SnakeOilCheck earns commissions from qualifying purchases made through affiliate links. Our editorial content and meta-analyses are produced independently.

Creatine: Does It Actually Work?

Last reviewed: 2026-02-20 00:00:00 +0000 UTC

🔬 Meta-analysis verified

SnakeOilCheck earns commissions from qualifying purchases made through affiliate links on this site. Our editorial content and meta-analyses are produced independently and are not influenced by affiliate relationships.

✓ WORKS

Strong evidence for strength and muscle mass from massive meta-analyses. Emerging evidence for cognition, especially in older adults and vegetarians.

Confidence
92/100 Strong
🔬 Meta-analysis verified

Increases strength and power

SMD = 0.35 Small effect [0.26, 0.44] 69 studies · 2827 people

69 randomized trials with nearly 3,000 people confirm creatine makes you stronger. The effect is …

Grade A

Builds muscle mass

SMD = 0.36 Small effect [0.25, 0.47] 143 studies · 5829 people

143 trials with nearly 6,000 people show creatine increases lean body mass. This is the largest …

Grade A
?

Improves cognition

SMD = 0.3 Small effect [0.07, 0.53] 16 studies · 804 people

16 trials with about 800 people show creatine may improve short-term memory. The effect is small but …

Grade B
?

Reduces exercise fatigue

SMD = 0.61 Moderate effect [, ] 14 studies · people

14 double-blind trials show creatine boosts your power output during repeated sprints. The effect is …

Grade B
?

Speeds exercise recovery

SMD = -1.09 Large effect [, ] 23 studies · 469 people

23 trials with 469 people show creatine lowers muscle damage markers (like CK) at 48-90 hours after …

Grade B

Reduces depression symptoms

SMD = -0.34 Small effect [-0.68, -0] 11 studies · 1093 people

11 trials with 1,093 people were pooled together, but the results are weak. The average improvement …

Grade C

Enhances sprint and high-intensity performance

WMD = 47.81 watts [15.55, 80.06] 69 studies · 1937 people

69 randomized trials with nearly 2,000 people show creatine improves sprint and high-intensity …

Grade A

Supports bone health

SMD = Minimal effect [, ] 5 studies · 1093 people

Two meta-analyses and a 2025 editorial all reach the same conclusion. Creatine doesn't improve bone …

Grade C

Improves blood sugar control

SMD = 0.05 Minimal effect [-0.53, 0.63] 9 studies · people

The only meta-analysis (9 studies) found creatine doesn't improve fasting blood sugar or insulin …

Grade C

Supports heart failure patients

SMD = Minimal effect [, ] 7 studies · 243 people

7 small trials with 243 heart failure patients have tested creatine. It's safe. It may improve …

Grade C

Protects against traumatic brain injury

SMD = Minimal effect [, ] 1 studies · 39 people

This is mostly hype right now. Only one human trial exists. It was a small, unblinded pilot study in …

Grade D

Boosts testosterone and DHT

SMD = Minimal effect [, ] 12 studies · people

This claim is mostly internet hype. One single study from 2009 (20 rugby players) found creatine …

Grade D

Improves skin health and reduces wrinkles

SMD = Minimal effect [, ] 0 studies · people

All the evidence for creatine and skin is about rubbing creatine cream on your face, not swallowing …

Grade D

What Is Creatine?

Creatine is a natural compound your body makes from amino acids. Your muscles store it as phosphocreatine and use it for quick energy during intense efforts like lifting weights or sprinting.

You get some creatine from meat and fish. But supplementing with creatine monohydrate raises your muscle stores about 20-40% above what food alone provides.

It’s one of the most studied supplements in history. We’re talking about hundreds of randomized controlled trials. The data on creatine is stronger than almost any other supplement you can buy.

We analyzed 3 major meta-analyses covering 228 individual studies. Here’s what the numbers actually show.

The Evidence, Claim by Claim

The sections below break down each major claim with real data from published meta-analyses. Every effect size, confidence interval, and study count comes from peer-reviewed research.

Strength and Power

This is where creatine shines brightest. Kazeminasab et al. (2025) pooled 69 randomized controlled trials with nearly 3,000 participants. The result: creatine significantly improves maximal strength when combined with resistance training.

The effect size (SMD 0.35) is small to medium. In practical terms, that means a few extra reps or a few extra pounds on your lifts. It won’t double your bench press. But it adds up over months of training.

The evidence quality is as good as it gets in nutrition research. 69 RCTs is a massive body of evidence.

Muscle Mass

Pashayee-Khamene et al. (2024) conducted the largest creatine meta-analysis ever. They pooled 143 RCTs with nearly 6,000 participants looking at body composition.

Creatine increases lean body mass (SMD 0.36). It doesn’t reduce fat mass. Total body weight goes up, partly from water retention inside muscle cells.

Here’s the important part: long-term studies using DEXA scans confirm that the gains aren’t just water. Real muscle tissue grows. The water retention happens early and levels off. The muscle growth continues as long as you keep training.

Cognition

This one is newer and less certain. Xu et al. (2024) analyzed 16 RCTs with about 800 participants looking at cognitive outcomes.

Short-term memory improved with a small effect size (SMD 0.30, p = 0.01). That’s statistically significant but modest.

The interesting finding: the effect is stronger in older adults and vegetarians. This makes sense. Your brain uses creatine for energy. People with lower baseline levels (vegetarians don’t get dietary creatine) seem to benefit more.

This claim gets a “maybe” verdict. The data is encouraging, but 16 studies with moderate heterogeneity isn’t enough to call it settled science.

Dosage

How much: 5g per day. That’s it. No need to get fancy.

Loading phase: Optional. Research used loading protocols of 20g per day (split into 4 doses) for 5-7 days to saturate faster. Most studies used 5g daily, which reaches the same levels in 3-4 weeks.

Timing: Doesn’t matter much. Take it whenever you’ll remember. Some research suggests post-workout may be slightly better, but the difference is tiny.

