Essays on Hospital Infrastructure Investment

Abstract

In this dissertation, I present three essays related to infrastructure investment by acute care hospitals. When hospitals use revenue to invest in their facilities, there is an opportunity cost in terms of staffing, technology, community benefits, and other possible forms of investment. While some level of improvement to facilities may be necessary due to depreciation, infrastructure investment is also a potential way to attract (or maintain) patient demand through signals of high quality and pleasant patient experience. Over $30 billion in construction is invested by hospitals annually in the United States, but little is known on whether this investment impacts demand and/or quality. I provide the first comprehensive look at hospital construction projects, using novel data from California. In Chapter 1, I provide a descriptive overview of significant construction projects. I identify nearly seventy major projects, defined as changing a hospital footprint through either complete rebuilding, additions to general acute care, or service-specific expansions. These projects range in cost from over $2.5 million to add capacity at a small community hospital to nearly $600 million for an urban hospital replacement. Hospitals undertaking these projects are more likely to have above-average volumes of discharges and are less likely to have for-profit ownership structure. They additionally have higher market share than facilities not undergoing a major construction project in my sample period. My results reveal that major infrastructure investment is a non-random occurrence and influenced by hospital and local conditions. I analyze the impact of major construction projects on demand for hospital care in Chapter 2. I perform a difference-in-difference analysis of admissions at the hospital and market level linking data from California’s state discharge dataset. At the hospital level, I find a significant increase in own-hospital admissions after a facility undergoes a major construction project. The impact is especially large among patients with private insurance, which is significant given these insurers typically offer the most generous reimbursement. At the market level, I find strong evidence that admission growth results from market expansion as opposed to purely business-stealing, though I cannot definitively rule out competitive effects. Preliminary evidence suggests market growth occurs particularly among elective procedures. These results suggest certificate-of-need law repeal may not have dramatic anti-competitive effects, but supplier-induced demand remains a concern. Finally, in Chapter 3, I look at the impact of major construction projects on the quality of care received. High quality care is both a policy goal and a potential justification for undergoing construction. I examine quality in terms of both patient experiences, measured using survey data, and clinical outcomes, measured using administrative data. I again employ difference-in-difference analysis for estimation. I find significant improvement across a variety of patient experience measures after a hospital construction project is completed. Among clinical measures, however, there is no significant impact. My results suggest that at least one dimension of quality improves because of construction. This finding raises equity concerns for improving quality at safety net hospitals, given their financial situation makes it more difficult for investment in infrastructure.