How Much Creatine Per Day for Muscle, Bone and Brain

Why the Standard Dose Comes from Muscle

Ask almost anyone how much creatine to take and you will hear the same answer: three to five grams a day. That number is not wrong, but it is worth knowing where it came from. According to Dr. Darren Candow, a professor at the University of Regina who has published more than 140 papers on nutrition, exercise and muscle aging, nearly every dosing recommendation in circulation was worked out for one purpose only: filling skeletal muscle. Roughly 95 percent of the body’s creatine sits in muscle, and for four decades that is where the research looked.

Candow explains that the picture has widened considerably. Creatine is now being studied for bone, brain, cardiovascular and immune health, and even in children and during pregnancy. His central argument is that muscle is only one of the tissues that use creatine, and the amount needed to saturate muscle may not be the amount needed to reach bone or brain. That is why he personally takes more than the standard recommendation, and why he thinks the question “how much creatine per day” now deserves a more layered answer than it used to.

This guide walks through what Candow said, tissue by tissue, and checks each claim against the published research. The short version is that the muscle dose is settled science, the bone dose is unresolved, and the brain dose is an active and promising research question rather than an established fact.

How Creatine Works in Muscle

Creatine’s job in muscle is to help regenerate ATP, the molecule that powers contraction. Candow describes phosphocreatine as a buffer that keeps ATP available for longer during hard efforts, which lets a person work at a higher intensity before slower energy systems have to take over. Anyone doing explosive or repeated high intensity work, whether that is weightlifting, sprinting or interval training, is the classic beneficiary.

He points to a pattern that many lifters will recognise. In a four set leg press, the first set usually feels the same with or without creatine, because resting stores are enough. The difference shows up in sets two, three and four, where supplemented individuals can grind out more repetitions. Over weeks, that extra volume adds up to greater physiological adaptation. Candow also notes that creatine speeds recovery between sets and even between individual contractions, so the same workout can be completed in less time with more benefit.

On mechanism, Candow describes two additional effects beyond raw energy supply. The first is neuromuscular: creatine appears to improve recruitment of type II muscle fibres, the fast, powerful fibres that decline with age. The second is cellular: creatine speeds the return of calcium into the sarcoplasmic reticulum after a contraction, which shortens relaxation time and helps muscle proteins engage again quickly. He is careful to add that creatine does not directly stimulate muscle protein synthesis, which surprises many people. Instead it appears to work through satellite cells, growth factors and a reduction in protein breakdown.

The research broadly supports this account. The International Society of Sports Nutrition position stand concludes that creatine monohydrate is the most effective nutritional supplement currently available for increasing high intensity exercise capacity and lean body mass during training (Kreider et al., 2017). A 2024 update of pooled trials in older adults found that creatine combined with strength training produced greater gains in lean tissue mass and upper body strength than training alone, though not in lower body strength (Candow et al., 2024). Candow himself flags an important caveat about the phrase “lean body mass”: in most studies it is measured by DXA, which captures connective tissue, water and organs alongside muscle, so it is a proxy rather than a direct measurement of muscle.

One nuance from the interview deserves emphasis. Candow says the best evidence for creatine’s anti inflammatory and recovery effects actually comes from endurance events such as marathons and triathlons, where markers of inflammation rise sharply and creatine appears to blunt that rise. Ordinary weight training, with its short bouts and rest periods, does not seem catabolic enough to show the same recovery advantage. He describes this as one of the findings that surprised him most.

Loading Versus Daily Dosing

The loading protocol most people have heard of, 20 grams a day for about a week, dates to the early 1990s and was designed for athletes who needed muscle saturation quickly. Candow explains that it works, but it also pulls water into cells rapidly, which is why some people experience bloating, temporary weight gain or digestive discomfort. His lab generally does not use it for that reason.

The alternative is simply to take a smaller daily dose for longer. According to Candow, three to five grams a day will fully saturate muscle in roughly three to four weeks, and after that the muscle is full regardless of whether a person loaded first. He recommends never dropping below one gram in a single serving, because very small amounts do not reach useful blood concentrations, but doses can otherwise be split across the day. For people who are prone to stomach upset, he suggests two and a half grams in the morning and two and a half in the evening, ideally with food, since a meal containing carbohydrate or protein appears to improve absorption and reduce irritation.

The published guidance matches this. The ISSN position stand describes both approaches as effective: loading with about 20 grams a day for five to seven days followed by three to five grams, or taking three to five grams from the start and reaching similar muscle levels in three to four weeks (Kreider et al., 2017). The expert consensus paper on creatine misconceptions likewise states that a loading phase is not required (Antonio et al., 2021).

