Creatine Side Effects: What Is Real, What Is Water, and What Is One Study

Creatine side effects in healthy adults are stomach upset from a large single dose and a rise in body water. Kreider, Mol Cell Biochem 2003, tracked 98 athletes for 21 months on a 69-marker panel and reported no adverse effect on health markers. That trial was open-label, and the hair question rests on one 2009 study of 20 men.

Most of what follows is reassuring, which is an awkward thing for a page like this to be. So the part that is not reassuring goes first.

The hair question has not been settled and this page is not going to tell you it has. One study found a hormone change and did not look at hair. One study looked at hair, in 38 men, for twelve weeks. Neither of those is an answer, and the sections below say so in the same words we would use if the finding went the other way.

And one thing on this page is not for sale everywhere: Base Work does not ship to New York addresses. The reason and the mechanics are in the shipping policy, and they are restated at the bottom of this page.

The short version, effect by effect

Every row below is expanded further down with its sample size, its population and its funding attached. Read the table for the shape and the sections for the caveats, because on this subject the caveats are most of the content.

Does creatine make you bloated?

Two different things get called bloating and only one of them is creatine. Gastrointestinal fullness is almost always a large single dose sitting undissolved, and it responds to splitting the dose, warmer water and food. The other is water. Creatine draws water into the muscle cell, and a 2009 systematic review found no reduction in total body water as a proportion of body mass.

The distinction matters because the two have opposite fixes. A stomach that objects to 20 g in one hit is answered by taking 5 g four times instead, or by not loading at all. Water inside a cell is not answered by anything, because it is not a fault.

Base Work is a 5 g serving. A loading protocol is four of those in a day — roughly 0.3 g per kilogram of bodyweight, split, for five to seven days, which is the figure the 2017 position stand in J Int Soc Sports Nutr prints. Nearly every stomach complaint filed against creatine was filed against that week rather than against the daily serving that follows it.

You do not have to run that week at all. J Appl Physiol, 1996: 20 g a day for six days and 3 g a day for 28 days produced the same roughly 20% rise in muscle creatine. Loading is a shortcut, not a requirement, and skipping it removes the commonest cause of the complaint this section is about. The full comparison is in how much creatine per day.

Does creatine make you hold water, and where does the water go?

Across 61 controlled trials pooled in 2025, creatine taken alongside resistance training raised body mass by 0.89 kg (95% CI 0.76 to 1.01). Part of what those trials record as tissue is intracellular water, because creatine draws water into the muscle. That caveat is mandatory in our own citation bank, and it is the honest content of the phrase water retention.

Two things follow from it, and the second is the one that gets left out.

The first: the scale moves, and it moves early. Under a kilogram, on average, across sixty-one trials. If a number on a scale is the measurement you are steering by, that is the size of the step you should expect and it is not a change in fat.

The second: it is not dehydration. The systematic review with meta-analyses in J Athl Train, 2009 — 10 qualifying studies out of 95 screened — found creatine did not reduce total body water as a proportion of body mass. Water moving into the cell is not water leaving the body, and the two get conflated constantly. That review is unpacked in does creatine dehydrate you.

One disclosure belongs on the 2025 pooled figure, because it is the kind of thing we would want to know. Two of its authors are long-standing figures in the sports-nutrition society whose position stands are quoted elsewhere on this page, with extensive documented supplement-industry research funding across their careers. The finding is a pooled estimate from 61 trials rather than one team's result, which is why it is here; the disclosure travels with it anyway.

Is creatine bad for your kidneys?

Creatine converts spontaneously into creatinine, which is the exact number a kidney panel reports — so a higher reading is expected on creatine and does not by itself mean kidney damage. A 2019 meta-analysis in J Renal Nutr found creatine did not alter plasma urea, and a 2025 systematic review in BMC Nephrol reached the same conclusion for healthy populations. This is about healthy adults only. Tell whoever ordered the blood test that you supplement.

