Does creatine help with dementia? What the research does and doesn't show

The evidence doesn't support creatine, at any dose, as a way to prevent or treat dementia or Alzheimer's disease in people. The only human data is one small, uncontrolled pilot trial, and a 2026 scoping review searching the entire literature found no controlled trial of creatine in dementia has ever been run.

What the evidence doesn't support

It doesn't support creatine as something that prevents dementia from developing, slows the decline once it starts, or substitutes for anything a doctor would recommend for a diagnosed case. It also doesn't support a specific dose, because no controlled human trial exists to test one against. What follows is a description of the handful of studies that do exist — where they were done, in whom, and what they actually measured — not a case for or against taking creatine for this reason.

What the trials found

A 2026 scoping review searched PubMed, Embase, Scopus and the Cochrane trial registry for anything testing creatine against a cognitive outcome in older adults or in animal models of dementia. Out of 8,317 records screened, six studies qualified: one test-tube study, four animal studies, and a single human pilot. No randomized controlled trial of any kind turned up. That's the honest state of the field as of this review.

The one human study is a pilot trial run at the University of Kansas: 20 people with a diagnosis of Alzheimer's disease took 20 g of creatine a day — four times a typical daily maintenance dose — for 8 weeks, with no placebo group to compare against. Nineteen of the 20 kept up at least 80% compliance. Brain creatine, measured directly with MR spectroscopy, rose 11% by the end. Scores on some cognitive tests — a composite measure, a word-list task, an oral reading test, and an attention task — were higher at 8 weeks than at the start. Because there was no placebo arm, the researchers themselves describe this as feasibility data, not proof of an effect: with nothing to compare against, practice effects (people getting better at a test simply from taking it twice) can't be ruled out.

Animal research is where most of the actual evidence sits, and it doesn't point in one direction. Some rodent studies found creatine supported brain energy metabolism and reduced signs of neuroinflammation in models bred to develop dementia-like changes. One rodent study found the opposite for a specific memory task: spatial memory got worse, not better. A separate animal study found a sex difference worth taking seriously — female rodents showed a cognitive benefit from creatine, while males trended toward harm — and the human pilot showed a parallel split in how men's and women's brains responded biochemically. Nobody has tested whether that sex difference holds in a human dementia trial.

A 2023 review from the same research group that ran the pilot trial makes the mechanistic case for testing creatine in Alzheimer's disease — brain energy metabolism is disrupted early in the disease, and mouse studies had shown creatine supporting that system — but the review states outright that no human trial existed at the time it was written. The 2025 pilot above is the first.

Questions people ask

Does creatine help with Alzheimer's?

No trial has tested that question with a placebo group to compare against. The only human data is an 8-week pilot in 20 people that measured feasibility and brain creatine levels, not whether creatine changes the course of the disease.

Can creatine prevent dementia?

Nothing in the current research tests prevention. Every study that exists looked at people who already had a diagnosis, or at animals bred to develop dementia-like brain changes — not at whether creatine keeps a healthy brain from declining in the first place.

Is there a creatine and dementia study people can point to?

Yes, one: a 2025 pilot trial in 20 Alzheimer's patients, run without a placebo group, testing 20 g a day for 8 weeks. It's described in detail above, including exactly what it did and didn't measure.

What dose of creatine was used in the Alzheimer's pilot?

20 g a day — about four times the 3-5 g maintenance dose most fitness research and product labels use. No lower or higher dose has been tested in a dementia population, so there's nothing to compare it against.

When to talk to a doctor

Memory changes, confusion or a suspected dementia diagnosis are a reason to see a doctor, not to start a supplement. Creatine is not a treatment for dementia or Alzheimer's disease, and nothing in the small amount of research above changes that. If you or someone you're caring for is showing signs of cognitive decline, a medical evaluation — not a supplement aisle — is the right next step.

Related reading: creatine benefits: what the research does and doesn't show, what is creatine monohydrate, and creatine dosage.