Does Ginkgo Biloba Work? The Independent Evidence Is Mostly Null

By Erin Rose · Updated July 17, 2026 · Methodology

Educational summary — not medical advice. Ginkgo is not a proven memory booster; clear it with your clinician if you take anticoagulants/antiplatelets or have a seizure disorder, and stop it 1–2 weeks before surgery.

Start with the null — because it's the strongest evidence

Ginkgo is unusual: it's been tested a lot, and the biggest, most independent tests came back negative. The Cochrane systematic review — the conservative benchmark — concluded the evidence that ginkgo helps cognition or dementia is "inconsistent and unreliable" (Birks 2009). More decisively, the two landmark prevention trials, designed to see whether ginkgo could stop dementia from developing, both failed at an adequate dose over years: GEM gave 240 mg/day to 3,069 older adults for ~6 years and found a hazard ratio of 1.12 — no protection (DeKosky 2008); GuidAge gave the same dose to 2,854 adults for 5 years and found no significant reduction in Alzheimer's (Vellas 2012). That's the headline: ginkgo does not prevent dementia, and there's no good evidence it boosts memory in healthy people.

The narrow signal that survives — with its conflict disclosed

It isn't uniformly null. Meta-analyses of the EGb 761 extract at 240 mg/day in people who already have mild-to-moderate dementia found benefits for cognition and daily function — a weighted mean difference around −2.86 on cognition and a benefit for activities of daily living (Tan 2015), and a pooled cognition effect of about SMD −0.52 described as dose-dependent (Gauthier 2014). Two honesty flags belong right here. First, this is treatment, not prevention, and it's EGb 761-specific — not a property of generic ginkgo. Second, the Gauthier analysis has manufacturer-affiliated authorship, which we disclose rather than bury. So the fair reading is a modest, narrow benefit for one extract in one clinical population — not a general cognitive enhancer.

What the trials show — by outcome

Read together: the independent evidence for the most-marketed ginkgo claims — prevention, general memory, tinnitus, circulation — is null. What's left is a modest, extract-specific treatment signal in dementia, and even that comes with a disclosed conflict. If you're going to use ginkgo, do it for that narrow reason, at 240 mg/day of EGb 761, with eyes open.

Safety: the bleeding tension, seizures, and surgery

Ginkgo is generally well tolerated, but three cautions are worth stating precisely — without overclaiming.

A note on the contaminant, kept bounded: ginkgo's ginkgolic acids are allergenic and cytotoxic, which is why pharmacopeias cap them at <5 ppm (Boateng 2022) — but they were cytotoxic, not mutagenic in the assays tested (Berg 2015), so this is a contaminant to limit, not a proven genotoxin. It's the reason to prefer an extract with a disclosed or inherent ginkgolic-acid limit — see EGb 761 vs generic.

Who it's most reasonable for

• People with diagnosed mild-to-moderate dementia weighing a modest adjunct — specifically EGb 761 at 240 mg/day — with realistic expectations and a clinician in the loop.
• Not healthy adults chasing a memory or focus boost — the evidence doesn't support that use.
• Not for preventing dementia at any dose — two large trials at 240 mg/day found no protection.

Skip or get medical advice first if you take anticoagulant/antiplatelet drugs, have a seizure disorder, have surgery planned (stop 1–2 weeks before), or are pregnant or breastfeeding.

Frequently asked questions

Related

• Ginkgo: what it is & who it's for
• EGb 761 vs generic "24%/6%" — why the narrow signal is one specific extract
• Ginkgo dosage guide — 120–240 mg/day + a dose calculator
• Best ginkgo — every product by quality tier and cost per serving

Sources