This is the accessible text file for GAO report number GAO-04-228 
entitled 'Medicaid: Improved Federal Oversight of State Financing 
Schemes Is Needed' which was released on March 16, 2004.

This text file was formatted by the U.S. General Accounting Office 
(GAO) to be accessible to users with visual impairments, as part of a 
longer term project to improve GAO products' accessibility. Every 
attempt has been made to maintain the structural and data integrity of 
the original printed product. Accessibility features, such as text 
descriptions of tables, consecutively numbered footnotes placed at the 
end of the file, and the text of agency comment letters, are provided 
but may not exactly duplicate the presentation or format of the printed 
version. The portable document format (PDF) file is an exact electronic 
replica of the printed version. We welcome your feedback. Please E-mail 
your comments regarding the contents or accessibility features of this 
document to Webmaster@gao.gov.

This is a work of the U.S. government and is not subject to copyright 
protection in the United States. It may be reproduced and distributed 
in its entirety without further permission from GAO. Because this work 
may contain copyrighted images or other material, permission from the 
copyright holder may be necessary if you wish to reproduce this 
material separately.

Report to the Committee on Finance, U.S. Senate: 

United States General Accounting Office: 

GAO: 

February 2003: 

Medicaid: 

Improved Federal Oversight of State Financing Schemes Is Needed: 

Oversight of State Financing Schemes: 

GAO-04-228: 

GAO Highlights: 

Highlights of GAO-04-228, a report to the Committee on Finance, U.S. 
Senate 

Why GAO Did This Study: 

For years, some states have taken advantage of a loophole in Medicaid 
law that allows them to claim billions of dollars in excessive federal 
matching funds by exploiting the “upper payment limit” (UPL), which is 
intended to be a ceiling on federal cost sharing. Congress and the 
Centers for Medicare & Medicaid Services (CMS) acted to curtail UPL 
financing schemes through law in 2000 and regulation in 2001. CMS 
recognized that some states had developed a long-standing reliance on 
UPL funds. The law and regulation authorized transition periods of up 
to 8 years for states to phase out excessive UPL claims.

GAO was asked to examine CMS’s oversight of nursing home UPL 
arrangements, including the status of and the basis for transition 
period decisions.

What GAO Found: 

CMS has granted transition periods to 18 states for phasing out 
excessive claims for federal Medicaid funds obtained through UPL 
financing schemes. Eight states were granted 1- or 2-year transition 
periods, seven were granted 5-year transitions, and three states—
Nebraska, Pennsylvania, and Wisconsin—were granted the maximum of 8 
years. The law permits 8-year transition periods for qualifying states 
with UPL financing schemes relating to a payment provision established 
on or before October 1, 1992. Although permissible under the law, CMS’s 
decisions to grant 8-year transition periods to two of the three states 
were not consistent with the agency’s stated purpose for the UPL 
regulation and transition policy, which targeted arrangements with 
problematic characteristics and states with a long-standing budgetary 
reliance on excessive federal funds. Neither Nebraska nor Wisconsin had 
such arrangements or budgetary reliance until after 1997 and 2000, 
respectively (see figure). Under their 8-year transition periods, these 
states can claim about $633 million more in federal matching funds than 
they could have claimed under shorter transition periods consistent 
with the stated purpose of CMS’s regulation and transition policy.

CMS has strengthened its oversight of state UPL schemes, including 
forming a team to coordinate its reviews, drafting internal guidelines 
for reviewing state methods, and conducting financial reviews that have 
identified hundreds of millions of dollars in improper claims. CMS has 
not focused its reviews on the states with the largest arrangements,
however, or instructed states on appropriate methods for calculating 
their UPLs. GAO’s analysis of six states’ UPL methods found variations 
and concerns suggesting that states may be overstating their UPL 
claims. Although efforts by Congress and CMS have narrowed the UPL 
loophole, it has not been eliminated. States can and do continue to 
claim excessive federal matching funds through UPL arrangements, using 
them for non-Medicaid purposes or to inappropriately increase the 
federal share of Medicaid program expenditures.

What GAO Recommends: 

CMS concurred with GAO’s recommendations that the agency improve its 
oversight of UPL arrangements, including expediting its financial 
reviews, establishing uniform guidance for states, and improving state 
reporting. CMS, Nebraska, and Wisconsin disagreed with GAO’s 
recommendation that CMS reassess its decisions to grant those two 
states an 8-year transition period. GAO is suggesting that Congress 
consider ending, under certain circumstances, the 8-year transition 
periods for states with excessive nursing home UPL arrangements. 

www.gao.gov/cgi-bin/getrpt?GAO-04-228.

To view the full product, including the scope and methodology, click on 
the link above. For more information, contact Kathryn G. Allen at 
(202) 512-7118.

[End of section]

Contents: 

Letter: 

Results in Brief: 

Background: 

Eighteen States Have Received Transition Periods, with Three States 
Receiving the Maximum of 8 Years: 

CMS's Basis for Granting 8-Year Transition Periods to Two States Was 
Not Consistent with Its Stated Objectives: 

CMS Has Strengthened Its Oversight of UPL Arrangements, but Concerns 
Remain: 

Conclusions: 

Matter for Congressional Consideration: 

Recommendations for Executive Action: 

Agency and State Comments and Our Evaluation: 

Appendix I: Comparison of Federal Funds under 8-Year and Shorter 
Transition Periods: 

Appendix II: Scope, Methodology, and Analysis of Selected State UPL 
Calculations: 

Appendix III: Comments from the Centers for Medicare & Medicaid 
Services: 

Appendix IV: Comments from the State of Michigan: 

Appendix V: Comments from the State of Nebraska: 

Appendix VI: Comments from the State of New York: 

Appendix VII: Comments from the State of Oregon: 

Appendix VIII: Comments from the State of Wisconsin and GAO's Response: 

GAO's Response to the State of Wisconsin's Comments: 

Appendix IX: GAO Contact and Staff Acknowledgments: 

GAO Contact: 

Acknowledgments: 

Related GAO Products: 

Tables: 

Table 1: Transition Period Assignments for States' Nursing Home UPL 
Arrangements: 

Table 2: Benefits and Shortcomings of CMS Actions to Strengthen 
Oversight of UPL Arrangements: 

Table 3: Status of CMS Financial Management Reviews of Nursing Home UPL 
Arrangements, as of October 2003: 

Table 4: Selected States' Use of Funds Generated through UPL 
Arrangements: 

Table 5: Excessive Federal Payments, State Fiscal Years 2001-2009, for 
Two States under Different Transition Periods: 

Figures: 

Figure 1: Overview of How States Can Use the UPL to Make Supplemental 
Payments beyond Actual Medicaid Payments: 

Figure 2: What Happens to an Excessive UPL Claim of $100 Million a Year 
under Each of the Four Transition Periods: 

Figure 3: Federal Share of Supplemental Payments in Nebraska and 
Wisconsin through SFY 2001: 

Abbreviations: 

BIPA: Medicare, Medicaid, and SCHIP Benefits Improvement and Protection 
Act of 2000: 

CBO: Congressional Budget Office: 

CMS: Centers for Medicare & Medicaid Services: 

HCFA; Health Care Financing Administration: 

HHS: Department of Health and Human Services: 

IGT: intergovernmental transfer: 

OIG: Office of Inspector General: 

NIRT: National Institutional Reimbursement Team: 

SFY: state fiscal year: 

UPL: upper payment limit: 

United States General Accounting Office: 

Washington, DC 20548: 

February 13, 2004: 

The Honorable Charles Grassley: 
Chairman: 
The Honorable Max Baucus: 
Ranking Minority Member: 
Committee on Finance: 
United States Senate: 

In searching for ways to generate additional revenues, many states have 
turned to certain creative financing schemes for Medicaid, the federal-
state partnership that finances health care for an estimated 53 million 
low-income people. Although each state must pay a share of Medicaid 
expenditures, the federal government pays the larger share--on average, 
57 percent--calculated according to a matching formula defined in 
statute.[Footnote 1] Over the years, some states have taken advantage 
of the flexibility built into the Medicaid program by devising 
financing schemes that inappropriately boost the federal share of 
program expenditures, through use of Medicaid upper payment limit (UPL) 
provisions, resulting in excessive federal payments. The UPL is the 
upper bound on what the federal government will pay as its share of the 
Medicaid costs for different classes of covered services, and it often 
exceeds what states actually pay providers for services. Some states 
exploited the UPL loophole by paying nursing homes and hospitals owned 
by local governments much more than the established Medicaid payment 
rate, and requiring the providers to return the excess payments to the 
state. States that used the UPL loophole have collected billions of 
excessive federal dollars since the mid-to late 1990s. Few states 
specifically acknowledged how they spent these funds, but officials in 
some states have reported uses including education or health care 
programs besides Medicaid.

Over the years, we and others have reported concerns with states' 
ability to take advantage of the UPL loophole and other financing 
schemes.[Footnote 2] As the UPL financing schemes came to light, 
Congress and the Health Care Financing Administration (HCFA), the 
federal agency administering the Medicaid program, took action through 
statute and regulation to curtail states' ability to claim excessive 
federal funds through these UPL financing schemes.[Footnote 3] HCFA 
initiated policy changes to restrict states' UPL arrangements in an 
October 2000 proposed regulation.[Footnote 4] The Medicare, Medicaid, 
and SCHIP Benefits Improvement and Protection Act of 2000 (BIPA) 
directed HCFA to issue a final regulation to limit states' ability to 
claim excessive federal matching funds through UPL 
arrangements.[Footnote 5] BIPA also required that HCFA's final 
regulation--established in January 2001[Footnote 6]--allow for 
transition periods as long as 8 years, during which time excessive UPL 
payments would be phased out. Because some states may have come to rely 
on these excessive federal funds, the length of a state's transition 
period was based in part on how long the state had in place a UPL 
arrangement meeting certain specified criteria.

This report addresses your questions about the criteria and process 
used by the agency, renamed the Centers for Medicare & Medicaid 
Services (CMS), to authorize transition periods. You asked us to 
examine states' UPL payment schemes that received the maximum 
transition period of 8 years and involved payments to nursing homes, 
which generally represent the highest-cost UPL arrangements.[Footnote 
7] In addition, you raised questions about CMS's monitoring of the UPL 
arrangements still allowed under the new 2001 regulation. In response 
to your request, this report addresses the following questions: 

* What is the status of CMS's activity in establishing transition 
periods for states to phase out UPL arrangements involving excessive 
payments to nursing homes?

* Did CMS have a sound basis for its decisions to grant maximum-length 
(8-year) transition periods for UPL arrangements involving excessive 
payments to nursing homes?

* Is CMS's continuing oversight of UPL arrangements sufficient to 
ensure that claims submitted by states are calculated appropriately and 
comply with Medicaid requirements?

To determine the status of CMS's actions in establishing transition 
periods, we interviewed CMS and state officials and reviewed documents 
pertaining to CMS's criteria, calculation, review, and preliminary 
decisions for the 18 states granted transition periods to phase out 
their nursing home UPL payment arrangements. To evaluate CMS's basis 
for approving maximum-length transition periods for arrangements 
involving payments to nursing homes, we reviewed the statutory language 
that authorized the 8-year transition period and related CMS 
regulations, policies, and memoranda, discussed transition period 
decisions with CMS officials, and reviewed CMS documentation, including 
written responses to our questions about specific transition period 
decisions. We also reviewed the nursing home UPL arrangements in the 
three states that received 8-year transition periods: Nebraska, 
Pennsylvania, and Wisconsin. To assess CMS's efforts to oversee state 
UPL arrangements, we analyzed the UPL calculations of six states. Four 
of these states--Michigan, New York, Oregon, and Washington--received 
5-year transition periods, and two states--Pennsylvania and Wisconsin-
-received 8-year transition periods. We selected these states because, 
of the 10 states with 5-or 8-year transition periods, these six have 
the largest nursing home UPL arrangements. Finally, we used our past 
work to provide historical information on the nature of state financing 
schemes, and reviewed studies conducted by CMS, the Office of the 
Inspector General (OIG) of the Department of Health and Human Services 
(HHS), state auditors, and others. We conducted our work from October 
2002 through January 2004 in accordance with generally accepted 
government auditing standards.

Results in Brief: 

CMS has granted provisional transition periods to 18 states for phasing 
out UPL schemes involving excessive federal payments for nursing homes. 
Of the 18, CMS determined that 3 states were eligible for the maximum 
8-year transition period, 7 for 5-year transitions, and 8 for 1-or 2-
year transitions. CMS determined that UPL arrangements in 8 other 
states were not eligible for any transition period because the 
arrangements were too recent. CMS officials told us that none of these 
decisions were final and that all were subject to continuing review. 
CMS has had to rely on limited historical information on states' 
arrangements to assign transition periods, and on state-provided 
information to determine how much federal funding each state would be 
allowed to claim during the transition periods. On the basis of data 
submitted by states, we estimate that the 10 states with nursing home 
UPL arrangements that have 5-or 8-year transition periods could receive 
about $9 billion in excessive federal matching funds during their 
transition periods. CMS's reviews of the validity of states' claims 
have been hampered by states' slow and incomplete responses to the 
agency's requests for information. Transition periods have already 
ended in 8 states, and 5-year transition periods are scheduled to end 
in 2005.

CMS's decisions to grant 8-year transition periods to two of the three 
states that received them--Nebraska and Wisconsin--were not consistent 
with the purpose the agency identified for the UPL regulation and for 
transition periods in the preamble to the January 2001 final rule. 
Although these decisions were permissible under the statute, neither 
Nebraska nor Wisconsin had the type of long-standing arrangement in 
place that the agency said the regulation was designed to curtail or a 
long-standing budgetary reliance on excessive payments. In Wisconsin's 
case, CMS's action is particularly troublesome because, as we earlier 
reported, Wisconsin's UPL financing scheme, established in 2001, should 
not have been approved.[Footnote 8] Wisconsin's arrangement was 
established after the agency had already taken action to curtail such 
practices. Over the 8-year transition periods, Nebraska and Wisconsin 
are eligible to receive about $633 million more in excessive federal 
matching funds than they would have been eligible for under shorter 
transition periods based on the purposes of CMS's regulation and 
transition period policy as discussed in the preamble. In contrast to 
Nebraska and Wisconsin, the third state granted an 8-year transition 
period, Pennsylvania, had a long-standing UPL arrangement that has 
generated large federal payments for the state and, therefore, had in 
place the type of problematic arrangement CMS described.

CMS has taken a number of steps to strengthen its oversight of state 
UPL arrangements, including forming a team to coordinate and bring 
uniformity to the agency's review of the arrangements, and drafting 
internal guidelines for reviewing state UPL calculations. CMS's newly 
established financial management reviews have identified hundreds of 
millions of dollars in improper UPL payments, allowing the agency 
either to avoid paying states inappropriately or to recoup the 
overpayment from states in those cases where the state had already been 
paid. For example, CMS has begun action to recoup over $200 million 
from Louisiana and Missouri because financial reviews determined that 
these states had claimed more in federal matching funds through their 
UPL arrangements than allowed under the 2001 regulation. Despite these 
positive results from its financial reviews, CMS has not focused its 
reviews on the states with the largest UPL arrangements. Furthermore, 
although CMS established internal guidelines for its financial 
management reviews, it has not issued guidance for states' use on 
appropriate methods for calculating their UPLs, contributing to the 
possible overstatement of states' claims for federal matching funds 
under the 2001 UPL regulation. The agency has also not developed a 
process that will assure states' compliance with its requirement that 
states report UPL payments made to individual nursing homes and other 
facilities. Despite CMS's new regulation, states still can--and do---
claim excessive federal matching funds and use them for non-Medicaid 
purposes or to inappropriately increase the federal share of Medicaid 
program expenditures.

We believe Congress and CMS should continue legislative and 
administrative efforts to preclude states' ability to claim excessive 
federal Medicaid payments and to inappropriately shift Medicaid costs 
to the federal government. We suggest that Congress consider a 
recommendation that remains open from our prior work, that is, to 
prohibit Medicaid payments to government-owned facilities that exceed 
costs. Further, in light of concerns that CMS's current policy for 
granting states an 8-year transition period could further erode the 
fiscal integrity of the Medicaid program and significantly increase 
future Medicaid expenditures, we also suggest that Congress consider 
ending the 8-year transition period for states with excessive nursing 
home arrangements, with a consideration for states that have 
demonstrated a long-standing budgetary reliance on excessive federal 
funds. This report also contains recommendations for the Administrator 
of CMS to take actions to improve the agency's oversight of state 
Medicaid financing arrangements. These actions include reassessing its 
decisions granting 8-year transition periods to Nebraska and Wisconsin, 
establishing guidance for states on appropriate methods for calculating 
their UPLs, improving its requirements for states to report on UPL 
arrangements, and giving priority to financial management reviews for 
states with the largest UPL arrangements and active transition periods.

In commenting on a draft of this report, CMS did not concur with our 
recommendation that it reconsider its initial decisions to grant 
Nebraska and Wisconsin 8-year transition periods. CMS believes that its 
current policy supports the intent of the UPL regulations regarding 
appropriate federal Medicaid spending as well as congressional intent 
to allow for an 8-year transition period. While we acknowledge that 
CMS's current transition period policy and decisions are legally 
permissible, we do not believe they are consistent with the objectives 
the agency identified when it issued the regulations. Because CMS is 
maintaining that its 8-year transition period policy and decisions are 
appropriate, we have elevated this issue for Congress to consider, as 
indicated above. Other than those recommendations related to its 
transition period policy and decisions, CMS generally concurred with 
our recommendations to improve its oversight, including to establish 
guidance for states and to improve its requirements for state reporting 
on their UPL arrangements.

We also provided a draft of this report to the states of Michigan, 
Nebraska, New York, Oregon, Pennsylvania, Wisconsin, and Washington. 
All the states but Washington provided comments. Nebraska and Wisconsin 
disagreed with our recommendation that CMS reconsider its initial 8-
year transition period decisions. Nebraska stated that it complied in 
good faith with the Medicaid regulations and that reinterpreting the 
decision to grant the state an 8-year transition period would be 
unacceptable. Wisconsin stated that it was eligible for an 8-year 
transition period under the criteria established by BIPA and provided 
extensive comments explaining the basis for this view. We acknowledge 
that CMS's current transition period decisions are permissible under 
the statute, and have clarified our report accordingly. However, 
because of the fiscal impact of these decisions and concerns that CMS's 
policy could open the door for other states to claim that they qualify 
for an 8-year transition period, we maintain that CMS should reconsider 
its decisions to grant 8-year transition periods to these states.

Michigan, New York, Oregon, and Pennsylvania provided more limited 
comments and did not take issue with our recommendations. These states 
also provided technical comments on their specific UPL methodologies, 
which we incorporated as appropriate.

Background: 

Title XIX of the Social Security Act authorizes federal funding to 
states for Medicaid, which finances health care for certain low-income, 
aged, and disabled individuals. States have considerable flexibility in 
designing and operating their Medicaid programs, but they must comply 
with federal requirements specified in Medicaid statute, regulations, 
and program directives. Each state administers its Medicaid program in 
accordance with a state plan approved by the Secretary of HHS, who has 
delegated approval authority to CMS. States' Medicaid plans specify, 
for example, the services to be provided and how the state will 
calculate the amount it will pay for these covered services. Any 
program changes a state wishes to make, including establishing UPL 
arrangements, must be submitted for CMS approval as a state plan 
amendment. The Social Security Act requires state plans to meet various 
requirements related to payments for Medicaid-covered care and 
services. Among other things, plans are to ensure that payments are 
consistent with efficiency, economy, and quality of care.[Footnote 9] 
The federal government pays a specified share of each state's Medicaid 
payments on the basis of a cost-sharing formula established under the 
Social Security Act.

UPLs Set Maximum Federal Cost Sharing for Medicaid Services: 

UPLs are the federal government's way of placing ceilings on federal 
financial participation in a state's Medicaid program.[Footnote 10] The 
UPL is tied to the amount that Medicare, the federal program that 
provides health coverage for seniors and some disabled persons, pays 
for comparable services. Thus, if the average Medicare rate for 1 day's 
care in a nursing home were $150, and a state provided a total of 
10,000 days of nursing home care under its Medicaid program, the UPL 
for that service would be $1.5 million. If the Medicaid cost-sharing 
formula called for the federal government to provide 60 percent of this 
state's Medicaid expenditure, the maximum the federal government would 
pay would be $900,000. In this way, the UPL places an upper limit 
beyond which federal matching funds will not be provided.

In practice, states' Medicaid rates are often lower than Medicare 
rates, creating a potential gap between the UPL and the amount states 
actually spend to provide services to Medicaid beneficiaries. For 
example, if a state actually paid for Medicaid-covered nursing home 
care at $100 per day instead of the $150 per day Medicare rate, the 
state's actual spending for 10,000 days of nursing home care would 
total $1 million, not $1.5 million. Under the Medicaid cost-sharing 
formula, the state would therefore receive federal matching funds for 
60 percent of $1 million, not 60 percent of $1.5 million. Under the 
Medicaid program, however, states may make payments to providers 
separate from and in addition to those made using standard Medicaid 
rates, and the federal government will share in those payments up to 
the maximum allowed under the UPL. Such supplemental payments might be 
made, for example, to compensate a specific facility for higher-cost 
patients or for meeting an important community need, such as providing 
a high volume of care to people on Medicaid (see fig. 1).

Figure 1: Overview of How States Can Use the UPL to Make Supplemental 
Payments beyond Actual Medicaid Payments: 

[See PDF for image]

[End of figure]

Some States Have Exploited UPLs to Inappropriately Increase Federal 
Matching Funds: 

Over the years, some states exploited UPLs to claim excessive federal 
matching funds by paying government-owned facilities at rates much 
higher than established Medicaid rates. States used a method known as 
"aggregation," whereby they estimated the UPL for a given class of 
service, such as nursing home care, on the basis of all facilities in 
the state that provided that service, including state-owned, local-
government-owned, and private facilities. By combining, or aggregating, 
the UPL for services provided by nursing homes in all three categories 
of ownership, for example, and determining the difference between the 
combined UPL and what a state actually paid for Medicaid-covered 
nursing home services statewide, the state could pay this large 
difference to only a few government-owned facilities, claim federal 
matching dollars for the payment, and require those facilities to 
return most or all of the excessive payment.[Footnote 11] If, for 
example, the difference between the standard Medicaid payment rate and 
the UPL for private nursing homes was $100 million, $20 million for 
local-government-owned nursing homes, and $50 million for state-
government-owned nursing homes, the state could make a payment equaling 
the aggregated difference of $170 million to a few government-owned 
nursing homes and then claim federal matching funds for the aggregated 
amount. Although, in some cases, states agreed to let the nursing homes 
keep a portion of the excess funds, typically the nursing homes were 
required to return the bulk of the payments to the state.

In 1987, HCFA acted to limit states' ability to use the UPL loophole. 
In response to states' manipulation of the UPL to make extremely large 
payments to state-owned facilities, HCFA created a separate UPL for 
state-owned facilities. HCFA's regulation, however, allowed states to 
aggregate the UPLs of local-government[Footnote 12] and private 
facilities. Abusive arrangements involving local-government-owned 
facilities, like the prior abusive arrangements with state-owned 
facilities, became prominent in the mid-1990s and shared three 
characteristics: 

* The basis of the states' payment was the combined UPL for private and 
local-government facilities. The states aggregated the dollar amounts 
in the UPL gap--that is, the difference between states' Medicaid 
payments for a class of services and Medicare's payments for the same 
services--for all local-government-owned and privately owned 
facilities providing, for example, nursing home services. Similar to 
the arrangements involving state-owned facilities, the states then paid 
one or more counties, on behalf of a few local-government nursing 
homes, a sum based on the UPL gap, in addition to and separate from 
payments at the states' standard Medicaid payment rates, which allowed 
the states to claim excessive federal matching dollars.

