This dissertation includes four chapters in environmental and health economics. In the first chapter, I study the impact of coal plant closures on Medicare Part D purchases. Much of the existing literature on the impacts of air pollution on health focus on morbidity or severe outcomes. A small and newer literature has focused on morbidity and less severe outcomes through the analysis of prescriptions. I add to this literature by showing that a coal plant closure leads to a -21.9 days decrease in prescription pills per month, a -12.4% decrease in total cost, and -$8.31 decrease in out-of-pocket costs per beneficiary per month for those living within 50 miles of the coal plant closure. My analysis also focuses on longer-term impacts as compared to most papers that analyze daily or weekly variation in air pollution. Studying longer-term impacts allows us to study how individuals may adapt to more apparent changes in pollution. In the second chapter, my coauthors and I study how children’s mental health was impacted by school closures during the COVID-19 pandemic. During the 2020-2021 academic year, school districts across the United States differed substantially in the number of in-person school days. Using the staggered nature of these openings, we estimate the impact of school reopenings on children’s mental health diagnoses and expenditures on mental health related diagnoses. Statewide variation in California in the amount of in-person learning created a unique quasi-experiment to study this important question. Using a difference-in-differences analysis, we find a 1.2 percentage point drop in monthly prevalence of mental health diagnoses (95%CI: -1.59 to -0.74), and a 10.6% (95%CI: -13.4%, -7.8%) drop in related healthcare spending related to the reopening of a school. We find that anxiety and depression comprise most of the differential in diagnoses between students residing in reopened and still closed districts. These effects are especially strong among girls.
In the third chapter, my coauthors and I study the association between wildfire smoke and asthma related utilization among children in California. This study investigates the effect of wildfire PM2.5 on asthma further by documenting the effect of PM2.5 on inhaler, inpatient, and outpatient utilization. We find that the cost of inhalers, inpatient, and outpatient utilization account for more than half of asthma costs related to wildfire smoke exposure. Using claims data from a large health plan, we construct a sample of Census tract by day utilization rates for asthma related outcomes and combine this with data on PM2.5 associated with wildfire smoke. Our sample consists of children with commercial health insurance who resided in California between January 2013 and December 2020. We utilize Poisson regression models with linear and binned measures of wildfire smoke exposure and contemporaneous and lagged exposure variables. We find that an additional 10 µg/m3 of PM2.5 is associated with a 1.44% increase in asthma related outpatient utilization, a 4.1% increase in inhaler fills, and a 5.4% increase in asthma related ED utilization. This equates to an average cost between $338 per 100,000 individuals for each additional 10 µg/m3 of PM2.5 of wildfire smoke from all 5 clinical settings.
Lastly in the fourth chapter, my coauthors and I study the association between wildfire smoke exposure and susceptibility to contracting COVID-19. In this paper, we determine the association between past wildfire smoke exposure and the susceptibility to and severity of COVID-19 cases in the months following exposure to smoke. The study population consists of commercially and Medicaid-insured individuals residing in California from June 2020 to January 2021. We find that exposure to an above median number of wildfire smoke days is associated with a 0.60 percentage point (pp) (95% CI, 0.20 pp to 0.099 pp; p= 0.003) increase in the likelihood of having a positive COVID-19 lab test or diagnosis relative to a mean of 2.73% and a 0.072 pp (95% CI, 0.058 pp to 0.086 pp; p< 0.001) increase in the likelihood of a COVID-19 related ED visit relative to a mean of 0.33%. These results add to the growing literature showing the detrimental effects of wildfire smoke on human health and display the need for mitigation procedures and support for at-risk individuals.