Development and internal validation of an early inpatient risk score for in-hospital mortality in acute pancreatitis: A retrospective cohort study

Introduction: Early identification of patients at high risk of mortality remains a clinical challenge in acute pancreatitis. Existing prognostic tools are often complex or require repeated assessments, limiting their routine use. We aimed to develop and internally validate an early admission-phase risk score for predicting in-hospital mortality in patients with acute pancreatitis. Methods: We conducted a retrospective cohort study of consecutive adult patients admitted with acute pancreatitis between 2020 and 2024. Acute pancreatitis was defined according to the revised Atlanta criteria. Multivariable logistic regression was used to develop a prediction model for in-hospital mortality using routinely available demographic, clinical, and laboratory variables obtained within the first 24 hours of hospitalization. Missing data were handled using multiple imputation. Model performance was assessed by discrimination and calibration, with internal validation performed using bootstrap resampling and temporal validation conducted in patients admitted during 2023–2024. Results: A total of 1,041 patients were included, of whom 53 (5.1%) died during hospitalization. The final model incorporated age, early multi-organ dysfunction within 24 hours of admission, C-reactive protein, and urea levels. The model showed good discrimination (area under the receiver operating characteristic curve, 0.81) and good agreement between predicted and observed in-hospital mortality probabilities. Bootstrap internal validation showed minimal optimism, and temporal validation in patients admitted during 2023–2024 confirmed stable model performance. A simplified risk score derived from the model stratified patients into low-to-intermediate-risk and high-risk categories, with substantially higher observed in-hospital mortality in the high-risk group. Conclusions: We developed and internally validated an admission-phase risk score for predicting in-hospital mortality in patients with acute pancreatitis using readily available clinical variables. The proposed score may support early inpatient risk stratification within the first 24 hours of hospitalization. However, external validation in independent cohorts across diverse healthcare settings, patient populations, and pancreatitis etiologies is required before broader clinical application.