Almost every version of this fear starts with a printout rather than a symptom. Someone takes creatine, has routine bloods done, and the creatinine line comes back above the reference range with a flag beside it.

Here is the mechanism, in one sentence, because it is the sentence nobody prints. Creatinine is the breakdown product of creatine. Taking more creatine puts more creatinine into circulation, so the number the laboratory reports goes up. The number went up because of what you swallowed, not because the filter changed.

That is not a reassurance we invented. The 2019 meta-analysis — de Souza e Silva and colleagues, J Renal Nutr 2019;29(6):480–489, 290 non-duplicated studies screened, 15 carried into qualitative synthesis and 6 into quantitative — found no change in plasma urea, and records the creatinine rise as an artefact of supplementation rather than evidence of dysfunction. A newer systematic review in BMC Nephrol in 2025 reached the same overall conclusion for healthy populations.

Three limits on that, and they are the whole of the qualifier our bank attaches to it.

• Healthy adults. None of this extends to a person with kidney disease, and nothing on this page should be read as if it did.
• It is not a clean bill of health. The defensible sentence is that a higher creatinine reading is expected on creatine and does not by itself mean damage. It is not that creatine is safe for your kidneys, and we are not going to write the second one because the studies do not support it.
• Say it before the test, not after. If a kidney panel is booked, tell the clinician who ordered it that you take creatine. It costs one sentence and it is the difference between a flagged result and a flagged result with an explanation.

If you already have a kidney condition, the answer to whether you should take creatine is not on this page and will not be. Ask the clinician who manages it.

Is creatine bad for your liver?

The strongest document available is a 21-month study of 98 college athletes on a 69-item panel of serum, whole-blood and urinary health markers, which reported no adverse effect against non-users training the same way. It was open-label. There is no liver-specific synthesis in the banks this page is written from, and that is a smaller answer than most pages give.

The study is Kreider and colleagues, Mol Cell Biochem 2003;244:95–104. Ninety-eight Division I-A college football players, tracked for twenty-one months, with blood and 24-hour urine collected at nine timepoints — 0, 1, 1.5, 4, 6, 10, 12, 17 and 21 months. The creatine group took 15.75 g a day for five days and then averaged 5 g a day. Against non-users under the same training load, long-term creatine did not adversely affect markers of health status.

Now the part that makes this an honest section rather than a comfortable one. That is a general clinical-marker panel, not a liver study. The kidney question has two dedicated syntheses of its own, quoted above. The liver question has this panel and nothing dedicated that we hold.

So the accurate sentence is: the broad health-marker evidence over 21 months is unremarkable, and there is no liver-specific body of evidence in front of us to summarise. A page that turns that into creatine is safe for your liver has added something the documents do not contain. If you have a liver condition, that is a question for the clinician managing it.

Does creatine cause hair loss?

The evidence is one small study and it has not replicated. Clin J Sport Med 2009, 20 college-aged rugby players: DHT rose 56% during loading and stayed 40% above baseline through maintenance, and testosterone did not change. It measured a hormone, not hair. A 2025 randomised trial with 38 completers measured hair directly over 12 weeks and found no difference. That is not the same as proof of safety, and this page will not write that sentence.

This question gets asked more than any other on this list, so it gets the longest answer and the most exact one.

What the 2009 study actually did

van der Merwe, Brooks and Myburgh, Clin J Sport Med 2009;19(5):399–404. Twenty college-aged rugby players. Seven days of loading at 25 g of creatine a day with 25 g of glucose, then fourteen days of maintenance at 5 g a day with 25 g of glucose, against a glucose placebo.

Dihydrotestosterone rose 56% after the loading week and remained 40% above baseline through maintenance. Serum testosterone did not change. That is the finding. It is the entire finding.

What the study did not do is the part that matters and the part that gets dropped in transit. It did not measure hair. Not density, not shedding, not follicle counts, not a photograph. It measured a hormone — the one the whole hair-loss worry is built around — in twenty men, over three weeks. And in the years since, no study has replicated the DHT finding.

