Objective: We have recently published a study on new-onset asthma in a large population in northern Europe using a modified job exposure matrix (N-JEM) to better reflect exposure assignment in these countries. The aim of this paper was to investigate how the N-JEM differs in exposure assignment and asthma risks from an already established JEM.
Method: The study comprised 6253 men and 7031 women from northern Europe, born 1945-1973, who had answered both a screening (1989-1992) and a follow-up questionnaire (1999-2001). During the study period (1980-2000), there were 136 men and 293 women with new-onset asthma. Hazard ratios of new-onset asthma were calculated for both JEMs using Cox regression models. The analyses were made separately for men and women and were also stratified for atopy. Cohen's kappa (κ) was used to show agreements in exposure assignment (yes/no) between the JEMs. Population attributable risks (PARs) were calculated as well.
Results: The agreement in exposure assignment between the JEMs was substantial for the group 'any exposure' to asthma agents (κ = 0.78). The agreement between comparable exposure groups in the JEMs varied from κ = 1.00 (pharmaceutical product antigens, textile dust, cleaning agents) to κ = 0.27 (low molecular weight agents). Significant increased asthma risks were seen for men exposed to isocyanates and accidental peak exposure with both JEMs. With the N-JEM, increased asthma risks were seen for men exposed to plant-associated antigens (all and non-atopic), epoxy compounds (all and non-atopic), and acrylates (non-atopic). With the other JEM, increased asthma risks were seen in men and women exposed to 'possible exposure to irritant gases or fumes' (all and non-atopic), a group classified as having low asthma risk. Men and women exposed to cleaning agents also showed significant asthma risks with both JEMs. PAR with the N-JEM was 14.3% for men and 6.6% for women, compared with 12.9% and 8.3% with the other JEM.
Conclusions: Acrylates, epoxy compounds, and isocyanates are three exposure groups in the modified asthma JEM that might better reflect exposure situations in northern Europe than the already established JEM. Exposure to 'possible exposure to irritant gases or fumes', a low asthma risk group in the established JEM, seems to be a group with high asthma risk in northern Europe. It is important to continuously update JEMs, which are based only on occupational titles, in order to find new risk groups and to better reflect changes in work exposures when old risks disappear and new emerge.
Keywords: Cohen’s kappa; asthma specific job exposure matrices; exposure assignment; hazard ratios; new-onset asthma; occupational asthma; population attributable risk.