Objective

2

To analyze the predisposed traditional Chinese medicine （TCM） constitution and risk factors of diabetic macrovascular complications， and to construct a clinical risk prediction model.

Methods

2

A total of 275 hospitalized diabetic patients admitted to Wuyi Hospital of Traditional Chinese Medicine， Jiangmen City， affiliated with Jinan University， from December 2023 to June 2024 were included， of whom 173 patients with imaging-confirmed macrovascular disease were assigned to the observation group and 102 patients without macrovascular disease to the control group. TCM constitution types were assessed using a standardized questionnaire. Binary logistic regression analyses were performed to identify influencing factors and independent risk factors for diabetic macrovascular complications. Data were imported into R4.3.2 software， and a nomogram prediction model was constructed using the rms package based on the identified risk factors. Model performance was evaluated using calibration curves， receiver operating characteristic （ROC） curves and area under the curve （AUC）， goodness-of-fit test， and decision curve analysis （DCA）.

Results

2

Univariate and multivariate logistic regression analyses indicated that

qi

deficiency constitution was the predisposition for diabetic macrovascular complications （

P

<

0.05）. Gender， age， history of hyperuricemia， and history of hypertension were independent risk factors （

P

<

0.05）. The prediction model based on multivariate analysis was expressed as： Logit

P

=age×0.093+history of hypertension×1.033+history of hyperuricemia×0.958+gender×0.989+

qi

deficiency constitution×0.926－6.741. Incorporating the predisposed constitution and independent risk factors into R software， a nomogram for diabetic macrovascular complications was constructed. The ROC curve analysis showed that the AUC of this prediction model for predicting the incidence of diabetic macrovascular complications was 0.82（95%

CI

：0.764-0.871）， indicating good discrimination. The calibration curve closely aligned with the ideal curve， suggesting good agreement between predicted and observed outcomes. The Hosmer-Lemeshow test showed χ

2

=8.389，

P

=0.396， indicating good model fit. Decision curve analysis demonstrated clinical utility， with better performance than any s

ingle risk factor.

Conclusion

2

Qi

deficiency constitution is closely associated with diabetic macrovascular complications. The nomogram based on

qi

deficiency， gender， age， history of hyperuricemia， and history of hypertension provides valuable clinical reference for predicting the risk of diabetic macrovascular complications.