The trial that did measure hair

One exists, and it is recent. J Int Soc Sports Nutr, 2025 — a 12-week randomised controlled trial. Forty-five resistance-trained males aged 18 to 40 were randomised and 38 completed: 5 g a day of creatine monohydrate against 5 g of maltodextrin. Hair was assessed directly, by trichogram and FotoFinder — hair density, follicular unit count, cumulative hair thickness. There were no group-by-time interactions for any hormone or any hair outcome (p > 0.05). No difference in DHT, no difference in the DHT to testosterone ratio, no difference in any hair parameter.

It is the first study to assess hair follicle health directly following creatine supplementation. It is also small and short, and both of those belong in the same breath as the result:

• 38 completers is a small sample. A null result in 38 people is a weak null.
• Twelve weeks is short. Androgenetic alopecia develops over years, not over a season.
• Nobody was screened for genetic predisposition. The population in which DHT is thought to matter is the population the trial could not identify.

So what is the answer

The honest one is short and it is not the one either side of this argument wants.

One 2009 study of twenty men found a hormone change and did not look at hair. One 2025 study looked at hair in thirty-eight men for twelve weeks and found nothing. Neither settles it.

We are not going to write creatine does not cause hair loss, because that sentence would be doing exactly what the 2009 study is accused of doing in reverse — taking a small result and stretching it into a general claim. And we are not going to dismiss the 2009 paper either, because it is a real, published, randomised finding and dismissing it would be the same move with the other sign.

If you are predisposed to pattern hair loss and this possibility bothers you, the position we would give a friend is that the question is open, the evidence is thin in both directions, and a dermatologist is better placed to weigh it for your particular case than any supplement brand is. Including this one.

Does creatine cause cramps or dehydration?

A 2009 systematic review with meta-analyses in J Athl Train, 10 qualifying studies out of 95 screened, found creatine did not reduce total body water as a proportion of body mass, did not impair thermoregulatory responses during exercise in the heat, and did not increase the incidence of heat-related illness.

Lopez, Casa, McDermott, Ganio, Armstrong and Maresh, J Athl Train 2009;44(2):215–223. Fifteen of ninety-five articles examined thermoregulation or hydration; ten met the inclusion criteria and were pooled. All three outcomes were null.

That is three separate versions of the same folk belief tested at once and none of them survived: not the drop in body water, not the impaired heat response, not the rise in heat illness. It is a 2009 finding and nothing in either citation bank contradicts it, which is the reason this section is four paragraphs rather than fourteen. The longer version is in does creatine dehydrate you.

What that review does not do is promise you will never cramp. Cramping has its own causes and creatine is not measurably one of them. If you cramp on creatine, you would probably have cramped without it.

Stomach upset, and the three things that cause it

Almost every gastrointestinal complaint filed against creatine traces to one of three things, and all three are fixable without stopping.

1. Too much in one sitting. A loading dose is roughly 0.3 g per kilogram of bodyweight per day — twenty to twenty-five grams for most men. Taken as one hit that is a lot of anything. The 2017 position stand splits it into four doses for exactly this reason.
2. Undissolved powder. Creatine that has not gone into solution sits heavily. Warmer water, a proper stir, and drink it before it settles out.
3. An empty stomach. Take it with food for a week and see whether the problem disappears. Usually it does.

If none of that works, drop to the bottom of the maintenance range and stay there. Three grams a day for 28 days reached the same roughly 20% rise in muscle creatine as 20 g a day for six days — J Appl Physiol, 1996. The low dose is not a compromise. It is the same destination on a slower road.

Acne, and whether creatine is a steroid

On acne: no trial in front of us measured it. Not one entry in either citation bank this page is written from reports a skin outcome from a creatine trial. We are not going to reason from the 2009 hormone finding to a skin outcome, because that is precisely the inference the 2009 authors did not make and the one this page spends a whole section refusing. If a page tells you creatine does or does not cause acne, ask which trial measured it.

