United States: CMS Administrator Decision Upholds Medicare Program’s Calculation Of DSH Payments

On May 11, 2006, the Administrator ("Administrator") of the Centers for Medicare & Medicaid Services ("CMS") modified the decision issued by the Provider Reimbursement Review Board (the "PRRB") in Baystate Medical Center v. Mutual of Omaha Insurance Company. Earlier, in its decision issued on March 17, 2006, the PRRB held that there were numerous inaccuracies and mistakes in CMS’ calculation of the Medicare Disproportionate Share Hospital ("DSH") percentage for Baystate, and ordered the Medicare program to correct these mistakes. The Administrator, however, found that CMS used the "best available data," and that any errors in CMS’ calculations would not likely have a material effect on the hospital’s reimbursement.

The Administrator’s decision may have a major impact on the numerous other DSH appeals currently pending before the PRRB.

Under the Social Security Act (the "Act"), hospitals that serve a disproportionately high percentage of low income patients are entitled to receive DSH payments.

The most commonly used method for a hospital to qualify for the DSH payment is based on a complex statutory formula under which payment is based on the level of the hospital’s DSH patient percentage, i.e., the sum of two fractions (expressed as a percentage): the "Medicare fraction" and the "Medicaid fraction." The Medicare and Medicaid fractions are added together to determine if the hospital is entitled to receive DSH payments and, if so, the amount of such payments. The dispute at issue in the Baystate case involved the calculation of the hospital’s Medicare fraction.

The Medicare fraction is also often referred to as the "SSI fraction" because it captures the number of Medicare beneficiaries who are also eligible for Supplemental Security Income ("SSI") benefits under the Act. The Medicare fraction is computed by dividing the number of patient days for patients who were entitled to both Medicare Part A benefits and SSI benefits (the numerator of the Medicare fraction) by the total number of patient days for patients entitled to benefits under Medicare Part A (the denominator of the Medicare fraction).

In order to calculate the numerator of the Medicare fraction, CMS obtains a data file from the Social Security Administration ("SSA") that includes a list of eligible SSI recipients (the "SSI file"). CMS then matches information from the SSI file against its own Medicare Part A entitlement information (contained in the Medicare Provider Analysis and Review ["MEDPAR"] file) to determine the number of Medicare/ SSI days for a particular hospital in a particular federal fiscal year. The denominator of the fraction is calculated by CMS based on Medicare claims data. CMS then notifies the hospital and its fiscal intermediary of its calculation.

Baystate, located in Springfield, Massachusetts, filed an appeal with the PRRB challenging the accuracy of CMS’ calculation of the hospital’s Medicare fraction for the fiscal years ended September 30, 1993, 1994, 1995 and 1996. Baystate alleged that its Medicare fraction was understated because both the SSI file and the MEDPAR file contained inaccurate or incomplete information, and that the process of matching the data in these files was flawed for the following reasons:

1. CMS failed to use proper patient identifiers;
2. the SSI data file omitted various categories of individuals entitled to SSI benefits; and
3. the MEDPAR data either included Medicare inpatient data that should have been excluded, or excluded Medicare inpatient data that should have been included.

The PRRB held a six-day hearing in 2004, and issued its decision on March 17, 2006, concluding that the methodology used by CMS and the fiscal intermediary ("FI") to calculate the Medicare DSH percentage was inaccurate. The PRRB found that: (1) there is no Medicare statute or regulation that prevents the recalculation of the DSH percentage, and (2) the statute and regulations require CMS to calculate the DSH percentage in an accurate manner.

The PRRB found that the data used by CMS to calculate the DSH percentage was flawed and incomplete in many respects. In particular, the PRRB concluded that the process used by CMS to match the SSI eligibility data against the MEDPAR data was flawed, and that this process may deflate a hospital’s Medicare fraction and DSH percentage. Furthermore, it determined that the data used by CMS to calculate the Medicare fraction and DSH percentage was not the best available data. Specifically, the PRRB noted the following systemic errors:

• The omission of inactive or "stale" records (i.e., SSI records of recipients whose eligibility for benefits terminated prior to the time when the SSA transmitted its data to CMS, or SSI records that were deleted due to space limitations);

• The omission of "forced pay" cases when an individual appeared to be ineligible for SSI benefits according to automated payment records, but the individual actually received the payment manually; i The omission of hold and suspense cases, where an individual was entitled to SSI benefits but payment was temporarily withheld or suspended;

• The omission of retroactive awards when an individual became entitled to SSI benefits retroactively;

• The omission of non-cash benefits when an individual was not entitled to cash payments of SSI benefits but was, nonetheless, entitled to noncash SSI benefits;

• The failure to use multiple identifiers to identify SSI recipients and to match MEDPAR data against SSI eligibility data using the only unique identifier available, namely, the individual’s social security number; and

• The loss of SSI eligible days attributable to inappropriately matching MEDPAR data against SSI eligibility data using non-unique identifiers that often change during the year.

The PRRB also found that the SSI data obtained from the SSA contained historical errors that were within CMS’ control to correct. In particular, it found that at least as early as 1993, and until February, 1996, CMS knew that SSI data omitted stale records (e.g., deceased persons) and that CMS did little or nothing to fix this problem.

