Effects of Job Crafting and Leisure Crafting on Nurses' Burnout: A Machine Learning-Based Prediction Analysis

Area of Science:

• Nursing
• Occupational Health
• Psychology

Background:

• Nurse burnout significantly impacts job performance, healthcare quality, and team cohesion.
• Few studies have simultaneously examined job crafting and leisure crafting in relation to nurse burnout.
• Understanding the differential effects of these crafting behaviors is crucial for developing targeted interventions.

Purpose of the Study:

• To assess the prevalence of job crafting, leisure crafting, and burnout among nurses.
• To analyze the influence of job crafting and leisure crafting on nurse burnout using machine learning models.

Main Methods:

• A multicentre cross-sectional survey involving 1235 nurses from four Chinese tertiary hospitals.
• Data collected using the Maslach Burnout Inventory-General Survey, Job Crafting Scale, and Leisure Crafting Scale.
• Analysis performed using logistic regression, support vector machine, random forest, and gradient boosting tree algorithms.

Main Results:

• Nurses reported mild to moderate burnout and moderate to high levels of job and leisure crafting.
• Machine learning models demonstrated good predictive power for burnout (AUC 0.809–0.821), with gradient boosting tree performing best.
• Job crafting emerged as the strongest predictor of burnout, followed by leisure crafting.

Conclusions:

• Job crafting and leisure crafting are beneficial preventative strategies for reducing nurse burnout.
• Nurse managers should promote opportunities for both job and leisure crafting to support nurse well-being.
• Implementing these crafting strategies can help mitigate burnout and enhance the nursing workforce.