Form: Creatine monohydrate. Don’t waste money on HCl, buffered, or ethyl ester versions. They cost more and don’t work better. Look for Creapure certification for purity.

With food? Not required. Taking it with carbs may slightly improve uptake, but the effect is small.

Who Should Be Careful With Creatine

Creatine is one of the safest supplements available. But a few groups should be cautious.

If you have kidney disease, talk to your doctor first. Creatine is safe for healthy kidneys, but compromised kidneys may not handle the extra load.

Creatine raises creatinine levels on blood tests. This is a known measurement artifact, not kidney damage. Tell your doctor you take creatine before routine bloodwork so they don’t misinterpret the results.

Stay well hydrated. Creatine pulls water into your muscles. Drink enough water throughout the day, especially during exercise.

If you’re taking NSAIDs regularly or any drugs that stress the kidneys, be extra careful about hydration. The combination isn’t dangerous for most people, but don’t ignore it.

The Bottom Line

Creatine is the real deal. It’s one of the few supplements where the evidence is overwhelming.

For strength and muscle, the data is about as strong as it gets in supplement research. The pooled evidence from 143 RCTs is consistent. If you lift weights and want to get stronger or build more muscle, creatine appears to help.

The cognitive benefits are promising but need more research. If you’re older or don’t eat meat, you might notice a mental boost. But don’t buy creatine for brain power alone just yet.

At roughly $0.12 per day, it’s also one of the cheapest supplements that actually does something. Monohydrate is the way to go. Skip the expensive “advanced” forms.

The Evidence, Claim by Claim

Increases strength and power ✓ Works

Effect Size SMD = 0.35 95% CI [0.26, 0.44]
Studies 69 2827 participants
Consistency I² = 62% τ = 0.280
Prediction Interval [-0.20, 0.90] Range of expected effects in new studies

69 randomized trials with nearly 3,000 people confirm creatine makes you stronger. The effect is small to medium in size but very consistent. It works for bench press, squat, leg press, and other compound lifts.

This is one of the most well-supported claims in sports nutrition. 69 RCTs is a massive evidence base. Moderate heterogeneity means results vary somewhat, but the direction is clear. Creatine improves maximal strength.

View full statistical analysis
Forest plot for creatine-strength
Forest plot. Each square is one study (size = weight). The diamond is the pooled effect. The dashed line marks zero (no effect).
Funnel plot for creatine-strength
Funnel plot. Symmetric = low publication bias concern. Hollow circles = imputed studies from trim-and-fill analysis.

Publication Bias Assessment

Egger's Test z = —, p = 0.08 slight asymmetry detected but not statistically significant
Records identified (n = 1240) Records screened (n = 680) Records excluded (n = 540) Full-text reports assessed (n = 140) Reports excluded (n = 71) Studies included in meta-analysis (n = 69)
PRISMA flow diagram showing study selection process.

Builds muscle mass ✓ Works

Effect Size SMD = 0.36 95% CI [0.25, 0.47]
Studies 143 5829 participants
Consistency I² = 58% τ = 0.320
Prediction Interval [-0.27, 0.99] Range of expected effects in new studies

143 trials with nearly 6,000 people show creatine increases lean body mass. This is the largest creatine meta-analysis ever done. The effect is small to medium. Fat mass doesn't change. Some of the weight gain is water, but real muscle growth happens too.

143 RCTs make this one of the most studied supplement effects in existence. The lean mass gains are real, not just water. DEXA-measured studies confirm actual muscle tissue increases. Body weight goes up more than lean mass alone, so some water retention is part of the picture.

View full statistical analysis
Forest plot for creatine-muscle-mass
Forest plot. Each square is one study (size = weight). The diamond is the pooled effect. The dashed line marks zero (no effect).
Funnel plot for creatine-muscle-mass
Funnel plot. Symmetric = low publication bias concern. Hollow circles = imputed studies from trim-and-fill analysis.

Publication Bias Assessment

Egger's Test z = —, p = 0.12 no significant asymmetry detected in funnel plot
Records identified (n = 2100) Records screened (n = 1150) Records excluded (n = 880) Full-text reports assessed (n = 270) Reports excluded (n = 127) Studies included in meta-analysis (n = 143)
PRISMA flow diagram showing study selection process.

Improves cognition ? Maybe

Effect Size SMD = 0.30 95% CI [0.07, 0.53]
Studies 16 804 participants
Consistency I² = 52% τ = 0.240
Prediction Interval [-0.18, 0.78] Range of expected effects in new studies

16 trials with about 800 people show creatine may improve short-term memory. The effect is small but statistically significant. It seems to work better for older adults and people who don't eat meat. Vegetarians and vegans may benefit most because their baseline creatine levels are lower.

The evidence is promising but not yet rock-solid. 16 RCTs is decent, but moderate heterogeneity and the prediction interval crossing zero mean some future studies might not find a benefit. The subgroup effects (older adults, vegetarians) are interesting but need more research.

View full statistical analysis
Forest plot for creatine-cognition
Forest plot. Each square is one study (size = weight). The diamond is the pooled effect. The dashed line marks zero (no effect).
Funnel plot for creatine-cognition
Funnel plot. Symmetric = low publication bias concern. Hollow circles = imputed studies from trim-and-fill analysis.

Publication Bias Assessment

Egger's Test z = —, p = 0.22 no significant publication bias detected
Records identified (n = 520) Records screened (n = 285) Records excluded (n = 240) Full-text reports assessed (n = 45) Reports excluded (n = 29) Studies included in meta-analysis (n = 16)
PRISMA flow diagram showing study selection process.

Reduces exercise fatigue ? Maybe

Effect Size SMD = 0.61 95% CI [%!f(<nil>), %!f(<nil>)]
Studies 14 %!d(<nil>) participants
Consistency I² = %!f(<nil>)% τ = %!f(<nil>)

14 double-blind trials show creatine boosts your power output during repeated sprints. The effect is moderate (delta 0.61). But here's the catch. It doesn't actually slow down how fast you fatigue. You're more powerful across all sprints, but you still lose power at the same rate. For endurance fatigue, a separate meta-analysis of 13 studies found zero benefit.

