Abstract:

Objective To analyze the risk factors of delirium in patients with malignant tumors and construct a nomogram prediction model. Methods A total of 312 patients with malignant tumors admitted to the Department of Oncology in Shengjing Hospital of China Medical University from January 2019 to December 2020 were selected. The patients were divided into non⁃delirium group and delirium group according to the DMS⁃ 5 diagnostic criteria. The clinical data of the two groups were analyzed and compared. The independent risk factors for delirium were determined by multivariate logistic regression analysis，which were further utilized to establish the nomogram prediction model of the risk of delirium in patients with malignant tumor. The calibration curve and ROC curve were used to evaluate the accuracy and discrimination of the nomogram prediction model. Results The incidence of delirium was 14.42%（45/312）. Multivariate logistic regression analysis showed age，intracranial lesions，anxiety，pain，and Pittsburgh sleep quality index were independent risk factors for delirium in patients with malignant tumors（P < 0.05）. The prediction nomogram model was constructed based on the independent risk factors mentioned above. In internal validation，calibration curve showed that the prediction curve was basically consistent with the actual observation curve，and ROC curve analysis showed that the area under the curve AUC was 0.919. Conclusion This study constructed a nomogram model to predict the risk of delirium in patients with malignant tumors. It could make early prediction of the risk of delirium in patients with malignant tumors，and help clinical medical staff to early screen the high⁃risk groups of delirium among patients with malignant tumors.

Key words:

malignant tumor, delirium, nomogram