Exposure to PAHs of workers on contaminated sites: a modelling approach to assess the occupational contribution – the SOLEX study

Abstract

The SOLEX study sought to assess the occupational contribution to PAHs exposure among employees working on former gasworks sites. Four categories of volunteers were selected according to their job profile: office, mixed, outdoor, and remediation (workers close to a soil desorber); during each of the two study periods (November 1997 and June 1998) a reference group was also recruited. Personal exposure was estimated by 1‐HOP measurements in urine samples taken before and after the shift each day of the study week. All participants completed a questionnaire describing their PAH exposure from sources other than job‐related. Each exposure job category is described by the median of the individual medians of 1‐HOP concentrations across all subjects in the group. Multiple linear regression models assessed the association between 1‐HOP urine concentrations and the occupational and general environmental exposure variables. The statistical models' fit were good (R ² = 0.87 in November 1997 and 0.75 in June 1998). Belonging to the outdoor and desorber job categories and smoking were the key exposure predictors for both periods. Physical activity while at work, and commuting time (using an automobile) were also predictive variables in November 1997. Compared to the reference group, working outdoors and near the soil desorber represented 70–99% of the total PAHs exposure among non‐smokers, and 54–97.6% among smokers. A measurable excess of PAH exposure was observed only among employees directly involved in the soil remediation process; soil pollution was responsible for nearly all the exposure of these subjects, smoking representing a smaller fraction.

Keywords:

• 1‐hydroxypyrene
• multiple linear regression
• contaminated soil
• occupational exposure

Acknowledgements

This research was supported by the French Environment Ministry (contract EN97C20 and EN97C22) and Gaz de France. The authors want to thank the volunteers and the doctors of the occupational medical services of Gaz de France. They also want to thank David Poizeau for his major contribution to the statistical analysis.

Notes

[n]: Number of volunteers.

*: Difference with the reference group p< 0.05.