SUBJECT: New Hampshire Department of
Health and Human Services

DECISION

The New Hampshire Department of Health and Human Services (New Hampshire) appealed the determination of the Centers for Medicare & Medicaid Services (CMS) disallowing $30,327,984 in Medicaid funding that New Hampshire claimed for payments made in March 2002 to county-owned nursing homes for services rendered during the state's fiscal year 2000. New Hampshire made the payments under a provision in its state Medicaid plan that permits reimbursement of county nursing homes in excess of the rates established in New Hampshire's state plan, on the basis that the nursing homes serve a high percentage of patients with medically intensive needs. CMS disallowed New Hampshire's claim on the grounds that the payments were not made in accordance with New Hampshire's approved state plan.

We find that New Hampshire made the disallowed payments pursuant to a new calculation methodology that was not consistent with its state plan and which dramatically increased the total amount paid to the counties, most of which was returned to the state. Under the circumstances before us, New Hampshire's interpretation of its plan to permit the new methodology was not reasonable, and we thus sustain the disallowance.

Law and regulations

General Medicaid provisions

Title XIX of the Social Security Act (Act), known as Medicaid, provides for joint federal and state financing of medical assistance for certain needy persons. See also 42 C.F.R. § 430.0. States which establish a Medicaid program are required to submit a state plan meeting all federal requirements. Section 1902 of the Act. To receive federal financial participation (FFP), a state must claim the costs of medical assistance in accordance with its approved Medicaid state plan. Section 1903(a) of the Act; 42 C.F.R. § 430.10. "The State plan contains all information necessary for [CMS] to determine whether the plan can be approved to serve as the basis for Federal financial participation (FFP) in the State program." 42 C.F.R. § 430.10; see also Virginia Dept. of Medical Assistance Services, DAB No. 1838 (2002). The plan must provide that it will be amended whenever necessary to reflect changes in federal law. 42 C.F.R. 430.12(c); New Jersey Dept. of Human Services, DAB No. 1652 (1998). FFP is available in state Medicaid payments to a long term care facility that has a valid provider agreement in effect. California Dept. of Health Services, DAB No. 1732 (2000).

CMS regulations on Medicaid reimbursement for inpatient hospital long term care facility services are at 42 C.F.R. Part 447, Subpart C. FFP is not available in a state's expenditures for such services that are in excess of the amounts allowable under this subpart. 42 C.F.R. § 447.257. The Medicaid state plan must comprehensively set out the methods and standards used by the state agency to set payment rates. 42 C.F.R. § 447.252. A state's Medicaid payments must be consistent with efficiency, economy, and quality of care. Section 1902(a)(30) of the Act; 42 C.F.R. § 447.250(b). The state must make assurances, satisfactory to CMS, that certain requirements are met. 42 C.F.R. § 447.253(a). Section 447.253(b)(2) provides that one of the assurances the state must make is that the proposed payment rate will not exceed the "upper payment limits" specified in section 447.272. The upper payment limit for each group of health care facilities (i.e., hospitals, nursing facilities and intermediate care facilities) is the aggregate amount that can be reasonably estimated would have been paid to that group of facilities for those services under Medicare payment principles. 42 C.F.R. § 447.272. Additionally, aggregate payments to each group of state-operated facilities may not exceed the amount that can reasonably be estimated would have been paid under Medicare payment principles. 42 C.F.R. § 447.272; Oklahoma Dept. of Human Services, DAB No. 1575 (1996). The federal share of payments made at a rate higher than authorized in the approved state plan is considered an overpayment subject to recovery by CMS. Section 1903(d) of the Act; California, DAB No. 1474; Missouri Dept. of Social Services, DAB No. 1229 (1991); California Dept. of Health Services, DAB No. 1007 (1989); Ohio Dept. of Public Welfare, DAB No. 637 (1985).

Generally, the Board gives deference to a state's interpretation of its own state plan, so long as that interpretation is an official interpretation and is reasonable in light of the language of the plan as a whole and the applicable federal requirements. Missouri Dept. of Social Services, DAB No. 1412 (1993); South Dakota Dept. of Social Services, DAB No. 934 (1988). However, the Board has held that states must follow the methods and standards set out in their Medicaid plans, and may not change their plans unilaterally. California, DAB No. 1474; Massachusetts Dept. of Public Welfare, DAB No. 853 (1987).

Background on the upper payment limits

New Hampshire made the disallowed payments pursuant to state plan provisions permitting payment to county facilities above Medicaid rates, but within the upper payment limits. Some background on the upper payment limits and their application by states is helpful to our later analysis.

