Evidence Trace, complete trial record
In plain terms
In plain terms: Creatine Monohydrate has strong evidence from human clinical trials. The benefit shown in those trials is muscle strength with resistance training in older adults; graded per use, and the bone density claim is the one that failed if that describes your situation.
Creatine Monohydrate
A
Replicated: replicated trials on a qualifying endpoint
The endpoint this grade is based on: Objectively measured functional outcome, Muscle strength with resistance training in older adults; graded per use, and the bone density claim is the one that failed
The evidence as the record states it
Graded by use rather than as one letter, because the four things creatine is marketed for do not share an evidence base. Strength holds Grade A by the functional route, on two independent randomized trials that measured strength with an instrument: Brose et al. at McMaster University (J Gerontol A Biol Sci Med Sci, 2003; PMID 12560406) found a greater gain in isometric knee-extension strength than placebo in 28 adults over 65 across 14 weeks of resistance training, and Chrusch et al. at the University of Saskatchewan (Med Sci Sports Exerc, 2001; PMID 11740307) found greater leg-press and knee-extension one-repetition maximum gains than placebo in men of mean age 70 across 12 weeks of training. Both trials are small. Two pooled analyses agree on direction, Forbes et al. (Nutrients, 2021; PMID 34199420) and Chilibeck et al. (Open Access J Sports Med, 2017; PMID 29138605), but they share an author (Candow), so they are not a second independent line of evidence. Naddafha et al. (J Int Soc Sports Nutr, 2026; PMID 42141930), from a separate group, pooled seven randomized trials in 608 postmenopausal women and adds the dose condition: benefit appeared at 5 g per day or more WITH resistance training, and trials at 3 g per day or less without it were null. The limit belongs beside the letter: the two largest and longest trials, at 237 and 200 participants over 2 years, found no effect on their own strength measures, and the pooled leg-press figure rests on 3 trials and 111 participants. Candow et al. (Appl Physiol Nutr Metab, 2015; PMID 25993883) is a single randomized trial in healthy older adults in the same strength population. Bone was tested directly and at length and failed: Chilibeck et al. (Med Sci Sports Exerc, 2023; PMID 37144634) randomized 237 women for 2 years and Sales et al. (J Gerontol A Biol Sci Med Sci, 2020; PMID 31257405) randomized 200 for 2 years, both null on bone mineral density, against one 12-month trial of 33 analyzed participants that was positive (PMID 25386713). Prokopidis et al. (Nutr Rev, 2023; PMID 35984306) found improved memory overall (SMD 0.29) with substantial heterogeneity and a null result in younger adults. Lyoo et al. (Am J Psychiatry, 2012; PMID 22864465) is a single unreplicated trial in women with major depressive disorder taking an SSRI. Grade A is scoped to strength; the other three uses carry their own letters.
PMID 12560406 PMID 11740307 PMID 34199420 PMID 29138605 PMID 42141930 PMID 37144634 PMID 31257405 PMID 25386713 PMID 35984306 PMID 22864465 PMID 41558805 PMID 29704637 PMID 25993883
Evidence by specific use
Muscle strength with resistance training in older adults
Grade A
One-repetition maximum is an instrument-recorded functional endpoint, which is the second route to Grade A on this site. Two mixed-sex meta-analyses in older adults, which share an author, and a 2026 meta-analysis from a separate group confined to postmenopausal women agree on direction, and the women-specific pooled figure is a leg-press 1RM gain of 7.5 kg (95 percent CI 2.2 to 12.8, I-squared 0 percent). The size is modest and one caveat belongs on the record rather than in a footnote: the two largest and longest individual trials, at 237 and 200 participants over 2 years, found no effect on their own 1RM measures, and the pooled leg-press result rests on 3 trials and 111 participants. DXA lean mass is a surrogate and caps at B by rule; it is not what this letter rests on. Read in October 2026 after an outside audit: Kazeminasab et al. (Nutrients, 2025; PMID 40944139), 69 trials across all ages, found the strength gain significant in younger adults and not significant in the older-adult subgroup; Davies et al. (JPEN J Parenter Enteral Nutr, 2024; PMID 38417175), 33 trials in older adults and adults with chronic disease, found upper-body and handgrip strength higher on creatine. A non-significant subgroup in an all-ages pooling is not a failed replication, and independent groups still agree on direction, so the letter holds at A with the caveat stated.
