2002 Progress Report: Multiple Pollutants and Risk of Emergency Department Visits for Cardiorespiratory Outcomes in AtlantaEPA Grant Number: R829213
Title: Multiple Pollutants and Risk of Emergency Department Visits for Cardiorespiratory Outcomes in Atlanta
Investigators: Tolbert, Paige , Flanders, Dana , Klein, Mitchel , Lyles, Robert , Metzger, Kristina , Mulholland, James , Peel, Jennifer , Russell, Armistead G. , Ryan, P. Barry , Todd, Knox , Waller, Lance
Current Investigators: Tolbert, Paige , Flanders, Dana , Klein, Mitchel , Lyles, Robert , Metzger, Kristina , Mulholland, James , Peel, Jennifer , Russell, Armistead G. , Ryan, P. Barry , Sarnat, Stefanie Ebelt , Sarnat, Jeremy , Todd, Knox , Waller, Lance
Institution: Emory University
EPA Project Officer: Chung, Serena
Project Period: January 1, 2002 through December 31, 2004 (Extended to December 31, 2006)
Project Period Covered by this Report: January 1, 2002 through December 31, 2003
Project Amount: $1,238,940
RFA: Health Effects of Particulate Matter (2001) RFA Text | Recipients Lists
Research Category: Air Quality and Air Toxics , Particulate Matter , Health Effects , Air
The main objective of this research project is to investigate the role of specific air contaminants and the interrelationships among them in producing acute exacerbations of certain cardiac and respiratory conditions. To do this, the study period of an ongoing epidemiologic study will be doubled and multipollutant questions in the resulting unique and powerful database will be addressed. The study is a time-series investigation of cardiac and respiratory emergency department (ED) visits in relation to daily measures of air quality, including detailed measurements of particulate matter (PM) components being conducted at the station in operation for the Aerosol Research and Inhalation Epidemiology Study (ARIES) in downtown Atlanta.
The study period for the previous epidemiologic study, part of the Study of Particulates and Health in Atlanta (SOPHIA), conducted in conjunction with ARIES, was August 1, 1998, through August 31, 2000. Operation of the ARIES station is being extended for an additional 2 years, through August 31, 2002, and in this project, we will extend outcome data collection for this period and conduct epidemiologic analysis of the combined data. Outcome data will be obtained from the 31 hospitals in the metropolitan Atlanta area participating in the SOPHIA study. Using the diagnostic codes for the visits, multiple cardiac and respiratory case groups will be formed. For each group, time-series analysis will be performed to examine the relationship of pollutant levels to daily counts of the outcome. Pollutants of a priori interest being measured at the ARIES station include: the criteria gaseous pollutants, oxygenated hydrocarbons, multiple-size fractions of PM-including PM10, coarse PM, PM2.5, a number of ultrafine particles (10-100 nm)-and selected components of PM2.5: sulfates, nitrates, acidity, water-soluble metals, elemental carbon, and organic matter. Control for copollutants and assessment of interactions among them will be achieved through standard modeling approaches as well as through innovative methods. In addition, the impact of spatial variability on the epidemiological results will be assessed.
With more than 1 million ED visits per year, this study may be the largest study of its type (i.e., single-city ED study with speciated PM data). This investigation will make advances in disentangling effects of PM from the effects of other pollutants and will contribute to our understanding of the effects of exposure to PM in the presence of other pollutants. Moreover, the detailed air quality data available to the proposers will allow assessment of the role of components of PM (e.g., sulfates, water-soluble metals) and size fractions in the multipollutant analysis.
In the first year of this grant, we worked with the 41 hospitals in the metropolitan Atlanta area to obtain data on ED visits for the period 2000-2002 for use in our study. To date, cooperation from the hospitals has been as good as or better than it was for the previous study period (through 2000), for which we were able to obtain participation from 31 of 41 hospitals (comprising 80 percent of the ED visits in Atlanta). The process of getting the required approvals and eventually the data entailed multiple mailings and telephone calls to hospital CEOs and other officials, as well as the information services staff. As batches of new ED data came in from the participating hospitals, we reviewed and edited the data. This entails examination of the distributions of the various data elements and other descriptive analyses. We compared data across hospitals and within hospitals for the earlier period and the current period and continued the analysis of in-house data. As we collected new data for the period 2000-2002, we continued to analyze the data we have in house for the period through 2000. The additional analyses that we have been performing on these data will help to inform and expedite analyses of the data through 2002 when they are available.
Analyses of the relationship of air quality to cardiovascular disease through 2000 have been submitted for publication. Cardiovascular disease visits were associated with NO2, CO, PM2.5, elemental carbon, and organic carbon, which may be acting as markers of traffic exhaust. The effect of ambient pollution on cardiovascular conditions appeared to be rapid, as the strongest associations tended to be observed with pollution levels on the same day as the ED visits. In addition, we assessed the spatial variability of the different pollutants of interest in collaboration with our Georgia Tech colleagues. We worked on methodological issues, and conducted extensive work on modeling issues relating to the time-series air pollution analyses that we will conduct for this research project, including the optimal method of model selection. In addition, we developed an approach to account for measurement error in our study. Finally, we were one of two teams that independently reported the problem of variance underestimation using the standard software for generalized additive models (Klein, et al., 2002; Ramsay, et al., 2002) this past year. We subsequently developed an improved variance estimator that performed well in simulations; this work has been submitted for publication.
In the coming year (Year 2), we will perform the following activities: (1) complete the assembly, editing, and descriptive analysis of hospital data for the period 2000-2002; (2) review and clean the new air quality data from the ARIES station for the period 2000-2002; (3) review database of other monitoring data (Aeromatic Information Retrieval System [AIRS] and Assessment of Spatial Aerosol Composition in Atlanta [ASACA]) assembled and edited by Georgia Tech collaborators for the period 2000-2002; (4) perform epidemiologic analysis of new 2-year data and compare results with those for the period 1998-2000 (using ARIES and ASACA data) and 1993-2000 (using AIRS data); (5) perform initial epidemiologic analysis of data for the combined periods (1998-2002 for ARIES and ASACA data and 1993-2002 for AIRS data); and (6) continue to work on multipollutant modeling strategy, as well as our approach for assessing impact of measurement error.
Journal Articles on this Report : 1 Displayed | Download in RIS Format
|Other project views:||All 107 publications||31 publications in selected types||All 29 journal articles|
||Flanders WD, Klein M, Tolbert P. A new variance estimator for parameters of semiparametric generalized additive models. Journal of Agricultural, Biological, and Environmental Statistics 2005;10(2):246-257.||