As noted above, a state's aggregate Medicaid payments to each group of health care facilities (i.e., hospitals, nursing facilities and intermediate care facilities) for inpatient hospital and long term care facility services may not exceed the amount that can reasonably be estimated would have been paid for those services under Medicare principles. 42 C.F.R. §§ 447.272, 447.253(b)(2). In addition, the version of section 447.272 in effect during FY 2000, the time period for which New Hampshire's claim applies, required that aggregate payments to each group of state-operated facilities also not exceed the amount that can reasonably be estimated would have been paid for those services under Medicare payment principles. This additional limitation was intended to address the practice by which some states differentially applied their payment systems to state-owned facilities so that they could pay (and claim FFP in) excessive Medicaid payments to those facilities above their actual costs, without exceeding, in the aggregate, the upper payment limit for all such facilities, state-owned and private. 52 Fed. Reg. 28,141, 28,144-45 (1987). To correct this situation, CMS in 1987 revised the upper payment limits regulation to require that state-operated facilities be grouped separately for purpose of determining the aggregate upper payment limits. 52 Fed. Reg. 28,141, 28,144-45, 28,147 (1987).

However, requiring separate aggregation of payments to state-owned and non-state-owned facilities did not alleviate the problem CMS cited in revising the regulation. Instead, CMS found states continuing to claim excessive reimbursement without exceeding the aggregate upper payment limits, because private facilities continued to be grouped with non-state governmental facilities (i.e., county-owned facilities) for the purpose of determining the limits. As CMS explained,

By developing a payment methodology that sets rates for proprietary and nonprofit facilities at lower levels, States can set rates for county or city facilities at substantially higher levels and still comply with the current aggregate upper payment limit. The Federal government matches these higher payment rates to public facilities. Because these facilities are public entities, funds to cover the State share may be transferred from those facilities (or the local government units that operate them) to the State, thus generating increased Federal funding with no net increase in State expenditures. This is not consistent with the intent of statutory requirements that Medicaid payments be economical and efficient.

65 Fed. Reg. 60,151, 60,152 (2000). This situation, CMS determined, "creates a financial incentive for States to overpay non-State-operated government facilities because States, counties, cities and/or public providers can, through this practice, lower current State or local spending and/or gain extra Federal matching payments." Id.

Accordingly, in January 2001, CMS again revised section 447.272, this time to establish separate aggregate upper payment limits for non-state owned or operated facilities. 66 Fed. Reg. 3147, 3175-76 (January 12, 2001). In the preamble to the proposed revision, CMS cited findings and testimony before Congress by the Office of the Inspector General and the General Accounting Office, to the effect that:

The States did not base the enhanced payments on the actual cost of providing services or increasing the quality of care to the Medicaid residents of the targeted nursing facilities.

The counties involved in the enhanced payment process used little or none of the enhanced payments to provide services to Medicaid residents. Instead, the counties returned these funds to their original source. That is, the funds were returned to the State's general funds or used to repay loans that were made to initiate the transaction, or both.

The States were clear winners in that they were able to reduce their share of Medicaid costs and cause the Federal government to pay significantly more than it should for the same volume and level of Medicaid services. The Federal share of the enhanced funding went into State accounts and, in some cases, could be used for any purpose.

Some States effectively recycled the Federal funds received from these enhanced payments to generate additional Federal matching funds.

Similarly, the GAO testified that current arrangements violate the basic integrity of Medicaid as a joint Federal/State program. By taking advantage of a technicality, these financing schemes allow States, in effect, to replace State Medicaid dollars with Federal Medicaid dollars.

65 Fed. Reg. 60,152-53 (October 10, 2000). The new restriction was to be phased in through a series of transition periods. 66 Fed. Reg. 3176 (January 12, 2001). CMS subsequently extended the transition period applicable to aggregate payments to long term care facilities. 66 Fed. Reg. 17,657 (April 3, 2001); 66 Fed. Reg. 46,397 (September 5, 2001). Before the Board, CMS provided a September 2000 GAO statement before Congress and an October 2001 GAO report detailing the history of these practices and their deleterious effects on Medicaid financing. CMS Exhibits (Exs.) 6, 7.

Background of the disputed payments, and the parties' arguments

New Hampshire's state Medicaid plan permits New Hampshire to reimburse "non-State operated governmental nursing facilities" (i.e., county-owned facilities) for long term care services at levels above the Medicaid per diem rates specified in its state Medicaid plan. The basis for the additional reimbursement, which began in 1994 and is called the "proportionate share incentive adjustment" or "ProShare" payment, is that county nursing facilities "provide care to many severely medically involved patients requiring an extraordinarily intensive and costly level of care." State Plan, Att. 4.19D, § 9999.8.e, TNs 96-6, 94-7; New Hampshire Exs. F, Y; CMS Ex. 8. The amount available for ProShare payments is effectively the difference between the amount that the county facilities receive under the plan's Medicaid rate, and the amount they would receive using Medicare payment principles, subject to the Medicare aggregate upper payment limit. ⁽¹⁾ At issue in this appeal are the methodology used to calculate the Medicare payment amount, and, to a lesser extent, the timing of the ProShare payment.

