Medicaid Waivers: HHS Approvals of Pharmacy Plus Demonstrations  
Continue to Raise Cost and Oversight Concerns (30-JUN-04,	 
GAO-04-480).							 
                                                                 
Under section 1115 of the Social Security Act, the Secretary of  
Health and Human Services may waive certain Medicaid requirements
for states seeking to deliver services through demonstration	 
projects. By policy, these demonstrations must not increase	 
federal spending. GAO has previously reported concerns with HHS's
approval process. GAO was asked to provide information on a new  
Medicaid section 1115 demonstration initiative called Pharmacy	 
Plus, intended to allow states to cover prescription drugs for	 
seniors not otherwise eligible for Medicaid. GAO reviewed the (1)
approval status of state proposals, (2) extent to which HHS	 
ensured that demonstrations are budget neutral, (3) basis for	 
savings assumptions, and (4) federal and state steps to evaluate 
and monitor the demonstrations. 				 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-04-480 					        
    ACCNO:   A10716						        
  TITLE:     Medicaid Waivers: HHS Approvals of Pharmacy Plus	      
Demonstrations Continue to Raise Cost and Oversight Concerns	 
     DATE:   06/30/2004 
  SUBJECT:   Budget controllability				 
	     Budget outlays					 
	     Disadvantaged persons				 
	     Drugs						 
	     Elderly persons					 
	     Federal/state relations				 
	     Health care programs				 
	     Health insurance cost control			 
	     Program evaluation 				 
	     State-administered programs			 
	     Waivers						 
	     Demonstration projects				 
	     HHS Pharmacy Plus Initiative			 
	     Medicaid Program					 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Product.                                                 **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************
GAO-04-480

United States General Accounting Office

GAO

Report to the Committee on Finance,

                                  U.S. Senate

June 2004

MEDICAID WAIVERS

    HHS Approvals of Pharmacy Plus Demonstrations Continue to Raise Cost and
                               Oversight Concerns

GAO-04-480

Highlights of GAO-04-480, a report to the Committee on Finance, U.S.
Senate

Under section 1115 of the Social Security Act, the Secretary of Health and
Human Services may waive certain Medicaid requirements for states seeking
to deliver services through demonstration projects. By policy, these
demonstrations must not increase federal spending. GAO has previously
reported concerns with HHS's approval process.

GAO was asked to provide information on a new Medicaid section 1115
demonstration initiative called Pharmacy Plus, intended to allow states to
cover prescription drugs for seniors not otherwise eligible for Medicaid.
GAO reviewed the (1) approval status of state proposals, (2) extent to
which HHS ensured that demonstrations are budget neutral, (3) basis for
savings assumptions, and (4) federal and state steps to evaluate and
monitor the demonstrations.

GAO recommends that the Secretary of HHS strengthen the processes for
approving and overseeing Pharmacy Plus and other Medicaid section 1115
demonstrations. HHS concurred with several recommendations for
strengthening demonstration approval and oversight but disagreed that
review criteria should be clarified and applied to already-approved
demonstrations. GAO maintains that the criteria for HHS's approvals should
be clear and consistently applied.

www.gao.gov/cgi-bin/getrpt?GAO-04-480.

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.

June 2004

MEDICAID WAIVERS

HHS Approvals of Pharmacy Plus Demonstrations Continue to Raise Cost and
Oversight Concerns

From January 2002 through May 2004, HHS reviewed Pharmacy Plus proposals
from 15 states and approved four: Florida, Illinois, South Carolina, and
Wisconsin. These demonstrations offer prescription drug coverage to
low-income seniors not otherwise eligible for Medicaid. HHS denied
proposals from Delaware and Hawaii as inconsistent with demonstration
guidelines; most of the rest were not under active review because HHS had
not determined how new Medicare prescription drug legislation will affect
proposed or operating Pharmacy Plus demonstrations. Over 5 years, the four
approved demonstrations will provide prescription drug coverage to half a
million low-income people age 65 or older, at a projected cost of about
$3.6 billion, of which the federal share would be about $2.1 billion.

HHS has not adequately ensured that the four approved demonstrations will
be budget neutral, that is, that the federal government will not spend
more with the demonstrations than without them. HHS approved the
demonstrations' 5-year spending limits using projections of cost and
beneficiary enrollment growth that exceeded benchmarks that HHS said it
considered in assessing budget neutrality, specifically, states' recent
average growth rates and projections for Medicaid program growth
nationwide. Neither HHS's negotiations with the states nor its rationale
for approving higher growth rates is documented. Using the benchmark
growth rates, GAO estimates that none of the four demonstrations will be
budget neutral and federal spending may increase significantly, for
example, by more than $1 billion in Illinois and $416 million in Wisconsin
over 5 years.

Unrealistic savings assumptions also contribute to demonstration spending
limits that are not likely to be budget neutral. States assumed that
keeping low-income seniors healthy-thus preventing them from spending down
their financial resources on health services and "diverting" them from
Medicaid eligibility-would generate sufficient savings to offset the
increased costs of providing a new drug benefit. GAO found neither state
experience nor other research to support such savings. Without
statespecific evidence, HHS approved savings assumptions for the four
states ranging from $480 million to $2 billion per state over 5 years. Had
more conservative assumptions been used to estimate demonstration savings,
the proposals likely could not have been approved as budget neutral.

Efforts by the states and HHS to evaluate and monitor the Pharmacy Plus
demonstrations are in their early stages. The four states have taken few
steps to put their own required evaluation plans into practice, and an
independent evaluation contracted by HHS and started in October 2002 is
scheduled to report in September 2005. In the interim, HHS has not ensured
that all states meet requirements for progress reporting on the
demonstrations. The information that states have submitted is often
insufficient for determining whether the demonstrations are operating as
intended, and this shortcoming will limit HHS's oversight capability.

Contents

  Letter

Results in Brief
Background
HHS Has Reviewed 15 Pharmacy Plus Proposals and Approved 4
HHS Has Not Ensured That Approved Demonstrations' Spending

Limits Will Be Budget Neutral
Pharmacy Plus Savings Assumptions Not Well Supported
States Have Taken Few Steps to Evaluate Demonstrations, and

HHS Has Not Ensured Sufficient or Timely Progress Reporting
Conclusions
Recommendations for Executive Action
Agency and State Comments and Our Evaluation

1

                                                                       3 7 10

17 24

29 35 36 37

Appendix I 	Calculating a State's Pharmacy Plus Spending Limit for All
Medicaid Seniors

Appendix II 	Denied, Withdrawn, and Pending Pharmacy Plus Demonstration
Proposals as of May 2004

Appendix III Comments from the Department of Health and

  Human Services 49

GAO's Response to the Department of Health and Human Services'
Specific Comments on GAO's Findings 59

Appendix IV Comments from the State of Florida

  Appendix V Comments from the State of Illinois 63

Appendix VI 	Comments from the State of Wisconsin 65
GAO's Response to the State of Wisconsin's Specific Comments 70

Appendix VII GAO Contact and Staff Acknowledgments

74

GAO Contact 74 Acknowledgments 74

Related GAO Products

Tables

Table 1: Highlights of Pharmacy Plus Demonstrations Approved as of May
2004 Table 2: HHS-Approved, State Historical, and CMS Actuary Growth

Rates Table 3: HHS-Approved and Benchmark 5-Year Spending Limits Table 4:
States' Projections of Medicaid Senior Populations after 5

Years, with and without HHS-Approved Pharmacy Plus Demonstrations

                                       11

                                     20 23

                                       25

Figure

Figure 1: Steps to Calculate Projected 5-Year Without-Demonstration Costs

Abbreviations

CBO Congressional Budget Office
CMS Centers for Medicare & Medicaid Services
FPL federal poverty level
HHS Department of Health and Human Services
HIFA Health Insurance Flexibility and Accountability
MMA Medicare Prescription Drug, Improvement, and

Modernization Act of 2003 OMB Office of Management and Budget PACE
Pharmaceutical Assistance Contract for the Elderly PACENET Pharmaceutical
Assistance Contract for the Elderly

Needs Enhancement Tier SCHIP State Children's Health Insurance Program SSA
Social Security Act

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. However, 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.

United States General Accounting Office Washington, DC 20548

June 30, 2004

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

Under section 1115 of the Social Security Act (SSA), the Secretary of
Health and Human Services may waive certain statutory requirements for
Medicaid-the joint federal and state program financing health care for
low-income families, certain seniors, and disabled individuals-in
connection with experimental, pilot, or demonstration projects that are
likely to promote program objectives.1 Because of the projects'
experimental nature, the Department of Health and Human Services (HHS)
requires demonstrations authorized under section 1115 to include
measurable objectives and an evaluation component. In addition, since the
early 1980s, HHS has required states to show that their proposals for
section 1115 demonstrations are "budget neutral" for the federal
government: that is, a proposed demonstration cannot raise federal
expenditures beyond what they would be under a state's existing program.

