THE CELLULAR BATTERY // CREATINE MONOHYDRATE
REVISION LOG — 2 ENTRIESMOST RECENT: 2026-10-08
- 2026-10-08CORRECTION — re-audit. This dossier was audited again from the primary sources. The graded claim is now stated: creatine with resistance training increases strength and lean mass. Cognition had been graded "real but conditional" (aging, sleep loss, plant-based diets); that grade is withdrawn. The EU food-safety authority found no established effect in healthy adults (EFSA 2024), which this dossier had not cited. The 2024 meta-analysis we relied on reported larger benefit at ages 18–60, not in older adults as we wrote; above 60, its attention subgroup was not significant, and its pooling was judged to double-count participants (EFSA 2024; Citherlet 2026). The memory effect in adults aged 66–76 came from the authors' re-analysis after a published critique, not from a corrected paper. We wrote that vegetarians benefit more cognitively in six places in the text and in our first changelog entry, and that they do not in three; the largest trial found they do not (Sandkühler 2023). The mood paragraph cited a 2026 review without a reference and gave a trial dose we could not confirm; that trial's reference was removed and the paragraph now cites two reviews (Jeryous Fares 2026; Eckert 2025). The effect-by-domain figure and the evidence-vs-marketing chart carried values not found in the sources and were removed. The kidney section now cites two meta-analyses, states that GFR estimated from creatinine reads lower, and no longer calls the kidney concern "debunked". The cognition row for aging and sleep loss was split into two rows, both moved from Tier A to Tier B; for aging, a randomised 24-week trial in older women found no effect (Alves 2013); the kidney row left the verdict and became a separate safety note; and this page's Tier B and C definitions were rewritten (B: "Positive results in small or preliminary human trials; not established."; C: "Tested in an adequate trial in this group; no significant effect on the primary outcome.") Side effects and the smaller effect in women were added. All references in the source list now link to PubMed with their literal titles. Calibrated certainty was revised from 80 (E80 · C70 · I92) to 77 (E78 · C70 · I84). This entry also records three edits made without a log entry: on 2026-08-06 the curation policy changed from a possible affiliate commission to no commission; on 2026-08-20 it changed from "our only revenue is subscription" to revenue "only from the operators we serve"; on 2026-10-07 the protocol reference was relabelled "(under re-audit)"; it was removed when the protocol reopened on 2026-10-08.
- 2026-07-15Initial gold-standard dossier issued: phosphocreatine–ATP mechanism, effect-by-domain forest-plot figure, evidence-vs-marketing chart, O.B.M. deep-analysis, and 7 peer-reviewed references. Graded CLINICAL (Tier A) for muscle/strength & safety; cognition stated as real-but-conditional (aging, sleep-loss, plant-based), not a universal nootropic. Kidney-harm myth documented as debunked.
Every entry above was written when the change was made, not reconstructed afterwards. Where we were wrong, the wrong version is named.
About two-thirds of muscle creatine is stored as phosphocreatine, which the cell uses to regenerate ATP during short, hard effort (Kreider 2017). This dossier grades one claim: with resistance training, creatine adds strength and lean mass beyond training alone. In older adults, that holds for strength and lean mass (Devries 2014; Liu 2025); in adults under 50, for strength, significantly in men and not in women (Wang 2024). Cognition is a separate question: the EU food-safety authority concluded that an effect in healthy adults has not been established (EFSA 2024), and the largest creatine–cognition trial up to 2023 found at most a small, borderline effect (Sandkühler 2023). In the trials that measured kidney filtration directly with Cr-EDTA, there was no significant change; GFR estimated from creatinine read 10.75 mL/min lower (de Souza Almeida 2026).
This dossier grades one claim: creatine with resistance training increases strength and lean mass beyond training alone. Cognition and mood are graded separately; kidney safety is reported as a safety note, outside the grading scale.
About two-thirds of muscle creatine is stored as phosphocreatine, which muscle uses to regenerate ATP in the first seconds of hard effort (Kreider 2017). Supplementing raises muscle creatine about 20%, quickly with a 20 g daily load or over four weeks at 3 g a day (Hultman 1996).
For cognition, the EU food-safety authority concluded that an effect in healthy adults has not been established (EFSA 2024). We had graded it "real but conditional"; that grade is withdrawn.
The Muscle CaseTIER A
With resistance training, creatine adds strength beyond training alone, and lean mass in older adults. In older adults: fat-free mass, chest press and leg press, and the 30-second chair stand (Devries 2014, 357 participants); lean mass and lower-limb strength with SMDs near 0.3, upper-limb strength not significant, most trials at about 0.1 g/kg a day (Liu 2025, 8 RCTs, 482 participants). In adults under 50: +4.4 kg upper-body and +11.4 kg lower-body strength, from trials mostly in men (Wang 2024).
In women the gains are smaller or slower: no significant strength gain under 50, from 49 women (Wang 2024); in older women, upper-body strength improved overall and muscle mass did not, with lower-body strength gains only in trials of 24 weeks or more (Santos 2021); after menopause, a small lean-mass gain appeared with at least 5 g a day plus training, and none at 3 g or less without training (Naddafha 2026).
The Brain CaseTIER B / C
For healthy adults, the EU food-safety authority concluded that a cause-and-effect relationship with cognition has not been established (EFSA 2024). The largest creatine–cognition trial up to 2023, 123 analysed adults on 5 g a day for six weeks, found at most a small, borderline effect, and vegetarians did not benefit more (Sandkühler 2023).