The PRRB found that Baystate was not required to quantify the financial impact of each of the identified flaws, nor was it required to show an exact number of incorrectly counted days. It concluded that the impact of the inaccuracies in the DSH calculation was likely to be significant; and it would not be significantly burdensome for CMS to redesign its computer programs used to calculate a hospital’s DSH percentage in a manner that would capture accurate information and accurately match SSI data with MEDPAR data.

The PRRB reversed the FI’s determination of Baystate’s DSH percentage and remanded the case to the FI to recalculate the hospital’s Medicare DSH percentage in a manner consistent with the PRRB’s decision.

The Administrator rejected the PRRB’s conclusion that the Medicare regulations allow for recalculation of the Medicare fraction based on updated or corrected data. Instead, the Administrator found that the regulations preclude the recalculation of the Medicare fraction based on updated or later data. The Administrator noted that CMS had made policy decisions balancing the need to reduce administrative burdens and the need for timely, accurate data, and that it concluded that the Medicare fraction was not subject to updating. The Administrator held that since the PRRB was bound by the regulations, it was not authorized to order any recalculation of the Medicare fraction based on updated or corrected data.

The Administrator further rejected the PRRB’s finding that CMS was required to conduct a precise calculation for the Medicare fraction and that the Medicare fraction could be corrected with later data. The Administrator noted that under the PRRB’s decision, the Medicare fraction would be always subject to recalculations based on later updated data, and there was a possibility that it would never be permanently determined.

Instead, the Administrator concluded that CMS’ calculation of the Medicare fraction only needed to be based on the "best available data," and that the calculation is acceptable if it is "reasonably accurate but not perfect." Furthermore, the Administrator stated that the policies and methods used by CMS resulted in "an acceptable rate of error, to the extent of being an almost nonexistent rate of error."

The Administrator then turned to the hospital’s several challenges to the data used by CMS in calculating the numerator and denominator of the Medicare fraction. At the PRRB hearing, the hospital had submitted substantial evidence intended to show that CMS did not use the best data in calculating the Medicare fraction.

The Administrator rejected the hospital’s argument that it was inappropriate for CMS to calculate the denominator of the hospital’s Medicare fraction based on MEDPAR data, rather than the Provider Statistical and Reimbursement Report ("PS&R") data. The Administrator noted that the preamble to the final regulations implementing the DSH adjustment, as well as later regulatory preambles, stated that the MEDPAR file should be used as the data source for the denominator.

The Administrator also rejected Baystate’s argument that there were systemic errors in the data used by CMS and in CMS’ matching process. The Administrator held that Baystate failed to prove: (1) that the data used by CMS in its calculation of the Medicare fraction was not the best data; and (2) that the hospital’s reimbursement was affected by these errors. Thus, the Administrator held that the hospital was not entitled to a recalculation of its Medicare fraction.

The Administrator also questioned the arrangement between Baystate and its Medicare consultants. The Administrator noted that the consultants had not been proffered as expert witnesses. However, they offered opinion testimony, similar to that which would have been offered by an expert, as to the effects of certain alleged CMS data flaws on the DSH percentage. The consultants had a 35% contingency fee arrangement and paid for all of the legal and consulting fees and expenses in relation to Baystate’s DSH appeals. The Administrator expressed the belief that such agreements are generally found to be against public policy and void, that the agreement had to be taken into consideration when evaluating the consultants’ credibility, and found that the PRRB failed to weigh the credibility of the consultants’ opinion testimony in light of the financial interest they had in the outcome of the litigation.

In light of the Administrator’s other findings, particularly that Baystate had not shown the existence of substantial data errors or that any of the alleged data errors had an effect on the Medicare fraction, it is unclear whether the Administrator’s comments on the agreement between the hospital and its consultants had any effect on his decision in this case.

In its decision, the PRRB also concluded that it would not be administratively burdensome for CMS to redesign its computer program to recalculate the Medicare fraction to correct the claimed errors with updated data. The Administrator, however, found that the PRRB had failed to consider the wider implications of its decision in this case, despite its knowledge of the cases presently pending before it on the DSH issue and the fact that the hospital had not been able to demonstrate any financial harm sufficient to justify relief. The Administrator also found that the PRRB had incorrectly assessed the administrative burden that would result from its ruling, and that the ruling could cause a significant, if not debilitating, disruption in CMS’s administration of the complex Medicare program.

Finally, the Administrator concluded that the PRRB’s standard, which the Administrator interpreted as requiring an errorless match of data in calculating the Medicare fraction, was inconsistent with the Medicare Prospective Payment System and the efficient administration of the Medicare program, and ruled that the hospital was not entitled to a recalculation of its Medicare fraction for the cost years in question.

The hospital has appealed the Administrator’s decision to federal district court. The ultimate decision in Baystate will likely have a significant impact on the numerous other individual hospital and group appeals pending before the PRRB on this and similar issues. Any hospital that receives DSH payments, as well as those "bubble" hospitals that are close to meeting the minimum threshold to receive DSH payments, should carefully consider the benefits of challenging CMS’ calculation of the hospital’s Medicare fraction and DSH percentage, and preserve its appeal rights pending the ultimate decision in Baystate.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.