FitProven
Evidence Breakdown
Based on 5 studies
THE BIGGER QUESTION
Whether this one statement holds is settled above. What to actually do about it is a wider question, weighed across every claim that bears on it.
- Should women take creatine? — this claim is one of 3 weighed there
Evidence map
For & against, at a glance
Double-Blind RCT
Pro
Brooks SJ et al. · 2023Journal of the International Society of Sports NutritionRandomized, double-blind, placebo-controlled trial in 13 female collegiate dancers (7 creatine at 0.1 g/kg/day, 6 placebo) over 6 weeks, with significant group × time interactions favoring creatine for total body water (32.2 → 32.7 kg, p = 0.024), body mass (58.3 → 59.0 kg, p = 0.010) and DXA lean mass (39.8 → 41.5 kg, p = 0.020), while placebo was flat or declined. Fat mass was unchanged (p = 0.393). The authors explicitly attribute part of the apparent lean-mass gain to the extra body water — real but small (~0.5 kg of fluid, ~0.7 kg of scale weight), and distributed as tissue water rather than measured as visible swelling.
0.21
Randomized, double-blind, placebo-controlled trial in 13 female collegiate dancers (7 creatine at 0.1 g/kg/day, 6 placebo) over 6 weeks, with significant group × time interactions favoring creatine for total body water (32.2 → 32.7 kg, p = 0.024), body mass (58.3 → 59.0 kg, p = 0.010) and DXA lean mass (39.8 → 41.5 kg, p = 0.020), while placebo was flat or declined. Fat mass was unchanged (p = 0.393). The authors explicitly attribute part of the apparent lean-mass gain to the extra body water — real but small (~0.5 kg of fluid, ~0.7 kg of scale weight), and distributed as tissue water rather than measured as visible swelling.
Design Double-Blind RCT (0.85) × quality 0.25 = impact 0.21
View sourceDouble-Blind RCT
Neutral
Powers ME et al. · 2003Journal of Athletic TrainingRandomized, double-blind, placebo-controlled trial in 32 resistance-trained adults (16 men, 16 women) over 28 days: 25 g/day loading for 7 days, then 5 g/day for 21 days, with body water measured directly by deuterium-oxide and sodium-bromide dilution rather than estimated by impedance. Creatine significantly raised muscle creatine concentration, body mass (+1.31 kg) and total body water (+2.04 L, +4.86%), and muscle creatine uptake in women did not differ from men — so the "water weight" half of the claim is real. The distribution half is not: the authors' stated conclusion is that **"fluid distribution was not changed"**. The intracellular-water increase (+1.13 L, +4.62%) was **not statistically significant**, extracellular water showed no significant change, and the ICW rise accounted for 55.4% of the TBW gain — close to the ~2/3 that is normal for body water, i.e. the extra fluid distributed roughly the way body water always does. There is no evidence here of fluid piling into the subcutaneous compartment that would make someone look puffy, but neither is there a demonstrated significant intracellular shift.
0.60
Randomized, double-blind, placebo-controlled trial in 32 resistance-trained adults (16 men, 16 women) over 28 days: 25 g/day loading for 7 days, then 5 g/day for 21 days, with body water measured directly by deuterium-oxide and sodium-bromide dilution rather than estimated by impedance. Creatine significantly raised muscle creatine concentration, body mass (+1.31 kg) and total body water (+2.04 L, +4.86%), and muscle creatine uptake in women did not differ from men — so the "water weight" half of the claim is real. The distribution half is not: the authors' stated conclusion is that **"fluid distribution was not changed"**. The intracellular-water increase (+1.13 L, +4.62%) was **not statistically significant**, extracellular water showed no significant change, and the ICW rise accounted for 55.4% of the TBW gain — close to the ~2/3 that is normal for body water, i.e. the extra fluid distributed roughly the way body water always does. There is no evidence here of fluid piling into the subcutaneous compartment that would make someone look puffy, but neither is there a demonstrated significant intracellular shift.
