Commute patterns, residential traffic-related air pollution, and lung cancer risk in the prospective UK Biobank cohort study.

Introduction: Commuting exposes millions of people to carcinogens from traffic-related air pollution (TRAP) but is seldomly considered in epidemiologic studies of lung cancer. In the prospective United Kingdom (UK) Biobank cohort study, we investigated associations between commute patterns, residential nitrogen dioxide concentrations (NO; a surrogate for TRAP), and lung cancer risk.

Methods: We analyzed 234,124 employed participants at baseline (2006-2010). There were 493 incident lung cancer cases diagnosed over an average 7-year follow-up. Subjects were cross-classified into exclusive categories of commute mode (automobile, public transportation, walking, cycling, active mixture, and other mixture) and frequency (regular: 1-4, often: ≥5 work-bound trips/week). Annual average residential NO concentrations in 2005-2007 were estimated with land use regression. Multivariable Cox regression was used to estimate associations between commute patterns, NO quartiles, and incident lung cancer. We conducted analyses stratified by NO (>, ≤median = 28.3 µg/m) and potential confounders such as sex and smoking.

Results: Compared to regular automobile use, commuting often by public transportation was associated with increased lung cancer risk (hazard ratio (HR) = 1.58, 95% confidence intervals (CI):1.08-2.33). Additionally, we found a positive exposure-response relationship with residential NO (HR = 1.21, 95 %CI: 0.90-1.62; HR = 1.48, 95 %CI: 1.10-1.99; HR = 1.58, 95 %CI: 1.13-2.23; p-trend = 3.1 × 10). The public transportation association was observed among those with higher NO (p-interaction = 0.02). Other commute categories were not associated with risk.

Conclusions: Commuters residing in high-NO areas who often use public transportation could have elevated lung cancer risk compared to regular automobile users. These results warrant investigations into which component(s) of public transportation contribute to the observed association with increased lung cancer risk.