"From age 25 onwards, we lose 1% of collagen per year." The phrase is everywhere, on creams, supplements and social media, and it deserves a real answer rather than a simple yes or no. Because collagen loss with age is one of the best-measured phenomena in dermatology over the past half-century, and the figure of 25 years doesn't come from any study at all.
This guide explains where the 1% comes from, what skin cells actually do as they age, why menopause changes the equation, what accelerates the loss, and what the three meta-analyses on oral collagen measured: at what dose, over how long, and with what limitations.
Reality. Skin collagen decreases continuously with age, measured since 1975 in a large number of healthy subjects; this is where the figure of approximately 1% per year comes from. Convention. Nothing is triggered at age 25: the decline is progressive from the start of adulthood. Mechanism. Fibroblasts age, 82 versus 56 ng/ml of procollagen between ages 20 and 80, and lose their attachment tension; the sun fragments the matrix and perpetuates a vicious cycle.
Menopause. 48% more collagen in the skin of women treated with hormones, at the same age. Oral collagen. Three meta-analyses, encompassing 19, 26 and 14 trials, measure improved hydration and elasticity at 90 days, with modest effects and noted biases. The most impactful factors remain sun exposure, tobacco and sugar.
Where does the "1% per year" come from
Skin collagen, dermis thickness and collagen density of the forearm were measured in a large number of normal subjects, to serve as reference for future studies. Skin collagen decreased with age and was lower in women at all ages. There is a direct relationship between skin collagen and dermis thickness.
Shuster S, Black MM, McVitie E. Br J Dermatol 1975;93(6):639-643. DOI: 10.1111/j.1365-2133.1975.tb05113.x
It was this study, conducted in Newcastle, that established the reference. It measured skin collagen, not estimated it: by biopsy, in healthy subjects of all ages. And it found a decline continuous throughout adult life, with no break, whose slope corresponds to approximately 1% per year, the figure adopted since by the entire industry. It also found that women have less skin collagen than men at all ages, which partly explains why signs of skin aging often appear earlier in them.
It comes from no study. The measured decline is progressive and begins at the start of adulthood, when growth ends. 25 years is simply the age when the plateau of youth has passed and when the slope becomes visible on a curve, and it's a convenient round number for marketing. Reality is less clear-cut and more interesting: it's not a switch, it's a slope, and what matters is what makes it steeper or less steep.
What skin cells do as they age
The reduction in synthesis of collagen types I and III is characteristic of chronologically aged skin. The reduction due to fibroblast aging was demonstrated by lower production of type I procollagen by fibroblasts isolated from skin of young subjects, 18 to 29 years old, compared to subjects over 80 years old: 82 versus 56 ng/ml. A reduction in mechanical stimulation was evidenced by morphological studies comparing dermal sections: a higher percentage of cell surface attached to collagen fibers, 78 versus 58 percent, and more extensive cell spreading in young skin than in aged skin. The authors conclude that the reduction in synthesis reflects at least two mechanisms: fibroblast aging and a lower level of mechanical stimulation.
Varani J, Dame MK, Rittie L, et al. Am J Pathol 2006;168(6):1861-1868. DOI: 10.2353/ajpath.2006.051302
The fibroblasts are the skin's collagen factories. The University of Michigan team showed that they weaken in two ways. First, they age themselves: removed from the skin and cultured under the same conditions, those from an 80-year-old person produce one-third less procollagen than those from a 25-year-old person. Next, they lose their tension : a fibroblast only produces collagen if it is attached to fibers that pull it, and in aged skin it finds fewer of them.
The same team described, in a review published in Archives of Dermatology, how these two mechanisms chain together in a vicious cycle. With age and sun exposure, enzymes called metalloproteinases, fragment the matrix. Fibroblasts can no longer attach to the fragments, they collapse, and a collapsed cell produces little collagen and many degradation enzymes. Fragmentation maintains collapse, which maintains fragmentation. The authors note that treatments with proven efficacy—topical retinoids, laser, injected hyaluronic acid—all work by stimulating the production of new, non-fragmented collagen, to which cells can reattach to regain their tension.
