Does creatine cause hair loss? Two trials read here do not show it. A 2025 randomised trial, by its authors’ account the first to measure hair directly, gave men 5 g/day for 12 weeks and found no significant difference from placebo in DHT or any hair outcome. The 2009 study behind the worry saw DHT rise but never measured hair.
This is an evergreen explainer. On 18 October 2026 this desk read both trials’ PubMed abstracts and the full text of the 2025 paper by Lak, Forbes, Antonio, Tinsley and seven co-authors (PMID 40265319), plus the 2009 abstract by van der Merwe, Brooks and Myburgh at Stellenbosch University. They are research papers, not a ruling on any product.
Does creatine cause hair loss? The two DHT trials side by side
| van der Merwe et al. 2009 | Lak et al. 2025 | |
|---|---|---|
| Journal | Clinical Journal of Sport Medicine | Journal of the International Society of Sports Nutrition |
| Who | 20 college-aged male rugby players, in season | 45 resistance-trained men aged 18-40 recruited; 38 completed |
| Design | Double-blind, placebo-controlled crossover, 6-week washout | Double-blind, randomised, creatine group vs placebo group |
| Dose | 25 g/day creatine plus 25 g glucose for 7 days, then 5 g/day for 14 | 5 g/day creatine monohydrate vs 5 g/day maltodextrin |
| Length | 3 weeks | 12 weeks |
| DHT | Up 56% after loading; still 40% above baseline after maintenance | No significant group difference |
| DHT:T ratio | Up 36% after loading; up 22% after maintenance | No significant group difference |
| Testosterone | Did not change | Total rose and free fell in both groups, independent of creatine |
| Hair measured | No | Yes: Trichogram and FotoFinder density, follicular unit count, cumulative thickness |
That last row is the whole argument. The hair-loss worry is a chain: creatine raises DHT, DHT thins hair. The 2009 trial tested the first link only, for three weeks, starting with a loading dose five times the 5 g/day maintenance dose. Its authors concluded that further investigation was warranted. They did not report a single hair outcome, because they never measured one.
Why do the articles disagree on 56% vs 41%?
Because they quote different moments of the same trial. Cleveland Clinic’s July 2024 explainer gives the 56% rise after seven days of loading, which matches the abstract’s first figure. Examine’s FAQ gives 41% at three weeks. The abstract’s own three-week figure is 40%, after the 14-day maintenance phase. Use the abstract’s numbers: 56% at day 7, 40% at day 21.
Cleveland Clinic also says no study has been able to recreate the 2009 results. That page predates the 2025 trial, so the trial best placed to test that statement was not yet in it. Examine says that until early 2025 the 2009 study was the only trial that had tested creatine’s effect on DHT, which makes the 2025 paper the second.
Creatine and hair loss study: what the 2025 trial settles, and what it does not
At 5 g/day, the same dose as the 2009 maintenance phase, the trial found no group-by-time interaction for any hormone or hair outcome. Total testosterone rose by 124 ng/dL on creatine and 216 on placebo, and free testosterone fell in both, with the abstract calling those changes independent of supplementation. Its authors call the result “strong evidence against the claim” that creatine contributes to hair loss. Strong evidence is not proof, and the paper does not claim proof.
The limits are worth knowing before you quote it at the gym:
- Small and short. The full text says 20 were allocated to creatine and 19 to placebo, one creatine participant dropped out, and 19 per group were analysed. One placebo participant missed the final hair assessment. Twelve weeks says nothing about years of use.
- No loading phase. It did not repeat the 25 g/day week that produced the 2009 spike, so the two trials do not test the same protocol.
- Healthy trained men only. The paper describes its sample as healthy young males aged 18-40, so nobody over 40 and no women.
- Disclosed industry ties. The abstract lists one author advising Creapure and receiving creatine donations from it, another as CEO and co-founder of the ISSN and an adviser to Creapure among other brands, and a third consulting for supplement makers. Disclosed is the right word; it is not a finding against the data.
Can creatine cause hair loss in females?
Neither trial enrolled women, so neither can answer this, and anyone who extends either result to women is guessing. Our explainer on what creatine does for women covers what the research does say there. If your worry about creatine is the scale rather than the hairline, see does creatine make you gain weight, and for the ingredient itself, our creatine monohydrate review.
FAQ
Does creatine increase DHT?
In one trial, yes; in the other, no. The 2009 rugby study reported DHT up 56% after a week at 25 g/day and still 40% above baseline after two weeks at 5 g/day. The 2025 trial, at 5 g/day for 12 weeks with a placebo group, found no significant difference in DHT or the DHT-to-testosterone ratio.
If I stop taking creatine, will my hair grow back?
Neither trial tested stopping, so neither can answer that. The 2025 trial found no significant difference from placebo in hair density, follicular units or thickness, which means it found nothing to reverse. If you are shedding, the 2025 trial points away from your tub as the first suspect, and a doctor is the person to find the real one.
Is creatine safe for women?
These two trials cannot tell you. The 2009 study enrolled male rugby players and the 2025 study enrolled resistance-trained men aged 18-40, so no woman was measured in either. For what the research says about creatine across women’s life stages, read our separate explainer rather than stretching a men-only hair trial to cover it.
Nothing here is medical advice, and none of it rules on a specific tub. Two small trials, one three weeks long and one twelve, are the direct DHT evidence this desk read on the hair-loss question. If your hairline is changing, a dermatologist will look at your scalp, not your shaker bottle, and that is the right order to do it in.