Predicting Burnout and Emotional Exhaustion in South African Nurses Using Machine Learning
Author Information
Author(s): Van Zyl-Cillié Maria Magdalena, Bührmann Jacoba H., Blignaut Alwiena J., Demirtas Derya, Coetzee Siedine K.
Primary Institution: North-West University
Hypothesis
Can machine learning models accurately predict the risk factors causing burnout and emotional exhaustion among nursing staff in South Africa?
Conclusion
Machine learning models can accurately predict feelings of burnout and emotional exhaustion among nurses in South Africa using full survey data.
Supporting Evidence
- The gradient booster classifier model had the highest accuracy score for predicting burnout and emotional exhaustion.
- Fatigue was identified as the strongest predictor of burnout and emotional exhaustion.
- Models using full survey data outperformed those using only demographic data.
Takeaway
This study used computers to find out what makes nurses feel really tired and stressed at work, helping to figure out how to make their jobs better.
Methodology
The study used supervised machine learning models developed from survey responses to identify predictive factors for burnout and emotional exhaustion.
Potential Biases
Potential bias may arise from self-reported data and the specific context of the healthcare environment during the pandemic.
Limitations
The study's data collection coincided with varying levels of the COVID-19 pandemic, which may affect the generalizability of the findings.
Participant Demographics
Nurses from medical-surgical units across South Africa, including both registered and enrolled nurses.
Statistical Information
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website