§ Claim under review · Efficacy
"A 2025 systematic review and meta-analysis of 23 randomised trials (1,474 people) found that high-quality studies showed no significant effect of collagen supplements on skin hydration, elasticity, or wrinkles, while low-quality, industry-funded studies reported benefits."
Verdict
Mostly accurate
Confidence
HighSummary
This one checks out on the facts, with one important piece left out. The study is real: a 2025 review in The American Journal of Medicine pooled 23 randomised trials with 1,474 people, and it did report that trials rated highest quality, and trials without industry funding, showed no significant effect on skin hydration, elasticity or wrinkles, while industry-funded and lower-rated trials reported benefits. What the post's headline sentence leaves out is that when all 23 trials were pooled together, the result was positive for all three measures. The null findings appear only once the authors split the set. Worth knowing too that "high quality" here meant a perfect 5 out of 5 on a 5-point scale in a set averaging 4.22, so the bar was narrow, and that a larger 2026 review covering 113 trials and nearly 8,000 people did find improvements in skin elasticity and hydration while finding no clear effect on wrinkles. The two reviews genuinely disagree and that disagreement has not been resolved. This is general information only and not medical or dermatological advice.
The readings
key figures from the evidencetrials in 2025 Myung and Park collagen meta-analysis
participants in 2025 collagen meta-analysis
average Jadad quality score across included trials
Why this verdict
Evidence
The paper described in the claim exists and the numbers match. The abstract states that a total of 23 RCTs with 1474 participants were analyzed, and that in a meta-analysis of all 23 RCTs, collagen supplements significantly improved skin hydration, elasticity, and wrinkles.
It then reports that in the subgroup meta-analysis by funding source, studies not receiving funding from pharmaceutical companies revealed no effect of collagen supplements for improving skin hydration, elasticity, and wrinkles, while those receiving such funding did. On the quality axis, the published PDF states that in the subgroup meta-analysis by methodological quality, high-quality studies revealed no significant effect . Trade coverage of the paper summarises the same two subgroup findings, reporting that trials not funded by pharmaceutical or supplement companies showed no benefit while those with industry backing reported positive effects, and that when only higher-quality studies were considered, no significant impact was observed across hydration, elasticity or wrinkles . The authors' stated conclusion is blunt: there is currently no clinical evidence to support the use of collagen supplements to prevent or treat skin aging.
On how quality was defined, the paper's methods say the grouping rested on the RoB 2 or Jadad scale, and a combination of methodological quality and funding status, high-quality and not funded versus low-quality and funded , and that studies scoring 5 on the five-point Jadad scale were considered high-quality and the remaining studies were classified as low-quality, as the average score across all included studies was 4.22 . The paper itself declared Funding: None and no conflict of interest for either author .
Later and larger syntheses reach a different headline. A 2026 umbrella review from Anglia Ruskin University brings together evidence from 16 systematic reviews, 113 randomised controlled trials and almost 8,000 participants around the world , and reported noted improvements in skin elasticity rated high certainty, SMD 1.01, 95% CI 0.44 to 1.59, across 1217 participants . Press coverage of that review reports that collagen supplements can improve skin elasticity and hydration when taken consistently over time but do not stop wrinkles . A separate 2026 meta-analysis found oral collagen supplementation improved skin hydration, SMD 0.71, 95% CI 0.08 to 1.34, and elasticity, with very high statistical heterogeneity of 90.4 percent .
Findings
✓ What's accurate 6
- The paper exists and is correctly identified by year and design. Myung and Park published a systematic review and meta-analysis in The American Journal of Medicine in 2025.
- The two numbers are exact. 23 RCTs and 1,474 participants is what the abstract reports.
- The high-quality subgroup finding is reported as the claim states. When only higher-quality studies were considered, no significant impact was observed across hydration, elasticity or wrinkles.
- The funding-source split is reported as the claim states. Trials without pharmaceutical company funding showed no effect on the three outcomes, while trials with such funding did.