* States' payments for the aggregated UPL gap resulted in total 
payments that were much higher than the facilities' actual costs. Since 
UPL arrangements were based on an aggregated UPL gap for all private 
and local-government facilities, but payments were made to only a few 
facilities, the payments have been huge--vastly higher than the 
payments those facilities would have received under the standard 
Medicaid payment rate. For example, our earlier work found that in one 
state, the average daily federal matching payment for a Medicaid 
patient in a county nursing home increased by 17 fold, from $54 per day 
to about $969 per day.[Footnote 13] Such UPL arrangements resulted in 
sums that were many times greater than the facilities' actual costs of 
providing Medicaid services.

* States' payments for the aggregated UPL gap were temporary because 
most or all of the excessive payments were immediately returned to the 
state. Because they are public entities, local-governments (typically 
counties) could return the payment to the state through what is known 
as an intergovernmental transfer.[Footnote 14] In some instances, the 
temporary payment was made through a bank transaction between the state 
and local government, with the "round-trip" of funds (from the state, 
to the local government, and back to the state) completed within a few 
hours.[Footnote 15] The local governments were in many cases allowed to 
keep a small portion of the UPL payment.

Establishing UPL arrangements using local-government and private 
facilities allowed states to claim billions of dollars in excessive 
federal Medicaid matching funds, without any increase in state Medicaid 
expenditures. States used the additional federal dollars for non-
Medicaid purposes and for financing the states' share of Medicaid 
expenditures. The total dollars involved were substantial: the 
Congressional Budget Office (CBO) concluded in January 2001 that UPL 
schemes were the principal factor behind the 9 percent growth in 
Medicaid spending in 2000, the largest spending increase of the 
previous 7 years.[Footnote 16]

HCFA and Congress Took Steps to Limit States' UPL Schemes: 

Responding to escalating costs from such arrangements, HCFA in October 
2000 proposed a regulation that would eliminate states' ability to 
aggregate UPLs across private and local-government providers. Under 
this regulation, a separate UPL was proposed for nonstate government 
facilities (those owned by local governments), a limit that would 
effectively end states' ability to aggregate UPLs for local-government 
and private providers.

For UPL schemes that were already in place, HCFA's proposed October 
2000 regulation contained transition periods that would phase out 
prohibited payments over time rather than eliminate them immediately. 
HCFA proposed such transition periods because it recognized that some 
states, as part of their overall state budgets, had come to rely on the 
additional money they were receiving through these schemes. The 
proposed transition periods extended payments to states for as long as 
5 years. The length of the transition period a state would receive was 
based on how long it had operated the type of UPL arrangement that 
would be prohibited under the proposed regulation, with the longest-
standing arrangements receiving the 5-year transition time. During the 
transition, states could continue to receive excessive federal matching 
funds from these arrangements--those prohibited by the 2001 UPL 
regulation--at a gradually declining rate.

BIPA directed HCFA to finalize its October 2000 proposed regulation and 
also established an 8-year transition period,[Footnote 17] which states 
could receive if they met two conditions. First, a state must have had 
a Medicaid plan payment arrangement (specifically, a provision or 
methodology) in place on or before October 1, 1992.[Footnote 18] 
Second, the payment provision or methodology must have provided for 
payments in excess of the new UPL limits under certain identified time 
periods.[Footnote 19]

HCFA issued its final UPL regulation on January 12, 2001. The 
regulation narrowed the UPL loophole by establishing separate UPLs for 
private, state, and local-government facilities. The January 2001 
regulation also created three transition periods--8 years, 5 years, and 
2 years. The agency later added a 1-year transition period for states 
with arrangements established around the same time as the regulation 
was published.[Footnote 20] Transition periods were based on the date 
the state's payment provision or methodology went into effect, with 
long-standing arrangements (those with earlier effective dates) 
receiving longer transitions. During its transition period, a state 
must gradually phase out excessive payments. The length of the 
transition period is important, because a state granted an 8-year 
transition period receives substantially more federal funding than it 
would receive under shorter periods. Figure 2 compares the amounts for 
a hypothetical excessive federal payment of $100 million that is now 
prohibited. In this example, a state would receive a total of $575 
million in excessive federal funds under an 8-year period, $350 million 
under a 5-year period, $200 million under a 2-year period, and $100 
million under a 1-year period. Amounts are greatest under the 8-year 
period because the state receives 3 full years of payment at 100 
percent of the excessive amount, plus decreasing payments for 5 
additional years. HCFA estimated that, even with such transition 
periods, its 2001 regulation would reduce payments to all states with 
any existing UPL arrangement (nursing home, inpatient hospital, or 
outpatient hospital) by $55 billion over a 10-year period.[Footnote 21]

Figure 2: What Happens to an Excessive UPL Claim of $100 Million a Year 
under Each of the Four Transition Periods: 

[See PDF for image]

Note: All dollar figures are in millions. Percentages of excessive 
federal payments are those allowed under the 2001 UPL regulation. 
Percentages for 2-year, 5-year, and 8-year transition periods were 
established on January 12, 2001, and the percentage for the 1-year 
transition period was established on September 5, 2001.

[End of figure]

Even under the January 2001 regulation, however, states can generate 
excessive federal matching payments beyond what they would claim using 
their established Medicaid payment rates. States can do this by 
aggregating the payments for all local-government nursing homes under 
one UPL, making payments to one or more counties on behalf of just a 
few nursing homes, and requiring that the excessive payments be 
returned to the state. States' use of such arrangements for nursing 
homes continues to grow, even though the newer arrangements involve 
fewer dollars because UPL aggregation is limited to local-government 
facilities. CMS has reported a more than fivefold growth in the number 
of states with UPL nursing home arrangements, from 5 states before 1999 
to 26 states as of January 2003.

Eighteen States Have Received Transition Periods, with Three States 
Receiving the Maximum of 8 Years: 

As of October 2003, CMS had determined that 18 of the 26 states with 
UPL arrangements involving nursing homes were eligible for transition 
periods to phase out excessive payments. CMS determined that 3 of the 
18 states were eligible for the maximum 8-year transition period, 7 
were eligible for 5-year transitions, and 8 were eligible for 1-or 2-
year transitions. Eight additional states were found not eligible for 
transition periods because their UPL arrangements were too new to 
qualify (see table 1).

Table 1: Transition Period Assignments for States' Nursing Home UPL 
Arrangements: 

8-year: Nebraska; Pennsylvania; Wisconsin; 

5-year: Alabama; Michigan; New Hampshire; New York; North Dakota; 
Oregon; Washington; 2-year: Iowa; Kansas; Louisiana; Missouri; 
New Jersey; South Dakota; Tennessee; 

1-year: Virginia; 

Not eligible: Arkansas; Colorado; Georgia; Indiana; Kentucky; 
Mississippi; Montana; South Carolina. 

Source: CMS.

[End of table]

Until 2001, the agency did not require states to specifically report 
how many federal dollars they claimed through UPL arrangements. Also, 
while states needed to obtain HCFA regional office approval for their 
payment methodologies, these payment provisions were not necessarily 
identified in the state plan amendments submitted by states, or by 
HCFA, as a "UPL arrangement." Consequently, CMS had limited historical 
information on states' UPL arrangements to use to assign transition 
periods and to determine how much in federal funds each state would be 
allowed to claim. To supplement the information it had, CMS requested 
documentation from states on how they calculated their UPLs and on 
their estimates of what they could claim during their transition 
periods. This information was requested when CMS informed states in 
early spring 2002 of the transition periods it had assigned them.

CMS's reviews of state UPL payment calculations took months to 
complete, and some were still under way as of October 2003, in part 
because some states' responses to CMS's requests were not received for 
weeks or even months.[Footnote 22] For example, one state did not 
respond to CMS's February 2002 letter and data request until July 16, 
2003--more than a year later. According to CMS officials, state 
responses were often incomplete or failed to provide the detailed 
information the agency requested, which also delayed CMS reviews of 
state UPL calculations and excess payment estimates. As of January 
2004, CMS had not completed reviews of UPL calculations for 3 of the 18 
states with nursing home UPL arrangements and an assigned transition 
period. These included 2 that had active transition periods and large 
UPL arrangements, New York and Oregon.

CMS officials told us that they consider all the transition period 
decisions provisional and subject to revision based on the findings of 
continuing reviews. CMS officials stated that they do not intend to 
issue final transition period determinations; instead, through ongoing 
reviews of state UPL calculations and estimated transition period 
payments, the agency will communicate directly with states to make any 
necessary changes. CMS also intends to monitor actual state spending 
and recoup any payments beyond the allowable limits. According to 
preliminary UPL estimates submitted by 10 states with UPL arrangements 
and either 5-or 8-year transition periods, these states expect to claim 
a total of about $9 billion in excessive federal matching funds during 
their transition periods.[Footnote 23]

By October 2002, transition periods for eight states had ended, even 
though CMS had not completed its initial reviews of all these states' 
UPL methodologies and transition period assignments.[Footnote 24] The 
transitions of the eight states granted 1-or 2-year transition periods 
were mandated to end on September 30, 2002, or earlier.[Footnote 25] 
The seven states with 5-year transition periods must be in compliance 
with the 2001 regulation by state fiscal year (SFY) 2006; thus, SFY 
2005 will be the last year that these states can claim a portion of the 
excessive amount that they have claimed in the past. The three states 
granted 8-year transition periods must be in compliance by October 1, 
2008.

CMS's Basis for Granting 8-Year Transition Periods to Two States Was 
Not Consistent with Its Stated Objectives: 

CMS's basis for granting 8-year transition periods to two of three 
states--Nebraska and Wisconsin--was not consistent with the objectives 
the agency identified for the UPL regulation and transition periods in 
the preamble to the January 2001 regulation. When developing and 
issuing its January 2001 UPL regulation, the agency repeatedly stated 
that transition periods were to address states with UPL arrangements 
with certain problematic characteristics and a long-standing budgetary 
reliance on excessive federal funds. Yet neither Nebraska nor Wisconsin 
had long-standing problematic arrangements or a long-standing budgetary 
reliance on excessive UPL payments from them. In contrast, the decision 
to grant Pennsylvania an 8-year transition period was consistent with 
the purpose described by CMS of the UPL regulation and transition 
periods. CMS made its initial transition period decisions without 
establishing and conveying how it would interpret the statutory 
language for granting states an 8-year transition period. Granting 
Nebraska and Wisconsin 8-year transition periods will result in 
significant federal government outlays. Under the 8-year phase-out, 
Nebraska and Wisconsin are slated to receive about $633 million more in 
federal matching funds than they would have received under shorter 
transition periods more consistent with the agency's stated objectives. 
As of September 2003, the states had already claimed about $497 million 
of these excessive federal funds.

CMS's Transition Period Decisions for Nebraska and Wisconsin Are 
Inconsistent with Stated Regulatory Objectives: 

Although permissible under the law, CMS's granting 8-year transition 
periods to Nebraska and Wisconsin is not consistent with the objectives 
the agency identified in the preamble to its January 2001 regulation, 
as neither state had in place the type of long-standing excessive 
payment arrangement the agency intended to curtail.[Footnote 26] In 
explaining the basis for the changes imposed by the final regulation, 
HCFA emphasized that a key objective of the regulation was to limit 
states' ability to gain excessive federal matching payments through 
their UPL arrangements, specifically on the basis of aggregating UPLs 
of private and local-government facilities. The regulation was needed, 
according to HCFA, because: 

[i]t had become apparent that the existing regulations created a 
financial incentive for States to overpay non-State government-owned or 
operated facilities because, through this practice, States, counties, 
and cities were able to effectively lower net State or local 
expenditures for covered services and gain extra Federal matching 
payments.

The agency further explained that: 

[b]y developing a payment methodology that set rates for proprietary 
and nonprofit facilities at lower levels, States were able to set rates 
for county or city facilities at substantially higher levels and still 
comply with the existing aggregate upper payment limit. The Federal 
government matched these higher payment rates to public facilities. 
Because these facilities are public entities, funds to cover the State 
share were 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.[Footnote 
27]

According to HCFA, such practices contributed to rapid growth in 
Medicaid spending and were not consistent with the statutory 
requirements that Medicaid payments be economical and 
efficient.[Footnote 28] In the preamble to the regulation, HCFA also 
stated that its paramount interest for issuing the regulation was 
protecting the fiscal integrity of the Medicaid program and that its 
proposed transition periods balance state budget issues with that 
protection. HCFA also emphasized that a transition policy was needed to 
recognize (1) that immediate implementation of the new UPLs could 
disrupt budget arrangements for states that had relied on the federal 
funds generated by such arrangements, and (2) that the length of the 
transition period should be based on how long a state had had an 
excessive UPL payment arrangement in place. HCFA reported that Congress 
affirmed in BIPA the use of budgetary reliance as the basis for 
granting transition periods of different lengths. Specifically, in 
response to commenters' opposition to having different transition 
periods based on the effective date of a state plan amendment, the 
agency stated in the preamble to its regulation that: 

[t]he reliance concept is applicable because these funds have been 
built into State and provider budgets for longer periods of time. We 
note also that in enacting a third transition period for States with 
excessive payment methodologies in place on or before October 1, 1992, 
the Congress has ratified our approach to establish transition periods 
based on a "reliance concept."[Footnote 29]

Although Nebraska and Wisconsin did have supplemental payment 
arrangements for nursing homes in place as of October 1, 1992, neither 
state had an arrangement that aggregated UPLs of local-government and 
private nursing homes, made extremely large payments to units of local 
governments on behalf of a few locally owned nursing homes, or required 
the local governments to return the bulk of payments to the 
state.[Footnote 30] In contrast, Pennsylvania did have a long-standing 
UPL arrangement, that is, an arrangement that aggregated payments and 
resulted in large payments to local providers that exceeded their 
costs.

Nebraska's UPL History: 

Nebraska did not establish the type of UPL arrangement considered 
problematic until January 1998. CMS nonetheless determined that the 
state qualified for an 8-year transition period on the basis of a 
supplemental payment provision that the state established in September 
1992.[Footnote 31] Nebraska's 1992 supplemental payment provision was 
designed to allow certain local-government-owned nursing homes to 
receive payments exceeding the state's normal cap on Medicaid nursing 
home payments, although payments under this arrangement could not 
exceed the nursing homes' costs of providing Medicaid services. 
Nebraska's 1992 payment provision differed significantly from the UPL 
arrangement that HCFA was trying to curtail in that it did not 
aggregate UPLs across local-government and private nursing homes. It 
also did not make large payments to a few nursing homes and require 
that they be returned to the state. Using the January 1998 UPL 
arrangement as the basis for the transition period decision, Nebraska 
would have qualified for a 5-year transition period.

Wisconsin's UPL History: 

Although Wisconsin did not establish the type of UPL arrangement 
considered problematic until 2001, in February 2002 CMS determined that 
the state qualified for an 8-year transition period on the basis of a 
small supplemental payment provision the state established in SFY 1985. 
Wisconsin's 1985 payment provision enabled the state to claim federal 
matching funds for some expenditures made by county nursing homes for 
Medicaid patients that were not covered by the state's standard 
Medicaid payment rates. These supplemental payments were relatively 
small. For example, in SFY 1992, Wisconsin's supplemental payments 
totaled $15 million--resulting in a federal matching share of about $9 
million--and when combined with standard Medicaid payments to the 
nursing homes still totaled less than the nursing homes' costs for 
their Medicaid patients. Moreover, the 1985 payment provision did not 
aggregate UPLs across nursing homes as the basis for the nursing homes' 
payments; rather, the state capped total payments and payments to 
individual nursing homes at an amount set annually, typically far below 
the Medicaid upper limit. Finally, county nursing homes were not 
required to return supplemental payments to the state, and the state 
did not retain any of the federal funds claimed through the 1985 
provision.[Footnote 32] In 2001, however, Wisconsin established a UPL 
arrangement that did aggregate the UPLs of local-government and private 
nursing homes in order to make extremely large payments to three 
counties on behalf of five nursing homes--totaling $637 million--which 
allowed the state to generate and retain about $373 million in federal 
funds.[Footnote 33] As a result, before 2001 Wisconsin had no reliance 
on excessive UPL payments that would justify an 8-year transition 
period, given the purpose stated by CMS.

In 2001, we reported that HCFA should not have approved Wisconsin's 
2001 arrangement because the state did not seek approval for the 
arrangement until after the agency had taken steps to curtail such 
arrangements through regulation.[Footnote 34] In September 2001, HCFA 
determined that the arrangement it approved for Wisconsin should 
receive a 16-month transition period, because the state's arrangement 
was recently established.[Footnote 35] But in February 2002, the agency 
determined that Wisconsin qualified for an 8-year transition period. 
CMS officials indicated that they changed their position on Wisconsin's 
transition because the state had informed them, in November 2001, that 
the new arrangement continued a provision that had been in place before 
October 1992. After discussions with the state, CMS agreed that the 
2001 UPL arrangement continued the state's 1985 payment arrangement and 
that Wisconsin therefore had a long-standing UPL arrangement that would 
qualify the state for an 8-year transition. In our view, Wisconsin's 
2001 UPL arrangement differs substantially from the 1985 arrangement, 
and granting the state an 8-year transition period is inconsistent with 
the agency's stated objectives in curtailing excessive UPL schemes.

Pennsylvania's UPL History: 

In contrast to Nebraska and Wisconsin, Pennsylvania had the type of UPL 
arrangement identified as problematic by CMS that predated October 
1992. As a result, CMS's granting an 8-year transition period to 
Pennsylvania is consistent with the objectives for the UPL regulation 
and transition periods that the agency identified when issuing the 
final regulation. The state provided information that described a 1992 
arrangement and showed that Pennsylvania claimed federal matching funds 
based on an aggregated UPL for public and private providers, and made 
unusually large payments to 23 of the state's 47 county-operated 
nursing homes in amounts that exceeded their Medicaid costs.

Recent Payment Spikes Reflect a Lack of Long-standing Budgetary 
Reliance on UPL Arrangements: 

Given that Nebraska and Wisconsin started their aggregated UPL 
arrangements in 1998 and 2001, respectively, these states lacked long-
standing budgetary reliance on excessive UPL funds. The date when a 
state first established an aggregated UPL arrangement is important, 
because the beginning of such an arrangement typically produces a sharp 
increase, or spike, in supplemental payments made on behalf of local-
government nursing homes and, consequently, in the federal matching 
share of such payments. Such an increase represents the earliest point, 
in our view, that a state could have established a budgetary reliance 
on excessive matching dollars from UPL arrangements. For example, 
federal payments for supplemental payments to local-government nursing 
homes in Nebraska increased from about $830,000 in 1997 to nearly $53 
million in 1999--the first full-year of the state's UPL arrangement. 
Indeed, the federal share of state supplemental payments to county 
nursing homes spiked dramatically in both Nebraska and Wisconsin long 
after 1992, in 1998 and 2001, respectively (see fig. 3).

Figure 3: Federal Share of Supplemental Payments in Nebraska and 
Wisconsin through SFY 2001: 

[See PDF for image]

Note: Nebraska established its supplemental payment provision in 
September 1992 (SFY 1993), but did not make payments until SFY 1994. 
Wisconsin started its supplemental payment provision in SFY 1985, and 
we obtained payment information back to SFY 1990.

[End of figure]

Because of the 8-year transition periods granted to Nebraska and 
Wisconsin, these two states can obtain more in federal matching 
payments during their transition periods than they had received before 
the 2001 regulation. In our view, this is further evidence that the 
states did not have long-standing budgetary reliance on excessive 
federal funds and that 8-year transition periods are not warranted. Our 
analysis of historical and projected claims data estimates that, 
together, Nebraska and Wisconsin are eligible for at least $490 million 
more in federal matching funds during their transition periods than 
they actually obtained in the 9 years before the January 2001 
regulation. Specifically, Wisconsin is eligible for $936 million in 
excessive federal matching funds from UPL payments during its 8-year 
transition period, which is $385 million more than the $551 million it 
obtained from SFY 1992 through SFY 2000. Similarly, Nebraska is 
eligible for $248 million in federal matching funds from excess UPL 
payments during its 8-year transition period, which is $105 million 
more than it obtained from SFY 1994 through SFY 2000.[Footnote 36]

CMS Made Transition Period Decisions without Establishing How It Would 
Apply Broad Statutory Language: 

Before making its initial transition period decisions, CMS did not 
establish or convey to the states how it would interpret and apply the 
statutory criteria for a state to receive an 8-year transition period. 
According to the agency, its policy for making this determination was 
still being developed during the course of our review. Under BIPA, a 
state could qualify for an 8-year transition period if the UPL payment 
provision that resulted in a state's not complying with the January 
2001 regulation was in place on or before October 1, 1992, including a 
"successor" or "subsequent amendment" to an earlier payment provision 
or methodology. To qualify for an 8-year transition period, CMS 
informed us that states must have had a supplemental payment provision 
in place on or before October 1, 1992--that is, a provision allowing 
for enhanced or supplemental payments to providers. CMS officials made 
initial transition period decisions by assessing how long the UPL 
payment arrangements known to exist in 2000 or 2001 had been in place. 
Thus, CMS's 8-year transition period determinations depended on whether 
a current UPL payment provision was a "successor" or "subsequent 
amendment" to a 1992 supplemental payment provision or methodology. 
Before it notified states of their assigned transition periods in early 
spring 2002, CMS did not establish how it would determine whether a 
state's latest UPL arrangement was a "successor" of an arrangement the 
state had had in place on or before October 1, 1992.[Footnote 37]

During our review, CMS's explanation of how the agency determined 
whether a provision was a "successor" changed. In March 2003, CMS 
officials advised us that they considered a state's latest UPL payment 
provision to be a successor of an earlier qualifying payment provision 
if changes made by the state in later years were consistent with the 
basic payment principles of the provision in place on or before October 
1, 1992, and that payment methodologies that were not identical to the 
original methodology might not be considered successors. CMS stated in 
writing that the agency would consider changes updating the state's 
Medicaid plan for inflation and other cost factors to be successor 
provisions. CMS officials informed us that other changes, including 
changes in payment provisions, would not qualify as successor 
provisions. This interpretation, however, was not consistent with the 
agency's initial decisions to grant Nebraska and Wisconsin 8-year 
transition periods, because in 1992 these states had payment provisions 
in place that differed significantly from the UPL arrangements that 
they established in 1998 and 2001, respectively.

When we asked for clarification, CMS stated that the agency was still 
evaluating its definition of a successor provision. In August 2003, CMS 
officials informed us in writing that the description it gave us 
earlier was too restrictive. Instead, CMS wrote that provisions for 
payments to the same provider type would embody the same basic 
principles as the original provision and that any modification to the 
original supplemental payment provision would constitute a successor, 
as long as payments were made to the same type of provider. CMS further 
explained its evolving policy and interpretation of the law. The agency 
said that its preliminary interpretation of BIPA potentially 
disqualified any amendments from consideration as successors and could 
potentially "render meaningless" BIPA's provisions creating the 8-year 
transition period. CMS officials indicated that as the agency developed 
its interpretation, officials considered various interpretations 
including those that, when applied consistently to all states, resulted 
in no states qualifying for an 8-year transition period and in all 
states qualifying. We find the approach adopted by CMS troublesome 
because it links current problematic UPL arrangements with 
significantly different payment provisions that may not have been 
problematic.