Initially, the state plan provided for facilities to each receive a single ProShare payment, to be made in June, the end of the state's fiscal year. New Hampshire Exs. D, Y. In 1996, with CMS's approval, New Hampshire amended its state plan to permit each year's ProShare payment to be made in two payments, "an interim payment in March or April, and a final payment in June of each State Fiscal Year." TN 96-07; New Hampshire Ex. F; CMS Ex. 8. According to New Hampshire, the two-payment system was adopted because counties desired to have access to the ProShare funds earlier than the end of the fiscal year. New Hampshire Ex. B, ¶ 4. The plan amendment provided for the ProShare payments to be determined as follows:

The interim payment shall be based upon the estimated Medicare Cost Limit as estimated by the fiscal intermediary. The payment to each facility is in proportion to the facility's Medicaid days, during the cost reporting period used to set its current rates, relative to the sum of Medicaid days for all eligible facilities. The total funds for incentive adjustment payments shall be established each State Fiscal Year subject to the anticipated level of all nursing facility payments within the year and to the payment limits of 42 CFR 447.272. The calculation of the payment limit of 42 CFR 447.272 was computed as follows:

The calculation was a comparison of the rates Medicare would pay for these facilities using the Medicare ratesetting methodology and the rates New Hampshire Medicaid actually is paying these facilities. The variance in the per diem rates was calculated per facility, an average variance determined and that variance applied to the total Medicaid days.

To determine the Medicare rate, the factors used by Medicare were determined and applied:

• Location (by county) in one of the two MSA's [metropolitan statistical areas] or the Non-MSA area;



• Inflation factor as supplied by the NH/VT Medicare fiscal intermediary; and



• The Medicare Cost Limit or the estimate of the Limit if that is what the fiscal intermediary provides (including labor, & non-labor components, wage index and Per Diem Add-on) as supplied by the NH/VT Medicare fiscal intermediary.

To allow comparison of like costs, to this rate was added the actual capital cost and ancillary cost per patient day as determined by Medicaid ratesetting. The variance between that calculated total Medicare rate and the actual New Hampshire Medicaid rate was then calculated as a per diem, and an aggregate variance calculated.

State Plan Att. 4.19D, § 9999.8.e.3, TN 96-6; New Hampshire Ex. F; CMS Ex. 8.

Under New Hampshire law, the following portions of the aggregate ProShare funds that counties receive for their nursing facilities must be returned to the state through intergovernmental transfers: 50% of the first $12.5 million, 100% of the second $12.5 million, and 75% of the remainder (amounts over $25 million). New Hampshire Rev. Stat. Ann. § 167:18h (2002) (eff. 1998); CMS Ex. 4.

The ProShare payments for FY 2000

In June 2000, New Hampshire made aggregate ProShare payments to its counties for its state FY 2000 of $20 million, of which the counties transferred $13.75 million back to the state. ⁽²⁾ State correspondence indicates that both transfers were to occur on the same date. ⁽³⁾ CMS Ex. 3.

At issue here, however, are ProShare payments made on March 27, 2002 for state FY 2000 of approximately $60 million, in which New Hampshire claimed 50% FFP, leading to the disallowance and this appeal.

The March 2002 payments were made because New Hampshire recalculated the amount of the FY 2000 ProShare payment using a new Medicare payment methodology that Congress adopted for skilled nursing facility services in the Balanced Budget Act of 1997, Public Law No. 105-33, 111 Stat. 251. Previously, nursing facilities were paid for Medicare services according to a retrospective, reasonable cost-based system. The new law utilized a prospective payment system, which provides for payment in accordance with prospectively determined per diem rates. The rates are derived from classifying SNF residents into 44 resource utilization groups (RUGs) that account for the relative resource use of different patient types. See 63 Fed. Reg. 26,252-316 (1998).

New Hampshire indicated that the effort to recalculate the FY 2000 ProShare payments was prompted by CMS's request that New Hampshire recalculate the ProShare payments for FYs 1998 and 1999, using more up-to-date, current-year data regarding the actual costs that nursing homes had incurred during those fiscal years, instead of estimated or prior year bed or patient days, as had been the case previously. New Hampshire Ex. B, ¶ 5; CMS Ex. 2, ¶ 10. These calculations (which did not employ the newer Medicare RUGs methodology) resulted in the determination that New Hampshire had overpaid its counties for FYs 1998 and 1999 by approximately $2 million and $1.17 million, respectively. New Hampshire Ex. I. New Hampshire agreed to adjust its claims for FFP to refund the federal share of the overpayments.

Beginning in May 2001 New Hampshire representatives worked with CMS officials to recalculate the FY 2000 ProShare payments utilizing the Medicare RUGs methodology; by August 2001, New Hampshire had determined that its counties were entitled to further ProShare payments of $60,655,969, in addition to the $20 million paid to the counties in June 2000. New Hampshire Ex. I; CMS Br. at 3. Adjusting for the ProShare overpayments for FYs 1998 and 1999, New Hampshire determined that $57,476,204 was due to the counties. Id. New Hampshire requested a supplemental grant award of 50% FFP in correspondence to CMS dated February 21, 2002. By letter dated February 26, 2002, the CMS Associate Regional Administrator advised that he was recommending disapproval of New Hampshire's request. In a letter dated June 26, 2002, CMS disallowed $30,327,984 (50% of $60,655,969) that New Hampshire claimed in a quarterly expenditure report submitted on April 29, 2002. ⁽⁴⁾ New Hampshire Exs. N, O, P.

