Abstract
BACKGROUND: Brain tumor patients experience high rates of symptom cluster distress (SCD), severely impacting quality of life and recovery. Nurse-led predictive tools are lacking. OBJECTIVE: To explore the determinants of SCD in brain tumor patients, analyze the risk indicators of SCD, construct a nomogram model, and validate the model. METHODS: Using a convenience sampling method, 300 brain tumor patients admitted to the Department of Neurosurgery at a tertiary-grade A cancer hospital from April 2023 to January 2025 were selected as the study subjects. Univariate and multivariate logistic regression analyses were performed to identify risk factors for SCD. Patients were categorized into a low symptom distress (SD) group (SD < 4 scores) and a high SD group (SD ≥ 4 scores) based on the presence of symptom-related distress. Multivariate logistic regression analysis was employed to investigate the factors associated with SCD in brain tumor patients. The risk prediction nomogram model for SCD in these patients was developed using R software (version 4.3.1) with the rms package. The prediction effect and degree of fit of the nomogram model were evaluated via receiver operating characteristic (ROC) curves, calibration curves, and the Hosmer-Lomoshow goodness-of-fit test. RESULTS: A prediction model was constructed based on eight influencing factors-age, payment method, disease duration, Karnofsky Performance Status (KPS), financial toxicity (COST), self-efficacy (GSES), and medical coping mode (MCMQ). The model's area under the ROC curve (AUC) was 0.813 (95% confidence interval [CI]: 0.765-0.862, P < 0.05), with a sensitivity of 0.717, specificity of 0.791, optimal cutoff value of 0.750, and Youden index of 0.508. The calibration curve analysis results indicated that the calibration curve of the column graph model for predicting SCD in brain tumor patients was a straight line with a slope close to 1. According to the Hosmer-Lemeshow goodness-of-fit test, the incidence of SCD in brain tumor patients predicted by the pillar-line model did not differ significantly from the actual incidence (P = 0.061). CONCLUSION: The model constructed in this study has a good prediction effect, which can provide a reference for medical staff to quickly identify the risk of SCD in brain tumor patients and take timely preventive management measures, thereby improving the effectiveness of symptom management.