Characterization and impact of ambient air pollutant measurement error in time-series epidemiologic studies

Time-series studies of ambient air pollution and acute health outcomes utilize measurements from fixed outdoor monitoring sites to assess changes in pollution concentration relative to time-variable health outcome measures. These studies rely on measured concentrations as a surrogate for population exposure. The degree to which monitoring site measurements accurately represent true ambient concentrations is of interest from both an etiologic and regulatory perspective, since associations observed in time-series studies are used to inform health-based ambient air quality standards.;Air pollutant measurement errors associated with instrument precision and lack of spatial correlation between monitors vary widely across pollutants and these errors have been shown to attenuate associations observed in health studies. Further, the impact of measurement error varies depending on the type of error present, with classical error resulting in greater attenuation than Berkson error, which is expected to yield unbiased effect estimators. Characterization and adjustment for air pollution measurement error can improve effect estimates in time-series studies.;Measurement error due to instrument precision and spatial variability was characterized for ambient air pollutants in Atlanta. Error was modeled for daily measures of 12 air pollutants using measurements from collocated monitoring sites to characterize instrument precision and data from multiple study area monitors to estimate population-weighted spatial variance. Simulations of instrument and spatial error were generated for each pollutant, added to a reference pollutant time-series, and used in a Poisson generalized linear model of air pollution and cardiovascular emergency department visits. This method allows for pollutant-specific quantification of impacts of measurement error on health effect estimates, both the assessed strength of association and its significance.;The impact of these measurement error sources is affected by both the amount and the type of error. Regarding the latter, error simulations ranged in type from purely classical to purely Berkson, as defined on a log scale. To inform on the type and amount of error present in Atlanta measurements, air pollutant concentrations were simulated over the 20-county metropolitan area for a 6-year period, incorporating several distribution characteristics observed in measurement data. Spatial and temporal autocorrelation as well as trends for season, day-of-week and distance from downtown were modeled. The simulated concentration field was then used to characterize the amount and type of error due to spatial variability in ambient concentrations. The impact of use of different exposure metrics in a time-series epidemiologic study was assessed.;Finally, methodologies developed for the Atlanta area were applied to Dallas, Texas. Measurement error due to spatial variability was quantified for ambient monitoring site networks in Dallas with consideration for the impact of this error on a time-series study of Dallas that is currently underway. Differences in air pollution measurement error due to spatial variability between Atlanta and Dallas data were assessed and the impact of this measurement error on health associations in Dallas was discussed.