Background <p>The prognosis of patients with a concomitance of severe traumatic brain injury (sTBI) and acute respiratory distress syndrome (ARDS) is poor, and early identification of such patients can provide diagnostic and therapeutic assistance for clinical treatment. However, few studies have been conducted to identify the risk of ARDS in patients with sTBI. This study aimed to construct a risk prediction model for ARDS in patients with sTBI and evaluate its efficacy.</p> Methods <p>From 2016 to 2023, 502 patients diagnosed with sTBI were selected from the Affiliated Hospital of Yangzhou University. All participants were randomly allocated to either the training or validation group. Feature selection for constructing the prediction model and developing a nomogram was carried out using the least absolute shrinkage and selection operator (LASSO) and multivariable logistic regression analysis. The effectiveness and clinical relevance of the model were evaluated using receiver operating characteristic (ROC) curves, the area under the ROC curve (AUC), calibration curves, and the decision curve analysis (DCA).</p> Results <p>The study found that 32.9% of patients with sTBI developed ARDS. The model was established based on oxygen saturation measured by pulse oximetry (SpO₂), pneumonia, and fluid volume in the first 24&#xa0;h. The model showed good discriminative ability with AUC values of 0.841 for the training and 0.821 for the validation groups. Calibration curves demonstrated that the predicted results align well with the actual results. The DCA suggested that the nomogram could lead to clinically beneficial outcomes at a significant threshold.</p> Conclusions <p>The diagnostic nomogram for ARDS in sTBI patients demonstrated satisfactory predictive value, assisting clinicians in identifying high-risk patients for ARDS.</p> Trail registration <p>ChiCTR2400085916.</p>