Newly graduated nurses often experience high levels of psychological stress, sleep disturbance, fatigue, and burnout during the early transition into clinical practice. Early identification of burnout and retention risk may help improve mental well-being, workforce stability, and quality of patient care.
This longitudinal observational study aims to develop a non-invasive sleep-based prediction platform for assessing burnout and retention risk among postgraduate nurses. Participants will undergo repeated psychological assessments and non-contact sleep monitoring during the study period. Sleep-related physiological parameters, including sleep efficiency, sleep structure, heart rate variability, and respiratory variability, will be collected together with validated psychological questionnaires.
The study will further apply machine learning and artificial intelligence approaches to integrate longitudinal physiological and psychological data for risk prediction and early identification of burnout-related conditions. The findings may support future development of precision mental health monitoring and supportive management strategies for high-stress healthcare workers.