Skip to Main content Skip to Navigation

Matrices Emplois-Expositions : Exemples de création, validation et application

Abstract : In epidemiology of occupational risks, exposure assessment is commonly based on answers to a questionnaire, in particular to explore exposure to factors such as psychosocial and biomechanical work factors. This method of measure can suffer from bias: classification bias (including memory bias), bias in estimated associations with health variables (especially when exposure and disease are collected simultaneously). Job-exposure matrix (JEM) could be an alternative and a relevant exposure assessment method.Three main objectives for this thesis project were: 1) to create and validate a JEM ‘STRESSJEM’ using the data from the national SUMER surveys (waves of 2003 and 2010) for the assessment of a large panel of psychosocial work factors. The Classification And Regression Trees (CART) segmentation method was used; 2) to study the validity of biomechanical exposure scores based on the JEM ‘JEM Constances’ compared to self-reported assessment in the CONSTANCES cohort; 3) to apply the JEM Constances to perform etiological analyzes on the relationship between exposure to biomechanical factors during lifetime work and limitations in climbing stairs from the CONSTANCES cohort.The results of the first part of the thesis allowed to conclude that the JEM ‘STRESSJEM’ could constitute a complementary method for the evaluation of decisional latitude and shift work, or even other dimensions of the Karasek questionnaire, excluding psychological demand. In the second objective, we concluded that a cumulative exposure score based on the JEM Constances was adequate for assessing exposure to carrying heavy loads compared to self-reported data. The validity was rather moderate repetitive work and physically intense work. Etiological analysis has shown that cumulative work exposure to carrying heavy loads or kneeling across working life was associated with functional limitations in climbing stairs, especially among the less educated.The results presented in this thesis contribute to a better knowledge about what can be expected from JEMs, concerning both possibilities and limitations of this kind of approach, with research perspectives and applications.
Document type :
Complete list of metadata

Cited literature [140 references]  Display  Hide  Download
Contributor : ABES STAR :  Contact
Submitted on : Friday, September 4, 2020 - 3:48:11 PM
Last modification on : Sunday, June 26, 2022 - 9:21:37 AM
Long-term archiving on: : Friday, December 4, 2020 - 5:42:19 PM


Version validated by the jury (STAR)


  • HAL Id : tel-02930860, version 1


Laure Ngabirano. Matrices Emplois-Expositions : Exemples de création, validation et application. Santé publique et épidémiologie. Université Paris-Saclay, 2020. Français. ⟨NNT : 2020UPASS072⟩. ⟨tel-02930860⟩



Record views


Files downloads