Ginkgo has been used in China for centuries and sold across the West for decades. Few supplements have accumulated as much history, or as much research.
That research includes a trial of 3,069 older adults followed for six years. It’s worth knowing what it found.
The circulation hypothesis
The proposed mechanisms are specific, which is part of why the ingredient has held attention for so long.
Cochrane’s background summary lists them: increasing blood supply by dilating blood vessels, reducing blood viscosity, modifying neurotransmitter systems, and reducing the density of oxygen free radicals[1].
The circulation piece has laboratory support. Ginkgo extracts have been shown to produce cerebral vasorelaxation via a nitric oxide pathway[3], and improved haemorrheological properties — blood flow characteristics — have been reported in type 2 diabetes with retinopathy[3].
The reasoning follows naturally. Cognitive decline in older adults involves vascular contributions. An extract that improves cerebral blood flow might therefore slow it.
That’s a reasonable hypothesis. It was tested properly.
The largest prevention trial
The Ginkgo Evaluation of Memory study enrolled 3,069 community-dwelling participants aged 72 to 96 across six US academic medical centres between 2000 and 2008[2].
The design was rigorous. Randomised, double-blind, placebo-controlled. 120 mg of EGb 761 twice daily — 240 mg total — against identical-appearing placebo. Median follow-up 6.1 years, maximum 7.3[2].
Mean age at entry was 79.1 years, and 46% were women[2]. Because the dementia rate ran lower than expected early on, follow-up was extended in 2006, and closeout began only after 439 dementia outcomes had accrued[2].
The result: no effect on all-cause dementia, no effect on Alzheimer’s disease, and no effect on the rate of progression to dementia in participants with mild cognitive impairment[4].
A companion analysis published the following year examined cognitive decline rather than dementia diagnosis, using the 3MSE, ADAS-Cog, and domain scores for memory, attention, visual-spatial construction, language and executive function. Annual rates of decline did not differ between groups[5].
The pattern in the trial record
Cochrane’s 2009 review assessed 36 trials and reached a clear conclusion: the evidence that ginkgo has predictable and clinically significant benefit for people with dementia or cognitive impairment is inconsistent and unreliable[1].
Here’s the detail that reframed this for me.
Of the four most recent trials reporting results at that time, three found no difference from placebo, and one reported very large treatment effects in favour of ginkgo[1].
I would previously have read “mixed results” as meaning a modest real effect showing up unevenly. That’s not what this distribution looks like. Three nulls and one outlier is a different shape — one that points toward methodological variation rather than a moderate underlying effect.
Cochrane noted the surrounding conditions: many early trials used unsatisfactory methods, were small, and publication bias cannot be excluded[1]. A subgroup analysis restricted to Alzheimer’s disease — 925 patients across nine trials — showed no consistent pattern of benefit[1].
Most trials tested the same standardised preparation, EGb 761, at varying doses[1]. So the inconsistency isn’t explained by different products.
What the recent update shows
A more recent Cochrane assessment separates the picture by condition and by time.
In mild cognitive impairment, ginkgo showed little or no effect at six months on cognition, clinical status, or daily activities — moderate-certainty evidence[6].
In dementia, there were modest improvements at six months in clinical, cognitive and functional outcomes. But by twelve months the effects were absent or uncertain[6].
The overall conclusion: it remains unclear whether ginkgo improves overall clinical status in people with cognitive complaints, and for mild cognitive impairment there is no meaningful clinical benefit[6].
That six-versus-twelve-month pattern is worth holding onto. A benefit that appears early and then can’t be found is a different claim from a benefit that persists.
Guidelines don’t agree
This is one area where professional bodies have reached different conclusions from similar evidence.
The World Federation of Societies of Biological Psychiatry guidelines suggest a certain ginkgo extract may be used to treat dementia symptoms. The British Association for Psychopharmacology and the American Academy of Family Physicians concluded that its benefits are inconsistent and unreliable[7].
That disagreement is honest information rather than a problem to resolve here. When a body of evidence sits close to the threshold, different groups weighing it can land differently.
Safety, and one thing to check
On safety the record is comparatively reassuring. Cochrane found no significant difference between ginkgo and placebo in the proportion of participants experiencing adverse events[1].
Two things still warrant attention.
Bleeding risk. Ginkgo’s proposed mechanism includes reducing blood viscosity. If you take anticoagulant or antiplatelet medication, or have surgery scheduled, this belongs in a conversation with your clinician rather than being managed on your own.
Drug metabolism. An in vitro study in human hepatocytes found ginkgo inhibited the metabolism of paclitaxel, which could raise blood levels of that drug when used together[8]. Anyone on cancer therapy or other narrow-margin medication should raise ginkgo use specifically.