CMS initially disallowed New Hampshire's claim for FFP in the March 2002 ProShare payments on the grounds that it violated New Hampshire's state plan, which requires that "final" ProShare payments be made to county facilities in June of the fiscal year to which the payments relate. New Hampshire had already made "final" ProShare payments in June 2000, CMS argued, and its state plan did not permit further payments in March 2002. During the appeal, CMS broadened the basis for the disallowance, asserting that New Hampshire violated its plan in calculating the March 2002 payments, because the plan did not employ the prospective RUGs methodology, but instead specified the methodology by which New Hampshire had calculated the "final" June 2000 ProShare payments, which CMS did not question. ⁽⁵⁾

New Hampshire argued that the procedure it followed in applying the RUGs methodology to calculate the disallowed payments was consistent with the procedure specified in the state plan. Alternatively, New Hampshire argued that the language in its plan specifying how to calculate the ProShare payments applied by its terms only to the interim, not the "final" payments, and, moreover, was worded in the past tense, meaning that it described only how the first ProShare payments were calculated. New Hampshire argued that the preamble to applicable federal regulations regarding upper payment limits indicates that states have flexibility in selecting the method used to determine the upper payment limits.

In response to the original basis for the disallowance, New Hampshire argued that its state plan's reference to a "final" ProShare payment should not be interpreted to bar it from adjusting its claims for FFP so long as those adjustments are within the federal two-year deadline for claiming FFP, which was the case here. New Hampshire argued that the Board's recent decision in Virginia Dept. of Medical Assistance Services, DAB No. 1838 (2002), upheld the state's claim for FFP in payments made after the date for the "final" payment specified in the state's Medicaid plan and was applicable here. New Hampshire noted that it amended its state plan to provide for two ProShare payments as a matter of convenience for the facilities, so that they could have access to funds earlier.

We sustain the disallowance. We first find that New Hampshire did not apply the calculation methodology specified in its state plan when it employed the prospective RUGs methodology to recalculate the ProShare payments. We also find that New Hampshire could not reasonably interpret its plan to incorporate the new methodology. We then discuss why the Board's holding regarding the word "final" in Virginia is not applicable here.

• New Hampshire did not apply the calculation methodology specified in its state plan.
  

We agree with CMS that New Hampshire did not calculate the March 2002 payments in accordance with its state plan. The approved state plan provides specific instructions for calculating ProShare payments, which utilize the Medicare Cost Limit as supplied by the fiscal intermediary. ⁽⁶⁾ The plan first provides that the interim payment shall be based upon the estimated Medicare Cost Limit as estimated by the fiscal intermediary, and that the payment to each facility is in proportion to the facility's Medicaid days, during the cost reporting period used to set its current rates, relative to the sum of Medicaid days for all eligible facilities.

The final payment is calculated through "a comparison of the rates Medicare would pay for these facilities using the Medicare ratesetting methodology and the rates New Hampshire Medicaid actually is paying these facilities," with determination of the Medicare rates being based on the total funds available within "the payment limits of 42 CFR 447.272." The calculation utilizes the Medicare Cost Limit as supplied by the fiscal intermediary (or the estimate of the limit if that is what the intermediary provides). The cost limit is adjusted through the application of several factors specified in the plan. These factors include inflation, location, and "capital cost and ancillary cost per patient day as determined by Medicaid ratesetting." New Hampshire Ex. F; CMS Ex. 8. New Hampshire did not dispute that it employed this methodology to calculate the $20 million FY 2000 ProShare payments that it made in June 2000, as well as the ProShare payments for FYs 1998 and 1999, made under the 1996 plan amendment.

New Hampshire's calculation of the March 2002 ProShare payment employed a different methodology than the one specified in the state plan. That determination involved application of the prospective rate setting "RUGs" methodology, which, as noted above, entails classifying a nursing facility's residents into 44 groups based on their utilization of facility resources, or "case mix." The state plan contains no reference to this methodology.

New Hampshire argued that in applying the RUGs methodology, it followed the specific instructions prescribed in its state plan. We find that this argument is not persuasive. New Hampshire submitted a declaration from a Budget Officer in its Office of Health Planning and Medicaid asserting that application of the RUGs methodology followed the state plan. New Hampshire Ex. U. The essence of this argument is that both calculations - the prospective RUGs calculation and the calculation described in the state plan and previously used to calculate ProShare payments - utilized information provided by the Medicare fiscal intermediary relating to Medicare reimbursement rates applicable to individual nursing homes. The calculation permitted by the plan uses the "Medicare Cost Limit or the estimate of the Limit if that is what the fiscal intermediary provides." CMS Ex. 8. The calculation of the March 2002 payments utilized "relevant rate information reported as prospective acuity groups for each nursing home from the fiscal intermediaries." New Hampshire Ex. U, ¶ 5. In both cases, New Hampshire adapted the information to the specific nursing homes in its claim by applying various factors.

