Essays in Public and Health Economics

Abstract

This dissertation contains three chapters, each with the goal of providing evidence to better inform policymaking around challenges facing a different segment of the health care system in the United States. In Chapter 1, I study the home care industry for seniors with significant functional difficulties. This industry has seen dramatic growth over the past few decades, spurred in part by increased public financing through Medicaid. I investigate the effects of a policy that increased the use of formal home care among Medicaid-eligible seniors by more than 50%. I show that this expansion of formal home care primarily serves to replace informal care from family members, particularly spouses and daughters. For daughters, the decrease in care supplied is accompanied by an increase in labor supply: for every 2.4-3 women whose parent receives formal home care as a result of this policy, one additional daughter works full-time. In Chapter 2 (with Eric Barrette and Leemore Dafny). I investigate the effects of two policies intended to increase access to and quality of treatment for people with opioid use disorder (OUD): insurance parity requirements, and supply-side policy increasing the number of medication-assisted treatment (MAT) providers. Neither policy significantly increases the probability that patients with OUD receive any treatment, but both policies shift patients’ choices of treatment, in opposing directions. While parity causes substitution away from MAT, the standard of care, increases in the county-level share of physicians able to prescribe agonists are associated with substitution toward medication-assisted treatment. In Chapter 3, I study facility-level determinants of infection and mortality at nursing homes during the first wave of the COVID-19 pandemic. I find that staff neighborhood characteristics are strongly associated with facility deaths per bed: one standard deviation increases in average staff neighborhood population density, public transportation use, and non-white share were associated with 1.3, 1.4, and 0.9 additional deaths per 100 beds, respectively. These effect sizes exceed the effect of all facility management variables and even facility size, suggesting the importance of controlling staff community transmission to lessen the devastation in nursing homes.