Résumé
Context: Since March 2020, healthcare workers have been under continued stress due to the COVID-19 pandemic. Biological variables have been largely neglected in studies exploring burnout and psychopathology, such as depression, anxiety, and posttraumatic stress disorder among healthcare workers dealing with COVID-19. Methods: Of 467 Quebec, Canada healthcare workers who responded to our electronic survey three months after the beginning of the pandemic, 372 (which corresponds to 80%) furnished a sample of their hair. The survey included questionnaires on mental health outcomes and associated factors. The length of the hair sample was 3 to 6 cm, providing an estimation of the cortisol, a stress hormone, secreted in the three months preceding the pandemic and the three months following its onset. We applied a multivariable regression model and a receiver operating characteristic curve (ROC curve) to examine hair cortisol as an early indicator of burnout and mental health, combined with individual, occupational, social, and organizational factors. Results: As we anticipated, hair cortisol levels soared after the beginning of the pandemic (relative change: +38% [SD = 67%]). Burnout symptoms and cortisol were significantly associated, healthcare workers in the second quarter experiencing lower burnout odds. There was no association between cortisol change and depression, anxiety, and posttraumatic stress disorder symptoms. Introducing cortisol in our burnout model with individual-occupational-socio-organizational factors strengthens its predictability pronouncedly. Conclusion: A modest relative change in hair cortisol levels (second quarter of change) forecasted burnout in this population. A non-invasive biological indicator of stress exposure, hair cortisol levels, could profit from additional clinical or research efforts. The final aim would be to prevent and control burnout in health personnel facing a significant stress factor.