In the spring of 2025, PrimaGround commissioned an independent research group at the University of Lund to run a 12-week, randomized, placebo-controlled trial of our hydrolyzed marine collagen peptides at 10 g/day. The study enrolled 84 subjects (aged 35–60, BMI 19–28, all women), all of whom were screened to exclude known confounders: current collagen supplementation, hormonal therapy, recent significant weight changes.
The trial was registered before enrollment began. The full protocol, data set, and statistical analysis are publicly available on the OSF repository. We had no role in data collection or analysis. What we asked for was a fair test.
The pre-registered hypotheses
The trial tested three primary endpoints:
- Change in skin elasticity (measured by Cutometer dual MPA 580) after 12 weeks
- Change in self-reported joint discomfort during exercise (WOMAC subscale)
- Change in fasting plasma hydroxyproline (a collagen breakdown marker)
Secondary endpoints included sleep quality, perceived hair quality, nail brittleness, and serum CRP as a non-specific inflammatory marker.
What the data showed
The skin elasticity endpoint was the clearest signal: the active group showed a 17.4% increase in dermal viscoelasticity (R5 ratio) vs. 2.1% in placebo (p<0.001). The effect was most pronounced in subjects over 45.
The joint discomfort endpoint showed a smaller but still significant effect: 23% reduction in WOMAC pain subscale in the active group vs. 6% in placebo (p=0.02).
The hydroxyproline marker did not reach significance — which is consistent with most of the prior literature and is what we would have predicted.
The result that surprised us
The secondary endpoint we were most skeptical of — sleep quality (Pittsburgh Sleep Quality Index) — showed a small but statistically significant improvement in the active group (PSQI improvement of 1.3 points vs. 0.2 in placebo, p=0.04). We did not predict this. We are not yet sure what to make of it. There is a plausible mechanism — glycine, abundant in collagen peptides, is a precursor to serine and a known GABAergic modulator — but the effect size in our trial is consistent with the published literature on supplemental glycine for sleep, suggesting that the collagen-and-sleep finding may be a glycine-and-sleep finding in disguise.
We are running a follow-up trial in 2026 with a glycine-matched control arm to test this.
“The protocol was honest. The data was clean. That alone separates this from 90% of what gets called ‘clinical evidence’ in the supplement category.” — Dr. Eva Lindqvist, lead investigator, Lund University
What we changed as a result
Nothing about the formulation. The dose we tested is the dose we sell. The trial confirmed the skin and joint endpoints we built the product around — and gave us an unexpected lead to investigate next.
Full trial data, protocol, and analysis available at osf.io/example-pg-collagen-trial. Trial registered prospectively at ClinicalTrials.gov.