Strong evidence that creatine improves repeated sprint power. But the fatigue claim is misleading. Creatine raises the ceiling, not the slope. It doesn't reduce endurance fatigue at all. Gets a 'maybe' because the claim as stated is only half true.

View full statistical analysis
Forest plot for creatine-reduces-exercise-fatigue
Forest plot. Each square is one study (size = weight). The diamond is the pooled effect. The dashed line marks zero (no effect).
Funnel plot for creatine-reduces-exercise-fatigue
Funnel plot. Symmetric = low publication bias concern. Hollow circles = imputed studies from trim-and-fill analysis.

Publication Bias Assessment

Egger's Test z = —, p = — not reported in accessible abstract

Speeds exercise recovery ? Maybe

Effect Size SMD = -1.09 95% CI [%!f(<nil>), %!f(<nil>)]
Studies 23 469 participants
Consistency I² = 83% τ = %!f(<nil>)

23 trials with 469 people show creatine lowers muscle damage markers (like CK) at 48-90 hours after hard exercise. That sounds good. But it doesn't improve actual recovery outcomes. Muscle strength, soreness, range of motion, and inflammation all showed no significant improvement. And with long-term use, damage markers actually go UP because creatine lets you train harder.

The headline number (SMD -1.09) looks impressive, but it's fragile. Remove one outlier study and it's no longer significant. High heterogeneity (I2=83%) means results vary widely. No effect on soreness, strength recovery, or inflammation. The paradox of higher CK with chronic use is well-documented.

View full statistical analysis
Forest plot for creatine-speeds-exercise-recovery
Forest plot. Each square is one study (size = weight). The diamond is the pooled effect. The dashed line marks zero (no effect).
Funnel plot for creatine-speeds-exercise-recovery
Funnel plot. Symmetric = low publication bias concern. Hollow circles = imputed studies from trim-and-fill analysis.

Publication Bias Assessment

Egger's Test z = —, p = 0.55 no significant publication bias detected
Records identified (n = 2107) Records screened (n = 1865) Records excluded (n = 1824) Full-text reports assessed (n = 41) Reports excluded (n = 18) Studies included in meta-analysis (n = 23)
PRISMA flow diagram showing study selection process.

Reduces depression symptoms ✗ No Evidence

Effect Size SMD = -0.34 95% CI [-0.68, -0.00]
Studies 11 1093 participants
Consistency I² = 71% τ = %!f(<nil>)

11 trials with 1,093 people were pooled together, but the results are weak. The average improvement equals about 2.2 points on a depression scale. That's below the 3-point threshold doctors consider meaningful. Six of the 11 trials had high risk of bias. The largest trial (n=1,741) studied Parkinson's patients and found nothing. When you correct for publication bias, the effect shrinks further.

GRADE rates this evidence 'very low,' the lowest possible rating. Downgraded for risk of bias, inconsistency, imprecision, and publication bias. The most promising signal comes from one good RCT of creatine added to SSRIs in women (Lyoo 2012). But the overall evidence doesn't support the claim.

View full statistical analysis
Forest plot for creatine-reduces-depression-symptoms
Forest plot. Each square is one study (size = weight). The diamond is the pooled effect. The dashed line marks zero (no effect).
Funnel plot for creatine-reduces-depression-symptoms
Funnel plot. Symmetric = low publication bias concern. Hollow circles = imputed studies from trim-and-fill analysis.

Publication Bias Assessment

Egger's Test z = —, p = — publication bias inflates the apparent benefit
Records identified (n = 558) Records screened (n = 0) Records excluded (n = 0) Full-text reports assessed (n = 34) Reports excluded (n = 23) Studies included in meta-analysis (n = 11)
PRISMA flow diagram showing study selection process.

Enhances sprint and high-intensity performance ✓ Works

Effect Size WMD = 47.81 95% CI [15.55, 80.06]
Studies 12 1937 participants
Consistency I² = 48% τ = %!f(<nil>)

69 randomized trials with nearly 2,000 people show creatine improves sprint and high-intensity performance. Wingate peak power increased by about 48 watts. Vertical jump improved by 1.5 cm. A separate meta-analysis of 14 repeated sprint studies found creatine boosted mean power by a moderate amount (delta 0.61). Males benefit more than females for power outcomes.

Very strong evidence base. 69 RCTs with no significant publication bias. The ISSN calls creatine the most effective ergogenic supplement for high-intensity exercise. This is distinct from the 'strength' claim because it covers anaerobic sprint performance, Wingate tests, and repeated sprint ability specifically.

View full statistical analysis
Forest plot for creatine-enhances-sprint-performance
Forest plot. Each square is one study (size = weight). The diamond is the pooled effect. The dashed line marks zero (no effect).
Funnel plot for creatine-enhances-sprint-performance
Funnel plot. Symmetric = low publication bias concern. Hollow circles = imputed studies from trim-and-fill analysis.

Publication Bias Assessment

Egger's Test z = —, p = 0.07 no significant publication bias detected

Supports bone health ✗ No Evidence

Effect Size SMD = %!f(<nil>) 95% CI [%!f(<nil>), %!f(<nil>)]
Studies 5 193 participants
Consistency I² = %!f(<nil>)% τ = %!f(<nil>)

Two meta-analyses and a 2025 editorial all reach the same conclusion. Creatine doesn't improve bone mineral density. The largest trial (200 women, 2 years, double-blind) found zero effect. One study showed creatine slowed bone loss at the femoral neck specifically, but no other trial replicated this. The 2025 Osteoporosis International editorial explicitly states there's insufficient evidence to recommend creatine for osteoporosis.

The evidence is consistently null. When multiple meta-analyses agree that an effect doesn't exist, that's a strong signal. The one positive finding (Chilibeck 2015 femoral neck) hasn't been replicated in larger or longer trials. Creatine may help muscles, but it doesn't help bones.

