Creatine: What It Actually Does for Muscle, Brain and Aging

Creatine spent thirty years as the unglamorous powder at the back of the gym bag — well studied, cheap, and boring. In about two years it has been rebranded as a brain supplement, a menopause intervention, and an anti-aging compound. Sales spiked. So did the claims.

Some of that is deserved. Creatine is one of a very small number of supplements with genuinely good evidence behind it, and unlike most of the shelf it doesn't need a marketing department to look respectable. But the gap between what the muscle research supports and what the brain research supports is much wider than the current enthusiasm suggests — and researchers have started saying so in print.

Here is the claim-by-claim version, with the sample sizes attached.

What creatine actually is

Creatine is a compound your body already makes — roughly a gram a day, in the liver and kidneys — and that you eat in meat and fish, at maybe one to two grams a day on a typical omnivorous diet. Around 95% of it sits in skeletal muscle as phosphocreatine, where it functions as a rapid-recharge battery: it donates a phosphate group to regenerate ATP during the first several seconds of hard effort, before other energy systems catch up.

Supplementing raises muscle stores by roughly 20–40%, depending on where you started. That "depending on where you started" is the most useful sentence in this article. People with low baseline stores — vegetarians and vegans above all, since dietary creatine comes almost entirely from animal foods — tend to show the largest responses. People already eating a lot of meat have less room to move, which is part of why individual results vary so much and why some trials find effects that others miss.

Muscle: the strong case, honestly sized

This is where creatine earned its reputation, and it deserves it. But two meta-analyses of the same question reached noticeably different conclusions, and the more recent one is the more careful.

The 2017 analysis pooled 22 randomised trials covering 721 participants aged 57 to 70, training two to three days a week for anywhere from 7 to 52 weeks. Creatine plus resistance training produced 1.37 kg more lean tissue mass (95% CI 0.97–1.76) than training with placebo, along with greater gains in chest press and leg press strength.

The 2026 analysis looked at 11 randomised trials in adults 60 and over, used a three-level model that accounts for multiple outcomes drawn from the same study, and graded the certainty of each result using GRADE. It found:

Compared with resistance training alone. A confidence interval that crosses zero means the trials, taken together, cannot rule out no effect — it does not mean the effect is absent, only that the evidence isn't strong enough to claim it.

Read those two together rather than picking the one you like. Creatine's strength benefit on top of training is consistent and modest. Its effect on actual muscle mass is probably real but not nailed down. Its effect on how well an older person walks, rises from a chair and climbs stairs — the thing that matters most — has not been demonstrated to statistical significance yet.

And every one of these findings is creatine plus resistance training. If you are taking creatine and not lifting, you are taking the smaller half of the intervention and skipping the larger one. Our strength training guides are the other half of this page.

Bone: the finding nobody talks about

A 12-month randomised trial put postmenopausal women on supervised resistance training three days a week with either creatine at 0.1 g/kg/day or placebo. Forty-seven were randomised; 33 completed and were analysed — a small study, and worth holding loosely.

The creatine group lost femoral neck bone density at −1.2% over the year, against −3.9% in the placebo group, and showed an increase in femoral shaft periosteal width, a structural measure that predicts bone bending strength. Hip fracture is one of the genuinely life-altering events of later life, and the femoral neck is where it happens.

One trial is not a body of evidence, and this needs replication before anyone treats it as established. But it's a more interesting result than most of what creatine is currently being marketed for, and it is a reason the case for creatine is arguably stronger in postmenopausal women than in the young men it was originally studied in — a thread we pick up in creatine for women.

The brain story, and why it's being oversold

It is worth spelling out what happened here, because it's a useful case study in how a health claim inflates.

The underlying biology is genuinely plausible. Your brain uses a lot of energy, it runs the same phosphocreatine system your muscles do, and brain creatine levels drop under metabolic stress. If you accept that, "creatine might help the brain under stress" is a reasonable hypothesis rather than a marketing invention.

Then the evidence arrived, and it was thin. The 2024 sleep-deprivation study is the strongest piece and it used a single 25-gram dose in an acute setting. The systematic review that gets cited as proof of cognitive benefit in older adults contained six studies, four of which were dietary questionnaires, and three of the six were rated below "fair" on methodological quality. A memory meta-analysis drew a letter alleging its statistics double-counted participants. A cognition meta-analysis needed a corrigendum. The aging review drew a formal objection.

None of that means creatine does nothing for the brain. It means the evidence currently cannot carry the weight being put on it — which is precisely the argument the Journal of Nutrition published in 2025, calling out the gap between the sales figures and the science.

How to actually take it

If you want specific products, our creatine buying guide covers what to look for — but the short version is that monohydrate is monohydrate, and third-party testing matters more than branding.

References

Every figure above traces to one of these. Links go to the published record.