However, the information New Hampshire supplied regarding its calculations substantiates that it indeed employed a different methodology than the one specified in its state plan. The record indicates that adapting the intermediary's rate information under the prospective RUGs methodology entailed the application of factors that are not specified or addressed in the state plan. The Budget Officer's earlier declaration describes the detailed and elaborate process by which the RUGs classifications were applied to patient data from the county nursing facilities. New Hampshire Ex. G. The effort, which New Hampshire described as a complicated process, took place over the course of three months and resulted in the determination that county providers were entitled to $80 million in ProShare payments for FY 2000, some $60 million more than they received under the prior calculation methodology. New Hampshire Br. at 8; New Hampshire Ex. G. The activities that New Hampshire staff undertook to apply the RUGs classifications to its nursing home data are not described or addressed in New Hampshire's approved state Medicaid plan. Second, a summary of adjustments that New Hampshire made to its calculation of the March 2002 ProShare payments shows that the figure determined with the prospective RUGs methodology had to be reduced to account for the inclusion in the RUGs data of charges for services that may have been reimbursed independently. New Hampshire Ex. I. These services are described as prescriptions, lab, outpatient hospital, X-ray, and reserved patient days; together, they totaled over $15 million. Thus, using the prospective payment required New Hampshire to make adjustments that are not referred to in the state plan. Even though in the calculations at issue New Hampshire made those adjustments, the absence of reference to them in the state plan means that there is no assurance they would always be made or accurately applied and further demonstrates that the payment was not made in accordance with the state plan, as required by 42 C.F.R. § 430.10.

New Hampshire argued that the preamble to the upper payment limits provision at 42 C.F.R. § 447.272 afforded states the flexibility to utilize prospective payment methodologies in calculating the upper payment limits:

The new UPL regulations afford States some flexibility in calculating a reasonable estimate of what Medicare would have paid for Medicaid services. In formulating their own approach to computing the UPL, States have flexibility to use either Medicare principles of cost reimbursement or prospective payment systems as the foundation of their estimates.

66 Fed. Reg. 3148, 3153 (January 12, 2001).

This language does not permit New Hampshire to adopt a methodology different than the one approved by CMS as part of New Hampshire's state plan.

The preamble goes on to state that CMS was not changing the standards that it applies to the review of state estimates of the upper payment limits. Id. This statement confirms that the state's estimates are subject to CMS review; here, CMS gave its approval to the method of calculating the upper payment limits (and thus the ProShare payments) in the state plan, but not to the figures that New Hampshire derived through application of the prospective methodology not specified therein.

Additionally, the preamble language addresses use of a state's estimate of the upper payment limits for a different purpose than that to which it was applied here. The distinction is subtle, but important. As used in section 447.272, the upper payment limits, as estimated by the state subject to CMS's review, represent an outer limit, a maximum level, above which the payments to a group of facilities, calculated pursuant to rates specified in the state plan, may not rise. Within that limit, states must also assure that the rates are reasonable and adequate to meet the costs that must be incurred by "efficiently and economically operated providers to provide services in conformity with applicable State and Federal laws, regulations, and quality and safety standards." 42 C.F.R. § 447.253. Here, by contrast, the upper payment limits were used to determine the actual level of reimbursement that county facilities were entitled to receive, in substitution for and in excess of the Medicaid reimbursement rates otherwise specified in the state plan. The flexibility that CMS described in the preamble did not extend to permitting a state simply to reinterpret its approved plan to permit calculation of reimbursement rates using a prospective payment system that is not clearly adopted or referred to in the plan.

Having found that the methodology that New Hampshire employed to calculate the March 2002 ProShare payment was not the methodology specified in its state plan, we next discuss why, under established principles guiding state plan interpretation, we reject New Hampshire's argument that it reasonably interpreted its state plan to permit the new methodology.

• New Hampshire's interpretation of its state plan as permitting it to adopt the RUGs calculation methodology is not entitled to deference.
  

The Board has repeatedly cited the standards for determining the degree of deference to be accorded a state in the interpretation of its state Medicaid plan that it discussed in South Dakota Dept. of Social Services, DAB No. 934 (1988). First, the Board looks to the language of the plan. If the language is ambiguous, the Board will consider whether the state's proposed interpretation is reasonable, gives effect to the purpose of the plan and program requirements, and is supported by evidence of consistent administrative practice. South Dakota at 4; North Carolina Dept. of Human Resources, DAB No. 1631, at 29 (1997); California Dept. of Health Services, DAB No. 1474, at 3 (1994). In conducting this analysis, the Board gives greater scrutiny to payments made to state-owned facilities; as we explain below, that heightened scrutiny is warranted here. South Dakota at 5.

