Walk down any supplement aisle and collagen has taken over half the shelf. The pitch is always the same: your collagen declines after 40, so take this and get it back. Skin, hair, nails, joints, “glow” — one powder promising everything.

Here’s what that pitch quietly leaves out. Collagen is not one thing with one effect. The strongest evidence for it is narrow, specific, and comes with conditions attached. The evidence for the thing people actually buy it for — bigger muscles — is not just weak, it’s negative.

A 2026 systematic review of randomized controlled trials put hard grades on this. And the grades are the most useful thing in the whole collagen conversation.

No, we’re not selling you a tub. Every “best collagen” article on the first page is written by someone who sells collagen. This one walks through the actual trials, in both directions, so you can read the primary sources and decide for yourself.

Collagen for tendons and joints: the short answer

Collagen supplementation has real, evidence-graded benefits for tendon and joint tissue — under three conditions that almost no marketing mentions:

  1. Dose of 15–30 g per day. Lower doses (~5 g) did not produce between-group differences.
  2. Taken with vitamin C (≥50 mg). Vitamin C is a required cofactor for collagen synthesis.
  3. Combined with heavy resistance training (≥70% of 1RM). Without mechanical load, the supplementation does very little.

And the flip side, stated just as plainly: collagen does not build muscle. The same review found strong evidence (GRADE A) against any additive effect of collagen on muscle strength. If you’re taking collagen expecting it to do what whey does, you’re buying the wrong thing.

What collagen actually is (and why that matters after 40)

Collagen is the most abundant protein in your body — roughly a third of your total protein mass. It’s the structural scaffolding: ~65–80% of the dry weight of a tendon, two-thirds of the dry weight of the cartilage in your knees.

Type I is the big one — tendons, ligaments, bone, skin. Type II is about 90% of the collagen in cartilage. Type III sits in skin and blood vessels. Those type numbers matter, because different supplements target different ones.

Production declines with age. It starts dropping around 25–30, and after 40 you lose roughly 1% per year. By 80, you’re working with a fraction of what you had. That decline is part of why tendons stiffen, why joints get noisier, and why recovery from a hard session takes longer than it did at 25.

So the logic of supplementation seems obvious. But there’s a catch that the industry prefers not to dwell on: the collagen in a powder is not the collagen in your tendon. You digest what you swallow into amino acids and short peptides. Whether those get reassembled into new tendon tissue is a separate question — and one that turned out to depend heavily on what you do with your body while taking it.

The evidence for tendons — and the dose that decides it

This is where collagen has its strongest case, and it’s genuinely better than most people assume.

The Buchalski et al. (2026) systematic review in the Journal of Functional Morphology and Kinesiology pooled eight randomized controlled trials (n = 257). Every one of them combined supplementation with resistance or plyometric training, over 3–15 weeks. Here’s what came back:

✅ Tendon cross-sectional area (CSA)      GRADE A  for
   — 3 of 4 studies showed significantly greater
     increases in collagen groups vs placebo

✅ Tendon stiffness / Young's modulus      GRADE A  for
   — but ONLY at higher doses (15–30 g/day)
   — lower-dose studies (~5 g) showed within-group
     effects only, no between-group difference

❌ Muscle strength                         GRADE A  AGAINST
   — strength improved in every trial, but collagen
     added nothing beyond the training itself

❓ Muscle CSA / physical performance        GRADE C  (conflicting)

That dose split is the single most important detail in the post. The Jerger et al. (2022) trial in the Scandinavian Journal of Medicine & Science in Sports used 5 g/day of specific collagen peptides with 14 weeks of resistance training and did find greater Achilles tendon CSA gains than placebo. But the stiffest evidence for stiffness and modulus came from the 15–30 g studies. At ~5 g, you’re in the range where effects are inconsistent.

The mechanism people use to explain this is mechanotransduction: loading a tendon triggers a signalling cascade in the tendon cells that drives matrix protein production. Collagen supplementation supplies the amino-acid precursors. Load provides the signal. Take the powder away and the signal still fires but the raw material is thinner. Take the load away and you’re delivering bricks to a site where nobody’s building. That’s the whole post, honestly — load is the active ingredient, collagen is the supply line.

One more honest note: the review found improvements in eccentric rate of force development and deceleration impulse in one study — but no effect on gross explosive metrics like jump height. Tendon adaptations are real, but subtle, and they don’t automatically show up on the scoreboard.

The evidence for joints — weaker than you’d think

Joints are the other half of the collagen pitch, and here the picture is muddier.

The Martínez-Puig et al. (2023) review in Nutrients covers collagen for joint health and finds genuine improvements in joint pain and function in several trials — including trials with undenatured type II collagen (UC-II) at doses as low as 40 mg/day. That’s a striking contrast to the gram-level doses used for tendons: a completely different supplement format working on a completely different pathway, targeting cartilage rather than tendon.

