Creatine and Cognitive Function: The Meta-Analysis Flaw Behind the Popular Claim
A widely cited 2024 meta-analysis said creatine improves memory and processing speed. EFSA rejected the health claim, and a 2026 commentary explains the statistical error behind it.
Disclaimer: This article is for research and educational purposes only. It does not constitute medical advice. Consult a qualified healthcare professional before making any health-related decisions about supplementation.
The Claim Everyone Repeats
Search "creatine brain" and the same claim appears everywhere: a 2024 systematic review and meta-analysis found that creatine monohydrate improves memory, attention, and processing speed in healthy adults. The paper is real and peer-reviewed, and it has been cited approvingly across the supplement press ever since. What most of that coverage leaves out is that the health claim built on this meta-analysis was formally reviewed by Europe's food safety regulator and rejected, and that a peer-reviewed commentary published in April 2026, in the same journal as the original paper, lays out the specific statistical error that inflated its results.
This is exactly the kind of evidence-quality story worth walking through carefully, because the underlying question, whether standard-dose creatine supplementation meaningfully improves cognition in healthy adults, is a genuinely useful one for anyone using it as part of a nootropic stack. The honest answer is more qualified than the headline claim suggests, and the reasoning behind that qualification is itself instructive.
The Original Meta-Analysis
Xu C, Bi S, Zhang W, and Luo L published "The effects of creatine supplementation on cognitive function in adults: a systematic review and meta-analysis" in Frontiers in Nutrition in 2024, pooling 16 randomised controlled trials with 492 participants aged 20.8 to 76.4 years (Xu C et al., Front Nutr 2024;11:1424972). The pooled results reported statistically significant improvements in memory (standardised mean difference 0.31, 95% CI 0.18 to 0.44), attention time (SMD -0.31, 95% CI -0.58 to -0.03), and processing speed (SMD -0.51, 95% CI -1.01 to -0.01). Overall cognitive function and executive function did not reach significance. The authors concluded that creatine monohydrate "may confer beneficial effects on cognitive function in adults," while calling for larger trials.
On its face this looks like a reasonably solid evidence base: 16 trials, several hundred participants, multiple significant domains. The problem sits one level down, in how those domains were counted.
What EFSA Found When It Reviewed the Same Evidence
The European Food Safety Authority's Panel on Nutrition, Novel Foods and Food Allergens evaluated a proposed health claim for creatine and cognitive function and adopted its opinion on 30 October 2024. The Panel rejected the claim outright. Its stated reasoning went directly at the Xu et al. meta-analysis: "the meta-analysis conducted in the context of that systematic review pools the results of multiple related cognitive tests from the same studies, which are not independent from each other, to calculate a single effect estimate, leading to double-counting of participants in evidence synthesis and inflated sample sizes" (EFSA NDA Panel, EFSA Journal 2024;22(11):e9100). The Panel's overall verdict: "A cause-and-effect relationship has not been established between creatine supplementation and an improvement in cognitive function in one or more of its domains." It also noted that the positive signals seen in some trials used acute high doses (around 20 g/day for 5 to 7 days) that were not replicated at the standard 3 g/day maintenance dose typically used in consumer products, or with longer-term use.
That is a strongly worded rejection from a body whose entire function is scrutinising exactly this kind of evidence, and it went largely unremarked outside supplement-industry trade press at the time.
The 2026 Commentary: Naming the Statistical Error Precisely
In April 2026, Tom Citherlet published a formal commentary on the original Xu et al. paper in the same journal, Frontiers in Nutrition (Citherlet T, "Commentary: The effects of creatine supplementation on cognitive function in adults," Front Nutr 2026, published 10 April 2026, DOI 10.3389/fnut.2026.1716285). The commentary makes explicit what EFSA's opinion had stated more briefly: several of the trials feeding into the meta-analysis reported multiple correlated cognitive outcomes from the same participants, and the meta-analysis treated each of those outcomes as an independent effect size rather than as repeated measures on the same people.
Citherlet gives concrete examples. The Alves 2013 trials contributed at least seven separate memory subtests each, and the McMorris studies contributed four outcomes each, all drawn from the same participant pools. Counting each subtest as if it came from a different set of people means "the number of observations in the pooled analysis exceeds the number of unique randomised participants," which the commentary describes as a straightforward violation of the independence assumption that meta-analytic pooling depends on. The practical consequence is that the pooled confidence intervals are narrower and the effect sizes look more statistically secure than the underlying data actually support, precisely the kind of artefact that can turn a null or marginal finding into an apparently significant one. Citherlet notes that EFSA reached a parallel conclusion independently, for the same underlying reason.
None of this means creatine has no cognitive effects, and the commentary does not claim that. What it means is that the specific evidentiary basis most often cited for the claim, the pooled significant effect sizes in the 2024 meta-analysis, is weaker than it appears once the double-counting is accounted for. A regulator and an independent methodological critique arriving at the same conclusion through the same mechanism is a meaningfully different situation than a single dissenting opinion.
