Analysis of the Joint Distribution of Disproportionate Share Hospital Payments. Methodology


General Approach. To assess the validity of the hospital specific data, we compared total spending included in the CMS reports to DSH expenditure data reported by the states in the annual Financial Management Reports (HCFA-64). We also checked total expenditures from the hospital specific reports against the states' DSH payment limits established in the Balanced Budget Act (BBA) of 1997. To merge the Medicaid DSH payment data with our estimated Medicare DSH payments, we identified hospitals in the Medicaid reports using their Medicare provider numbers. Only two states, Michigan and North Carolina, put Medicare provider numbers on their hospital specific Medicaid DSH reports. We used the CMS On-line Survey and Certification Reporting System (OSCAR) and Provider of Service (POS) files, as well as AHA on-line Hospital Directory ( to match hospital names in the Medicaid reports with Medicare provider numbers.

Several states included a few individual hospitals in their Medicaid DSH reports that we could not identify with sufficient confidence to match them with their Medicare provider numbers. We created dummy Medicare provider numbers for these hospitals to keep them in the data set, but they could not be linked up with Medicare DSH payment information. In addition, eight states lacked hospital specific payment information for a much larger share of their Medicaid DSH payments.6 For these states, we created one dummy variable to account for the missing DSH payments to acute care facilities and a separate dummy variable to account for missing IMD DSH payments. Appendix A provides an explanation of state-specific reporting issues and how we handled them.

Estimating New DSH Funds. Unfortunately, no data are available on the precise amounts of net gains to hospitals from the Medicaid DSH program (new funds) in FY1998. We used estimates developed by Coughlin, Ku and Kim (2000) for FY1997 to construct possible measures of this parameter on the upper and lower bounds. We looked at four scenarios:

  1. All funds from the DSH program are new funds to the hospitals. While this is unlikely to be a correct assumption, it gives the upper possible bound on the amount of new funds. We assumed this measure equals our estimate of total DSH payments by the state.
  2. Only the federal share of DSH payments represents new money to facilities. To calculate this measure, we apply federal matching percentage to the DSH payments made by the state to each hospital.
  3. Only the portion of the federal share that was actually paid to hospitals represents new money that they receive. This scenario takes into account the possibility that states may retain some residual funds for their own use. We applied estimates of the percentages collected by states in residual funds in FY1997 uniformly to federal shares of all hospitals in the state. We recognize that the actual redistributions at the level of individual hospitals may be more complex and some may have received higher percentages of their federal shares than others. However, we have no basis for making other than an across-the-board estimate. For those states that were not covered by the survey conducted by Coughlin, Ku and Kim (2000), we assumed zero residual funds. The assumption seems to be reasonable, because even within the sample covered by the survey only 12 states out of 40 have state residual funds.
  4. Only the portion of the federal share that was actually paid to non-state hospitals represents new money to health care facilities in the state. This measure is an estimate of the lower possible limit on the new funds to hospitals. It takes into account the fact that DSH money may not add new funds to state hospitals because of possible offsetting diversion of other state financial assistance.

Actual Medicaid DSH Payments vs. Receivables, by State. In some cases, states make Medicaid DSH payments to facilities located in a different state. We assessed the magnitude of such payments and found that interstate DSH transfers are very small (Appendix B Table B.1). Overall, they constitute only 0.3% of the total FY1998 DSH payments. Because the funds reported as paid to hospitals located in another state are generally small, we assumed in our analyses that the DSH amounts paid by the states and amounts received by the facilities in the same state are equivalent.

6.  Alabama, Colorado, Georgia, Illinois, Indiana, Minnesota, New Jersey, and Pennsylvania.

View full report


"report.pdf" (pdf, 829.97Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®