analogue scale (a self-rating line) [2]
Zdzieblik et al. (2015)	Collagen peptides	15 g a day [3]	12 weeks, alongside a resistance training program [3]	53 older men with sarcopenia (age-related muscle loss) [3]	Silica as placebo [3]	Muscle mass and strength. Both groups improved; the collagen group improved more [3]
König et al. (2018)	Collagen peptides ("specific collagen peptides")	5 g a day [4]	12 months [4]	131 postmenopausal women with reduced bone density were enrolled [4]	Placebo [4]	Bone mineral density and bone markers. Spine T-score change was +0.1 vs -0.03 [4]
Lugo et al. (2016)	Undenatured type II collagen (UC-II) from chicken sternum cartilage [5]	40 mg a day (0.04 g, about one thousand times less than the gram amounts above) [5]	180 days [5]	191 volunteers with knee osteoarthritis [5]	Placebo, and a glucosamine plus chondroitin group [5]	Knee symptom score (WOMAC) [5]
Shaw et al. (2017)	Vitamin C-enriched gelatin, a different product from peptides	5 g or 15 g, in separate sessions [6]	6 minutes of rope skipping, repeated over 3 days [6]	8 healthy men in a crossover design [6]	Placebo [6]	A blood marker of collagen making. It doubled after 15 g, taken before exercise [6]

How to read the columns

Product type. Hydrolyzed collagen peptides, gelatin and UC-II are different products, made differently and used in different amounts. See Collagen Peptides vs Gelatin and Gelatin and Undenatured Type II Collagen.
Amount. Grams (g) and milligrams (mg) are a thousandfold apart. The UC-II trial used milligrams; the others used grams.
Outcome. An instrument reading (skin elasticity, a blood marker, a bone scan) is not the same as feeling or functioning better. Some rows measure one, some the other.
Length and group. A trial in postmenopausal women, athletes or older men tells you about those people for that length of time.

What summaries of many trials say about amounts

Reviews pool trials that used different amounts, so they do not give one figure.

Review	What it covered	Amount information in the abstract
de Miranda et al. (2021), skin	19 trials, 1,125 participants aged 20 to 70, 95% women [7]	Not stated. The authors concluded about 90 days of hydrolyzed collagen reduced skin aging signs [7]
Pu et al. (2023), skin	26 randomized trials, 1,721 participants [8]	Not stated in the abstract we read
Liang et al. (2024), osteoarthritis	35 trials, 3,165 participants; collagen derivatives did not raise the risk of withdrawal or adverse events versus control [9]	Not stated in the abstract we read

Proper evidence not found (we searched PubMed for collagen dose-response and head-to-head amount comparisons, and read the abstracts above): a trial that compared several amounts of the same collagen product across the same outcomes. The one comparison we read, Proksch, tested 2.5 g against 5.0 g for skin elasticity, and both were reported better than placebo. [1]

Frequently asked questions

Is there an official recommended amount of collagen?

Not in the sources we read. We did not find an RDA, adequate intake or upper limit for collagen supplements from a health body, and the trial amounts in the table are not official figures. [12]

Why do the amounts in the table range from milligrams to grams?

The products differ. UC-II was studied at 40 mg a day, while hydrolyzed peptides and gelatin were studied at several grams. [5][3][6] The numbers cannot be compared directly.

Does a bigger amount do more?

Proper evidence not found. The one amount comparison we read tested 2.5 g against 5.0 g for skin elasticity and reported both better than placebo. [1]

Can I use a study amount as my own plan?

No. Study amounts apply to the people, product and length studied. Ask a clinician or pharmacist about any personal decision, especially with medicines, allergies, pregnancy, kidney or liver disease or planned surgery. [12]

Do these results mean collagen works?

Not by themselves. Some rows measure instrument readings or blood markers rather than how people feel or function, and some trials were small. For the evidence by use, see Collagen for Skin, Joints and Bone.

More details: evidence, limits and cautions

The details behind this page: cautions, limits of the evidence and where sources disagree. Open any note to read it.

Funding and size, stated plainly

The Shaw trial had only eight participants, so it is a small experiment on a blood marker, not a test of injury prevention. [6]
In the Lugo trial, two of the three authors list a supplement company as their workplace. [5]
One author of the König trial lists a collagen research company as an affiliation. [4]
In the Clark trial, data from 97 of the 147 people randomized could be analyzed. [2]

We did not read funding statements for every trial. Where a row shows no funding note, that means "not described in the sources we read", not that there was none.

Safety and who should talk to a clinician first

Tolerability in these trials. Proksch reported no side effects in its study. [1] The osteoarthritis review found no higher risk of adverse events than control. [9] Short trials do not rule out rare problems.
Fish and shellfish allergy. Fish collagen can cause serious allergic reactions. A case report describes a child with fish allergy who had anaphylaxis after fruit-flavored gummies made with fish collagen. [10] If lips, mouth or throat swell, or breathing is difficult, call 911 (US) at once. See Collagen Allergy.
Kidney stones. Collagen contains hydroxyproline, which the body can turn into oxalate. Proper evidence not found in humans: our PubMed search found only a rat study, where rats given a large amount of hydroxyproline had high oxalate in the urine. [11] We do not transfer a rat result to people, but anyone with a stone history should ask a clinician. See Collagen, Hydroxyproline, Oxalate and Kidney Stone Questions.
Medicines, pregnancy, children, kidney or liver disease, surgery. Supplements may interact with medicines or carry risks with certain medical problems or surgery, and many have not been tested in pregnant women, nursing mothers or children. [12] See Medication Interactions, Pregnancy and Breastfeeding and Supplements Before Surgery. Talk to a clinician or pharmacist, and see Collagen and Medicines, Kidney or Liver Conditions.
Product quality. What is on the label may not be what is in the product. [12] See Quality Testing and Certifications.

References

Education, not medical advice. This page is for general information. It does not diagnose or treat any condition and is not a substitute for care from a qualified health professional. Natural does not automatically mean safe. Ask a clinician or pharmacist before changing your care, especially if you take medication, are pregnant or breastfeeding, or have a health condition.