Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)




Deborah Freund

Subject Categories

Economics | Public Policy


Chapter 1

Low-income, publicly insured admissions historically cost more to treat than the average patient. To ensure that hospitals are reimbursed an adequate amount for care of indigent populations, Medicare reimburses hospitals an additional percentage amount according to federally set financial schedule. The reimbursement cutoff is discrete: at fifteen percent of a disproportionate patient percentage, a hospital is reimbursed an extra 2.5 percent of the standard prospective payment rate. I extend a simple model of hospital quality as a function of insurance reimbursement increases to determine that under certain circumstances there exists a positive relationship between quality and reimbursement. I use Hospital Consumer Assessment of Healthcare Providers and Systems data to analyze hospital ratings around the fifteen percent disproportionate patient percentage cutoff and find that on average, hospital ratings increase by six percentage points. When a subsample of non-profit hospitals is analyzed, hospital ratings increase by an average of 6.5 percentage points, primarily driven by patient facility cleanliness and medical provider communication ratings.

Chapter 2

The Center for Medicare and Medicaid Services (CMS) created the Hospital Compare Program in 2003 to increase transparency between health care providers and consumers. Implemented in 2005, this transparency consists of hospitals' collecting and making publicly available a set of hospital quality score measures. The CMS induced participation by financially penalizing hospitals that did not publicly report a specific subset of these measures (called "starter" measures). Three years into the program, the penalty for non-reporting both the starter measures and other ("non-starter") measures was increased. I use a difference-in-differences methodology to analyze the effect of the increased CMS penalty on the likelihood that a hospital publicly reported its starter and non-starter measure scores. I find that the penalty had an economically and statistically insignificant effect on the probability that a hospital publicly reported its starter scores but a statistically significant eight percent effect (p < 0.01) on whether it reported its non-starter scores. These findings are robust to a series of alternative empirical specifications.

Chapter 3

In 2006, Massachusetts passed a health care reform which required individuals to purchase health insurance and provided subsidized health insurance to the poor. The reform greatly increased the proportion of the state population that was insured. In this study we use a large data set of private health insurance claims to analyze the effect of the increase in the number of insured on physician reimbursement. We find that reimbursement for well-infant visits rose by approximately 4 percent during the reform implementation period, but the increase did not persist. Reimbursement for well-adult visits and appendectomies remained unchanged. Triple difference estimates using appendectomies (for which demand is extremely inelastic) as an additional control group show a 2 percent rise in well-infant visit reimbursement during the implementation period and no effect afterwards or on well-adult visit reimbursement. Estimates imply a temporary increase in the cost of health services with relatively elastic demand following a large scale insurance mandate, such as the Affordable Care Act.


Open Access