Design Double-Blind RCT (0.85) × quality 0.70 = impact 0.60
View sourceDouble-Blind RCT
Neutral
Moore SR et al. · 2023NutrientsRandomized, double-blind, placebo-controlled trial in 30 moderately active women (15 creatine, 15 placebo) who completed a 5-day loading protocol (20 g/day) in both the follicular and luteal menstrual phases, with a ≥4-week washout. In the luteal phase creatine increased total body water (+0.83 L vs −0.62 L placebo, p = 0.021), intracellular fluid (+0.74 L, p = 0.041) **and extracellular fluid** (+0.46 L vs −0.19 L, p = 0.013); the follicular phase showed no differences. Despite the measurable fluid shifts, body mass did not differ by condition (p = 0.427) or phase (p = 0.528) — so short-term loading can move fluid, including extracellular fluid, without producing weight gain.
0.43
Randomized, double-blind, placebo-controlled trial in 30 moderately active women (15 creatine, 15 placebo) who completed a 5-day loading protocol (20 g/day) in both the follicular and luteal menstrual phases, with a ≥4-week washout. In the luteal phase creatine increased total body water (+0.83 L vs −0.62 L placebo, p = 0.021), intracellular fluid (+0.74 L, p = 0.041) **and extracellular fluid** (+0.46 L vs −0.19 L, p = 0.013); the follicular phase showed no differences. Despite the measurable fluid shifts, body mass did not differ by condition (p = 0.427) or phase (p = 0.528) — so short-term loading can move fluid, including extracellular fluid, without producing weight gain.
Design Double-Blind RCT (0.85) × quality 0.50 = impact 0.43
View sourceMeta-Analysis
Con
de Guingand DL et al. · 2020NutrientsSystematic review and meta-analysis of 29 studies (18 ergogenic, 11 therapeutic) covering 951 female participants aged 13–68 taking oral creatine monohydrate. Pooled body-weight change did not differ meaningfully between creatine and placebo groups (mean differences in the 1.2–1.4 kg range with confidence intervals crossing zero), and there was no significant excess of total adverse events (RR 1.24) or gastrointestinal disturbance (RR 1.09) — the symptom category that would capture bloating. This is the largest female-specific safety synthesis and it does not find the weight gain or the discomfort the claim predicts.
0.80
Systematic review and meta-analysis of 29 studies (18 ergogenic, 11 therapeutic) covering 951 female participants aged 13–68 taking oral creatine monohydrate. Pooled body-weight change did not differ meaningfully between creatine and placebo groups (mean differences in the 1.2–1.4 kg range with confidence intervals crossing zero), and there was no significant excess of total adverse events (RR 1.24) or gastrointestinal disturbance (RR 1.09) — the symptom category that would capture bloating. This is the largest female-specific safety synthesis and it does not find the weight gain or the discomfort the claim predicts.
Design Meta-Analysis (1.0) × quality 0.80 = impact 0.80
View sourceNarrative Review
Con
Antonio J et al. · 2021Journal of the International Society of Sports NutritionExpert narrative review from the ISSN addressing 11 common creatine misconceptions, including water retention. It concludes that any fluid increase is short-term and primarily intracellular during the first days of loading, and that across 5–10 week trials there is no significant change in intracellular water, extracellular water or total body water relative to muscle mass. Useful as a synthesis of the mechanism, but it is an expert review by creatine researchers rather than primary data, so it carries low weight here.
0.11
Expert narrative review from the ISSN addressing 11 common creatine misconceptions, including water retention. It concludes that any fluid increase is short-term and primarily intracellular during the first days of loading, and that across 5–10 week trials there is no significant change in intracellular water, extracellular water or total body water relative to muscle mass. Useful as a synthesis of the mechanism, but it is an expert review by creatine researchers rather than primary data, so it carries low weight here.