In January 2002, HHS announced the Medicaid Pharmacy Plus section
1115 demonstration initiative, offering states the opportunity to provide
a
prescription drug benefit to two groups-seniors (people age 65 or older)
and disabled individuals-whose incomes, although low, exceed levels
that would qualify them for full Medicaid eligibility. Under this
initiative,
HHS and the Centers for Medicare & Medicaid Services (CMS), the agency
within HHS that has primary responsibility for reviewing the
demonstration proposals, encourage states to test over 5 years whether
extending a drug benefit to seniors who are not eligible for Medicaid

1Section 1115 allows waivers of requirements in Medicaid, the State
Children's Health Insurance Program (SCHIP), and several other programs
authorized under SSA. See section 1115 (codified at 42 U.S.C. S: 1315
(2000)); see also section 2107(e) of SSA (codified at 42 U.S.C. S:
1397gg(e)) (regarding the applicability of section 1115 to SCHIP).

would maintain these seniors' health and hold down overall Medicaid
costs.2

Over the past decade, Congress and others have raised concerns about the
extent to which HHS has ensured that approved section 1115 demonstration
waivers promote the goals and fiscal integrity of both Medicaid and the
State Children's Health Insurance Program (SCHIP). In particular, Congress
has been concerned about HHS's waiver approval process and the federal
costs associated with some of the demonstrations. Our past work has found,
for example, that HHS's process for approving demonstrations is not always
clear or open to public input and that the department has not always
ensured the budget neutrality of approved demonstrations, thereby raising
federal expenditures.3

You asked us for information on the Pharmacy Plus initiative. We focused
our review on the following four questions:

1. 	How many states have applied for Pharmacy Plus demonstration waivers,
and what is the status of their proposals?

2. 	To what extent has HHS ensured that the approved demonstrations are
budget neutral to the federal government?

3. 	How well supported are states' assumptions about savings that may
accrue to Medicaid from the Pharmacy Plus demonstration waivers?

4. 	What steps are states and HHS taking to evaluate approved
demonstrations and to monitor if they are functioning as intended?

Our work is based on a review and analysis of Pharmacy Plus demonstration
waiver proposals considered and approved by HHS from January 2002 through
May 2004. Our analysis covers only demonstration proposals submitted in
response to HHS's Pharmacy Plus initiative.4 To

2Although CMS is the agency within HHS that has primary responsibility for
reviewing section 1115 demonstration waiver proposals, we refer to HHS
throughout this report as the primary program entity because the authority
to grant waivers for the demonstrations resides with the Secretary of HHS.

3A list of related GAO products appears at the end of this report.

4We do not include proposals to provide a prescription drug benefit to
low-income seniors under an amendment to an existing section 1115 Medicaid
managed care waiver, like those approved for Vermont and Maryland, because
such proposals are reviewed under different guidelines.

determine the status of demonstrations under this initiative, we analyzed
HHS data on all proposals it considered, including their number, outcomes,
and characteristics. For approved demonstrations, we analyzed the
applications as submitted by the states; HHS decision memorandums and
approval letters; the applications as ultimately approved; HHS's terms and
conditions for approved demonstrations; and, when available, the states'
plans (called operational protocols) for how the demonstrations will
operate. We also discussed the process of review and approval with
officials of the reviewing agencies-HHS, CMS, and the Office of Management
and Budget (OMB)-and we obtained information from officials representing
the states with approved demonstrations. To assess budget neutrality, we
obtained available budget justifications and documentation from state and
federal officials and discussed with them the budget negotiations
associated with each approved demonstration.5 To examine the assumptions
behind the initiative and the likelihood of associated savings, we
reviewed published literature and interviewed officials from the Kaiser
Commission on Medicaid and the Uninsured and the Congressional Budget
Office (CBO). We discussed plans for evaluating the approved
demonstrations with HHS and state officials. During our review, Congress
passed the Medicare Prescription Drug, Improvement, and Modernization Act
of 2003, which adds a drug benefit to Medicare, the federal program
providing health insurance for the majority of people age 65 or older
regardless of income.6 We considered the limited information available as
of May 2004 about the relationship between Pharmacy Plus and the act. We
conducted our work from December 2002 through June 2004 in accordance with
generally accepted government auditing standards.

                                Results in Brief

From January 2002 through May 2004, HHS reviewed Pharmacy Plus
demonstration waiver proposals from 15 states. It approved four
demonstrations (Florida, Illinois, South Carolina, and Wisconsin), denied
two (Delaware and Hawaii), and considered nine other proposals. The four
demonstrations, each approved for a 5-year period, cover most prescription
drugs and incorporate cost sharing by beneficiaries. Together, the four
demonstrations may enroll as many as half a million people age 65 or older
whose incomes are higher than states' limits for Medicaid

5To verify the accuracy of data included in state demonstration
applications, we discussed these data and their sources with state
officials.

6Pub. L. No. 108-173, 117 Stat. 2066.

eligibility. Combined 5-year state and federal spending for the four
approved demonstrations' new drug benefit is projected to total
approximately $3.6 billion, the federal share of which is estimated to be
about $2.1 billion. The two proposals that HHS denied were inconsistent
with Pharmacy Plus guidelines. One state had an existing program that
covered the same population proposed for the demonstration; the other
wanted to expand coverage to people with incomes above the initiative's
limit of 200 percent of the federal poverty level (FPL). As of May 2004,
most of the remaining nine demonstration proposals were not under active
review by HHS, primarily because the department had not determined how the
new Medicare legislation would affect the Pharmacy Plus initiative and
whether HHS would continue to review Pharmacy Plus demonstration
proposals.

HHS has not adequately ensured that the four approved Pharmacy Plus
demonstrations will be budget neutral, that is, that the federal
government will spend no more with the demonstrations than without them.
HHS has approved 5-year demonstration spending limits based on projections
of cost and beneficiary enrollment growth that exceeded benchmarks that
department officials said they considered in assessing states' proposals
for budget neutrality. These cost and enrollment growth benchmarks
incorporate the states' recent historical average growth rates and
projections developed by CMS's Office of the Actuary for Medicaid program
growth nationwide. HHS has not established written criteria for how it
reviews or approves state-proposed cost or enrollment growth rates against
these benchmarks. HHS's basis for approving state demonstrations' spending
limits as budget neutral is not clear, particularly for Illinois and
Wisconsin, for neither the department's negotiations with the states nor
its rationale for approving higher-than-benchmark rates of estimated
growth is documented. HHS's internal decision memorandums-which described
the factors that HHS, CMS, OMB, and others considered in reviewing the
demonstrations and which are not publicly available-did not provide the
rationale for the approved spending limits, and neither did the publicly
available demonstration approval letters. The approved 5-year spending
limit for Illinois is more than $2 billion higher than it would have been
had benchmark growth rates been applied; Wisconsin's approved spending
limit is $713 million more than it would have been with benchmark rates.
We estimate that over 5 years, with the approved demonstrations, the
federal government could spend over $1 billion more in Illinois, $416
million more in Wisconsin, $55 million more in Florida, and $42 million
more in South Carolina than without the demonstrations, thus not meeting
the stated policy of budget neutrality.

States' assumptions about savings that may accrue to Medicaid from the
Pharmacy Plus demonstrations are not well supported by state experience or
research. The four states with approved demonstrations assumed that the
cost of extending drug benefits to seniors now ineligible for Medicaid
would be offset by savings accrued because demonstration beneficiaries
would stay relatively healthy and therefore not deplete their income or
assets to Medicaid eligibility levels. HHS approved the four states'
savings assumptions-some projecting significant reductions in overall
Medicaid senior enrollment and ranging from $480 million to $2 billion per
state in combined federal and state spending over 5 years-without
state-specific data supporting these assumptions. One state's estimate of
savings, for example, was derived by determining how much the state needed
to save in order to demonstrate that its proposal would be budget neutral,
rather than how much the state could realistically expect to save. The
limited research available indicates that health care savings due to
improved access to prescription drugs are likely to be much less than what
states assumed and HHS approved. Had more conservative savings assumptions
been used to estimate the demonstrations' costs, the proposals likely
could not have been approved as budget neutral. Because federal liability
is capped by a 5-year spending limit for each demonstration, states will
be at risk if anticipated savings do not accrue and if their demonstration
spending reaches or exceeds the limits.

As of February 2004, efforts by the states and HHS to evaluate and monitor
the Pharmacy Plus demonstrations-and particularly states' efforts to begin
implementing their evaluation plans to address stated evaluation
objectives-were in their early stages. The four states with approved
demonstrations had taken few steps toward implementing the evaluation
plans required as a condition of demonstration approval: Florida and South
Carolina had not decided whether state officials or an outside entity
would conduct their evaluations. Illinois and Wisconsin officials believed
that participating in the independent evaluation contracted by HHS and
started in October 2002 would take the place of their own evaluations. HHS
officials, by contrast, told us that state evaluations were still
required. This independent evaluation is scheduled to report by September
2005. But in the interim, HHS has not ensured that each state submits in
its progress reports enough information for HHS to monitor that its
demonstration is functioning as intended, that the information is in a
form enabling comparisons across states, or that it is submitted in a
timely manner. This lack of information limits the department's oversight
capability. For example, Illinois did not submit required quarterly
progress reports during the demonstration's first year of operation, and
its annual report did not provide information that would allow HHS to
assess

whether the new drug benefit was enabling seniors to avoid enrollment for
full Medicaid benefits.

