# A Nomogram for Predicting Pulmonary Embolism in Silicosis Patients

## TL;DR

This study creates a prediction model to help doctors identify the risk of pulmonary embolism in patients with silicosis, an occupational lung disease.

## Contribution

A novel nomogram was developed and validated for predicting pulmonary embolism in silicosis patients.

## Key findings

- The nomogram was built using variables like mMRC, pectoralgia, and history of VTE.
- Internal validation showed the model can assist clinicians in decision-making for patient management.
- The model's performance was compared to the Padua score for pulmonary thromboembolism prediction.

## Abstract

As one of the most severe occupational diseases that prevention efforts have supported for several decades, silicosis is still a public health issue that lacks a prediction model for pulmonary embolism.

A total of 162 patients confirmed to have silicosis were all involved in a training cohort to construct a nomogram with the outcome diagnosed by the CTPA using logistic regression. Univariate and LASSO analyses were used to select variables for the nomogram.

mMRC, pectoralgia, history of VTE, active tumor, unilateral lower limb pain or edema, hormonotherapy, reduced mobility, and heart failure/respiratory failure were selected for the establishment of the nomogram for silicosis with pulmonary embolism.

A novel nomogram was developed to predict pulmonary embolism in silicosis patients. The internal validation indicated that clinicians could utilize this predictive model to help decision‐making and patient management.

This is a data‐driven and evidence‐based study to better understand the risk factors for silicosis patients with pulmonary thromboembolism. Nomogram‐based predictive models were developed and validated to forecast pulmonary thromboembolism probability for silicosis patients. The prediction effect of the Padua score on pulmonary thromboembolism was compared with that of the model established in this study