Table of Contents

ABSTRACT

The 2017 and 2018 wildfire seasons in British Columbia, Canada were unprecedented in their scale, with large areas burned. A previous study found that prenatal exposure to wildfire-related fine particulate matter (PM2.5) was associated with increased risk of otitis media by age one year. The objective of this study was to evaluate whether the observed association between prenatal wildfire-related PM2.5 exposure and otitis media in infants varies by neighborhood indicators of social vulnerability. The study used the British Columbia Perinatal Data Registry (BCPDR) to identify all singleton infants in utero during the wildfire seasons. Daily PM2.5 exposures throughout pregnancy and the first year of life were estimated using the Canadian Optimized Statistical Smoke Exposure Model (CanOSSEM) at a resolution of 5×5 km². All infants who had otitis media treated with amoxicillin by age one year were identified by linking administrative health databases. Logistic regression conditioned on the epidemiologic week of birth was used to estimate the odds ratio associated with prenatal wildfire-related PM2.5 exposure during each trimester. Results were stratified by quintiles of four different indicators of neighborhood vulnerability: economic dependency; ethnocultural composition; situational vulnerability; and residential instability. For every indicator, very high or unknown neighborhood vulnerability showed the strongest association between exposure during the second trimester of pregnancy and otitis media. Our findings suggest that neighborhood social vulnerability may modify the effect of prenatal wildfire smoke exposure, especially during the second trimester. They also highlight the need to focus on populations that are inequitably impacted by the growing threat of wildfire smoke.

Keywords: Environmental health, respiratory health, wildfire smoke, prenatal exposure, health disparities