HHS's process for approving Medicaid demonstrations under the Pharmacy
Plus initiative continues to raise some of the same cost and oversight
concerns raised by other Medicaid section 1115 waiver approvals over the
past decade, including a failure to adequately justify the basis for
states' spending limits. We are recommending that the Secretary of HHS
clarify criteria for reviewing and approving demonstration spending
limits, consider applying these criteria to the four approved Pharmacy
Plus demonstrations, and publicly document the basis for Medicaid section
1115 demonstration approvals to better ensure that approved demonstrations
will not raise costs to the federal government. We are also recommending
that the Secretary ensure that approved Pharmacy Plus and other Medicaid
section 1115 demonstrations fulfill the objectives stated in their
evaluation plans and more actively monitor approved Pharmacy Plus and
other Medicaid section 1115 demonstrations.

In commenting on a draft of this report, HHS concurred with our
recommendations related to evaluating and monitoring the section 1115
demonstrations and documenting the basis for demonstration approvals. HHS
did not concur with our recommendations that it clarify criteria for
reviewing and approving states' proposed demonstration spending limits and
consider applying those criteria to its approval decisions for the four
Pharmacy Plus demonstrations. HHS indicated that while review criteria are
important, they cannot always be strictly applied because of variations in
state Medicaid programs and demonstration proposals, and it also stated
that the four approved demonstrations were based on wellsupported budget
estimates of future state spending. We have on several occasions raised
concerns with HHS about the budget neutrality of particular Medicaid
section 1115 demonstrations, and the Pharmacy Plus demonstrations are no
exception. We acknowledge that variations in state demonstration proposals
justify some review flexibility but believe that HHS has not clearly
articulated or documented the rationale for its decisions in approving
Pharmacy Plus demonstrations. Such lack of clarity raises questions about
whether these demonstrations, involving billions of federal dollars, have
been reviewed consistently.

We also provided a draft of this report to Florida, Illinois, South
Carolina, and Wisconsin. Illinois and Wisconsin officials commented that
we overstated the demonstrations' financial risk to the federal government
in light of data showing that to date, the demonstrations were operating
well within their spending limits. Both states asserted that their
pharmacy

Background

demonstrations were providing a valuable benefit to seniors and would be
budget neutral. Although we do not dispute the health benefit to seniors
of expanded access to prescription drugs, demonstrating savings to the
Medicaid program as a result of this expanded access is a separate issue.
Assumptions about such potential savings are not well supported by
research or by data from the states, even though all the states except
Wisconsin operated state-funded pharmacy assistance programs before
applying for their demonstrations. Florida commented that the spending
limit approved for its demonstration was less than 1 percent above the
benchmark spending level. South Carolina and Wisconsin provided technical
comments that were incorporated in the report as appropriate.

Established in 1965 under title XIX of SSA, Medicaid is the nation's
health care financing program for low-income families and certain people
who are age 65 or older or disabled. The program accounted for about $244
billion in federal and state expenditures in fiscal year 2002 and covered
an estimated 53 million people.7 The states and the federal government
share Medicaid spending according to a formula that provides a more
generous federal match for states where per capita income is lower.8

Medicaid is an open-ended entitlement program, meaning that the federal
government is obligated to pay its share of expenditures for all people
and services covered under an HHS-approved state Medicaid plan. To qualify
for federal matching payments, state Medicaid programs are required by law
to cover certain categories of beneficiaries, including pregnant women and
children with family incomes below specific limits, as well as individuals
with limited income and assets who are age 65 or older or

7In fiscal year 2001, the latest year for which complete data are
available, beneficiaries who were aged, blind, or disabled represented
about 30 percent of those served by Medicaid but accounted for more than
65 percent ($141 billion) of Medicaid's total $216 billion in federal and
state spending. Beneficiaries age 65 and older accounted for 26.5 percent
of total Medicaid spending, or $57 billion; blind or disabled individuals
accounted for nearly 39 percent, or $84 billion.

8In fiscal year 2003, the federal share of individual states' Medicaid
expenditures ranged from 50 to 77 percent. Federal Medicaid matching rates
were increased temporarily by 2.95 percentage points from April 1, 2003,
through June 30, 2004, pursuant to title IV of the Jobs and Growth Tax
Relief Reconciliation Act of 2003. See Pub. L. No. 108-27, S: 401(a)(3),
117 Stat. 752, 764-765 (to be codified at 42 U.S.C. S: 1396d note).

disabled.9 State programs are also required to cover certain services,
including physician and hospital services and nursing home care. As long
as states meet federal requirements and obtain HHS approval for their
state Medicaid plans, they have considerable flexibility in designing and
operating their programs. For example, states may choose to expand
coverage to seniors whose incomes are above statutory limits, and all
states have opted to provide prescription drug coverage. In addition,
section 1115 of SSA permits the Secretary of HHS to waive certain
statutory requirements applicable to Medicaid to allow states to provide
services or cover individuals not otherwise eligible for Medicaid and to
provide federal funding for services and populations not usually eligible
for federal matching payments.10

The Pharmacy Plus initiative allows states to provide a prescription drug
benefit to certain Medicare beneficiaries, specifically seniors and
disabled people, with incomes at or below 200 percent of FPL.11 Typically,
Medicaid eligibility under an approved state plan provides access to all
state Medicaid-covered services, but eligibility under a Pharmacy Plus
demonstration covers only a prescription drug benefit.12 The premise
behind the initiative is that expanded access to medically necessary drugs
will help keep low-income seniors healthy enough to avoid medical expenses
that could cause them to "spend down" their resources to the

9Medicaid eligibility is determined by several factors, including
individual or family income, age, and eligibility for certain other
federal benefits. For example, although state Medicaid programs vary, most
people who are age 65 or older, or are disabled, qualify automatically for
Medicaid if their incomes and assets qualify them for cash assistance
under the federal Supplemental Security Income program.

10In fiscal year 2001, more than 20 percent of total federal Medicaid
spending was governed by the terms and conditions of section 1115 waivers
rather than by standard Medicaid program requirements. See Jeanne Lambrew,
Section 1115 Waivers in Medicaid and the State Children's Health Insurance
Program: An Overview (Washington, D.C.: Kaiser Commission on Medicaid and
the Uninsured, 2001).

11In 2003, the annual income that represented 200 percent of FPL was
$17,960 for an individual, $24,240 for a family of two.

12States are also required to ensure that beneficiaries receiving only the
prescription drug benefit have access to primary care services to assist
with medical management related to pharmacy products, although they are
not required to pay for these services. Most Pharmacy Plus beneficiaries
are seniors covered by Medicare, which generally covers hospital and
physician services.

point of Medicaid eligibility.13 The initiative assumes that budget
neutrality for pharmacy-only coverage can be achieved by savings to
Medicaid from fewer seniors' enrolling for full benefits, as well as from
improved access to prescription drugs, improved service delivery or
medication management, and better management of drug benefit costs.

Unlike some other section 1115 demonstration waivers, the Pharmacy Plus
initiative requires a participating state to accept a fixed spending limit
as part of its budget neutrality agreement with HHS. This spending limit-
sometimes called an aggregate spending limit or global budget cap- applies
not only to services and beneficiaries in the state's demonstration drug
program, but also to all services for all Medicaid seniors in the state.
The Pharmacy Plus budget neutrality approach limits the amount the federal
government will match for a demonstration according to expected growth in
both service costs and enrollment (see app. I). Once a state has reached
its Pharmacy Plus spending limit, it cannot receive additional federal
matching dollars for any Medicaid services for seniors in the state, nor
can the state restrict enrollment of seniors who qualify for full Medicaid
benefits.14 Under the Pharmacy Plus scenario, a state accepts the
financial risks inherent in a fixed budget cap for unanticipated changes
in both cost and enrollment growth. For some other section 1115
demonstrations, budget neutrality is based on a projected per capita cost
for each demonstration beneficiary. This other scenario sets a limit on
spending per person, but because federal matching funds are available for
all people who enroll, a state does not have to accept financial risk for
unexpected growth in enrollment.

13To be eligible for Medicaid, seniors must meet income and asset limits
specified by each state, within federal requirements. Some individuals
with income and assets too high to be immediately eligible may qualify for
coverage on a monthly basis if they incur such high medical costs that
they "spend down" to a qualifying income and asset level. Such spending
down often happens when people enter nursing homes and quickly deplete
their resources.

14According to the terms of approval for Pharmacy Plus demonstrations, to
maintain budget neutrality, states may not alter eligibility or benefits
for seniors who qualify for full Medicaid. States may pursue a variety of
cost-control measures for the drug-only expansion, including limiting
enrollment and increasing cost sharing for Pharmacy Plus beneficiaries.