Candow’s weight based alternative is 0.1 to 0.14 grams per kilogram of body mass, which works out to roughly seven to ten grams a day for a 70 kilogram person. He notes that a loading phase can still make sense in specific situations, such as an athlete facing competition within days or someone starting rehabilitation after surgery who needs muscle stores topped up fast.

Does Timing or Cycling Matter

Candow’s view on timing is relaxed. Unlike caffeine, creatine does not need to be taken at a specific moment to work, because the goal is saturation rather than an acute effect. His one preference is to take it within an hour before or after training, on the reasoning that muscle contraction opens more of the transport pathways that carry creatine into cells. He cites a 1992 study in which one exercised leg took up around 37 percent more creatine while the resting leg still gained about 25 percent, which illustrates that exercise helps but is not required.

On cycling, Candow has changed his position over the years. He used to think creatine only mattered on training days. He now recommends taking it every day, not because muscle needs it on rest days but because he hopes that on those days more of the supply reaches bone and brain. He is untroubled by the idea that some is excreted, calling creatine cost effective enough that a little waste is acceptable.

He also raises the caffeine question, where he admits he is in the minority. Candow separates his morning coffee from his first creatine dose by at least two hours. His reasoning is cellular: caffeine promotes calcium release from the sarcoplasmic reticulum while creatine promotes reuptake, and at caffeine doses above roughly 250 milligrams he believes the two may interfere over months of daily use. He acknowledges that other creatine researchers disagree and that the only training study on the question, his own, was underpowered. A review of the topic reached a similar conclusion: a direct interaction in the body is unlikely, the original concern rests on a small 1996 study, and any blunting effect may relate to opposing effects on muscle relaxation time or simply to digestive upset from taking both at once (Trexler and colleagues, 2015). A single cup of coffee at around 80 to 100 milligrams of caffeine is not something Candow worries about.

Creatine for Bone Health

Bone is where Candow’s enthusiasm and his honesty sit closest together. He explains two proposed mechanisms. Directly, creatine appears to energise osteoblasts, the cells that build bone, at least in cell and animal studies. Indirectly and more importantly in his view, creatine appears to reduce osteoclast activity, the process that breaks bone down. He compares this to a mild version of what bisphosphonate drugs do and says roughly 15 studies now show these anti resorptive effects.

Then comes the caveat he repeats several times: not a single study has shown that creatine increases bone mineral density. What his group has shown is a slower loss of density, particularly at the hip, and improvements in measures of bone geometry that predict bending strength. He also states plainly that creatine would never replace a pharmaceutical bone treatment, that its effects are small, and that the main driver of any bone benefit is the mechanical loading from resistance training. In his words, if his parents took eight grams a day and did not exercise, he would not expect anything to happen to their bones.

The trial record supports both the promise and the caution. In a one year study of postmenopausal women taking 0.1 grams per kilogram (about seven to eight grams) during supervised resistance training, the creatine group lost 1.2 percent of femoral neck density compared with 3.9 percent in the placebo group, and also showed increased femoral shaft width, a predictor of bone bending strength (Chilibeck et al., 2015). However, when the same group ran a larger two year trial in 237 postmenopausal women using a higher dose of 0.14 grams per kilogram alongside resistance training and walking, creatine had no effect on bone mineral density at the femoral neck, total hip or lumbar spine. It did preserve section modulus and buckling ratio at the femoral neck, which are geometric properties related to fracture resistance, and it improved walking speed (Chilibeck et al., 2023). A pooled analysis of trials in adults over 50 similarly found no effect of creatine plus resistance training on bone mineral density at any site (Forbes et al., 2018).

Candow’s remark that lower doses have failed is also verifiable. A Brazilian trial gave 200 postmenopausal women with osteopenia three grams a day for two years, without an exercise programme, and found no benefit for bone density, bone microarchitecture, lean mass or muscle function (Sales et al., 2020). Candow describes eight grams as the lowest dose that has shown any bone effect in his own work and speculates that five grams over a long enough period might help, but he is clear that this is speculation and that the study to test it would be expensive.

An honest evidence rating for bone, then, is early. There is a plausible mechanism, one positive one year trial, one larger two year trial that found geometric but not density benefits, and pooled analyses showing no density effect. Anyone taking creatine primarily for bone should treat it as a possible small addition to resistance training, not as a treatment.

