Ambient Ozone and Children's Acute Asthma in New York City: a Comparison of Time-Series and Case-Crossover Analyses with Spatially-Resolved Air Pollution Metrics

Abstract

Ambient Ozone and Children's Acute Asthma in New York City: a Comparison of Time-Series and Case-Crossover Analyses with Spatially-Resolved Air Pollution MetricsAbstract Number:2277 Jiang Zhou*, Jessie Shmool, and Jane Clougherty Jiang Zhou* University of Pittsburgh, United States, E-mail Address: [email protected] Search for more papers by this author , Jessie Shmool University of Pittsburgh, United States, E-mail Address: [email protected] Search for more papers by this author , and Jane Clougherty University of Pittsburgh, United States, E-mail Address: [email protected] Search for more papers by this author AbstractBackground Daily changes in ozone have been associated with childhood asthma emergency department (ED) visits and hospitalizations. Prior studies, however, have lacked high- resolution ozone exposure data, relying instead on regional concentration estimates from regulatory monitors, leading to exposure misclassification. Further, few studies compare results in time-series and case-crossover models.Objectives We aim to examine associations between highly spatially- and temporally-resolved ozone exposure estimates and ED visits and hospitalizations for childhood asthma, by combining New York City Community Air Survey (NYCCAS) ozone surface data with daily ozone trends from local EPA Air Quality System (AQS) monitors, to improve spatio-temporal exposure assignment and epidemiological analyses.Methods We compared time-series and case-crossover models for acute effects of ambient ozone using asthma admissions data – ED visits (n = 35,907) and hospitalizations (n = 8,009) for children aged 5 to 17 – and ambient ozone data for May through September from 2005 to 2011. Time-series models were adjusted for temporal trends, temperature, and day of the week. Conditional logistic regression models were used in case-crossover analysis. Excess risks were estimated for interquartile range (IQR) increases in ozone exposures.Results Similar acute effects of ozone on asthma events were observed in both time-series and case-crossover analyses, with similar lag structures. Overall, ozone was positively associated with asthma ED visits in both sexes, for one to six lag days, peaking at lag day 3 (excess risk = 9%; 95% CI: 6.4%, 11.8%). Among asthma hospitalizations, we observed significant effects of ozone for girls, but not boys.Conclusions Ambient ozone was significantly associated with asthma ED visits and hospitalizations. Models will be developed to include spatially-adjusted ozone level in the case-crossover model to capture spatial variation in daily ozone exposures.