Design Narrative Review (0.3) × quality 0.35 = impact 0.11
View sourceDouble-Blind RCT
Pro
Brooks SJ et al. · 2023Journal of the International Society of Sports NutritionRandomized, double-blind, placebo-controlled trial in 13 female collegiate dancers (7 creatine at 0.1 g/kg/day, 6 placebo) over 6 weeks, with significant group × time interactions favoring creatine for total body water (32.2 → 32.7 kg, p = 0.024), body mass (58.3 → 59.0 kg, p = 0.010) and DXA lean mass (39.8 → 41.5 kg, p = 0.020), while placebo was flat or declined. Fat mass was unchanged (p = 0.393). The authors explicitly attribute part of the apparent lean-mass gain to the extra body water — real but small (~0.5 kg of fluid, ~0.7 kg of scale weight), and distributed as tissue water rather than measured as visible swelling.
0.21
Randomized, double-blind, placebo-controlled trial in 13 female collegiate dancers (7 creatine at 0.1 g/kg/day, 6 placebo) over 6 weeks, with significant group × time interactions favoring creatine for total body water (32.2 → 32.7 kg, p = 0.024), body mass (58.3 → 59.0 kg, p = 0.010) and DXA lean mass (39.8 → 41.5 kg, p = 0.020), while placebo was flat or declined. Fat mass was unchanged (p = 0.393). The authors explicitly attribute part of the apparent lean-mass gain to the extra body water — real but small (~0.5 kg of fluid, ~0.7 kg of scale weight), and distributed as tissue water rather than measured as visible swelling.
Design Double-Blind RCT (0.85) × quality 0.25 = impact 0.21
View sourceDouble-Blind RCT
Neutral
Powers ME et al. · 2003Journal of Athletic TrainingRandomized, double-blind, placebo-controlled trial in 32 resistance-trained adults (16 men, 16 women) over 28 days: 25 g/day loading for 7 days, then 5 g/day for 21 days, with body water measured directly by deuterium-oxide and sodium-bromide dilution rather than estimated by impedance. Creatine significantly raised muscle creatine concentration, body mass (+1.31 kg) and total body water (+2.04 L, +4.86%), and muscle creatine uptake in women did not differ from men — so the "water weight" half of the claim is real. The distribution half is not: the authors' stated conclusion is that **"fluid distribution was not changed"**. The intracellular-water increase (+1.13 L, +4.62%) was **not statistically significant**, extracellular water showed no significant change, and the ICW rise accounted for 55.4% of the TBW gain — close to the ~2/3 that is normal for body water, i.e. the extra fluid distributed roughly the way body water always does. There is no evidence here of fluid piling into the subcutaneous compartment that would make someone look puffy, but neither is there a demonstrated significant intracellular shift.
0.60
Randomized, double-blind, placebo-controlled trial in 32 resistance-trained adults (16 men, 16 women) over 28 days: 25 g/day loading for 7 days, then 5 g/day for 21 days, with body water measured directly by deuterium-oxide and sodium-bromide dilution rather than estimated by impedance. Creatine significantly raised muscle creatine concentration, body mass (+1.31 kg) and total body water (+2.04 L, +4.86%), and muscle creatine uptake in women did not differ from men — so the "water weight" half of the claim is real. The distribution half is not: the authors' stated conclusion is that **"fluid distribution was not changed"**. The intracellular-water increase (+1.13 L, +4.62%) was **not statistically significant**, extracellular water showed no significant change, and the ICW rise accounted for 55.4% of the TBW gain — close to the ~2/3 that is normal for body water, i.e. the extra fluid distributed roughly the way body water always does. There is no evidence here of fluid piling into the subcutaneous compartment that would make someone look puffy, but neither is there a demonstrated significant intracellular shift.