In addition to arguing that it adhered to its plan in applying the prospective RUGs methodology, an argument that we rejected above, New Hampshire argued that it could reasonably interpret the plan as permitting the RUGs methodology because the plan provided a specific methodology only for calculating the interim ProShare payment. The plan's instructions for calculating the final payment are worded in the past tense, New Hampshire argued, and merely describe how New Hampshire calculated the initial ProShare payment in 1994, as a means of assuring that its Medicaid payment rates did not exceed the upper payment limits in 42 C.F.R. § 447.272.

Although use of the past tense does inject some ambiguity, the plan's calculation methodology appears in the plan section titled "proportionate share incentive adjustment," and is the only one specified for determining the payments. Additionally, New Hampshire did not dispute that it employed the payment methodology described in the plan to determine the ProShare payments for FY 2000 that it made in June 2000, as well as the payments in FYs 1998 and 1999; employment of this method was confirmed by the declaration of its accountant. New Hampshire Ex. U. Further, New Hampshire did not dispute CMS's assertion that it had never previously revisited its ProShare payments after having made the end-of-the-fiscal year payment called for by the plan. New Hampshire's interpretation here is thus not supported by its past administrative practice. Finally, when New Hampshire amended its state plan in 1996 to provide for interim and final payments, there was no RUGs methodology for Medicare payments for long term care facility services, so New Hampshire could not have intended that the RUGs methodology be used for determining final payments.

In rejecting New Hampshire's adoption of the prospective RUGs methodology, we cannot turn a blind eye to congressional and agency concern over the proliferation of state funding mechanisms that employ the aggregate upper limits provision to dramatically increase Medicaid reimbursement to county- and state-owned providers, which then transfer substantial portions of that reimbursement back to the state. We earlier cited testimony before Congress addressing the use of such funding mechanisms to boost federal Medicaid dollars that states receive, in excess of the actual cost of providing services. CMS Exs. 6, 7. As one federal court noted--

In particular, some states began to pay locally-owned facilities at rates exceeding their cost of serving Medicaid beneficiaries. These states would claim federal matching funds based on these "inflated" amounts, and the local facilities would then funnel some or all of the additional money back to the states through IGTs [inter-governmental transfers]. In some instances, the money that was funneled back to the states through IGTs was used to increase care to underserved populations--not necessarily Medicaid populations--or to otherwise improve healthcare in these states. In other instances, however, states were using this money for purposes completely unrelated to health care.

Ashley County Medical Center v. Thompson, 205 F.Supp.2d 1026, 1034 (D. Ark. 2002).

The October 2001 GAO report to Congress noted the conclusion of the Congressional Budget Office that such funding "schemes" were the most notable factor behind the then-recent increase in federal Medicaid spending, which was growing at a rate nine times that of the Medicaid population. GAO Report, at numbered page 1, CMS Ex. 7.

New Hampshire pointed out GAO's acknowledgment that such funding mechanisms, though regrettable, were not at the time impermissible. CMS also has not established that the regulations promulgated in January 2001 to curtail this practice by requiring discrete aggregation of payments to county facilities apply here. ⁽⁷⁾

However, the considerations discussed in the GAO materials that impelled CMS to issue those regulations guide us in assessing whether New Hampshire's interpretation of its plan is reasonable and gives effect to the purpose of the plan and program requirements. In conducting that analysis, the Board has held that it will give greater scrutiny to payments made to state-owned facilities. South Dakota at 5. Addressing such payments, the Board stated that the reason for this scrutiny is that increases in reimbursement rates for state-owned facilities do not raise the cost to the state of operating them, but rather benefit the state by increasing revenue from federal funds. Therefore, the state may be less motivated to scrutinize public facility rates carefully (or may interpret its plan more liberally in regard to public facilities) than to control private facility rates which impact state and federal funding sources similarly. See New York State Dept. of Social Services, DAB No. 1394, at 11 (1993) and cases cited therein. That concern applies equally here, where New Hampshire's interpretation had the effect of increasing its Medicaid funding without a proportionate increase in its costs.

New Hampshire's application of the prospective RUGs methodology nearly quadrupled its FY 2000 ProShare payments. New Hampshire did not dispute CMS's determination that of the $57,476,204 paid to the counties, $44,357,153 was "instantaneously" transferred back to the state. CMS Br. at 3. New Hampshire also failed to respond to the declaration of the CMS accountant that his analysis of Medicaid patient data revealed that patients requiring intensive levels of medical services were no more concentrated in county facilities than in privately-owned facilities, which calls into question the basis of the ProShare program. CMS Ex. 2, ¶ 8. New Hampshire's interpretation furthers neither the Medicaid program's purpose of financing medical assistance for eligible needy persons, nor its plan's purpose of compensating certain facilities for costs in excess of the Medicaid rates they otherwise receive. As such, New Hampshire's interpretation is not reasonable, and merits no deference.

