She’s partly right — and so is the research that disagrees with her. Dr Farah Moustafa, a dermatologist at Tufts University School of Medicine, made headlines in early 2026 arguing that collagen supplements aren’t a recommended treatment for skin ageing. A few months later, the largest umbrella review ever published on the subject — 113 randomised controlled trials, nearly 8,000 participants — called the evidence “cautiously positive.” Both positions are credible, defensible, and incomplete. That genuine scientific disagreement is more useful than a clean answer either way, because it tells you exactly where the uncertainty actually sits.
The sceptical case: what Dr Moustafa actually argues
Her position, published via Tufts University and widely covered in SciTechDaily and medical outlets, isn’t that collagen supplements are dangerous or worthless — it’s more measured than the headlines suggest. Three core points:
Collagen gets broken down before it’s absorbed. This is the foundational objection. Your body doesn’t absorb intact collagen molecules and route them to your skin. It breaks them down during digestion, the way it breaks down any protein. Eating collagen, in this framing, is no different from eating any other protein source — your body decides what to do with the amino acids, and there’s no guarantee they end up in your skin rather than your muscles, bones, or anywhere else.
The studies that show benefits tend to be industry-funded and lower quality. Dr Moustafa cites a 2025 meta-analysis of 23 randomised controlled trials, published in The American Journal of Medicine by Myung and Park. Its headline finding: when the authors separated trials by funding source, industry-backed studies reported positive effects on hydration, elasticity, and wrinkles — but higher-quality studies with non-industry funding found no significant benefit on any of those measures. That’s a real, peer-reviewed finding, not a fringe claim.
Better-studied alternatives exist. Topical retinoids, broad-spectrum sunscreen, adequate dietary protein and vitamin C, and not smoking all have stronger evidence bases for skin health than any oral collagen supplement. Her recommendation is that collagen can be considered alongside those options, not used instead of them.
The counter-case: what the larger evidence base shows
The sceptical position is credible. It’s also not the whole picture.
The umbrella review is bigger and newer. The 2026 ARU umbrella review — led by Professor Lee Smith at Anglia Ruskin University and published in Aesthetic Surgery Journal Open Forum — analysed 16 systematic reviews covering 113 RCTs and nearly 8,000 participants. Its conclusions: longer supplementation was associated with greater improvements in skin hydration and elasticity, and the overall evidence was “cautiously positive.” That’s a far larger evidence base than any single meta-analysis, including the one Dr Moustafa cites.
The “body just breaks it down” argument isn’t quite that simple. Hydrolysed collagen doesn’t survive digestion intact — Dr Moustafa is right about that. But it doesn’t break down into entirely generic amino acids either. Research has shown that specific bioactive dipeptides, particularly prolyl-hydroxyproline (Pro-Hyp), survive digestion, appear in human blood after oral collagen intake, and stimulate dermal fibroblast growth and hyaluronic acid synthesis in laboratory studies. The mechanism by which ingested collagen peptides could reach and affect skin cells isn’t hypothetical — it’s been documented, albeit mostly in cell studies rather than large clinical trials. That’s a meaningful gap between “the body just destroys it” and “there’s no plausible mechanism.”
A second independent review agrees on the important parts. A separate meta-analysis by Batool et al., published in the Journal of Cosmetic Dermatology in 2026, found meaningful improvements in hydration (from as early as 4 weeks) and elasticity (mainly at 12 weeks) — but, importantly, did not find robust evidence that collagen reduces wrinkles or increases dermal density. Two independent groups reaching the same core conclusion — modest hydration and elasticity benefits, but not the wrinkle reduction that marketing emphasises — is harder to dismiss than one.
The industry-funding question deserves an honest look from both sides
Dr Moustafa’s point about industry funding is real, and the Myung and Park meta-analysis is a legitimate piece of research. But the supplement industry’s rebuttal isn’t entirely empty either. The Collagen Stewardship Alliance argued that several studies classified as “independent” in that meta-analysis actually had commercial ties, and that the quality-rating criteria weren’t fully disclosed. The Gelatin Manufacturers of the World pointed out that the review pooled very different formulations — marine collagen, bovine collagen, and blended products — as if they were interchangeable, and that 17 of 23 included trials used lower doses (around 3g/day) common in Asian markets rather than the higher doses more typical in European formulations.
None of that invalidates the core finding. But it does mean the clean narrative — “only industry-funded studies show benefit” — is more complicated than it sounds.
The ARU umbrella review, drawing on a much larger pool of 113 trials, noted a related pattern: newer, higher-quality studies showed smaller elasticity gains but larger hydration gains compared with older ones. That could mean formulations are improving, study design is getting better, or publication bias is being squeezed out. Probably some combination of all three.
The honest middle ground
Both sides are partly right, and neither settles it.
Dr Moustafa is right that collagen supplements aren’t a first-line treatment for skin ageing, the evidence base has real quality problems, industry funding is a legitimate concern, and sunscreen plus retinoids have a stronger evidence trail. She’s also right that the supplement market is poorly regulated — most products lack third-party verification, and marine-sourced collagen carries a methylmercury contamination risk she reasonably flags.
The positive reviews are right that modest improvements in hydration and elasticity show up consistently across multiple independent analyses, at a scale (thousands of participants across dozens of trials) that’s hard to explain away entirely. The bioactive-dipeptide mechanism is real, documented, and more specific than “the body just destroys it like any protein.”
What neither side has proven: that collagen supplements produce the dramatic, visible anti-ageing results the marketing implies. The ARU review’s own lead author said it plainly: “Collagen is not a cure all, but it does have credible benefits when used consistently over time.” The Journal of Cosmetic Dermatology meta-analysis went further — hydration and elasticity improved, but wrinkle reduction and dermal density did not survive rigorous analysis. Those are the two outcomes supplement marketing leans on hardest, and they’re the two that didn’t hold up.
What this means practically
If you’re already using sunscreen, eating well, and considering a retinoid, a collagen supplement is — at best — a modest addition, not a replacement. The evidence says “might help a little with hydration and elasticity, over weeks” rather than “turns back the clock.” Expect consistency to matter more than dose, based on what both the ARU review and the Batool et al. meta-analysis found.
If you’re choosing between a collagen supplement and sunscreen, buy the sunscreen. That’s not controversial on either side of this debate.
Our evidence roundup covers the specific trial data for the peptides used in the jelly sticks we track. The ARU umbrella review piece goes deeper into what 113 trials actually found. And our where-to-buy guide compares six UK-available options if you’ve weighed this up and decided to try one.