Dissertation Title: "Essays in Health Economics"This dissertation consists of three chapters that relate to the following broad areas in health economics: care provision for vulnerable populations, challenges in efficient insurance market functioning and the value of continuity of care.
Chapter 1: The Impact of Federally Qualified Health Centers on Youth Outcomes.
Health events that occur in youth such as adolescent pregnancy often have an enormous impacton adult outcomes. Adolescents are generally well-covered by health insurance but may not have access to care for other reasons. Using the large and staggered geographic expansion of Federally Qualified Health Centers (FQHCs) in the last two decades, this paper studies the impact of these community-based providers on teen pregnancy rates and high school graduation rates. Openings are associated with a 10 percent drop in the teen birth rates in an area. Declines are larger in counties with more than one opening and among low-income populations. I find no statistically significant effect on educational attainment overall but a 17 percent decline in the proportion of women who did not complete high school in areas where FQHC openings had the largest effects on the teen birth rate. These findings highlight the potential of community-based institutions to impact opportunities for youth.
Chapter 2: Physician Handoffs and Patient Mortality (with Prof Anupam Bapu Jena)
Transitions of patient care, or handoffs, have primarily been studied and associated with adverse events and errors among physician trainees. The relationship between handoffs and patient outcomes among physicians who have completed training is unknown, however, even though similar concerns apply outside the trainee setting. Using a 20% random sample of Medicare fee-for-service beneficiaries hospitalized during 2008-2012 and treated by a hospitalist, we analyzed whether patient 30-day mortality varied according to date of admission relative to the treating physician’s last working day. Admission towards the end of a physician’s shift block is predictive of a higher likelihood of handoff among otherwise similar patients, thus providing a setting for quasi-experimental analysis. We find that handoffs are associated with a 1.2 percentage point increase in 30-day mortality, with larger effects among high risk patients. This suggests a need for systematic measurement and evaluation of handoff processes within individual hospitals.
Chapter 3: The Extremely Under and Overcompensated in Individual Health Insurance Markets (with Prof Thomas McGuire)
This study seeks to investigate and describe the characteristics of individuals grossly underpaid and overpaid post risk adjustment and test the potential for adverse selection in the individual health insurance market. Using the 2016 HHS-HCC Risk Adjustment model software, we modeled risk-adjustment payments and analyzed residuals (the difference between payments and spending). Potential for supply-side and demand-side selection were explored through the measurement of the persistence of residuals and the correlation of individual expected spending and spending residuals, respectively. We found that the residual distribution is right-tailed but has a significant left tail as well and high spending variance HCCs are represented in both groups. Extreme residuals were highly persistent; the persistently underpaid spend disproportionately on specialty drugs while the persistently overpaid were frequently coded with transplant HCCs. The strong persistence of extreme residuals points to the potential for selection. Attention to the grossly over and underpaid individuals may lead to directions for improvement in plan payment systems. In addition, the role of specialty drugs in contributing to the persistence of residuals is worthy of more discussion.