HHS Has Reviewed 15 Pharmacy Plus Proposals and Approved 4

As of May 2004, HHS had approved four states' Pharmacy Plus demonstration
proposals, denied two, and considered proposals from nine other states.
All four approved demonstrations-Florida, Illinois, South Carolina, and
Wisconsin-are to operate for 5 years, during which time they might enroll
a total of half a million low-income individuals age 65 or older for the
new prescription drug coverage. HHS denied two demonstration proposals,
from Delaware and Hawaii, because they were not consistent with Pharmacy
Plus guidelines. Of the remaining nine proposals, one was withdrawn by the
state and others have been on hold since fall 2003, when Congress was
considering Medicare prescription drug legislation. At the time we
completed our work, legislation providing a new drug benefit through
Medicare had been enacted, but HHS had not determined how the new drug
program would affect the Pharmacy Plus initiative.

HHS Approved Four Pharmacy Plus Demonstrations

HHS has approved Pharmacy Plus demonstrations for low-income seniors in
four states: Florida, Illinois, South Carolina, and Wisconsin.15 As of May
2004, all four demonstrations had been implemented and under way for at
least 17 months: Illinois', Florida's, and Wisconsin's demonstrations were
implemented in 2002, South Carolina's in 2003 (see table 1). Together, the
four approved demonstrations are projected to enroll as many as 527,800
individuals for Medicaid prescription drug benefits only; as of April
2004, they reported combined enrollment of nearly 372,200 people.
Illinois' demonstration is the largest, with expected enrollment for the
drug benefit of more than 250,000 seniors over 5 years. As of April 2004,
more than 192,600 people were enrolled in Illinois' demonstration, the
majority of them moved into the Medicaid program from an existing
state-funded pharmacy assistance program.16

15HHS approved a fifth demonstration proposal, from Indiana, in April
2003, but the state declined to accept the terms of HHS's approval,
primarily because of the state's concerns about the spending limit for all
Medicaid seniors. Instead, Indiana proposed a different budget approach
that would not require the state to project and commit to a spending limit
covering all Medicaid seniors. Indiana's alternative proposal, submitted
May 2003, was one of the nine proposals under consideration by HHS as of
March 2004.

16Illinois' state-funded program covered seniors and people with
disabilities with incomes up to 250 percent of FPL. Participants who were
65 or older with incomes at or below 200 percent of FPL were automatically
enrolled in the demonstration on June 1, 2002. Eligible seniors with
incomes from 201 to 250 percent of FPL and people with disabilities
continue to be covered by the state-funded program.

  Table 1: Highlights of Pharmacy Plus Demonstrations Approved as of May 2004

State and demonstration status
(in order of approval) Description

Illinois

Approved January 2002
Implemented June 2002
Request for amendment submitted
March 2003

Projected enrollment: As many as 256,500 seniors with incomes at or below
200 percent of FPL; 192,600 participants enrolled as of April 2004.
Seniors with incomes at or below 200 percent of FPL who were participating
in Illinois' previous state-funded program were automatically enrolled in
the demonstration.

Spending limit for all Medicaid seniors: $14.0 billion in federal and
state Medicaid funding over 5 years; prescription drug benefit represents
$1.4 billion.

Benefits and cost sharing: Covers most prescription and some
over-the-counter drugs recommended by physicians. Seniors with private
insurance may choose to enroll for a $25 monthly rebate program. No cost
sharing for participants with incomes below 100 percent of FPL;
participants at or above 100 percent of FPL pay $1 for generic and $4 for
brand-name drugs. All participants pay 20 percent coinsurance after their
annual drug benefits under the program exceed $1,750.

State program: Before the demonstration, Illinois' state-funded pharmacy
assistance program enrolled about 170,000 participants. The state
continues to operate a state	funded program covering prescription drugs
for specified conditions for seniors with incomes from 201 to 250 percent
of FPL and for people with disabilities and incomes up to 250 percent of
FPL. Amendment sought to expand coverage to eligible seniors with incomes
at or below 250 percent of FPL; amendment was pending as of March 2004.

Florida

Approved July 2002
Implemented August 2002
Request for amendment submitted
September 2003

Projected enrollment: As many as 58,500 seniors with incomes from 88 to
120 percent of FPL; 54,400 participants enrolled as of April 2004.

Spending limit for all Medicaid seniors: $16.7 billion in federal and
state Medicaid
funding over 5 years; prescription drug benefit represents $477 million.

Benefits and cost sharing: Covers all prescription drugs up to a monthly
benefit limit of
$160 per participant. Except for some classes of drugs, such as mental
health and HIV
antiviral therapies, brand-name drugs are limited to four per month,
although physicians
are allowed to request exceptions. Participants pay $2 for generic drugs,
$5 for drugs on
the state's preferred drug list, and $15 for other brand-name drugs.

State program: Before the demonstration, Florida's state-funded pharmacy
assistance
program enrolled about 9,000 participants. Replaced by the demonstration,
the state	
funded program had the same income eligibility criteria as the
demonstration but a lower
benefit limit: $80 per participant per month. Amendment sought to expand
eligibility to
seniors with incomes at or below
200 percent of FPL and to add a prescription drug discount program;
amendment was
pending as of March 2004.

State and demonstration status
(in order of approval) Description

Wisconsin

Approved July 2002 Implemented September 2002

Projected enrollment: As many as 146,800 seniors with incomes at or below
200 percent of FPL; 70,300 participants enrolled as of April 2004.

Spending limit for all Medicaid seniors: $8.4 billion in federal and state
Medicaid
funding over 5 years; prescription drug benefit represents $919 million.

Benefits and cost sharing: Covers prescription drugs, including insulin.
Participants
pay an annual enrollment fee of $30 and those with incomes above 160
percent of FPL
pay an annual deductible of $500. Participants pay $5 for generic and $15
for brand	
name drugs (those with incomes from 160 to
200 percent of FPL begin co-payments after meeting the required
deductible).

State program: Wisconsin did not have a state-funded pharmacy assistance
program
for seniors before this demonstration. When the state implemented the
demonstration, it
also began offering state-funded pharmacy benefits to seniors with incomes
from 201 to
240 percent of FPL.

South Carolina

Approved July 2002 Implemented January 2003 Projected enrollment: As many
as 66,000 seniors with incomes up to 200 percent of FPL; 54,900 enrolled
as of April 2004.

Spending limit for all Medicaid seniors: $5.0 billion in federal and state
Medicaid funding over 5 years; prescription drug benefit represents $764.7
million.

Benefits and cost sharing: Covers generic and, when no generic is
available, brand	name prescription drugs, over-the-counter drugs
prescribed by physicians, insulin and other self-injected drugs, and
syringes. Except for medications for specified conditions, including
behavioral health disorders, cardiac disease, cancer, HIV/AIDS, and
terminal or life-threatening diseases, coverage limited to four
prescriptions or refills per month. Participants pay an annual deductible
of $500, plus $10 for generic drugs; $15 for brand	name drugs; and $21 for
drugs requiring prior authorization.

State program: Before the demonstration, South Carolina's state-funded
pharmacy assistance program enrolled about 41,000 participants.
Demonstration replaces that program and expands eligibility to seniors
with incomes from 175 up to 200 percent of FPL.

Source: GAO analysis of state and HHS documents.

All the demonstrations except Florida's are approved to enroll seniors
with incomes at or below 200 percent of FPL, the maximum eligible income
established in HHS's Pharmacy Plus guidance.17 As approved, Florida's
demonstration covers seniors with incomes from 88 to 120 percent of FPL,
but in September 2003, the state submitted an amendment to expand income
eligibility to 200 percent of FPL. Illinois also applied in March 2003 to
amend its approved demonstration to expand eligibility, in its case to
include seniors with incomes at or below 250 percent of FPL. The terms of
Illinois' demonstration approval

17Although HHS has identified the initiative's target population as
seniors and people with disabilities who have incomes at or below 200
percent of FPL, none of the states with approved pharmacy demonstrations
have chosen to include people with disabilities.

specifically permit the state to seek this amendment, as long as the state
submits data supporting its ability to cover this expansion population at
no additional cost to the federal government. As of March 2004, HHS was
reviewing both amendments.

Projected 5-year costs vary among the four approved demonstrations. For
Florida, Illinois, South Carolina, and Wisconsin, total combined federal
and state Medicaid spending on the new drug benefit alone is expected to
be more than $3.6 billion over 5 years, of which the federal share would
be approximately $2.1 billion. The combined federal and state Medicaid
spending limits for the four demonstrations-for services to all Medicaid
seniors in the four states-would total $44 billion over 5 years, with a
federal share of at least $25 billion. The estimated 5-year costs solely
for the drug benefit range from $477 million in Florida to $1.4 billion in
Illinois, and combined 5-year federal and state spending limits (based on
projected costs for services to all Medicaid seniors) range from $5.0
billion in South Carolina to $16.7 billion in Florida.