Shorter Transition Periods Would Have Resulted in Less Federal 
Spending: 

Under the regulation allowing a state with an 8-year transition period 
to claim its full 2001 UPL payment until 2004, Nebraska and Wisconsin 
have been able to generate significantly more in excessive federal UPL 
payments than they could have under shorter transition periods 
consistent with CMS's stated purpose for the 2001 regulation. Had CMS 
based transition periods on the date that a state actually established 
a problematic UPL arrangement, the transition periods assigned would 
have been shorter, and allowable federal matching dollars would have 
totaled $551 million, which is $633 million less than the nearly $1.2 
billion that CMS is allowing the two states to claim. Through SFY 2003, 
the two states had already claimed about $497 million more than they 
would have been able to receive under shorter transition periods. The 
two states are eligible to claim a total of about $135 million more in 
federal matching funds during the remaining years of their 8-year 
transition periods that they would not be able to claim under shorter 
transition periods.

Nebraska would have received $102 million less in federal funding under 
a 5-year transition period, during which the state would have had to 
reduce its excessive payments (by 25 percent per year over 4 years, 
instead of 15 percent per year over 6 years). The payments allowed in 
Wisconsin are even more troubling, because in 2001 we found HCFA's 
approval of Wisconsin's UPL arrangement unjustified. In approving the 
arrangement, and then granting Wisconsin an 8-year transition period, 
the agency made the state eligible for about $936 million in excessive 
federal funds. Most of the federal funds have already been 
claimed.[Footnote 38] (See app. I for a more detailed description of 
the year-to-year differences in excessive federal payments.): 

A review by the HHS OIG concluded that long transition periods were not 
needed, and estimates by CBO point to significant federal savings if 
transition periods were ended sooner. On the basis of its reviews of 
UPL arrangements in six states in 2000, the OIG concluded that 
transition periods in HCFA's January 2001 regulation were longer than 
needed for states to adjust their spending to achieve the lower UPL 
ceilings established by the new regulation.[Footnote 39] The OIG 
recommended that CMS seek authority to eliminate or reduce transition 
periods, but CMS did not concur. Savings from shortening transition 
periods could be significant. According to estimates from CBO in 2003, 
the federal government could save $2.8 billion in 2004 and $7.3 billion 
over 5 years if transition periods were stopped after 2003 and all 
states were required to comply with the January 2001 regulation 
beginning in 2004.[Footnote 40]

CMS Has Strengthened Its Oversight of UPL Arrangements, but Concerns 
Remain: 

CMS has taken action to improve its oversight of state UPL 
arrangements, but its efforts do not go far enough to ensure that 
states' claims are for Medicaid-covered services provided to eligible 
beneficiaries. Such oversight is important because, even though CMS's 
regulations significantly narrow the UPL loophole, states can still 
claim excessive federal matching dollars for nursing home payments that 
exceed the facilities' actual costs. CMS's recent efforts include 
forming a team to centralize and coordinate the review of state UPL 
arrangements, bringing more uniformity to the process. The agency has 
also conducted financial management reviews of some states' UPL 
arrangements and identified and disallowed some inappropriate claims. 
CMS has not, however, completed detailed financial management reviews 
of states with the largest UPL arrangements or issued guidance 
instructing states how to appropriately calculate the UPL. Our review 
of UPL estimates in six states found wide variances in calculations and 
identified concerns about the accuracy of state estimates. Further, all 
six states are still using federal Medicaid funds obtained through UPL 
arrangements for non-Medicaid purposes or to increase the federal share 
of Medicaid expenditures over the established federal matching rate.

UPL Loophole Has Narrowed, but States Can Still Generate Excessive 
Federal Funds: 

Because states can aggregate payments to all local-government nursing 
facilities under one UPL, they can still generate substantial excessive 
federal matching payments beyond their standard Medicaid claims. Our 
analysis of CMS estimates indicates that under the January 2001 
regulation, states could still generate about $2 billion annually in 
federal matching funds through UPL arrangements with nursing homes by 
making payments that are substantially higher than a facility's cost of 
providing services. In Wisconsin, for example, CMS's January 2001 
regulation would still allow the state to claim federal matching funds 
on the basis of the UPL for all county-owned nursing homes in the 
state. The state could continue to make large UPL payments to a few 
counties participating in its UPL arrangement and generate federal 
matching funds for as much as $41 million more than the nursing homes' 
actual cost of providing services.

CMS Has Strengthened Oversight, but Issues Remain in Four Areas: 

CMS has taken a variety of actions to strengthen its oversight of state 
UPL nursing home arrangements (see table 2).[Footnote 41] It has, for 
example, formed a special team to coordinate review of UPL transition 
periods and other UPL arrangements, drafted guidelines for internal use 
in reviewing UPL arrangements, conducted financial management reviews 
in a few states to help ensure UPL payments were proper, and 
established some UPL-reporting requirements. Although these actions 
demonstrate CMS's efforts to curb UPL abuses, our review of several 
state UPL programs identified areas where stronger agency oversight is 
needed. For example, CMS has not issued guidance to states to help 
ensure that they use appropriate methods to calculate their UPLs, and 
it has not completed financial reviews of some states with the largest 
arrangements and longest transition periods. CMS also has not 
standardized its reporting requirements, but instead has asked states 
to provide information as part of its ongoing transition period 
reviews.

Table 2: Benefits and Shortcomings of CMS Actions to Strengthen 
Oversight of UPL Arrangements: 

CMS action: Formed National Institutional Reimbursement Team; 
Benefit: Coordinates CMS review and approval of initial UPL 
applications and amendments, conducts reviews of states' claims for 
federal matching funds during transition periods; 
Shortcomings: The volume and complexity of UPL submissions and related 
workload has resulted in delayed reviews and questionable decisions.

CMS action: Developed draft guidelines for reviewing UPL methodology; 
Benefit: Establishes guidelines and some consistent standards that CMS 
reviewers can use to assess state UPL methods and calculations; 
Shortcomings: Guidelines not finalized and similar guidance not yet 
provided to states.

CMS action: Conducted financial management reviews; 
Benefit: Verifies the accuracy of state UPL claims; 
identifies flaws in states' methods for calculating their UPLs and 
recommends corrective action; 
Shortcomings: Reviews limited in number and conducted mainly in smaller 
states, rather than in states with large UPL claims and long transition 
periods.

CMS action: Established annual and quarterly UPL reporting 
requirements; 
Benefit: Provides information that could be used to better monitor 
state UPL programs; 
Shortcomings: Annual reporting requirement not yet implemented; 
quarterly reports provide only aggregate data, rather than facility-
specific information needed to monitor payment levels. 

Source: GAO analysis of CMS actions.

[End of table]

National Institutional Reimbursement Team: 

To promote more timely and consistent review of state UPL arrangements, 
CMS formed the National Institutional Reimbursement Team (NIRT) in July 
2002. This reimbursement team centralizes the reviews and approvals of 
state proposals to change or institute new payment methods for 
institutions, including nursing homes and hospitals. As of September 
2003, the centralized team consisted of six members from CMS 
headquarters and four from its regional offices. Its major 
responsibilities are to make transition period decisions and reviews 
and to validate states' claims for federal matching funds during their 
transition periods.

The team has uncovered a number of inappropriate UPL claims. For 
example, in New Jersey, the team determined that the state's 2-year 
transition period ended in September 2002, but the state continued to 
make UPL payments above the limits established by the January 2001 rule 
and claimed $238 million in UPL payments that the team denied. In 
addition, the team identified and rejected a number of states' claims 
for retroactive payments. Some states attempted to retroactively change 
the way they calculated their UPLs so they could increase payments for 
prior periods, specifically for SFY 2000, the base year for transition 
periods. If such changes were made to the base year, increased payments 
would also carry forward and enable the states to make higher payments 
throughout their transition periods. But changes that do not comply 
with the state's current Medicaid plan are prohibited. The team 
prevented three states--New Hampshire, Oregon, and Washington--from 
claiming $30 million, $46 million, and $265 million, respectively, in 
federal matching funds for prohibited retroactive payments.[Footnote 
42]

Although NIRT has brought greater consistency to the oversight process 
and identified a number of unallowable UPL claims, it has been 
challenged by the volume, complexity, and variety of state UPL 
methodologies, as well as by its other responsibilities to review all 
changes that states propose involving payments to institutional 
providers, including hospitals and nursing homes.[Footnote 43] The 
team's heavy workload has contributed to delays and raises concern 
about the comprehensiveness of CMS's review and approval of states' 
claims during transition periods. For example, despite the team's 
reviews, we identified concerns with two states' methodologies that the 
team's review did not challenge. Our analysis of Pennsylvania's UPL 
methodology found shortcomings in the state's calculations; 
specifically, the state had overestimated its base-year ceiling by 
about $11 million, which could allow the state to claim more than $55 
million in excess federal matching funds during its 8-year transition 
period.[Footnote 44] In Michigan, NIRT reviewers did not identify an 
error in the state's 2000 UPL methodology, which used a higher wage 
index (from 2001) than the index that was actually in place for 2000. 
This error increased Michigan's UPL estimate by more than $8 million, 
allowing the state to claim almost $5 million more in federal matching 
funds in SFY 2000.

During our review, team officials told us that they had a new effort 
under way to hold states more accountable for how the federal funds are 
spent. Starting in August 2003, for any state plan amendments that 
proposed changing how the state would pay nursing homes or other 
institutions, the team requested information that had not previously 
been required. According to officials, the reimbursement team began 
asking states with proposed state plan amendments to provide detailed 
descriptions of the state's payment provision, including sources of 
state matching funds for supplemental payments; the extent that total 
payments under the new payment provision would exceed providers' costs; 
how the state would use the additional funds; and whether the state 
required payments to providers to be returned to the state (and if so, 
how the state planned to spend such funds). Team officials indicated 
that they would disapprove any proposed plan amendments from states 
that did not adequately respond to the team's requests for this 
information. As of October 2003, the team had asked 30 states with 
proposed state plan amendments to provide additional information, and 
the agency was in the process of receiving and reviewing states' 
initial responses. At that time, decisions about whether to approve or 
disapprove the plan amendments had not yet been made. CMS officials 
said that all states would be asked to provide detailed information on 
their payment provisions--irrespective of whether the state had 
qualified for a transition period.

Guidelines for Calculating a UPL: 

CMS has developed draft internal guidance for reviewing state UPL 
arrangements, but it has not finalized it or issued guidance for 
states, setting out acceptable methods for states to calculate their 
UPLs. Under the Medicaid reimbursement process, the federal 
contribution to a state's UPL payments is based on the state's estimate 
of its UPL; inflating the UPL, whether intentional or not, could bring 
a state a financial windfall. In August 2002, CMS's Division of 
Financial Management issued a draft UPL financial management review 
guide for the agency's internal use when auditing state UPL 
arrangements. The guide was designed to provide some instructions on 
performing a financial management review of Medicaid supplemental 
payments, and related UPL arrangements, and has been used by NIRT and 
reviewers who have examined states' UPL arrangements. The reimbursement 
team prepared a section of the guide on suggested approaches for 
calculating a UPL. According to the guide, unique circumstances could 
require the use of additional methods and procedures, determined by the 
reviewer's professional judgment and reviewed by the team.

To determine their UPLs, states are required to make reasonable 
estimates of what Medicare would pay for similar services. Although CMS 
allows states some flexibility in how they calculate their UPLs, its 
financial management review guide states that, generally, any 
methodology states elect to use should incorporate at least three 
factors: (1) consideration of any geographic variation within the state 
in Medicare rates, (2) determination of the appropriate Medicare 
payment rate for the medical and resource needs of comparable Medicaid 
patients, and (3) adjustment for the different services covered by 
Medicaid and Medicare. To determine if states were using UPL 
methodologies that incorporated such factors, we applied Medicare 
payment principles as well as standards from CMS's financial management 
review guide to review the UPL calculation methodologies used by six 
states that operate large UPL arrangements and have long transition 
periods.

Our examination of methodologies in the six states--Michigan, New York, 
Oregon, Pennsylvania, Washington, and Wisconsin--revealed concerns 
with UPL calculations in each state, including widely varying and 
potentially inaccurate calculation methods. In our view, these 
variations and potential inaccuracies can result in overestimates of 
the UPL. Concerns we identified with states' methods included 
overestimates of the number of Medicaid residents served and the use of 
incorrect Medicare rates for service locations. Oregon, for example, 
projected and used in its SFY 2000 UPL calculation about 163,000 more 
nursing home resident days than nursing homes actually provided. As a 
result, we estimate that the state's claim for SFY 2000 was inflated by 
$6 million. If this overestimate is not corrected, the state may 
receive at least another $22 million during its transition period, 
because SFY 2000 establishes the excessive payment amount allowed 
during that transition period. We also found that states used a variety 
of different approaches to estimate their UPLs. For example, states 
used different methodologies in calculating appropriate Medicare rates 
to account for changes that occurred when Medicare moved from a cost-
based reimbursement system to a prospective payment system starting in 
1998.[Footnote 45]

In addition, five of the six states we reviewed did not use a method 
for determining the appropriate Medicare payment rate, based on the 
medical and resource needs of comparable Medicaid patients, that was 
consistent with CMS's guide.[Footnote 46] Our analysis of the one state 
that did comply with CMS's method found that this approach, as 
expected, resulted in a lower and more accurate estimate of the UPL 
than the approach used by the other five states. Requiring states to 
follow a standard methodology could have helped prevent many of the 
discrepancies we identified. Appendix II summarizes our analysis of the 
methods that the states in our review used to calculate their UPLs.

Financial Management Reviews: 

CMS has also improved its ability to monitor state UPL programs by 
conducting a number of financial management reviews. CMS's Division of 
Financial Management, in conjunction with the regional offices, 
conducts annual audits, also known as financial management reviews, of 
state Medicaid expenditures and claims. Although examining UPL 
arrangements is not a mandatory item for every financial management 
review, in recent years CMS selected a number of states for focused UPL 
audits.[Footnote 47] Regional financial auditors familiar with states' 
financial management systems and expenditure levels perform these 
reviews. The reviews conducted to date have identified millions of 
dollars in unallowable UPL claims, and CMS has begun to recover 
improperly paid federal matching funds. For example, to prevent state 
UPL payments from escalating during transition periods, CMS's January 
2001 regulation limited payments to SFY 2001 levels. Recent financial 
management reviews in Louisiana and Missouri have identified federal 
matching payments of $116 million and $87 million, respectively, that 
CMS has determined were improper increases over their maximum allowed 
payments. CMS has initiated action to recoup these excessive payments.

Although these financial reviews have identified millions of dollars in 
potentially unallowable payments, CMS's financial management reviews 
have not been targeted to those states with the longest transition 
periods and largest claims. As of October 2003, reviewed states were 
those with 1-or 2-year transition periods. None of the 10 states with 
transition periods of 5 or 8 years had been reviewed (see table 3). 
While there is merit to focusing reviews on those states whose 
transition periods ended first, it is also important to ensure that 
states claiming large amounts of excessive funds for a long period of 
time--in particular states with 8-year transition periods that can 
claim 100 percent of their excessive payments for the first 3 years of 
the transition--have their UPL payment methods and claims reviewed for 
accuracy.

Table 3: Status of CMS Financial Management Reviews of Nursing Home UPL 
Arrangements, as of October 2003: 

Transition period: 8-year; State: Nebraska; 
Financial management review of nursing home arrangement: No; 
Financial management review of nursing home arrangement: 
Federal share of state UPL payments in SFY 2001 (dollars in millions): 
$55.

Transition period: 8-year; State: Pennsylvania; 
Financial management review of nursing home arrangement: No; 
Financial management review of nursing home arrangement: 
Federal share of state UPL payments in SFY 2001 (dollars in millions): 
$816.

Transition period: 8-year; State: Wisconsin; 
Financial management review of nursing home arrangement: No; 
Financial management review of nursing home arrangement: 
Federal share of state UPL payments in SFY 2001 (dollars in millions): 
$355.

Transition period: 5-year; State: Alabama; 
Financial management review of nursing home arrangement: No; 
Financial management review of nursing home arrangement: 
Federal share of state UPL payments in SFY 2001 (dollars in millions): 
$51.

Transition period: 5-year; State: Michigan; 
Financial management review of nursing home arrangement: No; 
Financial management review of nursing home arrangement: 
Federal share of state UPL payments in SFY 2001 (dollars in millions): 
$402.

Transition period: 5-year; State: New Hampshire; 
Financial management review of nursing home arrangement: No; 
Financial management review of nursing home arrangement: 
Federal share of state UPL payments in SFY 2001 (dollars in millions): 
$16.

Transition period: 5-year; State: New York; 
Financial management review of nursing home arrangement: No; 
Financial management review of nursing home arrangement: 
Federal share of state UPL payments in SFY 2001 (dollars in millions): 
$496.

Transition period: 5-year; State: North Dakota; 
Financial management review of nursing home arrangement: No; 
Financial management review of nursing home arrangement: 
Federal share of state UPL payments in SFY 2001 (dollars in millions): 
$26.

Transition period: 5-year; State: Oregon; 
Financial management review of nursing home arrangement: No; 
Financial management review of nursing home arrangement: 
Federal share of state UPL payments in SFY 2001 (dollars in millions): 
$96.

Transition period: 5-year; State: Washington; 
Financial management review of nursing home arrangement: No; 
Financial management review of nursing home arrangement: 
Federal share of state UPL payments in SFY 2001 (dollars in millions): 
$87.

Transition period: 2-year; State: Iowa; 
Financial management review of nursing home arrangement: Yes; 
Financial management review of nursing home arrangement: 
Federal share of state UPL payments in SFY 2001 (dollars in millions): 
$190.

Transition period: 2-year; State: Kansas; 
Financial management review of nursing home arrangement: Yes; 
Financial management review of nursing home arrangement: 
Federal share of state UPL payments in SFY 2001 (dollars in millions): 
$101.

Transition period: 2-year; State: Louisiana; 
Financial management review of nursing home arrangement: Yes; 
Financial management review of nursing home arrangement: 
Federal share of state UPL payments in SFY 2001 (dollars in millions): 
$415.

Transition period: 2-year; State: Missouri; 
Financial management review of nursing home arrangement: Yes; 
Financial management review of nursing home arrangement: 
Federal share of state UPL payments in SFY 2001 (dollars in millions): 
$134.

Transition period: 2-year; State: New Jersey; 
Financial management review of nursing home arrangement: Yes; 
Financial management review of nursing home arrangement: 
Federal share of state UPL payments in SFY 2001 (dollars in millions): 
$445.

Transition period: 2-year; State: South Dakota; 
Financial management review of nursing home arrangement: No; 
Financial management review of nursing home arrangement: 
Federal share of state UPL payments in SFY 2001 (dollars in millions): 
$32.

Transition period: 2-year; State: Tennessee; 
Financial management review of nursing home arrangement: No; 
Financial management review of nursing home arrangement: 
Federal share of state UPL payments in SFY 2001 (dollars in millions): 
$76.

Transition period: 1-year; State: Virginia; 
Financial management review of nursing home arrangement: Yes; 
Financial management review of nursing home arrangement: 
Federal share of state UPL payments in SFY 2001 (dollars in millions): 
$179.

Transition period: None[A]; State: Arkansas; 
Financial management review of nursing home arrangement: No; 
Financial management review of nursing home arrangement: 
Federal share of state UPL payments in SFY 2001 (dollars in millions): 
$9.

Transition period: None[A]; State: Colorado; 
Financial management review of nursing home arrangement: No; 
Financial management review of nursing home arrangement: 
Federal share of state UPL payments in SFY 2001 (dollars in millions): 
$6.

Transition period: None[A]; State: Georgia; 
Financial management review of nursing home arrangement: No; 
Financial management review of nursing home arrangement: 
Federal share of state UPL payments in SFY 2001 (dollars in millions): 
$68.

Transition period: None[A]; State: Indiana; 
Financial management review of nursing home arrangement: No; 
Financial management review of nursing home arrangement: 
Federal share of state UPL payments in SFY 2001 (dollars in millions): 
$7.

Transition period: None[A]; State: Kentucky; 
Financial management review of nursing home arrangement: No; 
Financial management review of nursing home arrangement: 
Federal share of state UPL payments in SFY 2001 (dollars in millions): 
$17.

Transition period: None[A]; State: Mississippi; 
Financial management review of nursing home arrangement: No; 
Financial management review of nursing home arrangement: 
Federal share of state UPL payments in SFY 2001 (dollars in millions): 
$11.

Transition period: None[A]; State: Montana; 
Financial management review of nursing home arrangement: No; 
Financial management review of nursing home arrangement: 
Federal share of state UPL payments in SFY 2001 (dollars in millions): 
$4.

Transition period: None[A]; State: South Carolina; 
Financial management review of nursing home arrangement: No; 
Financial management review of nursing home arrangement: 
Federal share of state UPL payments in SFY 2001 (dollars in millions): 
$0.1. 

Source: CMS.

[A] These states' UPL arrangements were too new to qualify for a 
transition period; federal share applies to federal fiscal year 2001, 
2002, or 2003.

[End of table]

Annual and Quarterly UPL Reporting Requirements: 

Until recently, states were not required to report separately on the 
amount of Medicaid funds they were claiming for nursing home UPL 
arrangements. As a result, information on the number of states with 
such arrangements and on the total amount of funds claimed through 
these arrangements has not been readily available. Starting in 2001, 
CMS established two reporting requirements to help remedy this lack of 
information. First, as of October 2001, CMS required all states with 
UPL arrangements to report quarterly the total amount they were 
claiming in Medicaid funds for their arrangements. Second, to 
facilitate the monitoring of UPL programs in states with transition 
periods, CMS incorporated into a January 2002 regulation an annual UPL 
reporting requirement for the duration of each state's transition 
period. Specifically, states that are eligible for a transition period 
are required to report to CMS the total Medicaid payments made to each 
facility, as well as a reasonable estimate of the amount that would be 
paid for the services under Medicare payment principles.[Footnote 48]

Under the October 2001 reporting requirement, states are reporting the 
total amount of their UPL claims on a quarterly basis. Although these 
reports provide summary information on a statewide basis, they do not 
identify the amount paid to each nursing home--important information 
that could help CMS oversee changes in claimed amounts and in the 
potential for continued aggregated payments to a few nursing homes. 
Quarterly reporting on payments made to each nursing home could also 
allow CMS to detect dramatic changes in supplemental payments, conduct 
a more timely review of states' payments, and take more immediate 
action than waiting for the results of annual reports.

As of October 2003, CMS had not taken steps to implement the 2002 
reporting requirement to collect facility-specific information on a 
consistent and continuing basis from states with transition periods. 
For example, CMS has not provided guidance to states as to the required 
format for these reports or the time frames for reporting. According to 
CMS officials, the agency has gathered some payment information on 
individual nursing homes as part of its transition period decision-
making process, but it has not established a standard format or 
reporting time frames. Such an ad hoc approach does not ensure that CMS 
gathers the information it needs to monitor states' UPL arrangements 
during transition periods.

