On this page
Key takeaways
● The König 2018 RCT (131 postmenopausal women with reduced BMD) showed statistically significant BMD gains at the lumbar spine and femoral neck after 12 months of 5 g/day collagen peptides versus placebo (1).
● The dose is modest — 5 g/day. Higher doses have not been shown to produce more effect for bone.
● Bone turnover markers moved in the expected direction: P1NP (formation) up, CTX-1 (resorption) down.
● A four-year open-label follow-up replicated the trend, with progressive BMD gains.
● Non-substitute rule: hydrolysed collagen is not a substitute for prescribed osteoporosis therapy (bisphosphonates, denosumab, romosozumab, SERMs) in diagnosed disease.
● Independent replication is limited. Most of the bone literature funnels through one research programme funded by Gelita.
Quick answer
Yes, for postmenopausal bone density, at 5 g/day for 12 months, based on one well-conducted RCT with a four-year follow-up. The König 2018 trial showed measurable BMD gains at the lumbar spine and femoral neck in postmenopausal women with T-score ≤ −1. The result is striking because the dose is modest and the outcome is objective (DXA-measured BMD, not subjective ratings). The main caveats: the evidence base is narrow (mostly one research programme, industry-funded) and hydrolysed collagen is not a substitute for prescribed osteoporosis therapy in diagnosed disease. It is a reasonable adjunct to vitamin D, calcium, weight-bearing exercise, and — where indicated — pharmacotherapy.
The anchor trial — König 2018 in detail
Almost the entire clinical evidence base for hydrolysed collagen and bone density rests on one well-conducted trial published in Nutrients in 2018 by Denise König and colleagues at the University of Freiburg (1). It is worth walking through in detail because the result is unusual and because it is what supplement marketing quotes.
Design. 131 postmenopausal women with reduced bone mineral density (T-score ≤ −1 at either the lumbar spine or femoral neck) were randomised to either 5 g/day of specific collagen peptides or 5 g/day of matched maltodextrin placebo. Supplementation continued for 12 months. Blinding was maintained throughout. The primary outcome was change in bone mineral density measured by DXA scanning at the lumbar spine and femoral neck. Secondary outcomes included bone-formation marker P1NP and bone-resorption marker CTX-1.
Result. The collagen group showed statistically significant increases in BMD versus placebo at both the lumbar spine and the femoral neck over 12 months. The bone turnover markers moved in the expected direction: P1NP (a marker of osteoblast activity and new collagen formation in bone matrix) rose in the collagen group; CTX-1 (a marker of bone resorption by osteoclasts) declined. The combination of BMD gains plus favourable shifts in independent turnover markers strengthens the interpretation that collagen supplementation was affecting real bone metabolism, not just DXA measurement noise.
Follow-up. A four-year open-label extension of the trial reported progressive BMD gains in participants who continued supplementation, though open-label follow-ups carry different methodological considerations than the primary blinded phase.
Why the dose is worth noting — 5 g/day was enough
Two things about this result are unusual. First, the dose. The trial used 5 g/day — a modest amount, roughly half the 10 g scoops that dominate commercial products. Higher doses have not been shown to produce more effect for bone outcomes. This is one indication where the marketing pressure toward larger scoops is not evidence-aligned.
Second, the outcome. BMD measured by DXA scanning is a hard, quantitative outcome much less susceptible to expectation bias than subjective skin hydration ratings or self-reported pain scales. Combined with the objective turnover markers, this makes the result harder to dismiss as a placebo effect than would be the case with softer endpoints. If you are supplementing collagen for postmenopausal bone specifically, 5 g/day is what the evidence supports — not the standard 10 g scoop. This can meaningfully affect cost.
The mechanism — signalling to osteoblasts
Bone is a dynamic tissue in constant turnover, with osteoblasts building new bone matrix and osteoclasts resorbing existing bone. In healthy young adults, formation and resorption are balanced. In postmenopausal women, oestrogen decline accelerates resorption, tipping the balance and producing net bone loss over time. Osteoporosis therapy works by either suppressing resorption (bisphosphonates, denosumab) or stimulating formation (romosozumab, teriparatide).
The proposed mechanism for hydrolysed collagen operates on the formation side. Ingested collagen peptides — chiefly Pro-Hyp and Hyp-Gly dipeptides that survive intestinal digestion — reach the bloodstream and can distribute to bone tissue. There, they appear to stimulate osteoblast differentiation and activity, upregulating the transcription factor Runx2 (the master regulator of osteoblast differentiation) and increasing collagen matrix synthesis. Bone organic matrix is roughly 90% collagen; more matrix formation supports more mineralisation. The P1NP rise and CTX-1 decline observed in König 2018 are consistent with this mechanism operating in the expected direction.
The peptides themselves are not incorporated into bone. They are signalling to bone cells to do more of their normal work.
