Memory gaps in graduates a 'stroke warning sign'

Friday December 12 2014

“People with memory problems who have a university education could be at greater risk of a stroke,” BBC News reports. The hypothesis is that the gaps in memory could be the result of reduced blood flow to the brain, which may then trigger a stroke at some point in the future.

Researchers documented memory complaints and occurrences of stroke in a group of 9,152 adults aged over 55 living in the Netherlands, for an average of 12.2 years.

This showed that replying “yes” to the question “Do you have any memory complaints” was associated with a 20% higher risk of stroke overall compared with those saying “no”. This relative risk was higher in people categorised as educated – holding a university degree or higher vocational qualification.

This shouldn’t be interpreted as meaning that having a higher education increases your risk of stroke. The explanation put forward by the experts was that people with a higher education may have higher levels of cognitive awareness, so they may be more likely to be aware of worsening.

The study had a number of strengths, such as its long follow-up. However, its results were only significant using a subjective, self-reported measure of mental ability. A more objective assessment showed no link. There are many potential explanations for this, including the possibility that higher education compensates in some way.

Still, knowing what we do about blood flow and brain function, a link between memory problems and stroke is plausible.

Where did the story come from?

The study was carried out by researchers based in the Netherlands and was funded by a range of medical, science and academic research funding bodies from the Netherlands and the European Commission. There were no conflicts of interest reported.

The study was published in Stroke, a peer-reviewed Journal of the American Heart Association. The study has been published on an open-access basis so it is free to read online or download as a PDF.

BBC News reported the study accurately and although it outlined potential implications of the results, it did not discuss any of its limitations.

What kind of research was this?

This was a cohort study (Rotterdam Study) investigating whether memory complaints earlier in life were linked to occurrence of a stroke in later life.

The researchers say that people with cognitive impairment – some impairment of their brain’s ability to function – are at a higher risk of having a stroke.  

A stroke is a serious and potentially fatal condition in which the flow of blood to the brain is disrupted. Strokes have two main causes

  • a blood clot blocks the supply of blood to the brain (ischaemic stroke)
  • bleeding occurs inside the brain, usually due to a weakened blood vessel bursting (haemorrhagic stroke)

Both types of stroke can occur in people with cardiovascular disease:

The research team wanted to know whether there were early signs of cognitive impairment, such as memory lapses, that could help them identify people at higher risk of stroke. If they knew who the high risk people were, they could focus efforts on minimising their risk, potentially preventing some occurrences of stroke. 

What did the research involve?

The researchers documented memory complaints and any occurrences of stroke in a group of 9,152 adults over 55 living in Rotterdam, the Netherlands, for an average of 12.2 years.

Trained investigators interviewed all participants at home. The presence of subjective memory complaints was assessed by the question, “Do you have memory complaints?” Cognitive function was assessed using the standard objective measure of the Mini-Mental State Examination. This assesses orientation, memory, attention, language, and visuospatial construction (the ability to recognise a pattern or set of objects and then replicate the pattern or set). It wasn’t clear when the assessment of memory complaints occurred, or whether it was reported over time.

Once enrolled in the study, participants were left to their own devices while the researchers were notified of any reports of stroke in the following years.

People who already had a stroke or had dementia at enrolment into the study were excluded. The number of participants available for analysis was 9,152.

The research team analysed links between memory complaints and incidence of stroke. They also looked at whether education level influenced this link. The analysis took account of a range of known confounders for stroke risk, including:  

  • age
  • gender
  • smoking
  • body mass index
  • blood cholesterol levels
  • diabetes
  • blood pressure and blood pressure medications
  • ability to perform activities of daily living – a type of older age disability test

What were the basic results?

Over the study period 1,134 strokes occurred, average follow up was 12.2 years.

One of the main findings was that people reporting subjective memory complaints were 20% more likely to have a stroke than those who didn’t (Hazard ratio [HR] 1.20, 95% Confidence interval [CI] 1.04 to 1.39). However, this result was not found using the more objective measure of mental ability, the Mini-Mental State Examination. Better point scores on the test were not significantly linked to stroke occurrence (HR 0.99, 95% CI 0.95 to 1.02). These figures come from the analyses that took account of the largest list of confounders.

The second important finding was that level of education was significantly influencing the results. Subjective memory complaints were linked with stroke only in those with high education – defined in this study as higher vocational education or university training (HR 1.39, 95% CI 1.07 to 1.81).

Participants with missing information tended to be older, had more memory complaints, more likely to be female and had slightly worse scores on the assessments of mental abilities. These people were still included in the analysis.

How did the researchers interpret the results?

The researchers concluded “Subjective memory complaints might be an early indicator of stroke risk, especially in highly educated individuals”.


This study showed that highly educated people who notice memory complaints in themselves may be more likely to develop stroke than those who don’t, over an average of 12 years.

The study had a number of strengths, such as its population-based prospective design and availability of data on more than 9,000 participants at baseline with a long follow-up. However, there were also a number of limitations that weaken the strength of the conclusions.

It was not clear whether the memory complaints were assessed just once at the start of the study or an ongoing basis. Some people may report memory complaints that are only temporary, while others not initially reporting complaints may have done so in later years. This could have altered the results, but is only likely to have had a minor influence.

The results were only significant using the subject measure of mental ability. It would be interesting to explore whether other subjective and objective assessments show a link or not. Results tend to be more reliable if there is consistency between different measures of the same thing, objective or subjective. This was not the case in this study. 

The Mini-Mental State Examination is known to be less sensitive in well-educated patients. Possibly a different type of test is required.

Although the study adjusted for a range of confounders, it is difficult to exclude the possibility that residual confounding by measurement error or unmeasured factors biased the results to an unknown degree.

Data on depression and depressive symptoms was not available. The researchers highlight that this was a major limitation, “because it has been suggested that the associations with subjective indicators of health, especially memory, may be confounded by prevalence of depression”.

The bottom line is that this study suggests a link between memory complaints in the highly educated and stroke but does not prove one causes the other. The study authors point to a plausible biological explanation but this was not tested in this study.

The results may warrant further investigation and confirmation in different studies, using different ways of assessing memory. If the link is real, we’d expect to see somewhat consistent results across different measures. Based on this study we can’t say that educated people with memory complaints are definitely at a higher risk of stroke.

Nevertheless, vascular dementia (where reduced blood flow to the brain causes cognitive dysfunction) and stroke are both linked to the same underlying cardiovascular disease process, so a link between memory problems and stroke is plausible – particularly for this specific type of dementia.

Methods you can use to reduce your stroke risk include eating a healthy diet, taking regular exercise, quitting smoking if you smoke, moderating your consumption of alcohol. Read more about stroke prevention.

Analysis by Bazian
Edited by NHS Choices