Quantitative exposure assessment in community-based studies

Objective
The quality of exposure assessment has been shown to have large implications on epidemiological results. For hazard identification, qualitative or semi-quantitative exposure assessment is sufficient, while risk quantification requires more detail. We describe a quantitative exposure assessment approach for five lung carcinogens (i.e. asbestos, chromium, nickel, polycyclic aromatic hydrocarbons (by its proxy benzo(a)pyrene (BaP)) and respirable crystalline silica (RCS)) to be applied within the SYNERGY project - a pooled analysis of lung cancer case-control studies.
Methods
Over 350,000 occupational measurements, collected from national and institutional databases or paper records existing in research institutes in Europe and Canada, were gathered in the ExpoSYN database. Statistical empirical prediction models were developed to create a quantitative job-exposure matrix (SYN-JEM).
Results
SYN-JEM provided job-, region-, and time-specific exposure estimates. Because of a limited number of measurements a simpler predictive model was used for estimating BaP, i.e. without using random effects for region and job. Decreasing time trends were observed for all selected agents, ranging from -1.2% per year for personal BaP and nickel concentrations to -10.7% for asbestos (before bans were implemented). Regional differences from the full statistical model (corrected for jobs, measurement year, sampling duration, etc.) ranged from a factor 3 for chromium to 10 for asbestos.
Conclusion
Statistical modelling of occupational exposure measurements enabled us to predict job-, region-, and time-specific exposure levels for four (asbestos, chromium, nickel and RCS) out of five considered lung carcinogens. The use of quantitative data should be encouraged for use in community-based studies, allowing for more advanced analyses.