Skin and hair in menopause: the 30% collagen myth
Women lose 30% of their collagen in the first 5 years of menopause: the figure is everywhere, and it appears in no original publication. The real data, from Brincat's work in 1987, is about 1 to 2% per year. What estrogen really does to your skin, why the time elapsed since your menopause predicts the state of your collagen better than your age, the hair loss that affects one postmenopausal woman in two, and the honest hierarchy of what works: daily sun protection and retinoids (randomized trials), quitting smoking (strong association), collagen peptides (fragile signal), topical estrogen and hormone therapy (not validated for skin). In French with English subtitles.
Watch on YouTube (full description and references) →
Read the written and sourced version of this topic →
What you will understand
One figure circulates everywhere as soon as skin and menopause come up: "women lose 30% of their collagen in the first five years". You find it on supplement websites, in magazines, in advertising. Going back to the original literature, it appears in no primary publication. It is an extrapolation that became a "fact" by dint of being copied, and it mostly serves to sell.
The real data exists, and it is older: Brincat's work, in 1987, describes a decline of about 1 to 2% per year in skin collagen. Over five years, that makes 5 to 10%, not 30%. The nuance is not cosmetic: it completely changes the degree of urgency being sold to you.
The most interesting observation in that work lies elsewhere. What best predicts the state of your collagen is not your chronological age, it is the number of years elapsed since your menopause. Your skin is a hormonal organ: it carries estrogen receptors on keratinocytes, fibroblasts, melanocytes, hair follicles and sebaceous glands.
On the hair side, a 2022 cross-sectional study found female pattern hair loss in 52.2% of the postmenopausal women studied. But after adjustment, only BMI remained statistically associated: you therefore cannot claim that menopause causes hair loss. That is the kind of precision marketing removes.
The video ends with the most useful part: the honest hierarchy of evidence. What is demonstrated by randomized trial, what is only an association, and what is not validated at all. Without fear, without guilt, and without inventing anything.
The key points to remember
- The "30% of collagen in 5 years" figure appears in no original publication. The real data, from Brincat 1987, is about 1 to 2% per year.
- Your skin is a hormonal organ: estrogen receptors on keratinocytes, fibroblasts, melanocytes, hair follicles and sebaceous glands.
- The decline in collagen follows menopausal age (the years elapsed since menopause) more than chronological age.
- Female pattern hair loss: 52.2% in a 2022 cross-sectional study, but after adjustment only BMI remained significant. Association, not causation.
- The best demonstrated measure for skin: daily sun protection. It is the only long-term randomized anti-aging trial (Nambour, 4 and a half years).
- Topical retinoids have proved themselves in a randomized trial. The 0.025% concentration is as effective as 0.1%, and irritates less.
- Tobacco is strongly associated with premature wrinkling. The association is observational, the mechanism is consistent.
- Collagen peptides in powder form: a favorable signal in a meta-analysis, but the heterogeneity of the studies was criticized in the same journal, and industry funding is frequent.
- Hormone therapy does not have skin among its recognized indications (NAMS 2022). The best trial on topical estrogen is negative.
Video chapters
- 00:00 What you see in the mirror
- 00:54 Your skin is a hormonal organ
- 01:53 The 30% figure: back to the sources
- 02:50 The real data: 1 to 2% per year
- 03:47 What that really changes
- 04:46 Your hair: 1 woman in 2
- 05:41 What works for hair
- 06:38 The best demonstrated anti-aging measure
- 07:39 Retinoids
- 08:32 Tobacco
- 09:29 Collagen powder
- 10:29 Hormones and skin
- 11:30 The hierarchy of evidence
- 12:29 Recap
The scientific sources cited
Every claim in this video rests on a verified reference. Here is the complete list, with the PubMed links. This work is covered in detail in the written, sourced version of this topic.
