Using Explainable Supervised Machine Learning to Predict Burnout in Healthcare Professionals.

Area of Science:

• Occupational Health
• Health Informatics
• Psychology

Background:

• Healthcare professional (HCP) burnout is a significant issue with complex causes.
• The COVID-19 pandemic exacerbated burnout among HCPs.
• Limited research has applied machine learning (ML) to predict HCP burnout during the pandemic.

Purpose of the Study:

• To develop and evaluate ML models for predicting HCP burnout.
• To identify key predictors of burnout in HCPs during the COVID-19 pandemic.

Main Methods:

• A survey was administered to 450 HCPs during the COVID-19 pandemic.
• Survey data included demographic characteristics and work system factors.
• Machine learning models were utilized to predict burnout, with the best model evaluated using the area under the receiver operating curve (AUC).

Main Results:

• The highest performing ML model achieved an AUC of 0.81.
• Eight key features significantly predicted burnout: excessive workload, inadequate staffing, administrative burden, professional relationships, organizational culture, values and expectations, intrinsic motivation, and work-life integration.

Conclusions:

• Machine learning can effectively predict HCP burnout.
• Understanding key burnout predictors is crucial for targeted interventions.
• Findings support evidence-based resource allocation to mitigate burnout and enhance care quality.