Design Double-Blind RCT (0.85) × quality 0.70 = impact 0.60
View sourceDouble-Blind RCT
Neutral
Moore SR et al. · 2023NutrientsRandomized, double-blind, placebo-controlled trial in 30 moderately active women (15 creatine, 15 placebo) who completed a 5-day loading protocol (20 g/day) in both the follicular and luteal menstrual phases, with a ≥4-week washout. In the luteal phase creatine increased total body water (+0.83 L vs −0.62 L placebo, p = 0.021), intracellular fluid (+0.74 L, p = 0.041) **and extracellular fluid** (+0.46 L vs −0.19 L, p = 0.013); the follicular phase showed no differences. Despite the measurable fluid shifts, body mass did not differ by condition (p = 0.427) or phase (p = 0.528) — so short-term loading can move fluid, including extracellular fluid, without producing weight gain.
0.43
Randomized, double-blind, placebo-controlled trial in 30 moderately active women (15 creatine, 15 placebo) who completed a 5-day loading protocol (20 g/day) in both the follicular and luteal menstrual phases, with a ≥4-week washout. In the luteal phase creatine increased total body water (+0.83 L vs −0.62 L placebo, p = 0.021), intracellular fluid (+0.74 L, p = 0.041) **and extracellular fluid** (+0.46 L vs −0.19 L, p = 0.013); the follicular phase showed no differences. Despite the measurable fluid shifts, body mass did not differ by condition (p = 0.427) or phase (p = 0.528) — so short-term loading can move fluid, including extracellular fluid, without producing weight gain.
Design Double-Blind RCT (0.85) × quality 0.50 = impact 0.43
View sourceMeta-Analysis
Con
de Guingand DL et al. · 2020NutrientsSystematic review and meta-analysis of 29 studies (18 ergogenic, 11 therapeutic) covering 951 female participants aged 13–68 taking oral creatine monohydrate. Pooled body-weight change did not differ meaningfully between creatine and placebo groups (mean differences in the 1.2–1.4 kg range with confidence intervals crossing zero), and there was no significant excess of total adverse events (RR 1.24) or gastrointestinal disturbance (RR 1.09) — the symptom category that would capture bloating. This is the largest female-specific safety synthesis and it does not find the weight gain or the discomfort the claim predicts.
0.80
Systematic review and meta-analysis of 29 studies (18 ergogenic, 11 therapeutic) covering 951 female participants aged 13–68 taking oral creatine monohydrate. Pooled body-weight change did not differ meaningfully between creatine and placebo groups (mean differences in the 1.2–1.4 kg range with confidence intervals crossing zero), and there was no significant excess of total adverse events (RR 1.24) or gastrointestinal disturbance (RR 1.09) — the symptom category that would capture bloating. This is the largest female-specific safety synthesis and it does not find the weight gain or the discomfort the claim predicts.
Design Meta-Analysis (1.0) × quality 0.80 = impact 0.80
View sourceNarrative Review
Con
Antonio J et al. · 2021Journal of the International Society of Sports NutritionExpert narrative review from the ISSN addressing 11 common creatine misconceptions, including water retention. It concludes that any fluid increase is short-term and primarily intracellular during the first days of loading, and that across 5–10 week trials there is no significant change in intracellular water, extracellular water or total body water relative to muscle mass. Useful as a synthesis of the mechanism, but it is an expert review by creatine researchers rather than primary data, so it carries low weight here.
0.11
Expert narrative review from the ISSN addressing 11 common creatine misconceptions, including water retention. It concludes that any fluid increase is short-term and primarily intracellular during the first days of loading, and that across 5–10 week trials there is no significant change in intracellular water, extracellular water or total body water relative to muscle mass. Useful as a synthesis of the mechanism, but it is an expert review by creatine researchers rather than primary data, so it carries low weight here.
Design Narrative Review (0.3) × quality 0.35 = impact 0.11
View sourceTap any node to expand its detail.
