Every year, hundreds of diabetic patients in the United States die because they were not able to afford their daily shots of life-sustaining insulin and chose to, out of sheer desperation, ration their insulin prescription so that it lasted for as many days as they could stretch it. For such individuals, uncompensated care that hospitals provide is their sole source of medical care.
Uncompensated care is the broad umbrella term for all the medical services for which the provider receives no financial compensation whatsoever. In general, hospitals face rising uncompensated care expenses. In 2019, hospitals reported 2.4 times the uncompensated care expenditures recorded in 1995, at $41.61 billion. Uncompensated care expenses strain the bottom lines of hospitals.
Simultaneously increasing is the trend of standalone hospitals joining forces with large established health systems in an attempt to maintain or bolster their financial condition. I sought to determine whether, and to what extent, the dollar amount of hospital uncompensated care would change after such standalone hospitals join health systems. Second, I wanted to assess if this relationship changes depending on the percentage of uninsured populations of states.
I found that hospitals affiliated with health systems provide more uncompensated care than those that are not, but joining, in itself, has no effect on the dollar amount of uncompensated care hospitals provide. This relationship does not change depending on a state’s percentage of uninsured population. Thus, community members relying on uncompensated care for all their medical needs remain unaffected despite hospitals more readily joining health systems.
The data I analyzed were derived from longitudinal datasets published by the Centers for Medicare and Medicaid Services (CMS) and the Dartmouth Atlas of Healthcare. Information regarding the amount of uncompensated care provided by NFP hospitals is collected every year by the Centers for Medicare and Medicaid Services (CMS) and published as the “Healthcare Cost Reporting Information System” (HCRIS) database. Data on the NFP hospitals that joined health systems, and which particular health systems they joined, were obtained from the dataset called “Hospital Research Data”, which is published by the Dartmouth Atlas of Healthcare. I considered the years 2014, 2015, 2016, and 2017 in my analysis. My investigation only considered not-for-profit (NFP) hospitals that operated as critical access hospitals or other acute care hospitals. I adjusted my uncompensated care amounts to account for fluctuations in inflation. I performed regression analysis to ascertain the relationship between NFP hospitals joining health systems and the subsequent change in the uncompensated care amounts they provide as a result of the consolidation.
I found that out of the total 2,702 hospitals observed, 1,788 were already part of health systems as of 2014, which is the starting time period for my consideration. In addition, 780 hospitals never joined a system by the end of my sample in 2017. These two groups served as control groups since their system affiliation did not change during the analysis period. A total of 57 hospitals joined a system in 2015, 38 in 2016, and 39 in 2017.
As illustrated by Figure 1, hospitals that were always in a system (joined in, or prior to, 2014) or joined a system during my analysis period provided significantly higher amounts of uncompensated care compared to similar NFP hospitals that never joined a health system, as of 2017. This suggests that the types of hospitals that are generally system members provide a greater amount of uncompensated care.
Table 1 shows average hospital uncompensated care without adjusting for hospital fixed effects. Hospitals in systems provided $5.1 million more in annual uncompensated care, and this difference is statistically significant. This indicates that hospitals in health systems do tend to provide more uncompensated care, consistent with the difference across groups in Figure 1. However, when the hospital fixed effects are included in Table 2 to determine whether uncompensated care changes when NFP hospitals join systems, I find no statistically insignificant results. That is, I cannot conclude with utmost certainty that system-belonging-NFP hospitals’ relatively high provision of uncompensated care can be attributed to them joining a health system.
Thus, joining a health system, in itself, does not result in an increased provision of uncompensated care but other factors exclusively correlated with either the independent (health system affiliation) or dependent (uncompensated care amount) variables could, in fact, cause this increase.
The second finding of my research is that there is no correlation between the change in uncompensated care spending of NFP hospitals after they joined health systems and the percentage of uninsured populations of the state in which a hospital is located.
The key takeaway from my research is that health system membership has no statistically significant impact on the amount of hospital uncompensated care provided.
My findings are reassuring, in that while uncompensated care provision does not increase, it does not decrease either because of hospitals joining health systems.
This means that uninsured community members may still be able to receive healthcare services and would likely not be turned away even if the hospital becomes a member of a health system.
This also means that the government will not be forced to intervene to provide a safety net program for those who can no longer access hospital uncompensated care.
Karunya Surendar is a senior at Indiana University.
Leave a Reply