The impact of market forces and public health insurance on inpatient care

Abstract

This dissertation considers both private and public-sector influences on inpatient care, focusing first on the impact of strategic hospital behavior on entry into new procedure markets, and second on the effect of changes in Medicaid and Medicare on hospital care. Chapter 1, "Entry Deterrence in Hospital Procedure Markets: A Simple Model of Learning-by-Doing," investigates whether incumbent hospitals threatened by entry in profitable procedure markets take advantage of learning-by-doing in these markets to erect barriers to entry. By focusing on incumbent behavior following a positive shock to the profitability of a procedure, and comparing this behavior across markets with different levels of entry-deterrence incentives, I am able to detect limited evidence consistent with entry deterrence through learning-by-doing in three case studies: electrophysiological studies, liver transplants, and prostatectomy. Chapter 2, "Does Public Insurance Improve the Efficiency of Medical Care? Medicaid Expansions and Child Hospitalizations," addresses the relationship between health insurance availability and the nature and frequency of hospitalization. Together with co-author Jonathan Gruber, I find that the Medicaid expansions from 1983 to 1996 were associated with a 22% decline in "avoidable hospitalizations," hospitalizations that can potentially be averted by timely outpatient care. However, the increased insurance coverage had a larger, offsetting impact on other types of hospitalizations, yielding a 10% overall increase in child hospitalizations.

(cont.) The effects on intensity of care once in the hospital are ambiguous, but the data show that more children were treated in for-profit facilities, and fewer in public institutions as a result of the expansions in Medicaid. Chapter 3, "Hospital Responses to Changes in Average Reimbursement Rates: An Assessment of a Natural Experiment," explores the effect of increased reimbursement to hospitals on billing practices (specifically, "upcoding") and intensity of care. Because the hospital industry is highly-regulated and predominantly not-for-profit, standard theories of firm behavior may not apply to hospitals, yielding ambiguous a priori predictions of hospital responses to reimbursement changes. My empirical analysis suggests that large increases in reimbursement for particular diagnoses were not met with increased spending on care for patients in those diagnoses. If upheld in future research, this finding has important implications for providers of health insurance, both public and private. Accounting for one-third of health expenditures, and over 4 percent of GDP overall, the hospital sector is critical both to healthcare and to the economy at large. Understanding hospital behavior will require additional investigation of competitive practices as well as public interventions.