This dissertation is composed of three essays that consider the determinants and persistence of health care spending and how policies that control increasing health care costs affect the distribution of health care spending in the U.S. In the first essay, I study the association between education and health care spending for a set of health conditions amenable to self-management. Empirical findings from estimated health expenditure models reveal strong inverse relationships between education and health care spending among elderly adults with hypertension and/or asthma. Additionally, I find that greater educational attainment is associated with a reduced likelihood of being in the top 5% of health care spenders for elderly adults with hypertension and nonelderly adults with diabetes, and also with less severe conditions. The second essay assesses how the distribution of family out-of-pocket health care spending has been affected by changes in recent cost-sharing to understand the effectiveness of the risk protection function of private health insurance against high medical care expenses. The results suggest that families who rely more on health care because of one or more their member's existing health conditions are most affected by changes in cost sharing during the period 2001-2005 and the increased exposure to out-of-pocket spending occurrs primarily for families at higher percentiles of the out-of-pocket spending distribution, thus reducing the "return" to risk protection from holding private health insurance. The final essay examines the dynamics of out-of-pocket health care spending by looking at the persistence of such spending among Medicare beneficiaries. The findings suggest that having a certain chronic condition or a health shock clearly increases the probability of out-of-pocket health care spending persistence. Additionally, having an existing health insurance that supplements Medicare coverage or the acquisition of a new supplementary health insurance has a significant impact on the probability of persistence.

The United States spends a larger percentage of GDP on healthcare than any other OECD nation, and yet it performs poorly on measures of access, process quality and outcomes relative to other wealthy nations. One hypothesis that may explain the relatively poor performance of the United States' healthcare system per dollar spent is that markets for hospitals, physician services, and insurance are highly consolidated and lack competition. This dissertation contributes to the body of literature seeking to measure healthcare market structure and assess the relationship between this market structure and unit prices, overall spending, and healthcare quality. The first paper in this dissertation, "The Association between Hospital-Physician Vertical Integration and Outpatient Physician Prices Paid by Commercial Insurers: New Evidence," demonstrates market-level associations between vertical integration among hospitals and physicians and higher prices for outpatient care. The second essay, "How Hospital-Owned Physician Organizations are Associated with Healthcare Prices, Expenditures, and Quality," analyzes vertical integration and prices, quality, and total patient spending, building on the findings of the first essay using physician and hospital level claims data. This study finds that vertical integration is associated with higher levels of annual spending for patients attributed to vertically integrated providers, but interpretation of association between vertical integration and outpatient prices is confounded by pre-intervention trends. Analysis of CMS hospital outcome measures does not show associations between vertical integration at a hospital level and quality. Finally, the third essay, "Automated Delineation of Hospital Market Boundaries in California," explores geographic market definition in healthcare, an important topic in research and antitrust action, while assessing the application of community detection methods in this field. This study finds that community detection methods group hospitals with higher accuracy than other geographic markets as measured by patient flows and may offer promise for merger screening, research on market concentration, and research on geographic variation in healthcare. The final chapter reviews overarching limitations of the dissertation, outlines directions for future research, and comments on potential policy approaches to promote competition and address the symptoms that may result from highly consolidated healthcare markets

As medical care becomes an increasingly large share of Gross Domestic Product, understanding the mechanisms for how and why medical care spending is rising becomes increasingly important. Such an evaluation should consider the productivity relationship between medical care and health. An evaluation of medical productivity involves the measurement of medical care input prices, disease treatment output prices, and the productive relationship between medical care inputs and disease treatment health outcomes. Medical care price measurement is complicated by the heterogeneity of services, the role of insurance in negotiating prices, rapid technological advancements in medical care and limited availability of transaction price data. Health outcome prices are difficult to construct because of the difficulty in measuring health outcomes, the heterogeneity of health outcomes, and the messy relationship between consumption goods and health. Finally, in addition to accurate input and output price measurement, a productivity assessment requires a measurable causal relationship between medical care services and health outcomes. To date, all of these requirements have been insurmountable hurdles to assessing the productivity of medical care for the entire United States economy. This dissertation uses the Medical care Expenditure Panel Survey to address the necessary requirements for evaluating the productivity of medical care. The second chapter constructs regional medical care price indices using transaction prices that control for service type heterogeneity. The data employed in the analysis associates the observed medical care spending with the diseases the spending is used to treat. This association is exploited in the third chapter, which constructs medical care treatment prices for twelve of the major health conditions in the United States. The fourth chapter compares the productivity of medical care services used to produce disease treatment health outcomes across insurance types.

Patients treated at in-network facilities can involuntarily receive services from out-of-network professionals, which may result in a "surprise bill." As of June 2019, fewer than half of states protect patients from surprise out-of-network medical bills, and there are no federal policies enacted to protect patients. Moreover, payment for out-of-network medical services contribute to rising health care costs in the United States. This dissertation is comprised of three essays addressing surprise out-of-network medical bills and out-of-network health care provider payment. The first essay quantifies the prevalence and magnitude of potential surprise medical bills in ambulatory surgery centers (ASCs) and describes the characteristics of providers and health plans involved. This observational study of commercial claims identifies possible surprise out-of-network bills in one-in-twelve ASC episodes. These potential bills average $1,100 per episode and are predominately generated by anesthesiologists, registered nurse anesthetists, and independent laboratories. These findings indicate that consumer protection policies are needed to address surprise out-of-network billing in ASCs. The second essay examines the early effects of California's recent policy addressing surprise medical billing (AB-72) on the dynamics among physician, hospital, and insurer stakeholders. This case study identifies that an out-of-network payment standard set at payer-specific local average commercial negotiated rates has changed the negotiation dynamics between hospital-based physicians and payers. Leverage has shifted in favor of payers, and physicians reported that this experience of decreased leverage is exacerbating provider consolidation. Thus, this study finds that out-of-network payment limits can influence payer-provider bargaining. The third essay projects the potential impacts of an out-of-network hospital payment limit on negotiated in-network payments by private health plans. This study estimates the effects of three proposed out-of-network payment limits for hospital care - 80% of billed charges, average private prices, and 125% of Medicare - on negotiated in-network prices and total payments for hospital care in 2017. The results suggest that a strict out-of-network payment limit set at 125% of Medicare could achieve reductions in hospital payments similar to more drastic reforms, such as Medicare for All and public plan buy-in programs. This dissertation demonstrates that policies to address surprise out-of-network billing must be comprehensive in the scope of services, settings, and patient populations they cover to effectively protect patients. It also demonstrates that policies to address out-of-network billing impact the underlying contracting dynamics of the health care market and can influence the amount paid to providers both out-of-network and in-network.