Date of Award




Document Type


Degree Name

Doctor of Philosophy (PhD)


Department of Environmental Health Sciences

Content Description

1 online resource (vii, 88 pages) : color illustrations, color maps.

Dissertation/Thesis Chair

Shao Lin

Committee Members

Greg Nemet, Scott Sheridan, Fangqun Yu, Kai Zhang, Wangjian Zhang


Case-crossover, Chronic lower respiratory disease, Diurnal temperature range, GEOS-Chem, Psychoactive substance use, Ultrafine particles, Air, Particulate matter, Air quality, Lungs, Respiratory organs, Global temperature changes

Subject Categories

Environmental Health | Epidemiology | Public Health


Background: Air pollution leads to more than four million premature deaths worldwide and nearly 200,000 in the US each year. While there is extensive research relating PM2.5 to hospital visits for various health outcomes, there is relatively little research on ultrafine particles (UFPs), a particularly small and dangerous air pollutant. Another important exposure is temperature due to its association with dehydration, heat stress, and cardiovascular-related hospital admissions. However, relatively little research has been conducted on diurnal temperature range (DTR), or the range of temperature within a single day, which is an important indicator of climate change. Importantly, the interaction between temperature and air pollution is understudied yet potentially significant, as both exposures cause significant systemic stress.Purpose: This project helps fill existing knowledge gaps by examining the lagged health effects of UFPs on traditional (chronic lower respiratory diseases, Aim 1) and non-traditional air pollution-related health outcomes (mental health disorders and psychoactive substance use, Aim 2). This project also explores the multiplicative interaction between UFP concentration and DTR, which is innovative in the field. Finally, this project examines how electricity production contributes to local asthma rates (Aim 3). Methods: Aims 1 and 2 are individual-level analyses that utilize a case-crossover design. In addition, these aims use conditional logistic regressions to estimate the excess risk of visiting the emergency department for a host of health outcomes while controlling for atmospheric conditions, important co-pollutants, and time-varying variables. Aim 3 utilizes a cross-sectional design and uses zip-code level population and poverty data to calculate asthma rates around each type of power station. Results: Aim 1) Each interquartile range increase in ultrafine particle exposure led to a peak risk of a respiratory-related emergency department visit on lag 0-6 (1.8%, 95% CI: 1.6-2.0%). The highest risk was in the subtype emphysema (lag 0-5: 4.2%, 95% CI: 0.2-8.4%), followed by asthma, chronic bronchitis, other chronic airway obstructions, and unspecified bronchitis (lag 0-6, excess risk range: 1.5-2.0%). We also found significant interactions with seasons (especially fall), mild thermal conditions (temperature/RH <=90th percentile), for children (<18), and among men (lag 0-6, excess risk range: 1.0-2.8%). Aim 2) CLRDs experienced the largest increased excess risk (ER), 7.34% (95% CI: 5.86, 8.84) on lag 0-6, followed by drug dependency (5.14%), drug psychoses (5.01%), mental health and behavioral disorders (MHDs) combined (3.55%), alcoholic psychoses (3.35%), and psychoactive substance use combined (2.47%). CLRDs also experienced the largest increased excess risk associated with DTR on 0-6 (2.04%, 95% CI: 1.67-2.41), followed by alcohol dependence (0.94%). CLRDs, again, experienced the most significant UFP-DTR interaction result on lag 0-6 (1.044), followed by drug psychoses (1.032), psychoactive substance use (1.015), alcoholic psychoses (1.018), and MHDs (1.014). Aim 3) Populations living within 3-miles of battery power stations experienced 28.0 times as many asthma cases (95%CI: 27.6-28.4) as expected after controlling for population, followed by petroleum (RR: 19.3), natural gas (RR: 7.6), hydroelectric (RR: 3.6), solar (RR: 2.3), biomass (RR: 1.7), wind (RR: 0.6), flywheels (RR: 0.4), and nuclear (RR: 0.2). Asthma risk ratios were highest in the poorest quartiles of zip codes compared to the richest for petroleum (RR: 15.5, 95% CI: 15.4-15.6 vs. RR: 0.3, 95% CI: 0.3-0.3), natural gas (RR: 6.6 vs. RR: 1.8), and solar (RR: 3.0 vs. RR: 1.0) power stations. Asthma rates fell 12.5% faster in coal plant-containing zip codes than the state average. Conclusion: In this study, UFP exposure increased CLRD, MHD, and substance use-related ED visits. Conversely, the DTR results were mixed, and more research is required in this area. In addition, UFP-DTR associations for alcohol-related outcomes were significantly different from drug-related outcomes. Information regarding UFP and DTR-health associations may be useful for surveillance and healthcare efforts during low air quality days. Future power plant-health analyses should control capacity factors, installed pollution technologies, population age, proximity to roadways, proximity to other stationary sources of air pollution, smoking status, atmospheric conditions, and air pollution dispersion. Finally, the efforts to link power stations, UFPs, and health outcomes are immature but promising.