Development of a Clinical Risk Score Prediction Tool for 5-, 9-, and 13-Year Risk of Dementia

IMPORTANCE Although researchers have devoted substantial efforts, money, and time to studying the causes of dementia and the means to prevent it, no effective treatment exists yet. Identifying preclinical risk factors of dementia could help prevent or delay its progression. OBJECTIVE To develop a point risk score prediction model of dementia. DESIGN, SETTING, AND PARTICIPANTS This study used a large UK population-based prospective cohort study conducted between March 13, 2006, and October 1, 2010. Data analysis was performed from June 7 to September 15, 2021. Individual analyses of time end points were concluded at the first dementia diagnosis during the follow-up period. The data were split into training and testing data sets to separately establish and validate a prediction model. MAIN OUTCOMES AND MEASURES Outcomes of interest included 5-, 9-, and 13-year dementia risk. Least absolute shrinkage and selection operator and multivariate Cox proportional hazards regression models were used to identify available and practical dementia predictors. A point risk score model was developed for the individual prediction of 5-, 9-, and 13-year dementia risk. RESULTS A total of 502505 participants were selected; the population after exclusions for missing data and dementia diagnosis at baseline was 444 695 (205187 men; mean [SD] age, 56.74 [8.18] years; 239 508 women; mean [SD] age, 56.20 [8.01] years). Dementia occurrence during the 13 years of follow-up was 0.7% for men and 0.5% for women. The C statistic of the final multivariate Cox proportional hazards regression model was 0.86 for men and 0.85 for women in the training data set, and 0.85 for men and 0.87 for women in the testing data set. Men and women shared some modifiable risk and protective factors, but they also presented independent risk factors that accounted for 31.7% of men developing dementia and 53.35% of women developing dementia according to the weighted population-attributable fraction. The total point score of the risk score model ranged from -18 to 30 in men and -17 to 30 in women. The risk score model yielded nearly 100% prediction accuracy of 13-year dementia risk both in men and women. CONCLUSIONS AND RELEVANCE In this diagnostic study, a practical risk score tool was developed for individual prediction of dementia risk, which may help individuals identify their potential risk profile and provide guidance on precise and timely actions to promote dementia delay or prevention.