Abstract
BACKGROUND: The primary objective of this study is to identify independent risk factors associated with secondary pulmonary hypertension (PH) in silicosis patients through retrospective statistical analysis and to develop a clinically applicable short-term predictive model. METHODS: We retrospectively analyzed the clinical data of 199 silicosis patients, among whom 98 had PH and 101 did not. PH was defined as an estimated systolic pulmonary artery pressure (sPAP) > 40 mmHg. A short-term clinical prediction model for PH in hospitalized silicosis patients was constructed by identifying independent risk factors through logistic regression analysis. RESULTS: Logistic regression analysis identified the modified Medical Research Council (mMRC) score, silicosis stage, forced vital capacity (FVC), oxygenation index, and eosinophil ratio as independent risk factors for the development of pulmonary hypertension in silicosis patients. A clinical prediction model for male patients was constructed using binary multifactorial logistic regression incorporating these factors: P = e(x)/(e(x)+1), where X=-14.958 + 0.943 ⋅ silicosis stage + 0.479 ⋅ mMRC score + 0.471 ⋅ eosinophil ratio + 2.549 ⋅ oxygenation index + 0.863 ⋅ FVC. The Hosmer-Lemeshow test indicated a good fit for this model (P = 0.231). The plotted receiver operating characteristic curve yielded an area under the curve of 0.908, accompanied by a Youden’s index of 0.704, a specificity of 87.9%, and a sensitivity of 82.5%. CONCLUSIONS: The model demonstrates strong predictive performance for predicting the risk of PH in male patients with silicosis. In clinical practice, paying attention to key factors—including mMRC, silicosis stage, FVC, oxygenation index, and eosinophil ratio—along with the implementation of targeted interventions for high-risk patients, is essential to improving quality of life and long-term prognosis. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12890-026-04242-x.