Background: Existing dementia risk scores require collection of additional data from patients, limiting their use in\npractice. Routinely collected healthcare data have the potential to assess dementia risk without the need to collect\nfurther information. Our objective was to develop and validate a 5-year dementia risk score derived from primary\nhealthcare data.\nMethods: We used data from general practices in The Health Improvement Network (THIN) database from across\nthe UK, randomly selecting 377 practices for a development cohort and identifying 930,395 patients aged 60ââ?¬â??95 years\nwithout a recording of dementia, cognitive impairment or memory symptoms at baseline. We developed risk algorithm\nmodels for two age groups (60ââ?¬â??79 and 80ââ?¬â??95 years). An external validation was conducted by validating the model on\na separate cohort of 264,224 patients from 95 randomly chosen THIN practices that did not contribute to the\ndevelopment cohort. Our main outcome was 5-year risk of first recorded dementia diagnosis. Potential predictors\nincluded sociodemographic, cardiovascular, lifestyle and mental health variables.\nResults: Dementia incidence was 1.88 (95 % CI, 1.83ââ?¬â??1.93) and 16.53 (95 % CI, 16.15ââ?¬â??16.92) per 1000 PYAR for those\naged 60ââ?¬â??79 (n = 6017) and 80ââ?¬â??95 years (n = 7104), respectively. Predictors for those aged 60ââ?¬â??79 included age, sex,\nsocial deprivation, smoking, BMI, heavy alcohol use, anti-hypertensive drugs, diabetes, stroke/TIA, atrial fibrillation,\naspirin, depression. The discrimination and calibration of the risk algorithm were good for the 60ââ?¬â??79 years model; D\nstatistic 2.03 (95 % CI, 1.95ââ?¬â??2.11), C index 0.84 (95 % CI, 0.81ââ?¬â??0.87), and calibration slope 0.98 (95 % CI, 0.93ââ?¬â??1.02). The\nalgorithm had a high negative predictive value, but lower positive predictive value at most risk thresholds.\nDiscrimination and calibration were poor for the 80ââ?¬â??95 years model.\nConclusions: Routinely collected data predicts 5-year risk of recorded diagnosis of dementia for those aged 60ââ?¬â??79, but\nnot those aged 80+. This algorithm can identify higher risk populations for dementia in primary care. The risk\nscore has a high negative predictive value and may be most helpful in ââ?¬Ë?ruling outââ?¬â?¢ those at very low risk from further\ntesting or intensive preventative activities.
Loading....