select search filters
roundups & rapid reactions
before the headlines
Fiona fox's blog

expert reaction to observational study looking at tea and coffee consumption and stroke/dementia risk

An observational study published in PLOS Medicine looks at consumption of tea and coffee and risk of developing stroke, dementia, and poststroke dementia.


Prof Tara Spires-Jones, UK Dementia Research Institute Programme Lead and Deputy Director, Centre for Discovery Brain Sciences, University of Edinburgh, said:

“This study examined data from large numbers of people looking at whether consuming tea and coffee are associated with risk of stroke or dementia.  The research is interesting and robustly conducted, but as the authors point out, it is not possible to know from this kind of data whether the tea and coffee drinking are the cause of reduced risk of stroke or dementia. There could be other factors in people that drink tea and coffee that are the real influencers of disease risk. More work needs to be done to understand the potential biological links between tea and coffee drinking and stroke/dementia.”


Prof David Llewellyn, Professor of Epidemiology and Digital Health at the University of Exeter, said:

“The relationship between coffee consumption dementia, stroke and neuroimaging findings in the UK Biobank study have already been reported in our publication in Nutritional Neuroscience earlier this year. Drinking coffee in moderation is associated with a lower risk of dementia, but the highest risk of dementia is associated with heavy coffee consumption. We found that drinking more than six cups of coffee a day was associated with 53 per cent increase in dementia risk in comparison with people drinking one or two cups a day.

“The new study extends our previous analyses by demonstrating a similar relationship with tea drinking and clarifying the association with stroke. However, it is questionable why they concluded that drinking tea and coffee may reduce the risk of stroke and dementia. Indeed, if people already drink tea or coffee in moderation then increasing their consumption further could potentially increase their risk of having a stroke or developing dementia. The potential harm linked with heavy consumption is particularly clear for coffee.

“A more balanced conclusion may be that drinking tea and coffee in moderation is associated with a lower risk of stroke and dementia. In theory people who don’t drink tea or coffee at all may benefit from starting to drink in moderation, and heavy consumers of tea and coffee may benefit from drinking less. Whether this really makes a difference to the risk of having a stroke or developing dementia is not yet known.”


Dr Charlotte Mills, Lecturer in Nutritional Sciences at the University of reading, said:

“This is an interesting and well-conducted study that shows a link between drinking tea and/or coffee and having a lower risk of stroke or dementia.

“The research only shows a link and doesn’t show that the tea or coffee causes the reduced risk of disease. It may be that there are other factors at work.

“However, the finding is consistent with other research showing a link between drinking tea and coffee and other health benefits, such as reduced risk of heart attack or developing type-2 diabetes.

“One limitation of this work is that the results do not appear to consider milk consumption as a confounding factor. Typically in the UK most people drink tea and coffee with added milk, and there is some evidence that milk consumption might also reduce the risk of stroke.

“Both tea and coffee contain large quantities of natural bioactive chemicals, particularly polyphenols, which can also be found in blueberries and cocoa. These have previously been shown to provide many health benefits, including cutting the risk of neurodegenerative diseases. It is possible that these polyphenols in the tea or coffee may be the cause of the benefits observed in this research.”


Dr Rosa Sancho, Head of Research at Alzheimer’s Research UK, said:

“For most of us, our risk of dementia depends on the complex interaction of our age, genetics and lifestyle. We know that stroke increases the risk of developing vascular dementia. Understanding which aspects of our lifestyle have the greatest effect on our brain health is key to empowering people to make informed decisions about their lives.

“Studies like this one are not able to pinpoint cause and effect, and while the researchers attempted to control for other factors that could affect a person’s risk of stroke and vascular dementia, no firm conclusions can be made about whether tea or coffee cause this lower risk. Participants only reported tea and coffee consumption at the beginning of the study, and there is no data on long-term habits, so it’s not clear how relevant the findings are to long-term brain health.

“While previous studies have looked at associations between tea and coffee consumption and better brain health, there has been inconsistency in findings. Future research with participants of a range of ages and ethnicities will be needed to fully understand what types of dementia and stroke are associated with tea and coffee drinking. Participants in this study reported themselves to be mainly White British (96%), therefore we cannot infer an association that is relevant to everyone in the UK.”


Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:

“In an observational study, like this new piece of research, people aren’t allocated by the researchers to drink particular amounts of tea and coffee. They just drink what they’d drink anyway. The researchers record that, and they can then calculate how the participants’ chances of having a stroke or a dementia diagnosis are related to how much coffee or tea they said they drank.

“The big problem is that, if the researchers find an association between beverage consumption and disease risk, they can’t say whether any differences in risk are actually caused by the coffee or tea consumption. The researchers themselves make that admirably explicit. It happens because there are very many other differences between people who drink different amounts of tea and coffee, apart from the differences in their tea and coffee consumption. Some of these other differences might be the real cause of any differences in stroke or dementia risk, and not the tea or coffee at all. That’s particularly likely with diseases like strokes and dementia, where the risks are known to be related to many kinds of factor.

