08:43
creatine
The claim that creatine causes hair loss traces back almost entirely to a single small, unreplicated 2009 hormone study that never even measured hair; the one trial that has since directly tested hair outcomes (April 2025) found no effect at 5g/day over 3 months, so the video concludes current evidence does not support creatine causing hair loss, while leaving higher-dose and longer-duration exposure as open questions.
The hair-loss claim originates from a 2009 randomized trial of 20 college-age male rugby players given creatine (25g/day loading week, then 5g/day maintenance for two weeks) or placebo.
In that 2009 trial, total testosterone did not change, but DHT (dihydrotestosterone, a testosterone byproduct) rose 56% after the loading week and remained 40% elevated after the full 3 weeks.
Mechanistically, testosterone is converted to DHT by an enzyme; DHT binds the androgen receptor in hair follicles, causing miniaturization (thinner, weaker hair, shortened growth cycle) — but only in people with genetic susceptibility to male pattern baldness.
The 2009 trial never measured hair itself — no hair counts, density, or thickness — only blood hormone levels.
Even with the reported increase, all DHT values in the 2009 trial stayed within the normal clinical range (about 0.98 rising to 1.5 then 1.4, versus a normal range up to 2–3).
The DHT elevation reported in 2009 has never been reproduced by any other trial, as far as the video's research found.
A 2021 paper reviewing frequently asked questions on creatine concluded there is no convincing evidence linking creatine to hair loss.
That 2021 paper raised 'regression to the mean' as a possible statistical explanation: the placebo group started with numerically higher DHT (1.26) than the creatine group (0.98) before the trial began, a non-significant difference that may have exaggerated the apparent divergence over time.
As of 2021, the causal question had never been tested directly — the field lacked both strong evidence for and against the hair-loss link.
In April 2025, the first (and so far only) trial to directly test hair outcomes was published: a randomized controlled trial of 38 men given 5g/day creatine or placebo for 3 months.
The 2025 trial found no significant difference between groups in free testosterone, total testosterone, DHT, or hair parameters (hair count, density, thickness).
The 2025 trial ran 3 months — four times longer than the original 2009 rugby trial — strengthening its evidentiary weight.
A key limitation of the 2025 trial: it used a flat 5g/day dose throughout and did not replicate the original 25g/day loading-week protocol from the 2009 rugby trial, leaving open whether a higher loading dose could have an effect.
The video frames its overall conclusion as provisional: the only direct trial to date shows no effect at 5g/day for 3 months, but higher doses or longer exposure remain untested.
The video is an explicit follow-up to a prior video on creatine (covering muscle mass, fat mass, strength, and kidney safety), prompted by viewer questions about hair loss.
The entire hair-loss hypothesis rests on one small, unreplicated data point (a hormone-only measurement in 20 rugby players) that never actually assessed hair — a striking gap between the strength of the underlying evidence and how widely the claim circulated.
The 'regression to the mean' explanation offered by the 2021 paper reframes the 2009 study's headline finding (a 56% DHT increase) as potentially an artifact of unlucky baseline randomization rather than a true drug effect.
Percentage increases can be misleading in isolation: DHT rose 56%, but the absolute values throughout the 2009 trial stayed within the normal physiological range — a distinction the video treats as important to the interpretation.
The 2025 trial, though designed largely as a direct response to the 2009 rugby trial, curiously did not replicate that trial's high-dose loading phase — so it tests a milder protocol than the one that originally raised the concern, meaning the specific loading-dose scenario is still unaddressed.
The video treats the underlying question as still scientifically open on dose and duration, resisting a flat 'no' despite the negative 2025 result — the negative finding narrows the claim rather than closing it.
«So, we published a video on creatine last week, and a lot of you guys asked about creatine and hair loss. Does creatine supplementation cause hair to fall out?»
— 00:00
«And so the whole link to hair loss has to do with these hormones because testosterone itself doesn't cause baldness. But this DHT, dihydrotestosterone, has been linked to male pattern baldness.»
— 01:36
«One thing to point out is in this trial, they never reported any hair measurements. They didn't look at whether they were losing hair or not. They just looked at the hormone levels.»
— 02:52
«the normal range of DHT goes all the way up to two or three depending on the lab. So all of these values are well within the normal range.»
— 03:20
«this paper concludes that current evidence does not indicate that creatine raises the total or free testosterone levels or DHT or causes hair loss or baldness.»
— 05:35
«And at the end of the three months there was no significant difference in free testosterone levels, total testosterone levels, DHT levels or hair parameters.»
— 06:41
«this is a little annoying because this trial is designed largely to test this idea coming from the rugby trial, largely as a response to that rugby trial. It would have been nice if they did it the exact same way, right?»
— 07:35
«So the conclusion is the only test we have so far and it's a randomized trial looking at this question doesn't show an effect of creatine supplementation on hair loss at least with five grams a day for 3 months.»
— 08:02
Reception
Audience shares genuine hair loss concerns from personal experience but appreciates the scientific analysis and thoughtful discussion of confounding variables.
The video builds its case incrementally and transparently, surfacing its own key source's weaknesses (no hair measurements, unreplicated finding, possible regression-to-the-mean artifact, and a dosing mismatch in the follow-up trial) rather than asserting a simple verdict, and lands on 'no compelling evidence' while explicitly flagging higher doses and longer exposure as untested.
08:43