Where the Mechanism Is Genuinely Strong, Even If the Population-Level Claim Isn't
The biological rationale for creatine affecting brain function is not in dispute, and it is worth separating from the question of whether standard daily supplementation moves the needle on cognition in healthy people. Creatine and phosphocreatine form a rapid-buffering energy shuttle that regenerates ATP during periods of high metabolic demand, the same phosphocreatine system discussed in more general terms in our review of mitochondria and cognitive performance. Neurons are energetically expensive cells, and a depleted phosphocreatine pool under metabolic stress is a plausible mechanism by which creatine availability could matter for cognition specifically under conditions that draw down brain energy reserves.
The clearest human demonstration of that mechanism in action is not a chronic low-dose supplementation trial but an acute, high-dose one. Gordji-Nejad and colleagues gave 15 healthy adults a single high oral dose of creatine monohydrate (0.35 g/kg) or placebo during 21 hours of partial sleep deprivation, and used phosphorus and proton magnetic resonance spectroscopy to directly track brain phosphocreatine, ATP, inorganic phosphate, and pH alongside a cognitive test battery (Gordji-Nejad A et al., "Single dose creatine improves cognitive performance and induces changes in cerebral high energy phosphates during sleep deprivation," Sci Rep 2024;14:4937). Creatine improved word memory task performance by 10.3% relative to placebo and processing speed on language, logic, and numeric tasks by 16 to 29%, while the MRS data showed it prevented the sleep-deprivation-induced drop in brain pH and stabilised the phosphocreatine-to-inorganic-phosphate ratio, with total cerebral creatine rising measurably in the left medial parietal region. This is a small trial (n=15) and needs replication, but it pairs a behavioural result with a direct, objective measurement of the proposed mechanism in the same participants, which is a stronger design than most of the trials feeding the disputed meta-analysis. It is also directly relevant to anyone using creatine around acute sleep restriction, a state where phosphocreatine depletion appears to be the operative variable; our review of sleep architecture and cognition covers the broader cost of sleep debt on cognitive performance that this kind of intervention is trying to offset.
The pattern across the literature, in other words, looks less like "creatine helps cognition" and more like "creatine helps restore brain energetics specifically when they are acutely taxed," with the evidence for a standing daily benefit in well-rested, healthy adults resting on a meta-analysis that both a regulator and an independent statistical critique have identified as methodologically compromised.
Practical Implications
Treat the "creatine improves memory" headline claim with real skepticism. The 3 g/day maintenance-dose, chronic-supplementation case for cognitive enhancement in healthy adults rests substantially on a meta-analysis that EFSA rejected as a basis for a health claim and that a 2026 peer-reviewed commentary has shown double-counts correlated outcomes from the same participants as if they were independent.
The acute, high-dose, sleep-deprivation evidence is more credible, not less relevant. Effects observed at higher single doses (around 20 g/day short-term, or the 0.35 g/kg single dose used by Gordji-Nejad's group) under conditions of genuine metabolic stress have a more direct mechanistic demonstration behind them than the chronic low-dose literature.
Creatine's other benefits are unrelated to this specific dispute. Nothing here challenges creatine's substantially better-supported evidence base for muscular strength, power output, and exercise performance; the concern here is narrowly about the cognitive-enhancement claim in healthy, rested adults.
Evidence quality matters more than an impressive-looking citation count. Sixteen trials and 492 participants sounds substantial until the unit-of-analysis problem is accounted for, a reminder that meta-analyses are only as reliable as the independence of the data points feeding them, and that a regulator's rejection of a claim is itself a data point worth reading rather than a footnote.
Summary
A 2024 meta-analysis reported that creatine monohydrate improves memory, attention, and processing speed in adults, a finding that has circulated widely as settled evidence. EFSA reviewed the same evidence in October 2024 and rejected the associated health claim, explicitly citing double-counting of correlated cognitive outcomes as inflating the meta-analysis's apparent precision. A formal commentary published in April 2026 in the same journal names the error in detail, with specific examples of trials contributing multiple non-independent outcomes to the pooled estimate. The mechanistic case for creatine affecting brain energetics remains genuinely strong, and a well-designed acute high-dose trial under sleep deprivation, with direct spectroscopy evidence of the proposed mechanism, supports a real effect under conditions of metabolic stress. What the current evidence does not support is the specific, widely repeated claim that standard daily creatine supplementation reliably improves cognition in healthy, rested adults. That claim's primary evidentiary basis has now been formally rejected by a regulator and methodologically dismantled by an independent commentary, and the honest position is to treat it as unresolved rather than established.
Primary sources: Xu C, Bi S, Zhang W, Luo L. "The effects of creatine supplementation on cognitive function in adults: a systematic review and meta-analysis." Frontiers in Nutrition 2024;11:1424972; EFSA Panel on Nutrition, Novel Foods and Food Allergens (NDA). "Creatine and improvement in cognitive function: Evaluation of a health claim pursuant to article 13(5) of regulation (EC) No 1924/2006." EFSA Journal 2024;22(11):e9100, adopted 30 October 2024; Citherlet T. "Commentary: The effects of creatine supplementation on cognitive function in adults: a systematic review and meta-analysis." Frontiers in Nutrition 2026, published 10 April 2026; Gordji-Nejad A, Matusch A, Kleedorfer S, Patel HJ, Drzezga A, Elmenhorst D, Binkofski F, Bauer A. "Single dose creatine improves cognitive performance and induces changes in cerebral high energy phosphates during sleep deprivation." Scientific Reports 2024;14:4937. This article summarises published research and a regulatory opinion; it is not a substitute for individualised medical advice.