New Hampshire argued that CMS effectively acknowledged the reasonableness of New Hampshire's plan interpretation by working closely with New Hampshire staff during the effort to recalculate the ProShare payment using the perspective RUGs methodology. New Hampshire reported almost weekly communications with CMS representatives over a three-month period in 2001, at the end of which the CMS accountant provided a spread sheet reflecting the new calculations, which became the basis of the disallowed payment.

To the extent that New Hampshire is raising an argument of equitable estoppel, that argument must fail. The Board has noted that the prevailing view in the federal courts is that equitable estoppel does not lie against the federal government, if indeed it is available at all, absent at least a showing of affirmative misconduct. See, e.g., Minnesota Dept. of Human Services, DAB No. 1791 (2001) and cases cited therein. More importantly, two of the traditional elements required to assert estoppel - reasonable reliance and detriment to the opposing party - are not present here. Id. The CMS Associate Regional Administrator informed New Hampshire well in advance of the March 27, 2002 payments to the counties that New Hampshire would not be receiving any additional funds for FY 2000 based on the recalculated numbers, first in a phone call shortly after New Hampshire reported the recalculated numbers in August 2001, and then in a letter dated February 26, 2002. New Hampshire Exs. J, at ¶ 3; N; O. The CMS accountant who worked with New Hampshire during the recalculation indicated that the effort was undertaken to demonstrate the impact of applying the new methodology to ProShare. He stated that New Hampshire neither asked for nor received approval to apply the prospective RUGs methodology retroactively to FY 2000, and that New Hampshire officials were told that approval of any claims would be up to the CMS central office, an assertion confirmed by contemporaneous internal correspondence among New Hampshire officials. CMS Ex. 2, ¶¶ 4, 11; New Hampshire Ex. I. New Hampshire cannot reasonably complain of any detriment resulting from its having worked with CMS staff to develop the revised payment.

• Other arguments, including Board precedent permitting payment after the "final" payment specified in a state plan, do not warrant permitting New Hampshire to apply the new calculation methodology.

As noted above, the original basis of the disallowance of the March 2002 payments was that New Hampshire had already made the "final" payments permitted by its plan in June 2000, and the plan did not provide for any subsequent payments.

There is merit to the argument that the plan's reference to a "final" payment does not necessarily prohibit otherwise appropriate adjustment of timely payments to providers (and consequent claims for FFP). As New Hampshire noted, the Board recently held that a similar plan provision referring to "final" provider payments did not bar a state's claim for FFP in payments made after the date for the final payments specified in the state's Medicaid plan. Virginia Dept. of Medical Assistance Services, DAB No. 1838 (2002). However, the circumstances and holding in Virginia do not carry the weight of analogy which New Hampshire accords it and do not support New Hampshire's claim here.

Virginia concerned enhanced reimbursement that Virginia paid (in addition to and beyond the quarterly payments provided for in its state plan) to two specific facilities, based on the unreimbursed costs of serving Medicaid and uninsured individuals. The Board found that the plan's language was ambiguous because it described the payment as "final" while at the same time permitting adjustments to reflect federal limits, and that Virginia had consistently interpreted its plan to permit subsequent enhanced payments. The Board also found that CMS's position -- that Virginia could collect overpayments from providers, but not make upward adjustments after the "final" payments -- would have required that Virginia make excessive and premature enhanced payments to the two hospitals at the same time as it made the "final" payments, when it lacked the cost data necessary to determine the enhanced payments. These payments would necessarily have been subject to downward adjustment as actual cost data became available. The hospitals would thus have been paid funds that they would later have to return, and Virginia would have been forced to claim federal funding knowing that it would have to return a significant portion of that funding at some point in the future. The Board found this roundabout procedure to be fiscally unsound for the federal government and the Medicaid program.

None of those circumstances exist here. There has been no showing that New Hampshire had ever previously interpreted its plan to permit ProShare payments for a given fiscal year beyond the dates specified in its plan; New Hampshire did not contest CMS's assertion that New Hampshire had never performed retroactive adjustments on ProShare payments subsequent to the dates for payment called for in both versions of the state plan. Upholding the disallowance here also does not present any of the illogical consequences that the Board found persuasive in Virginia. Finally, in Virginia, there was no allegation that the state failed to follow its state plan in calculating the disallowed payments.

New Hampshire also complained that CMS acted inconsistently in interpreting the plan to bar ProShare payments after the "final" payment deadline, while seeking post-deadline refunds of FFP in ProShare overpayments for FYs 1998 and 1999. New Hampshire did not dispute the overpayment determinations for FYs 1998 and 1999, and that issue is not before us. We note, however, that the determinations for those years were simply adjustments based on admittedly more accurate data (which, arguably, should have been used in the first place), and did not come about through the adoption of a new calculation methodology not specified in the approved state plan: there is no allegation that in calculating the overpayments New Hampshire disregarded the instructions in its state plan in favor of the prospective RUGs methodology used to recalculate the ProShare payments for FY 2000. We do not hold that a state is precluded from making routine upward or downward adjustments to prior period payments to reflect more accurate data consistent with the state plan, or from claiming FFP in those adjustments, so long as the claims are made in accordance with the state Medicaid plan, are timely, and meet other applicable requirements. ⁽⁸⁾

Conclusion

For the reasons explained above, we sustain the disallowance.

