Objective To analyse the associations between life-time occupational exposures to diesel, leaded and unleaded gasoline engine emissions and lung cancer.

Methods Our case-control study enrolled 1503 lung cancer cases and 1198 population controls between 1996 and 2001 in Montreal. Occupational exposure to diesel, leaded and unleaded gasoline engine emissions was assessed using a combination of subject-reported job and tasks history and expert assessment. Exposure status “Ever or never exposed”, duration, frequency and concentration of exposure based on qualitative assessment were assigned to each participant. Lifetime average exposure and cumulative exposure were derived as semi-quantitative indices. Using multivariate logistic and generalised additive logistic regression analyses, we evaluated the association of lung cancer with the mentioned exposures. Odds ratios (OR) were calculated for an index increase between exposure to environmental levels and exposure levels at the 90th percentile of the entire population.

Results In models including only one type of engine emissions at a time, only diesel showed an association with lung cancer (OR for average exposure: 1.19, 95% CI: 1.03, 1.37). When restricting the analysis to participants exposed to one of the three types of engine emissions exclusively, cumulative exposure to leaded gasoline, unleaded gasoline, and diesel engine emissions were associated with odds ratios of lung cancer of 2.11 (95% CI: 1.25, 3.56), 0.66 (95% CI: 0.25, 1.72) and 1.09 (95% CI: 0.95, 1.24), respectively. The inclusion of the three types of engine emissions in one model, using either multivariate logistic or generalised additive logistic regression, yielded similar estimates to those obtained for segments of the population with increased exposure to only one type of engine emissions.

Conclusion Exposure to diesel and unleaded gasoline engine emissions confounded the association of leaded gasoline engine emissions and lung cancer. Increases in cumulative exposure to leaded gasoline were associated with an increased risk of lung cancer.