But it would be dishonest to stop there. STAT News ran a blunt review of the collagen-for-arthritis evidence and flagged the same pattern that shows up any time a supplement gets big money behind it: the smallest, shortest trials reported the largest benefits. Trials running 10–13 weeks showed bigger effects than trials running 24–48 weeks. Trials using 2–10 g/day claimed larger effects than trials using 10–40 g/day. And in the largest trials, symptom improvement landed under five points on a 100-point scale — a difference patients describe as feeling “much the same.”

A signal that shrinks as study quality rises is not a strong signal. And as STAT pointed out, the absence of one large, well-run trial is itself informative: the companies selling this product have the money to run one, and haven’t.

So: joint evidence exists, is directionally positive, and is notably softer than the tendon evidence. Treat it as plausible, not settled.

So what should a man over 40 actually do?

If you’re going to supplement collagen, do it in the only configuration the evidence supports:

1. Get the dose right, or don’t bother. 15 g/day minimum, and don’t expect ~5 g to do the job it can’t do. This is also the honest reason collagen is expensive: you’re buying grams, and you need a lot of them.

2. Take it with vitamin C. The review specifies ≥50 mg alongside. A glass of orange juice or a standard vitamin C tablet clears this easily. It’s a cofactor for collagen synthesis, not a bonus.

3. Time it around heavy loading. Take it 30–60 minutes before training, then actually train heavy — ≥70% of your 1RM. The 2026 review’s conclusion is explicit that the benefit appears when combined with high-intensity resistance training. Collagen taken while sitting on the couch is expensive urine with a nice label.

4. Give it months, not weeks. Tendon tissue remodels slowly. The trials that worked ran 12–15 weeks and often longer. Tendons have poor blood supply and adapt at a fraction of the speed of muscle. Anything promising tendon results in two weeks is selling something else.

5. Know what you’re not buying. Not muscle. If your goal is hypertrophy, the money and the effort both point at hitting your protein target with a complete protein — and if you’re weighing powders, the beef-versus-whey evidence is the place to check the label. Collagen is a targeted tool for connective tissue, not a general protein.

The decade framing

There’s a real, specific, unglamorous win here for men in their 40s and 50s: your tendons are what keep you lifting for the next thirty years, and tendon injuries are the thing that most reliably ends a training decade early. A tweaked Achilles or a grumbling elbow at 45 can cost you six months. That’s a bigger threat to your long game than a slightly suboptimal protein powder.

So the value of collagen isn’t the “glow.” It’s that heavy loading plus adequate supply might let you keep loading heavy — which is how you keep muscle, bone, and capacity into your 60s. The collagen isn’t the intervention. The training is the intervention, and collagen is one input that helps the connective tissue keep up with it.

And that will keep being true whether or not the next bottle promises a better glow. Buy grams, take vitamin C, lift heavy, wait a season. Skip the rest of the story.

Sources

  • Buchalski A, Jeanfavre M, Altorelli C, Leff G. (2026) — “Collagen Supplementation on Tendon-Related Structural and Performance Outcomes: A Systematic Review.” Journal of Functional Morphology and Kinesiology, 11(1):130. PMID 41900537. Eight RCTs (n = 257). GRADE A for tendon CSA and stiffness at 15–30 g/day with vitamin C ≥50 mg and RT ≥70% 1RM; GRADE A against an effect on muscle strength; GRADE C for muscle CSA and physical performance.
  • Jerger S, et al. (2022) — “Effects of specific collagen peptide supplementation combined with resistance training on Achilles tendon properties.” Scandinavian Journal of Medicine & Science in Sports. 14-week trial; 5 g/day specific collagen peptides produced significantly greater Achilles tendon CSA gains than placebo.
  • Martínez-Puig D, et al. (2023) — “Collagen Supplementation for Joint Health: The Link between Composition and Clinical Evidence.” Nutrients. Reviews joint pain/function improvements, including undenatured type II collagen (UC-II) at 40 mg/day.
  • Collagen decline with age — Collagen production begins declining around age 25–30 at approximately 1% per year after 40 (Arthritis Foundation, “Are Collagen Supplements Helpful for Arthritis?”).
  • Tendon composition — Collagen constitutes approximately 65–80% of tendon dry weight (Kannus 2000; Goes et al. 2020, cited in the systematic review literature).
  • Mechanotransduction — Mechanical loading of tendon tissue creates a signalling cascade increasing matrix protein production and tendon hypertrophy (Svensson et al. 2016).
  • Counterweight — the soft joint evidence — STAT News (2022), “The painfully weak evidence on taking collagen to treat arthritis”: small/short trials reported larger effects than large/long trials, and the largest trials found improvements under 5 points on a 100-point scale.