PMID 12560406 PMID 11740307 PMID 34199420 PMID 29138605 PMID 42141930 PMID 40944139 PMID 38417175
Bone mineral density in postmenopausal women
Grade D
The positive result people cite is the smallest and shortest trial: 12 months, 33 analyzed, attenuated femoral neck loss (Chilibeck et al., Medicine and Science in Sports and Exercise, 2015; PMID 25386713). It did not replicate. Two 2-year trials at 200 and 237 participants found nothing on density at any site, and the 2026 pooled review (Naddafha et al., Journal of the International Society of Sports Nutrition, 2026; PMID 42141930) reports bone density unchanged. One qualification belongs here and it is not a rescue: the 2023 trial did find maintained section modulus and buckling ratio at the femoral neck, which are geometry rather than density. This grade is about density, which is what the claim is about.
PMID 37144634 PMID 31257405 PMID 25386713 PMID 42141930
Memory and cognition in older adults
Grade C
Overall SMD 0.29 (95 percent CI 0.04 to 0.53) with I-squared of 66 percent. The effect sits almost entirely in adults aged 66 to 76 (SMD 0.88) and is absent in those aged 11 to 31 (SMD 0.03, -0.14 to 0.20). C rather than higher because the signal is not consistent across populations, which is mixed trial data regardless of how the endpoint is measured. One point matters for how this use is marketed: the review reports that sex did NOT influence the findings, so there is no women-specific cognitive effect here to claim.
PMID 35984306
Major depressive disorder, as an adjunct to SSRI treatment
Grade C
A single positive trial with no independent replication, which is the definition of C on this site. The 2026 systematic review of creatine across mental disorders found five randomized trials in total (238 participants), of which this one and a psychotherapy-augmentation trial were positive, while trials in female adolescents and in bipolar depression found no difference from placebo. Two of 17 participants in one trial experienced hypomania or mania. The endpoint is a depression rating scale, which is self-report and caps at B in any case.
PMID 22864465 PMID 41558805
Why Grade A and not higher
Grade A requires a hard clinical endpoint or objectively measured functional outcome, independently replicated across at least two trials. Creatine Monohydrate meets that standard based on the trials cited above. A downgrade to Grade B would follow if replication attempts produced null results, if the effect size proved substantially smaller in a broader population, or if the studied dose proved impractical to sustain.
What this grade covers and what it does not
Grade A by the replicated instrument-measured function route: one-repetition maximum is a reading off a bar, not a questionnaire, and independent groups agree: randomized trials at McMaster University (Brose et al., 2003; PMID 12560406) and the University of Saskatchewan (Chrusch et al., 2001; PMID 11740307) each found greater strength gains with creatine than with placebo during resistance training in adults over 65, and a 2026 pooled analysis from a separate group (PMID 42141930) agrees. The two earlier pooled analyses on this record share an author, so they are not counted as independent of each other. WHAT THIS GRADE DOES NOT COVER: the A belongs to ONE of the four things creatine is sold for, and the other three carry their own letters on this same record. It attaches to strength in older adults taking at least 5 g per day WHILE resistance training. Trials at 3 g per day or less, without training, were null, so the exercise is not optional context, it is part of what was tested. Bone mineral density was tested directly, in two trials running two years each, and failed; that use is a D. Memory and depression sit at C. Nothing here grades creatine for lifespan, and no trial has asked.
What would move this grade
A null result in a well-powered independent replication trial, or evidence that the effect does not hold at a practically sustainable dose, would prompt downgrade to Grade B.
How to track whether this is working
Grip strength, lean mass (DEXA if available), cognitive assessment, creatinine (distinguishing from true GFR decline)