When they applied, three of the four states with approved demonstrations
already operated state-funded pharmacy assistance programs for seniors.
Most beneficiaries eligible for these programs are also eligible for
Pharmacy Plus coverage. HHS allows the states to subsume all or a portion
of an existing program under a demonstration, as long as the states'
demonstrations propose to expand either the number of beneficiaries or the
scope of drug coverage.18 In other words, the state may not simply secure
federal matching dollars for the costs of an existing state-funded drug
program with no expansion. To meet this condition, states with approved
demonstrations either raised income eligibility thresholds or expanded the
scope of drug coverage beyond that of their existing state programs. For
example, Florida doubled its maximum monthly benefit from $80 to $160 per
person, and South Carolina expanded eligibility to include seniors with
incomes from 175 through 200 percent of FPL. Illinois' demonstration
offered a more comprehensive

18The Pharmacy Plus application form and standard terms and conditions for
three of the four approved demonstrations do not include a specific
maintenance-of-effort requirement. Only Illinois, the first Pharmacy Plus
demonstration approved, has such a maintenance-ofeffort requirement. HHS
officials told us that the department's policy regarding state maintenance
of effort has been evolving. Since the Illinois approval, the officials
told us that they have asked states to demonstrate that they are expanding
their programs.

drug benefit than its state-funded program did.19 Wisconsin did not
previously have a state-funded pharmacy assistance program for seniors.20

HHS Denied Two Demonstration Proposals

In 2003, HHS denied Pharmacy Plus demonstration proposals from two states,
Delaware and Hawaii. (See app. II for descriptions of denied, withdrawn,
and pending proposals.) Delaware's proposal was denied primarily because
HHS required that the state expand beyond the existing state-funded
program and limit coverage to seniors with incomes at or below 200 percent
of FPL. Delaware's state-funded pharmacy assistance program already
covered seniors and disabled adults with incomes up to 200 percent of FPL
or whose prescription drug costs exceeded 40 percent of their annual
incomes. For this reason, the state could not expand either eligibility or
coverage and stay within Pharmacy Plus guidelines. Although Delaware
proposed adding a pharmacy benefit management component to monitor
appropriate prescription use and to control costs, HHS found this proposed
change to the existing program insufficient.21

Hawaii proposed to make prescription drugs available at the discounted
Medicaid rate to state residents of all ages with family incomes at or
below 300 percent of FPL. This benefit was to be funded through
participant cost sharing, manufacturer rebates, and a fixed state
contribution of $1 per prescription. HHS's denial was based primarily on
the request to cover individuals with incomes up to 300 percent instead of
200 percent of FPL. Other reasons for the denial included the proposed
coverage for all state residents, instead of targeting seniors and people
with disabilities, and the

19Illinois' state-funded program began in 1985 to cover prescriptions for
cardiovascular disease and has expanded over the years to include drugs
for several conditions, including arthritis, diabetes, Alzheimer's
disease, cancer, glaucoma, lung disease, and Parkinson's disease.

20Wisconsin began offering state-funded drug benefits to seniors with
incomes from 200 up to and including 240 percent of FPL at the same time
that it implemented its demonstration.

21A Delaware Medicaid official expressed concern that HHS's denial was
inconsistent with its guidelines, which, as set forth in the application
form, indicate that states may be allowed to provide enhanced pharmacy
benefit management services as one option for expanding state-funded
programs under the demonstration. HHS officials told us that the denial of
Delaware's proposal would not be reconsidered because the proposed
expansion was not large enough, and approving it would simply make the
state-funded program eligible for federal matching payments.

minimal state financial participation of $1 per prescription in the first
year of the demonstration.22

HHS Has Considered Nine Other Demonstration Proposals

From January 2002 through May 2004, HHS considered Pharmacy Plus
demonstration proposals from nine other states: Arkansas, Connecticut,
Indiana, Maine, Massachusetts, Michigan, New Jersey, North Carolina, and
Rhode Island. As of May 2004, eight were still pending; one proposal, from
Massachusetts, had been withdrawn. Most proposals would cover seniors with
incomes at or below 200 percent of FPL; several would also cover adults
with disabilities. The drug benefits would generally be comprehensive and
require participant cost sharing, which in some cases would include an
annual enrollment fee and 20 percent co-payment for each prescription. All
but one of the states with pending proposals have state-funded pharmacy
assistance programs that they propose to include in whole or in part in
their demonstrations. (App. II describes these demonstration proposals.)

As of May 2004, most of the pending proposals were not under active review
by HHS primarily because the department had not determined the effect of
the Medicare prescription drug legislation on the Pharmacy Plus
demonstration proposals. HHS officials told us in October 2003 that
Arkansas, Rhode Island, and Indiana officials had asked that review of
their states' proposals be put on hold until after Congress had completed
consideration of the Medicare legislation. At that time HHS was still
reviewing a proposal from North Carolina but regarded proposals from four
other states as inactive because longtime negotiations with those states
had reached an impasse. Connecticut and New Jersey, for example, already
had broad state-funded drug coverage for seniors with incomes up to 200
percent of FPL. In such cases, HHS has been unwilling to approve federal
financing for existing state-funded programs.

Medicare Prescription Drug Legislation May Affect Pharmacy Plus Initiative

The Medicare Prescription Drug, Improvement, and Modernization Act of 2003
(MMA) will provide seniors access to a Medicare-covered prescription drug
benefit and will likely affect how HHS and the states manage the Medicaid
Pharmacy Plus initiative. This law gives Medicare

22A Hawaii Medicaid official reported that the agency was unable to
fundamentally change the nature of the proposal during HHS review because
the proposal had been outlined in state law.

beneficiaries the opportunity to enroll for prescription drug coverage to
begin on January 1, 2006, and, as an interim measure, the opportunity to
enroll for Medicare-endorsed drug discount cards beginning in June 2004.
It also directs HHS to establish effective coordination between Medicare
plans and state Medicaid and pharmacy assistance programs and to establish
a commission to address these and other transition issues.23 In 2006, the
Medicare drug benefit will replace Medicaid as the primary source of
prescription drug coverage for low-income seniors who would have been
eligible for both full benefits under Medicaid and drug benefits under
Medicare plans.24 Under MMA, individuals with limited assets and incomes
below 150 percent of FPL will be eligible for federal subsidies to assist
with the drug benefit's cost-sharing requirements.25 But because Pharmacy
Plus demonstrations in Illinois, South Carolina, and Wisconsin cover
individuals with incomes above 150 percent and at or below 200 percent of
FPL regardless of other assets, some current demonstration beneficiaries
may not qualify for these subsidies. Pharmacy Plus beneficiaries are
likewise ineligible for the Medicare drug discount cards.26

23Specifically, the law establishes a State Pharmaceutical Assistance
Transition Commission to develop a proposal for addressing the
transitional issues facing statefunded pharmacy assistance programs and
their participants because of implementation of the new Medicare
prescription drug benefit. Members of the commission are to include
representatives of states operating pharmacy benefit programs, interested
organizations, private insurers, and others. The commission is required to
submit a detailed proposal, including specific legislative or
administrative recommendations, to Congress and the President by January
1, 2005. See Pub. L. No. 108-173, S: 101(a), 117 Stat. 2071, 2128-31
(adding sections 1860D-23 and 1860D-24 to SSA) (to be codified at 42
U.S.C. S:S: 1395w-133 and 1395w-134); see also section 106, 117 Stat.
2168-9.

24See Section 103(c), 117 Stat. 2158 (amending section 1935 of SSA) (to be
codified at 42 U.S.C. S: 1396n-5).

25For example, individuals with incomes below 135 percent of FPL generally
will be entitled to a full premium subsidy; individuals with incomes up to
150 percent of FPL will be entitled to an income-related premium subsidy
determined on a sliding scale. In addition, after reaching an
out-of-pocket threshold of $3,600, individuals with incomes below 135
percent of FPL will have no co-payment for covered drugs, while those with
incomes below 150 percent of FPL will pay either $2 or $5 for each drug.
See Section 101(a), 117 Stat. 2107-9 (adding sections 1860D-14(a)(1) and
(2) to SSA) (to be codified at 42 U.S.C. S:S: 1395w-114(a)(1) and (2));
see also section 101(a), 117 Stat. 2077-9 (adding section 1860D2(b)(4) to
SSA) (to be codified at 42 U.S.C. S: 1395w-102(b)(4)).

26See Section 101(a), 117 Stat. 2132-3 (adding section 1860D-31(b) to SSA)
(to be codified at 42 U.S.C. S: 1395w-141(b)).

HHS Has Not Ensured That Approved Demonstrations' Spending Limits Will Be
Budget Neutral

As of May 2004, HHS indicated it was considering how enactment of the new
law would affect Pharmacy Plus demonstrations and proposals.27 Officials
from the four states with approved demonstrations told us in December 2003
that they were uncertain how the law would affect their demonstrations,
but they had no plans to end the demonstrations early. After the Medicare
prescription drug benefit begins in 2006, some demonstrations could be
discontinued or modified. Early termination could have an impact on the
demonstrations' budget neutrality, which often depends on savings in later
years to offset higher start-up costs. Officials in Illinois and Florida
indicated in December 2003 that their pharmacy demonstrations might be
converted to state-funded programs in 2006.

