Out-of-Hospital Cardiac Arrests and Fine Particulate Matter (PM2.5) during 2015-2017 California Wildfires

Abstract

OPS 10: Wildfires, Room 210, Floor 2, August 26, 2019, 10:30 AM - 12:00 PM Background: The burden of California wildfires is increasing, yet cardiovascular health impacts of wildfire smoke are not well understood. Methods: We examined the relationship of wildfire smoke exposures and cardiac arrest occurring outside a hospital setting using categorical estimates of wildfire-related PM2.5 (light, medium, heavy smoke) derived from satellite-imagery in the National Oceanic Atmospheric Association’s Hazard Mapping System Smoke Product. Surveillance data of EMS-attended out-of-hospital cardiac arrests (OHCA) in adults (≥ 35 years of age) was provided by the Cardiac Arrest Registry to Enhance Survival for 14 California counties during wildfire months, May – October, 2015-2017. We applied conditional logistic regression incorporating a spline for heat index in a case crossover design (matching smoke exposures with OHCA cases and 4 referent day controls from 1, 2, 3, and 4 weeks prior to case date). We modeled exposures for lags 0-3, stratified by gender, age (35-64 and ≥ 65), and SES (percent living below poverty). Results: We observed positive relationships between heavy smoke levels (>22 µg/m3) and OHCA, consistent across lags 0-3, while light and medium smoke often had negative associations. These relationships were more pronounced in the lower SES group, which experienced impacts across all exposure levels, while the higher SES group tended to have elevated ORs only with heavy smoke, and even showed deficits with light and medium smoke. Conclusion: This study of cardiac arrest occurring outside a hospital setting may capture associations not represented in previous studies based on health care utilization. Consistent with many environmental exposures, the lower SES population appeared to be more vulnerable to the effects of wildfire smoke. Further investigation can help explain the differences in effects by SES and gender, and inform interventions most relevant to vulnerable populations during wildfire events. This abstract does not necessarily reflect USEPA policy.