Risk score for prediction of 10 year dementia risk in individuals with type 2 diabetes: a cohort study

Background Although patients with type 2 diabetes are twice as likely to develop dementia as those without this disease, prediction of who has the highest future risk is difficult. We therefore created and validated a practical summary risk score that can be used to provide an estimate of the 10 year dementia risk for individuals with type 2 diabetes. Methods Using data from two longitudinal cohorts of patients with type 2 diabetes (aged >= 60 years) with 10 years of follow-up, we created (n=29 961) and validated (n=2413) the risk score. We built our prediction model by evaluating 45 candidate predictors using Cox proportional hazard models and developed a point system for the risk score based on the size of the predictor's beta coefficient. Model prediction was tested by discrimination and calibration methods. Dementia risk per sum score was calculated with Kaplan-Meier estimates. Findings Microvascular disease, diabetic foot, cerebrovascular disease, cardiovascular disease, acute metabolic events, depression, age, and education were most strongly predictive of dementia and constituted the risk score (C statistic 0.736 for creation cohort and 0.746 for validation cohort). The dementia risk was 5.3% (95% CI 4.2-6.3) for the lowest score (-1) and 73 3% (64.8-81.8) for the highest (12-19) sum scores. Interpretation To the best of our knowledge, this is the first risk score for the prediction of 10 year dementia risk in patients with type 2 diabetes mellitus. The risk score can be used to increase vigilance for cognitive deterioration and for selection of high-risk patients for participation in clinical trials.