Creatine works in women too, but the effect is small

Creatine is among the most-studied sports supplements in existence, yet trials conducted specifically in women are scattered across different sports, training states and outcome types, making the overall picture hard to read. A previous systematic review gathered 27 female-specific studies but could only describe the results narratively; the average size of the effect across related outcomes was never calculated. This review set out to put a number on it: how much does creatine actually help women who exercise or train, across performance, physiology and body composition?

The reviewers searched PubMed, Scopus, Web of Science, SPORTDiscus and CINAHL for experimental studies published between 1 January 1996 and 5 March 2026, identifying 13,244 records. After screening, 160 reports were read in full and 34 studies with 692 female or female-eligible participants entered the systematic review. Twenty-one of those studies contributed 170 comparable creatine-versus-control effects from 424 analyzable participants. These were combined using multilevel statistical models designed to handle multiple related outcomes from the same study, with additional checks on reconstructed data, and the certainty of each estimate was formally graded.

The overall pooled effect was small and positive: a Hedges g of 0.23, with a plausible range of 0.09 to 0.37. Performance outcomes came in at g of 0.30 and body composition at 0.24, though the body-composition model mixed outcomes that do not all point in one favourable direction. The physiological estimate was genuinely uncertain at 0.10, with a range that crossed zero. Certainty was rated low for performance and body composition and very low for the overall and physiological evidence, driven by risk-of-bias concerns and wide variation in protocols and outcomes.

This analysis converts a previously narrative body of female-specific creatine research into a quantitative estimate for the first time, and the answer, small positive effects on selected strength, power, high-intensity and lean-mass-related measures, broadly matches expectations from the wider creatine literature. What it cannot yet do is say which sports, protocols or outcomes benefit most, because the underlying studies vary too much and their methodological quality is mixed. Larger, better-standardised female-specific trials are the obvious next step.

For women who train, this is a reason to think creatine's benefits are not a male-only story: the pooled female-specific data leans positive across strength, power and lean-mass measures. At the same time, the effects are small and the reviewers' own confidence in them is low, so expectations should be sized accordingly. Nothing here identifies a best protocol or a guaranteed outcome; it establishes direction, not magnitude.