View full statistical analysis
Forest plot for creatine-supports-bone-health
Forest plot. Each square is one study (size = weight). The diamond is the pooled effect. The dashed line marks zero (no effect).
Funnel plot for creatine-supports-bone-health
Funnel plot. Symmetric = low publication bias concern. Hollow circles = imputed studies from trim-and-fill analysis.

Publication Bias Assessment

Egger's Test z = —, p = — not reported
Records identified (n = 0) Records screened (n = 0) Records excluded (n = 0) Full-text reports assessed (n = 0) Reports excluded (n = 18446744073709551611) Studies included in meta-analysis (n = 5)
PRISMA flow diagram showing study selection process.

Improves blood sugar control ✗ No Evidence

Effect Size SMD = 0.05 95% CI [-0.53, 0.63]
Studies 9 %!d(<nil>) participants
Consistency I² = 22% τ = %!f(<nil>)

The only meta-analysis (9 studies) found creatine doesn't improve fasting blood sugar or insulin resistance. The one standout trial (Gualano 2011) showed creatine plus exercise lowered HbA1c by 1.1% in type 2 diabetics. But that's a single trial with 25 people. One study in vegetarians actually showed creatine WORSENED glucose handling. The evidence only looks promising when creatine is combined with exercise in diabetics, and even that needs replication.

The meta-analysis is clear. No significant effect on fasting glucose (SMD 0.05) or HOMA-IR (SMD -0.38). The one impressive result (Gualano 2011 HbA1c reduction) is a single small trial. The mechanism via GLUT-4 translocation is biologically plausible but only activates with exercise. Don't take creatine for blood sugar control.

View full statistical analysis
Forest plot for creatine-improves-blood-sugar-control
Forest plot. Each square is one study (size = weight). The diamond is the pooled effect. The dashed line marks zero (no effect).
Funnel plot for creatine-improves-blood-sugar-control
Funnel plot. Symmetric = low publication bias concern. Hollow circles = imputed studies from trim-and-fill analysis.

Publication Bias Assessment

Egger's Test z = —, p = — not reported in publicly accessible abstract

Supports heart failure patients ✗ No Evidence

Effect Size SMD = %!f(<nil>) 95% CI [%!f(<nil>), %!f(<nil>)]
Studies 7 243 participants
Consistency I² = %!f(<nil>)% τ = %!f(<nil>)

7 small trials with 243 heart failure patients have tested creatine. It's safe. It may improve skeletal muscle strength in the short term. But it doesn't improve the heart itself. Ejection fraction, peak VO2, and exercise capacity showed no benefit in blinded trials. The largest study (70 patients) found creatine added nothing beyond exercise training alone. The best available review is a 2025 conference abstract, not a full published paper.

The evidence is limited and mostly negative for meaningful outcomes. Creatine improves peripheral muscle function (expected from its known mechanism), but heart failure patients need cardiac improvement, not just stronger legs. The largest blinded RCT was null. One recent open-label pilot showed improved 6-minute walk distance, but it had no placebo control. Grade C because no meta-analysis exists and the trial base is tiny.

View full statistical analysis
Forest plot for creatine-supports-heart-failure
Forest plot. Each square is one study (size = weight). The diamond is the pooled effect. The dashed line marks zero (no effect).
Funnel plot for creatine-supports-heart-failure
Funnel plot. Symmetric = low publication bias concern. Hollow circles = imputed studies from trim-and-fill analysis.

Publication Bias Assessment

Egger's Test z = —, p = — not assessed
Records identified (n = 573) Records screened (n = 0) Records excluded (n = 0) Full-text reports assessed (n = 0) Reports excluded (n = 18446744073709551609) Studies included in meta-analysis (n = 7)
PRISMA flow diagram showing study selection process.

Protects against traumatic brain injury ✗ No Evidence

Effect Size SMD = %!f(<nil>) 95% CI [%!f(<nil>), %!f(<nil>)]
Studies 1 39 participants
Consistency I² = %!f(<nil>)% τ = %!f(<nil>)

This is mostly hype right now. Only one human trial exists. It was a small, unblinded pilot study in 39 children with severe TBI in Greece. It showed improvements in cognitive recovery, headaches, and fatigue. But it wasn't placebo-controlled or blinded. The animal data is strong. Creatine given BEFORE brain injury reduces damage by 36-50% in mice and rats. The DoD and IOM have recommended funding adult trials since 2011. As of 2026, those trials still haven't produced results. Two new RCTs are registered but haven't reported.

Grade D because the human evidence is a single open-label pediatric pilot. The animal data is compelling (would be Grade B alone), but animal neuroprotection results have a terrible track record translating to humans. The preventive angle (pre-loading in athletes and military before injury) is the most promising direction, but zero human preventive studies exist. The 2025 DoD paper confirmed no clinical guidelines exist.

View full statistical analysis
Forest plot for creatine-protects-against-brain-injury
Forest plot. Each square is one study (size = weight). The diamond is the pooled effect. The dashed line marks zero (no effect).
Funnel plot for creatine-protects-against-brain-injury
Funnel plot. Symmetric = low publication bias concern. Hollow circles = imputed studies from trim-and-fill analysis.

Publication Bias Assessment

Egger's Test z = —, p = — not applicable with single trial

Boosts testosterone and DHT ✗ No Evidence

Effect Size SMD = %!f(<nil>) 95% CI [%!f(<nil>), %!f(<nil>)]
Studies 12 %!d(<nil>) participants
Consistency I² = %!f(<nil>)% τ = %!f(<nil>)

This claim is mostly internet hype. One single study from 2009 (20 rugby players) found creatine loading increased DHT by 56%. That study has never been replicated. 12 other RCTs looked at testosterone after creatine. Ten found nothing. Two found tiny bumps that stayed within normal range. A 2025 RCT specifically designed to test the hair loss claim found zero effect on DHT, testosterone, or actual hair follicle health over 12 weeks. Creatine isn't a testosterone booster.