Sleep loss: under 21 hours without sleep, a single 0.35 g/kg dose partly offset the decline in 15 young adults (Gordji-Nejad 2024); 0.2 g/kg in 29 adults had a smaller effect (Gordji-Nejad 2026). Aging: a re-analysis found a memory benefit (SMD 0.80) in a small subgroup of trials in adults aged 66–76, while the overall effect was no longer significant (Prokopidis 2023); against it, a 24-week randomised trial in older women, which the EU authority judged more rigorous than a non-randomised trial in young adults that had reported a working-memory benefit, found no change in any cognitive measure (Alves 2013; EFSA 2024). A 2024 meta-analysis reported a memory benefit (Xu 2024); its only age analysis, on attention, found no significant effect above 60 in three trials, and its pooling was judged to double-count participants (EFSA 2024; Citherlet 2026).
In depression the picture is unsettled. A 2026 systematic review of five small trials rested its promise on one trial with an SSRI started together with creatine and one with psychotherapy; the trials in adolescent girls and in bipolar depression found no difference from placebo (Jeryous Fares 2026). A 2025 meta-analysis of 11 trials, in people with and without depression, found an average effect below what patients notice, at very low certainty (Eckert 2025). In an open study, both patients with bipolar depression developed hypomania or mania (Roitman 2007).
Is creatine a muscle supplement with evidence, or a brain booster with marketing?
Meta-analyses agree in direction: with training, more strength, and more lean mass in older adults; smaller or slower in women (Devries 2014; Wang 2024; Santos 2021).
Liu et al. — creatine with resistance training in the aged: systematic review & meta-analysis (Eur Rev Aging Phys Act, 2025) ↗No established effect in healthy adults (EFSA 2024); a narrow signal under acute sleep loss (Gordji-Nejad 2024); a randomised 24-week trial in older women found no effect (Alves 2013).
EFSA NDA Panel — creatine and improvement in cognitive function: health-claim evaluation (EFSA Journal, 2024) ↗In the trials that measured filtration with Cr-EDTA — a few small trials, including people with type 2 diabetes — there was no significant change; GFR estimated from creatinine read 10.75 mL/min lower across the trials that used it (de Souza Almeida 2026). A second meta-analysis found no significant change in GFR, without separating the method (Naeini 2025). Data in kidney disease remain thin (Longobardi 2023).
de Souza Almeida et al. — creatine supplementation and kidney health: systematic review & meta-analysis (Int Urol Nephrol, 2026) ↗With training, the evidence supports creatine for strength and, in older adults, for lean mass. For cognition in healthy adults, an effect has not been established. On measured kidney filtration, the trials found no significant change, while estimates based on creatinine read lower.
In the trials that measured kidney filtration with Cr-EDTA — a few small trials, including people with type 2 diabetes — there was no significant change; GFR estimated from creatinine read 10.75 mL/min lower across the trials that used it. The review rated its overall GFR evidence moderate certainty (GRADE) (de Souza Almeida 2026).
A second meta-analysis found no significant change in GFR, without separating the method (Naeini 2025). Data in people with kidney disease remain thin (Longobardi 2023).
Tell whoever orders your bloodwork that you take creatine, so a creatinine-based result is read correctly.
3–5 g of creatine monohydrate a day (Hultman 1996; Kreider 2017). Pair it with resistance training, as the strength trials did, most at about 0.1 g/kg a day (Liu 2025). Tell whoever orders your bloodwork: GFR estimated from creatinine reads lower (de Souza Almeida 2026). There is no established reason to take it for cognition if you are healthy (EFSA 2024).
OCCABUZZ sells no supplement and earns no commission on creatine; our revenue comes only from the operators we serve, never from what we grade. Buy plain monohydrate. The evidence drives the grade.
- ›Take 3–5 g of creatine monohydrate daily (Hultman 1996; Kreider 2017). Loading — 20 g a day as four 5 g servings for 5–7 days (Kreider 2017) — reaches the same muscle level sooner (Hultman 1996); a single 10 g serving raised diarrhea over placebo (Ostojic 2008).
- ›Buy plain monohydrate. Monohydrate is the form evaluated in the vast majority of efficacy studies (Kreider 2017) and in all 16 cognition trials of Xu 2024.
- ›Pair it with resistance training: the strength and lean-mass trials did, most at about 0.1 g/kg a day (Liu 2025). There is no established reason to take it for cognition if you are healthy (EFSA 2024).
For an adult who lifts, creatine adds strength (Devries 2014; Wang 2024) and, in older adults, lean mass, with an SMD near 0.3 (Liu 2025). The effect is smaller or slower in women (Wang 2024; Santos 2021). The daily habit has a known cost: about 1 in 6 users in the largest creatine–cognition trial up to 2023 reported a side effect, against 1 in 25 on placebo (Sandkühler 2023).
Creatine is not an established nootropic: the EU food-safety authority concluded that a cause-and-effect relationship with cognition has not been established in healthy adults, including in the sleep-loss and older-women studies it reviewed (EFSA 2024). The strength and lean-mass results come from trials that paired it with resistance training (Devries 2014; Liu 2025). Its role in depression is unsettled: a 2025 meta-analysis of 11 trials, in people with and without depression, found an average effect below what patients notice, at very low certainty (Eckert 2025). In an open study, both patients with bipolar depression developed hypomania or mania (Roitman 2007). Data in people with kidney disease remain thin (Longobardi 2023). The grade is CLINICAL for the muscle claim only.
Peer-reviewed trials — sample size, effect size, stage.
Applies to healthy apex, not only to clinical deficit.
Survives a demanding calendar. Zero executive friction.
Cleared all three layers. Protocol-grade.
The score is a geometric mean — a single failed layer collapses it. Excellence in two cannot rescue a gap in the third. That is why hype scores low and proven, feasible, broadly-applicable work scores high. The restraint is the product.