Abstract: Objective To develop a dynamic nomogram prediction model for hypoactive delirium risk in ICU patients and verify its predicted effect. Methods A total of 430 patients were selected from the ICU of a tertiary Grade A hospital in Chongqing from October 2022 to November 2023. The Confusion Assessment Methodfor Intensive Care Unit and Richmond Agitation-Sedation Scale were used for delirium assessment and classification. Independent predictors were determined through univariate analysis and multivariate logistic regression analysis,and a dynamic nomogram prediction model was developed. The area under the receiver operating characteristic curve and Hosmer-Lemeshow goodness of fit were used to test the discrimination and calibration of the model,and the internal and external verification of the model was conducted. Results The predictors were electrolyte disorders（OR=2.350）,indwelling catheters≥2（OR=3.529）,use of diuretics（OR=0.342）,Glasgow Coma Scale score（OR=0.183）,concentration of C-reactive protein（OR=1.006）,and urea concentration（OR=1.063）. In modeling group,the AUC was 0.941（95%CI：0.916~0.966）,the sensitivity was 87.9%,the specificity was 90.3%,and the Hosmer-Lemeshow test was P=0.415. In validation group,the AUC was 0.897（95%CI：0.837~0.956）,the sensitivity was 79.6%,the specificity was 86.5%,and the Hosmer-Lemeshow test result showed that P=0.450. Conclusion The dynamic nomogram prediction model constructed in this study can effectively predict the probability of hypoactive delirium in ICU patients,providing an effective tool for medical staff to scientifically predict the occurrence of hypoactive delirium,and is convenient for clinical use.

Key words: Intensive Care Units, Hypoactive Delirium, Risk Factors, Nomogram, Nursing Care