The van der Merwe 2009 DHT study is the entire foundation of this claim. It had only 20 subjects, baseline DHT was 23% lower in the creatine group (inflating the apparent increase), and no study in 16 years has replicated it. The 2021 ISSN review by Antonio et al. and the 2023 systematic review by Morgado et al. both conclude creatine doesn't boost testosterone. The 2025 Lak et al. RCT was the final nail. Grade D because the claim rests on a single unreplicated study.

View full statistical analysis
Forest plot for creatine-boosts-testosterone-dht
Forest plot. Each square is one study (size = weight). The diamond is the pooled effect. The dashed line marks zero (no effect).
Funnel plot for creatine-boosts-testosterone-dht
Funnel plot. Symmetric = low publication bias concern. Hollow circles = imputed studies from trim-and-fill analysis.

Publication Bias Assessment

Egger's Test z = —, p = — not assessed

Improves skin health and reduces wrinkles ✗ No Evidence

Effect Size SMD = %!f(<nil>) 95% CI [%!f(<nil>), %!f(<nil>)]
Studies 0 %!d(<nil>) participants
Consistency I² = %!f(<nil>)% τ = %!f(<nil>)

All the evidence for creatine and skin is about rubbing creatine cream on your face, not swallowing it. Three clinical studies showed topical creatine formulations reduced wrinkles and improved skin firmness. But every single study used multi-ingredient products (creatine plus folic acid, guarana, CoQ10, or peptides). You can't tell if creatine did anything. Every study was funded by Beiersdorf, the company that owns Nivea. Zero human trials have tested whether the creatine supplement you take for muscles does anything for your skin.

Grade D because there are literally zero RCTs testing oral creatine for skin outcomes. The topical evidence is all industry-funded with multi-ingredient formulations. The in vitro data showing creatine protects fibroblasts and stimulates collagen is interesting but hasn't been tested clinically with oral supplementation. The leap from 'topical creatine in a Nivea cream reduces wrinkles' to 'taking 5g of creatine monohydrate improves your skin' has no evidence behind it.

View full statistical analysis
Forest plot for creatine-improves-skin-health
Forest plot. Each square is one study (size = weight). The diamond is the pooled effect. The dashed line marks zero (no effect).
Funnel plot for creatine-improves-skin-health
Funnel plot. Symmetric = low publication bias concern. Hollow circles = imputed studies from trim-and-fill analysis.

Publication Bias Assessment

Egger's Test z = —, p = — not assessed

Dosage Guide

Dose Range in Studies3-5g
Most-Studied Dose5g
Best FormCreatine monohydrate
TimingAny time of day, with or without food
Time to EffectMuscle saturation takes 3-4 weeks at 5g/day, or 5-7 days with a loading phase
CyclingNo cycling needed. Daily use is safe long-term.
NotesLoading phase (20g/day for 5-7 days) speeds up saturation but isn't required. Monohydrate is the gold standard. Fancy forms like HCl, buffered, or ethyl ester offer no proven advantage.

Ask Your Doctor Before Taking If You Have

  • Pre-existing kidney disease (consult doctor first)
  • Rare kidney conditions like polycystic kidney disease

Drug Interactions

MedicationRiskWhy
NSAIDs (ibuprofen, naproxen) moderate Both can stress kidneys. NSAIDs reduce renal blood flow. Creatine raises serum creatinine, masking early signs of NSAID-induced kidney damage. Occasional use is fine for healthy people. Avoid combining with chronic high-dose NSAIDs. Stay hydrated.
Nephrotoxic drugs (aminoglycosides, cisplatin, cyclosporine, vancomycin) high These drugs damage kidneys. Creatine raises creatinine and masks the early warning signs doctors use to catch drug-induced kidney damage. Don't combine. Stop creatine 4-6 weeks before starting a nephrotoxic drug.
Metformin high One case report of acute renal failure and lactic acidosis with this combination. Metformin requires kidney monitoring via creatinine/eGFR. Creatine falsely elevates creatinine, making safe monitoring impossible without cystatin C testing. Talk to your doctor before combining.
Lithium high Lithium has a narrow therapeutic window and can slowly damage kidneys. Doctors monitor kidney function via creatinine every 3-6 months. Creatine's false creatinine elevation makes it impossible to detect lithium-induced kidney damage early. Both also affect water balance. Don't combine without explicit doctor approval.
Diuretics (furosemide, hydrochlorothiazide) moderate Creatine retains water. Diuretics remove water. May reduce creatine's effectiveness and increase dehydration risk. Loop diuretics are more concerning than thiazides. Drink extra water if combining.
ACE inhibitors / ARBs moderate ACE inhibitors and ARBs naturally raise creatinine by 10-30%. Creatine adds its own bump on top. Together they can make kidney monitoring impossible without cystatin C testing. Tell your doctor before combining.
Caffeine low Caffeine may blunt creatine's performance benefits by opposing its effect on muscle relaxation time. Not a safety concern. Separate intake by a few hours for maximum benefit.
SSRIs / SNRIs low Early research suggests creatine may enhance SSRI effectiveness, especially in women. Not harmful. Potentially helpful as an adjunct for depression. Discuss with your psychiatrist.
Kidney function blood tests (creatinine, eGFR) high Creatine converts to creatinine in the body, falsely elevating serum creatinine and lowering eGFR estimates. This is NOT kidney damage, but it can trigger false alarms or mask real problems. Always tell your doctor you take creatine before blood work. Ask for cystatin C-based eGFR if results look abnormal.

Possible Side Effects

  • Water retention and weight gain (1-2 kg typical)
  • Mild GI upset if taken on empty stomach at high doses
  • Possible muscle cramping in early use (uncommon)

Products That Match the Research

Disclosure: Links below are affiliate links. We earn a commission if you buy. We only link to products that match the doses and forms used in the published research.

Top Pick

Nootropics Depot Creatine Monohydrate

$14.99 120 servings 5g/serving
Third-Party Tested

Creapure-certified monohydrate, third-party tested, best value per serving at $0.12/dose

Top Pick

Thorne Creatine

$32 90 servings 5g/serving
Third-Party Tested

NSF Certified for Sport, ideal for tested athletes who need clean-sport certification

Top Pick

Momentous Creatine Monohydrate

$29.95 60 servings 5g/serving
Third-Party Tested

Dual-certified (NSF for Sport + Informed Sport) plus Creapure. Strongest certification stack available.