If you’re in Korea
Ginkgo occupies an unusual position here, and it’s worth being precise about the categories.
Ginkgo leaf extract is available as a prescription and over-the-counter medicine in Korea, not only as a supplement. That means some products containing it are regulated as pharmaceuticals, with the review process that entails, while others are sold in the supplement aisle. The category a product sits in changes what claims it may carry and what oversight it received.
If you’re buying it here, check which one you’re holding. And if you’re already taking a ginkgo-containing medicine prescribed by a doctor, adding a ginkgo supplement on top is a duplication worth raising rather than assuming is harmless.
For any supplement version, the recognised functional wording can be looked up on the official food safety portal[9] — and, as elsewhere in this series, that wording is likely narrower than the advertising around it.
Closing
Ginkgo is a case where long use and plausible mechanism met a well-designed trial, and the trial didn’t confirm the expectation.
What I found most instructive wasn’t the null result itself. It was the shape of the wider literature — three recent trials finding nothing and one finding a great deal. I’d been reading “mixed evidence” as a weak yes. Looked at properly, that distribution says something closer to: the studies aren’t measuring the same thing reliably.
None of this means ginkgo does nothing for anyone. It means the claim that it predictably prevents cognitive decline has been tested at scale and wasn’t supported, and that a supplement with real drug interactions deserves the same conversation any medicine would.
At a Glance
- Proposed mechanisms: dilating blood vessels, reducing blood viscosity, modifying neurotransmitters, reducing free radicals
- The GEM trial randomised 3,069 adults aged 72–96 to 240 mg EGb 761 daily, median follow-up 6.1 years
- Result: no effect on all-cause dementia, Alzheimer’s disease, or progression from mild cognitive impairment
- A companion analysis found no difference in annual rates of cognitive decline
- Cochrane’s review of 36 trials: benefit is inconsistent and unreliable; publication bias cannot be excluded
- Of the four most recent trials at that time, three found no difference and one found very large effects
- Recent update: in dementia, modest gains at 6 months but absent or uncertain by 12 months; in MCI, no meaningful benefit
- Adverse events did not differ from placebo, but bleeding risk and drug metabolism interactions warrant a clinician conversation
※ This article is for general information and does not replace medical advice, diagnosis or treatment. Ginkgo may affect bleeding and can interact with medication, including chemotherapy agents. If you take anticoagulants or antiplatelet drugs, have surgery scheduled, or are on any prescribed medication, consult a clinician before use. Concerns about memory or cognition warrant medical assessment rather than self-treatment.
References
- Birks J, Grimley Evans J. “Ginkgo biloba for cognitive impairment and dementia”, Cochrane Database of Systematic Reviews (CD003120), https://www.cochrane.org/evidence/CD003120_there-no-convincing-evidence-ginkgo-biloba-efficacious-dementia-and-cognitive-impairment
- “Ginkgo biloba for Prevention of Dementia: A Randomized Controlled Trial”, PMC (GEM design, follow-up and participant characteristics), https://pmc.ncbi.nlm.nih.gov/articles/PMC2823569/
- “Ginkgo biloba for preventing cognitive decline in older adults: a randomized trial”, PubMed (cited mechanism studies: cerebral vasorelaxation and haemorrheology), https://pubmed.ncbi.nlm.nih.gov/20040554/
- “The Ginkgo Evaluation of Memory (GEM) study”, ScienceDirect (summary of null findings on dementia and MCI progression), https://www.sciencedirect.com/science/article/abs/pii/S1551714406000085
- DeKosky ST, et al. “Ginkgo biloba for prevention of dementia: a randomized controlled trial”, JAMA 2008;300(19):2253-62, https://pubmed.ncbi.nlm.nih.gov/19017911/
- “Ginkgo biloba: Weak evidence supporting big claims” (summary of recent Cochrane findings by condition and timepoint), American Council on Science and Health, https://www.acsh.org/news/2026/03/30/ginkgo-biloba-weak-evidence-supporting-big-claims-50034
- “Ginkgo Biloba & Your Brain”, Cognitive Vitality, Alzheimer’s Drug Discovery Foundation (guideline positions), https://www.alzdiscovery.org/cognitive-vitality/ratings/ginkgo-biloba
- “Ginkgo Biloba for cognitive impairment and dementia” — discussion citing Etheridge et al. on paclitaxel metabolism, ResearchGate, https://www.researchgate.net/publication/23808798_Ginkgo_Biloba_for_cognitive_impairment_and_dementia
- Food Safety Korea (MFDS), recognised functional ingredient and approved wording lookup, https://www.foodsafetykorea.go.kr/