“Researchers can make statistical adjustments that attempt to allow for differences in other factors, and the researchers on this study did do that. That still can’t settle the issue of what causes what, though, because the web of cause and effect is too complex. And that’s basically why we still don’t know lots of things about possible health effects of tea and coffee, despite all the studies that have been done over the years. It’s close to impossible to do other most other types of study on these things, so we’re stuck with observational studies and their inevitable limitations.

“The researchers mention other limitations too – for instance, tea and coffee consumption were recorded only when people first joined the study, and might have changed during the 11 year average time their health was being recorded.

“What did the researchers find? A stated aim of the study was to look at different combinations of tea and coffee consumption on stroke and dementia risk, but I don’t think that completely worked out. It’s true that, as the press release says, the research found a 32% reduction in stroke risk, in a given time, in people who drank 2-3 cups of coffee and also 2-3 cups of tea a day, compared to those who drank no coffee or tea. But there’s quite a lot of statistical uncertainty about that figure. So there’s no implication that somehow 2-3 cups of each a day is the ‘best’ combination.

“What is interesting, though, and not very obvious from the press release, is that the researchers found associations between stroke and dementia risk that did not increase or decrease uniformly with tea and coffee consumption. Instead, what generally happened is that the risk of stroke or dementia was lower in people who drank reasonably small amounts of coffee or tea compared to those who drank none at all, but that after a certain level of consumption, the risk started to increase again until it became higher than the risk to people who drank none.

“For coffee drinking, the risk of a stroke was lower for people who drank between about half a cup a day and about five cups a day, than for people who drank no coffee at all. The lowest risk was for people who drank about 2-3 cups a day, as the press release says. But once the coffee consumption got up to 7 or 8 cups a day, the stroke risk was greater than for people who drank no coffee, and quite a lot higher than for those who drank 2 or 3 cups a day. The position is a bit different with dementia risk – the press release is right to say that the lowest dementia risk estimates were again for those who drank 2 or 3 cups of coffee a day, but the margins of error mean that it’s not at all clear whether the risk is really any different from the dementia risk for those who drank no coffee at all. In people that drank more than 2 or 3 cups a day, the estimated risk of dementia started to look higher, and people who drank more than about 8 cups seemed definitely to be at higher dementia risk than people who drink no coffee.

“How much difference is there in the risks of stroke or dementia for people who drink smallish amounts of tea and coffee, compared to those who drink none? Of the people in this study, who, as the researchers make clear, aren’t really typical of the UK or any other population, for every 1,000 people who drank no coffee at all, about 30 had a stroke during the 11-year follow-up. Another group of 1,000, who were similar to the non-coffee drinkers, but instead drank 2 or 3 cups of coffee a day, would have had, on the basis of this research, between 25 and 28 strokes over the same period – so rather fewer. The difference is not immense, though strokes are generally quite common and serious, and a difference of this size deserves thinking about. But because it’s an observational study, we still can’t be sure that the difference is caused by coffee consumption. The calculations for tea and stroke turn out to be similar. In this study, for every 1,000 people who drank no tea at all, about 31 had a stroke during the follow-up. A similar group of 1,000 who drank 2 or 3 cups of tea a day would have between 25 and 28 strokes over the same time – again we can’t say that the reduction is caused by the tea.

“The dementia risk for the people in this study was less than the stroke risk, because generally they weren’t yet into the time of life when dementia risk really increases a lot. For those who drank no coffee, about 15 in every 1,000 had a new diagnosis of dementia during follow-up. The number of newly diagnosed dementias was also about 15 in every 1,000 who drank no tea. If a similar group of 1,000 people drank 2 or 3 cups of coffee instead of none, the number having a new dementia diagnosis would be between 12 and 14, so a reduction but not a large one. If a similar group to the non-tea-drinkers instead drank 2 or 3 cups of tea a day, then there would be between 12 and 15 dementia diagnoses – so a small reduction, or possibly no reduction at all.”



‘Consumption of coffee and tea and risk of developing stroke, dementia, and poststroke dementia: A cohort study in the UK Biobank’ by Yuan Zhang et al. was published in PLOS Medicine at 7pm UK TIME on Tuesday 16 November 2021.



Declared interests

Prof Kevin McConway: “I am a Trustee of the SMC and a member of its Advisory Committee.  My quote above is in my capacity as an independent professional statistician.”

Prof Tara Spires-Jones: “I have no conflicts of interest with this study.”

Dr Rosa Sancho: “no conflicts of interest.”

None others received.

in this section

filter RoundUps by year

search by tag