The independent-replication caveat
The evidence base for hydrolysed collagen and bone is essentially one research programme. The König 2018 trial was funded by Gelita AG, the ingredient supplier that manufactures the specific collagen peptides used in the trial. Several follow-up trials from the same Freiburg research group have reported related outcomes (2), but broad independent replication by unrelated research groups has been limited.
This does not mean the result is wrong. Well-conducted industry-funded trials often replicate. But a single research programme is a fragile foundation for a clinical claim, even when the trial is well-conducted. A reader relying on hydrolysed collagen for bone protection should hold the confidence level appropriately — this is a promising, striking result, not a settled finding. One partial replication of the direction, in a different context: Argyrou 2020 examined collagen peptides added to standard calcium and vitamin D in postmenopausal women with osteopenia and reported favourable bone turnover marker changes (5).
The non-substitute rule — critical to internalise
This section requires special emphasis because misreading it could produce real clinical harm.
Hydrolysed collagen is not a substitute for prescribed osteoporosis therapy in patients with diagnosed osteoporosis (T-score ≤ −2.5) or with established fracture risk. Bisphosphonates (alendronate, risedronate, zoledronate), denosumab, romosozumab, and SERMs (raloxifene) deliver fracture-risk reduction of a magnitude that hydrolysed collagen cannot match. The König 2018 BMD gains, while striking, are smaller than typical pharmacotherapy effects and — critically — have not been validated against fracture endpoints. BMD is a surrogate marker; what actually matters clinically is whether you have a hip fracture at 75. That endpoint has not been tested for collagen.
Where collagen fits: as an adjunct to standard care in postmenopausal women, particularly in osteopenia (T-score between −1 and −2.5), where the König 2018 study population sits and where pharmacotherapy may not yet be indicated. Vitamin D, calcium, weight-bearing exercise, and — where clinically indicated — pharmacotherapy remain the foundation (4). Collagen at 5 g/day is a reasonable additional intervention on top of that foundation.
If you have diagnosed osteoporosis and are considering discontinuing prescribed pharmacotherapy in favour of collagen, do not. Discuss any change to your osteoporosis regimen with the prescribing physician. This is not a hypothetical concern; the marketing sometimes implies that collagen is a natural alternative to "harsh drugs." It is not an equivalent alternative.
Where hydrolysed collagen sits in the bone-care lever stack
● Vitamin D + calcium + weight-bearing exercise — the foundational trio. All bone RCTs, including König 2018, were conducted against a background of adequate calcium and vitamin D. Do this first.
● Impact and resistance exercise — meta-analyses show meaningful BMD effects from consistent progressive resistance training and impact-loading exercise. Larger effect sizes than any supplement.
● Prescribed osteoporosis pharmacotherapy where clinically indicated. Non-substitutable by supplementation.
● Hydrolysed collagen at 5 g/day — a legitimate adjunct on top of the above. Evidence-aligned dose. Modest expected effect. Not a substitute.
● Other supplements with weaker bone evidence — magnesium, vitamin K2, isoflavones. Individual reasoning.
Dose and duration for bones
The trial-validated dose is 5 g/day. There is no clear evidence advantage for higher doses in bone outcomes specifically. If bone is your primary reason for supplementing, 5 g/day is what the evidence supports; a standard 10 g scoop can effectively be split (5 g in the morning, 5 g not used) or a smaller-scoop product chosen. If you are supplementing with combined bone, skin, and joint goals, 10 g/day covers all indications reasonably.
Duration is what makes bone different from skin and joints. Bone remodelling operates on a slow timescale. Meaningful BMD change requires 12 months of consistent daily supplementation, per König 2018. Turnover markers may shift earlier (weeks to months), but the DXA-measurable BMD change is a year-long protocol. Discontinuing after three or six months because "nothing feels different" would be premature.
Who this is reasonable for
● Postmenopausal women with osteopenia (T-score between −1 and −2.5), as an adjunct to vitamin D, calcium, and weight-bearing exercise. This is the König 2018 population and the strongest candidate group.
● Postmenopausal women with normal BMD who want to slow age-related bone loss as a proactive measure. Reasonable, with the caveat that trial evidence in this specific population is thinner.
● Women approaching menopause who want to build bone reserve. Mechanistically plausible; trial evidence in this transition population is limited.
● Older adults of either sex with age-related bone density decline, particularly those with fracture history. The König 2018 evidence is postmenopausal-female-specific, so applying it to older men is extrapolative but not unreasonable given the mechanism.
Where this is not the primary intervention: anyone with diagnosed osteoporosis on prescribed pharmacotherapy (collagen is an adjunct, not a substitute); anyone whose bone-care foundation (vitamin D, calcium, weight-bearing exercise) is not yet in place (address that first).
