What the ginkgo trials actually found
Ginkgo has more human trial data behind it than almost anything else in this category, which makes it unusually easy to check.
The Ginkgo Evaluation of Memory study randomised just over 3,000 older adults to 120 mg of standardised ginkgo extract twice daily or placebo. Its published conclusion was that ginkgo did not result in less cognitive decline.
The 2009 Cochrane review described the evidence for a predictable, clinically significant benefit as inconsistent and unreliable. Ginkgo remains a reasonable ingredient to include; it is not a reason to expect a result.
Why ginkgo is worth checking
Before this post went live, Dr. James Bateman III, MD, MPH, a physician, checked it against the CogniMary label and the sources it cites. Most ingredients in this category have a thin trial record: a few small studies, often industry-funded, usually short. Ginkgo is the exception. It has decades of trials, a standardised extract used across most of them, and one very large publicly funded study.
That makes it the best test case in the category. If a well-studied ingredient at a proper dose over a long period does not produce an effect, it tells you something about how much to expect from the less studied ones.
The GEM study
The Ginkgo Evaluation of Memory study is the one that matters most. It enrolled just over 3,000 adults aged 72 to 96, randomised them to 120 mg of standardised extract twice daily or to placebo, and followed them for years.
The dose — 240 mg a day — is at the top of the range used in the literature. The duration was long. The sample was large. It was, by the standards of this field, about as good a test as anyone has run.
The published conclusion
“The use of G. biloba, 120 mg twice daily, did not result in less cognitive decline in older adults with normal cognition or with mild cognitive impairment.”
That is a null result at a high dose in a large sample over a long period. It is not a small study that failed to find an effect for lack of power.
What Cochrane concluded
The 2009 Cochrane review of ginkgo for cognitive impairment and dementia pooled the trial literature and described the evidence that ginkgo has a predictable and clinically significant benefit as inconsistent and unreliable.
“Inconsistent” is doing real work in that sentence. It does not mean every trial was negative. It means the trials disagree with each other in ways that are hard to explain, which is the signature of an effect that is either very small or not there.
Cochrane reviews are updated as new evidence appears, and later work on ginkgo exists. We are citing the 2009 review because it is the version we were able to open and read in full, and we would rather cite what we read than summarise what we did not.
The dose question
There is a common defence of ginkgo that goes: the trials used the wrong dose, or the wrong extract, or ran too short. On ginkgo specifically, that defence is weaker than usual.
| The objection | What GEM had |
|---|---|
| The dose was too low | 240 mg a day, the top of the studied range |
| The extract was not standardised | EGb 761, the standardised extract used across the literature |
| Too short | Followed for years |
| Too small | Just over 3,000 participants |
| Wrong population | Older adults, which is exactly who these products target |
None of the usual escape hatches is available. That is what makes GEM such an important result and such an inconvenient one.
The fair reading
It would be overreaching to say ginkgo does nothing. Here is what we think the evidence actually supports.
- For preventing cognitive decline in older adults, the best trial found no effect. That is a specific claim with a specific answer.
- For people already diagnosed with dementia, the picture is more mixed, and some reviews report small benefits with low certainty. That is a different population from a healthy buyer.
- Ginkgo is well tolerated in most trials, with the bleeding-risk caution being the main practical concern.
- It is a reasonable ingredient to include in a formula. It is not a reason to expect a result.
The last two lines are the honest position, and it is not the position any ginkgo product's marketing takes.
What this means on a label
If ginkgo is the best-evidenced ingredient in a formula and its best trial was null, the formula needs something else carrying it.
On CogniMary's label, ginkgo sits second in the blend, so its ceiling is 317.5 mg — enough for a researched dose. It is one of the few ingredients there that could be at a studied amount. What it cannot do is deliver a result the studies did not find.
That is the pattern worth carrying into any purchase in this category: check the amount first, then check what the amount was shown to do. The second question is the one most buyers never reach.
- Snitz, O'Meara, Carlson et al. — 2009Ginkgo biloba for Preventing Cognitive Decline in Older Adults: A Randomized Trial — JAMA · PMID 20040554 · source
- Birks & Grimley Evans — 2009Ginkgo biloba for cognitive impairment and dementia — Cochrane Database of Systematic Reviews · CD003120.pub3 · source
- U.S. Government Publishing Office21 CFR 101.36 — Nutrition labeling of dietary supplements — eCFR.gov · source
Why one null trial matters this much
A single negative trial does not usually settle anything. It is worth explaining why GEM carries more weight than most.
It was designed to find an effect
GEM was a prevention trial: enrol people without dementia, dose them for years, and see whether fewer of them decline. That design gives an effect the best chance to appear.
Not a failure to look
It was publicly funded
Industry funding is associated with more favourable results across the supplement literature. A publicly funded trial has no commercial stake in either answer.
No stake in the outcome
It used the standard extract
EGb 761 is the material most of the ginkgo literature is built on. A null result on a non-standard preparation would prove less.
The same material as the positive studies
It was large enough to detect a small effect
Just over 3,000 participants. A null result in a small trial means “we could not tell”. A null result in a large one means “if there is an effect, it is smaller than this”.
The difference that matters
Put those four together and GEM is about as informative as a negative result in this field gets. That is why it appears on this site more than once.
What ginkgo is still used for
It would be a distortion to leave the impression that ginkgo is worthless. Three things are worth saying on the other side.
- In people already diagnosed with dementia, some reviews report small benefits with low certainty. That is a different population from a healthy buyer and a weaker grade of evidence, but it is not nothing.
- It is generally well tolerated across a very large trial literature, with bleeding risk being the practical caution rather than a frequent adverse event.
- The mechanism is real. Ginkgo does things in the body. The question GEM answered is whether those things add up to less cognitive decline, not whether the plant is inert.
- It is one of the few ingredients in this category with enough data to be wrong about. Most have too little evidence to produce a clear negative.
That last point deserves emphasis. Ginkgo looks bad here partly because it has been studied properly. An ingredient with no large trial behind it is not better supported; it is less examined.
How this changes what to look for
The useful lesson from ginkgo is not about ginkgo. It is about how to read the phrase “clinically studied” on any label.
| What the label says | What it can mean |
|---|---|
| “Clinically studied ingredient” | A trial exists. It may have been negative |
| “Backed by research” | Research exists. It may be in mice |
| “Shown to support memory” | One trial, possibly small, possibly in a diagnosed population |
| “Used for centuries” | No trial is being claimed at all |
None of those four phrases tells you the direction of the evidence, and all four are common. The only way through is to look up the ingredient and the dose, which takes one search and is what this whole site is arguing for.
Ginkgo is the best possible demonstration, because it is simultaneously the most studied ingredient in the category and one of the ones with the clearest negative result.
Where this leaves the category
Ginkgo is the best-studied ingredient in cognitive supplements, and its largest trial was null. It is worth sitting with what that implies for everything with less evidence behind it.
It does not mean nothing in the category works. It means the base rate for a large, well-run trial finding a clear cognitive benefit in healthy adults is low, and that a product making confident claims is not usually making them from a stronger position than ginkgo's.
The practical consequence is to lower what you expect and to make the purchase cheap to be wrong about: one bottle, a noted start date, and a return window you have written down.