Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)

Graduate Group

Health Care Management & Economics

First Advisor

Guy David


The first chapter examines the effect of Maryland’s Global Budget Revenue (GBR) payment system on hospital volume. Under this payment system, the rate regulating authority prospectively determines each hospital's revenue budget based on historic utilization to slow total hospital spending growth in the state. Theory suggests that hospitals will meet this target by reducing volume only when they have exhausted their ability to adjust reimbursement rates. Due to Maryland’s unique history of hospital payment, estimation of causal effects has been limited by availability of suitable comparison groups. This paper develops three measures of in state hospital-level exposure--hospitals with above median growth in predicted disease burden, hospitals with above median predicted growth in hospital inpatient service lines, and hospitals without revenue exclusions. For the two exposure measures based on growth, I apply a trend-break difference-in-difference under the a priori assumption that hospital volume would continue along pre-intervention trends in the absence of the payment system change. I find no evidence that the GBR is associated with any divergence of hospital inpatient volume from pre-intervention trends. The second chapter studies the effect of regional systems of care on health care delivery and outcomes for time critical illnesses of acute stroke and ST-elevation myocardial infarction (STEMI) in Philadelphia, Pennsylvania. Regionalization in Philadelphia trades off small increases in patient transportation time for improved access to appropriate treatment at the receiving hospital. This paper leverages the differential impact of regionalization on neighborhoods where the nearest hospital is not designated as a regional center of care. In these treated areas, emergency medical services (EMS) personnel bypass the nearest hospital to transport acute stroke and STEMI patients to the nearest regional center of care. Using a difference-in-differences empirical approach, this paper estimates the effect of regionalization on the probability of admission to a non-designated facility and a number of short-term utilization and outcome measures. Regionalization is effective in channeling volume to designated facilities, but the effects of regionalization are both condition and market specific.