Readily available data collected by GPs could be used to predict dementia risk, the BMC Medicine study found, but experts have advised caution before making the indicator available in a clinical setting.
Researchers at University College London assessed anonymised data from over 1m GP records to calculate the individual Dementia Risk Score (DRS) of each patient – a predictor of their likelihood to develop dementia over the next five years.
Data was obtained for patients aged 60-95 from a random selection of 377 practices using data from The Health Improvement Network (THIN) database. Those with recorded dementia, cognitive or memory problems and precursor conditions were excluded.
The patients were then monitored over a five-year period to assess whether they later developed dementia.
The researchers claim that the score was able to predict people aged 60-79 with a very low risk of developing dementia over the next five years with an accuracy of 85%. Around one in every 10 people identified went on to develop the condition.
However, the risk score was less effective at predicting those at a high risk of dementia or risk for patients over 80. This latter point suggests that it may be reasonable to test for dementia in patients over 80 ‘on the basis of age alone’, the researchers said.
The scores draw on several factors contained within the record, including history of depression, stroke, alcohol intake, diabetes, irregular heart rate, weight loss, smoking and high BP.
Factors such as exercise, education level and family history were not taken into account, as these are not routinely collected by GPs.
Other dementia risk scores have been developed in the past, but the DRS is the first that uses data that is already recorded as a matter of course – which makes it a cheaper and easier option. The assessment could also be automated, the researchers said.
The researchers said the score could be used to ‘rule out’ patients with very low risks from further testing or intensive preventative activities and could allow GPs to make ‘more proactive diagnoses’ of dementia by catching the disease earlier.
‘Routinely collected healthcare data has potential in assessment of dementia risk in clinical populations, without collecting further information from patients,’ they said.
‘Further research should be undertaken to explore the performance of the DRS in different settings and populations, including variations in performance in areas where the prevalence, detection and recording of dementia by GPs is very low or very high.
‘We also need to further understand how the tool might be used in practice, the ethical implications and what the impact of this might be for older people, clinicians and the potential costs for health services.’
Clare Walton of the Alzheimer’s Society said the score must be further refined before being rolled out to a clinical setting. Being able to identify patients at high risk of dementia would be ‘more useful’, she added.
‘In my opinion, the score didn’t perform well enough to be used in a clinical setting,’ she said. ‘Unless the algorithm is further refined, dementia risk would be incorrectly predicted too often and could cause more harm than good.
‘Before we start routinely giving people a risk score in clinics, we need to look into the impact this information will have on wellbeing and on the standard of care that people receive.’