What we still don't know
● Whether the König 2018 result replicates broadly outside the Freiburg research programme. Independent replication remains limited.
● Whether the BMD gains translate to fracture risk reduction. This is the clinically important endpoint. It has not been tested for collagen.
● Whether the effect applies as strongly to older men, premenopausal women, or non-osteopenic populations. Trial evidence is postmenopausal-osteopenic-female-specific.
● Whether combining collagen with prescribed osteoporosis pharmacotherapy produces additive benefit or the drugs alone are sufficient. Not tested.
● What happens on discontinuation after multi-year supplementation. Whether BMD gains persist, revert, or plateau is unknown.
Bottom line
Hydrolysed collagen at 5 g/day for 12 months produces measurable bone mineral density gains in postmenopausal women with osteopenia, based on one well-conducted RCT with a four-year open-label follow-up. The dose is modest, the outcome is objective, and the bone-turnover markers move in the expected direction. The main caveats are that the evidence base rests largely on one research programme and that BMD is a surrogate for fracture risk, which has not been tested. Take collagen at 5 g/day as an adjunct to the bone-care foundation — vitamin D, calcium, weight-bearing exercise, and prescribed pharmacotherapy where indicated. It is never a substitute for prescribed osteoporosis treatment in diagnosed disease. For the full pillar picture, see our complete guide.
Frequently asked questions
Does hydrolysed collagen actually help bones?
Yes, based primarily on the König 2018 RCT (1). Postmenopausal women taking 5 g/day for 12 months showed statistically significant BMD gains at the lumbar spine and femoral neck versus placebo. A four-year follow-up reported progressive gains. Independent replication is limited. Take this as promising evidence, not settled.
How much collagen should I take for bone density?
5 g/day is the trial-validated dose. Higher doses have not been shown to produce more bone effect. If you take a standard 10 g scoop for other goals, you are still covered; if bone is your primary reason, 5 g is what the evidence supports and can meaningfully affect cost.
How long until hydrolysed collagen affects bone density?
12 months. Bone remodelling operates slowly, and König 2018 assessed at 12 months. Turnover markers may shift earlier. Discontinuing before a year would not give the intervention a fair test.
Can hydrolysed collagen replace my osteoporosis medication?
No. Hydrolysed collagen is not a substitute for prescribed osteoporosis therapy in diagnosed osteoporosis. Bisphosphonates, denosumab, romosozumab, and SERMs deliver fracture-risk reduction of a magnitude collagen cannot match. Discuss any change to your osteoporosis regimen with your prescribing physician.
Is hydrolysed collagen useful for men with low bone density?
Mechanistically plausible; trial evidence is largely postmenopausal-female-specific. Extending to older men or women pre-menopause is reasonable extrapolation with lower confidence. Consider as an adjunct to standard bone-care recommendations rather than a primary intervention.
Do I need to take collagen with calcium and vitamin D?
The König 2018 trial was conducted against a background of adequate calcium and vitamin D intake. If your diet does not cover these adequately, address that first — the foundation matters more than the adjunct. Collagen on top of the foundation is where the evidence supports it, not as a stand-alone bone strategy.
Does the type of collagen matter for bone?
Bovine collagen (Type I and III) is what the König 2018 trial used. Marine collagen is chemically similar and mechanistically comparable, though bone-specific trials with marine collagen are limited. Type II collagen (UC-II) is a different product for a different indication (joint cartilage) and does not apply here.
References
1. König D, Oesser S, Scharla S, Zdzieblik D, Gollhofer A. Specific collagen peptides improve bone mineral density and bone markers in postmenopausal women. Nutrients 2018. https://pubmed.ncbi.nlm.nih.gov/29337906/
2. Zdzieblik D, Jendricke P, Oesser S, Gollhofer A, König D. The influence of specific bioactive collagen peptides on body composition and muscle strength in middle-aged, untrained men: a randomized controlled trial. J Int Soc Sports Nutr 2021. https://pubmed.ncbi.nlm.nih.gov/34965865/
3. Kumar S, Sugihara F, Suzuki K, Inoue N, Venkateswarathirukumara S. A double-blind, placebo-controlled, randomised, clinical study on the effectiveness of collagen peptide on osteoarthritis. J Sci Food Agric 2015. https://pubmed.ncbi.nlm.nih.gov/24852756/
4. Cosman F, de Beur SJ, LeBoff MS, et al.. Clinician's guide to prevention and treatment of osteoporosis. Osteoporos Int 2014. https://pubmed.ncbi.nlm.nih.gov/25182228/
5. Argyrou C, Karlafti E, Lampropoulou-Adamidou K, et al.. Effect of calcium and vitamin D3 supplementation with and without collagen peptides on bone turnover in postmenopausal women with osteopenia. J Musculoskelet Neuronal Interact 2020. https://pubmed.ncbi.nlm.nih.gov/32117065/