Clinical nurses’ work-life balance prediction due to patient safety incidents using classification and regression tree analysis: A secondary data analysis

Abstract

Background Patient safety incidents lead to performance difficulties for nurses when providing nursing practice. This affects work-life balance and causes second and third-victimization. This study predicts factors affecting clinical nurses’ work-life balance due to patient safety incidents using classification and regression tree analysis techniques. Methods This study was a secondary analysis of data from a cohort research project, which used a descriptive survey for data collection. Participants comprised 372 nurses. Data were collected using SurveyMonkey, a mobile-based survey software solution, from January to September 2021. Data included the general characteristics of clinical nurses, second damage, second damage support, third damage, and work-life balance. The data were analyzed using Lasso. Variables with low explanatory power were excluded, thereafter, the variables selected by Lasso were analyzed with a classification and regression tree model to predict work-life balance. Results A regression tree was applied to predict work-life balance using seven variables—education level, marital status, position, physical distress, second-victim support, turnover intentions, and absenteeism (selected through Lasso analysis). After pruning, at tree size four, when turnover intentions were < 4.250, physical distress < 2.875, and second-victim support < 2.345, the predicted work-life balance was 3.972. However, when turnover intentions were < 4.250, physical distress < 2.875, and second-victim support ≥ 2.345, then the predicted work-life balance was 2.760. Conclusions This study’s results can be used as fundamental data for formulating workforce risk management strategies, such as managing each nurse’s occupational stress. Ultimately, they can help improve organizational culture to prevent the recurrence of additional patient safety incidents.