- Brincat M et al., BMJ, 1983. PubMed 6416400
- Brincat M et al., Obstet Gynecol, 1987. PubMed 3120067
- Brincat M et al., Br J Obstet Gynaecol, 1987. PubMed 3828252
- Affinito P et al., Maturitas, 1999. PubMed 10656502
- Castelo-Branco C et al., Maturitas, 1992. PubMed 1345134
- Brincat M et al., Br J Obstet Gynaecol, 1985. PubMed 3978054
- Sumino H et al., Endocr J, 2004. PubMed 15118265
- Thornton MJ, Dermatoendocrinol, 2013. PubMed 24194966
- Raine-Fenning NJ et al., Am J Clin Dermatol, 2003. PubMed 12762829
- Ashcroft GS et al., Am J Pathol, 1999. PubMed 10514397
- Yoon HS et al., Acta Derm Venereol, 2014. PubMed 23722352
- Rzepecki AK et al., Int J Womens Dermatol, 2019. PubMed 30997378
- NAMS Position Statement, Menopause, 2022. PubMed 35797481
- Calleja-Agius J, Brincat M, Gynecol Endocrinol, 2012. PubMed 21970508
- Kamp E et al., Clin Exp Dermatol, 2022. PubMed 35727900
- Chaikittisilpa S et al., Menopause, 2022. PubMed 35357365
- Birch MP et al., Br J Dermatol, 2001. PubMed 11251562
- van Zuuren EJ et al., Cochrane Database Syst Rev, 2016. PubMed 27225981
- Blume-Peytavi U et al., J Am Acad Dermatol, 2011. PubMed 21700360
- Bergfeld W et al., J Drugs Dermatol, 2016. PubMed 27391639
- Hughes MC et al., Ann Intern Med, 2013. PubMed 23732711
- Weiss JS et al., JAMA, 1988. PubMed 3336176
- Griffiths CE et al., Arch Dermatol, 1995. PubMed 7544967
- Koh JS et al., Int J Dermatol, 2002. PubMed 11895509
- de Miranda RB et al., Int J Dermatol, 2021. PubMed 33742704
- Peres G, Int J Dermatol, 2022 (methodological critique). PubMed 34196407
- Pullar JM et al., Nutrients, 2017. PubMed 28805671
Frequently asked questions
Where does the 30% collagen figure really come from?
From nowhere, strictly speaking. Going back to Brincat's publications from the 1980s, you find a cross-sectional comparison between treated and untreated women, and a decline estimated at about 1 to 2% per year. The "30% in 5 years" is a review extrapolation, repeated until it became an apparent fact.
Does collagen powder work?
The signal exists: a meta-analysis of 19 randomized trials reports an improvement in hydration, elasticity and wrinkles at around 90 days. But a critique published in the same journal points to the heterogeneity of the included studies, and many are funded by industry. It is a fragile signal, not solid proof.
Does hormone therapy improve the skin?
It is not a recognized indication. The 2022 NAMS position does not include skin among the indications for hormone therapy. Some older trials show an increase in collagen, but a skin benefit does not on its own justify treatment. It is a shared medical decision, taken for other reasons.
What about estrogen cream on the face?
The best available randomized trial is negative: on sun-exposed skin, neither wrinkles nor elasticity were improved, and there was in addition an induction of MMP-1, an enzyme that degrades collagen. The positive trials are old and small.
What should I do first for my skin after 50?
If you were to remember only one thing: daily sun protection. It is the only anti-aging measure supported by a long-term randomized trial. Then come topical retinoids, then quitting smoking. This ranking is not a prescription: discuss it with the professional who knows your file.
Does menopause really make hair fall out?
Caution is needed, despite a high frequency. A 2022 cross-sectional study found female pattern hair loss in 52.2% of the postmenopausal women studied, which corresponds roughly to one woman in two. But after adjustment, only body mass index remained statistically associated. You therefore cannot claim that menopause causes hair loss: it is an association, not a demonstrated relationship.
What is the best demonstrated anti-aging measure?
Daily sun protection, and by far. It is the only anti-aging measure supported by a long-term randomized trial, the Australian Nambour trial, run over four and a half years. Then come topical retinoids, also evaluated in a randomized trial, with this useful detail: the 0.025% concentration is as effective as 0.1% while irritating less. The choice remains to be discussed with a professional.
Why do the years since menopause count more than age?
Because the skin is a hormonal organ. It carries estrogen receptors on keratinocytes, fibroblasts, melanocytes, hair follicles and sebaceous glands. In Brincat's work, what best predicted the state of skin collagen was not chronological age but the number of years elapsed since menopause. The decline described there is about 1 to 2% per year.