THREE ESSAYS ON MARYLAND'S GLOBAL BUDGET REVENUE PROGRAM AND HOSPITAL-BASED NEONATAL CARE

Abstract

Health care spending is a major concern in the United States. State and federal governments have been engaged in a number of health care system reform initiatives designed to contain costs by regulating both price and quantity. Comprehensive evaluations of these initiatives are crucial for policymakers reshaping and expanding reforms. This dissertation evaluates the impact of Maryland's Global Budget Revenue (GBR) program, one of the most innovative statewide hospital payment reforms, on birth-related hospital utilization. The GBR program was designed to provide incentives for hospitals to reduce high-cost services and substitute them for lower-cost population health investments. This is largely accomplished by capitating annual budgets. This dissertation evaluated the effects of GBR on high-cost neonatal services, especially the neonatal intensive care unit (NICU). I examine heterogeneous treatment effects with respect to observable clinical needs and financial incentives. In Chapter One, I provide an overview of Maryland’s GBR program and introduce the conceptual framework. In Chapter Two, I examine the impact of GBR on NICU admissions and infant mortality. I explore the heterogeneity of treatment effects by infant health risk. Chapter Three expands the analysis to broader birth-related hospital services by investigating the impact of GBR on length of stay (LOS), the total cost of care, and utilization of specific high-cost services. Chapter Four departs from GBR and examines NICU utilization related to another critical source of financial incentive – health insurance type. Chapter Five concludes the dissertation. I find that Maryland's GBR program led to a substantial decline in NICU admissions, which was mainly driven by the decrease in admissions of relatively healthy infants, and there are no changes in the infant or neonatal mortality rate. The GBR program is also associated with declines in LOS and high-cost services used for infants. Finally, I observe that infant, maternal, and state characteristics explain the variations in NICU care across insurance type for high-risk infants but not for relatively low-risk infants. My findings provide positive evidence on implementing global hospital budget programs and shed light on the economic incentives affecting NICU care.