Some States Use Funds from UPL Arrangements for Non-Medicaid Purposes 
or to Inappropriately Increase the Federal Share of Medicaid 
Expenditures: 

Despite efforts to improve oversight of state UPL schemes, there is no 
assurance that states are not using excessive federal Medicaid UPL 
matching funds for non-Medicaid purposes or to inappropriately increase 
the federal share of Medicaid expenditures. Federal Medicaid matching 
funds are intended for Medicaid-covered services for eligible 
individuals on whose behalf payments are made.[Footnote 49] But states 
reported continuing to use UPL arrangements to obtain federal matching 
funds that are spent for non-Medicaid purposes or to effectively 
increase those states' federal matching rates. Each of the six states 
we examined with large UPL arrangements reported using the proceeds 
from its UPL arrangement for a variety of Medicaid and non-Medicaid 
purposes. Similar to past reported uses of UPL funds,[Footnote 50] some 
states in our review put excessive funds from UPL arrangements into the 
state's general fund, which the state may or may not use for Medicaid 
purposes. States also used excessive federal funds obtained through UPL 
arrangements to pay for the state's share of its Medicaid program. In 
this way, federal funds are "recycled" to generate additional federal 
funds, effectively increasing those states' federal match rate. In 
Wisconsin, for example, we estimate that by obtaining excessive federal 
matching payments and using these funds as the state share of other 
Medicaid expenditures, the state effectively increased the federal 
matching share of its total Medicaid expenditures from 59 percent to 68 
percent in SFY 2001. States often provide a small amount of the total 
federal UPL match to the county nursing homes on whose behalf it was 
claimed but retain the majority of the funds. Table 4 provides further 
information on the reported uses of UPL funds in the six states we 
reviewed.

Table 4: Selected States' Use of Funds Generated through UPL 
Arrangements: 

State: Michigan; Use: Funds generated by the state's UPL arrangement 
are deposited in the state's general fund but are tracked separately as 
a local fund source. These local funds are earmarked for future 
Medicaid expenses and used as the state match, effectively recycling 
federal UPL matching funds to generate additional federal Medicaid 
matching funds.

State: New York; Use: Funds generated by the state's UPL arrangement 
are deposited into its Medical Assistance Account. Proceeds from this 
account are used to pay for the state share of the cost of Medicaid 
payments, effectively recycling federal funds to generate additional 
federal Medicaid matching funds.

State: Oregon; Use: Funds generated by the state's UPL arrangement are 
being used to finance education programs and other non-Medicaid health 
programs. UPL matching funds recouped from providers are deposited into 
a special UPL fund. Facing a large budget deficit, a February 2002 
special session of the Oregon legislature allocated the fund balance, 
about $131 million, to finance kindergarten to 12th grade education 
programs. According to state budget documents, the UPL funds are being 
used to replace financing from the state's general fund.

State: Pennsylvania; Use: Funds generated by the state's UPL 
arrangement are used for a number of Medicaid and non-Medicaid 
purposes, including long-term care and behavioral health services. In 
SFY 2001-2003 the state generated $2.4 billion in excess federal 
matching funds, of which 43 percent was used for Medicaid expenses 
(recycled to generate additional federal matching funds), 6 percent was 
used for non-Medicaid purposes, and 52 percent was unspent and 
available for non-Medicaid uses (does not total 100 percent because of 
rounding).

State: Washington; Use: Funds generated by the state's UPL arrangement 
are commingled with a number of other revenue sources in a state fund. 
The fund is used for various state health programs, including a state-
funded basic health plan, public health programs, and health benefits 
for home care workers. A portion of the fund is also transferred to the 
state's general fund. The fund is also used for selected Medicaid 
services and the state's SCHIP program, which effectively recycles the 
federal funds to generate additional federal Medicaid matching funds.

State: Wisconsin; Use: Funds generated by the state's UPL arrangement 
are deposited in a state fund, which is used to pay for Medicaid-
covered services in both public and private nursing homes. Because the 
state uses these payments as the state share, the federal funds are 
effectively recycled to generate additional federal Medicaid matching 
funds.

Sources: CMS and states.

[End of table]

Further Action Would Help Address Continuing Concerns with UPL Schemes: 

Although Congress and CMS have taken significant steps to help curb 
inappropriate UPL arrangements, the growing use of state UPL nursing 
home arrangements poses continuing concerns about states' ability to 
aggregate payments under the UPL. An outstanding recommendation from an 
earlier GAO report would, if implemented, strengthen CMS oversight of 
UPL programs and help mitigate these concerns. In August 1994, we 
recommended that Congress consider prohibiting Medicaid payments that 
exceeded actual costs for any government-owned facility.[Footnote 51] 
The recommendation was aimed at eliminating states' ability to 
aggregate payments to multiple nursing homes (regardless of category of 
ownership) under one UPL and to make large payments that exceed a 
facility's costs, by essentially creating facility-specific limits. In 
September 2001, the OIG made a similar recommendation, that CMS 
implement facility-specific upper limits based on facility costs. At 
that time, CMS agreed that facility-specific limits may be the most 
effective approach to ensure that UPL payments are reasonable, but the 
agency also stated that, if possible, it wanted to maintain states' 
flexibility in how they paid facilities. CMS indicated that it did not 
want to impose facility-specific limits until it was clear that its 
January 2001 regulation had not solved the problem. CMS indicated that 
if this approach proved inadequate, it would consider additional 
measures, such as facility-specific limits.

Conclusions: 

The actions that Congress and CMS have taken to limit states' ability 
to inappropriately claim federal Medicaid funds through UPL financing 
schemes have helped strengthen the fiscal integrity of Medicaid's state 
and federal partnership. With BIPA, and CMS's implementing regulation, 
the UPL loophole has been narrowed. CMS has subsequently further 
strengthened its oversight of states' continuing UPL arrangements, 
including launching a major initiative in August 2003 aimed at holding 
states more accountable for using Medicaid funds only for Medicaid 
purposes.

At the same time, problems with federal oversight of state financing 
schemes remain. Although CMS described transition periods as a 
mechanism for assisting states that had relied on excessive UPL funds 
to adjust to the new and reduced limits, CMS had little historical 
information on, and uncertain criteria for, its decisions before it 
assigned states their transition periods. We found that, while 
permissible under the statute, CMS's decisions to grant Nebraska and 
Wisconsin an 8-year transition period were inconsistent with the 
agency's stated objectives because neither state's circumstances 
suggest a budgetary reliance on excessive UPL funds for long periods. 
In our view, these decisions did not reflect a balancing of Medicaid's 
fiscal integrity with the state budget issues as cited by CMS. Further, 
CMS's current transition period policy raises broader concerns about 
the program's fiscal integrity in the future, as states without any 
long-standing budgetary reliance on federal funds from UPL arrangements 
may claim that they qualify for an 8-year transition period and 
continue to submit claims for excessive funds from the federal 
government. These decisions establish a questionable precedent with 
unknown ramifications for the Medicaid budget.

In addition to the lack of information CMS has on states' UPL 
arrangements and concerns about its transition period policy and 
decisions, our work points to other concerns with current financing 
arrangements and the limits of CMS's ability to oversee them. Our work 
and the financial reviews CMS has conducted indicate that states 
continue to submit improper claims for federal matching funds. Claims 
are based on widely varying and sometimes inaccurate methods for 
estimating what Medicare would pay, a problem stemming from the 
agency's lack of guidance to states on allowable methods for 
calculating their UPLs. Although our work identified some overpayments 
to states based on inappropriately calculated UPLs, the full extent of 
problems with states' methods is unknown, in part because CMS has not 
completed its reviews of all states with long transition periods and 
with large claims for federal matching payments. In addition, although 
CMS has improved UPL reporting requirements, such reporting is still 
limited. This dearth of information and the complexities of states' 
financing schemes are likely to continue to challenge and complicate 
the agency's oversight role.

Given the continuing oversight challenges and significant financial 
risks to the federal government that remain through these arrangements, 
the UPL loophole should be closed altogether. Although the federal 
government may need to have an upper payment limit or some other means 
to help ensure that Medicaid payments are economical and efficient, any 
limit should be just that: an upper bound to what states can pay to 
individual facilities for Medicaid-covered services and for which they 
can receive federal reimbursement. Consistent with this approach and 
with sound fiscal policy, federal matching payments should be based on 
the lower of the established limit or facilities' actual Medicaid 
costs. Because the UPL provision continues to allow states to claim 
hundreds of millions of dollars above what they actually pay for care, 
we believe the earlier action we recommended to Congress--that it 
consider prohibiting Medicaid payments that exceed costs to any 
government-owned facility--is still valid and would help to further 
safeguard federal Medicaid funds.

Matter for Congressional Consideration: 

We believe Congress should continue its efforts to close the UPL 
loophole and prevent further claims from arrangements that undermine 
the fiscal integrity of the Medicaid program. In addition to 
reiterating our previous recommendation to Congress to limit Medicaid 
payments to providers' costs, we believe action is required to address 
the impact of CMS's transition policy and decisions on program 
integrity. We suggest that Congress consider ending the 8-year 
transition periods for states with excessive nursing home UPL 
arrangements, with a consideration for any state that has demonstrated 
a long-standing budgetary reliance on the federal funds.

Recommendations for Executive Action: 

To protect the fiscal integrity of the Medicaid program, we recommend 
that the Administrator of CMS take the following two actions: 

* establish criteria for making transition period decisions that are 
consistent with the objectives described in CMS's January 2001 UPL 
regulation, and: 

* reconsider the agency's initial decisions to grant Nebraska and 
Wisconsin 8-year transition periods.

To further improve UPL oversight, we also recommend that the 
Administrator of CMS take the following three actions: 

* establish uniform guidance for states, which would set forth 
acceptable methods to calculate UPLs;

* expedite the financial management reviews of states with UPL 
arrangements, assigning high priority to reviews of states with 5-and 
8-year transition periods, including those we identified as having 
methodological problems; and: 

* improve state reporting on UPL arrangements, such as implementing the 
current requirement for states with transition periods to report 
payments on a facility-specific basis, and requiring such reports for 
all states with a UPL arrangement.

Agency and State Comments and Our Evaluation: 

We provided a draft of this report for comment to CMS and the states of 
Michigan, Nebraska, New York, Oregon, Pennsylvania, Wisconsin, and 
Washington. CMS and all the states except Washington provided comments. 
Pennsylvania provided a technical comment. Comments from Michigan, 
Nebraska, New York, Oregon, and Wisconsin are found in appendixes IV 
through VIII, respectively.

CMS's Comments and Our Evaluation: 

CMS generally concurred with our recommendations to improve its UPL 
oversight, including establishing uniform guidance for states to 
calculate their UPL and strengthening its requirements for states to 
report UPL activity. CMS concurred in part with our recommendation to 
expedite financial management reviews of states with 5-and 8-year 
transition periods, indicating that it had made more progress than our 
draft reflected, but also agreeing to review those states that it had 
not yet reviewed. CMS concurred with our recommendation that it 
establish criteria for making transition period decisions that are 
consistent with the objectives described in issuing its January 2001 
UPL regulation, but the agency also commented that it had already done 
so. CMS did not concur with our recommendation that it reconsider its 
initial decisions to grant Nebraska and Wisconsin 8-year transition 
periods. CMS stated that its current policy and transition period 
decisions reflect both the intent of the UPL regulation and 
congressional intent to allow for an 8-year transition period.

CMS also stated that its transition period policy and decisions are 
legally supportable and, in summarizing its concerns, provided a 
detailed explanation of the legal basis for its current policy and 
decisions. We acknowledge that CMS's 8-year transition period policy 
and its initial transition period decisions are permissible under the 
statute, and we have clarified our report accordingly. CMS also stated, 
however, that it considered various interpretations of BIPA's 
provisions for an 8-year transition period, suggesting that the 
provisions were susceptible to alternative interpretations. We disagree 
that CMS has already established transition period criteria in a manner 
consistent with the objectives it identified in the preamble to the 
regulation implementing BIPA to help ensure the fiscal integrity of the 
Medicaid program. We note that, although CMS maintains that it has 
applied its current policy consistently in making its decisions, its 
decisions were made before the agency defined key provisions of its 
interpretation--in particular, what constitutes a "successor" payment 
provision. We are concerned that CMS's current policy could invite more 
states to claim that they qualify for an 8-year transition period, 
because they had a supplemental payment provision in place on or before 
October 1, 1992, particularly given the scant information CMS has 
available on states' historical payment arrangements. Because CMS 
disagrees with our view, we believe Congress should consider addressing 
this issue.

CMS also commented on our draft report's treatment of the agency's 
basis for and explanations of its transition period policy and 
decisions. CMS expressed concerns that the draft report ignored, almost 
entirely, the basis for and explanations of CMS's policy 
interpretations and decisions. CMS also stated that the report 
inaccurately suggested that the agency changed policy interpretations 
in the middle of UPL implementation in order to provide preferential 
treatment to certain states. We do not believe or report that CMS 
changed its policy to favor certain states. Moreover, we did not ignore 
CMS's explanations of its evolving policy; in fact, as CMS pointed out, 
we obtained and analyzed considerable oral and written documentation 
throughout our review in order to understand CMS's position. Further, 
we believe that it was appropriate to report on the evolution of CMS's 
policy during our review. CMS's policy development occurred months 
after states had been informed of their transition periods, and more 
than 2 years after BIPA was enacted. Our review, and CMS's evolving 
policy and interpretations, took place months after the agency had sent 
letters to Nebraska, Wisconsin, and Pennsylvania, notifying them that 
they would receive an 8-year transition period.[Footnote 52] Our 
concern was, and remains, that the agency was developing key aspects of 
its policy, such as the definition of a "successor" provision, after it 
notified states about their transition periods. We acknowledge that CMS 
staff had a very complex challenge in making transition period 
decisions. But we suggest that, given the substantial financial impact 
of these decisions, CMS should have developed its policy position and 
obtained and reviewed states' documentation before it notified states 
of their initial transition periods and allowed states to claim funds 
based on these decisions. In response to CMS's concern that we do not 
adequately reflect the agency's current policy, we have modified the 
report to do so.

In its cover letter, CMS also expressed four general concerns: 

* The stated objectives included in the draft report differ from those 
in the letter notifying the Secretary of the initiation of the 
assignment. Our final reporting objectives differed from the questions 
posed at the start of our assignment because we agreed with our 
congressional requesters, soon after we initiated work, to broaden the 
scope of work to include additional states and to provide a broader 
perspective on CMS's oversight of state UPL arrangements. We 
communicated these revised objectives to CMS staff at the time these 
revisions occurred and at meetings held during our review.

* The illustrations in the draft report of UPL-related spending in 
Nebraska and Wisconsin do not fairly and accurately represent the 
amounts allowed by CMS. The illustrations of Nebraska and Wisconsin's 
past payment levels (figure in "Highlights" and fig. 3 in the report) 
are based on payment amounts reported to us by the states and CMS. 
Consequently, we believe our illustrations are fair and accurate in the 
context they are presented--to summarize the federal share of states' 
historical supplemental payments.

* The report does not consider the limited CMS resources available to 
determine transition periods and amounts. We believe our report 
captures the challenges faced by CMS staff responsible for making UPL 
transition period decisions and policy and for reviewing state plan 
amendments--including tremendous workload and limited personnel (see 
pp. 28-29). We also report that CMS had limited historical information 
on states' UPL arrangements for assigning transition periods and 
determining how much in federal funds each state would be allowed to 
claim, and that some states' responses to CMS's requests for 
information were not received for weeks or months (see p. 14-15).

* The report does not acknowledge that GAO staff told CMS officials 
that the agency's interpretation of two statutory terms was reasonable 
and legally supportable. We have clarified our report to explicitly 
acknowledge that, while CMS's policy and decisions were permissible 
under BIPA, they departed significantly from the stated objectives of 
the agency's UPL regulations.

CMS's written comments appear in appendix III. CMS also provided 
technical comments, which we considered and incorporated as 
appropriate.

Nebraska's and Wisconsin's Comments and Our Evaluation: 

Nebraska and Wisconsin also disagreed with our recommendation that CMS 
reconsider its initial 8-year transition period decisions. Nebraska 
stated that it complied in good faith with the Medicaid regulations and 
that reinterpreting the decision to grant the state an 8-year 
transition period would be unacceptable. Wisconsin stated that it was 
eligible for an 8-year transition period under the criteria established 
by BIPA and provided an extensive explanation of the basis for this 
view. (Nebraska's comments appear in app. V, and Wisconsin's comments 
and our detailed response appear in app. VIII.): 

In addition to disagreeing with our position that its 8-year transition 
period was unjustified, Wisconsin commented on our approach for 
estimating the state's UPL payments during state fiscal year 2000, 
stating that its methodology was appropriate and that the one we used 
was flawed. We disagree that our methodology was in error, since other 
states had taken the same approach and CMS had approved their 
methodologies. We agree, however, that Wisconsin's methodology could 
also be considered reasonable, and we have revised our report 
accordingly. Our concern remains that, because CMS has not defined what 
constitutes a "reasonable" methodology, states' methods vary widely, 
and states are free to use methods designed to maximize their UPL 
payments. This lack of consistency formed the basis for our 
recommendation that CMS establish uniform guidance for states that 
would set forth acceptable methods to calculate UPLs, and CMS 
concurred. Wisconsin also disagreed with our concern that HCFA should 
not have approved the state's UPL arrangement when it was initially 
proposed in 2001. The state asserts that its 2001 state plan amendment 
was legal and appropriate and disagrees with the conclusion of our 2001 
report that CMS's approval of the arrangement was unjustified. 
[Footnote 53] We disagree, and refer to our response to the state's 
comments in our 2001 report.

New York's Comments and Our Evaluation: 

New York's comments discussed the concept of local-government cost 
sharing through intergovernmental transfers (IGT). New York explained 
that its Medicaid state plan has, since its inception, included local 
government cost sharing and that IGTs of local-government funds to 
finance Medicaid services is integral to the state's Medicaid financing 
statutes. We agree that IGTs are a legitimate tool used by state and 
local governments to carry out their shared governmental functions, 
including collecting revenues and making expenditures for government 
services. Moreover, the Medicaid statute allows local governments to 
contribute up to 60 percent of the state's share of Medicaid 
expenditures. We do not take issue with New York's use of IGTs to 
enable local governments to transfer funds to the state. We do, 
however, disagree with the state's implication that its UPL financing 
scheme is appropriate simply because the IGT mechanism is legally 
established. New York also provided several technical comments that we 
incorporated where appropriate. New York's comments appear in appendix 
VI.

Other States' Comments and Our Evaluation: 

Three other states (Michigan, Oregon, and Pennsylvania) provided 
comments summarized below.

* Michigan expressed concern that the error in its UPL methodology that 
we report is not substantive and said that the error has been corrected 
for subsequent years; it recommended deleting the audit finding. 
Michigan further said that its methodology did not account for an 
increase in Medicare rates in SFY 2000 that would have offset the 
overestimated claim based on its calculation. While we agree that this 
error did not have a substantial impact on the state's claims, we 
maintain that the state's error was relevant to our report, which 
includes concerns with CMS's oversight of the accuracy of states' 
claims for UPL-related payment.

* Oregon did not take issue with the report's general substantive 
findings, except for the characterization that states were engaged in 
unauthorized activities. The state commented that CMS's oversight of 
state UPL arrangements was consistent with the regulations because it 
afforded states some flexibility in calculating a reasonable estimate 
of what Medicare would have paid for Medicaid services. We acknowledge 
that state flexibility is an integral aspect of the Medicaid program, 
allowing states broad discretion to establish priorities to cover and 
pay for populations and services.

* Pennsylvania provided a single technical comment about the number of 
county-owned nursing homes, which we incorporated into our final 
report.

As arranged with your offices, unless you publicly announce the 
contents of this report earlier, we plan no further distribution of 
this report until 30 days after its issuance date. At that time, we 
will send copies of this report to the Secretary of Health and Human 
Services, the Acting Administrator of the Centers for Medicare & 
Medicaid Services, and other interested parties. We will also make 
copies available to others upon request. In addition, the report will 
be available at no charge on the GAO Web site at http: //www.gao.gov.

If you or your staff have any questions, please contact me at (202) 
512-7118. Another contact and other major contributors are included in 
appendix IX.

Signed by: 

Kathryn G. Allen: 
Director, Health Care--Medicaid and Private Health Insurance Issues: 

[End of section]

Appendix I: Comparison of Federal Funds under 8-Year and Shorter 
Transition Periods: 

We assessed the difference between what Nebraska and Wisconsin are 
slated to receive under the 8-year transition periods they have been 
granted, and what they would have received under shorter transition 
periods consistent with CMS's stated objectives for the 2001 
regulation. Specifically, we compared annual federal matching funds 
claimed by Nebraska and Wisconsin under 8-year transition periods with 
those the states would have received had CMS based its decisions on 
when the states actually started the arrangements giving rise to 
excessive federal payments (see table 5).

Because Nebraska's upper payment limit (UPL) arrangement started in 
1998, the state would qualify for a 5-year transition period. Under 
this transition period, the state would have 1 year less than under the 
approved 8-year transition period to claim the full excessive payment 
before starting to phase it out, and the state would have to phase out 
the excessive payment more quickly. As a result, the state would be 
eligible for $102 million less in federal matching payments.

On the basis of when Wisconsin's UPL arrangement began, its transition 
period would decrease from 8 years to 16 months. Because the state did 
not have a problematic UPL arrangement on or before October 1, 1992, 
and given CMS's stated objectives for the January 2001 regulation and 
transition periods, in our view the state should have received no more 
than the 16-month transition period.[Footnote 54] In that event, the 
state would have had to stop claiming federal matching funds from its 
UPL arrangement on November 5, 2001, making it eligible for about $531 
million less in federal matching funds.[Footnote 55]

Wisconsin had, as of the end of state fiscal year (SFY) 2003, already 
claimed most of the excessive funds it can claim during its transition 
period--more than $895 million in federal matching funds--for the 3-
year period from SFY 2001 through 2003. As illustrated in table 5, the 
amount in excessive federal payments drops for Wisconsin after SFY 2003 
and continues to decline in subsequent years. This large drop occurs 
because excessive payments through SFY 2003 are based on the amount of 
excessive payments in SFY 2001, the first year of Wisconsin's UPL 
arrangement. In contrast, starting in SFY 2004, the allowed excessive 
payment amount is a percentage of excessive payments in SFY 2000, the 
year that Wisconsin had a much smaller supplemental payment 
arrangement.

Table 5: Excessive Federal Payments, State Fiscal Years 2001-2009, for 
Two States under Different Transition Periods: 

Dollars in millions: 

State: Nebraska; Transition period: 8-year assigned by CMS; 
State fiscal year: 2001: $45; 
State fiscal year: 2002: $45; 
State fiscal year: 2003: $30; 
State fiscal year: 2004: $38; 
State fiscal year: 2005: $31; 
State fiscal year: 2006: $25; 
State fiscal year: 2007: $18; 
State fiscal year: 2008: $11; 
State fiscal year: 2009: $4; 
Total: $248.

State: Nebraska; Transition period: 5-year[A]; 
State fiscal year: 2001: 45; 
State fiscal year: 2002: 45; 
State fiscal year: 2003: 23; 
State fiscal year: 2004: 22; 
State fiscal year: 2005: 11; 
State fiscal year: 2006: 0; 
State fiscal year: 2007: 0; 
State fiscal year: 2008: 0; 
State fiscal year: 2009: 0; 
Total: 146.

