Burnout protective patterns among oncology nurses: a cross-sectional study using machine learning analysis

Oncology nurses face unique and intense demands due to the nature of their work, caring for patients with life-threatening illnesses. The emergence of professional burnout among these nurses is influenced by several factors, highlighting the importance of identifying protective and risk factors to mitigate its impact. This study aims to identify burnout profiles and protective socio-demographic and work-related patterns associated with reduced burnout among oncology nurses.

A cross-sectional study was conducted with 150 oncology nurses at a specialized hospital exclusively dedicated to adult oncology treatment in Portugal. Data collection included a self-administered questionnaire incorporating the validated Portuguese version of Maslach Burnout Inventory (MBI). Statistical analyses were performed using SPSS and machine learning tools, specifically KMeans clustering and Random Forest algorithms.

Six protective patterns against burnout were identified, characterized by conditions of permanent contracts, work-life balance, and supportive work environments. Moreover, factors such as holding management roles and being a parent of two or more children might even be protective in some circumstances, suggesting a nuanced relation between personal and professional factors. Machine learning analyses made apparent the unpredictability of burnout and highlighted the critical role of protective factors in mitigating its impact.

This study underscores the importance of resilience-building strategies and promoting protective factors, such as job stability, learned experience, and adequate rest, to reduce burnout risk among oncology nurses. Future research should validate these findings through hypothesis-driven analyses to inform targeted and context-specific burnout prevention programs.

Not applicable.