Wenqing Zhu,1,* Shuoyuan Tan,2,* Zhitong Zhou,1 Miaomiao Zhao,3 Yingquan Wang,4 Qi Li,5 Yang Zheng,6 Jianwei Shi7,8 

1Tongji University School of Medicine, Tongji University, Shanghai, People’s Republic of China; 2School of Public Health, Shanghai Jiaotong University School of Medicine, Shanghai, People’s Republic of China; 3School of Clinical Medicine, Shanghai University of Medicine and Health Sciences, Shanghai, People’s Republic of China; 4Department of NCD Surveillance, Division of Chronic Non-Communicable Diseases and Injury, Shanghai Municipal Center for Disease Control & Prevention, Shanghai, People’s Republic of China; 5Department of Vital Statistics, Shanghai Municipal Center for Disease Control & Prevention, Shanghai, People’s Republic of China; 6Division of Chronic Non-Communicable Diseases and Injury, Shanghai Municipal Center for Disease Control & Prevention, Shanghai, People’s Republic of China; 7Department of General Practice, Yangpu Hospital, Tongji University School of Medicine, Shanghai, People’s Republic of China; 8Department of Social Medicine and Health Management, School of Public Health, Shanghai Jiaotong University School of Medicine, Shanghai, People’s Republic of China

*These authors contributed equally to this work

Correspondence: Jianwei Shi, Email shijianwei_amy@126.com

Introduction: Cardiovascular diseases (CVD) represents a leading cause of morbidity and mortality worldwide, including China. Accurate prediction of CVD risk and implementation of preventive measures are critical. This study aimed to develop a short-term risk prediction model for CVD events among individuals aged ≥ 60 years in Shanghai, China.
Methods: Stratified random sampling recruited elderly individuals. Retrospective data (2016– 2022) were analyzed using Lasso-Cox regression, followed by a multivariable Cox regression model. The risk scoring was visualized through a nomogram, and the model performance was assessed using calibration plots and receiver operating characteristic curves.
Results: A total of 9,636 individuals aged ≥ 60 years were included. The Lasso-Cox regression analysis showed male gender (HR=1.482), older age (HR=1.035), higher body mass index (HR=1.015), lower high-density lipoprotein cholesterol (HR=0.992), higher systolic blood pressure (HR=1.009), lower diastolic blood pressure (HR=0.982), higher fasting plasma glucose (HR=1.068), hypertension (HR=1.904), diabetes (HR=1.128), and lipid-lowering medication (HR=1.384) were related to higher CVD risk. The C-index in the training and validation data was 0.642 and 0.623, respectively. Calibration plots indicated good agreement between predicted and actual probabilities.
Conclusion: This short-term predictive model for CVD events among the elderly population exhibits good accuracy but moderate discriminative ability. More studies are warranted to investigate predictors (gender, high-density lipoprotein cholesterol, systolic blood pressure, diastolic blood pressure, hypertension, and lipid-lowering medication) of CVD incidence for the development of preventive measures.

Keywords: cardiovascular disease, elderly population, predictive model, short-term risk, Lasso-Cox regression, China