Roadway Proximity and Lung Cancer in Participants of the NIH-AARP Cohort in California

Abstract

Introduction: Outdoor air pollution is classified as a cause of lung cancer. Certain traffic-related pollutants decay rapidly with distance from roadways, yet few studies have evaluated risks by residential proximity to roads. Distance-based exposure surrogates may help to identify the pollutant components that drive lung cancer risk. Methods: From residential addresses at enrollment in 1995 and U.S. Census TIGER/Line® files, we generated traffic proxies for 143,223 California participants in the NIH-AARP Diet and Health Study as distances (m) to the nearest multi-lane roadway (A1: limited access highways, A2: other major highways, A3: secondary roads). With 4,054 incident cases and follow-up through 2011, we used Cox models to estimate lung cancer risks in relation to distances categorized as 0-49m, 50-199m, and ≥200m, and continuously (per 100m). Models were adjusted for age, sex, race/ethnicity, smoking status and intensity, education, and family history of cancer. We also compared associations in the historically heavily polluted tri-county area of Los Angeles, Orange, and Riverside (N=53,362) to the rest of the state (N=86,654). Results: Overall, 3% of California participants and 12% of cases lived within 200m of any A1-A3 road. We observed no association with lung cancer overall for individuals residing ≤49 or 50-199m compared to ≥200m from these roads, nor for continuous distances. However, risks of small cell carcinoma (SCC) of the lung were elevated, with associations stronger for those in closest proximity (HR[≤49m]=1.71,95%CI=1.09-2.67 and HR[50-199m] =1.23,95%CI=0.93-1.63). SCC risks ≤49m were also higher in the tri-county area (HR=2.93,95%CI=1.59-5.43) than in the rest of the state (HR=1.12,95%CI=0.58-2.18). Conclusions: These findings suggest that risk of SCC increases in close proximity to roadways. Validation of exposures with traffic and pollutant estimates is warranted.