Associations between fatigue, burnout, and risk behaviors in mental health occupational therapists: a cross-sectional study of coping and substance use

BACKGROUND: Occupational therapists working in mental health services face sustained emotional demands, high workload, and clinical complexity, which may increase burnout risk. While Maslach's three-dimensional model remains the gold standard for conceptualization, a gap persists in integrating physiological factors such as fatigue and behavioral risk outcomes within a theoretically derived framework in occupational therapy. OBJECTIVE: To analyze the associations between fatigue and burnout among mental health occupational therapists in Spain, and to examine the relationships among emotional exhaustion, depersonalization, and substance use within a theoretically derived model. METHODS: A cross-sectional observational study was conducted with a nationwide sample of 663 occupational therapists. Burnout was assessed using the Maslach Burnout Inventory, fatigue with the Chalder Fatigue Scale, and coping strategies with the COPE-28. Correlational analyses, hierarchical regressions, and structural equation modeling were performed. RESULTS: Fatigue was significantly associated with higher emotional exhaustion and depersonalization, and lower personal accomplishment. Sleep and concentration problems were strongly associated with emotional exhaustion. The structural model revealed associations consistent with a theoretically derived sequence linking fatigue, emotional exhaustion, depersonalization, and substance use. Problem-focused coping was associated with lower depersonalization without reducing baseline fatigue. CONCLUSIONS: Burnout in mental health occupational therapists may be understood within a theoretically derived framework in which fatigue is associated with emotional exhaustion and, in turn, with depersonalization and substance use. These findings provide preliminary evidence consistent with a theoretically derived sequence of associations, although causal inferences cannot be established due to the cross-sectional design. CLINICAL TRIAL NUMBER: Not applicable.