乙型肝炎肝硬化病人抑郁障碍预测模型的构建OA

Objective:To analyze the risk factors for depression in patients with hepatitis B-related cirrhosis(HBC),to establish a decision tree prediction model,and to formulate nursing interventions to reduce the risk of depression in HBC patients.Methods:A total of 200 HBC patients admitted to our hospital from January 2023 to January 2025 were selected as study subjects.Logistic regression analysis was used to identify risk factors influencing the occurrence of depression in HBC patients.A decision tree model for depression occurrence was constructed using the Chi-squared automatic interaction detection(CHAID)algorithm.The predictive performance of the decision tree model was evaluated by plotting the receiver operating characteristic(ROC)curve and calculating the area under the curve(AUC).Results:According to the Hamilton Depression Scale(HAMD)scores,all patients were divided into the depression disorder group with 58 cases and the non-depression disorder group with 142 cases.Logistic regression analysis showed that the education level,family support,psychological resilience,disease duration,and Child-Pugh classification were all risk factors for depression in HBC patients(P<0.05).The decision tree model identified 5 explanatory variables,namely Child-Pugh classification,psychological resilience,family care level,disease duration,and educational attainment.Among them,the Child-Pugh classification was the most significant influencing factor for the occurrence of depressive disorders in HBC patients.The AUC of the decision tree model was 0.930.After ten-fold cross-validation,the average AUC of the model was 0.914,which showed a relatively small difference from the original model's AUC.There was no significant overfitting in the model.The generalization ability was good.The Hosmer-Lemeshow test results showed that there was no statistically significant difference between the predicted risk of depression disorder by the model and the actual disease status of the patients.The calibration was good.Using HAMD≥17 as the diagnostic threshold for depression,the model showed a sensitivity of 87.93%,specificity of 91.55%,and Youden's index of 0.795.Conclusions:Education level,family support,psychological resilience,disease duration,and Child-Pugh classification are important risk factors for depression in HBC patients.The decision tree model based on these factors demonstrates high predictive value.Clinical practice should develop targeted nursing interventions addressing the above risk factors to reduce the incidence of depression in HBC patients.