HHS has not adequately ensured that the spending limits it has approved
for Pharmacy Plus demonstrations will be budget neutral-in other words,
that the federal government will spend no more under the demonstrations
than without them. For all four demonstrations, HHS approved 5-year
spending limits based on projections of cost and beneficiary enrollment
growth that exceeded benchmarks that department officials told us they
considered in assessing the reasonableness of states' demonstration
proposals. These cost and enrollment growth benchmarks incorporate states'
historical experience and expectations for Medicaid program growth
nationwide. The discrepancies between the growth benchmarks and the
approved growth rates were greatest for Illinois and Wisconsin. Neither
HHS's negotiations with the states nor the department's rationale for
approving higher-than-benchmark growth rates is well documented. Had HHS
based the 5-year demonstration spending limits on the benchmark growth
rates, the federal share of approved spending would be considerably lower,
particularly for Illinois and Wisconsin: specifically,

27In commenting on a draft of this report, HHS said that it expected less
need for Pharmacy Plus demonstrations when Medicare coverage for
prescription drugs is expanded under MMA and that states would need to
decide if they want to continue their demonstrations.

$1 billion lower in Illinois and $416 million lower in Wisconsin.28 For
Florida and South Carolina, the federal share of approved spending would
have been $55 million and $42 million lower, respectively.

HHS Approved Projected Growth Rates Exceeding Benchmarks

HHS based the Pharmacy Plus demonstration spending limits it approved on a
range of estimated future growth rates for cost per beneficiary and for
enrollment, which in some cases exceeded benchmarks29 the department told
us it considered in assessing the reasonableness of states' proposals. A
standard Pharmacy Plus application form developed by HHS and a technical
guidance document are the chief sources of criteria and formal guidance to
states for developing demonstration proposals. But HHS has not established
written criteria for how it reviews and approves the growth rates that
states propose. These growth rates are key elements in the budget
neutrality negotiations between states and the federal government because
higher rates result in more generous spending limits, which represent the
federal government's agreed-on maximum spending for all the states'
Medicaid seniors during the demonstrations. An inappropriately high
spending level can represent a higher federal liability than warranted.

The process used by HHS and the states to determine whether states'
proposed Pharmacy Plus demonstrations will be budget neutral requires
comparing two cost estimates: (1) projected 5-year costs of a state's
existing Medicaid program for seniors ("without-demonstration costs") and
(2) projected 5-year costs of the state's existing program plus the drug
benefits and beneficiaries added by the demonstration ("withdemonstration
costs"). These calculations factor in projected growth in

28We have reported that by allowing Illinois to include impermissible
costs in its projected spending under the Medicaid program without the
demonstration, HHS approved a Pharmacy Plus demonstration in the state
that was not budget neutral, inappropriately raising allowed costs for the
project by $275 million. We recommended to the Secretary of HHS that the
agency ensure that valid methods are used to demonstrate budget neutrality
and that it apply such methods to adjust the federal spending commitment
under Illinois' demonstration. See U.S. General Accounting Office,
Medicaid and SCHIP: Recent HHS Approvals of Demonstration Waiver Projects
Raise Concerns, GAO-02-817 (Washington, D.C.: July 12, 2002). As of March
2004, HHS had not adjusted the state's spending limit.

29We use the term "benchmark" throughout this report to describe the cost
and Medicaid enrollment growth rates HHS considered when making its
approval decisions for Pharmacy Plus demonstrations. We use "benchmark"
rather than "criterion" because, in contrast to its practice in approving
some other section 1115 demonstrations, HHS did not have written cost or
enrollment growth criteria for Pharmacy Plus.

costs and enrollment each year. As long as projected with-demonstration
costs do not exceed projected without-demonstration costs, the
demonstration can be approved as budget neutral. As a result, the
projected costs of a state's existing, without-waiver Medicaid program for
seniors effectively sets the spending limit for all services provided to
all Medicaid seniors in the state for the 5-year demonstration term.
Appendix I outlines the basic steps HHS follows in setting Pharmacy Plus
demonstration spending limits.

To determine budget neutral spending limits for the pharmacy
demonstrations, HHS officials told us they consider the following for
estimating growth in costs and enrollment through the course of the
demonstrations:

o  	For cost growth per beneficiary, similar to guidelines for other types
of section 1115 demonstrations,30 HHS seeks to approve a growth rate equal

to the lower of either the state's historical average annual growth in
perbeneficiary cost (that is, the average annual rate for the 5 years
before the demonstration proposal) or the nationwide projected growth
rate, developed by CMS's Office of the Actuary, for Medicaid cost per
beneficiary age 65 or older.31

o  	For enrollment growth, HHS considers the state's historical average
annual growth in enrollment as a starting point and, to a lesser extent,
the CMS Actuary's nationwide rate, but it allows states to present a
rationale for a higher rate that anticipates rising future enrollments.

HHS's approved growth rates in some cases exceeded these benchmarks (see
table 2). For per-beneficiary cost growth rates in Florida, Illinois, and
Wisconsin, HHS did not approve the lower of either the state's historical

30The Health Insurance Flexibility and Accountability (HIFA) initiative,
for example, another Medicaid section 1115 demonstration waiver
initiative, specifies a similar benchmark for cost growth per beneficiary
in its standard application form. The HIFA initiative provides two options
for projecting cost growth per beneficiary: the Medical Care Consumer
Price Index, developed by the Bureau of Labor Statistics, or a
state-specific projection of what the program would have cost without the
demonstration. For the second option, HHS policy is to apply the lower of
state-specific experience-using 5 years of Medicaid data-or the CMS
Actuary's Medicaid baseline for the eligible groups covered by the
demonstration.

31The CMS Actuary's growth projections in effect during budget discussions
for the four approved Pharmacy Plus demonstrations were a 6.3 percent
annual increase in cost per beneficiary and a 1.8 percent annual increase
in enrollment.

average rate or the CMS Actuary's rate of 6.3 percent. Similarly, for
beneficiary enrollment growth rates, HHS approved rates for Illinois and
Wisconsin that exceeded both the states' historical experience and the CMS
Actuary's 1.8 percent projected annual growth rate. In Illinois' case, the
approved rate for beneficiary enrollment growth-5 percent per year over
the 5-year demonstration-was considerably higher than the state's 5-year
historical average enrollment growth of 1.6 percent per year.

Table 2: HHS-Approved, State Historical, and CMS Actuary Growth Rates

 Annual growth rates in percent Cost growth rates Enrollment growth rates State

HHSapproved

                                                     State historical average

CMS Actuary

HHSapproved

                                                     State historical average

CMS Actuary

                        Florida   6.5       6.5      6.3      1.4      1.4    
                     Illinois a   5.5       4.5      6.3      5.0      1.6    
               South Carolina b   6.3      10.0      6.3      1.0      0.7    
                      Wisconsin   6.3       5.4      6.3      2.0       0.01c 

Source: GAO analysis of HHS and state documents.

aThe state historical average rates for Illinois (4.5 percent cost growth
and 1.6 percent enrollment growth) reflect updated information provided by
the state to HHS shortly before its approval decision on January 28, 2002.
This information does not appear in the demonstration application that was
submitted on July 31, 2001.

bFor South Carolina, historical average rates reflect 3 years of data,
rather than the generally required 5 years, because cost and enrollment
data for Medicaid seniors were incomplete for 2 of the 5 years (data for
certain long-term care waivers for seniors were not included).

cThis figure is lower than the 0.12 percent enrollment growth rate
Wisconsin proposed in its demonstration application. We included 5 years
of data in calculating our average rate for the state, while the state
itself averaged 3 years of data, excluding 2 years when enrollment
declined slightly.

Basis for Approved Spending Limits Not Clear or Well Documented

HHS's basis is unclear for approving growth rates higher than the
benchmarks in some cases, particularly for approving higher enrollment
growth rates for Illinois and Wisconsin. The department's negotiation
process with these two states, during which officials reached agreement on
allowed growth rates, was not documented, nor was its rationale for
approving rates that differed from the lower of state historical
experience or the CMS Actuary's projections. In particular, HHS's internal
decision memorandums-which described the factors that HHS, CMS, OMB, and
others considered in reviewing the demonstrations and which are not
publicly available-did not provide the rationale for the approved spending
limits, and neither did the publicly available demonstration approval
letters.

