Abstract: Objective: To develop a risk prediction model of demoralization syndrome, and to provide a basis for its prevention. Methods: A total of 376 inpatients with cancer were included from the oncology department of a tertiary grade A hospital in Dalian from November 2021 to June 2022 by convenience sampling, multivariate Logistic regression analysis was used to analyze the risk factors of demoralization syndrome, a nomogram was constructed for visual display, and a receiver operating characteristic curve was used to test the prediction effects of the model. Results: The incidence of demoralization syndrome in cancer patients was 44.9%, and a total of 6 independent risk factors were screened for demoralization syndrome by multivariate Logistic regression analysis. The area under the predicted ROC curve was 0.973 (P<0.001). The maximum Youden index was 0.850, the sensitivity was 0.891, the specificity was 0.959, the slope of the drawn calibration curve was close to 1, and the predicted value of the model was in good agreement with the actual observed value consistency. Conclusion: The constructed model in this study is with good differentiation and accuracy, which can provide evidence for medical staff in oncology department to identify patients with demoralization syndrome in an early stage.

Key words: cancer; demoralization syndrome; correlative factors; risk prediction model; nomogram