Menopause, the second factor
Skin biopsies were taken from 29 postmenopausal women who had not received hormone therapy and from 26 women treated with estrogen and testosterone implants for two to ten years. The average hydroxyproline content, and thus collagen, in the skin was 48 percent higher in treated women than in untreated women, matched for age. This difference was significant. The implication is that estrogen or testosterone, or both, prevent the decline in skin collagen that occurs with age and protect the skin as they protect bone in postmenopausal women.
Brincat M, Moniz CF, Studd JW, et al. Br Med J 1983;287(6402):1337-1338. DOI: 10.1136/bmj.287.6402.1337
At equal age, nearly half more collagen in the skin of women whose hormones were maintained. This result, published over forty years ago, tells us two things. That the drop in estrogen is, after age itself, the factor that weighs most heavily on skin collagen, which explains the frequent feeling of "aging all at once" around menopause. And that the slope of 1% per year is not a uniform inevitability: it is steeper in the years following menopause, then stabilizes. Our premenopause guide details this period.
What accelerates the loss
| Factor | What it does | Level of evidence |
|---|---|---|
| Sun | Activates enzymes that fragment collagen; at the heart of the vicious cycle described by theArchives of Dermatology | High, it is the primary external factor |
| Menopause | The drop in estrogen removes a signal that supports synthesis; 48% difference measured | High, biopsies |
| Tobacco | Reduces dermal oxygenation and activates the same degradation enzymes as sun exposure | High, clinical and histological data |
| Excess sugar | Glycation: sugars bind to collagen fibers and stiffen them | Moderate, established mechanism, skin effect less quantified |
| Poor diet | Insufficient proteins and lack of vitamin C, an essential cofactor for enzymes that assemble collagen | Moderate; frank vitamin C deficiency prevents synthesis |
This table gives the order of priorities, and it is not the marketing one. Before any supplement, facial sun protection and smoking cessation are what most change the trajectory. A diet sufficient in proteins and vitamin C is the foundation upon which all collagen synthesis rests, with or without supplementation.
Oral collagen: three meta-analyses
Eligible studies were randomized, double-blind, placebo-controlled trials evaluating oral hydrolyzed collagen supplementation and reporting at least one of the following outcomes: wrinkles, hydration, elasticity, and skin firmness. Nineteen studies were included, totaling 1,125 participants aged 20 to 70 years, of whom 95 percent were women. Pooled analysis showed favorable results for hydrolyzed collagen compared to placebo on hydration, elasticity, and wrinkles, with the first two confirmed in subgroup analysis. The authors conclude that taking hydrolyzed collagen for 90 days is effective in reducing signs of skin aging.
de Miranda RB, Weimer P, Rossi RC. Int J Dermatol 2021;60(12):1449-1461. DOI: 10.1111/ijd.15518
A second meta-analysis, published in 2023 in Nutrients by a Cochrane Taiwan team, is the largest: 26 randomized trials and 1,721 participants. It confirms improvements in hydration and elasticity compared to placebo, and provides two useful nuances. The effect on hydration varies according to the source of collagen and the duration of intake; the effect on elasticity does not differ according to source. And its authors note several biases in the included trials, concluding that larger trials remain necessary. A third meta-analysis, covering 14 trials and 967 participants published between 2017 and 2023, finds the same results on hydration and elasticity at 12 weeks.
They measurehydration andelasticityusing devices, and wrinkles in the first one. They do not measure "rejuvenation," nor an increase in dermal collagen stock, nor a long-term effect beyond three months. The effects are statistically significant and clinically modest. And a significant portion of the included trials are funded by manufacturers, which the authors of the 2023 meta-analysis reflect in their note on biases. Oral collagen has real data; it doesn't turn back the clock twenty years.
