Collagen: what the trials actually show.

Collagen is a $2.5B annual category. Most of it doesn't do what the label says.

Here's the research, sorted by what's actually been demonstrated in placebo-controlled human trials:

What works (modest evidence)

Hydrolyzed collagen peptides, 2.5-10g/day, 8+ weeks. Skin elasticity improves by 7-9% vs placebo in pooled meta-analysis. Wrinkle depth decreases by similar magnitude. Choi et al., 2019 meta-analysis covering 11 RCTs.

Mechanism: ingested peptides don't go to skin directly (digestion breaks them to amino acids). They appear to act as a signal to fibroblasts to produce more endogenous collagen. The signal is real but modest.

Vitamin C, oral, 75-200mg/day. Required cofactor for collagen synthesis. Most people are sufficient, but borderline-deficient women show measurable skin elasticity improvement on supplementation. Pullar et al., 2017 — Nutrients.

What doesn't work

Topical collagen creams. Collagen molecules are too large to cross the stratum corneum. They sit on skin until you wash off. This has been known since the 1980s. The category exists because the word "collagen" sells.

Marine vs bovine vs chicken collagen — no meaningful differential outcome data. The "type I" vs "type II" marketing is overstated; the body breaks both to amino acids.

Collagen "boosters" with proprietary blends. Almost universally weak. The category that works is hydrolyzed peptides at adequate dose. Everything else is variation on a theme.

What works better than collagen for skin

Three interventions out-perform any collagen product in head-to-head trials:

1. Daily SPF 30+. UV is the dominant variable in skin aging. Single best intervention. Hughes et al., 2013 — 4.5 year RCT showed 24% less skin aging on daily SPF group vs sporadic-SPF control.

2. Tretinoin (or retinoid). Prescription strength. Increases collagen synthesis in skin directly (vs ingested collagen which only signals). Dermal effect at 6-12 months. Kligman et al., 1986 and 40 years of follow-up trials.

3. Estrogen, oral or topical. Skin has estrogen receptors. Skin thickness drops ~30% in the first 5 post-menopausal years and reverses with HRT. (Brincat et al., 2005 review). The skin benefit isn't usually marketed because HRT is prescribed for symptoms, not cosmetics — but it's real.

What this means practically

Collagen peptides aren't snake oil, but they're modest. If you're going to spend on the category, spend on the peptides at adequate dose (~10g/day), give it 8-12 weeks, and don't expect more than 7-10% improvement in elasticity metrics.

If you're not already doing the higher-leverage interventions — daily SPF, retinoid, and (if appropriate) HRT — those come first. Collagen is dessert.

On the affiliate disclosure

This site has affiliate links to two collagen products. Both are hydrolyzed peptides at adequate dose, both are in the price range that doesn't require a markup explanation. I take a commission if you buy through the link. I don't take a commission to write favorably about them.

If you don't want to use the affiliate link, the same products are available on Amazon at similar prices. The information is the same either way.

Next week: empty nest. The friendship architecture question for women in their 50s.

Alexander After Forty Feel Reader-funded. Research-led. No supplement-brand sponsorships.