This site is intended for health professionals only

At the heart of general practice since 1960

Risk score predicts 10-year risk of dementia in elderly with diabetes

A simple risk score could allow GPs to identify older patients with diabetes who are at high risk of developing dementia within 10 years.

US researchers who developed the score said it could be used to target preventative treatments towards those most at risk, although experts questioned how much use it would be given the current lack of proven effective preventative interventions for dementia.

The score – reported this month in the Lancet Diabetes & Endocrinology – is based on the patient’s age, education level and diabetes-related complications, such as severe hyper- or hypoglycaemia, microvascular disease or stroke.

The researchers developed the score by analysing records of 29,961 patients with diabetes, with an average age of 70.6 years, and selecting risk factors that most strongly predicted the onset of dementia over 10 years. They then assigned points to each risk factor according to its predictive strength, giving a range of 14 categories ranging from a score of -1 to a score of 12–19.

Dementia risk increased linearly across categories and patients with the highest score (12–19) were 37 times as likely to develop dementia over 10 years as those with the lowest (-1). Higher scores were also associated with a shorter time to dementia onset, with an average time of 5.2 years for the highest category compared with 9.8 years in the lowest-risk category.

The score predicted dementia just as accurately when the researchers tested it out in a separate cohort of 2,413 patients, with a summary C-statistic of 0.746 compared with 0.736 in the original cohort.

Lead author Dr Rachel Whitmer, from the Kaiser Permanente Division of Research in California, said: ‘Early detection of patients with type 2 diabetes who are at increased risk of dementia could help to develop and target preventive treatment, and our scoring system has the potential to change clinical care by giving clinicians a simple and accurate way of predicting the risk of dementia in older people with type 2 diabetes.’

But an editorial accompanying the study questioned the premise for using such a risk score until interventions are proven to be effective for preventing dementia.

Author Dr Anna-Maija Toppanen, from the University of Eastern Finland, said: ‘Generally risk scores might be useful in the identification of  individuals who should be monitored for disease symptoms, selection of high-risk individuals for clinical trials, targeting of preventive interventions towards those at greatest risk, and assessment of the effectiveness of an intervention at reducing the risk of future illness.

‘[This score] might be useful for clinicians for the first purpose, but clinical trial data on effective preventive interventions on dementia are currently lacking.’

Lancet Diabetes & Endocrinology 2013; online 20 August

Have your say