Skip to main content

Risk factors for stroke

Monday 21 June 2010

Just 10 lifestyle choices and medical conditions account for the majority of strokes, the_ Daily Mail_ reported. The newspaper said a study had found that 80% of cases were caused by high blood pressure, smoking, a fat stomach, poor diet and lack of exercise. Another 10% could be attributed to diabetes, excessive alcohol consumption, stress and depression, heart disorders and a higher concentration of molecules in the blood (apolipoproteins) that are involved in the transportation of bad cholesterol.

The newspaper report is accurate and importantly highlights that the majority of the risk factors for stroke are modifiable. In other words, they are a risk that can be changed.

This was a well-conducted international study. The researchers say that in future, they will address some of the study’s shortcomings, including recruiting an additional 10,000 stroke patients and matching controls, making the results more robust and allowing comparison between countries. Their conclusion that "targeted interventions that reduce blood pressure and smoking, and promote physical activity and a healthy diet, could substantially reduce the global burden of stroke", seems a sensible one.

Where did the story come from?

The study was carried out by researchers from McMaster University and other medical and academic institutions around the world, including in China, India, Uganda, Mozambique, Colombia and Australia. The study was funded by the Canadian Institutes of Health Research, the Heart and Stroke Foundation of Canada, the Canadian Stroke Network, Pfizer, Merck, AstraZeneca and Boehringer Ingelheim. The study was published in the peer-reviewed medical journal The Lancet.

The Daily Mail has covered this research appropriately.

What kind of research was this?

This story is based on the INTERSTROKE study, a multinational case control study in 6,000 people recruited from 22 countries between 2007 and 2010. The study’s aim was to determine which factors are associated with stroke and how much risk each factor contributes. To do this, researchers compared the exposures in 3,000 people who had an acute first stroke against a control group that had no history of stroke (matched for age and sex).

What did the research involve?

People who had experienced their first acute stroke were recruited from 84 centres in 22 countries. Patients were included after an acute stroke within five days of stroke symptoms first appearing or from when they were last seen without symptoms, within 72 hours of admission to hospital and if a brain scan (CT or MRI) was planned for within one week of initial diagnosis.

Stroke patients were asked to complete a questionnaire at the beginning of the study. For those who were incapable of this, a proxy respondent was identified (spouse or first-degree relative living in the same home). A control was identified for each case and matched in terms of sex and age (within five years). The control group was based either in hospital or in the community and had no history of a stroke. The researchers identified which type of stroke the participant had (ischaemic [blood clot] or intracerebral haemorrhagic [bleed into the brain]), based on clinical assessment and neuroimaging (CT or MRI).

The questionnaire assessed the participants’ risk factors, measuring key vascular risk factors, including hypertension (high blood pressure) or diabetes, anthropometrical measurements (waist, hip circumference, height and weight), physical activity, diet, alcohol, smoking and psychosocial factors.

To assist with defining hypertension, blood pressure and heart rate were recorded on three occasions for all cases. Physical activity and diet risk scores (including that associated with drinking) were calculated depending on responses to the questionnaires. Smoking was categorised as current, former or never. Depression was also ranked. Other measures, such as blood glucose and cholesterol, were assessed.

The researchers then used statistical models to determine which factors were linked with the risk of stroke. All of the findings took into account geographical region, gender, age and all potential confounders that had been measured. The researchers were interested in the association between stroke (all strokes and ischaemic and haemorrhagic stroke) and the following factors: hypertension, smoking status, diabetes mellitus, physical activity, diet, psychosocial factors, abdominal obesity, alcohol intake, and apolipoprotein concentrations (protein molecules linked to the transport of good and bad cholesterol). For each of these factors, the researchers calculated the population attributable fraction (PAF), an assessment of how much each factor contributes to the overall risk of the outcome.

What were the basic results?

Several factors were associated with an increased risk of stroke: history of hypertension, current smoking, waist-to-hip ratio, diet risk score (increased risk associated with increased consumption of red meat, organ meats or eggs, fried foods and cooking with lard), lack of physical activity, history of diabetes mellitus, more than 30 alcoholic drinks per month or binge drinking, psychological stress, other heart problems, and cholesterol. A model indicated that these factors accounted for 90% of the risk for all types of stroke. The following table shows the level of increased risk with each factor (individually these do not add up to 90% because of interactions between variables).

When they looked at the two different types of stroke separately, all of these factors were linked with ischaemic stroke. Only hypertension, smoking, waist-to-hip ratio, diet and alcohol were linked to intracerebral haemorrhagic stroke.

The PAFs for stroke for this population were calculated as follows:

  • 51.8% - Hypertension (self-reported history of hypertension or blood pressure >160/90mmHg)
  • 18.9% - Smoking status
  • 26.5% - Waist-to-hip ratio
  • 18.8% - Diet risk score
  • 28.5% - Regular physical activity
  • 5% - Diabetes mellitus
  • 3.8% - Alcohol intake
  • 4.6% - Psychosocial stress
  • 5.2% - Depression
  • 6.7% - Cardiac causes (atrial fibrillation, previous MI, rheumatic valve disease, prosthetic heart valve)
  • 24.9% - Ratio of ApoB to ApoA (reflecting cholesterol levels)

How did the researchers interpret the results?

The researchers note that their findings suggest that "10 risk factors are associated with 90% of the risk of stroke". They say that targeting interventions to reduce blood pressure and smoking and to promote physical activity and a healthy diet could substantially reduce the burden of stroke.


This is an important study that quantifies the contribution of different factors to the overall risk of stroke. As the researchers note, it provides information on the relative importance of different factors, building on the findings of previous epidemiological studies. Finding that hypertension is the biggest risk factor for all types of stroke is important because, like many other factors, it is a modifiable risk that can be addressed by appropriate medication and lifestyle changes. They say that this is particularly important in low-income settings as screening programmes need relatively little training, and resources and interventions are inexpensive.

It also confirmed that stopping smoking greatly reduces the risk of stroke, and that intake of fish and fruits were the dietary components most associated with risk reduction. Surprisingly, there was no association between vegetable intake and stroke. The researchers say that this needs further exploration. The researchers also highlighted inconsistencies in the research base, and point out the following limitations of their study:

  • A case-control design, as employed here, has several biases, including recall bias (participants’ responses being affected by their own memory or personal biases) and problems with selecting participants. The people in this study were all in hospital, so the findings may not be relevant for those with less or more severe strokes.
  • A reliance on hospital records for the type of stroke participants had. This potentially differs between countries.
  • Importantly, they note the small sample size and state that in the next phase they will include an additional 10,000 case-control pairs. This will be large enough to allow them to analyse patients by region and provide more information on how this profile of risk differs according to geography.

The researchers say that in subsequent phases of their research, they will address some of these shortcomings.

Overall, this study provides key information about the relative importance of the risk factors for overall stroke and for the different types of stroke. The conclusion that "targeted interventions that reduce blood pressure and smoking, and promote physical activity and a healthy diet, could substantially reduce the global burden of stroke", seems sensible.

Analysis by Bazian
Edited by NHS Website

Links to the headlines

Ten causes of a stroke: Scientists pinpoint triggers for 90% of cases.

Daily Mail, 21 June 2010

Links to the science

O'Donnell MJ, Xavier D, Liu L et al.

Risk factors for ischaemic and intracerebral haemorrhagic stroke in 22 countries (the INTERSTROKE study): a case-control study.

The Lancet 2010, Early Online Publication, June 18