Creatine and the Brain Under Stress

The brain is the tissue Candow expects to dominate creatine research for the next two or three decades. He explains that the brain is unusual: it makes its own creatine, and the blood brain barrier is relatively resistant to creatine arriving from the bloodstream. Astrocytes, the large glial cells that filter what enters neurons, lack the creatine transporter, so uptake from supplements is blunted. A healthy, well rested brain with no metabolic disease is producing enough on its own and, as Candow puts it, tells circulating creatine that it is not needed.

That changes under stress. Candow’s consistent theme is that creatine helps most when the brain is metabolically strained: sleep deprivation, jet lag, hypoxia, concussion, heavy mental fatigue, and possibly aging. He summarises the human evidence as split, with some trials showing nothing, but with the positive results clustering in populations that are fatigued, sleep deprived or older. Small in stature but consuming around 20 percent of daily energy, the brain has more to gain when its energy supply is under pressure.

The evidence supports the stress framing surprisingly well. A cross sectional Brazilian study using magnetic resonance spectroscopy found that vegetarians and omnivores had the same brain creatine content despite very different dietary intake, which confirms that the brain relies on its own synthesis under normal conditions (Solis et al., 2014). Candow cited this study, though it was published in 2014 rather than about four years ago as he recalled. On the other side of the coin, a six week trial giving healthy young adults 10 or 20 grams a day found no improvement in cognitive performance or prefrontal oxygenation (Dose-Response of Creatine Supplementation on Cognitive Function in Healthy Young Adults, 2023). In other words, an unstressed young brain does not obviously benefit, exactly as Candow predicts.

The strongest positive evidence comes from sleep deprivation. A German crossover trial gave 15 participants a single dose of 0.35 grams per kilogram (about 25 grams for a 70 kilogram person) or placebo during 21 hours of sleep deprivation. Creatine improved processing speed and memory tasks, with effects visible at three hours, peaking around four hours and lasting up to nine hours, and spectroscopy showed corresponding changes in brain energy metabolites (Scientific Reports, 2024). Candow described this study accurately. A follow up from the same group tested a lower single dose of 0.2 grams per kilogram and found a smaller but still measurable reduction in cognitive decline during sleep deprivation, which suggests the acute effect is dose dependent (Nutrients, 2026). The lead author has publicly cautioned against people taking such large single doses at home, so this should be read as a proof of concept rather than a protocol.

A systematic review and meta analysis of randomised trials in healthy people found that creatine supplementation improved memory, with the effect most pronounced in older adults, and did not find that dose, duration or sex changed the result (Prokopidis et al., 2023). That fits Candow’s suggestion that aging is itself a form of metabolic stress on the brain and that as little as five grams a day has shown some potential in older adults.

Where the evidence is thinner is on the exact dose needed to raise brain creatine stores over time. Candow says the best imaging studies used 20 grams a day for at least a week, that a single study using around four grams took three months to show accumulation, and that a German study comparing two, four and ten grams showed the largest percentage increase at ten grams. The 20 gram figure is consistent with the imaging literature summarised in the memory meta analysis. The four gram and ten gram studies could not be located and verified for this article, so those specific figures should be treated as Candow’s account rather than confirmed findings. His personal practice, ten grams a day rising to 20 during travel or sleep loss, is presented in the interview as informed opinion rather than a recommendation backed by trials.

Sleep, Concussion and Mood

Three narrower brain topics came up, each with a different strength of evidence.

On sleep, Candow described a study in young women who resistance trained twice a week for six weeks and slept close to an hour longer on training days when taking creatine. The published trial confirms the direction of the finding but differs on details. The 21 participants took five grams of creatine plus five grams of maltodextrin, not ten grams of creatine, and total sleep increased significantly on training days but not overall. Strength improved in both groups with no difference between creatine and placebo, so Candow’s statement that creatine improved strength in this study is not supported by the paper (Aguiar Bonfim Cruz et al., 2024). Candow himself notes that no mechanism was measured and that the result runs opposite to anecdotal reports of insomnia on high doses, so this is a single small trial and nothing more.

On concussion, Candow argues that people in contact sports should take creatine daily as a preventive measure, based on rodent studies showing faster recovery when creatine was on board before injury. He cited one human study of children given 0.4 grams per kilogram immediately after head trauma. That study exists but was conducted in Greece, not Scotland, and it was an open label pilot in 39 children and adolescents with moderate to severe traumatic brain injury. Children who received creatine for six months had shorter post traumatic amnesia, shorter intensive care stays and better outcomes in cognition, self care and communication, with no side effects reported (Sakellaris et al., 2006). Because there was no placebo and no blinding, and because the dose and population were unusual, the finding is encouraging but far from definitive. A larger placebo controlled trial in adults with persistent concussion symptoms is planned but has not reported.

