Summary and findings of the EPA and CDC symposium on air pollution exposure and health

Abstract

. The symposium brought together health and environmental scientists to discuss the state of the science and the cross-jurisdictional and methodological challenges in conducting air pollution epidemiology, environmental public health tracking and accountability research. The symposium was held over 2 days and consisted of technical presentations and breakout group discussions on each of the three principal themes of this meeting: (1) monitoring and exposure modeling information, (2) health effects data and (3) linkage of air quality and health data for research, tracking and accountability. This paper summarizes the symposium presentations and the conclusions and recommendations developed during the meeting. The accompanying two papers, which appear in this issue of the Journal, provide more in-depth discussion of issues pertinent to obtaining and analyzing air pollution exposure and health information. The symposium succeeded in identifying areas where there are critical gaps of knowledge in existing air pollution exposure and health information and in discovering institutional or programmatic barriers, which impede accessing and linking disparate data sets. Several suggestions and recommendations emerged from this meeting, directed toward (1) improving the utility of air monitoring data for exposure quantification, (2) improving access to and the quality of health data, (3) studying emerging air quality and health issues, (4) exploring improved or novel methods for linking data and (5) developing partnerships, building capacity and facilitating interdisciplinary communication. The meeting was successful in promoting an interdisciplinary dialogue around these issues and in formulating strategies to support these recommended activities. Finally, this symposium subsequently led to strengthening and initiating new partnerships or interactions between the EPA, CDC, States, academia and the research community at large.

A Decade of Environmental Public Health Tracking (2002-2012)

“…This event appeared somewhat pivotal, bringing together the Environmental Protection Agency, CDC, and many Environmental Public Health Tracking stakeholders together to address air quality and health issues while developing strategies for building capacity across agencies and communities. Following the conference, McKone et al 65 and Ozkaynak et al 67 summarized and reported that workgroup participants’ recommendations included obtaining better exposure information for evaluating health effects and improving data linkage and data integration in exposure assessment approaches.…”

Development and evaluation of alternative approaches for exposure assessment of multiple air pollutants in Atlanta, Georgia

“…37 Although air quality and exposure models have the ability to introduce variability not present in CS measurements, the potential to introduce greater uncertainty in the resultant health effect estimates due to modeling error must be considered. 4 When regional pollutants (PM 2.5 , SO 4 , O 3 ) are of interest, CS measurements may be sufficient to reflect spatial variability, especially for time-series or case-crossover studies over large urban or metropolitan scales, because of the limited local-scale spatial variability of these pollutants at the ZIP code level, and due to the strong temporal correlation between CS measurements and either hybrid or exposure model estimates. However, in studies of local pollutants (EC, CO, and NO x ), both air quality modeling and exposure modeling may need to be considered in order to represent spatial variability adequately.…”

“…12 The modified approach took ambient CS monitor hourly measurements for each pollutant and removed local source contributions as modeled by AERMOD (see exposure metric (iii) below) to infer hourly estimates of regional BG pollution at each monitoring site, later interpolated to ZIP code centroids as described below. Hourly measurement data from six NO x monitors, four CO monitors, 14 O 3 monitors, and five PM 2.5 monitors were used in this study; two PM 2.5 composition monitors provided 24-h measurements of EC and SO 4 . For details of locations of monitors used for creating BG estimates see Figure 1.…”

“…[1][2][3][4][5] Fixed-site monitors may be sufficient for representing ambient concentrations for pollutants with limited spatial and temporal heterogeneity. For pollutants with local source impacts, the concentrations measured at CS monitors may not represent intra-urban variation in air pollution levels.…”

Engaging academia to advance the science and practice of environmental public health tracking

“…In collaboration with the U.S. Environmental Protection Agency (EPA), we have developed a hierarchical Bayesian model to predict daily PM 2.5 concentrations on a small scale (Vaidyanathan et al, 2013). Further we have hosted several workshops with EPA, academia, and state and local partners to identify and collaborate on common methodological challenges (Matte et al, 2009; Ozkaynak et al, 2009) and have evaluated the use of different data sources and methods for tracking health outcomes, exposures, and hazards (Beale et al, 2010; Rabito et al, 2007; Talbot et al, 2009; Wartenberg et al, 2008; Wilhelm et al, 2008; Young et al, 2008). …”