HHS and state officials told us that Illinois and Wisconsin used a variety
of arguments to convince the department that their situations warranted
higher enrollment growth rates. But the states provided little specific
documentation to HHS or to us to support these arguments. For example:

o  	Illinois asserted that its projected annual enrollment growth rate for
the demonstration years from 2002 through 2007 should be significantly
higher than its 5-year average historical growth rate of 1.6 percent,
because income eligibility levels for seniors in its Medicaid program
increased from 41 to 100 percent of FPL from July 2000 through July 2002.
As support, the state provided HHS with updated Medicaid enrollment
data-which were more recent than those included in the original
demonstration application and showed increased growth rates for seniors
compared with earlier years-but these rates were still lower than the 5
percent HHS approved and did not raise the historical average to 5
percent.32 The state did not provide documents with actuarial projections
of the estimated number of people expected to enroll in Medicaid because
of the change in eligibility criteria. Illinois justified applying the 5
percent annual growth rate to all 5 years of the pharmacy demonstration by
providing a chart showing that enrollment in a different state program,
SCHIP, had grown more than 5 percent per year on average for 3 years after
that program's eligibility criteria were expanded. In our view, however,
Illinois' SCHIP enrollment experience with children does not provide a
reasonable basis for predicting enrollment by seniors in the Pharmacy Plus
demonstration.

o  	Wisconsin asserted that its projected annual enrollment growth rate
for the demonstration years should be significantly higher than either its
5-year unadjusted historical growth rate of 0.01 percent or the 0.12
percent rate based on 3 years of historical data reported in its
application because of the anticipated effects of a nationwide Social
Security Administration mail outreach program to low-income Medicare
beneficiaries. This outreach program informed seniors enrolled in Medicare
about other benefits, including Medicaid assistance for Medicare
cost-sharing

32Illinois provided HHS with updated information on senior enrollment in
the state's Medicaid program, including growth of 1.3 percent between
state fiscal years 1999 and 2000 and 4.2 percent between state fiscal
years 2000 and 2001. These updated figures are incorporated in the 5-year
historical average enrollment growth rate of 1.6 percent, covering state
fiscal years 1997 through 2001.

requirements, for which they might qualify.33 Wisconsin officials told us
they proposed a 4 percent future annual enrollment growth rate for seniors
in the expectation that this outreach program, along with factors
including an aging population and the economic downturn, would increase
Medicaid enrollment. According to HHS, Wisconsin did not document any
projections of how many newly eligible Medicaid individuals could be

prompted to enroll after the Social Security Administration outreach
mailing. Instead, it submitted information based on a review of a similar
outreach effort in Minnesota. According to Wisconsin state officials,
during negotiations HHS proposed 1 percent as a more reasonable growth
rate, and HHS and state officials agreed to an approved enrollment growth
rate of 2 percent per year. Our related work suggests that Wisconsin may
be justified in claiming some increase in Medicaid enrollment as a result
of the outreach program, but the effect appears to be less than 1 percent.
Notably, although the Social Security Administration mail outreach program
was nationwide, HHS did not consider its effects when approving enrollment
rates for other states.34

Demonstration Spending Limits Would Be Lower if Based on Benchmark Rates

Application of benchmark rates for projected per-beneficiary cost and
enrollment growth would have produced lower spending limits for all four
approved Pharmacy Plus demonstrations (see table 3). Benchmark-based
limits on combined federal and state spending would be approximately $3
billion lower over 5 years than what HHS approved for the four
demonstrations, and the federal share alone would come to about $1.6
billion less. The higher-than-benchmark growth rates HHS approved for
Illinois and Wisconsin accounted for most of these differences. Had the

33The Medicare, Medicaid, and SCHIP Benefits Improvement and Protection
Act of 2000 requires the Social Security Administration to notify Medicare
beneficiaries nationwide about other benefits, including Medicaid, for
which they could potentially be eligible. See Pub. L. No. 106-554, App. F,
S: 911(a), 114 Stat. 2763, 2763A-583 (adding section 1144 to SSA) (to be
codified at 42 U.S.C. S: 1320b-14). See U.S. General Accounting Office,
Medicare Savings Programs: Results of Social Security Administration's
2002 Outreach to Low-Income Beneficiaries, GAO-04-363 (Washington, D.C.:
Mar. 26, 2004).

34See GAO-04-363. In 2002, the Social Security Administration outreach
program sent letters to about 16.4 million low-income Medicare
beneficiaries nationwide, yielding an estimated 74,000 additional
individuals-a further increase in enrollment of 0.5 percent-enrolling in
Medicaid programs that help low-income Medicare beneficiaries pay their
Medicare premiums and, in some cases, their deductibles and coinsurance.
On the basis of a sample, we estimated that Wisconsin experienced a
statistically significant increase in additional Medicaid enrollment after
the Social Security Administration mailing, although at 0.4 percent, the
response was slightly less than the U.S. average. The Social Security
Administration outreach program continued on a smaller scale in 2003 and
thereafter.

spending limit for Illinois' demonstration, in particular, been based
strictly on the benchmark rates, combined federal and state spending would
have been almost $2.2 billion, or 15 percent, lower, and the federal
government's liability under the demonstration (at the state's 50 percent
federal matching rate) lower by more than $1 billion. The difference is
less pronounced for Wisconsin, where the approved federal and state
spending limit exceeds what it would have been had benchmark rates been
applied by about $713 million, translating into about $416 million in
additional federal spending.

Table 3: HHS-Approved and Benchmark 5-Year Spending Limits

                              Dollars in millions

Dollar difference (percentage Federal share of States HHS-approved
Benchmark difference)a differenceb

                     Florida   $16,669    $16,575      $94 (0.6)          $55 
                    Illinois      14,047   11,880    2,167 (15.4)       1,083 
              South Carolina       4,962   4,902       60 (1.2)     
                   Wisconsin       8,378   7,666       713 (8.5)          416 
                      Totala   $44,056    $41,023    $3,033 (6.9)      $1,596 

Source: GAO analysis of HHS and state documents.

Notes: Figures reflect total approved federal and state spending limits
for all Medicaid services for seniors over the 5 years of each
demonstration, including spending on new Pharmacy Plus prescription drug
benefits. We calculated benchmark spending limits using, for cost growth
per beneficiary, the lower of either the state's historical average cost
growth rate or the CMS Actuary's projected rate, and for enrollment
growth, the state's unadjusted 5-year historical average enrollment growth
rate, except for South Carolina, where we used the average of 3 years'
data.

aDollar differences and totals are based on numbers before rounding.

bFederal share calculated by applying fiscal year 2003 federal Medicaid
matching rates for each state to the dollar difference for that state:
Florida, 58.8 percent; Illinois, 50 percent; South Carolina, 69.8 percent;
and Wisconsin, 58.4 percent.

The spending limits HHS approved for Illinois and Wisconsin exceed
estimates based on consistent application of the benchmark growth rates by
15.4 percent and 8.5 percent, respectively.35 The limits approved for

35We calculated Wisconsin's benchmark spending limit based on the 0.01
percent average annual enrollment growth rate for 5 years of historical
data, which included 2 years when enrollment declined slightly. Had the
0.12 percent average annual enrollment growth rate been used, based on 3
years of historical data as the state proposed in its demonstration
application, Wisconsin's combined federal and state spending limit would
have been $7.9 billion, about 5.7 percent, or $477 million, below what HHS
approved.

Florida and South Carolina, while not budget neutral compared with the
benchmark spending estimates, reflect relatively small differences.
Florida's approved spending limit exceeds the benchmark estimate by less
than 1 percent-$94 million of a 5-year approved federal and state spending
limit of nearly $16.7 billion-and South Carolina's approved spending limit
exceeds the benchmark by 1.2 percent, or $60 million.

CBO has similarly reported that Pharmacy Plus demonstrations are likely to
increase federal Medicaid spending. Before passage of MMA, CBO estimated
that the Pharmacy Plus demonstrations would add about $18 billion to
federal Medicaid spending over the 10 years from 2004 through 2013.
According to CBO officials, the agency considered a range of scenarios for
how the initiative might grow with new demonstration approvals and
estimated the initiative's overall effect on Medicaid spending.36 The
officials told us that CBO did not include any of the demonstrations'
projected savings in its analysis because it did not find the argument
that savings would occur convincing.

Neither data from state experience nor other research supports the savings
assumptions necessary for budget neutrality in the Pharmacy Plus
demonstrations. In developing their demonstration proposals, states
assumed that keeping low-income seniors healthy-thus preventing them from
spending down their financial resources on health services and "diverting"
them from Medicaid eligibility-would generate savings to help offset the
increased costs of providing a new drug benefit. Without statespecific
evidence, HHS approved savings assumptions negotiated with the states,
including significant projected reductions in Medicaid senior enrollment.
But the limited research available suggests that potential health care
savings due to improved access to prescription drugs are likely to be much
less than the levels the states assumed and HHS approved. Had more
conservative savings assumptions been used to estimate the demonstrations'
costs, the proposals likely could not have been approved as budget
neutral. Moreover, concerns have arisen about what actions states might
take to control spending on behalf of seniors if estimated

Pharmacy Plus Savings Assumptions Not Well Supported

36According to CBO officials, the agency did not analyze the cost impact
of individual Pharmacy Plus demonstrations but estimated potential impact
on Medicaid spending overall under various growth scenarios. CBO estimated
that roughly 1.2 million people would join the Medicaid rolls for the
prescription drug benefit only. See U.S. Congressional Budget Office, An
Analysis of the President's Budgetary Proposals for Fiscal Year 2004: An
Interim Report (Washington, D.C.: March 2003).

savings do not accrue and states reach or exceed their spending limits
under the demonstrations.