State: Nebraska; Transition period: Difference[B]; 
State fiscal year: 2001: [Empty]; 
State fiscal year: 2002: [Empty]; 
State fiscal year: 2003: 8; 
State fiscal year: 2004: 16; 
State fiscal year: 2005: 20; 
State fiscal year: 2006: 25; 
State fiscal year: 2007: 18; 
State fiscal year: 2008: 11; 
State fiscal year: 2009: 4; 
Total: $102.

State: Wisconsin; Transition period: 8-year assigned by CMS; 
State fiscal year: 2001: 300; 
State fiscal year: 2002: 298; 
State fiscal year: 2003: 297; 
State fiscal year: 2004: 13; 
State fiscal year: 2005: 10; 
State fiscal year: 2006: 8; 
State fiscal year: 2007: 6; 
State fiscal year: 2008: 4; 
State fiscal year: 2009: 0.4; 
Total: 936.

State: Wisconsin; Transition period: 16-month[C]; 
State fiscal year: 2001: 300; 
State fiscal year: 2002: 105; 
State fiscal year: 2003: 0; 
State fiscal year: 2004: 0; 
State fiscal year: 2005: 0; 
State fiscal year: 2006: 0; 
State fiscal year: 2007: 0; 
State fiscal year: 2008: 0; 
State fiscal year: 2009: 0; 
Total: 405.

State: Wisconsin; Transition period: Difference[B]; 
State fiscal year: 2001: [Empty]; 
State fiscal year: 2002: 194; 
State fiscal year: 2003: 297; 
State fiscal year: 2004: 13; 
State fiscal year: 2005: 10; 
State fiscal year: 2006: 8; 
State fiscal year: 2007: 6; 
State fiscal year: 2008: 4; 
State fiscal year: 2009: 0.4; 
Total: $531.

Total[D]; Transition period: 8-year assigned by CMS; 
State fiscal year: 2001: 345; 
State fiscal year: 2002: 343; 
State fiscal year: 2003: 327; 
State fiscal year: 2004: 51; 
State fiscal year: 2005: 42; 
State fiscal year: 2006: 33; 
State fiscal year: 2007: 24; 
State fiscal year: 2008: 15; 
State fiscal year: 2009: 5; 
Total: 1,184.

Total[D]; Transition period: Shorter; 
State fiscal year: 2001: 345; 
State fiscal year: 2002: 149; 
State fiscal year: 2003: 23; 
State fiscal year: 2004: 22; 
State fiscal year: 2005: 11; 
State fiscal year: 2006: [Empty]; 
State fiscal year: 2007: [Empty]; 
State fiscal year: 2008: [Empty]; 
State fiscal year: 2009: [Empty]; 
Total: 551.

Total[D]; Transition period: Difference[B]; 
State fiscal year: 2001: 0; 
State fiscal year: 2002: 194; 
State fiscal year: 2003: 304; 
State fiscal year: 2004: 28; 
State fiscal year: 2005: 31; 
State fiscal year: 2006: 33; 
State fiscal year: 2007: 24; 
State fiscal year: 2008: 15; 
State fiscal year: 2009: 5; 
Total: $633. 

Source: GAO analysis of state payment data and estimates of UPL 
payments during transition periods.

[A] The 5-year transition period is based on GAO analysis showing that 
the state established a UPL arrangement on January 1, 1998, which 
qualifies it for a 5-year transition period under CMS's UPL regulation.

[B] Differences based on numbers before rounding.

[C] The 16-month transition period is based on GAO analysis showing 
that the state established a UPL arrangement on February 7, 2001, which 
indicates that the originally assigned 16-month transition period 
should be maintained.

[D] Totals may not add because of rounding.

[End of table]

[End of section]

Appendix II: Scope, Methodology, and Analysis of Selected State UPL 
Calculations: 

An upper payment limit (UPL) represents the maximum amount the federal 
government will pay as its share of a state's Medicaid 
expenditures.[Footnote 56] Because UPLs are based on the amount 
Medicare would pay for similar services, a state must develop a 
reasonable methodology to (1) identify nursing home services provided 
to Medicaid residents, (2) generate a reasonable estimate of what 
Medicare would have paid for equivalent services, and (3) account for 
differences in covered services between Medicaid and Medicare and 
determine the payment adjustments needed. Because an inflated UPL can 
generate excessive federal matching funds, it is important that states 
use sound methods for calculating their UPLs. Although CMS has 
developed a draft financial review guide as a basis for the agency's 
audits of state calculations, it has not prescribed a standard UPL 
methodology that states must use. According to agency officials, CMS 
deliberately decided to recognize the variation among state approaches 
and to allow states some flexibility to develop methods appropriate to 
their situations. With this flexibility in mind, we examined the extent 
to which states used appropriate methods to calculate their UPLs. We 
made this assessment by comparing states' methodologies and results 
with Medicare payment principles and with CMS's draft internal UPL 
review guide.

For our analysis, we obtained data and documentation from six states 
that have large nursing home UPL arrangements and received either an 8-
year or a 5-year transition period--Michigan, New York, Oregon, 
Pennsylvania, Washington, and Wisconsin. We generally focused our 
review on state methods for state fiscal year (SFY) 2000 because it is 
the base year for determining the sum of excessive payments that will 
be phased out during the transition period, but we also reviewed SFY 
2001 methods to determine whether selected states had made changes. 
Methods used in SFY 2001 are important because that particular year 
established the amount that certain states--those with long transition 
periods--could claim before the phase-out of excessive payments began. 
We based our analysis on the information that was provided to us by 
states and CMS during our review.

We found widely varying methods, potentially inaccurate UPL 
calculations, and other errors as well. Some states, for example, used 
incorrect Medicare rates, such as rates from the wrong year, when 
computing their UPL; another potentially overstated the number of 
Medicaid nursing home resident days by using estimates that were higher 
than the actual days provided. These methods could have inflated 
states' UPLs--and subsequent claims for federal matching funds--by tens 
of millions of dollars. Further, because SFY 2000 errors applied to 
states' base year for their transition periods, excessive federal 
claims could continue throughout their transition periods, leading to 
continued excessive federal outlays for those states with UPL 
arrangements in place.

Incorrect Medicare Payment Rates: 

All six states we examined adopted methods that, in our view, were not 
consistent with Medicare payment principles. Under Medicare, nursing 
homes receive a prospective daily rate to cover most services provided 
to a resident during each day of a covered stay. The rate is adjusted 
for the resident's expected care needs and therapy, as determined by 
the resident's assignment to one of Medicare's 44 different payment 
groups. In addition, Medicare payment rates are adjusted by area wage 
indexes to account for geographic variations in costs. For UPL 
purposes, an accurate Medicare equivalent should be based on the number 
of Medicaid resident days in each of Medicare's 44 payment groups. 
States should also adjust their UPL calculations by the appropriate 
Medicare wage index. In all six states we examined, we identified 
instances in which states' UPL methodologies made general assumptions 
that, in our view, raised concerns and potentially resulted in 
inaccurate Medicare equivalents and inflated UPLs.

* Five states we examined did not use Medicaid resident days--the most 
precise method for determining a Medicare equivalent--in their UPL 
calculations. Only one state, Washington, calculated a weighted 
Medicare rate using Medicaid resident days. The five remaining states 
used less accurate measures, either the number of Medicaid residents or 
the number of Medicaid nursing home beds. In our view (and suggested by 
CMS's internal financial review guide), the use of Medicaid resident 
days is a more accurate basis for determining the weighted Medicare 
payment rate because it accounts for different lengths of stay in each 
of the payment groups. In addition, one state, Wisconsin, 
inappropriately changed its UPL methods in SFY 2001, by estimating its 
UPL on the assumption that all Medicaid residents fell into a single 
Medicare payment group. This approach was problematic because the state 
selected a higher payment group (indicating that all residents needed 
skilled care) even though state data showed that 60 percent of Medicaid 
residents actually required lower levels of care.

* Another error we identified involved the use of incorrect Medicare 
payment data, such as Medicare wage indexes in different areas. Each 
year, CMS updates Medicare nursing home data and determines a wage 
index for each geographic area in a state.[Footnote 57] Three states -
-Michigan, Wisconsin, and Washington--did not correctly apply updated 
Medicare payment data in their UPL estimates. In Michigan's SFY 2000 
estimate, the state's UPL methodology adjusted Medicare payment rates 
using federal fiscal year 2001 wage indexes, rather than the lower wage 
indexes applicable in SFY 2000. As a result, Michigan's UPL estimate 
was inappropriately inflated by more than $8 million, allowing the 
state to claim almost $5 million more in federal matching funds than it 
should have in SFY 2000 and, if the error is not corrected, more than 
$17 million more during its 5-year transition period. In Wisconsin, the 
state's SFY 2000 estimate used outdated Medicare payment data, which 
lowered the state's UPL estimate by about $3 million. In Washington, 
the state's SFY 2001 estimate used the wage index and Medicare payment 
rates applicable in rural areas for nursing homes located elsewhere, 
which had a different index and payment rates.

Overstated Medicaid Nursing Home Resident Days: 

Although most states we examined did not use the number of resident 
days in calculating the Medicare payment rates for the Medicaid nursing 
home residents served, states did incorporate an estimate of resident 
days in determining the amount of the aggregate UPL and the amount 
available for an excessive UPL payment. Specifically, after the states 
calculated the difference between Medicare and Medicaid per diem rates, 
they multiplied this difference by an estimate of the number of 
Medicaid resident days provided by nursing homes in the states to 
determine the amount they could pay and claim under their UPL 
arrangement. While states may have used the best available data at the 
time, changes in the actual number of resident days provided could lead 
to inflated UPL estimates. For example, we found that one state--
Oregon--determined its UPL claims on the basis of an estimated number 
of Medicaid resident days that was significantly higher than the actual 
number of days provided by nursing homes that year. In its base year 
estimate, Oregon relied on an estimated number of Medicaid nursing home 
resident days that was nearly 163,000 days more: 

than the actual provided days, allowing that state to claim more than 
$6 million in additional federal matching funds. If this error is not 
corrected, Oregon could claim another $22 million during its 5-year 
transition period.[Footnote 58]

[End of section]

Appendix III: Comments from the Centers for Medicare & Medicaid 
Services: 

DEPARTMENT OF HEALTH & HUMAN SERVICES 
Centers for Medicare & Medicaid service:

DATE: JAN 14 2004:

TO: Kathryn G. Allen:

Director, Health Care-Medicaid and Private Health Insurance Issues 
General Accounting Office 

FROM: Dennis G. Smith:

Acting Administrator:

Centers for Medicare & Medicaid Services:

SUBJECT: MEDICAID: Improved Federal Oversight of State Financing Schemes 
Is Needed (GAO-04-228):

We appreciate the opportunity to respond to the November 25, 2003, 
draft report entitled, MEDICAID: Improved Federal Oversight of State 
Financing Schemes Is Needed (GAO-04-228). The General Accounting Office 
(GAO) provided this draft report to the Centers for Medicare & Medicaid 
Services (CMS) for review and comment prior to issuing the report in 
final form. You indicated that the comments would be reflected in the 
final report. Upon review of the November 25, 2003, draft report, CMS 
is very concerned with GAO's findings regarding CMS' implementation of 
the upper payment limit (UPL) Federal regulations.

The findings and recommendations included in this draft report are 
based on the GAO's unique interpretation of the statutory language 
included in Medicare, Medicaid, and SCHIP Benefits Improvement and 
Protection Act of 2000 (BIPA 2000) and the GAO's inference of preamble 
language included in the UPL regulations. Moreover, the GAO's findings 
and recommendations ignore, almost entirely, the basis for and 
explanations of CMS' policy interpretations and decisions, which were 
provided to you both orally and in writing over the 13 months that you 
investigated CMS implementation of the UPL regulations. The only 
reference to CMS policy interpretation included in this draft report 
suggests that CMS changed policy interpretations in the middle of the 
UPL implementation in order to provide preferential treatment to 
certain states. We believe this assertion is an inaccurate statement 
and does not take into account any of the explanation provided to you 
both orally and in writing. Finally, we note the following general 
observations:

(i) The stated objectives included in this draft report are different 
from the objectives stated at the October 1, 2002, entrance conference 
and included in the September 11, 2002, letter sent by the GAO to the 
Secretary of Health and Human Services, yet the draft report contains 
no explanation of such change in objectives;

(ii) The illustrations included in the draft report of UPL-related 
spending in Nebraska and Wisconsin do not fairly and accurately 
represent the amounts allowed by CMS;

(iii) To the extent that the draft report criticizes CMS for the length 
of time taken in determining UPL transition periods and transition 
amounts, GAO fails to consider the limited resources available to CMS 
in administering the UPL regulations; and,

(iv) The draft report fails to acknowledge that GAO's own audit staff 
stated during the November 4, 2003, exit conference, that CMS' 
interpretation of the statutory phrase, "Medicaid payment provision in 
place prior to October 1, 1992," and of the statutory terms "successor 
provisions" and "subsequent amendments" were both reasonable and 
legally supportable.

Consequently, we have several significant comments with regard to the 
findings and recommendations included in this draft report. Upon your 
consideration of such comments, we urge you to re-consider your 
conclusions and recommendations regarding qualification for the 8-year 
transition under the UPL regulations.

Attachments:

OVERVIEW:

The CMS is concerned about GAO's apparent misunderstanding of CMS 
implementation of the new UPL regulation. The CMS has been clear with 
GAO staff regarding the evolution of CMS policy development during our 
UPL implementation and GAO's concurrent investigation. The rationale 
behind the evolution of CMS policy during implementation of the UPL 
rules was openly shared with, and carefully explained to, GAO staff via 
numerous e-mail communications, conference calls, and formal written 
communications.

It is important to note that this review took place simultaneously to 
CMS' development and implementation of the UPL regulations. These 
provisions are highly complex in both the areas of policy and 
information gathering. In addition, as you are aware from the internal 
status reports provided by CMS, we have had to request additional 
information on several occasions and in several states in order to 
fully evaluate the states' UPL calculations. The data gathering has 
been a lengthy process and often includes the collection and evaluation 
of significant amounts of data.

In the draft report, GAO has asserted that CMS changed its policy 
interpretations when in fact CMS was simply undertaking the process to 
develop its policy interpretation. In response to the substantial 
number of questions raised by the GAO during your review, CMS explained 
that we continue to develop policy to implement the UPL regulations. 
The detailed information provided to CMS under the UPL transition 
review has assisted in developing legally supportable policies. 
Specifically, in response to the questions raised during the March 18, 
2003, conference call with your staff (see Attachment 1), CMS explained 
the basis for the evolution of CMS policy regarding the definition of 
"subsequent amendments/successor provisions." As we continued to review 
Medicaid institutional reimbursement state plan language, we informed 
GAO staff that our internal, preliminary interpretation potentially 
disqualified any amendments from being considered successors and had 
the potential to render meaningless the BIPA 2000 provision on the 8-
year transition period. This explanation was formally provided to you 
on August 14, 2003, in response to your July 7, 2003, request for 
confirmation of how CMS is implementing the UPL transition periods (see 
Attachment 2).

The policy interpretation that CMS has now adopted interprets the terms 
"successor provisions" and "subsequent amendments" consistently with 
the interpretation of what constitutes a payment provision in place 
prior to October 1, 1992. Specifically, CMS interprets the BIPA 2000 
phrase, "payment methodology in place prior to October 1, 1992" to mean 
an "enhanced" Medicaid payment provision or a "supplemental" Medicaid 
payment provision. The CMS does not consider a regular Medicaid payment 
rate provision to qualify under this statutory phrase. In addition, CMS 
interprets the BIPA 2000 phrases "successor provisions" and "subsequent 
amendments" to mean "enhanced" or "supplemental" Medicaid payment 
provisions to the same type of providers as the qualifying payment 
methodology in place prior to October 1, 1992. Specifically, a 
successor provision and a subsequent amendment must be an enhanced or 
supplemental Medicaid payment provision to the same provider type 
(e.g., county nursing 
facilities) that was eligible for an enhanced or supplemental Medicaid 
payment prior to October l, 1992. Under CMS' current interpretation, 
successor provisions and subsequent amendments do not have to contain 
reimbursement formulas that are identical to the payment methodology in 
place prior to October 1, 1992.

We believe that our current policy supports the intent of the UPL 
regulations regarding appropriate Federal Medicaid spending as well as 
Congressional intent to allow for an 8-year transition period. 
Moreover, CMS has uniformly applied this policy to all states. Based on 
our preliminary review of historical state spending, we had indicated 
that we believed there were five states (including hospital and nursing 
home UPL arrangements) that had the potential to qualify for 8-year 
transition periods. As we developed our interpretation of both the 
"payment provision in place prior to October 1, 1992," language and the 
"successor provision" language, we considered various interpretations 
including those, that when applied consistently to all states, resulted 
in no state qualifying for an 8-year transition and those that resulted 
in every state qualifying for an 8-year transition period. Based on the 
information we have received from states to date, we believe a 
consistent application of our current policy would likely result in 
five states qualifying for 8-year transition periods through either a 
hospital or nursing facility supplemental payment program. This 
represents neither a significant expansion nor reduction from our 
initial estimate of the number of states qualifying for an 8-year 
transition period.

Your report recommends that CMS reassess its decision to grant 8-year 
transition periods to Nebraska and Wisconsin, establish guidance for 
states on appropriate methods for calculating their UPLs, and give 
priority to financial management reviews for states with the largest 
UPL arrangements. As will be explained in detail under the "TECHNICAL 
CORRECTIONS/COMMENTS" and "RECOMMENDATIONS" sections of this response, 
we believe:

(i) The CMS has developed sound, legally supportable policies with 
regard to the 8-year transition under the UPL regulations and has 
appropriately determined that Nebraska, Pennsylvania, and Wisconsin as 
qualify for an 8-year transition period, based on a nursing facility 
supplemental payment provision in place prior to October 1, 1992;

(ii) The CMS has been providing on-going and direct guidance to all 
states through the implementation of the UPL regulations and through 
CMS' Medicaid institutional reimbursement state plan amendment review 
process; (iii) In allocating resources to the determination of 
transition periods and amounts, the CMS has directed its staff, 
whenever possible, to give priority to the determination of these 
issues for states with the largest UPL arrangements, to ensure that 
these states can incorporate CMS findings into their budgeting plans.

RECOMMENDATIONS:

GAO Recommendation:

The Administrator of CMS establish criteria far making transition 
period decisi are consistent with the objectives described in its 
January 2001 UPL regulation.

CMS:

We concur. As carefully detailed in the response to this draft report, 
CMS has already done so.

GAO Recommendation:

CMS reconsider its initial decisions to grant Nebraska and Wisconsin 8-
year transi periods.

CMS:

We do not concur. The CMS has developed a reasonable and legally 
supportable interpretation of 8-year transition qualification under the 
statute and UPI.. regulation and GAO has stated that both Wisconsin and 
Nebraska meet those criteria.

GAO Recommendation:

CMS establish uniform guidance for states, which would set forth 
acceptable methods to calculate their UPLs.

CMS:

We concur. The CMS does not necessarily agree with the GAO's definition 
of a reasonable estimate of the UPL. Nor does CMS believe that an 
exhaustive "laundry" list of acceptable methods can be compiled that 
would address every payment methodology to every provider in every 
state. Hence, CMS will issue guidance on the characteristics and 
principles underlying acceptable methods of calculating UPLs, along 
with extensive examples. However, CMS believes a certain degree of 
state flexibility is warranted.

GAO Recommendation:

CMS expediting the financial management reviews of states with UPL 
arrangements, with a high priority on reviews ofstates with S-and 8 -
year transition periods, including those we identified as having 
methodological errors.

CMS:

We concur in part. The CMS has made significantly more progress than is 
articulated by the GAO in this draft report. However, CMS agrees to 
perform financial management reviews in the 5-year and 8-year states 
not currently under review by the HHS OIG.

GAO Recommendation:

CMS improving its requirements for states far reporting UPL 
arrangements, such as implementing its current requirement for states 
with transition periods to report payments on a facility-specific 
basis, and requiring such reports for all states with a UPL 
arrangement.

CMS:

We concur. The CMS will re-evaluate current UPL reporting requirements 
and CMS will further consider GAO's recommendation for improved UPL 
reporting requirements. 

[End of section]

Appendix IV: Comments from the State of Michigan: 

STATE OF MICHIGAN:

JENNIFER M. GRANHOLM, GOVERNOR
DEPARTMENT OF COMMUNITY HEALTH: 
LANSING: 

JANET OLSZEWSKI:
DIRECTOR:

December 23, 2003:

Ms. Kathryn G. Allen, Director:

Health Care --Medicaid and Private Health Insurance Issues 
United States General Accounting Office:

Washington, DC 20548:

Dear Ms. Allen:

Pursuant to your letter of December I, 2003, 1 am submitting comments 
on the GAO report entitled "MEDICAID - Improved Federal Oversight of 
State Financing Schemes is Needed".

On pages 26 and 39 it was noted that Michigan "made adjustments for 
geographic variations in labor costs that incorporated inappropriately 
high wage rates. The higher rates overstated the state's UPL by more 
than $8 million, allowing excess federal claims of almost $5 million.":

We have reexamined our long term care upper payment limit (UPL) 
calculation for fiscal year 2000 (the year on which the phase-out 
amounts were based) and have determined that our initial calculations 
were substantially correct. There was a labeling problem in our 
spreadsheet where the label describing the wage index number was 
incorrect as stated in the GAO analysis. However, the column of figures 
which drove the UPL calculations contained (with one minor exception) 
the correct wage index numbers and, therefore, there was no substantial 
error in Michigan's UPL calculation for fiscal year 2000. This same 
labeling error was discovered by an OIG audit of our fiscal year 2003 
calculation but, again, the column driving the calculations contained 
the correct number. These labels have been corrected for our 2003 and 
2004 submissions to CMS.

It should also be noted that the Medicare RUGS rates were raised 
significantly by CMS for the period covering the final six months of 
fiscal year 2000. If those changes had been factored into the Michigan 
UPL calculation, the Medicare upper payment limit for fiscal year 2000 
would have increased by about $65 million.

Given our analysis of the two points discussed above, I recommend that 
the audit findings related to Michigan in the GAO report be deleted. 
There was no error of substance and a disallowance or reduction in 
federal funding is not appropriate. Thank you for the opportunity to 
review and comment on the draft report.

Sincerely,

Signed by: 

Janet Olszewski: 
Director: 

[End of section]

Appendix V: Comments from the State of Nebraska: 

NEBRASKA HEALTH AND HUMAN SERVICES SYSTEM:

DEPARTMENT OF SERVICES 
DEPARTMENT OF REGULATION AND LICENSURE 
DEPARTMENT OF FINANCE AND SUPPORT:

STATE OF NEBRASKA 
MIKE JOHANNS, GOVERNOR:

December 8, 2003:

Kathryn G. Allen, Director:

Health Care - Medicaid and Private Health Insurance Issues 
United States General Accounting Office:

Washington, D.C. 20548:

Dear Ms. Allen:

We have reviewed draft report GAO-04-228 entitled "MEDICAID: Improved 
Federal Oversight of State Financing Schemes Is Needed" as requested in 
your correspondence of December 1, 2003. The report reviews the actions 
of the federal Centers for Medicare & Medicaid Services (CMS) in 
determining the transition period for phase-out of state Upper Payment 
Limit (UPL) arrangements. The Nebraska Medicaid Program has complied in 
good faith with regulations established by the federal Department of 
Health and Human Services and administered by CMS. A reinterpretation 
of the federal decision to grant Nebraska an eight-year transition 
period would be unacceptable.