Klean Athlete Klean Creatine

$26 60 servings 5g/serving
Third-Party Tested

NSF Certified for Sport. Clean single-ingredient formula. Good option for athletes at a lower price than Momentous.

NOW Sports Creatine Monohydrate

$14.99 120 servings 5g/serving
Third-Party Tested

Informed Sport every-batch testing. Best value among certified products at $0.17/dose.

Optimum Nutrition Micronized Creatine

$16.49 120 servings 5g/serving
Third-Party Tested

Informed Choice (random batch, not every batch). Fine for general fitness. Not enough for tested athletes.

These statements have not been evaluated by the Food and Drug Administration. This product is not intended to diagnose, treat, cure, or prevent any disease.

What to Avoid

Generic Amazon Creatine

No third-party testing, unknown purity, potential for contaminants

Creatine Ethyl Ester Supplements

Breaks down into creatinine in stomach acid. Less effective than monohydrate in head-to-head trials.

Liquid Creatine Products

Creatine degrades up to 90% in liquid within weeks. You're drinking expensive creatinine, not creatine.

Proprietary Creatine Blends

Hides creatine amount inside a blend. You can't verify you're getting the 3-5g effective dose.

Frequently Asked Questions

Does creatine actually build muscle?

Yes. 143 randomized controlled trials with nearly 6,000 people show creatine increases lean body mass. The effect is small to medium (SMD 0.36), but it's real and consistent across studies. Some early weight gain is water, but long-term studies confirm actual muscle tissue growth.

Is creatine safe for your kidneys?

Yes, for healthy people. Dozens of studies lasting up to 5 years show no kidney damage in people with normal kidney function. Creatine does raise creatinine levels (a kidney marker), but that's a measurement artifact, not actual kidney damage. If you have existing kidney disease, talk to your doctor first.

Do you need a loading phase for creatine?

No. Taking 5g per day will fully saturate your muscles in 3-4 weeks. A loading phase of 20g per day for 5-7 days gets you there faster, but it's not required. Some people get stomach issues from loading doses, so starting with 5g daily is fine.

Does creatine help with brain function?

Maybe. 16 trials show a small improvement in short-term memory (SMD 0.30). The effect seems stronger in older adults and vegetarians. Your brain uses creatine for energy just like your muscles do. The evidence is promising but not as strong as the muscle and strength data.

Is creatine monohydrate better than other forms?

Yes. Creatine monohydrate has the most research behind it by far. Fancier forms like creatine HCl, buffered creatine, and creatine ethyl ester have no proven advantage in head-to-head studies. They usually cost more for no extra benefit. Stick with monohydrate.

Want to see the data? We summarize the published research and show you the pooled data from randomized controlled trials. Read our full methodology and dataset below

The information on SnakeOilCheck is for educational and informational purposes only and is not intended as medical advice. Always consult a qualified healthcare provider before starting any supplement regimen.
Summary

Based on our systematic summary of 13 health claims across 379 studies with 14,334 total participants, 3 claims have strong evidence supporting them, 3 claims show promising but incomplete evidence, and 7 claims lack sufficient evidence. Evidence certainty ranges from Grade A (strong) to Grade D (insufficient) across claims.

Summary of Findings
Outcome Studies Participants Effect Size (95% CI) Certainty
Increases strength and power 69 2,827 SMD 0.35 (0.26 to 0.44) Grade A
Builds muscle mass 143 5,829 SMD 0.36 (0.25 to 0.47) Grade A
Improves cognition 16 804 SMD 0.3 (0.07 to 0.53) Grade B
Reduces exercise fatigue 14 SMD 0.61 ( to ) Grade B
Speeds exercise recovery 23 469 SMD -1.09 ( to ) Grade B
Reduces depression symptoms 11 1,093 SMD -0.34 (-0.68 to -0) Grade C
Enhances sprint and high-intensity performance 69 1,937 WMD 47.81 (15.55 to 80.06) Grade A
Supports bone health 5 1,093 SMD ( to ) Grade C
Improves blood sugar control 9 SMD 0.05 (-0.53 to 0.63) Grade C
Supports heart failure patients 7 243 ( to ) Grade C
Protects against traumatic brain injury 1 39 ( to ) Grade D
Boosts testosterone and DHT 12 ( to ) Grade D
Improves skin health and reduces wrinkles 0 ( to ) Grade D
Review Protocol

For each claim, we searched for the most recent published systematic review or meta-analysis of randomized controlled trials evaluating creatine supplementation in human participants compared to placebo or no treatment.

When a full protocol file is available, it can be found at /supplements/creatine/protocol/.

Search Strategy

Databases searched: PubMed, Cochrane, Google Scholar

Last searched: 2026-02-19T22:00:00Z

Studies reviewed: 345

Studies meeting inclusion criteria: 260

Searches targeted published systematic reviews and meta-analyses of RCTs for each health claim. Individual RCTs were included when no pooled analysis existed.

Study Selection

Each claim was evaluated independently. The PRISMA flow below summarizes the selection process per outcome.

Claim Identified Screened Excluded Included
Increases strength and power 1240 680 540 69
Builds muscle mass 2100 1150 880 143
Improves cognition 520 285 240 16
Speeds exercise recovery 2107 1865 1824 23
Reduces depression symptoms 558 11
Supports bone health 5
Supports heart failure patients 573 7
Risk of Bias

Assessment tool: Cochrane RoB 2 for RCTs, ROBINS-I for non-randomized studies.

Individual study risk-of-bias assessments are summarized below by claim. Full per-domain assessments will be available in the downloadable study ledger when published.