On depression, Candow was careful to say that no study has tested creatine alone; it has only ever been an adjunct to antidepressants or therapy, and it could never replace them. The key trial randomised 52 women with major depressive disorder to escitalopram plus either five grams of creatine or placebo for eight weeks. The creatine group improved faster, with differences visible by week two, and the remission rate at eight weeks was 52 percent versus 26 percent (Lyoo et al., 2012). Candow’s account matches the trial. He points to reduced brain creatine and bioenergetics in depression as the plausible mechanism, and a large randomised trial is still needed.

For neurodegenerative disease, Candow was blunt: the totality of evidence shows little. He mentioned a five year trial that found no benefit. That trial was the NINDS Long Term Study 1 in Parkinson’s disease, run across 45 sites in the United States and Canada, not Germany as he recalled, and it gave 10 grams a day, not a small dose. It enrolled 1,741 people with early Parkinson’s disease and was stopped for futility with no difference in clinical decline between creatine and placebo (NINDS Exploratory Trials in Parkinson Disease Investigators, 2015). Candow’s suggestion that the dose was small does not hold for this trial, which used the same amount he personally takes. His broader point stands: creatine has not been shown to alter the course of established neurodegenerative disease, and prevention trials in people with early cognitive decline are only now getting under way.

Who Responds Most

Candow repeatedly returns to the idea that creatine helps most where the body is under strain or where baseline stores are low. Several populations follow from that.

Vegans and vegetarians get almost no creatine from food, since it is found in meat, poultry and fish and only in trace amounts in plants. Candow says they have substantially lower muscle creatine than omnivores and respond exceptionally well to supplementation. An eight week trial confirms this: vegetarians started with significantly lower muscle creatine, and those who supplemented during resistance training gained more total creatine, phosphocreatine, lean tissue and total work capacity than omnivores who supplemented (Burke et al., 2003). As noted above, this advantage does not automatically extend to the brain, where vegetarians and omnivores start from the same place.

Women, and particularly postmenopausal women, are the population in which Candow says his group has seen the best bone results, alongside improvements in strength, agility and balance. He urges women who are worried about water retention to skip the loading phase and start at two to three grams split across the day. A sex difference does appear in his own data on protein breakdown: creatine reduces markers of whole body breakdown in young males but not in females, and he speculates that oestrogen or progesterone may explain this. That mechanism is unresolved.

Older adults face what Candow calls anabolic resistance, and he argues that creatine, like protein, may need to be dosed higher with age. Two Canadian pooled analyses found that creatine plus resistance training improved functional tasks such as rising from a chair in older adults, but Candow acknowledges that no such benefit has been shown without exercise. He is emphatic that movement is foundational and creatine is the addition.

Children are the most controversial group. Candow’s position is that creatine is a nutrient the body already makes, that children and adolescents in published studies have not shown adverse effects, and that he would give it to his own children. He mentioned proposed bans in some states and considers them unsupported by evidence. A review of 13 studies in healthy children and adolescents found performance benefits and no adverse events, while noting that the evidence base is small (Jagim and Kerksick, 2021). The ISSN considers supplementation acceptable for younger athletes under supervision, with a balanced diet and recommended doses (Kreider et al., 2017). On pregnancy, Candow was clear that human data are limited to questionnaires, that ethics approval is difficult, and that larger trials are needed before anything can be said with confidence.

Safety, Kidneys and Common Myths

Candow calls creatine monohydrate the safest and most effective ergogenic aid available, and the position stands agree. The ISSN states that there is no compelling evidence that short or long term use, up to 30 grams a day for five years, harms otherwise healthy people (Kreider et al., 2017).

The kidney concern is the one he hears most. His explanation is that creatine is metabolised to creatinine, which doctors use to estimate kidney filtration. Someone supplementing will often show mildly elevated blood creatinine, which can be misread as reduced kidney function. When they stop, the number falls, which is sometimes taken as proof that creatine was harming the kidneys when it was simply a measurement artefact. He advises telling your doctor you take creatine before any blood work. Two independent meta analyses back this reading. One found that creatine did not alter serum creatinine or urea to a clinically meaningful degree and did not induce renal damage at the doses and durations studied (de Souza e Silva et al., 2019). A 2025 meta analysis found a small statistically significant rise in serum creatinine but no significant change in glomerular filtration rate, and concluded that creatine is likely safe for kidney function in healthy people at standard doses (Kabiri Naeini et al., 2025). People with existing kidney disease are a separate question and should speak to a physician.

