Text

theses

Exposure to ambient fine particulate matter (PM2.5) is associated with multiple negative health outcomes. Studies investigating these associations commonly use PM2.5 concentrations measured at outdoor, central-site monitors to estimate exposure. Because people spend the majority of time indoors, however, the variable efficiency with which ambient PM2.5 penetrates and persists indoors is a source of error in epidemiologic analyses. This error generally results in an underestimation of health effects, hampering the detection of associations between ambient PM2.5 exposures and the risk of health outcomes. To reduce this error, practical methods to model indoor concentrations of ambient PM2.5 are needed. This dissertation contributes to exposure science by advancing existing models of residential exposure to ambient PM2.5 and by improving the robustness and accessibility of these tools. First, drivers of variability in the fraction of ambient PM2.5 found indoors (F) are identified and the potential for this variability to explain observed heterogeneity in PM-mediated health-effect estimates is explored. Next, a physically-based mass-balance model and modeling tools that account for variability in human activity patterns (e.g. time spent in various indoor and outdoor environments) are used to compute ambient PM2.5 exposures that account for the modification of PM2.5 with outdoor-to-indoor transport in order to explore whether the use of these refined exposure surrogates reduces error and bias in epidemiologic analyses. Subsequently, this outdoor-to-indoor transport model is evaluated and refined using measured indoor and outdoor PM2.5 concentrations and air exchange rates, providing a practical and robust tool for reducing exposure misclassification in epidemiologic studies. Finally, the volatility basis set is used for the first time to study shifts in the gas-particle partitioning of ambient organics with transport indoors. This dissertation provides guidance regarding measurements and data most critically needed to facilitate the prediction of refined exposure surrogates in large epidemiological studies and, thus, informs the design of future sampling campaigns and epidemiologic studies. It enables a better accounting of ambient particle penetration into and persistence in the indoor environment and constitutes an important advancement in the efforts to reduce exposure error in epidemiologic studies and to elucidate relationships between PM2.5 exposure and adverse health outcomes.

ETD_5330

Ph.D.

Includes bibliographical references

by Natasha Hodas

doi:10.7282/T3GB22CD

ETD doctoral

The author owns the copyright to this work.