Sincerely,

Signed by: 

Robeft J. Seiffert, Administrator: 
Medicaid Division:

CC: Stephen B. Curtiss: 

[End of section]

Appendix VI: Comments from the State of New York: 

STATE OF NEW YORK 
DEPARTMENT OF HEALTH:
Coming Tower:
The Governor Nelson A. Rockefeller Empire State Plaza: 
Albany, New York 12237:
Antonia C. Novello, M.D., M.P.H., Dr. P.H. Commissioner:
Dennis P. Whalen:
Executive Deputy Commissioner:

January 8, 2004:

Ms. Kathryn G. Allen:

Director, Health Care - Medicaid and Private Insurance Issues 
United States General Accounting Office 
Washington, DC. 20548:

Dear Ms. Allen:

Thank you for submitting a draft copy of the GAO report, Medicaid: 
Improved Federal Oversight of State Financing Schemes is Needed, for 
our review and comment.

Enclosed are our comments on the report and an overview of New York's 
upper payment limit calculation. Please note that our comments refute 
many of the report's findings regarding New York State. Therefore, we 
expect that the final GAO report will more accurately reflect the 
State's calculation and use of its nursing home Upper Payment Limit.

If you have any questions concerning our comments, please contact Mark 
H. Van Guysling, Assistant Director, Division of Health Care Financing 
at (518) 474-6350.

Sincerely,

Signed by: 

Sandra Pettinato: 
Deputy Director Office of Medicaid Management:

Enclosures:

NEW YORK STATE RESPONSE TO THE GENERAL ACCOUNTING OFFICE (GAO) DRAFT 
REPORT #04-228:

GAO Finding: Funds generated by New York's UPL arrangement are 
deposited into its Medical Assistance Account. Proceeds from this 
account are used to pay for the state share of the cost of Medicaid 
payments; effectively recycling federal funds to generate additional 
federal Medicaid matching funds.

NYS Response:

Since its inception New York's Medicaid State Plan has included local 
government cost sharing. Intergovernmental transfers (IGTs) from 
counties and the City of New York to the State are integral to New 
York's Medicaid financing statutes. This form of financing of the non-
federal share of Medicaid expenditures is permitted by federal law 
(Section 1902(a)(2) of the Social Security Act (SSA)) as long as the 
State dollars make up, in the aggregate, at least 40 percent of the 
non-federal share of expenditures for which federal financial 
participation is sought under SSA Section 1903. The State and local 
shares of Medicaid payments vary by category of service and by certain 
recipient-specific factors. Further, State statutes sometimes 
establish state and local funding arrangements that are specific to 
particular payments under the State Plan. For the payments that are the 
subject of the GAO finding, State statute provides that local 
governments will assume 100 percent of the non-federal cost, which is 
derived from funds appropriated by local governments' legislative 
bodies. This expenditure is clearly eligible for federal matching funds 
pursuant to 42 CFR 433.51 (b) and is entirely consistent with the 
Congressional intent to permit States to fund a portion of the non-
federal share of Medicaid expenditures with funds appropriated by local 
governments. State statute establishing these payments also requires an 
additional transfer of funds from local governments to the State's 
local assistance medical assistance account. These funds represent a 
Medicaid shares adjustment between the State and local governments that 
complies with the non-federal share local government contribution 
limitations established by Section 1902(a)(2). Transfer of these funds 
in no way negates the initial local government expenditure that is 
eligible for federal matching funds.

The State makes payments for services under the approved State Plan 
directly to providers. State laws create no repayment obligation on the 
part of nursing facilities with regard to funds received under the 
approved State Plan, though associated transactions between some public 
providers and their local governments were noted in their audited 
annual financial statements in previous years. The State is not 
authorized or required to monitor the financial relationships that may 
exist between local governments and their public nursing facilities. 
Further, we believe it would be inappropriate to do so for the purpose 
of Medicaid claiming.

GAO Finding: The federal share of New York State UPL payments in SFY 
2001 was $520,000,000, which has not been subject to a financial 
management review by CMS.

NYS Response:

While the room under the upper payment limit for SFY 2001 was 
calculated to be in excess of $1 billion, only $991.5 million in 
payments were made to county sponsored nursing facilities for IGT. 
Therefore the federal share of the New York State IGT payments was only 
$495.75 million. It is unclear where the $520 million was derived from.

GAO Finding: For example, none of the states accurately determined 
Medicare rates when calculating their UPLs, because they did not 
correctly account for changes that occurred when Medicare moved from 
cost based reimbursement to prospective rates starting in 1998.

NYS Response:

New York's UPL calculations have been based on the formula as described 
in the Federal Registers beginning with the interim final rule 
published 5112198 which includes changes that occurred when Medicare 
moved from cost based reimbursement to prospective rates.

GAO Finding: The methodology for New York's SFY 2000 UPL estimate was 
also inaccurate because it used Medicare, not Medicaid, residents to 
calculate a weighted Medicare payment rate; this substitution likely 
inflated the calculated Medicare equivalent and the estimated UPL 
because Medicare patients generally fall into higher Medicare payment 
groups than do Medicaid residents:

NYS Response:

New York's UPL calculation utilizes only patient data for Medicaid 
residents in determining the number of residents that fall into each of 
the 44 PPS payment groups. The only place Medicare residents were used 
was in determining the Medicare Federal rate, which was needed in the 
calculation to estimate the facility specific rate. (See attached 
explanation of the 2000 Upper Payment Limit Calculation.):

GAO Finding: None of the six states we examined accounted correctly for 
Medicare's move from retrospective cost-based reimbursement to 
prospective rates. Over the three years following Medicare's 
implementation of prospective nursing home payments in 1998, many 
nursing homes were paid an amount that blended prior cost-based rates 
with new prospective per diem rates. None of the six states' 
methodologies properly accounted for the transition by using facility 
specific blending rates specified by Medicare regulations. Some states 
applied a single blending factor for all nursing homes, and others made 
improper adjustments in applying blending rates. New York and 
Pennsylvania, for example, selectively blended rates only for those 
nursing homes that would generate higher Medicare payment rates - 
regardless of whether or not the facility was actually blending.

NYS Response:

Following the formula as published in the 5/12/98 Federal Register, New 
York used facility specific blended rates for the three years following 
the implementation of the Prospective Payment System. The Federal 
Register published 7/30/99 specified that for any facility that was 
advantaged, an immediate transition to the Federal rate for those 
facilities was allowed. The full Federal rate was used instead of the 
blended rate. NYS made the assumption that all facilities that were 
advantaged moved to the Federal rate.

GAO Finding: We identified two states, New York and Oregon, which 
developed their UPL based on an estimate of the number of Medicaid 
resident days that was significantly higher than the actual number of 
days that were provided by nursing homes that year. While the states 
may have used the best available estimates at the time, in our view, 
states should be held accountable for reconciling estimates with the 
actual services provided. Left uncorrected, this error could increase 
their federal claims throughout their transition periods. New York 
relied on an estimate of the number of Medicaid nursing home resident 
days that was nearly 600,000 higher than the actual number of days 
where residents were served in the state's nursing homes, allowing the 
state to claim more than $9 million in additional federal matching 
funds. If the discrepancy is not corrected, the state could claim an 
additional $35 million over its five-year transition period.

NYS Response:

We do not have enough information to accurately respond to this 
finding. Specifically, we need more information regarding the 600,000 
patient day discrepancy. What period does it relate to and how does it 
relate to the $9 million overage?

GAO Finding: Our estimate of overstated claims are conservative for 
both states. Even with the corrections we made, we believe the states' 
methodologies continue to overstate UPLs in New York and Oregon because 
they assume that all residents stay in nursing homes for an entire 
year. The discrepancies would be greater had more accurate information 
been available on the residents' actual average length of stay.

NYS Response:

New York State uses actual patient days reported in certified cost 
reports. These days take into consideration length of stay.

2000 UPPER PAYMENT LIMIT CALCULATION:

The upper payment limit calculation determines how much lower the 
Medicaid rates being paid in New York State are than what they would 
have been if they had been calculated using Medicare principles.

MEDICAID RATE:

The Medicaid rates used in this calculation were the January 1, 2000 
Medicaid rates.

MEDICAID RATE USING MEDICARE PRINCIPLES:

The rates used in this calculation are 50 percent of the federal rate 
plus 50 percent of the facility-specific rate.

FEDERAL RATE:

The federal rate is based on the new Medicare Prospective Payment 
System (PPS). National prices for 44 PPS RUG groups were first wage 
adjusted for New York State areas. The prices and the wage indices used 
are contained in the Federal Register published 7/30/99. Patient data 
from nursing facilities' 1997 MDS+ data was then matched to patient 
data from the second quarter 1997 PRI data to determine which patients 
in the MDS+ file were Medicaid residents. 1997 data was chosen to use 
since it was the most complete set of both MDS+ and PRI data that was 
available at the time. It has been found that using the case mix from 
1997 instead of a more current period does not materially affect the 
calculation. Once the Medicaid residents were determined, the Medicaid 
residents for each facility were then categorized into one of 44 RUG 
groups. Each facility's federal rate was calculated by multiplying the 
number of Medicaid residents in each RUG group by the applicable price 
for each group, then dividing the sum of the 44 groups by the total 
number of Medicaid residents. An inflation factor was then applied to 
adjust for the 4.5 month difference in the midpoints between the period 
the initial prices were based on (7/1/98-9/30199) and the period the 
initial Medicaid rates were based on (1/1/99 - 12/31/99). The inflation 
factor used was 1.00675, calculated as [4.5/12X (the revised 1999 
market basket of 2.8 percent from the 7/30/99 Federal Register minus 1 
percent)].

FACILITY-SPECIFIC RATE:

In general, the facility-specific rate is:

A) For facilities participating in the Nursing Home Case Mix and 
Quality (NHCMQ) demonstration project, it is the rate paid under the 
demonstration project.

B) For facilities not participating in the demonstration project, it is 
the rate calculated under the previous routine ceiling methodology.

Since these rates were calculated for Medicare, New York State did not 
have a Medicaid equivalent. An assumption was made that the percentage 
relationship between the Medicare facility-specific rate and the 
Medicare federal rate would be the same in the Medicaid calculation. 
Therefore, a Medicaid facility-specific rate had to be estimated by 
using this relationship and applying it to the Medicaid federal rate. A 
facility-specific rate percentage was calculated by dividing the 1999 
Medicare facility-specific rate (see below) by the 1999 Medicare 
federal rate (see below). This percentage was then multiplied by the 
1999 Medicaid federal rate, calculated in accordance with applicable 
regulations, to arrive at the estimated 1999 facility-specific rate for 
Medicaid. The 2000 facility-specific rate was then calculated by 
multiplying the 1999 facility-specific rate by the full 2000 market 
basket rate of 2.9 percent as published in the 7/30/99 Federal 
Register. If a facility-specific rate percentage could not be 
determined for a particular facility due to the unavailability of data, 
an average facility-specific rate percentage was used. The average 
facility-specific rate percentage used depended on whether or not the 
facility participated in the demonstration project. The average for 
facilities participating in the demonstration project was 1.1158. The 
average for facilities not participating in the demonstration project 
was .9738.

1999 Medicare facility-specific rate:

Facilities participating in the demonstration project -:

If the Medicare Part A Intermediary was Empire Medicare Services, the 
rate used was the facility-specific per diem rate on Line 5 of the 1997 
NHCMQ demonstration computation. The rate on Line 5 includes an 
inflation factor of 1.031532. If the Medicare Part A Intermediary was 
United Health Care, the rate used was the total amount reimbursed on 
Line 25 divided by the total demonstration program days on Line 2 of 
the 1997 calculation of the NHCMQ demonstration reimbursement 
settlement. This per diem was then multiplied by an inflation factor of 
1.031532.

Facilities not participating in the demonstration project: 

If the Medicare Part A Intermediary was Empire Medicare Services, the 
rate used was the facility-specific per diem rate on Line 20 of the 
1995 rate computation. The rate on Line 20 includes an inflation factor 
of 1.071430. If the Medicare Part A Intermediary was United Health 
Care, (or Empire Medicare Services and the 1995 computation of the 
facility-specific per diem rate was not available) the rate used was 
the 1995 Part A Medicare rate trended to 1999 using an inflation factor 
of 1.071430.

1999 Medicare federal rate:

The Medicare federal rate was calculated exactly the same as the 
federal rate in the 1999 Upper Payment Limit calculation, except that 
Medicare residents were used instead of Medicaid residents.

CALCULATION OF UNADJUSTED ROOM UNDER THE UPPER PAYMENT LIMIT:

The Medicaid rates were weighted by estimated 2000 Medicaid days to 
develop a 2000 statewide average Medicaid rate. The Medicare rates were 
weighted by estimated 2000 Medicaid days to develop a 2000 statewide 
average Medicare rate. The 2000 statewide average Medicaid rate was 
$166.56 and the 2000 statewide average Medicare rate was $200.51. The 
difference of $33.95 was multiplied by total Medicaid days to arrive at 
the unadjusted room under the limit of $1,097.5M. The methodology in 
this analysis has been applied to all Nursing Facilities in New York 
State. 

[End of section]

Appendix VII: Comments from the State of Oregon: 

December 19. 2003:

Oregon:

Theodore R. Kulongoski, Governor:

Department of Human Services 
Health Services 
Office of Medical Assistance Programs 
500 Summer Street NE, E49 
Salem, Oregon 97301-1079 
Voice - (503) 945-5772 
Fax - (503) 373-7689 
TTY - (503) 378-6791:

Kathryn G. Allen:

Director, Health Care-Medicaid and Private Health Insurance Issues: 
United States General Accounting Office: 
Washington, D.C. 20548:

RE: GAO-04-228 Oversight of State Financing Schemes:

Dear Ms. Allen:

Thank you for the opportunity to comment on the above reference draft 
report entitled MEDICAID: Improved Federal Oversight of State Financi 
Schemes is Needed.

Oregon does not take issue with the general substantive findings 
delineated in the report except the general characterization that 
states were engaged in unauthorized activities regarding transactions 
in the Nursing Facility Proportionate Share (PPS) program. Findings in 
the report state that the PPS program was casually overseen by CMS. 
Implementation by CMS had been done consistent with the preamble 
statement of the implementing regulations to the program,

"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, State 
have flexibility to use either Medicare principles of cost 
reimbursement or ment prospective payment systems as the foundation of 
their 2001 estimates." 66 Fed. Reg. 3148 (January 12, 2001.)
 
States remitted their State Plan Amendments, methodologies and claims 
to CMS within prescribed protocols, and these were reviewed and acted 
upon by CMS within that defined framework. While we understand that GAO 
now believes that framework to be flawed, it was the correct framework 
to be followed by states. Hence, the characterization that states 
actions were outside of authority we find to be inaccurate.

Pursuant to existing rules, Oregon filed a required State Plan 
Amendment and received approval of that amendment. In accordance with 
the methodology approved, Oregon estimated its entitlement to PPS and 
requested reimbursement from CMS. CMS reviewed the requests and 
rendered payment without issue.

Again, we want to thank you for the opportunity to comment on this 
report.

Sincerely,

Signed for: 

Lynn Read: 

Administrator: 

[End of section]

Appendix VIII: Comments from the State of Wisconsin and GAO's 
Response: 

Note: GAO comments supplementing those in the report text appear at the 
end of this appendix.

DIVISION OF HEALTH CARE FINANCING:
1 WEST WILSON STREET 
P O BOX 309 MADISON W1 53701-0309:

Telephone: 608.266-8922:
FAX: 698-266-1096:
TTY: 608-261-7798:
dhfs.wisconsin.gov:

State of Wisconsin: 
Jim Doyle, Governor: 

Department of Health and Family Services: 
Helene Nelson, Secretary: 

December 23, 2003:

Kathryn G. Allen, Director: 
Health Care - Medicaid and Private Health Insurance Issues: 
United States General Accounting Office: 
Washington, DC 20548:

Dear Ms. Allen:

On December 1, 2003, the Wisconsin Medicaid Program received from the 
General Accounting Office (GAO) a draft report entitled MEDICAID: 
Improved Federal Oversight of State Financing Schemes is Needed (GAO-
04-228). In its cover letter, GAO indicated it wished to receive 
comments regarding the draft report by December 23, 2003. Enclosed 
please find Wisconsin's comments regarding the GAO draft report.

The GAO draft report concludes that Wisconsin should be subject to an 
Upper Payment Limit transition period of only 16 months rather than the 
8 years granted by CMS. As a result, GAO asserts that Wisconsin is not 
entitled to receive over 5500 million in federal Medicaid funds it 
would have received under the 8-year transition period. GAO is 
incorrect on both counts, as further explained in Wisconsin's comments.

These comments are being sent via email, fax and first class mail. 
Please note there are four enclosures that you will receive with the 
faxed and mailed versions. Because the enclosures are not available in 
electronic fonnat, you will not receive them with the emailed version.

Thank you for the opportunity to review and comment on GAO's draft 
report.

Sincerely,

Signed by: 

Mark B. Moody: 
Administrator:

Enclosures:

WISCONSIN'S COMMENTS REGARDING GAO-04-228:

The GAO report (GAO-04-228) concludes that Wisconsin should be subject 
to an Upper Payment Limit transition period of only 16 months rather 
than the 8 years granted by CMS. As a result, GAO asserts that 
Wisconsin is not entitled to receive over $500 million in federal 
Medicaid funds it would have received under the 8-year transition 
period. GAO is incorrect on both counts, as further explained below.

Wisconsin is entitled to "Tier 3" transition status under BIPA.

In October 2000 HHS published a proposed rule changing the upper 
payment limit (UPL) test for Medicaid payments to local government 
institutional facilities. The proposed new UPL measure would be the 
aggregate of what would be paid under Medicare payment principles for 
services provided in local government facilities. The existing UPL 
measure was the aggregate of what would be paid under Medicare payment 
principles for services provided in all facilities. HHS believed that 
states were taking advantage of the existing UPL measure to make unduly 
large Medicaid payments to local government facilities, and claiming 
the federal share for those payments. The proposed new UPL rule 
contained two transition provisions, one allowing payments exceeding 
the new UPL measure until September 30, 2002, and the other allowing 
payments exceeding the new UPL measure, in declining amounts, through 
state fiscal year 2005. [NOTE 1]

In December 2000 Congress enacted BIPA.[NOTE 2] In BIPA, Congress 
directed HHS to adopt a new UPL rule based on the October 2000 proposed 
rule, but further required that the rule provide for a third transition 
period. The third transition provision would allow certain states to 
make payments exceeding the new UPL measure, in declining amounts, 
until September 30, 2008. [NOTE 3] BIPA describes a state qualifying 
for this transition period as follows:

A State described in this paragraph is a State with a State medicaid 
plan payment provision or methodology (including a payment provision or 
methodology approved under a waiver of such plan) which--

(A) was approved, deemed to have been approved, or was in effect on or 
before October 1, 1992 (including any subsequent amendments or 
successor provisions or methodologies and 
whether or not a State plan amendment was made to carry out such 
provision or methodology after such date) or under which claims for 
Federal financial participation were filed and paid on or before such 
date; and:

(B) provides for payments that are in excess of the upper payment limit 
test established under the final regulation required under subsection 
(a) (or which would be noncompliant with such final regulation if the 
actual dollar payment levels made under the payment provision or 
methodology in the State fiscal year which begins during 1999 were 
continued). [NOTE 4]

There are thus two conditions a state must meet to qualify for the Tier 
3 transition period. The first is that the state must have had a 
payment methodology in effect on or before October 1, 1992. The second 
condition is that that methodology or a successor to that methodology 
must have provided for payments exceeding the UPL under HHS' new rule 
at the time the new UPL rule takes effect.

A "State medicaid plan payment provision or methodology" within the 
meaning of the first Tier 3 condition is a provision or methodology for 
making supplemental payments to local government facilities that 
complied with the then-existing UPL measure but might have exceeded the 
UPL measure under the rule proposed by HHS in October 2000. Wisconsin 
meets the first condition because it has had a provision for making 
supplemental payments to local government nursing homes in its state 
plan since 1986.

It is clear from correspondence between HCFA and Wisconsin Medicaid 
relating to Wisconsin's state plan for state fiscal year (SFY) 1987 
(July 1, 1986-June 30, 1987) that section 3.775 of Wisconsin's nursing 
home payment methodology meets the first condition for Tier 3 status 
under BIPA. In particular, a January 30, 1987, letter [NOTE 5] from 
Wisconsin to HCFA indicated that a HCFA central office employee had:

... expressed concern that under the terms o^ the amendment, 
Wisconsin's payments to nursing homes could approach the Medicare upper 
limit. He suggested that, as an assurance that the Medicare upper limit 
not be exceeded, the State compute the supplemental funds under this 
provision and provide that the facility's Medicaid rate including the 
supplemental payment cannot exceed the payment it would receive under 
Medicare principles. [He] would also approve a system wherein the state 
computes the aggregate payment to government facilities under Medicare 
principles and assures that total rates of facilities receiving 
payments under this program does not exceed the Medicare aggregate 
payment amount.

[Wisconsin] cannot make either of these changes to the amendment for 
the reasons outlined below:

1. Wisconsin determines the Medicare upper limit in the aggregate, and 
not on a facility-by-facility or category-of-ownership basis. This 
approach is justified under 42 CFR 447.253(b)(2), which reads: "The 
Medicaid agency's estimated average proposed payment rate is reasonably 
expected to pay no more in the aggregate for inpatient hospital 
services or long term care facilities services than the amount that the 
agency reasonably estimates would be paid for the services under the 
Medicare principles of reimbursement." (emphasis added). We have not, 
and do not intend to apply the Medicare upper limit on a facility-
specific or ownership-specific basis.

[Emphasis by underscoring in 1987 letter; emphasis by italics added.]

Based on the above passages from correspondence regarding Wisconsin's 
SFY 1987 state plan, it is clear that the only limit on total Medicaid 
payments to local government nursing facilities recognized under the 
SFY 1987 version of section 3.775 is what all facilities, government-
owned and privately owned, would have been paid in the aggregate under 
Medicare principles. Wisconsin rejected HCFA's request for an assurance 
that supplemental payments to government facilities would not exceed 
what government facilities alone would have been paid under Medicare 
principles. Ultimately, the only UPL assurance Wisconsin made in this 
regard was that:

... rates under the Wisconsin Medical Assistance Program for 1985-86 
including supplements paid under s. 3.775 of the Nursing Home Methods 
paid no more in the aggregate for long term care facility services than 
we reasonably estimate would be paid for the services under Medicare 
principles of reimbursement. [NOTE 6]

Notably, Wisconsin did not make the assurance requested by HHS that 
aggregate Medicaid payments to government facilities would not exceed 
what government facilities would have been paid in the aggregate under 
Medicare principles. This approach of using the aggregate Medicare 
upper limit for all facilities, private and government, as the Medicaid 
UPL for payments to local government facilities is precisely what HHS 
intended to prohibit in its proposed new UPI, rule in October 2000.

