Date of Award

January 2024

Document Type


Degree Name

Master of Public Health (MPH)


School of Public Health

First Advisor

Kai Chen


Background: As climate and land use change amplify the frequency and severity of wildfire-conducive environmental conditions, wildfire smoke pollution becomes an increasingly pressing public health concern. Understanding how wildfire smoke particulate matter impacts health, and whether it affects different endpoints than those targeted by all-source particulate matter, is vital to our understanding of current and future climate-related health risks.

Methods: A space-time-stratified case-crossover study was conducted to analyze the respiratory, cardiovascular, and mental health impacts of wildfire smoke and nonsmoke PM2.5 exposure on emergency department (ED) visits in Nevada. Total ED visits from 2016-2019 were provided by the Nevada Department of Health and Human Services, and daily PM2.5 concentrations were obtained from modeled national estimates. Nonsmoke PM2.5 absolute values and smoke PM2.5 percentile values were used as exposure inputs. Relative risk of disease was calculated using a conditional Poisson regression controlling for temperature, humidity, and federal holidays, and stratified by year, month, day of the week, and zip code. Single-day lags of up to six days were included in analysis.

Results: We observed the strongest positive associations between one-percentile increases in smoke PM2.5 and same-day asthma (RR = 1.267, 95% CI = 1.164–1.379), 2-day lagged cerebrovascular (RR = 1.183, 95% CI = 1.100–1.272), 3-day lagged schizophrenia (RR = 1.260, 95% CI = 1.169–1.359), and 3-day lagged substance abuse (RR = 1.135, 95% CI = 1.069–1.204) ED visits. Associations between smoke PM2.5 and chronic obstructive pulmonary disease (COPD), non-cerebrovascular circulatory diseases, and anxiety were inconclusive or negative. For each 1 ug/m3 increase in nonsmoke PM2.5, we observed elevated risk of all-respiratory, asthma, and COPD cases, with magnitude increasing at greater daily lags (6-day lag RR = 1.017, 95% CI = 1.015–1.019; 5-day lag RR = 1.021, 95% CI = 1.018–1.024; and 6-day lag RR = 1.018, 95% CI = 1.015–1.021, respectively). Nonsmoke PM2.5 was also significantly associated with hypertensive disease (HD) (6-day lag RR = 1.006, 95% CI = 1.003–1.008) and substance abuse (2-day lag RR = 1.009, 95% CI = 1.006–1.012). We observed null or negative associations between nonsmoke PM2.5 and cerebrovascular, heart, and anxiety-related diseases.

Conclusion: Our study indicates that wildfire smoke and nonsmoke PM2.5 have the greatest effects on different health outcomes. Wildfire smoke PM2.5 was most strongly associated with increased risk of schizophrenia and cerebrovascular disease, while nonsmoke PM2.5 was most strongly associated with all-respiratory, COPD, and hypertensive diseases. We also show significant and lasting effects of both smoke and nonsmoke PM2.5 on substance use disorders.


This thesis is restricted to Yale network users only. It will be made publicly available on 05/07/2025