Document resource
Objective This study aimed to develop and evaluate a job-exposure matrix (JEM) specific to healthcare workers and to compare estimates of known exposure-outcome associations using two exposure methods: the developed JEM and self-reported data.Material and Methods JEM SOIGNANCES was constructed using data from healthcare workers within the CONSTANCES cohort, according to job titles and sectors of activity, across 24 exposures covering organizational, psychosocial, physical, chemical and biological factors. The JEM was developed in a training dataset and its internal performance was assessed in a validation sample using AUC and Cohen’s Kappa. Association with relevant health outcomes (like pain, depressive symptoms, hypertension) were evaluated using logistic regression and compared to estimates obtained from self-reported exposure data as a reference.Results Of the 24 included exposures, 15 demonstrated satisfactory internal performance, with fair-to-moderate agreement with self-reported data, such as physical effort at work, ionizing radiation exposure, carrying heavy loads, shift work and formaldehyde exposure. The remaining exposures showed poor-to-moderate discriminatory power and poor agreement. Comparing association estimates, 10/16 associations based on JEM organizational exposures aligned with estimates based on self-reported data in direction and magnitude. For biomechanical exposures, 11/15 associations showed overlapping confidence intervals between the two methods. Other exposures generally underestimated associations.Conclusion JEM Soignances provided good internal performance, with good agreement with self-reported data for organizational and biomechanical exposures but not for psychosocial exposure. Further studies are needed to test its usage for research, though information tools could be developed based on it.