Evidence
PRO (1)
PRO Double-Blind RCTn=130.25 Brooks SJ, Candow DG et al. (2023)
Over 6 weeks of 0.1 g/kg/day creatine, significant group x time interactions for total body water (32.2 to 32.7 kg, p = 0.024), body mass (58.3 to 59.0 kg, p = 0.010) and DXA lean mass (39.8 to 41.5 kg, p = 0.020); fat mass unchanged (p = 0.393). About 0.5 kg of extra fluid and 0.7 kg of scale weight
Randomized, double-blind, placebo-controlled trial in 13 female collegiate dancers (7 creatine at 0.1 g/kg/day, 6 placebo) over 6 weeks, with significant group × time interactions favoring creatine for total body water (32.2 → 32.7 kg, p = 0.024), body mass (58.3 → 59.0 kg, p = 0.010) and DXA lean mass (39.8 → 41.5 kg, p = 0.020), while placebo was flat or declined. Fat mass was unchanged (p = 0.393). The authors explicitly attribute part of the apparent lean-mass gain to the extra body water — real but small (~0.5 kg of fluid, ~0.7 kg of scale weight), and distributed as tissue water rather than measured as visible swelling.
Weighted 0.25 — n=13 completers split 7 creatine / 6 placebo — severely underpowered, and a significant group x time interaction from six people per arm is fragile however well the blinding was done. Independently funded (university undergraduate research office), double-blind, with DXA, and it is one of very few creatine hydration studies in women, which is why it is worth keeping at all. Under the n-under-10-per-arm rule this sits in the bottom band.
Funding: Office of Undergraduate Research, University of Idaho
Journal of the International Society of Sports Nutrition
AGAINST (2)
AGAINST Meta-Analysisn=9510.80 de Guingand DL, Palmer KR et al. (2020)
No excess of total adverse events (RR 1.24, 95% CI 0.51-2.98) or gastrointestinal disturbance (RR 1.09, 95% CI 0.53-2.24) vs placebo, and no meaningful difference in body-weight change; no deaths or serious adverse events
Systematic review and meta-analysis of 29 studies (18 ergogenic, 11 therapeutic) covering 951 female participants aged 13–68 taking oral creatine monohydrate. Pooled body-weight change did not differ meaningfully between creatine and placebo groups (mean differences in the 1.2–1.4 kg range with confidence intervals crossing zero), and there was no significant excess of total adverse events (RR 1.24) or gastrointestinal disturbance (RR 1.09) — the symptom category that would capture bloating. This is the largest female-specific safety synthesis and it does not find the weight gain or the discomfort the claim predicts.
Weighted 0.80 — 951 female participants across 29 studies — the largest female-specific creatine safety synthesis there is, independently funded with no conflicts. Discounted from the top band because the confidence intervals are wide (the adverse-event RR spans 0.51-2.98, so it cannot exclude real harm), adverse events were inconsistently defined and reported across the component studies, and safety endpoints were rarely their primary outcome.
Funding: none declared (no external funding; public infrastructure and NHMRC fellowship support only; authors declared no conflicts)
Nutrients
AGAINST Narrative Review0.35 Antonio J, Candow DG et al. (2021)
Concludes creatine-related fluid gain is short-term and primarily intracellular during loading, and that across 5-10 week trials there is no significant change in intracellular water, extracellular water or total body water relative to muscle mass
Expert narrative review from the ISSN addressing 11 common creatine misconceptions, including water retention. It concludes that any fluid increase is short-term and primarily intracellular during the first days of loading, and that across 5–10 week trials there is no significant change in intracellular water, extracellular water or total body water relative to muscle mass. Useful as a synthesis of the mechanism, but it is an expert review by creatine researchers rather than primary data, so it carries low weight here.
Weighted 0.35 — No sample size — this is a narrative review with no pooled data, so the field is omitted. The paper states no funding, but the competing-interests section is extensive: the lead author is CEO of the ISSN (which is sponsored by creatine manufacturers) and multiple co-authors sit on the scientific advisory board of Alzchem, a creatine manufacturer. Useful mechanistic synthesis, high risk of advocacy bias, no primary data — hence a low multiplier on an already low base.
