Food and diet

Reports that Marmite prevents dementia are laying it on a bit thick

"A daily slice of Marmite on toast may help prevent you getting dementia," the Daily Mail reports, with little justification.

A small study did find that Marmite had an effect on electrical activity in the brain, but there is no evidence this would prevent dementia.

The study involved 28 people in their early 20s. Researchers looked to see whether eating Marmite affected the brain's response to watching flickering images on a screen, measured by electroencephalogram (EEG) scans. This test is used as a measure of "brain cell excitability" in the visual cortex area.

Marmite contains vitamin B12 and glutamate which are thought necessary for the brain to produce GABA, which is thought to reduce brain cell excitability. Too little GABA may be a factor in epilepsy.

The study compared the effect of eating a teaspoon of Marmite each day for one month with eating peanut butter. Healthy volunteers were tested before and after the trial of Marmite or peanut butter. After a month, scans on the Marmite-eaters' brains showed lower levels of excitability.

The researchers suggest that boosting GABA levels through diet might contribute to treating epilepsy. However, there's no clinical evidence to support this indication, never mind the media speculation about dementia.

For those who fall into the "hate camp" when it comes to Marmite, other sources of vitamin B12 include meat and cheese.

Where did the story come from?

The study was carried out by researchers from the University of York and was funded by the Wellcome Trust and the Leverhulme Trust.

The Leverhulme Trust was set up by the founder of Lever Brothers (William Hesketh Lever) now Unilever, which manufactures Marmite. However, the trust says it does not seek to influence the topic or study design of research when it provides grants.

The study was published in the peer-reviewed Journal of Psychopharmacology.

Predictably, the UK media loved this story. The Daily Telegraph and Daily Mirror referred to Marmite "giving a boost" to the brain. Sky News said it "keeps the brain healthy" and The Sun said it "may prevent dementia."

The Daily Mirror's report is the most balanced and was the only one to point out the link between funding and Unilever. It concludes that: "an analysis by the Daily Mirror of the nine-page study [ie reading it] found zero references to dementia or Alzheimer’s disease".

Several media outlets reported that the study was done in men only; however the study makes clear that more than half of the participants were women. 

What kind of research was this?

This was a randomised controlled trial, which is a good way to see the effect of an intervention. Researchers wanted to see if the yeast extract reduced the brain's response to visual stimuli.

What did the research involve?

Researchers measured brain response to visual stimuli (flickering images on a screen) using EEGs, in 28 volunteers. They were randomly allocated to take a teaspoon of either Marmite or peanut butter a day, in addition to their usual diet. After a month they were tested again, and the results compared between the two groups.

The volunteers (10 men and 18 women) were all in their 20s. None had epilepsy (in case flickering images triggered a seizure), smoked, had nut allergies or used controlled substances.

The experiment used flickering images with a "control" task, in which volunteers had to estimate the difference in contrast between two wave forms. This allowed researchers to check the groups were concentrating on the screen equally.

Researchers used several variations of the task, including a "mask" variation which should reduce the effect on an area of the brain called the visual cortex.

Volunteers were asked to take a teaspoon of their allocated spread each day and record that they had done so. Statistical models were used to look for differences between the EEGs for those who'd eaten Marmite and those who'd eaten peanut butter.

The researchers also tested both spreads for levels of glutamate and B vitamins. A sub-group of Marmite-eaters was tested again two months later.

What were the basic results?

Researchers say that on average the people who ate Marmite had reduced levels of "evoked response" – activity in response to images – compared to their baseline results. The average response didn't change for those eating peanut butter.

Only evoked responses were changed – response to background levels of activity when viewing a blank screen were not affected. The volunteers' performance on the attention test was no different between the two groups, suggesting that Marmite-eaters and peanut butter eaters had concentrated on the screen to the same degree.

In the group of Marmite-eaters retested after two months, response levels were still lower than at baseline, but not as low as immediately after the month-long trial.

In the chemical analysis of the spreads, Marmite had three times as much vitamin B6, almost twice as much glutamate and 116 times as much vitamin B12 than the peanut butter.

How did the researchers interpret the results?

The researchers said their results were "consistent with an increase in the availability of GABA in visual areas of the brain."

They said that previous research has shown that people with epilepsy showed increased visual responses when tested using the same visual stimuli as used in this experiment. "This raises the possibility that dietary interventions geared towards increasing GABA concentration might reduce excitability to normal levels, and potentially alleviate some symptoms of the disorder," they say.

They suggest it could reduce the number of seizures and be particularly useful for people who can't take anti-epilepsy drugs, or whose medicines haven't controlled their seizures.

They conclude that additional studies would be required to determine which substance in Marmite might be responsible for the results.


This is an early investigative research study, and while some of the findings are interesting, it's a long way from showing that yeast extract spreads can help with conditions like epilepsy or other neurological disorders.

The study's strength is that it was carried out as a randomised controlled trial. However, its small size means we need to see the results replicated in larger studies to be sure they are not down to chance. We also need to see longer-term studies into the actual clinical effects of the changes measured. At this point, we don't know what effect – if any – the changes in brain response have on the people involved.

The study has no implications for people with dementia, or at risk of dementia. The lead researcher told NHS Choices: "We're a bit puzzled as to where the idea [that dementia is involved] has come from. Our study didn't test any patients and we don't have any reason to expect that Marmite would have any effect on dementia at this time."

It's also important to be aware that the suggested effects on epilepsy have not been tested on people with epilepsy. No-one with epilepsy should be tempted to stop taking their medicines in favour of Marmite.  

NHS Attribution