Objective

2

By exploring the influencing factors of readmission in patients with stable angina of coronary heart disease （CHD） based on real-world clinical data， to establish a risk prediction model of integrated traditional Chinese and western medicine， in order to provide a basis for early identification of high-risk populations and reducing readmission rates.

Methods

2

A prospective clinical study was conducted involving patients with stable angina pectoris of CHD， who were divided into a training set and a validation set at a 7∶3 ratio. General information， traditional Chinese medicine （TCM）-related data， and laboratory test results were uniformly collected. After a one-year follow-up， patients were classified into a readmission group and a non-readmission group based on whether they were readmitted. Univariate and multivariate logistic regression analyses were performed to identify independent risk factors for readmission. A risk prediction model of integrated traditional Chinese and western medicine was constructed and visualized using a nomogram. The model was validated and evaluated in terms of discrimination， calibration， and clinical decision curve analysis.

Results

2

A total of 682 patients were included， with 477 in the training set and 205 in the validation set， among whom 89 patients were readmitted. Multivariate logistic regression analysis identified heart failure history ［

OR

= 6.93， 95%

CI

（1.58， 30.45）］， wiry pulse ［

OR

= 2.58， 95%

CI

（1.42， 4.72）］， weak pulse ［

OR

= 3.97， 95%

CI

（2.06， 7.67）］， teeth-marked tongue ［

OR

= 4.38， 95%

CI

（2.32， 8.27）］， blood stasis constitution ［

OR

= 2.17， 95%

CI

（1.06， 4.44）］， phlegm-stasis mutual syndrome ［

OR

= 3.64， 95%

CI

（1.87， 7.09）］， and elevated non-high-density lipoprotein cholesterol ［

OR

= 1.30， 95%

CI

（1.01， 1.69）］ as influencing factors of readmission. These factors were used as predictor

s to construct a nomogram-based risk prediction model for readmission in patients with stable angina. The model demonstrated moderate predictive capability， with an area under the receiver operating characteristic curve （AUC） of 0.818 ［95%

CI

（0.781， 0.852）］ in the training set and 0.816 ［95%

CI

（0.779， 0.850）］ in the validation set. The Hosmer-Lemeshow test showed good calibration （χ² = 4.55，

P

= 0.80）， and the model's predictive ability was stable. When the threshold probability exceeded 5%， the clinical net benefit of using the model to predict readmission risk was significantly higher than intervening in all patients.

Conclusion

2

History of heart failure， teeth-marked tongue， weak pulse， wiry pulse， phlegm-stasis mutual syndrome， blood stasis constitution， and non-high-density lipoprotein cholesterol are influencing factors for readmission in patients with stable angina of CHD. A clinical prediction model was developed based on these factors， which showed good discrimination， calibration， and clinical utility， providing a scientific basis for predicting readmission events in patients with stable angina.