Nomogram prediction of caries risk among schoolchildren age 7 years based on a cohort study in Shanghai

Abstract

Objective Caries risk assessment tools are essential for identifying and providing treatment for individuals at high risk of developing caries. We aimed to develop a nomogram for the assessment and evaluation of caries risk among Chinese children. Methods We enrolled schoolchildren age 7 years from a primary school in Shanghai. Baseline information of participants was collected using a questionnaire completed by children’s caregivers. A nomogram of a novel prediction scoring model was established based on predictors detected in univariate and multivariate analyses. Predictive accuracy and discriminative ability of the nomogram were calculated using the concordance index (C index). The bootstrap method (1000 samples) was used to decrease overfitting. The net benefit of the model was validated using decision curve analysis. Results Overall, 406 children with complete information and two completed dental examinations were included in the final analysis. The nomogram based on logistic regression model coefficients demonstrated a C index of 0.766 (95% confidence interval: 0.761–0.771) for caries risk. The net benefit of the decision curve analysis was 38.6% at 55% threshold probability. Conclusion This nomogram model, derived using dietary habits, oral hygiene status, and caries experience, showed promising predictive ability to assess the caries risk among Chinese children.