- Treating quality and funding together is something the authors themselves did, not an invention of the post. The methods describe a combination of methodological quality and funding status, high-quality and not funded versus low-quality and funded.
- The post's own image text, which says some formulations show modest gains and that dramatic reversal is not supported, is consistent with the wider literature, including the 2026 umbrella review's finding of improvement in elasticity and hydration but not an end to wrinkles.
≈ What's misleading 4
- The claim sentence reports only the subgroup results and leaves out the paper's own headline pooled result. Across all 23 trials together, the supplements significantly improved skin hydration, elasticity, and wrinkles. A reader who sees only the subgroup half would not know that the main analysis was positive and that the null findings emerge when the set is split. The post's caption partly repairs this by mentioning modest gains in individual trials, but the claim as written does not.
- "High-quality" sounds like a settled grade and is not. In this paper it meant a perfect 5 out of 5 on the Jadad scale, against a set whose average score was 4.22, so most trials in the pool were close to the cut-off and fell on the "low-quality" side of it. That is a narrow and paper-specific threshold, not a general judgement that those trials were poor.
- The claim compresses "low-quality" and "industry-funded" into one group, as if the benefit-reporting trials were all both. The paper ran these as separate subgroups as well as a combined one, and trade bodies responding to the paper argue that the majority of the industry-funded studies scored 3 to 5 out of 5 on the Jadad scale. Those are interested parties and their objections are context rather than evidence, but the point that the two labels are not interchangeable stands on the paper's own methods.
- Presented in October 2026 with no date framing beyond "2025," the claim reads as the current state of the evidence. It is not the most recent or the largest synthesis. The 2026 umbrella review covering 16 systematic reviews, 113 randomised controlled trials and almost 8,000 participants reports improvements in elasticity and hydration.
? What's uncertain 4
- The exact effect sizes and confidence intervals within the high-quality subgroup, and how many of the 23 trials fell into it, could not be confirmed from the abstract and PDF excerpts available. A null result in a small subgroup can reflect too few participants rather than a true absence of effect, and without those numbers that distinction cannot be settled here.
- Whether the funding classifications Myung and Park assigned to individual trials are correct is contested. The challenges come from collagen industry bodies, so they carry an obvious interest, and no independent re-analysis correcting or confirming the classifications was found.
- The two syntheses disagree and the reason is not resolved. The 2025 paper splits by quality and funding and finds nulls; the 2026 umbrella review pools published meta-analyses and finds benefit. Pooling meta-analyses can carry forward the same biased trials the 2025 paper was trying to filter out, and the 2026 hydration estimate came with heterogeneity of 90.4 percent, meaning the trials disagree with each other a great deal.
- Dose and formulation are not pinned down anywhere in the claim. Collagen supplements differ widely in peptide source, molecular weight, dose and added ingredients such as vitamin C and hyaluronic acid, and a pooled figure across that mixture does not tell a reader what any specific product does.
Sources
7 of 7 linked to recordsMyung S-K, Park Y. "Effects of Collagen Supplements on Skin Aging: A Systematic Review and Meta-Analysis of Randomized Controlled Trials." The American Journal of Medicine 2025;138(9):1264-1277. DOI 10.1016/j.amjmed.2025.04.034. PMID 40324552
Same paper, full-text PDF (methods section on Jadad and RoB 2 subgrouping)
Ravindran/Smith et al. "Collagen Supplementation for Skin and Musculoskeletal Health: An Umbrella Review of Meta-Analyses on Elasticity, Hydration, and Structural Outcomes." Aesthetic Surgery Journal Open Forum 2026;8:ojag018, published 30 Jan 2026
Moltó-Balado P, Simeó-Monzó A. "Oral Collagen Supplementation for Skin Hydration and Elasticity." Journal of Aesthetic Medicine 2026;2(3):17
Tufts University / Farah Moustafa, dermatologist commentary, Jan 2026
NutraIngredients, "Industry reacts to meta-analysis...", 26 Aug 2025
Collagen Stewardship Alliance statement; Tosla Nutricosmetics white paper