Abstracts: ISEE 21st Annual Conference, Dublin, Ireland, August 25-29, 2009: Symposium Abstracts
Background and Objective:
We investigated effect modification by season and potential confounding by ambient pollen levels and circulating respiratory infections on short-term associations between ambient air pollutants and pediatric emergency department visits for asthma or wheeze.
Hospital-specific daily counts of emergency room visits for asthma or wheeze among children age 5–17 were collected from 41 Metropolitan Atlanta, USA, hospitals during 1993–2004 (n = 91,386 visits). Three-day moving averages of central monitoring station ambient air pollution measurements were calculated for the criteria gases, PM10, coarse PM, PM2.5, and several PM2.5 components. Risk ratios for the warm season (May-October) and cold season (November-April) were estimated via Poisson generalized linear models so as to approximate a case-crossover analysis (matching on year, month, and lag 0 max temperature in Celsius) while accounting for overdispersion. We included a cubic polynomial for day-of-season; indicator variables for hospital and day of week; the daily count of upper respiratory infections among children age 5–17 (aggregated over hospitals); and three-day moving averages of dew point and minimum temperature. We also investigated confounding by ambient levels of several pollen species.
We observed several significant positive associations during the warm season and few associations during the cold season. Ozone in particular was strongly associated with pediatric asthma emergency department visits in both seasons. In multipollutant models we observed evidence for effects of both ozone and traffic pollutants. Ambient pollen levels did not appear to confound these relationships. Risk ratio estimates tended to be closer to unity when we controlled for upper respiratory infections, although associations were present regardless.
Although we controlled tightly for temperature and temporal trends, we nevertheless observed several positive, statistically significant associations between ambient pollutants and pediatric emergency department visits for asthma and wheeze.