Claims that coffee prevents dementia are lukewarm at...

"Coffee really can help to prevent dementia: Just two cups a day 'cuts the risk of developing it by 36 per cent','' the Mail Online reports. But if you look closely at the research behind this report, the results are of borderline significance, meaning it is likely they were influenced by chance.

Researchers in the US found that postmenopausal women who consume an average of 261mg of caffeine a day (the equivalent of two to three cups of coffee) are at lower risk of cognitive impairment or probable dementia, than those who consume lower levels of caffeine.

The study followed 6,467 women for up to 10 years. Those consuming more caffeine were found to be at lower risk of having problems with cognitive functioning.

However, caffeine consumption was based on self-report and decaffeinated drinks were not specified, it was presumed all coffee, tea and cola drinks were caffeinated.

In addition, the results were modest to say the least. While researchers estimated the preventative effect was 26% (not 36% as widely reported in the media), according to the maths used by the researchers, the actual figure could be as low as 1%.

As the exact cause of dementia and in particular Alzheimer's disease is not known, there is no clear way to prevent the condition. There are things you can do that may reduce your risk, such as stopping smoking and reducing alcohol intake, eating a healthy, balanced diet, and staying physically fit and mentally active.

Where did the story come from?

The study was carried out by researchers at the University of Wisconsin-Milwaukee, Wake Forest School of Medicine, the Health Partners Institute for Education and Research in Minnesota, and Harvard Medical School, all in the US.

The study was funded by the National Heart, Lung and Blood Institute, at the National Institutes for Health, US. The authors do not report any sources of conflict.

The study was published in the peer-reviewed Journals of Gerontology, Medical Sciences, an open-access journal and is free to read online.

The quality of the UK media's reporting of the study was poor. Firstly, most of the headlines, such as the Mail's "Coffee really can help to prevent dementia," overstated the implications of the study's results.

Secondly, nobody highlighted that the main result in terms of dementia prevention only barely scraped the level required for statistical significance.

Finally, and most importantly, all of the UK media's sources reported a factual inaccuracy. The actual reduction in terms of hazard ratio was 26% not 36%.

The inaccuracy appears to originate with a press release from the EurekAlert! science news service. This suggests no UK media source bothered to read the actual study (which to be fair to EurekAlert! was linked to in its article) and instead just used the press release.

What kind of research was this?

This was a prospective cohort study which aimed to assess the relationship between caffeine intake and overall incidence of probable dementia or cognitive impairment in postmenopausal women. Previous research in animals has suggested a protective effect of caffeine and other components in coffee on brain function.

The researchers used data from a long running randomised controlled trial in the US called the Women's Health Initiative. Women aged 65 to 80 had been randomised to take oestrogen tablets or placebo between 1995 and 1999. They were followed up annually for up to 10 years.

As this was a cohort study using data from the trial, it can only show a link between one factor – in this case caffeine intake – and another – dementia and cognitive impairment. It cannot prove caffeine intake lowered the risk of dementia or cognitive impairment.

What did the research involve?

Researchers looked at data from 6,467 women who had provided self-reported caffeine data within six months of starting the study and had at least one follow-up cognitive assessment.

Global cognitive function was assessed annually by trained, certified technicians and interviewers using the 100-point Modified Mini Mental State (3MS) exam until 2007. Annual assessment after 2007 was done by the 40 point Telephone Interview for Cognitive Status-modified (TICSm).

Both the 3MS and the TICSm are well validated methods of assessing and measuring cognitive abilities.

Caffeine intake was based on self-report at the start of the study using a food frequency questionnaire (FFQ). Caffeine intake was estimated from questions on coffee, tea, and cola beverages, including frequency and serving size. It was presumed that intake of these drinks were of the caffeinated form as there was no specific question asking about decaffeinated forms.

The data was analysed according to caffeine intake and the time until incidence of probable dementia or cognitive impairment. The results were adjusted to take into account risk factors including:

age
education
body mass index
hormone therapy
race
sleep quality
depression
hypertension
prior cardiovascular disease
diabetes
smoking
alcohol consumption

What were the basic results?

In a sample of generally healthy postmenopausal women, during the 10 years of follow-up, 209 women received a classification of probable dementia and 179 of mild cognitive impairment.

Women who drank more than 172mg of caffeine per day (equivalent to just under two cups of coffee) had a 26% lower risk of probable dementia than those who drank less (adjusted hazard ratio (HR) 0.74, 95% confidence interval (CI) 0.56 to 0.99). As mentioned this result is borderline statistically significant.

They were also 26% less likely to have either probable dementia or mild cognitive impairment (HR 0.74, 95% CI 0.60 to 0.91).

The average amount of caffeine consumed in the women above the 172mg cut-off was 261mg, equating to about three cups of coffee or five cups of tea. The average amount consumed below the cut-off was 64mg.

How did the researchers interpret the results?

The researchers concluded that they showed a "lower risk of probable dementia or global cognitive impairment incidence in women with higher caffeine consumption, which are generally consistent with the literature."

Although more studies are needed to verify the consistency of reports, given that caffeine intake is easily modifiable, it is important to quantify its relationship with cognitive health outcomes not only from preventative stand point but also to better understand the underlying mechanisms and their involvement in dementia and cognitive impairment.

They further add "given that Alzheimer's disease prevalence is expected to quadruple by 2050, even a small reduction in age-related cognitive impairment or dementia burden would thereby have significant public health implications."

Conclusion

This study suggests there is a link between self-reported caffeine consumption and risk of developing probable dementia or some kind of cognitive impairment.

The findings could be important in leading to more research to investigate the mechanisms by which caffeine might provide protection against dementia and cognitive impairment. The hope being that such investigation may eventually lead to new forms of drug treatments.

However, there are several limitations of the study, including:

The level of caffeine was self-reported and may be inaccurate, particularly as it was presumed the reported intake of coffee, tea and cola were all caffeinated, which may not be the case.
As the actual levels of caffeine were not measured in the study, it may have been another component in coffee, tea of cola that was having a positive effect on the cognitive ability of the women.
The women in the sample were mostly white and generally highly educated and may not be representative of the general population which may have had an impact on the cognitive functioning scores.
Although some confounding factors were adjusted for, some were not, for example other aspects of women's diets and family history of dementia.
Breaking down level of caffeine consumed into more than two groups would have been useful to observe if cognitive function improves with caffeine consumed or if there is an optimum level of caffeine consumption somewhere in the middle.

Finally, the measures of cognitive function were not consistent throughout the study and therefore comparing women assessed using different instruments may lead to over- or under-estimation of cognitive impairment or probable dementia, causing inaccuracies in the conclusions.

Overall, it cannot be said that women consuming more caffeine are at a lower risk of developing cognitive impairment or probable dementia. More research is needed to identify whether there are mechanisms by which caffeine may be a protective factor in cognitive functioning.

If you are concerned about dementia then more effective methods of prevention (though not guaranteed) include exercise, healthy diet, quit smoking and moderate your consumption of alcohol. A good general rule is that what is good for heart tends to be good for the brain.