Dose, duration, timing
The dose. The trials included in the meta-analyses generally use between 2.5 and 10 g of hydrolyzed collagen per day, with the lowest doses corresponding to standardized short peptides. What matters on a label is the amount of hydrolyzed collagen and, when indicated, the fraction of short peptides. Our article on dosage details the trials dose by dose.
The duration. The two largest meta-analyses use 90 days ; the third measures at 12 weeks. Below eight weeks, nothing can be assessed, because dermal renewal is slow. A photo at baseline, same lighting, same angle, allows you to see what the daily mirror does not show.
The timing. The trials did not compare morning and evening: they administered a daily dose at the time chosen by the participant. What matters is consistency for at least eight weeks, not the time of day. Our article when to take collagen details what we know about peptide absorption and why consistency trumps timing.
10 g of type 1 marine collagen per day, the high dose from the trials, with 210 mg of hyaluronic acid and 80 mg of vitamin C, a cofactor for collagen synthesis. In powder form, to be mixed into a hot or cold beverage, one daily serving, 30 days per jar. For a 90-day course, three jars or subscription.
View Marine Collagen PowderEffects measured in trials: hydration and elasticity, at 90 days, modest. Does not replace sun protection or smoking cessation.
Choose what fits you best: the answer appears right below.
At this age, the decline is real but slow, and fibroblasts still work well. What makes the slope steep is sun and smoking, not lack of supplement. Daily facial sun protection, proteins and vitamin C in your diet. Oral collagen has data on hydration at all ages in trials, 20 to 70 years old, but it comes second.
Sun and smoking activate enzymes that fragment collagen and feed the vicious cycle described by theArchives of Dermatologyreview. No supplement compensates for this. Sun protection and smoking cessation are, by far, the two actions that most change the slope of that 1%.
Estrogen decline is the second factor after age, with 48% collagen difference measured between treated and untreated women. This is the period when oral collagen makes the most sense, in a 90-day course within the trial range, and it's also a period to discuss with a doctor, because the hormonal question isn't solved with a supplement.
Meta-analyses measure their effects at 90 days, and dermal renewal takes weeks. Three weeks tells you nothing. Take a photo now, same lighting and same angle, continue until twelve weeks without interruption, and compare. While you're at it, check the dose on the label: between 2.5 and 10 g per day in the trials.
This test provides guidance, it does not replace professional medical advice.
Frequently asked questions
Do we really lose collagen starting at 25 years old?
The loss is real, the 25-year figure is a convention. The reference study, published in 1975 in the British Journal of Dermatology, measured skin collagen on the forearm in a large number of healthy subjects: it decreases continuously with age throughout adult life, and it is lower in women at all ages. The figure of approximately 1% per year commonly cited comes from this study. There is no switch at 25: the decline is gradual, it begins at the start of adulthood, and 25 is simply the age when the plateau of youth has passed.
Why does skin produce less collagen as we age?
For two reasons measured by a University of Michigan study published in theAmerican Journal of Pathology. First, fibroblasts age: isolated from skin of people 18 to 29 years old, they produce 82 ng/ml of type I procollagen, compared to 56 in people over 80 years old. Second, they receive less mechanical stimulation: in young skin, 78% of their surface is attached to collagen fibers, compared to 58% in aged skin, and a less-tensioned cell produces less. Both mechanisms add up.
What is the vicious cycle of collagen fragmentation?
A review by the same team, published in Archives of Dermatologydescribes it this way. With age and sun exposure, enzymes called metalloproteinases fragment the collagen matrix. Fibroblasts can no longer attach to the fragments, they collapse, and a collapsed cell produces little collagen and many degradation enzymes. The imbalance perpetuates fragmentation, which perpetuates collapse, in a self-reinforcing cycle. This is the cycle that effective treatments seek to break by stimulating the production of new, non-fragmented collagen that cells can attach to.
Does menopause accelerate collagen loss?