Funding: none declared
Journal of the International Society of Sports Nutrition
NEUTRAL (2)
NEUTRAL Double-Blind RCTn=32INDUSTRY0.70 Powers ME, Arnold BL et al. (2003)
Body mass +1.31 kg and total body water +2.04 L (+4.86%) over 28 days (both significant vs placebo). Fluid DISTRIBUTION was unchanged: the +1.13 L (+4.62%) rise in intracellular water was NOT statistically significant, extracellular water did not change significantly, and the ICW rise accounted for 55.4% of the TBW gain — consistent with normal fluid distribution. Muscle creatine uptake did not differ between women and men
Randomized, double-blind, placebo-controlled trial in 32 resistance-trained adults (16 men, 16 women) over 28 days: 25 g/day loading for 7 days, then 5 g/day for 21 days, with body water measured directly by deuterium-oxide and sodium-bromide dilution rather than estimated by impedance. Creatine significantly raised muscle creatine concentration, body mass (+1.31 kg) and total body water (+2.04 L, +4.86%), and muscle creatine uptake in women did not differ from men — so the "water weight" half of the claim is real. The distribution half is not: the authors' stated conclusion is that "fluid distribution was not changed". The intracellular-water increase (+1.13 L, +4.62%) was not statistically significant, extracellular water showed no significant change, and the ICW rise accounted for 55.4% of the TBW gain — close to the ~2/3 that is normal for body water, i.e. the extra fluid distributed roughly the way body water always does. There is no evidence here of fluid piling into the subcutaneous compartment that would make someone look puffy, but neither is there a demonstrated significant intracellular shift.
Weighted 0.70 — n=32 (16 men, 16 women) over 28 days — small, but randomised, double-blind and placebo-controlled, with body-water compartments measured directly rather than inferred, and funded by a professional body and the NIH rather than a supplement maker. Held to 0.70 by the short duration and the narrow sample (young resistance-trained adults).
Funding: National Athletic Trainers' Association Research and Education Foundation, and NIH grant MO1 RR00847 to the University of Virginia GCRC (both public/independent). In-kind: "We also thank Experimental and Applied Sciences (EAS) Inc, Golden, CO, for supplying the Cr monohydrate used in the investigation."
Journal of Athletic Training
NEUTRAL Double-Blind RCTn=300.50 Moore SR, Gordon AN et al. (2023)
in the luteal phase creatine raised total body water (+0.83 L vs -0.62 L placebo, p=0.021), intracellular fluid (+0.74 L, p=0.041) and extracellular fluid (+0.46 L vs -0.19 L, p=0.013); no follicular-phase differences and no change in body mass
Randomized, double-blind, placebo-controlled trial in 30 moderately active women (15 creatine, 15 placebo) who completed a 5-day loading protocol (20 g/day) in both the follicular and luteal menstrual phases, with a ≥4-week washout. In the luteal phase creatine increased total body water (+0.83 L vs −0.62 L placebo, p = 0.021), intracellular fluid (+0.74 L, p = 0.041) and extracellular fluid (+0.46 L vs −0.19 L, p = 0.013); the follicular phase showed no differences. Despite the measurable fluid shifts, body mass did not differ by condition (p = 0.427) or phase (p = 0.528) — so short-term loading can move fluid, including extracellular fluid, without producing weight gain.
Weighted 0.50 — 43 enrolled but only 30 analysed (15 creatine, 15 placebo), and the design has a specific hole for a fluid-retention question: progesterone — a principal driver of fluid balance — was never measured, estrogen was measured only after supplementation, ovulation was not confirmed with kits, and the sample mixed naturally menstruating women with hormonal-contraceptive users. The creatine was donated by AlzChem and the senior author serves as a scientific advisor to AlzChem, so the industry tie points in the direction of the reassuring no-weight-gain finding.
Funding: no external funding; creatine monohydrate donated by AlzChem Trostberg GmbH
Nutrients