The Impact of Alternative Types of Post-Acute Medical Care: Stratified IV Analysis Using Nursing Home Exits
[Job Market Paper]
Approximately 40 percent of Medicare-covered inpatient hospitalizations are followed with post-acute care, mostly provided by home health services, nursing homes, and inpatient rehabilitation facilities. Despite substantial overlap in patient characteristics, these providers are reimbursed at very different rates according to expected care intensity, making the utilization of post-acute care a major driver of unexplained geographic variation in US health care spending. This paper compares the quality and cost of post-acute care across provider types, instrumenting with hospital-nursing home vertical disintegration to overcome selection bias. My results show that the termination of hospital-affiliated nursing homes increases the likelihood of both home health care and inpatient rehab, with the former a less intensive type of care and the latter a more intensive type of care. To address the violation of monotonicity in this multiple treatment setting, I propose stratified instrumental variables based on conditional first stage effects estimated with machine learning. My results demonstrate that compliers who switch from nursing homes to home care experience no change in health outcomes or other health care treatments, while those who switch from nursing homes to inpatient rehab cost Medicare twice as much with no improvement in health outcomes.
The Economic Impact of Healthcare Quality
IMF Working Paper No. 19/173, August 2019
We study the costs of hospitalizations on patients' earnings and labor supply, using the universe of hospital admissions in Denmark and full-population tax data. We evaluate the quality of treatment based on its ability to mitigate the labor market consequences of a given diagnosis and propose a new measure of hospital quality, the "Adjusted Earning Losses" (AEL). We document a sizeable heterogeneity in quality across Danish hospitals: AEL standard deviation is equal to approximately 10% of the average decline in labor earnings following a hospital admission. We show that AEL contains significant additional information relative to traditional measures and does not suffer from worse selection issues. We also document a large decline in the labor cost of hospitalizations over time, with large variations across diseases. We find that the average post-hospitalization reduction in labor earnings declined by 25% (50%) on the intensive (extensive) margin between 1998 and 2012.
Work in Progress
The Impact of Private Contracting on Health Care for the Disabled: Evidence from California's Medicaid Program
We study the impact of Medicaid Managed Care (MMC) on the health and utilization of the disabled beneficiaries. We examine a policy change in Medi-Cal (the Medicaid program in California) which mandates senior and disabled Medicaid beneficiaries to enroll in Medicaid Managed Care in a selection of counties where they previously could voluntarily choose whether or not to enroll in MMC. Our main empirical strategy is to exploit the fact that the policy change was implemented according to the birth month of beneficiaries from June 2011 through May 2012. Our main source of data is the administrative records of all inpatient hospital discharges and ER visits in California from 2009 through 2014. Overall, moving the seniors and persons with disabilities (SPD) from FFS to MMC appears to have caused disruption in access to care that manifests itself in an increase in use of the emergency room. For the entire SPD population, we also find a transitory increase in inpatient visits that are non-ER transfer from another hospital and a decline in visits that are non-ER and non-transfer. To examine heterogeneity, we look at individuals who were heavy utilizers of healthcare services prior to the reform. Overall, we find that increase in the number of outpatient ER visits was driven by individuals whose total number of inpatient hospitalizations and ER visits prior to the reform were above the median. We also observe an increase in mortality rates mainly concentrated among the heavy utilizers, suggesting that they may be particularly vulnerable to disruptions in care and shirking of managed care organizations.
Who Makes an Active Choice? Testing Models of Default Effects in Medicare Part D
We investigate the nature of active decision-making in the context of low-income ("dual-eligible") beneficiaries in Medicare Part D. In this setting, beneficiaries of Part D who receive low-income subsidies and do not make an active insurance plan choice when they qualify for Medicare are randomly assigned to a plan. The majority of beneficiaries fail to do so, making an understanding of this policy important for understanding the dual-eligible market. We combine administrative data from Medicare with an administrative measure of whether a beneficiary was enrolled through this process, which we can use as a measure of active decision-making that was largely unobserved in prior literature. We augment this data with individually-linked longitudinal administrative Medicaid data that predates beneficiaries' enrollment in Medicare. We use this rich data to measure the determinants of active choice behavior and find that the highest-risk beneficiaries make active choices at only slightly higher rates than those with lowest-risk, suggesting that the dearth of active choice is not as "rational" as is assumed in typical models. We further investigate the persistence of "inactive choice", including how it responds to health shocks, as well as other market and personal determinants. Our results highlight the difficulty beneficiaries have in navigating complex health insurance markets, and show that the downsides of providing social insurance through privatized market systems may be worse than existing theory would suggest.