Abstract TP176: Residential Proximity to High Traffic Roadways is Associated with Post-Stroke Mortality

Abstract

Background: Living in areas with higher pollution has been associated with incidence of ischemic stroke and higher rates of all-cause mortality. There is considerable interest in the relative role of traffic related pollution specifically, but less is known about the relationship between traffic related pollution and long term survival after stroke. We hypothesized that living closer to a busy roadway is associated with a higher rate of post-stroke mortality in the Boston area. Methods: We identified 1763 consecutive patients age ≥21 years admitted to Beth Israel Deaconess Medical Center with neurologist-confirmed ischemic stroke between April 1, 1999 and October 31, 2008 and residing in the Boston metropolitan area. The Social Security Death Index was used to identify deaths through June 2012. We obtained average daily traffic counts from the Massachusetts Department of Transportation and geocoded patients’ addresses to determine distance to the nearest roadway averaging ≥10,000 vehicles/day. Categories of residential distance were defined as ≤100 meters (m), 100 to ≤200 m, 200 to ≤400 m or >400 m from a busy roadway. Models were adjusted for demographics, medical history, and socioeconomic factors. Results: Among 1682 stroke patients with complete data, there were 826 deaths [median follow up = 4.6 years]. We observed a higher rate of post-stroke mortality among people living closer to high traffic roadways. Patients living ≤100 m from high traffic roadways had a 23% (95% CI: 2%, 49%) higher rate of post-stroke mortality than patients living >400 m away (Figure, p-trend=0.04). Conclusions: In this study, living close to a busy roadway was associated with an elevated mortality rate. This relationship remained statistically significant after adjustment for individual and neighborhood level factors, providing further evidence of the association between traffic-related pollution and mortality in a susceptible population.