Yian Zhan, Ying Li, Shan Yang
Based on the stress process theory, this study found that 91.86% of midwives experienced compassion fatigue at any level; 63.6% had moderate to severe levels-the threshold considered clinically significant for the purpose of the prediction model. Six independent predictors were identified: age, sleep quality, perceived recognition in the past month, frequency of traumatic events, and positive/negative coping. A nomogram prediction model was developed using these predictors. The model demonstrated good discriminative ability (AUC = 0.820, 95% CI: 0.763-0.876) and calibration (Hosmer-Lemeshow p = 0.672). This nomogram provides nursing managers with a practical, low-cost tool to identify midwives at high risk of compassion fatigue, providing a reference for targeted preventive strategies.
BACKGROUND: Compassion fatigue among midwives is prevalent and poses significant risks, adversely affecting both their personal well-being and the quality of maternal and infant care. However, a tailored, practical tool for predicting compassion fatigue risk in this population is lacking.
OBJECTIVE: To analyze factors influencing compassion fatigue in midwives, construct a predictive model for compassion fatigue risk, and provide evidence-based strategies for prevention.
METHODS: In this cross-sectional study, 209 midwives from 29 public hospitals in western China were enrolled via convenience sampling. Compassion fatigue was assessed using the Professional Quality of Life Scale, alongside measures of the Traumatic Stress Scale for Midwives, Social Support Rating Scale, 10-Item Connor-Davidson Resilience Scale, and Simplified Coping Style Questionnaire. Univariate analysis and multivariate logistic regression identified risk factors for compassion fatigue. A risk prediction model was constructed and a nomogram model was developed, followed by model evaluation and internal validation.
RESULTS: Six independent predictors were identified: age, sleep quality, perceived recognition in the past month, frequency of traumatic events, and positive/negative coping strategies. The model achieved an area under the receiver operating characteristic curve (area under the curve [AUC]) of 0.820 (95% confidence interval [CI]: 0.763-0.876). After internal validation via bootstrapping, the AUC decreased to 0.786 (95% CI: 0.779-0.815). The Hosmer-Lemeshow test yielded χ2 = 5.777, p = 0.672.
CONCLUSION: Based on the stress process theory, this study found that 91.86% of midwives experienced compassion fatigue at any level; 63.6% had moderate to severe levels-the threshold considered clinically significant for the purpose of the prediction model. Six independent predictors were identified: age, sleep quality, perceived recognition in the past month, frequency of traumatic events, and positive/negative coping. A nomogram prediction model was developed using these predictors. The model demonstrated good discriminative ability (AUC = 0.820, 95% CI: 0.763-0.876) and calibration (Hosmer-Lemeshow p = 0.672). This nomogram provides nursing managers with a practical, low-cost tool to identify midwives at high risk of compassion fatigue, providing a reference for targeted preventive strategies.
IMPLICATIONS FOR NURSING MANAGEMENT: This nomogram may assist managers in early identification of midwives at risk for compassion fatigue and facilitate risk-stratified preventive strategies, with external validation needed before broader clinical implementation. The findings highlight the actionable areas for intervention-including psychological support, trauma-informed care, sleep and workload management, and professional recognition practices-which may help mitigate compassion fatigue and promote workforce stability.