Objective To explore the influencing factors of postoperative hypoactive delirium（HD） in severe patients and establish a risk prediction model. Methods This study was a prospective analysis to facilitate 720 patients in the surgical intensive care unit (SICU) from October 2022 to July 2023 and January 2024 to July 2024 through hospital electronic medical record system, and intensive care confusion assessment (CAM-ICU) and Richmond agitation-sedation scale (RASS) to classify the modeling group into HD group (117 patients) and non-HD group (387 patients). General data, operation-related data and clinical data of ICU were collected, and the incidence of postoperative HD was studied and the univariate and multivariate regression analysis was conducted to explore the independent risk factors of postoperative HD in severe patients, establish a prediction model and conduct internal validation. The receiver operating characteristic (ROC) curve was drawn, and the predictive value of the predictive model for developing postoperative HD in severe patients was evaluated. Results The incidence of postoperative HD in severe patients was 23.2%, the highest proportion of all types of delirium (51.5%). Univariate and multivariate analysis of demographic and clinical characteristics showed that age, hypertension, diabetes history, surgery duration, mechanical ventilation, interleukin (IL) -6, acute physiology and chronic health score (APACHE Ⅱ score), muscle strength, daily living ability (ADL) and ICU stay length (all P 0.05); the other differences were not significant (all P0.05). Age [Ratio ratio (OR) 1.063, 95% confidence interval (CI) 1.038~1.088], operative time [OR 1.255, 95%CI 1.090~1.445], mechanical ventilation [OR 1.665, 95%CI 1.041~2.662], IL-6[OR 1001, 95%CI 1.000~1.002], muscle strength [OR 0.954, 95%CI 0.931~0.978] and ADL [OR 0.965, 95% CI 0.941~0.989] were independent risk factors for postoperative HD in severe patients (P 0.05). The ROC curve analysis showed that at the cut-off value was 0.239, the Youden index was the largest, the sensitivity and specificity were 72.1% and 73.0% respectively, and the area under the curve (AUC) predicting postoperative HD in severe patients was 0.791. Conclusions Age, diabetes, surgery duration, IL-6, muscle strength and ADL were independent risk factors for postoperative HD in postoperative patients. Using a prediction model can predict postoperative HD and help its prevention and diagnosis.