This site is intended for health professionals only


Risk score ‘could help rule out dementia’



Researchers have developed a risk score that uses information gathered routinely by GPs and could help predict a person’s risk of developing dementia over the next five years.

The team, from University College London, say that the score is accurate at ruling out the likelihood of dementia in ‘low-risk’ patients and could therefore help to reassure people worried about memory problems.

The Dementia Risk Score (DRS) has a high negative predictive value, and can predict that a person aged 60-79 will not develop dementia within five years with 85% accuracy.

But its positive predictive value was less accurate and the score could not identify those that would have a dementia diagnosis within five years, nor could it predict the risk for those aged over 80.

Concluding in the study, researchers said: ‘This risk score can be used to identify higher risk populations for dementia in primary care. The risk score has a high negative predictive value and may be most helpful in ”ruling out” those at very low risk from further testing.’

Dr Kate Walters, lead researcher, told the Times: ‘I am a GP and it is common to see people worried about developing dementia, so for some people it could be very reassuring.’

Researchers developed the DRS by collecting data from almost a million patients with no previous record of dementia and built a computer algorithm that used variables taken routinely by GPs, including health and lifestyle measurements, to predict the risk of dementia within a five-year follow-up period.

To repeat and validate their findings, the researchers then selected an additional 250,000 patients to test their algorithm on.

While age and sex were key predictors used in the measurement, social deprivation, smoking, BMI and depression were also found to contribute to the risk score.

But experts are cautious against using the DRS and suggest further research is required to determine its validity.

Dr Doug Brown, director of research and development at Alzheimer’s Society said: ‘For every 10 people classed as low risk with this method, at least one person could still go on to develop dementia. This level of uncertainty reduces the likelihood that the score would be used widely by doctors.

‘The fact that this test uses routine information gathered during a GP consultation is certainly advantageous, but before it is recommended for use in clinics, it would need to be backed up by further research.’

’Predicting dementia risk in primary care…’ has been published in BMC Medicine