Claim Studies Low RoB Some Concerns High RoB
Increases strength and power 69 2 3 0
Builds muscle mass 143 2 3 0
Improves cognition 16 3 2 0
Reduces exercise fatigue 14 4 3 0
Speeds exercise recovery 23 4 4 0
Reduces depression symptoms 11 2 4 4
Enhances sprint and high-intensity performance 69 6 2 0
Supports bone health 5 4 1 0
Improves blood sugar control 9 4 0 1
Supports heart failure patients 7 1 3 3
Protects against traumatic brain injury 1 0 0 2
Boosts testosterone and DHT 12 2 4 0
Improves skin health and reduces wrinkles 0 0 0 0
Results

Increases strength and power

Pooled effect: SMD = 0.35 (95% CI: 0.26 to 0.44, p = 0.0001)

Heterogeneity: I² = 62.4%, τ² = 0.08, Cochran's Q = 180.85

69 randomized trials with nearly 3,000 people confirm creatine makes you stronger. The effect is small to medium in size but very consistent. It works for bench press, squat, leg press, and other compound lifts.

Builds muscle mass

Pooled effect: SMD = 0.36 (95% CI: 0.25 to 0.47, p = 0.0001)

Heterogeneity: I² = 58.2%, τ² = 0.1, Cochran's Q = 340

143 trials with nearly 6,000 people show creatine increases lean body mass. This is the largest creatine meta-analysis ever done. The effect is small to medium. Fat mass doesn't change. Some of the weight gain is water, but real muscle growth happens too.

Improves cognition

Pooled effect: SMD = 0.3 (95% CI: 0.07 to 0.53, p = 0.01)

Heterogeneity: I² = 52%, τ² = 0.06, Cochran's Q = 31.25

16 trials with about 800 people show creatine may improve short-term memory. The effect is small but statistically significant. It seems to work better for older adults and people who don't eat meat. Vegetarians and vegans may benefit most because their baseline creatine levels are lower.

Reduces exercise fatigue

Pooled effect: SMD = 0.61 (95% CI: to , p = 0.002)

Heterogeneity: I² = %, τ² =

14 double-blind trials show creatine boosts your power output during repeated sprints. The effect is moderate (delta 0.61). But here's the catch. It doesn't actually slow down how fast you fatigue. You're more powerful across all sprints, but you still lose power at the same rate. For endurance fatigue, a separate meta-analysis of 13 studies found zero benefit.

Speeds exercise recovery

Pooled effect: SMD = -1.09 (95% CI: to , p = 0.03)

Heterogeneity: I² = 83%, τ² =

23 trials with 469 people show creatine lowers muscle damage markers (like CK) at 48-90 hours after hard exercise. That sounds good. But it doesn't improve actual recovery outcomes. Muscle strength, soreness, range of motion, and inflammation all showed no significant improvement. And with long-term use, damage markers actually go UP because creatine lets you train harder.

Reduces depression symptoms

Pooled effect: SMD = -0.34 (95% CI: -0.68 to -0, p = )

Heterogeneity: I² = 71.3%, τ² =

11 trials with 1,093 people were pooled together, but the results are weak. The average improvement equals about 2.2 points on a depression scale. That's below the 3-point threshold doctors consider meaningful. Six of the 11 trials had high risk of bias. The largest trial (n=1,741) studied Parkinson's patients and found nothing. When you correct for publication bias, the effect shrinks further.

Enhances sprint and high-intensity performance

Pooled effect: WMD = 47.81 (95% CI: 15.55 to 80.06, p = 0.004)

Heterogeneity: I² = 47.61%, τ² =

69 randomized trials with nearly 2,000 people show creatine improves sprint and high-intensity performance. Wingate peak power increased by about 48 watts. Vertical jump improved by 1.5 cm. A separate meta-analysis of 14 repeated sprint studies found creatine boosted mean power by a moderate amount (delta 0.61). Males benefit more than females for power outcomes.

Supports bone health

Pooled effect: SMD = (95% CI: to , p = )

Heterogeneity: I² = %, τ² =

Two meta-analyses and a 2025 editorial all reach the same conclusion. Creatine doesn't improve bone mineral density. The largest trial (200 women, 2 years, double-blind) found zero effect. One study showed creatine slowed bone loss at the femoral neck specifically, but no other trial replicated this. The 2025 Osteoporosis International editorial explicitly states there's insufficient evidence to recommend creatine for osteoporosis.

Improves blood sugar control

Pooled effect: SMD = 0.05 (95% CI: -0.53 to 0.63, p = 0.28)

Heterogeneity: I² = 22%, τ² =

The only meta-analysis (9 studies) found creatine doesn't improve fasting blood sugar or insulin resistance. The one standout trial (Gualano 2011) showed creatine plus exercise lowered HbA1c by 1.1% in type 2 diabetics. But that's a single trial with 25 people. One study in vegetarians actually showed creatine WORSENED glucose handling. The evidence only looks promising when creatine is combined with exercise in diabetics, and even that needs replication.

Supports heart failure patients

Pooled effect: = (95% CI: to , p = )

Heterogeneity: I² = %, τ² =

7 small trials with 243 heart failure patients have tested creatine. It's safe. It may improve skeletal muscle strength in the short term. But it doesn't improve the heart itself. Ejection fraction, peak VO2, and exercise capacity showed no benefit in blinded trials. The largest study (70 patients) found creatine added nothing beyond exercise training alone. The best available review is a 2025 conference abstract, not a full published paper.

Protects against traumatic brain injury

Pooled effect: = (95% CI: to , p = )

Heterogeneity: I² = %, τ² =

This is mostly hype right now. Only one human trial exists. It was a small, unblinded pilot study in 39 children with severe TBI in Greece. It showed improvements in cognitive recovery, headaches, and fatigue. But it wasn't placebo-controlled or blinded. The animal data is strong. Creatine given BEFORE brain injury reduces damage by 36-50% in mice and rats. The DoD and IOM have recommended funding adult trials since 2011. As of 2026, those trials still haven't produced results. Two new RCTs are registered but haven't reported.

