Never Smokers -- Are They More Sensitive to the Respiratory Health Effects of Ambient Air Pollution?

Background: Several studies show an association between ambient particulate matter (PM) and all-cause mortality. The Adventist Health and Smog 1 (AHSMOG-1) study (N=6,338) has previously found associations between ambient air pollution and incident chronic obstructive pulmonary disease (COPD) using the spatial interpolation method from the three nearest fixed monitoring stations to residence and workplace. However, few studies have assessed the risk of death among disease specific subgroups such as those with COPD. Objectives: The aims of this study were 1) to assess the effect of chronic exposure to ambient air pollutants on risk of all-cause mortality among persons with COPD and/or asthma. 2) To assess the effect of chronic exposure on ambient air pollutants (PM2.5 and O3) and the risk of non-cancer respiratory mortality among a new and larger cohort of Adventists from the Adventist Health and Smog Study 2 (AHSMOG-2). The study population consists of subjects who have never smoked and who have not changed their primary residence since enrollment in the study. This study specifically assessed chronic exposure to ambient particulate and gaseous pollutants (PM2.5 and O3) by utilizing AHSMOG-2 data and mortality data from National Death Index (NDI) for the years 2002-2007. Methods: Using data from the Adventist Health Study-2, a cohort of 97,000 US and Canadian subjects, a subgroup of 43,964 never smoking subjects who had not moved since enrollment, lived in the US, and who self-reported having been diagnosed with either COPD and/or asthma was selected. Further, these subjects had completed a 50 page baseline questionnaire which included demographics, medical history, time spent outdoors, exercise, smoking, and diet. A total of 102 subjects have died during an average of 3 years and 10 months of follow-up. The Cox proportional hazard model was used for survival analysis, adjusting for a number of potential confounders including age, race, education, body mass index (BMI), time spent outdoors, physical exercise, and dietary patterns. Inverse distance weighting (IDW) with interpolation to residence address was used to estimate individual level ambient air pollution. Cumulative monthly and averages from 12 months prior to enrollment until one month before censoring were used. Results and Conclusion: For each 10 μg/m3 increase in PM2.5, the hazard ratio (HR) for all-cause mortality was 2.70 (95% (Cl): 1.09-6.65) in the single-pollutant model and 2.23 (95 % Cl: 0.99-5.03) in the two-pollutant model with O3. Elevated, but not statistically significant associations were found with O3, either in single or two-pollutant models (HRs and 95% Cl were 1.32 (0.74-2.36) and 1.33 (0.79-2.25) respectively). For each 10 ppb increase in O3, the HR for any cause of death due to non-cancer respiratory diseases was 1.93 (95% Cl: 1.31- 2.85) in the single-pollutant model and 1.95 (95% Cl: 1.32-2.87) in the two-pollutant model with PM2.5. No associations were found with PM2.5 either in single or two-pollutant models (HRs and 95 % Cl were 1.12 (0.64-1.97) and 1.21 (0.69-2.12) respectively). These findings could have implications for policy regulations