Traffic Congestion as a Risk Factor for Mortality in Near-Road Communities: A Case-Crossover Study

Existing epidemiologic research on traffic largely neglects localized fluctuations. We leveraged finely resolved congestion data to investigate short-term associations with mortality in communities near roadways. We identified all nonaccidental, cardiovascular, cerebrovascular, and respiratory deaths (2009-2013) within 1 km of a highway in the Puget Sound region of Washington State. Using a case-crossover design, we examined the association of congestion 0-150 m, 151-300 m, and 301-1,000 m upwind of a decedent's home with mortality, adjusting for meteorology, holidays, and influenza activity. Among 9,449 deaths, we observed higher odds of cerebrovascular and respiratory mortality with greater upwind congestion, especially congestion near the decedent's home. For each 10-minute-km increase in upwind congestion within 150 m, the odds of cerebrovascular mortality were 1.08 (95% confidence interval (CI): 0.88, 1.33); within 151-300 m, the odds of cerebrovascular mortality were 1.05 (95% CI: 0.98, 1.12) times higher. We observed similar patterns for respiratory mortality, with 1.06 (95% CI: 0.76, 1.50) times higher odds of death with greater upwind congestion within 150 m and 1.02 (95% CI: 0.95, 1.10) times higher odds within 151-300 m. No increased odds of mortality were observed at greater distances, for overall mortality, or with downwind congestion. Unexpectedly, lower odds of cardiovascular mortality were suggested with greater congestion. This work demonstrates the use of nontraditional data to characterize the impacts of near-road exposures.