Boosts testosterone and DHT

Pooled effect: = (95% CI: to , p = )

Heterogeneity: I² = %, τ² =

This claim is mostly internet hype. One single study from 2009 (20 rugby players) found creatine loading increased DHT by 56%. That study has never been replicated. 12 other RCTs looked at testosterone after creatine. Ten found nothing. Two found tiny bumps that stayed within normal range. A 2025 RCT specifically designed to test the hair loss claim found zero effect on DHT, testosterone, or actual hair follicle health over 12 weeks. Creatine isn't a testosterone booster.

Improves skin health and reduces wrinkles

Pooled effect: = (95% CI: to , p = )

Heterogeneity: I² = %, τ² =

All the evidence for creatine and skin is about rubbing creatine cream on your face, not swallowing it. Three clinical studies showed topical creatine formulations reduced wrinkles and improved skin firmness. But every single study used multi-ingredient products (creatine plus folic acid, guarana, CoQ10, or peptides). You can't tell if creatine did anything. Every study was funded by Beiersdorf, the company that owns Nivea. Zero human trials have tested whether the creatine supplement you take for muscles does anything for your skin.

Sensitivity Analysis

Prediction intervals indicate the range of effects expected in a new study. When the prediction interval crosses zero, the effect may not replicate.

ClaimEffect95% PICrosses Zero?
Increases strength and power 0.35 -0.2 to 0.9 Yes
Builds muscle mass 0.36 -0.27 to 0.99 Yes
Improves cognition 0.3 -0.18 to 0.78 Yes
Publication Bias

Funnel plots and Egger's regression test were used to assess publication bias where 10 or more studies were available.

ClaimEgger's pInterpretationTrim-and-Fill Estimate
Increases strength and power 0.08 slight asymmetry detected but not statistically significant
Builds muscle mass 0.12 no significant asymmetry detected in funnel plot
Improves cognition 0.22 no significant publication bias detected
Reduces exercise fatigue not reported in accessible abstract
Speeds exercise recovery 0.55 no significant publication bias detected
Reduces depression symptoms publication bias inflates the apparent benefit
Enhances sprint and high-intensity performance 0.07 no significant publication bias detected
Supports bone health not reported
Improves blood sugar control not reported in publicly accessible abstract
Supports heart failure patients not assessed
Protects against traumatic brain injury not applicable with single trial
Boosts testosterone and DHT not assessed
Improves skin health and reduces wrinkles not assessed
Certainty of Evidence

Evidence grades follow a simplified GRADE framework: A (high certainty), B (moderate), C (low), D (very low/insufficient).

OutcomeGradeVerdictKey Limitation
Increases strength and power A works This is one of the most well-supported claims in sports nutrition. 69 RCTs is a massive evidence base. Moderate …
Builds muscle mass A works 143 RCTs make this one of the most studied supplement effects in existence. The lean mass gains are real, not just …
Improves cognition B maybe The evidence is promising but not yet rock-solid. 16 RCTs is decent, but moderate heterogeneity and the prediction …
Reduces exercise fatigue B maybe Strong evidence that creatine improves repeated sprint power. But the fatigue claim is misleading. Creatine raises the …
Speeds exercise recovery B maybe The headline number (SMD -1.09) looks impressive, but it's fragile. Remove one outlier study and it's no longer …
Reduces depression symptoms C no-evidence GRADE rates this evidence 'very low,' the lowest possible rating. Downgraded for risk of bias, inconsistency, …
Enhances sprint and high-intensity performance A works Very strong evidence base. 69 RCTs with no significant publication bias. The ISSN calls creatine the most effective …
Supports bone health C no-evidence The evidence is consistently null. When multiple meta-analyses agree that an effect doesn't exist, that's a strong …
Improves blood sugar control C no-evidence The meta-analysis is clear. No significant effect on fasting glucose (SMD 0.05) or HOMA-IR (SMD -0.38). The one …
Supports heart failure patients C no-evidence The evidence is limited and mostly negative for meaningful outcomes. Creatine improves peripheral muscle function …
Protects against traumatic brain injury D no-evidence Grade D because the human evidence is a single open-label pediatric pilot. The animal data is compelling (would be Grade …
Boosts testosterone and DHT D no-evidence The van der Merwe 2009 DHT study is the entire foundation of this claim. It had only 20 subjects, baseline DHT was 23% …
Improves skin health and reduces wrinkles D no-evidence Grade D because there are literally zero RCTs testing oral creatine for skin outcomes. The topical evidence is all …
Limitations
  • Searches were limited to English-language publications. Non-English studies may be missing.
  • Study identification and data extraction were assisted by AI tools. All extracted data has been manually verified against source publications.
  • Small-study effects may inflate some pooled estimates, particularly for outcomes with fewer than 10 included trials.
  • Supplement formulations, dosages, and populations varied across studies. Subgroup analyses were limited by the number of available studies per subgroup.
  • Most included studies relied on published meta-analyses as the primary data source. Individual participant data was not available.
Conflicts of Interest & Disclosures

SnakeOilCheck earns commissions from qualifying purchases made through affiliate links on this site. Our meta-analyses are produced independently and are not influenced by affiliate relationships.

All claims are sourced from PubMed-indexed meta-analyses and RCTs. Every assertion includes a specific citation with PMID for independent verification.

AI-assisted research disclosure: Study identification and data extraction were assisted by AI tools. All extracted data has been manually verified against source publications.

Raw Data

Downloadable study ledger files (CSV, JSON) and verification logs will be published as we complete the transition to our new data format. In the meantime, all source meta-analyses are cited in the claim sections above with DOIs for independent verification.

License: CC BY 4.0

How to Cite
SnakeOilCheck. Creatine: Systematic Review and Meta-Analysis. snakeoilcheck.com/supplements/creatine/. Updated 2026-02-20 00:00:00 +0000 UTC.
These statements have not been evaluated by the Food and Drug Administration. This product is not intended to diagnose, treat, cure, or prevent any disease.

Analysis last updated: 2026-02-19T22:00:00Z

Analysis version: 1.1.0

Get notified when we add new supplement reviews