On hair loss, Candow traced the myth to a single crossover study in rugby players in which a high loading dose raised DHT by around 57 percent, while remaining within the normal range, with no measurement of hair at all. He said a direct study was in review. That study has since been published: 45 resistance trained men took five grams a day or placebo for 12 weeks, with hair density, follicle count and thickness assessed directly alongside testosterone and DHT. There were no differences between groups on any hormone or hair outcome (Journal of the International Society of Sports Nutrition, 2025). The trial was short and did not screen for genetic predisposition, so it cannot rule out an effect over years in susceptible people, but it is the best evidence available and it points the same way as Candow’s expectation.

Other myths Candow addressed, and which the expert consensus paper also examined, include dehydration and cramping (creatine pulls water into muscle and if anything may protect against cramping), fat gain (the evidence points to a small reduction in body fat alongside lean mass gain), increased urination (no evidence that the molecule is a diuretic; extra fluid intake is the likelier cause), and blood pressure (no adverse effect seen) (Antonio et al., 2021). On body fat specifically, a meta analysis of 19 trials in adults over 50 found that creatine plus resistance training produced a 0.55 percentage point greater reduction in body fat than training alone, with a non significant half kilogram difference in fat mass, which matches Candow’s description of a real but very small effect (Forbes et al., 2019).

Areas where Candow was speculative and where this article found no verification include the immune system, cancer, cardiovascular function, male fertility and homocysteine. He described each as emerging, mixed or preliminary, and that characterisation should be respected.

Choosing and Taking Creatine

Candow’s advice on product choice is simple: creatine monohydrate, third party tested, ideally as the sole ingredient. He explains that monohydrate survives stomach acid, passes through the small intestine intact and is chemically identical to what the liver makes, which likely explains its safety record. Other forms such as hydrochloride, buffered or liposomal creatine must first deliver creatine to the blood to be useful, and none has been shown to be safer or more effective than monohydrate. He is sceptical of claims about superior bioavailability, since monohydrate is already close to fully absorbed. He also questions any product marketed as creatine that does not cause a small increase in cell water, since that osmotic effect is a feature of the molecule itself. Micronised monohydrate is simply a finer powder that dissolves more easily and works the same way. The misconceptions consensus reaches the same conclusion on alternative forms (Antonio et al., 2021).

For anyone who gets stomach upset, Candow’s practical steps are: start at around three grams a day split into two servings, take each with a meal, avoid dissolving it in plain water alone, and titrate up over a few weeks. He mixes ten grams into breakfast yoghurt with protein. Heat is not the problem people assume; warm liquids improve solubility and only prolonged high heat degrades creatine. He also flags two things to be cautious about: multi ingredient blends, where the creatine dose is often unknown and the other ingredients are untested in combination, and products of uncertain origin that lack third party verification.

A Practical Dosing Framework

Pulling the interview and the evidence together, the answer to how much creatine per day depends on what you are trying to reach.

For muscle strength, training volume and lean mass, three to five grams of creatine monohydrate daily, taken consistently, saturates muscle within about a month and is supported by decades of trials. A loading phase of 20 grams a day for five to seven days gets there faster but is optional and more likely to cause bloating. This is the strong evidence tier.

For bone, Candow’s trials used around eight to ten grams a day alongside resistance training. One year of that protocol slowed hip bone loss in postmenopausal women, but a larger two year trial found no density benefit, only preserved bone geometry, and three grams a day without exercise did nothing. Resistance training is the driver and creatine is at best a small addition. This is the early evidence tier.

For the brain, the honest summary is that creatine helps most when the brain is stressed. A large single dose has been shown to blunt the cognitive cost of a night without sleep, memory appears to improve with supplementation particularly in older adults, and creatine speeds antidepressant response as an adjunct in women with depression. Healthy young adults with adequate sleep show little benefit even at 20 grams a day. Whether a daily dose of ten grams reliably raises brain creatine over time, as Candow hopes, has not been established in trials that could be verified here. This is the promising tier for stressed or older brains and the unsupported tier for healthy young ones.

Candow’s own answer is to take more than the standard dose, accept that some is excreted, and hope the surplus reaches bone and brain. That is a reasonable personal choice given creatine’s safety record and low cost, but it is worth being clear that it is a hypothesis he is living by rather than a conclusion the evidence has yet reached. What the evidence does say with confidence is that three to five grams a day is effective for muscle, that creatine is well tolerated across a very wide range of doses and populations, and that exercise remains the foundation for every benefit discussed here.