Related Experiment Videos
Development and Internal Validation of Interpretable Machine Learning Models for Identifying Burnout Syndrome Among
Wei Hu1, Yunfan Ji2, Fengzhi Chai2
1Department of Nursing, First Affiliated Hospital of Jinzhou Medical University, Jinzhou, Liaoning, China, jzmu.edu.cn.
Journal of Nursing Management
|July 20, 2026
Summary
Intensive care unit (ICU) nurse burnout is predicted by psychological resilience and job satisfaction, alongside nursing stress and shift work. Machine learning models can identify these factors to prevent nurse burnout.
Area of Science:
- Nursing
- Healthcare Management
- Artificial Intelligence
Background:
- Nurse burnout poses risks to patient safety and healthcare quality.
- Identifying burnout correlates is crucial for intervention development.
Purpose of the Study:
- Develop and validate a machine learning model to identify burnout in intensive care unit (ICU) nurses.
- Determine key factors associated with ICU nurse burnout.
Main Methods:
- Surveyed 318 ICU nurses in China, collecting data on 34 potential predictors.
- Utilized machine learning algorithms, including random forest, with LASSO regression for predictor selection.
- Validated the model using AUC, calibration, and SHAP analysis.
Main Results:
- A nine-predictor random forest model achieved an AUC of 0.983.
- Psychological resilience and job satisfaction were key protective factors.
- Nursing stress, night shift frequency, and poor sleep quality were significant risk factors.
Conclusions:
- A validated machine learning model can identify ICU nurse burnout.
- Interventions should focus on organizational factors like staffing and support for resilience.
- Further validation in diverse settings is recommended.
