Background: Although critically ill pregnant women received specialized management in the Obstetric Critical Care Unit (OCCU), the disease spectrum and prognostic determinants still need to be further explored to optimize the early warning system. This study aims to systematically analyze the disease spectrum and identify risk factors for unfavorable prognosis in critically ill obstetric patients transferred to the OCCU.

Methods: A retrospective cohort study was conducted on 313 obstetric patients admitted to OCCU from 2015 to 2025. According to the prognosis (favorable, n = 222; unfavorable, n = 91), the patients were divided into two groups. The variables were screened using Least Absolute Shrinkage and Selection Operator (LASSO) regression and the independent risk factors were analyzed using multivariate logistic regression.

Results: Main transfer causes were postpartum hemorrhage (34.82%) and preeclampsia/eclampsia (21.73%). Independent risk factors for unfavorable prognosis included older age, lower gestational age, pre-existing heart disease, diabetes, hypertension, placental abruption, higher Acute Physiology and Chronic Health Evaluation II (APACHE II)/Sequential Organ Failure Assessment (SOFA) scores, and elevated D-dimer, Scr, and C-reactive Protein (CRP) levels (all p < 0.05). The integrated prediction model achieved an area under the curve (AUC) of 0.866 (95% CI: 0.819–0.913).

Conclusion: Specific comorbidities, disease severity scores and biomarkers are key prognostic factors. The model shows strong early warning and predictive capabilities.