Collagen benefits

Does collagen supplementation rejuvenate skin?

Skin is the most crowded area of collagen studies. Three meta-analyses ask the same question and look at sets of studies that are close to each other but not identical.

• de Miranda et al. (2021): 19 randomized, double-blind, controlled trials; 1,125 participants aged 20–70 (95% women). In the pooled analysis, hydrolyzed collagen was favorable compared with placebo for skin hydration, elasticity and wrinkles. The authors reported that 90 days of use was effective in reducing skin aging. ^[1]
• Pu et al. (2023): 26 RCTs, 1,721 participants; durations of 2–12 weeks. For skin hydration, the pooled effect size across 18 studies was 0.63 (95% CI 0.38–0.88), and elasticity was examined with 19 studies. Of the products, 14 were fish-derived, 1 bovine, 1 chicken and 2 porcine; in 9 studies the source was not stated. The hydration effect of chicken-derived collagen was weakest (−0.03; 95% CI −0.40 to 0.34). For hydration, the effect with use longer than 8 weeks (0.59) was larger than with shorter use (0.39). ^[2] The authors state that some studies had fewer than 40 participants and that differences in duration, form and content made comparison difficult; risk of bias was identified for missing outcome data in 13 studies, deviation from the intervention in 7 studies, selection of the reported result in 2 studies and the randomization process in 1 study. ^[2] The review was funded by a hospital grant; the authors reported no conflict of interest. ^[2]
• Myung and Park (2025): 23 RCTs, 1,474 participants. When all studies were taken together, there was significant improvement in hydration, elasticity and wrinkles. In the subgroup analysis by funding source, no effect was seen on any of the three outcomes in studies not funded by pharmaceutical companies, while a significant effect was found in funded studies. There was no significant effect in any area in high-quality studies; a significant improvement in elasticity was found in low-quality studies. The authors concluded that there is no clinical evidence supporting collagen supplements for preventing or treating skin aging. ^[3]

These three results are not contradictory, but they do not fit into a single sentence either. The pooled analyses are favorable; the separation by funding and quality shows in which studies the effect is concentrated. The relationship between funding source and result does not prove that a study is biased; it only says that the same effect was not seen in independently funded studies. The effects are at the level of “hydration and elasticity measurements”; how far they correspond to an improvement a person can notice in the mirror cannot be inferred from these analyses.

The assessment by the European Food Safety Authority (EFSA) of a “skin elasticity” claim for a specific collagen peptide product is covered in the main collagen guide.

Does collagen help osteoarthritis pain?

The 2019 meta-analysis by García-Coronado and colleagues examined oral collagen-derived supplements in osteoarthritis in placebo-controlled RCTs using WOMAC and VAS measures. The weighted mean difference was −8.00 for the total WOMAC score (95% CI −13.04 to −2.95) and −16.57 for VAS (95% CI −26.24 to −6.89). Among the WOMAC subscores, stiffness fell (−0.41; 95% CI −0.74 to −0.08); the pain subscore (−0.22; 95% CI −1.58 to 1.13) and the physical function subscore (−0.62; 95% CI −5.77 to 4.52) are not significant. ^[4] That is, while the total score and VAS fell, the WOMAC pain subscore did not; the conclusion “reduces pain” does not follow from this alone.

This meta-analysis evaluates collagen-based supplements together, not a single collagen product. Hydrolyzed collagen and undenatured type II collagen are examined at different amounts and with different assumed mechanisms; this difference is covered on the types page.

Does collagen help joint pain in athletes?

The 24-week study by Clark and colleagues was done in athletes with no sign of joint disease but with activity-related joint pain: 147 collegiate or club athletes (72 men, 75 women) received 10 g per day of collagen hydrolysate or a placebo containing xanthan. In the 97 people whose data could be analyzed, six measures were significant against placebo (for example, physician-assessed pain at rest −1.37 and −0.90; p = 0.025); the difference became more pronounced in the subgroup of 63 people with knee pain. ^[5]

EFSA examined this study within a manufacturer’s application: for none of the 15 outcome measures was there a significant difference when correction for multiple comparisons (Bonferroni-Holm) was applied; pain had decreased in all participants, and no imputation for missing data had been used in the statistical analysis. EFSA, also stating that animal and in vitro studies do not predict an effect in humans, concluded that a cause-and-effect relationship could not be established. ^[6] The same data look favorable in the uncorrected analysis and nonsignificant in the corrected analysis; the evidence grade on this page is therefore given as D (insufficient).

Does collagen increase bone density?

In the study by König and colleagues, 131 postmenopausal women with primary age-related bone density loss received 5 g per day of specific collagen peptides or placebo (maltodextrin); 102 completed the study, and the analysis was on an intention-to-treat (ITT) basis. Mean age was 64.3. At month 12, the spine T-score was +0.1 in the collagen group and −0.03 in the control (p = 0.030); the femoral neck T-score was +0.09 versus −0.01 (p = 0.003). P1NP increased in the collagen group; CTX-1 increased in the control group. ^[7] Part of the study was funded by a collagen manufacturer; the authors reported that planning, organization and data analysis were carried out only by the researchers and that there were no conflicts of interest. ^[7]

This is a single 12-month RCT, the differences are at the level of 0.1 in T-score, and it did not measure fracture risk. It is not evidence that collagen substitutes for other drug and nutritional measures for bone density.

Is collagen beneficial for hair and nails?

The data that can be reported for nails come from an open-label, single-center study without a control group: 25 participants took 2.5 g per day of specific collagen peptide for 24 weeks and were then followed for 4 weeks without treatment. A 12% increase in nail growth rate and a 42% decrease in the frequency of broken nails were reported; “overall clinical improvement” was assessed in 64% of participants. ^[9] Because there was no placebo or blinded assessment, these results cannot be attributed to collagen. The evidence reviewed for hair includes no controlled study.

Does collagen increase muscle mass?

In the RCT by Zdzieblik and colleagues, 53 sarcopenic men (mean age 72.2) performed guided resistance training three sessions per week for 12 weeks and received 15 g per day of collagen peptide or silica (placebo). After training, lean mass changed by +4.2 kg in the collagen group and +2.9 kg in the placebo group; quadriceps isokinetic strength by +16.5 Nm versus +7.3 Nm; and fat mass by −5.4 kg versus −3.5 kg (all p < 0.05). ^[8] The paper later received a correction: according to the 2025 correction, the manufacturer supplied product samples, paid participant fees and sample analysis, and one author is a co-inventor of a patent application made for the active ingredient (Br J Nutr 2025;134(5):440). ^[8]

In a smaller study looking at the same question in older women and comparing whey with collagen (22 healthy women, 30 g twice daily, 6 days), whey increased muscle protein synthesis more than collagen peptides; with collagen, long-term muscle protein synthesis did not rise significantly above basal levels in the resting and exercised leg. ^[10] The two studies measure different outcomes (body composition and strength versus rate of protein synthesis) and were done in different populations; together they do not support the conclusion “collagen builds muscle,” and they are consistent with the fact that “collagen peptides are not a complete protein.”

Who funded the collagen benefit studies?

The table below summarizes the funding source of each study according to its own disclosure; funding information does not invalidate a result, but it affects the reader’s interpretation.

This table does not say that collagen does not work; it shows that some of the most-cited positive studies were done with manufacturer support and with the manufacturer’s own product.

Summary of collagen benefits: strength of evidence

This page does not recommend a personal dose; the amounts used in collagen studies, as shown above, ranged from 2.5 g/day to 30 g twice daily depending on the study and product.