On whether it is a steroid: creatine monohydrate is one compound, sold as a dietary supplement. The 2025 consensus statement in Frontiers in Nutrition, from the sports-nutrition society whose dosing stand is quoted above, holds that creatine has a well-supported safety profile including with long-term supplementation. That is the strongest sentence available and it is a statement about a body of evidence, not a guarantee about a person.

Is creatine safe to take for years?

The longest health-marker document is 21 months, 98 athletes, 69 markers, open-label. Two separate two-year randomised trials in postmenopausal women have since run, one at 3 g a day without training and one at 0.14 g/kg a day with it; both were null on their primary bone outcome. A 2026 meta-analysis pooling seven trials reported adverse events mild and comparable to placebo, with renal indices unchanged.

Three documents carry that answer and each one is worth naming, because the useful thing about them is how long they ran rather than what they found.

Twenty-one months. The Kreider panel above — 98 athletes, 69 clinical markers, nine collection points. Its limitation is that it was open-label: nobody was blinded, and the comparison was against non-users training the same way rather than against a placebo arm.

Two years, twice. Sales and colleagues, J Gerontol A 2020;75(5):931–938: two hundred postmenopausal women with osteopenia, 3 g a day, no resistance training, twenty-four months. Bone density declined over time in both arms with no interaction between group and time, and bone markers, microarchitecture, falls and fractures were all unchanged. The authors' own conclusion is that the trial refutes the long-standing notion that creatine on its own does anything for bone over the long run. Note the design: no training arm and 3 g a day, so it does not say that creatine alongside training does nothing. That trial was part-funded by AlzChem AG, a creatine manufacturer, which makes a flat null over two years a stronger negative rather than a weaker one.

Then Chilibeck and colleagues, Med Sci Sports Exerc 2023;55(10):1750–1760: 237 postmenopausal women, 0.14 g/kg a day, two years of resistance training three days a week plus walking six days a week. No effect on femoral neck, total hip or lumbar spine bone mineral density. Funded by the Canadian Institutes of Health Research, not by industry.

Both of those are trials about bone, and both are null on their primary outcome. Neither says anything about outcomes other than bone. They are cited here for the exposure rather than the result: two independent two-year randomised trials, several hundred people, and nothing turned up that stopped either one.

The 2026 pooled safety line. A meta-analysis in J Int Soc Sports Nutr pooled seven randomised trials, 608 postmenopausal women, durations from 12 to 104 weeks, and reported adverse events mild and comparable to placebo with renal indices unchanged. Three caveats travel with it and all three are in the paper: only seven trials, risk of bias rated mostly some concerns, and the review was not prospectively registered — the authors say so themselves. Its author list also carries the industry-ties note described in the water section above.

Who should ask a doctor before taking creatine?

Anyone with a kidney condition, anyone pregnant or breastfeeding, anyone taking prescription medicine, and anyone under 18 — these are products for adults. There is no study on this page that would let us say more than ask a clinician, and a page that says more is inventing it.

That is the entire section and it is short on purpose. Four questions get asked here constantly and all four have the same answer.

• A kidney condition. Every renal finding quoted above is explicitly about healthy adults. None of it transfers.
• Pregnancy or breastfeeding. Nothing in either citation bank addresses it. Silence is the honest answer and a clinician is the right destination.
• Prescription medicine. A pharmacist can check an interaction in under a minute. We cannot check it at all.
• Under 18. TARE products are dietary supplements for adults 18 and over. A question about a teenager belongs to that teenager's clinician.

One more, because it is practical rather than medical: if a blood test is booked, say you take creatine before it is drawn. That is the sentence that turns a flagged creatinine line into an explained one.

Does the form of creatine change any of this?

A 2026 audit in Nutrients read the creatine trial literature itself: 357 reports covering 343 randomised trials. Monohydrate was the form in 275 of them, 80.2%. Five trials, 1.5%, compared more than one creatine formulation at all. The answer to whether another form is gentler is not that the alternatives lost. It is that almost nobody ran the comparison.