In numerous passages in the report, GAO suggests that in order for a 
state to have a "State medicaid plan payment provision or methodology" 
(and thus meet the first Tier 3 condition) the state must have had a 
long-standing practice of making extremely large supplemental payments 
to local government facilities.[NOTE 7] There is no support for this 
view in the language of BIPA that establishes the eligibility criteria 
for the Tier 3 transition period. Congress expressly required HHS to 
issue a final UPL rule that is based on the October 2000 proposed rule 
and that "provides for a transition period in accordance with 
subsection (b)." [NOTE 8] In turn, subsection (b) provides that:

The final regulation ... shall provide that, with respect to a State 
described in paragraph (3), the State shall be considered to be in 
compliance with the final regulation ... so long as, for each State 
fiscal year [that begins after September 30, 2002, and ends on 
September 30, 20081, the State reduces payments ... by 15 percent in 
the first such State fiscal year, and by an additional 15 percent in 
each of the next 5 State fiscal years. [NOTE 9]

As noted at page 1, above, a "State described in paragraph (3)" under 
BIPA is a State with a payment provision or methodology which "was 
approved, deemed to have been approved, or was in effect" on or before 
October 1, 1992 and "provides for payments that are in excess of the 
upper payment limit test established under the final regulation..." 
[NOTE 10] The use of the past tense in the first clause and present 
tense in the second clause of this provision makes clear Congress' 
intent not to limit Tier 3 transition status in the way suggested by 
GAO. Had Congress intended to limit Tier 3 to States with a 
longstanding practice of making supplemental facility payments that 
would have exceeded the new UPL test, it would not have used the 
present tense in clause (B). The BIPA Tier 3 language in no way 
suggests that in order to qualify for the Tier 3 transition period a 
State must have had a payment provision or methodology that, on or 
before October 1, 1992, provided for payments that would have exceeded 
the new UPL test.

Had Congress intended to require a state, in order to qualify for Tier 
3, to have had a long-standing practice of making supplemental payments 
to local government facilities that in fact exceeded what would be 
permitted under the new UPL rule, it would have done so expressly. The 
GAO report makes repeated reference to CMS' stated objectives for 
granting transition periods, 
as indicated in the preamble to the final regulations.[NOTE 11] 
However, CMS' objectives in creating the original 2 transition periods 
are not relevant in determining Congress' intent in mandating the Tier 
3 transition period. The federal legislation is clear on its face in 
this regard.

Moreover, even were the BIPA Tier 3 language not clear on its face, the 
legislative history of the provision makes Congress' intent plain:

The provision [BIPA sec. 705] also requires the final regulation to 
stipulate a third set of rules governing the transition period for 
certain states. This additional set of rules would apply to states with 
payment arrangements approved or in effect on or before October 1, 
1992, or under which claims for federal matching were paid on or before 
that date, and for which such payments exceed the UPLs established 
under the final regulation. For these states, a 6-year transition 
period would apply ...[NOTE 12]

The passage "such payments exceed the UPLs established under the final 
regulation" underscores Congress' directive to CMS' to apply the new 
UPL test to State supplemental payment methodologies that were in 
effect at the time the new UPL rule took effect. Otherwise, the passage 
would have read "such payments would have exceeded the UPLs established 
under the final regulation," or words to similar effect.

Finally, subsequent Congressional pronouncements admonish CMS not to 
stray from a literal interpretation of BIPA Tier 3 language. During 
2001 budget deliberations, the Senate Committee on Appropriations noted 
that the BIPA UPL provisions,

which included reasonable transition periods for States to adopt a 
revised standard for treatment of their programs, were adopted after 
intense bipartisan negotiations among the House, Senate, and 
administration. ... The Committee is pleased that the administration, 
in January and September 2001, supported this Medicaid upper payment 
limit agreement in the promulgation of Medicaid upper payment limit 
regulations. Further, the Committee 
reiterates its commitment to both the letter and spirit of this 
agreement, and directs the administration to maintain its course in 
complying with congressional intent. Any subsequent modifications 
should be done only after the administration has had an opportunity to 
assess the implementation of the new regulations and only in 
consultation with the States and their Medicaid programs, as well as 
the other stakeholders. [NOTE 13]

For the reasons stated above, Wisconsin meets the first condition, set 
forth in BIPA section 705:

(b) (3) (A), to qualify for Tier 3 transition to compliance with the 
new UPL rule.

Wisconsin also meets the second condition for qualifying for the Tier 3 
transition period. Under BIPA section 705 (b) (3) (B), there are two 
alternative ways to meet this second condition. Wisconsin meets it both 
ways.

One way to meet the second Tier 3 condition is to have a payment 
methodology that, at the time the new UPL rule took effect, provided 
for payments that are in excess of the UPL under the new rule. The new 
UPL rule took effect March 13, 2001. [NOTE 14] Wisconsin's state 
Medicaid plan for SFY 2001 (effective July 1, 2000 - June 30, 2001), 
which was in effect with respect to payments made on March 13, 2001, 
provided for payments to local government facilities substantially 
exceeding the UPI, provided for in the new rule that took effect on 
that date. (Indeed, GAO appears to concede this point in its draft 
report.):

The other way for a State to meet BIPA's second Tier 3 condition is for 
the State to have made payments for "the State fiscal year which begins 
during 1999" that would have exceeded the UPL under the new rule if the 
payments had been continued to the effective date of the new rule. In 
Wisconsin, "the State fiscal year which begins during 1999" is SFY 2000 
(July 1, 1999 - June 30, 2000). The September 23, 2002 letter from 
Wisconsin Medicaid Director Peggy Handrich to CMS' Cheryl Harris (see 
enclosed) establishes that Wisconsin Medicaid payments to local 
government facilities for Wisconsin SFY 2000 would have exceeded the 
UPI, under the new rule by over $25,000,000.

Though CMS accepted Wisconsin's calculations as to what the UPL under 
the new rule would have been in Wisconsin SFY 2000, GAO asserts in its 
draft report that Wisconsin underestimated this figure. GAO concludes 
that had Wisconsin accurately estimated this figure, Wisconsin's 
actual payments to local government facilities during SFY 2000 would 
not have exceeded the UPL. GAO estimates that payments under Medicare 
payment principles would have been at least $162.77 per bed day, as 
opposed to the $133.50 per bed day estimated by Wisconsin. This 
difference would more than offset the $25,000,000 excess calculated by 
Wisconsin for SFY 2000.

Because it was unclear from GAO's draft report the basis upon which GAO 
arrived at its estimate of $162.77 per bed day, Wisconsin requested 
that GAO provide Wisconsin all computations and other analysis leading 
to this estimate. GAO's response to this request demonstrates that in 
arriving at the $162.77 per bed day estimate GAO fundamentally 
misapplied applicable methodology for establishing UPLs. [GAO's 
response is attached as an appendix to this document.]

The fundamental flaw in GAO's approach to estimating Wisconsin's SFY 
2000 UPL relates to that part of the UPL calculation in which GAO used 
Medicare rates rather than Medicare rate-setting methodology in 
approximating what Medicare would have paid for Medicaid recipients 
residing in Wisconsin nursing homes. The governing federal rule 
provides, in pertinent part:

...aggregate Medicaid payments to a group of facilities within one of 
the categories described in paragraph (a) of this section may not 
exceed a reasonable estimate of the amount that would be paid for the 
services furnished by the group of facilities under Medicare payment 
principles in subchapter B of this chapter. [NOTE 15]

During Wisconsin's SFY 2000 Medicare reimbursed skilled nursing 
facilities for services provided to Medicare residents under the 
transition provisions of its new prospective payment system (PPS). 
During PPS transition, a facility's Medicare payment was computed as a 
weighted average of a "Federal payment" and a "facility-specific 
payment." The Federal payment was computed by applying a published 
schedule of 44 per diem rates varying according to the intensity of 
service needs of the facility's Medicare residents. The facility-
specific payment was computed by applying the facility's own historical 
average Medicare cost per diem to the current population of Medicare 
residents. The historical average cost was taken from the facility's 
1995 Medicare cost report inflated to the payment period. In this 
manner, Medicare payment principles at that time based a portion of the 
payment on each facility's historical cost per diem for the population 
being served (the facility-specific payment), with the remaining 
portion based upon a pricing schedule dependent upon the population's 
service needs (the Federal payment).

To apply this Medicare payment principle to a population of Medicaid 
residents in calculating the UPL for SFY 2000, Wisconsin computed a 
Medicaid Federal payment and a Medicaid facility-specific payment. 
Wisconsin computed the Medicaid Federal payment by applying the 
published Medicare rate schedule to the population of Medicaid 
residents according to each resident's service needs classification. 
GAO did not challenge this portion of the Wisconsin UPL calculation for 
SFY 2000.

Wisconsin computed the Medicaid facility-specific payment using the 
same methodology used to compute the Medicare facility-specific payment 
at that time. [NOTE 16] Specifically, Wisconsin derived the Medicaid 
facility-specific average costs using 1995 Medicaid cost report 
information (adjusted to remove non-Medicaid and developmentally 
disabled resident costs) inflated to the payment period.

Rather than applying Medicare payment principles to estimate the 
facility-specific payment component of Wisconsin's UPL, GAO used 
"actual rates used by Medicare to pay Wisconsin nursing homes in SFY 
2000." [Appendix, p. 1, par. 3. (Emphasis added.)] In applying Medicare 
payment principles to Medicaid residents, it is inappropriate to use 
actual Medicare facility-specific rates, since these rates are derived 
from average Medicare resident costs. Medicare resident costs per diem 
are typically much greater than Medicaid costs per diem. This is true 
because, in order to qualify for Medicare payment for nursing home 
care, a resident must have been admitted following an inpatient 
hospital stay, and the services the patient receives in the nursing 
home must have been necessitated by the condition for which the patient 
was hospitalized.[NOTE 17] Moreover, Medicare generally only covers the 
kinds of services in a nursing home that would be covered if furnished 
to a hospital inpatient. [NOTE 18] Medicare payment is made for a 
maximum of 100 days of post-hospitalization nursing home care. 
[NOTE 19] Unlike Medicaid, Medicare does not cover long-term care 
services in nursing homes. Thus, GAO's use of actual Medicare facility-
specific rates results in an upper payment limit for Medicaid residents 
that is excessive.

Wisconsin was reasonable and consistent in applying Medicare payment 
principles to the population of Medicaid residents. The resulting upper 
limit of $133.50 per patient day results from a direct application of 
those principles to the best available information about Medicaid 
nursing home residents. GAO's recomputed value of $162.77 results from 
a facility-specific component that is based upon Medicare resident 
average per diem costs. GAO's approach does not yield a reasonable 
payment for Medicaid residents, and, contrary to federal rules, is not 
based on Medicare payment principles.

While, as GAO notes, CMS has never published detailed guidance with 
respect to the Medicare UPL calculation, what federal guidance exists 
is consistent with Wisconsin's approach to calculating the UPL for SFY 
2000. The State Medicaid Manual emphasizes that federal UPL rules:

... permit a State greater discretion in determining if the requirement 
has been met and emphasize the State's flexibility to develop 
procedures for applying the upper limit test. They relieve the State of 
the burden of having to use the detailed cost finding principles 
required by Medicare or of complying with a prescriptive formula 
approach in ascertaining what would have been paid for such services 
under the Medicare principles of reimbursement. [NOTE 20]

That section goes on to state that "in determining if you meet the test 
of the upper payment limit, your estimate must use the Medicare 
principles in effect at the time," and that,

[i]n determining whether you comply with the Medicare upper limit ... 
you need not follow exactly every detailed procedure used to implement 
either of these principles in the Medicare program, so long as the 
principles are satisfied.

The repeated references in the State Medicaid Manual section to 
application of Medicare principles in calculating the UPL belies GAO's 
mechanistic approach of simply substituting Medicare facility-specific 
rates for Medicaid services, regardless of the applicability of those 
rates to the services received or the people who receive them. That 
GAO's approach is inconsistent with Medicare principles becomes even 
clearer when one considers that the UPL applies as well to Medicaid 
payments for services for which Medicare does not pay at all, such as 
the services provided in an Intermediate Care Facility for the Mentally 
Retarded (ICF/MR). Clearly, the UPL rule does not envision using a 
Medicare payment rate as the UPL measure for services Medicare does not 
cover.

For all of the reasons set forth above, GAO's conclusion that Wisconsin 
was overpaid by over $500 million in Federal Medicaid funds is 
incorrect.

Other Issues:

Wisconsin's SFY 2001 Supplemental Payment Methodology was Legal and 
Appropriate. At page 4 and again at pages 16 and 20, GAO contends that 
HCFA "should not have approved" Wisconsin's SFY 2001 methodology for 
making supplemental Medicaid payments to local government nursing 
homes. GAO bases its position in this regard on the fact that 
Wisconsin's SFY 2001 state plan was amended to include the supplemental 
payment methodology on February 7, 2001, while HCFA had published a 
rule which would prohibit such payments on January 12, 2001. GAO 
omitted the fact that the new rule did not take effect until March 13, 
2001, and applied only to state plan amendments submitted on or after 
that date. [NOTE 21] State plan amendments, like Wisconsin's, that were 
submitted prior to March 13, 2001 were reviewed by HCFA under the 
former UPL rule, which aggregated the UPI, measure across government 
and non-government facilities. Under that rule, HCFA had no alternative 
but to approve Wisconsin's SFY 2001 supplemental payment methodology.

Supplemental Payments to Nursing Homes in Wisconsin were Lawfully 
Financed. At pages 7 and 8 and elsewhere in its report GAO notes that 
"typically" supplemental Medicaid payments to local government 
facilities are returned to the state. Wisconsin does not engage in this 
practice. In Wisconsin, supplemental payments to local government 
facilities have been financed through a combination of certified public 
expenditures by the counties that operate the facilities and by 
intergovernmental transfers of public funds from the counties to the 
state. Both of these methods of funding Medicaid payments are expressly 
authorized in federal law. [NOTE 22] Indeed, Congress has expressly 
prohibited HHS from restricting States' use for Medicaid expenditures 
of funds derived from local taxes that are transferred from local units 
of government to the state. [NOTE 23]

Wisconsin Acted Reasonably in Estimating the UPL during the Transition 
from Cost-Based to Prospective Medicare Reimbursement. At page 26 GAO 
asserts that "none of the states accurately determined Medicare rates 
when calculating their UPLs, because they did not correctly account for 
changes that occurred when Medicare moved from cost based reimbursement 
to prospective rates starting in 1998." In a footnote to that text, GAO 
states that "[s]ome states applied a standard blending rate to all 
nursing homes, not accounting for the different blended rates that were 
specific to each facility. Two states selectively chose to use a 
standard rate only when it would generate the most federal dollars.":

Wisconsin's SFY 1999 and 2000 UPL calculations took into account 
Medicare's transition from a cost-based to a prospective rate system. 
In both years, facilities were grouped according to their status under 
the transition rules to the new payment system, some using 100% of the 
Federal PPS rate and others following the phase-in schedule from 
facility-specific rates to the Federal rate. The SFY 1999 calculation 
blended the old three-part upper limit (i.e., the minimum of routine 
service limit, allowable cost and private-pay rates) with the FFY 1999 
PPS rates. The FFY 1999 rates for the phase-in facilities were, in 
turn, 75% of facility-specific rate and 25% of the Federal rate.

For SFY 2000, the upper limit was a blend of the FFY 1999 and the FFY 
2000 PPS rates. For the phase-in facilities, the weights on the 
facility-specific rates were 75% in FFY 1999 and 50% in FFY 2000. The 
BBRA 1999 provision to allow the phase-in facilities to elect to move 
to 100% of the Federal rate starting with the cost report year on or 
following 1/1/2000 was also incorporated in the FFY 2000 portion of the 
upper limit calculation.

Using Federal Matching Funds for Other Medicaid Services is Legal and 
Appropriate. At pages 29 to 30, GAO asserts that states' use of federal 
matching funds to pay for additional Medicaid services "effectively 
increas[es] those states' federal match rate." hi particular, GAO notes 
that:

In Wisconsin, for example, we found that, by obtaining excessive 
federal matching payments and using these funds as the state share of 
other Medicaid expenditures, the state effectively increased the 
federal matching share of its total Medicaid expenditures from 59 
percent to 68 percent in SFY 2001.

States routinely use claimed federal Medicaid funds to pay for 
additional Medicaid services. This is entirely consistent with 
principles of federal Medicaid claiming. When a State makes a Medicaid 
expenditure with state and local public funds and claims its 
proportionate federal match for that expenditure, the federal matching 
funds received by the State are deposited in state appropriation 
accounts and may be expended in any manner directed by the state 
legislature. Spending those matching funds as state Medicaid 
expenditures for additional services is in no way inappropriate, and it 
has never been held that this practice affects the state's overall 
federal matching share.

Wisconsin Uses Federal Funds Generated by Supplemental Local Government 
Facility Payments Exclusively to Pay for Medicaid-Covered Services. In 
the chart on page 31 of the report GAO indicates that "[f]unds 
generated by the State's UPL arrangement are deposited in a state fund, 
which is used primarily to pay for Medicaid-covered services ... ." 
[Emphasis added.] Under Wisconsin law, federal Medicaid funds claimed 
based on supplemental payments to local government facilities are 
deposited in a trust fund and may only be used to pay for Medicaid 
services.

NOTES: 

[1] 65 FR 60151, October 10, 2000.

[2] The Medicare, Medicaid and SCHIP Benefits Improvement and 
Protection Act of 2000, Public Law 106-554. 

[3] This transition period is now commonly referred to as "Tier 3."

[4] Public Law 106-554, section 705 (b) (3).

[5] A copy is enclosed. This letter and the rest of the correspondence 
relating to the 1986 version of section 3.775 was provided to GAO 
during its investigation.

[6] March 17, 1987 letter from Christine Nye to Judith Stec, enclosed.

[7] See, e.g., p. 14: "CMS's granting 8-year transition periods to 
Nebraska and Wisconsin is not consistent with the objectives that the 
agency published in the preamble to its January 2001 regulation, as 
neither state had in place the type of longstanding excessive payment 
arrangement the agency intended to curtail." Page 15: "Although 
Nebraska and Wisconsin did have supplemental payment arrangements for 
nursing homes in place as of October 1, 1992, neither state had an 
arrangement that ... made extremely large payments to units of local 
governments on behalf of a few locally-owned nursing homes... ." [Also 
see accompanying footnote.] Page 16: "...Wisconsin had no reliance on 
excessive UPL payments prior to 2001 that would justify an 8-year 
transition period given the purpose stated by CMS." Page 17: "... 
Wisconsin's 2001 UPL arrangement differs substantially from the 1985 
arrangement, and granting the state an 8-year transition period is 
inconsistent with the agency's stated objectives in curtailing 
excessive UPL schemes.":

[8] BIPA, sec. 705 (a) (2).

[9] BIPA, see. 705 (b) (1). [Emphasis added.]

[10] BIPA, sec. 705 (b) (3) (A) and (B). [Emphasis added.]

[11] See, e.g., p. 4: "Over the 8-year transition periods, Nebraska and 
Wisconsin are eligible to receive about $633 million more in excessive 
federal matching funds than they would have been eligible for under 
shorter transition periods based on the stated purposed of CMS's 
regulation and transition policy." Page 9: "HCFA proposed such 
transition periods because it recognized that some states, as part of 
their overall state budgets, had come to rely on the additional money 
they were receiving through these schemes." Page 14: "CMS's granting 8-
year transition periods to Nebraska and Wisconsin is not consistent 
with the objectives that the agency published in the preamble to its 
January 2001 regulation, as neither state had in place the type of 
longstanding excessive payment arrangement the agency intended to 
curtail." Page 16: "... Wisconsin had no reliance on excessive UPI, 
payments prior to 2001 that would justify an 8-year transition period 
given the purpose stated by CMS." Page 17: " ... [G)ranting [Wisconsin) 
an 8-year transition period is inconsistent with the agency's stated 
objectives in curtailing excessive UPL schemes." Page 20: "Under the 
final rule allowing a state with an 8-year transition period ..., 
Nebraska and Wisconsin have been able to generate significantly more in 
excessive federal UP L payments than they could have under shorter 
transition periods consistent with the purpose of the 2001 regulation 
as stated by CMS.":

[12] House Report 106-1033, Committee on Conference, December 15, 2000; 
See also House Report 106-1004, Committee on Conference, October 26, 
2000.

[13] Senate Report 107-84, Senate Committee on Appropriations, October 
11, 2001.

[14] Final Rule, 66 FR 3148 at 3148.

[15] 42 CFR § 447.272 (b) [emphasis added].

[16] (See 63 FR 26251 at 26286, May 12, 1998.):

[17] 42 US.C. §1395x (i).

[18] 42 U.S.C. §1395f (a) (2) (B). 

[19] 42 U.S.C. §1395d (a) (2).

[20] State Medicaid Manual, § 6005.

[21] See State Medicaid Directors Letter #01-017.

[22] 42 CFR § 433.51 (b); 42 USC § 1396a (a) (2). 

[23] 42 USC Sec. 1396b (w) (6) (A)..

The following is our response to the State of Wisconsin's comments 
provided on December 23, 2003.

GAO's Response to the State of Wisconsin's Comments: 

Our responses to Wisconsin's comments are numbered below to correspond 
with the state's various points. Wisconsin generally contends that (1) 
the state's arrangement met the first BIPA condition required to 
receive an 8-year transition period (see Wisconsin's comments 1 through 
4 and our corresponding response), (2) the state's arrangement met the 
second BIPA condition required for an 8-year transition period (see 
Wisconsin's comments 5 and 6 and our corresponding response), and (3) 
the state's arrangement is legal and appropriate (see Wisconsin's 
comments 7 through 11 and our corresponding response). In view of the 
evidence we reviewed, we continue to believe that CMS's decision to 
grant Wisconsin an 8-year transition period was not consistent with the 
objectives stated in the preamble to the agency's final UPL rule.

1. We disagree with Wisconsin that we incorrectly concluded that a 16-
month transition period--rather than an 8-year transition period--was 
warranted for the state, with the result that the state was not 
entitled to receive more than $500 million in federal Medicaid funds. 
We continue to maintain that, because the state's 2001 UPL arrangement 
was established after HCFA had taken action to limit the arrangements, 
Wisconsin's arrangement should not have been approved in the first 
place. In addition, we did not conclude that CMS's decision was 
unlawful. Rather, we believe that if CMS's transition period decisions 
had been more consistent with the objectives stated in the preamble to 
the agency's January 2001 regulation, CMS's decisions would have better 
protected the fiscal integrity of the Medicaid program.

2. Wisconsin also explained at length the legal basis for its 8-year 
transition period in terms of the two conditions set forth in BIPA, 
including an interpretation of BIPAs reference to a "payment provision 
or methodology" that defines the phrase as a provision or methodology 
for making supplemental payments. We have revised the report to 
expressly acknowledge that CMS's transition period decision with 
respect to Wisconsin was permissible given the statutory language. But 
we disagree with the state's position that BIPA defines a "payment 
provision or methodology" as a supplemental payment provision. Neither 
the plain language of BIPA nor its legislative history refers to 
"supplemental payments" or otherwise provides a basis for the state's 
definition.

3. We disagree with the state's contention that correspondence between 
HCFA and Wisconsin's Medicaid program relating to the state's plan for 
state fiscal year (SFY) 1987 provides evidence that the state's 
methodology (and the supplemental arrangement it established) was the 
type of arrangement that the agency described in its proposed UPL rule 
in October 2000. The state implied that it had a problematic UPL 
arrangement before October 1992 because it used an aggregated UPL test 
in providing HCFA with the required assurance that state Medicaid 
payments did not violate the UPL requirement. From our review of the 
actual payment methodology that Wisconsin had in place in 1992, 
however, we concluded that the state's supplemental payment was small 
and not the type of arrangement HCFA said it was trying to address with 
its UPL regulation.

