Does collagen supplementation help skin, joints or bones?

Some trials report small gains in skin moisture and knee pain. Bone evidence rests on one 12-month trial, and funding bias is a concern.

Published October 2026

Short answer

Small, mixed benefits for skin and knee pain in trials, and thin evidence for bones. Collagen is not a reliable fix for any of the three. Evidence grade C (Limited). Meta-analyses and trials for skin and knee pain conflict, many are small or industry-linked, and bone rests on one trial; downgraded for inconsistency, indirectness and risk of bias.

Key takeaways

Meta-analyses of skin trials report better hydration, elasticity and wrinkles than placebo [1], but a 2025 review found no effect in high-quality trials or in trials without pharmaceutical funding [2].
A 2025 meta-analysis of 11 trials (870 people) found less knee osteoarthritis pain and better function with collagen, but results varied widely, and a 2019 meta-analysis found no significant effect on the pain subscore [3][4].
A 2025 trial of 68 people with knee osteoarthritis found no difference from placebo after 12 weeks [5].
One 12-month trial of 131 postmenopausal women found a small rise in spine and femoral neck bone density on collagen peptides versus placebo [6].
Side effects were low in the trials reviewed, but most trials had a high risk of bias and the review calls for longer studies [7].

The short version

Most trials of collagen we reviewed test skin and knee pain, and their results conflict [1][2][3][5]. Some trials and meta-analyses report small gains in skin moisture and knee pain. The bone evidence comes from a single trial of 131 women [6].

What the studies show

Skin. A 2021 meta-analysis of 19 placebo-controlled trials (1,125 people, 95% women) found that hydrolyzed collagen improved skin hydration, elasticity and wrinkles over 90 days [1]. A 2025 review of 23 trials (1,474 people) also found pooled gains. But trials without pharmaceutical funding showed no effect, and the high-quality trials showed no significant effect. The review's authors concluded there is currently no clinical evidence to support collagen for skin aging [2]. A 2026 review judged most of its 25 trials to have a high risk of bias and called the skin evidence preliminary [7].

Knee pain. A 2025 meta-analysis of 11 trials (870 people) found that collagen improved knee osteoarthritis pain and function, but results varied widely between trials [3]. A 2019 meta-analysis found that the total WOMAC score and its stiffness subscore improved, while the pain and function subscores did not differ significantly from placebo [4]. A 2016 trial of 191 people found that undenatured type II collagen, 40 mg a day, lowered WOMAC scores over 180 days compared with placebo [8]. Overall, the knee results are mixed: pooled meta-analyses favour collagen, but not every trial agrees. A 2025 trial of 68 people found no difference from placebo after 12 weeks [5]. In 139 athletes with activity-related knee pain, 5 g a day of specific collagen peptides for 12 weeks reduced pain during activity more than placebo, but pain at rest did not differ significantly [9].

Bones. In a 12-month trial of 131 postmenopausal women with age-related bone loss, spine T-scores rose by 0.1 on 5 g a day of specific collagen peptides and fell by 0.03 on placebo [6].

The nuances

Products and doses. The trials used different products: hydrolyzed collagen, specific collagen peptides and undenatured type II collagen [8][10]. Hydrolyzed collagen doses in the reviewed studies ranged from 2.5 to 10 g a day [10].

Absorption. In a crossover study of healthy adults, collagen hydrolysate produced measurable blood levels of hydroxyproline and related peptides [11]. That study measured blood levels only, so it does not show that the peptides reach skin, cartilage or bone.

Funding. Most earlier collagen trials were industry-sponsored, according to the background of a 2025 trial [5]. In the 2025 skin review, the benefit appeared only in trials funded by pharmaceutical companies [2]. Several authors of the 2024 absorption study were employed by Rousselot BV, BioTeSys GmbH or Triskelion [11].

Sports nutrition. In a retrospective subgroup of 20 older men who were overweight or obese, taken from a three-month trial of 32, a whey- and casein-based supplement blend produced larger gains in muscle mass and leg strength than a collagen-based blend [12]. Some authors were employed by Exerkine Corporation [12].

Safety

Side effects were low in the reviewed trials [7], and the 2025 knee trial reported none in either group [5]. A 2019 review noted that collagen supplements lack regulation for quality, absorption and efficacy [10]. None of the studies we read reported on pregnancy, breastfeeding, or people with kidney or liver disease, and we found no established upper limit. See the Collagen page for the full safety section.

What this means for you

The evidence hints at small benefits for skin moisture and knee pain, but it conflicts and much of it is industry-linked [2][3]. Bone evidence is a single trial [6]. Anyone considering collagen for joints or bones should discuss it with a doctor, especially if they take medicines or have a long-term health condition.

Go deeper

Collagen — Some trials report small gains in skin hydration and in knee pain, but results conflict and many trials are small or industry-linked.