Machine learning and SHAP interpretation to identify high-level compassion fatigue among operating room nurses: a multicenter cross-sectional study.

backgroundOperating room nurses (ORNs) are at high risk for compassion fatigue (CF), which significantly impairs individuals' well-being, undermines the stability of the nursing workforce, and jeopardizes patients' safety. The study aimed to analyze the prevalence and symptom characteristics of CF among ORNs, construct and compare predictive models using machine learning, and determine the relative contribution of distinct features to the models.

methodsThis is a multi-center cross-sectional study. The questionnaires used in the study included a sociodemographic questionnaire, the Professional Quality of Life Scale (ProQoL), the Patient Health Questionnaire-9 (PHQ-9), the Generalized Anxiety Disorder 7-item Scale (GAD-7), and the Pittsburgh Sleep Quality Index (PSQI). LASSO regression was used to select critical variables, and predictive models such as decision tree, logistic regression, random forest, SVM, and XGBoost were constructed and compared. SHapley Additive exPlanation (SHAP) were drawn to show the contribution of each feature to the models. SPSS version 26.0 and R software version 4.4.0 were used for statistical analyses.

resultsIn this study, a total of 1024 ORNs from 20 cities across China were recruited. According to ProQoL, 326 (31.8%) reported severe CF, 311 (30.4%) moderate CF, and the remaining 387 (37.8%) no or mild CF. Among the three dimensions, the incidence of secondary traumatic stress was most common (95.4%), followed by low compassion satisfaction (61.3%) and burnout (35.0%). In five machine learning-based predictive models, the RF model stood out with the highest AUC at 0.851 (95%CI: 0.795-0.907) in testing set. Following closely, the XGBoost model showed favorable efficacy with the AUC at 0.824 (95%CI: 0.769-0.879) in testing set, outperforming the remaining algorithms. The results of the two SHAP plots (RF and XGBoost) were consistent: depression, anxiety, self-mental health training, sleep quality, and length of service emerged as the five most significant contributors to the models.

conclusionThis study identified severe CF among ORNs, and the most serious symptom was secondary traumatic stress. The RF model exhibited the best performance in identifying high-level CF among ORNs, and SHAP improved the interpretability of the model. The findings of this study could help medical managers and researchers better understand CF and provide timely interventions for ORNs. CLINICAL TRIAL NUMBER: Not applicable.