Judith A. Ballard

Cecilia Sparks Ford

Marc R. Hillson
Presiding Panel Member

1. Determination of the level of Medicare reimbursement that the facilities would receive for Medicaid services necessarily has a hypothetical aspect, because Medicaid covers some services that Medicare does not. 48 Fed. Reg. 56,046, 56,054 (1983); 56 Fed. Reg. 48,826, 48,863-64 (1991).

2. It is not apparent from the record whether the June 2000 payments were preceded by the "interim" payment specified in the state plan.

3. CMS reported that ProShare payment calculations supported payments of $28 million for FY 2000, but that appropriations by New Hampshire's legislature limited the actual payment to $20 million. CMS Ex. 2, ¶ 7.

4. The record reflects some uncertainty as to the precise amount of New Hampshire's claim for FFP in the 2002 ProShare payments for FY 2000, and thus the amount of the disallowance. In correspondence to CMS dated February 21, 2002, New Hampshire indicated that it was seeking a total grant award increase in the amount of $57,476,204, with 50% FFP of $28,738,102. New Hampshire Ex. N. However, CMS's June 26, 2002 disallowance letter states that New Hampshire paid the county facilities $60,655,969 on March 27, and claimed 50% FFP, or $30,327,984, in a quarterly expenditure report submitted on April 29, 2002, for the period ended March 21, 2002. In its brief, CMS reported that New Hampshire's actual payments to the county facilities were $57,476,204 after adjustments for the return of the overpayments for FYs 1998 and 1999, but also that New Hampshire claimed FFP on the basis of a $60,655,969 payment. CMS Br. at 3. In its reply brief, New Hampshire did not dispute or refer to CMS's assertion that it claimed FFP in $60,655,969 in ProShare payments. We need not determine the exact amount of the disallowance, however, because there is no indication that New Hampshire actually received the disallowed federal funds, and New Hampshire thus does not have to refund any monies to CMS as a result of our decision sustaining the disallowance.

5. The Board has considerable discretion to permit the parties to clarify or amend their positions as long as the opposing party has the opportunity to respond to any change. See 45 C.F.R. §§ 16.1, 16.9, 16.13, 16.15, and 16.21; Mississippi Dept. of Human Services, DAB No. 1267 (1991); West Central Wisconsin Community Action Agency, DAB No. 861 (1987). New Hampshire responded to the amended basis for the disallowance in its reply brief and did not request an opportunity for further briefing.

6. The intermediary is an entity that has a contract with CMS to determine and make Medicare payments for Part A or Part B benefits payable on a cost basis and to perform other related functions. 42 C.F.R. § 400.202.

7. CMS argued that the March 2002 payments violated the upper payment limit regulation as revised January 12, 2001 because New Hampshire applied an aggregate upper payment limit for all non-state nursing facilities, (i.e., private and county facilities). 66 Fed. Reg. 3147, 3175-76. However, CMS failed to address the applicability of the transition period that was provided in the new regulation and subsequently amended. These transition periods effectively provided a phase-in of the aggregate limit for county facilities, based on the effective date of the state's plan, and the state fiscal years to which the limit applied. Thus, as originally published in January 2001, the regulation provided that:

For approved plan provisions that are effective after October 1, 1992 and before October 1, 1999, payments during the transition period may not exceed the following--

(1) For State FY 2003: State FY 2003 UPL + .75X.

(2) For State FY 2004: State FY 2004 UPL + .50X. (3) For State FY 2005: State FY 2005 UPL + .25X. (4) For State FY 2006; State FY 2006 UPL. . .

66 Fed. Reg. 3147, 3175-76. In these calculations, "X stands for the payments to a specific group of providers described in paragraphs (a)(2) and (a)(3) of this section [county and private providers] in State FY 2000 that exceeded the amount" that would have been paid under the upper payment limit if that limit had been applied to that year. Id. Other transition periods applied to plans with different periods of effectiveness. CMS did not address the applicability of this regulation to this disallowance, other than to assert that New Hampshire had not claimed that the transition periods applied. CMS provided no basis to conclude that the regulation would apply to New Hampshire's payment for a prior fiscal year, or to a payment made before the first year of the transition periods by which the regulation was implemented. CMS also made no determination of the amount by which New Hampshire's aggregate payments to county providers exceeded the upper payment limits applicable to county providers in the new regulations.

8. Because we find that the plan's reference to a "final" payment does not necessarily bar routine adjustments to prior payments, or timely claims for those adjustments, we do not address New Hampshire's argument that CMS had agreed that its plan amendment adopting the interim/final payment schedule was not a "significant change" requiring public notice. Our holding does not turn on the plan's adoption of the "final" wording, but on our determination that New Hampshire failed to follow its plan in applying the prospective RUGs methodology.