"Collagen Has NO Benefits For Skin Aging"

Brad's supplementation decision is grounded in his extensive reading of the evidence. He notes that while the meta-analytic headline was negative, the actual data paint a different picture. He walks through the fibroblast lab study showing that collagen peptides boost production of structural skin components, and multiple studies demonstrating that ingested collagen peptides appear in the blood and then in the skin. The burn study provides some evidence of collagen-specific benefits over general protein, though not in skin aging. He is transparent that a definitive study in skin aging is still needed, but given the safety and the consistent positive signals, he chooses to continue. The protocol amount is 12.5 g, likely a practical dose used in many studies.

Brad Stanfield, who had taken collagen for years, was initially stoked by the new abstract suggesting he could save money. Upon reading the full study, he discovered that the authors' conclusion hinged on two statistical artifacts: the conversion of diverse outcome measures into standard mean differences (SMD) and the resulting wide confidence intervals. For non-industry-funded studies, the SMD for skin hydration was 0.33 with a confidence interval of -0.06 to 0.73. Because the interval barely dipped below zero, the result was labeled non-significant, despite the majority of the range being positive. In the set of 9 high-quality studies, 7 showed positive outcomes, and among the 5 that were both high-quality and non-funded, 4 were positive. Thus, the data overwhelmingly points to benefit, but the pooled statistics blurred this. He argues that the authors missed the wood for the trees: the individual trials, including those of high quality and without industry funding, consistently report improvements in skin hydration, elasticity, and wrinkles. This disconnect between individual study results and meta-analytic conclusion is a critical lesson in evidence appraisal.

His decision rests on several pillars: First, the majority of high-quality, non-industry-funded studies still show benefit. Second, there is a plausible biological mechanism—collagen peptides stimulate fibroblasts to produce collagen, elastin, and proteoglycan, and ingested peptides reach skin via blood. Third, a burn-patient study directly compared collagen peptides to matching protein, and collagen accelerated wound healing, suggesting skin-specific benefits beyond mere protein intake. Fourth, he acknowledges the remaining uncertainty (no skin-aging matched-protein trial yet) but views the risk as minimal because collagen peptides have no known health risks. He emphasizes the need for more studies but sees the preponderance of evidence as sufficient to continue.