Yes, and it has been measured since 1983. A study published in the British Medical Journal compared skin biopsies from 29 untreated postmenopausal women and 26 women treated with estrogen and testosterone implants for two to ten years, matched for age: collagen content was 48% higher in treated women. The authors conclude that sex hormones prevent age-related decline in skin collagen, just as they protect bone. The drop in estrogen at menopause is therefore the second major factor after age itself.
What accelerates collagen loss?
Sunlight above all: ultraviolet rays activate the enzymes that fragment collagen, and theArchives of Dermatology places sun exposure at the heart of the degradation cycle. Tobacco, which reduces oxygen supply to the dermis and activates the same enzymes. The drop in estrogen at menopause, measured at a 48% difference in collagen between treated and untreated women. And, based on more indirect data, excess sugar through glycation of fibers, lack of sleep, and a diet poor in protein and vitamin C, an essential cofactor in collagen synthesis.
Is there evidence for oral collagen?
Three meta-analyses of randomized, double-blind trials measure this. The first, in theInternational Journal of Dermatology, groups 19 trials and 1,125 participants aged 20 to 70: hydrolyzed collagen improves hydration, elasticity, and wrinkles compared to placebo, and its authors conclude that 90 days of intake is effective. The second, in Nutrients, groups 26 trials and 1,721 participants and confirms hydration and elasticity, noting that the effect on hydration depends on the source and duration, and that several trials present biases. The third, on 14 trials and 967 participants, finds the same results at 12 weeks. The effects are real, measurable, and modest.
What dose and for how long?
Trials included in the meta-analyses generally use between 2.5 and 10 g of hydrolyzed collagen per day, and measure their effects between 8 and 12 weeks. The two largest meta-analyses support 90 days as the effective duration. Below eight weeks, no conclusions can be drawn: dermis renewal is slow. A photo at the start, same lighting, same angle, allows you to see what the daily mirror does not show.
What time of day should you take collagen?
The trials in the meta-analyses did not compare timing: they administered a daily dose at the time chosen by the participant. What matters is consistency over at least eight weeks, not the time of day. Our dedicated article on timing details what we know about peptide absorption and why consistency trumps timing.
What to remember about collagen loss after age 25?
That the loss is real and has been measured since 1975, continues throughout adult life, and that 25 is a convention, not a threshold. That fibroblasts age and receive less mechanical tension, and that sun-induced fragmentation perpetuates a vicious cycle. That menopause accelerates loss, with a measured 48% difference. That oral collagen has three meta-analyses on hydration and elasticity, at 90 days, with modest effects and noted biases. And that sun, tobacco, and sugar are the factors with the most impact.
Sources
The studies cited below were identified via PubMed.
- Shuster S, Black MM, McVitie E. The influence of age and sex on skin thickness, skin collagen and its density. British Journal of Dermatology, 1975;93(6):639-643. DOI
- Varani J, Dame MK, Rittie L, et al. Decreased collagen production in chronologically aged skin: roles of altered fibroblast function and defective mechanical stimulation. American Journal of Pathology, 2006;168(6):1861-1868. DOI
- Fisher GJ, Varani J, Voorhees JJ. Looking older: fibroblast collapse and its therapeutic implications. Archives of Dermatology, 2008;144(5):666-672. DOI
- Brincat M, Moniz CF, Studd JW, et al. Sex hormones and skin collagen content in postmenopausal women. British Medical Journal, 1983;287(6402):1337-1338. DOI
- de Miranda RB, Weimer P, Rossi RC. Effects of hydrolyzed collagen supplementation on skin aging: systematic review and meta-analysis. International Journal of Dermatology, 2021;60(12):1449-1461. DOI
- Pu SY, Huang YL, Pu CM, et al. Effects of oral collagen on skin aging: systematic review and meta-analysis. Nutrients, 2023;15(9):2080. DOI
- Dewi DAR, Arimuko A, Norawati L, et al. Impact of oral hydrolyzed collagen supplementation on skin regeneration: systematic review and meta-analysis. Cureus, 2023;15(12):e50231. DOI
