Objective To construct a nomogram model for predicting the risk of constipation in adult patients with cardiovascular disease (CVD).

Methods Adult patients with CVD admitted to Zhongshan Hospital, Fudan University from June 15, 2025 to August 15, 2025 were prospectively enrolled as the training set. Data from 290 adult patients with CVD collected between December 1, 2025, and February 10, 2026, from four tertiary hospitals across China were used as the external validation set. Risk factors for constipation were identified using the least absolute shrinkage and selection operator (LASSO) and Firth logistic regression, and a nomogram model was subsequently constructed. Receiver operating characteristic (ROC) curve and the area under the curve (AUC) were used to evaluate the discrimination ability of the model. Calibration curves and decision curve analysis (DCA) were used to assess calibration and clinical applicability. External validation was performed using the validation dataset.

Results A total of 780 adult patients with CVD were included in the training set, among whom 391 patients (50.1%) developed constipation. Eleven variables were selected by LASSO regression and Firth logistic regression to construct the nomogram model. The ROC curve showed that the AUC for predicting constipation in adult patients with CVD was 0.881. Calibration curves and DCA demonstrated good calibration and predictive performance of the model. A total of 290 patients were included in the external validation set, among whom 144 developed constipation. The AUC of the external validation set was 0.865, and both the calibration curves and DCA indicated satisfactory model performance.

Conclusions The nomogram model based on LASSO-Firth logistic regression demonstrated good predictive performance and may serve as a useful tool for identifying adult patients with CVD at high risk of constipation.