APSE conducted a statistical analysis to assess, at the state level, the association between the amount of uncompensated care (UCC) provided by hospitals in a state in a year and the number of individuals covered by Medicaid and the number of individuals who are uninsured. It then used the results of this analysis along with state-level projections of the numbers of uninsured and Medicaid-covered individuals to project how much lower UCC would be.
Hospital UCC in 2011 and 2012 was calculated from Hospital Cost Reports, as described in the main body of this report. These data were then aggregated to the state level in each year for use in the state-level analysis.
The numbers of individuals who were uninsured and who were covered by Medicaid in each state and in each year were obtained from estimates made by the U.S. Bureau of the Census.63 These estimates are based on survey data from the Current Population Survey.
To assess the association between UCC and the numbers of uninsured and Medicaid-covered individuals, we used the following panel-data model:
UCCs,t is the dollar amount of UCC provided by hospitals in state s and in year t (in billions of current dollars);
Medicaids,t is the number of individuals covered by the Medicaid program in state s and in year t (in millions);
Uninsureds,t is the number of individuals who were uninsured for the entire year in state s and in year t (in millions);
ϕs is a set of state fixed effects; and
θt is a set of year fixed effects.
The use of panel data and the inclusion of a set of state and year fixed effects is preferred over a simpler model assessing the cross-sectional association between UCC, Medicaid, and Uninsured because the latter model is more likely to be affected by omitted variable bias.
The results of this statistical model are reported in the Table below. The results indicate that a one-million increase in the number of people uninsured in a state is associated with a $0.344 billion increase in hospital UCC in that state. Similarly, a one-million increase in the number of people covered by Medicaid in a state is associated with a $0.292 billion decrease in hospital UCC in that state.
63 United States Census Bureau, online Health Insurance data. Available at http://www.census.gov/hhes/www/hlthins/