Development and validation of a nomogram prediction model for hypertension-diabetes comorbidity based on chronic disease management in the community

Abstract

Purpose: ​Develop and validate a nomogram prediction model for hypertension-diabetes comorbidities based on chronic disease management in the community. Patients and methods: The nomogram prediction model was developed in a cohort of 7200 hypertensive patients at a community health service center in Hongshan District, Wuhan City. The data were collected from January 2022 to December 2022 and randomly divided into modeling and validation groups at a 7:3 ratio. The Lasso regression model was used for data dimensionality reduction, feature selection, and clinical test feature construction. Multivariate logistic regression analysis was used to build the prediction model. Results: The application of the nomogram in the verification group showed good discrimination, with an AUC of 0.9205 (95% CI: 0.8471–0.9527) and a good calibration effect. Decision curve analysis demonstrated that the predictive model was clinically useful. Conclusion: This study presents a nomogram prediction model that incorporates age, waist-height ratio and elevated density lipoprotein cholesterol (HDL-CHOLESTEROL), which can be used to predict the risk of codeveloping diabetes in hypertensive patients.