That audit is the most useful single document in this whole subject and it is almost never quoted, so here it is in full. Nutrients 2026;18(16):2697, from the Faculty of Science at the University of Split — a meta-research study rather than a trial, MEDLINE and Embase searched from inception to 1 March 2024, 357 reports covering 343 randomised trials published between 1994 and 2023. It was not industry-funded.

Read the last row again. Five. And the paper records those five comparisons as frequently limited by non-equimolar dosing — the arms were not matched on the amount of creatine actually delivered, which is the one thing a form comparison has to get right. Only three of the five discussed bioavailability at all.

So when a label implies its form is absorbed better or sits easier, the honest reply is not that the claim has been disproved. It is that the comparison the claim depends on has been run five times across three decades of trials, and the audit records those five as frequently limited by unequal dosing. That is a different and more damaging answer, and it is the reason this brand sells the boring one.

What the few head-to-head trials do report is consistent. A three-arm, triple-blind, placebo-controlled trial in 31 team-sport athletes, 5 g a day for eight weeks, concluded that claims of hydrochloride superiority are unfounded — the long version is in creatine HCl versus monohydrate. A buffered form tested in 2012 produced no greater change in muscle creatine content than monohydrate, and creatine ethyl ester tested in 2009 was worse at raising serum and muscle creatine. Neither one beat it.

Delivery format is the newest version of the same question and it is the least tested. The 2026 gummy trial ran ten weeks in 32 female athletes at 5 g a day — and its comparison arm is described as a control group rather than a placebo group, with no blinding reported, which for outcomes like these is a serious limitation. That is the whole gummy evidence base, and it is unpicked in do creatine gummies work.

Base Work is creatine monohydrate. One ingredient, 5 g, sixty servings. Not because it is exotic but because it is the form 80.2% of the literature was run on, and this page has no interest in selling you a version of it that nobody tested.

What is in our tub, what our certificate covers, and where we will not ship it

Base Work is creatine monohydrate and nothing else. 5 g a serving, 300 g in the tub, 60 servings, taken daily. There is no loading week in the instructions and no clock attached to it, because the evidence does not support attaching one.

On timing specifically, two small trials point in opposite directions. One from 2013, n=19, found that taking it after training may be superior to before — the authors' own hedge. One from 2026, n=11 and described by its authors as exploratory, found more total load lifted when it was taken before. Two small trials, two directions. Total daily intake matters far more than the hour, which is the same conclusion when to take creatine reaches at greater length. Base Work is unflavoured so that it goes into whatever you were already drinking, at whatever hour you were already drinking it.

The certificate behind that powder is real, and it is smaller than the way this industry usually describes such things. It was issued by an independent laboratory that had nothing to do with making the product — Ethos Analytics, Phoenix, Arizona, signed by the laboratory director on 18 February 2026. The purity result is a good one: creatine monohydrate 99.87% against a specification of greater than 99.5%, by HPLC.

And here is what it does not cover, which is the part we would rather you heard from us. The customer of record on that document is Vox Nutrition, not TARE. The sample photographed on it is a sealed poly bag of bulk raw material from Accobio USA Inc., lot CM20251225A. The Serving Size field reads N/A. It is a test of an ingredient lot bought by the supplier, not of the tub we ship you, and no certificate in our hands tests the tub. The document is transcribed row by row on Testing & Certification, and what a certificate can and cannot prove is unpacked in what a certificate of analysis proves.

New York

We do not ship Base Work to New York addresses, or any bundle containing it. New York General Business Law §391-oo is the reason: it makes certain dietary supplements an 18-and-over product in that state and requires a delivery method that obtains an adult signature and government-issued photo identification at the door. Our fulfilment goes out as standard parcel, which is not that. Checkout will not complete to a New York address for this item, and the full statement is in the shipping policy. Night Work and Under Load ship there normally.

More in this cluster: what a trademarked creatine is, and what ours is, how much creatine per day, what creatine monohydrate actually does, whether creatine expires, and creatine for men over 30. Or read the documents themselves on Testing & Certification.