Three Essays in Health Economics

Abstract

This dissertation contains three essays that explore how nursing homes respond to public reporting, changes in regulation and payments, as well as heterogeneity across organizational forms. More specifically, I focus on three important issues in the nursing home industry: public reporting, corporate chains, and financial performance. Having implications on the quality of care in the industry, these issues play important roles in shaping long-term care in the United States.

The first chapter evaluates the effects of simplified report cards for nursing homes on their financial performance. Starting in 2008, Centers for Medicare and Medicaid Services adopted a five-star quality rating system for nursing homes to provide consumers an easier way to compare facilities and encourage facilities to improve quality of care. Using a unique dataset containing the unpublished underlying quality scores and relevant thresholds used to assign the number of stars for each nursing home from 2008 to 2013, I implement a regression discontinuity design to estimate the effects of star ratings on nursing homes’ financial performance. I find that both revenues and costs of the 5-star nursing homes are significantly higher than those of nearly identical 4-star facilities. Upon further investigation, the main effects are due to a shift toward a more favorable payer mix. In addition, the effects are stronger in more competitive markets. Overall, the results suggest that nursing homes do not gain financially by getting an additional star under the new public reporting system.

In the second chapter, I examine how practice patterns of nursing homes vary by chain affiliation. In 1998, Medicare implemented a Prospective Payment System (PPS) for post-acute care provided by nursing homes. Primarily based on therapy minutes, the PPS created an incentive for nursing homes to provide high-intensity therapy services. Accounting for more than 50% of nursing homes, chains may choose to affect affiliated facilities’ practice patterns. This paper aims to examine whether the selection of therapy treatment levels for nursing home residents differs between independent and chain-acquired nursing homes from 2003 to 2009, using a Difference-in-Differences estimation approach. I find that independent facilities acquired by chains experience a 2.92 percentage point increase in the proportion of residents with the highest therapy treatment level. These chain effects may last years after acquisition. In addition, I find that the main effects are mostly due to acquisitions by large chains and for-profit chains. However, the increase in treatment intensity does not lead to a shorter length of stay.

In the third chapter, I investigate the predictors for chain acquisitions of independent nursing homes, focusing on the financial aspects of acquired targets. The purpose of this study is to understand the motivations behind chain acquisitions of independent nursing homes, with a focus on financial performance of the targeted facilities. I use a facility-year level discrete time logit model to predict the probability of an independent nursing home being acquired by a chain from 2000 to 2010. I find that chains are more likely to acquire nursing homes with worse financial performance. These findings raise important issues about the federal government’s efforts to make ownership information transparent and to set up an effective transaction monitoring mechanism for the nursing home industry.