Association Between Environmental Quality and Lung Cancer Survival in the United States, 2000-2005

Abstract

Background/Aim: Lung cancer remains one of the most prevalent and lethal cancers in the United States. Individual environmental exposures have been associated with lung cancer incidence. However, the impact of cumulative environmental exposures on survival is not well understood and we aim to address this gap in literature. Methods: We estimated county level overall environmental quality in the United States using U.S. Environmental Protection Agency’s (USEPA) Environmental Quality Index (EQI). The EQI captures exposure to over 200 environmental factors across five environmental domains (air, water, land, sociodemographic, and built) for the years 2000-2005. For persons diagnosed with lung cancer from 2000-2005 (n=272,371), county-level smoking prevalence and individual level data on survival time (through Dec. 31, 2013) and co-variates (age, marital status, sex, histology, stage, race, surgery, radiation) from the Surveillance, Epidemiology, and End Results Program (SEER) were linked to the EQI based on place of residence. We modeled the EQI and associated domain-specific indices as quartiles (Q; Q4 worst environment) and used Cox Proportional Hazard models to estimate hazard ratios (HR) and 95% confidence intervals. We adjusted for individual-level covariates and stratified by stage at diagnosis (local, regional, distant) and rural-urban status. Results: Prior to stratification, overall EQI and individual domains showed mostly null or slight positive or negative associations comparing highest to lowest index quartile (EQIQ4: HR=0.92(0.89,0.95). Post stratification, we observed stronger positive associations for individual domains across different strata (e.g. Non-Metro Urban, Local Stage AirQ4: HR=1.21(1.11, 1.33); Metro Urban, Local Stage SociodemQ4: HR=1.08(1.03,1.13)). Most positive associations were observed for urban areas and at the localized cancer stage. Conclusions: Our results suggest a link between aspects of environmental quality and lung cancer survival with potential variation by rural-urban status and stage at diagnosis. This abstract does not necessarily reflect EPA policy.