Background To estimate the effect of mechanical exposure on musculoskeletal pain, there is a need to use valid and feasible exposure assessment methods. Direct measurements are not possible in large epidemiological studies and self-reported measures are prone to bias. Job exposure matrices (JEM) have been shown as a promising tool to assess work-related chemical exposures. Gender-specific JEM for mechanical exposures has, however, been used to a limited extent. The aim was therefore to develop and validate a gender-specific job exposure matrix for mechanical exposures in a representative sample of the Norwegian population.

Methods Data from the Living Conditions Survey on Work Environment, collected by Statistics Norway in 2006 and 2009, was randomly split and one half was used to construct the JEM and the other to test the matrix validity. Occupation was classified on the 4 digit level according to the 1998 version of the Norwegian Standard Classification of Occupations (STYRK98), based on the International Standard (ISCO-88). Mechanical exposure information was collected by personal telephone interviews and included exposures that increase musculoskeletal load.

Results The matrix represented 94% (329) of job titles in STYRK98. Among men, 41% of the job titles had ≥20 respondents while 45% of the job titles had ≤10 respondents. Among women, the corresponding proportions were 23% and 66%, respectively. After merging job titles with few respondents (≤10), based on similar work tasks and educational level, the number of occupational groups were 168 among men and 115 among women. Preliminary results show moderate agreement (kappa) between the group-based exposures and the individual-based exposures.

Conclusion To evaluate whether the constructed gender-specific JEM can be regarded valid for exposure assessment in large epidemiological studies, the JEM’s specificity and sensitivity and its predictive validity will be further analysed.