Construction and evaluation of a nomogram prediction model of atherogenesis risk in patients with type 2 diabetes mellitus
10.3969/j.issn.1006-2483.2024.05.013
- VernacularTitle:2型糖尿病患者动脉粥样硬化发生风险的列线图预测模型构建与评价
- Author:
Chaojun SHI
1
;
Zijun LIU
1
;
Yifan WANG
1
;
Weiqin CAI
1
;
Qi JING
1
;
Hongqing AN
1
;
Qianqian GAO
1
Author Information
1. Shandong Second Medical University , Weifang , Shandong 261053 , China
- Publication Type:Journal Article
- Keywords:
Atherosclerosis;
Type 2 diabetes;
Nomogram;
Prediction model
- From:
Journal of Public Health and Preventive Medicine
2024;35(5):56-59
- CountryChina
- Language:Chinese
-
Abstract:
Objective To analyze the risk factors influencing the occurrence of atherosclerosis in patients with type 2 diabetes, and to construct and evaluate a nomogram prediction model. Methods Multivariate logistic regression was used to analyze the risk factors of atherosclerosis in type 2 diabetes mellitus, and R software was used to build a nomogram prediction model. The accuracy and clinical validity of the model were verified by using H-L fit curve, area under ROC curve and calibration curve. Results The prevalence rate of atherosclerosis was 56.37%. Independent risk factors for atherosclerosis in type 2 diabetes mellitus (P<0.05) were body weight (OR=1.42，P<0.05), glycated serum protein (OR=1.35, P<0.05), lactate dehydrogenase (OR=1.17, P<0.05), alkaline phosphatase (OR=0.79, P<0.05), hyperlipidemia (OR=2.30, P<0.05), stroke (OR=4.20, P<0.05), coronary heart disease (OR=64.54, P<0.05), lower extremity artery disease (OR=24.52, P<0.05), and other endocrine diseases (OR=1.65 , P<0.05). The area under ROC curve was 0.91, the slope of the calibration curve was close to 1, and the H-L fit curve χ2=3.11. The internal verification result of the constructed nomogram prediction model was P=0.93. External verification of patients in the test set showed that the area under ROC curve was 0.91, indicating good differentiation and accuracy of the model. Conclusion The prediction model established by using the risk factors screened in this study has a high accuracy and differentiation, and medical staff can take effective prevention measures according to the individual factors of patients.
