This article is more than 5 years old. Information may no longer be current.

New test predicts dementia during routine family doctor visits

Information collected during routine primary care visits can be used to predict 5-year risk for dementia diagnosis in patients aged 60 to 79 years, but not those 80 and older, according to data published in BMC Medicine.

“There is a strong drive internationally for clinicians to be more proactive in dementia diagnosis,” Kate Walters, PhD, MSc, MMed, of the research department of primary care and population health, at University College London, and colleagues wrote. “There is, however, a limited evidence base for current approaches to dementia screening and case-finding, and further work needs to be completed to validate new methods across different settings, including primary care.”

To develop and validate a 5-year dementia risk score, derived solely from information routinely collected during family doctor visits, the researchers examined data from general practices across the United Kingdom in The Health Improvement Network (THIN) database. They selected a development cohort of 377 practices, identifying 930,395 patients aged 60 to 95 years without a prior history of dementia, cognitive impairment or memory symptoms at baseline. The researchers developed risk algorithm models for those 60 to 79 years, and those 80 to 95 years.

The researchers then validated their models using a separate cohort of 264,224 patients in 95 randomly chosen THIN practices that were not part of the development cohort. Their main outcome was 5-year risk of first recorded dementia diagnosis. Potential predictors included sociodemographic, cardiovascular, lifestyle and mental health variables.

In the development cohort, for patients aged 60 to 79 years, there were 6,017 new diagnoses in 3,205,190 person years at risk (PYAR), representing an incidence rate of 1.88 per 1,000 PYAR. Among patients aged 80 to 95 years, there were 7,104 new dementia diagnoses in 429,670 PYAR, for an incidence of 16.53 per 1,000 PYAR.

Predictors for those aged 60 to 79 included increasing age, female sex, calendar year, living in a deprived area, smoking, hazardous or harmful alcohol drinking and a history of stroke/transient ischemic attack (TIA), diabetes, coronary heart disease, atrial fibrillation, or current depression and antidepressant drug, anxiety drug, hypnotic drug and aspirin use.

Among patients aged 80 to 95 years, new dementia diagnoses were associated with age, female sex, history of stroke/TIA, diabetes, atrial fibrillation, hazardous or harmful alcohol drinking, statin prescriptions, current depression and antidepressant drug, anxiety drug, hypnotic drug and aspirin use. In addition, there was a small negative association with current smoking, BMI, systolic BP, antihypertensive drugs and NSAIDs, excluding aspirin, among this group.

When applied to the validation cohort, which had incidence rates similar to those in the development cohort, the researchers’ model performed well for patients aged 60 to 79 years in terms of discrimination, with a Royston’s D statistic of 2.03 (95% CI, 1.95-2.11), and Uno’s C index 0.84 (95% CI, 0.81-0.87). The calibration slope was 0.98 (95% CI, 0.93-1.02). However, the model performed poorly for patients aged 80 to 95 years, in terms of both discrimination and calibration.

“Our new dementia risk algorithm for 60- to 79-year-olds can be added to clinical software systems and a practice could, for example, run this risk model on all eligible people and offer those at risk more detailed testing or specific preventive management,” Walters and colleagues wrote. “ … 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.” – by Jason Laday

Disclosure: The authors report no relevant financial disclosures.