4. We disagree with Wisconsin regarding the factors we weighed in 
assessing CMS's decision to grant the state an 8-year transition 
period: first, that we erred in considering whether the state had a 
long-standing practice of making extremely large supplemental payments 
to local government facilities; and second, that we erred in 
considering the objectives identified by CMS for the new UPL regulation 
in the preamble to the final rule. Wisconsin stated that BIPA provides 
no support for our position. We acknowledge that BIPA does not refer to 
these specific factors, but both are key to determining whether a state 
had developed budgetary reliance on the excessive federal funds as a 
result of using UPL arrangements over a number of years--CMS's stated 
reason for granting transition periods.

5. We do not dispute that Wisconsin may have met BIPA's second 
condition for qualifying for an 8-year transition period. Regarding the 
first way to meet this condition--having payments that exceeded the UPL 
under the new regulation as of its effective date, March 13, 2001--we 
agree that the state made a substantial UPL payment before that date. 
In fact, the state completed a $637 million electronic transfer of 
funds to participating counties on March 12, 2001. Although this 
transfer for a Medicaid payment clearly exceeded the new UPL regulation 
and took place before the March 13, 2001, effective date, we question 
CMS's decision to treat this payment arrangement as a successor to a 
significantly different and smaller supplemental payment arrangement in 
place on or before October 1, 1992.

6. Regarding the second way for a state to meet BIPA's second 
condition--to have made payments for the state fiscal year beginning in 
1999 that would have exceeded the UPL under the new rule had the 
payments been continued--we disagree with Wisconsin's assertion that 
our approach to estimating the state's SFY 2000 UPL was fundamentally 
flawed. Our approach to estimating Wisconsin's UPL, which shows that 
the state complied with the new UPL and would not qualify for a 
transition period, applied the same types of principles used by other 
states whose UPL estimation methods were approved by CMS. We agree, 
however, that Wisconsin's methodology, while more conservative, could 
be considered reasonable. We have revised our report to recognize that 
either the state's approach or the one we used could be considered 
reasonable, given the lack of CMS guidance for appropriate methods for 
calculating the UPL.

7. Wisconsin asserts that its SFY 2001 supplemental payment methodology 
was legal and appropriate and disagrees with the conclusions from our 
earlier report that HCFA's approval of this arrangement was 
unjustified.[Footnote 59] We disagree with the state's contention that 
the agency's approval of the state's arrangement was appropriate given 
its stated objectives and regulation to limit such arrangements; we 
explained our position in greater detail in our previous report.

8. We disagree with Wisconsin's assertion that--because the 
supplemental payments were lawfully financed through the use of 
intergovernmental transfers (IGTs) and certified public expenditures 
(CPEs) from local governments such as cities and counties--the UPL 
payments are appropriate. We agree that IGTs and CPEs are appropriate 
and legal mechanisms for the transfer of funds between government 
entities. In addition, we found that the state's UPL arrangement serves 
to increase federal Medicaid matching payments without a commensurate 
increase in state expenditures, which in our view is not appropriate or 
consistent with the goals and fiscal integrity of the Medicaid program. 
We also find Wisconsin's description of its UPL arrangement misleading. 
Wisconsin maintains that it does not require the local government 
facilities to return the supplemental Medicaid payments to the state. 
While technically correct, Wisconsin counties first transfer the total 
amount of the UPL payment (federal and state share) to the state and 
then the state transfers the funds back to the county--with the same 
result that the state does not increase its share of Medicaid 
expenditures.

9. As stated in comment 6, we have revised our report to acknowledge 
that Wisconsin's method for accounting for the Medicare's transition 
from a retrospective cost-based reimbursement system to a prospective 
rate system may be considered reasonable in light of CMS's lack of 
guidance on how states should calculate their UPL.

10. We disagree with Wisconsin's contention that using federal matching 
funds obtained through UPL arrangements as the state's share of other 
Medicaid expenditures is appropriate. Under Wisconsin's arrangement, 
the state is not seeking federal reimbursement for payments made to 
local governments for Medicaid services provided to Medicaid-covered 
beneficiaries, as the state implies. Instead, the state is seeking 
reimbursement for a same-day wire transfer of a bank loan between 
county and state. As the state acknowledges, the claimed federal funds 
are placed in a trust fund, from which they are drawn upon as the state 
share for other Medicaid payments. The state thus increases federal 
Medicaid payments without a commensurate increase in its own payments. 
This practice effectively raises the federal government's share of 
Medicaid expenditures beyond the state's Medicaid-formula-based 
matching rate.

11. We have revised our report to incorporate Wisconsin's clarification 
that it uses the federal funds generated by its UPL arrangement to pay 
for Medicaid-covered services. It is important to clarify, however, 
that the funds are likely not being used for the Medicaid services 
provided to the Medicaid beneficiaries on whose behalf the funds are 
claimed. Rather, as we point out in comment 10, the federal funds are 
being used to effectively replace state funding for other Medicaid 
services and beneficiaries.

[End of section]

Appendix IX: GAO Contact and Staff Acknowledgments: 

GAO Contact: 

Katherine Iritani, Assistant Director, (206) 287-4820: 

Acknowledgments: 

In addition, Tim Bushfield, Ellen Chu, Helen Desaulniers, Behn Miller 
Kelly, Terry Saiki, and Stan Stenersen made key contributions to this 
report.

[End of section]

Related GAO Products: 

Medicaid: HCFA Reversed Its Position and Approved Additional State 
Financing Schemes. GAO-02-147. Washington, D.C.: October 30, 2001.

Medicaid: State Financing Schemes Again Drive Up Federal Payments. GAO/
T-HEHS-00-193. Washington, D.C.: September 6, 2000.

State Medicaid Financing Practices. GAO/HEHS-96-76R. Washington, D.C.: 
January 23, 1996.

Michigan Financing Arrangements. GAO/HEHS-95-146R. Washington D.C.: 
May 5, 1995.

Medicaid: States Use Illusory Approaches to Shift Program Costs to 
Federal Government. GAO/HEHS-94-133. Washington, D.C.: August 1, 1994.

FOOTNOTES

[1] In fiscal year 2002, the latest year for which data are available, 
the federal government paid $139 billion of the $244 billion spent on 
Medicaid. For each state, the federal and state shares are derived from 
a formula known as the federal medical assistance percentage, which is 
designed to reflect each state's capacity to finance Medicaid services. 
The federal share of each state's expenditures ranged from 50 percent 
to 76 percent in 2002.

[2] A list of related GAO products is provided at the end of this 
report.

[3] In June 2001, HCFA was renamed the Centers for Medicare & Medicaid 
Services (CMS). We continue to refer to HCFA where agency actions were 
taken under its former name.

[4] See 65 Fed. Reg. 60,151 (2000).

[5] See Pub. L. No. 106-554, App. F, §705(a), 114 Stat. 2763A-463, 575-
576 (2000).

[6] See 66 Fed. Reg. 3148 (2001) (codified at 42 C.F.R. Part 447 
(2002)).

[7] The other major UPL payment schemes funneling excessive federal 
matching funds to states involved hospitals.

[8] U.S. General Accounting Office, Medicaid: HCFA Reversed Its 
Position and Approved Additional State Financing Schemes, GAO-02-147 
(Washington D.C.: Oct. 30, 2001).

[9] 42 U.S.C. §1396a(a)(30)(A).

[10] HCFA established UPLs in response to the Omnibus Budget 
Reconciliation Act of 1980 and the Omnibus Budget Reconciliation Act of 
1981. See 46 Fed. Reg. 47,964 (1981); 48 Fed. Reg. 56,046 (1983). 

[11] Until 1987, separate UPLs existed for different classes of 
services (such as nursing home or inpatient hospital care), but within 
a service category, separate UPLs were not set for different types of 
providers; that is, local government-owned and private nursing homes 
qualified for the same UPL for a given service.

[12] Throughout this report we refer to local-government facilities to 
identify a general ownership group. When discussing certain state UPL 
arrangements involving only nursing homes owned by county governments, 
however, we use the term county nursing homes.

[13] U.S. General Accounting Office, Medicaid: State Financing Schemes 
Again Drive Up Federal Payments, GAO/T-HEHS-00-193 (Washington D.C.: 
Sept. 6, 2000).

[14] Intergovernmental transfers are an authorized financing mechanism 
in which states may use revenue from local governments to help fund the 
state share of allowable Medicaid expenditures. 

[15] In a variant of this practice, some states required one or a few 
counties to send the state an amount equal to the total amount the 
state determined it could pay under the UPL for nursing home services. 
The state then sent the money back to the counties as a Medicaid 
payment, claimed the federal share of the payment, and then kept those 
federal dollars for its own purposes.

[16] See Congressional Budget Office, The Budget and Economic Outlook: 
Fiscal Years 2002-2011 (Washington D.C.: 2001).

[17] On the basis of elapsed time between September 30, 2002, and an 
ending date of September 30, 2008, BIPA describes the longest 
transition period as a 6-year transition. CMS, however, officially 
refers to this transition as an 8-year transition period because the 
agency measures from the passage of BIPA to the end of the transition 
period and rounds up to 8 years. In this report, we also refer to this 
transition period as an 8-year period.

[18] Under the first BIPA requirement, a state must have had a Medicaid 
plan payment provision or methodology that "was approved, deemed to 
have been approved, or was in effect on or before October 1, 1992 
(including any subsequent amendments or successor provisions or 
methodologies and whether or not a State plan amendment was made to 
carry out such provision or methodology after such date) or under which 
claims for Federal financial participation were filed and paid on or 
before such date . . . ." See section 705(b)(3)(A), 114 Stat. 2763A-
576.

[19] Under the second BIPA requirement, the state's Medicaid plan 
payment provision or methodology must have provided "for payments that 
are in excess of the upper payment limit test established under the 
final regulation . . . (or which would be noncompliant with such final 
regulation if the actual dollar payment levels made under the payment 
provision or methodology in the State fiscal year which begins during 
1999 were continued)." See section 705(b)(3)(B), 114 Stat. 2763A-576. 
States' fiscal years are set by states and do not necessarily align 
with the federal fiscal year. Most state fiscal years start July 1 and 
end June 30. In applying this requirement, CMS determined whether 
states made excessive payments during two different time periods. The 
first period was the effective date of the January 12, 2001 final 
regulation, which was March 13, 2001, and the second was the state 
fiscal year that began during 1999 (SFY 2000).

[20] On September 5, 2001, CMS established a 1-year transition period 
for states that submitted plan amendments before March 13, 2001, that 
did not comply with the final regulation's new UPLs. See 66 Fed. Reg. 
46,397 (2001).

[21] This estimate did not take into account a later revision of the 
UPL regulation to reduce the upper limit for local-government hospitals 
from 150 percent to 100 percent, a reduction that CMS estimated would 
save an additional $9 billion in federal expenditures over 5 years.

[22] When CMS asked a state to provide information related to its 
transition period, the state was under no requirement to reply within 
any specified time frame.

[23] This estimate is based on states with 5-year transition periods 
(Alabama, Michigan, New Hampshire, New York, North Dakota, Oregon, and 
Washington) and 8-year transition periods (Nebraska, Pennsylvania, and 
Wisconsin). As of October 2003, CMS had not reviewed all the states' 
estimated claims during their transition periods; some of these 
estimates may therefore change after review.

[24] In commenting on a draft of this report, CMS told us in January 
2004 that its reviews of 7 of the 8 states were completed. CMS's 
initial transition period decisions did not change as a result of these 
reviews being completed. 

[25] Specifically, states granted a 1-year transition period could 
exceed the UPL until the later of two dates: either November 5, 2001, 
or 1 year from the approved effective date of the state plan provision 
authorizing the UPL arrangement. States granted a 2-year transition 
period were allowed to exceed the UPL until September 30, 2002. See 42 
C.F.R. §§ 447.272(e)(2)(ii)(A) and (D).

[26] Although not legally binding, the preamble to a federal regulation 
is a valuable tool for determining the meaning or underlying purposes 
of an agency's regulation. See Conn. Gen. Life Ins. Co. v. 
Commissioner, 177 F.3d 136 (3d Cir. 1999).

[27] See 66 Fed. Reg. 3149-3150 (2002).

[28] HCFA provided the same description of the UPL problem in a July 
26, 2000, letter to all state Medicaid directors and in the preamble to 
its October 10, 2000, proposed regulation, which Congress, through 
BIPA, instructed CMS to finalize.

[29] See 66 Fed. Reg. 3161-3162 (2002).

[30] It is important to identify the type of supplemental payment 
arrangement a state had in place on or before October 1, 1992, because 
states may have had approved supplemental payment arrangements that did 
not have the characteristics of problematic UPL arrangements, such as 
extremely large payments to just a few local-government facilities. For 
example, states may have made small supplemental payments to local-
government facilities to compensate them for higher-cost patients or 
for meeting an important community need, such as providing a high 
volume of care to Medicaid beneficiaries.

[31] Although the supplemental payment provision was effective 
September 1992, the state did not make supplemental payments until July 
1993.

[32] In 1994, Wisconsin changed how it claimed federal matching funds 
under its supplemental payment provisions. It did not, however, 
incorporate methods identified in the preamble to the January 2001 
regulation; that is, the state did not identify the maximum amount 
allowed by aggregating payments of county and private nursing homes or 
make extremely large payments to only a few county nursing homes. In 
2001, CMS reviewed the state's 1994 method (and the associated claims 
for federal funds based on that method), found the state's method to be 
inappropriate, and prohibited its use.

[33] Wisconsin first proposed its UPL arrangement on February 7, 2001, 
almost a month after the January 12, 2001, regulation was published. 
Even though this arrangement mimicked the practices that HCFA was 
explicitly working to limit, the agency approved the arrangement on May 
8, 2001. In Wisconsin's arrangement, three counties wired a total of 
$637 million to the state, and the state then wired most of the money 
back to the counties, creating an illusion of a Medicaid payment. The 
wire transfers were completed in one day--March 12, 2001--the day 
before the effective date of the new UPL regulation, which is one of 
the dates for a state to qualify for an 8-year transition period. 
Further, because the state retained $373 million of the federal funds 
generated and then used them as the state share of other Medicaid 
expenditures, we estimate that the state generated $222 million more in 
federal funds. The payment transactions were wired to the three 
counties on behalf of five county nursing homes. Of the federal funds 
generated, about $150,000 was paid to consultants and banks as 
transaction fees and charges related to the wire transfer; the 
remainder was retained by the state. 

[34] GAO-02-147.

[35] In September 2001, HCFA established a UPL regulation specifically 
for several states with newly established UPL arrangements, including 
Wisconsin, which shortened the time they could operate these 
arrangements. See 66 Fed. Reg. 46,397 (2001). For a further discussion, 
see GAO-02-147.

[36] Through SFY 2003, Wisconsin had already obtained $895 million, 
which is $344 million more than it obtained the 9 years before January 
2001. Nebraska had already obtained $120 million through SFY 2003, and 
this year will surpass the $143 million it obtained in the years its 
arrangement was in place before the 2001 regulation (from when payments 
were first made under the arrangement in SFY 1994 through SFY 2000).

[37] CMS allows states to qualify for an 8-year transition period on 
the basis of payment provisions in place in SFY 2000 or SFY 2001. We 
use the term "latest" to refer to provisions in either of these state 
fiscal years.

[38] In Wisconsin, federal funding declined dramatically when the 
phase-out portion of the transition period began on July 1, 2003.

[39] HHS, Office of the Inspector General, Review of Medicaid Enhanced 
Payments to Local Public Providers and the Use of Intergovernmental 
Transfers, A-03-00-00216 (Washington, D.C.: 2001).

[40] According to CBO, supporters of such an action argue that the 
reduction would treat states more equitably by not allowing some states 
to obtain more federal Medicaid funds than intended by statute; 
opponents counter that requiring faster compliance would cut federal 
payments to some states when they are already facing budgetary 
difficulties. Congressional Budget Office, Budget Options (Washington, 
D.C.: 2003), 133, http: //www.cbo.gov/
showdoc.cfm?index=4066&sequence=0&from=0 (downloaded Oct. 6, 2003).

[41] CMS has also taken action to curb excessive UPL arrangements 
involving hospitals. For example, CMS's January 2001 regulation also 
established transition periods to phase out excessive UPL payments for 
inpatient and outpatient hospital services.

[42] In January 2003, the Departmental Appeals Board upheld a CMS 
decision to disallow $30 million (federal share) of New Hampshire's SFY 
2002 UPL claims. CMS determined that the state improperly used a 
payment calculation methodology that was inconsistent with its approved 
state plan. See New Hampshire Department of Health and Human Services., 
DAB No. 1862 (2003).

[43] Before establishing NIRT, review and approval of state plan 
amendments related to institutional payment provisions, including UPL 
arrangements, were primarily a CMS regional office responsibility.

[44] An earlier OIG audit identified about $155 million in unallowable 
UPL claims because the state had claimed matching funds for UPL 
payments that were not supported by actual expenditures. HHS, Office of 
Inspector General, Review of Medicaid Claims for County Nursing 
Facility Supplementation Payments by the Commonwealth of Pennsylvania, 
A-03-00-00211 (Philadelphia, PA: 2001).

[45] Over the 3 years following Medicare's implementation of 
prospective nursing home payments in 1998, many nursing homes were paid 
an amount that blended prior cost-based rates with new prospective per 
diem rates. Two states (Washington and Pennsylvania) blended Medicare 
facility-specific cost-based rates with their estimated Medicare 
prospective payment rates for Medicaid residents. In contrast, two 
other states (New York and Wisconsin) took a more conservative 
approach, blending Medicaid facility-specific cost-based rates with 
their estimated Medicare prospective payment rates. For Wisconsin in 
SFY 2000, this conservative methodology was to the state's advantage, 
because by this method, the state's small supplemental payment exceeded 
the new UPL, thus appeared to meet one condition for an 8-year 
transition period. Had the state, like Washington and Pennsylvania, 
used the less conservative approach approved by CMS, the result would 
have been a payment below the new UPL, not qualifying the state for an 
8-year transition period. 

[46] These states calculated their UPL by using, as a basis for their 
estimate, the number of residents that each nursing home served, rather 
than the number of resident days for each level of service. CMS's draft 
guidance maintains that, if states use a weighted average methodology, 
UPLs should be calculated on the basis of patient days, by level of 
service, rather than the number of patients, which can result in 
overestimates of what Medicare would have actually paid for those 
services.

[47] At CMS's request, the HHS OIG in 2003 began conducting UPL 
financial reviews in seven states with nursing home UPL arrangements 
(transition period in parenthesis): Alabama (5 years), Colorado (none), 
Kansas (2 years), Michigan (5 years), Indiana (none), New York (5 
years), and Oregon (5 years). According to officials, the reviews were 
limited to SFY 2003 payments, and no review had been completed as of 
October 2003.

[48] CMS's January 2001 regulation contained reporting requirements 
applicable to hospitals that were non-state government owned or 
operated. Subsequently, a January 18, 2002, regulation expanded the 
reporting requirements to all facilities, including nursing homes that 
receive payments during transition periods. See 67 Fed. Reg. 2602 
(2002).

[49] See 42 U.S.C. § 1396.

[50] See, for example, HHS, A-03-00-00216, 2001; Teresa A. Coughlin and 
Stephen Zuckerman, States' Use of Medicaid Maximization Strategies to 
Tap Federal Revenues (Washington, D.C.: Urban Institute, 2002); Andy 
Schneider and David Rousseau, Upper Payment Limits: Reality and 
Illusion in Medicaid Financing (Washington, D.C.: Kaiser Commission on 
Medicaid and the Uninsured, 2002).

[51] U.S. General Accounting Office, Medicaid: States Use Illusory 
Approaches to Shift Program Costs to Federal Government, GAO/
HEHS-94-133 (Washington D.C.: Aug. 1, 1994).

[52] For example, letters to Nebraska and Wisconsin, with the stated 
purpose of notifying the state of its transition period, were sent in 
February 2002. We began our fieldwork in December 2002.

[53] GAO-02-147.

[54] We reported in 2001 that HCFA's approval of Wisconsin's 2001 
proposed transition period was unjustified (GAO-02-147).

[55] Furthermore, on the basis of our analysis of the state's 2000 UPL 
calculation, we determined that had the state not used a conservative 
approach in estimating its SFY 2000 UPL it would have no excess UPL 
payments from SFY 2000 to phase out between SFY 2004 and 2009, even 
under an 8-year transition period.

[56] Under the 2001 regulation, UPLs exist for different types of 
services, such as inpatient hospital, outpatient hospital, and nursing 
home services, as well as for different facility ownership types, such 
as private, state owned, and local-government owned.

[57] For Medicare nursing home payment purposes, each state is divided 
into specific urban areas, and a different adjustment is calculated for 
each area on the basis of each area's labor costs. In addition, areas 
not falling into one of the urban areas are classified as rural and 
have a separate wage index.

[58] Another state--New York--relied on an estimated number of Medicaid 
nursing home resident days from 2 years earlier. Our analysis of 
information provided by the state indicated that New York's estimate 
was overstated, however, the state provided subsequent information that 
suggested its estimate was conservative. We were unable to reconcile 
the difference in time to be included in this report.

[59] GAO-02-147.

GAO's Mission: 

The General Accounting Office, the investigative arm of Congress, 
exists to support Congress in meeting its constitutional 
responsibilities and to help improve the performance and accountability 
of the federal government for the American people. GAO examines the use 
of public funds; evaluates federal programs and policies; and provides 
analyses, recommendations, and other assistance to help Congress make 
informed oversight, policy, and funding decisions. GAO's commitment to 
good government is reflected in its core values of accountability, 
integrity, and reliability.

Obtaining Copies of GAO Reports and Testimony: 

The fastest and easiest way to obtain copies of GAO documents at no 
cost is through the Internet. GAO's Web site ( www.gao.gov ) contains 
abstracts and full-text files of current reports and testimony and an 
expanding archive of older products. The Web site features a search 
engine to help you locate documents using key words and phrases. You 
can print these documents in their entirety, including charts and other 
graphics.

Each day, GAO issues a list of newly released reports, testimony, and 
correspondence. GAO posts this list, known as "Today's Reports," on its 
Web site daily. The list contains links to the full-text document 
files. To have GAO e-mail this list to you every afternoon, go to 
www.gao.gov and select "Subscribe to e-mail alerts" under the "Order 
GAO Products" heading.

Order by Mail or Phone: 

The first copy of each printed report is free. Additional copies are $2 
each. A check or money order should be made out to the Superintendent 
of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or 
more copies mailed to a single address are discounted 25 percent. 
Orders should be sent to: 

U.S. General Accounting Office

441 G Street NW,

Room LM Washington,

D.C. 20548: 

To order by Phone: 	

	Voice: (202) 512-6000: 

	TDD: (202) 512-2537: 

	Fax: (202) 512-6061: 

To Report Fraud, Waste, and Abuse in Federal Programs: 

Contact: 

Web site: www.gao.gov/fraudnet/fraudnet.htm E-mail: fraudnet@gao.gov

Automated answering system: (800) 424-5454 or (202) 512-7470: 

Public Affairs: 

Jeff Nelligan, managing director, NelliganJ@gao.gov (202) 512-4800 U.S.

General Accounting Office, 441 G Street NW, Room 7149 Washington, D.C.

20548: