ginkgo bilobaCochrane reviewdementiamild cognitive impairmentsystematic reviewcognitive performancenootropicsEGb 761

Ginkgo Biloba for Dementia and MCI: What the 2026 Cochrane Review Found

A February 2026 Cochrane systematic review pooled 82 trials and 10,613 participants to test ginkgo biloba against placebo in dementia, mild cognitive impairment, multiple sclerosis, and subjective memory complaints. The results differ sharply by population.

28 October 202610 min read

Research context: This article discusses a peer-reviewed Cochrane systematic review. It is intended for informational and educational purposes only and does not constitute medical advice. Consult a qualified healthcare professional before making changes to diet, lifestyle, or supplementation.

The most comprehensive ginkgo review to date splits its verdict by population

Ginkgo biloba is one of the oldest and most widely sold cognitive supplements, usually as the standardized extract EGb 761, and the evidence behind it has accumulated unevenly for decades: some trials in people with dementia, some in healthy adults, mostly small, often with methods that made pooling difficult. On 5 February 2026, the Cochrane Database of Systematic Reviews published the most comprehensive attempt yet to settle what that evidence actually shows. A team led by Lisa Wieland pooled 82 randomized controlled trials covering 10,613 participants, searched through 18 November 2024, and analyzed ginkgo's effects separately across four distinct populations: people with subjective memory complaints, people with cognitive problems from multiple sclerosis, people with mild cognitive impairment (MCI), and people with diagnosed dementia (Wieland LS, Ludeman E, Chi Y, et al., "Ginkgo biloba for cognitive impairment and dementia," Cochrane Database of Systematic Reviews 2026;2:CD013661, DOI: 10.1002/14651858.CD013661.pub2, PMID: 41641880). The headline finding is not a single yes-or-no answer. It is four different answers, graded by Cochrane's own certainty-of-evidence framework, and they point in different directions depending on who is taking ginkgo and why.

What the review actually pooled

Of the 82 included trials, 72 trials covering 9,783 participants had data that could be extracted and combined. More than half the trials were conducted in China, and the authors report that only four studies met a low-risk-of-bias standard across every domain assessed, a caveat that applies to the entire body of evidence discussed below. Trials compared ginkgo against placebo, against usual treatment, against other medicines, or as an add-on to existing treatment. The review's key outcomes, assessed mainly at six months, were global clinical status, global cognition, activities of daily living, adverse events, and serious adverse events, each rated using the GRADE system that ranks evidence certainty from very low to high. That population-by-population structure, rather than a single pooled ginkgo-versus-placebo number, is what distinguishes this review from the head-to-head ranking of ten plant compounds in healthy older adults covered in our network meta-analysis review, which did not include people with diagnosed cognitive impairment.

Subjective memory complaints: the evidence is uncertain, and one trial flagged more harm

Three trials totaling 597 participants tested ginkgo in people reporting memory or thinking concerns without a formal diagnosis of impairment. The review rates the effect on global clinical status as uncertain, at very low certainty, meaning the available trials do not let Cochrane's reviewers draw a confident conclusion either way. One larger trial in this group, running three months, suggested that overall adverse events may be more common with ginkgo than placebo. That is a caution flag on a population that is, by definition, not cognitively impaired and would be taking ginkgo preventively.

Multiple sclerosis-related cognitive problems: probably no effect

Two small trials, 164 participants combined, tested ginkgo over three months in people with cognitive problems linked to multiple sclerosis. The review rates this evidence as moderate certainty and concludes ginkgo probably makes little or no difference to cognition in this group. No serious harms were attributed to ginkgo in these trials.

Mild cognitive impairment: probably little to no benefit

Twelve trials covering 1,913 participants with MCI showed, at moderate certainty, probably little to no effect of ginkgo on global clinical status, cognition, or activities of daily living at six months. Safety data told a reassuring story at this stage: seven trials with 991 participants found little or no difference in adverse events up to twelve months (low certainty), and three trials with 714 participants found little or no difference in serious adverse events (high certainty, the review's strongest rating anywhere in the analysis). This is the population where ginkgo's safety profile is best established and its cognitive benefit is least supported.

Dementia: small to moderate benefits, low certainty, and wide disagreement between trials

The largest and most clinically relevant group was dementia, with 13 trials covering 3,288 participants. Here the review reports that ginkgo may produce small to moderate benefits at six months on global clinical status, cognition, and activities of daily living, compared with placebo. That finding comes with two qualifiers the authors are explicit about: the certainty is rated low, and the individual trials disagreed with each other substantially, with heterogeneity statistics (I²) reaching as high as 96% on some outcomes, meaning the pooled estimate is an average across trials that often did not point the same way or by the same magnitude. On safety, nine trials with 2,746 participants found probably little or no difference in adverse events up to twelve months (moderate certainty), and six trials with 2,463 participants found serious adverse events may show little or no difference at six months (low certainty).

Why the same supplement produces four different verdicts

The pattern across these four groups shows that the population studied matters. The low-certainty benefit shows up specifically in people with diagnosed dementia. In MCI and multiple sclerosis-related cognitive problems, where impairment is present but less advanced or differently caused, the review found no detectable group-level benefit. The review did not test why. One possible explanation, which is our interpretation and not a finding of the review, is that cerebral blood flow and neuroinflammation, the two pathways ginkgo is proposed to act on, are more disrupted in dementia than in milder impairment. In subjective complaints, where no impairment has been diagnosed at all, the evidence does not support a benefit and raises a mild safety question instead. Readers tracking the broader landscape of tested and untested dementia-prevention strategies can see our Alzheimer's prevention research review for how ginkgo's modest, population-specific signal compares with other approaches aimed at modifiable risk and neuroprotection.

What the review does not establish

The authors are explicit about the limits of what 82 trials, mostly short and methodologically uneven, can show. Few dementia trials followed participants beyond six months, so the review has essentially nothing to say about whether any benefit persists, grows, or fades over a year or more of use. Only four trials across the entire review met a low-risk-of-bias standard in every domain, which means most of the included evidence carries some risk of overestimating or underestimating ginkgo's true effect. The geographic concentration of trials in China, more than half the total, raises questions about how directly the pooled results generalize to other populations and extract formulations, since the review does not report a separate analysis isolating one standardized product like EGb 761 from the full mix of ginkgo preparations tested. None of this means the dementia finding is wrong. It means the finding is a low-certainty signal that a future, better-designed trial could confirm, shrink, or reverse.

Key takeaways

A February 2026 Cochrane systematic review pooled 82 randomized trials and 10,613 participants to test ginkgo biloba separately in four populations. In people with dementia, ginkgo showed low-certainty evidence of small to moderate benefits at six months for global clinical status, cognition, and daily living skills, though trials disagreed with each other substantially. In people with mild cognitive impairment, the evidence pointed to probably little to no benefit, with a well-established safety profile. In multiple sclerosis-related cognitive problems, ginkgo probably made no difference. In people with subjective memory complaints but no diagnosed impairment, the evidence was too uncertain to draw a conclusion, and one trial suggested a higher rate of adverse events with ginkgo. Across all groups, only four of the 82 included trials met a low-risk-of-bias standard in every domain, and few dementia trials followed participants long enough to say anything about effects beyond six months.

FAQ

Does this review mean ginkgo biloba works for dementia?

It means that, pooling 13 trials and 3,288 people with dementia, ginkgo showed a small to moderate benefit at six months on global clinical status, cognition, and daily living skills compared with placebo. Cochrane rates this evidence as low certainty, and the individual trials varied widely in their results, with heterogeneity statistics reaching as high as 96% on some outcomes. It is a modest, uncertain signal in this specific population, not a confirmed treatment effect comparable to an approved dementia medication.

Does ginkgo help people with mild cognitive impairment?

The review found probably little to no effect on global status, cognition, or daily living skills in people with MCI, based on twelve trials and 1,913 participants rated at moderate certainty. The safety data in this group was the strongest and most consistent in the entire review, including a high-certainty finding of no meaningful difference in serious adverse events, so for MCI the evidence points to little general cognitive benefit in this population rather than to any safety problem.

Is ginkgo safe to take if you don't have diagnosed cognitive impairment?

The evidence here is the least reassuring of the four groups. In people with subjective memory complaints and no formal diagnosis, the review found the effect on clinical status too uncertain to interpret, and a larger three-month trial in this specific group suggested adverse events may be more common with ginkgo than placebo. That is a single trial's signal within a small, very-low-certainty body of evidence, not a settled safety finding, but it argues against treating ginkgo as a low-stakes preventive supplement for people who have not been assessed for actual cognitive impairment.

Why do the results differ so much between populations?

The review reports the four populations separately but does not test why they differ: a small to moderate, low-certainty signal appears in dementia, while no clear benefit appears in groups with milder or differently caused impairment. One possible explanation, which is our interpretation rather than a conclusion of the review, is that ginkgo's proposed mechanisms (cerebral blood flow and neuroinflammation) are more relevant in dementia. The authors do note that most trials were short, more than half were conducted in China, and only four trials across the whole review had a low risk of bias in every domain, all of which limit how far any single population's result can be generalized.

Primary source: Wieland LS, Ludeman E, Chi Y, Feinberg TM, Chen IH, Chen KH, Zhu Y, Wolverson E, Amri H. "Ginkgo biloba for cognitive impairment and dementia." Cochrane Database of Systematic Reviews 2026, Issue 2, Art. No. CD013661. DOI: 10.1002/14651858.CD013661.pub2. PMID: 41641880. Plain-language summary: Cochrane Evidence, CD013661. This article summarizes published, peer-reviewed research and is not a substitute for individualized medical advice.

Disclaimer

This article is for educational and research-context purposes only. It does not constitute medical advice and is not a substitute for consultation with a qualified healthcare practitioner. The Cochrane review discussed reflects published, peer-reviewed findings across trials of varying quality, not personalized clinical guidance. Ginkgo biloba can increase bleeding risk, particularly in combination with anticoagulant or antiplatelet medications. Always consult a doctor before beginning, modifying, or stopping any supplement, particularly if you have a diagnosed cognitive condition, take blood-thinning medication, are pregnant or breastfeeding, or take other prescription medications.