HHS-Approved Savings Assumptions Are Not Supported by State Experience

The approved Pharmacy Plus demonstrations count on expected savings based
on reductions in the projected number of seniors who will enroll in
states' Medicaid programs-ranging from a 3 percent reduction in Florida to
a 25 percent reduction in South Carolina over the demonstrations' 5 years.
The dollar amounts of combined federal and state savings projected under
these assumptions in the demonstrations' budget neutrality calculations
range from $480 million in Florida to $2 billion in Illinois (see table
4).

Table 4: States' Projections of Medicaid Senior Populations after 5 Years,
with and without HHS-Approved Pharmacy Plus Demonstrations

                 Projections                Florida  Illinois South Wisconsin 
                                                           Carolina 
      Total projected Medicaid seniors,     210,200 186,683 90,006     70,412 
       without demonstration, by year 5                             
    Total projected Medicaid seniors, with  204,300 145,240 67,946     57,297 
           demonstration, by year 5                                 
     Projected number of seniors diverted     5,900   41,442 22,060    13,115 
           from Medicaid by year 5                                  
     Annual reduction in Medicaid seniors         3             5 5     2.5-5 
                 (percentage)                                       
Cumulative 5-year reduction in Medicaid        3           22 25 
             seniors (percentage)                                   

Projected dollar savings due to diversion of seniors from Medicaid
by year 5 $480 million $2 billion $769 million $926 million

Source: GAO analysis.

Note: Analysis of state and HHS documents and of Jocelyn Guyer, The
Financing of Pharmacy Plus Waivers: Trade-offs between Expanding Rx
Coverage and Global Caps in Medicaid (Washington, D.C.: Kaiser Commission
on Medicaid and the Uninsured, 2003).

To project the extent to which Pharmacy Plus would reduce its new
enrollment of Medicaid seniors, and thus its total senior enrollment,
Florida made the relatively conservative assumption that the drug benefit
would enable seniors to avoid Medicaid eligibility for 1 year; after 5
years, the state's total projected number of Medicaid seniors would be
5,900 (3 percent) lower with the demonstration than without it. The other
states, in contrast, assumed that everyone diverted in each year of their
demonstrations would remain out of Medicaid throughout the full
demonstration period and would not, for example, enter a nursing home,
which often results in Medicaid eligibility. As a result, Illinois, South
Carolina, and Wisconsin projected reductions of nearly 20 percent or more
in Medicaid senior enrollments at the end of 5 years. Had these states
made more conservative assumptions-assuming, for example, as Florida chose
to, that providing access to prescription drugs would delay seniors'

entry into Medicaid by only 1 year instead of 5-their projected
withdemonstration costs would have exceeded projected withoutdemonstration
costs and would not have been budget neutral.

Although states' demonstration proposals aim to achieve savings by
expanding seniors' access to prescription drugs and improving their
health, in practice it appears that some states' estimates of expected
savings may have been derived in part by determining how much in savings
was needed to demonstrate budget neutrality. In their proposals, none of
the three states that previously had state-funded pharmacy assistance
programs (Florida, Illinois, and South Carolina) provided data from those
programs that specifically supported such high projected savings. Based on
conversations with Wisconsin health care financing officials and a review
of documents, we found that the state's demonstration savings estimates
were a residual of the budget-negotiating process, derived from
determining how much was needed in savings to demonstrate budget
neutrality, rather than from research or data about what was realistic.

Premise behind Approved Demonstrations Not Well Supported by Research

The premise that Pharmacy Plus demonstrations will generate savings by
keeping low-income seniors from becoming Medicaid-eligible is not
supported by research. In a previous report, we reviewed the research
studies cited in Illinois' demonstration proposal and found that they did
not sufficiently support the state's theory that a full drug benefit for
lowincome seniors would yield the projected level of savings.37 Although
these studies indicated that access to prescription drugs benefited people
in poor health, they all focused on people who already had specific
diagnosed conditions, such as diabetes, heart disease, or HIV, rather than
on a general population of seniors. An extensive 2003 review of research
examining drug coverage for low-income seniors found relatively few
studies about the effect on Medicaid spending of expanded access to a

37GAO-02-817.

broad prescription drug benefit.38 The one study this review considered
most relevant, conducted in the mid-1980s, assessed Pennsylvania's
statefunded program, Pharmaceutical Assistance Contract for the Elderly
(PACE), and found that despite high enrollment, Medicaid entry among PACE
participants was neither prevented nor delayed enough to have a
discernible effect on the state's overall Medicaid budget.39 Other studies
of broad prescription drug benefits for low-income seniors, including one
of New York's program, found some reductions in participants' health care
costs but mainly for inpatient hospital care, which, for people age 65 or
older, is covered by Medicare rather than Medicaid.40 Still other studies
in this review examined the more limited question of how access to
appropriate drugs affects people already suffering from specific
illnesses. Such research sheds little light on the cost-effectiveness of
offering comprehensive drug benefits to a broad population of low-income
seniors.

Some states that have not submitted Pharmacy Plus proposals examined the
diversion and savings assumptions behind the demonstrations and found that
they would not likely be realized. For example, in considering whether to
apply for a demonstration, Minnesota found a substantial risk that seniors
receiving only a drug benefit would eventually become Medicaid-eligible
over a 5-year follow-up period. In its optimal model, the study estimated
that to generate enough savings to offset the new drug costs, the risk of
Medicaid entry would have to be reduced by 50 percent

38Ellen O'Brien, "Will Prescription Drug Coverage for the Low-income
Elderly Pay for Itself? A Review of the Literature," Georgetown University
Health Policy Institute working paper, prepared for the Kaiser Commission
on Medicaid and the Uninsured, May 2003, cited with permission. See
www.hpi.georgetown.edu/pdfs/obrienrxcostoff.pdf, downloaded March 16,
2004. The author identified sources through a Medline search of the
research literature and a manual search of citations in published papers.
The analysis covered research literature on the effects of prescription
drug coverage on low-income seniors, access limits in Medicaid and other
insurance plans, studies of specific drugs or drug classes, and studies of
the cost-effectiveness of recent prescription drugs.

39Bruce Stuart and Daniel Lago, "Prescription Drug Coverage and Medical
Indigence among the Elderly," Journal of Aging and Health, vol. 1, no. 4
(1989), 452-469.

40In 2002 we reported that in assessing the cost of a Medicare
prescription drug benefit at that time, CBO, OMB, and CMS's Actuary did
not accept the premise that providing a prescription drug benefit to
low-income seniors would pay for itself (GAO-02-817). CBO raised several
reasons for caution, including that greater use of drugs among seniors
could increase the risk of side effects and adverse drug reactions, which
could in turn increase use of hospitals, emergency rooms, and other health
care services. CBO concluded that Medicare beneficiaries without any drug
coverage already consumed a large number of prescription drugs, and
therefore any savings due to increased access would likely be small.

for non-nursing-home enrollees and by 30 percent for those who become
eligible after entering a nursing home. Minnesota Medicaid officials
concluded that this scenario was not realistic and dropped the state's
Pharmacy Plus demonstration proposal.41

Pennsylvania also conducted a Pharmacy Plus demonstration feasibility
study for PACE and the related Pharmaceutical Assistance Contract for the
Elderly Needs Enhancement Tier (PACENET) programs, which together enrolled
about 270,000 seniors in 2002.42 The study found that to offset drug
benefit costs, the programs would need aggressive cost containment,
through such approaches as increased co-payments, reduced provider
reimbursements, and a preferred drug list. In addition, the study noted
that in states with generous drug benefits, savings from expansion to more
seniors are particularly difficult to realize because most beneficiaries
who would have avoided expensive nursing home care have already done so.
As of March 2004, Pennsylvania had not submitted a Pharmacy Plus
demonstration proposal.

Concerns about Effects on States if Savings Do Not Accrue

Although it is early in demonstration implementation, we and others have
raised concerns about how states may be affected if savings under Pharmacy
Plus do not accrue and the states' spending reaches or exceeds HHS's
approved spending limits. We noted in our July 2002 report that the
Illinois Pharmacy Plus demonstration, as approved, makes several risky
assumptions with regard to the extent of the expected savings.43 In such
cases the federal government would not be at financial risk, but the
states would be, because the spending limits cover services for all the
states' Medicaid seniors. Any expenditures for Medicaid seniors beyond the
demonstration's federally matched spending limit would be entirely the
state's responsibility. Officials in Florida and Wisconsin expressed
concerns that their demonstration spending limits, based on fixed rates of
growth projected over 5 years, could not be adjusted to reflect

41The results of this study by JEN Associates of Cambridge, Massachusetts,
were reported in a memorandum to the Minnesota Department of Human
Services, "Feasibility Analysis of PDP Expansion," dated March 7, 2002.

42Pennsylvania's PACE program has operated since 1984 and PACENET has
provided an expanded benefit to additional low-income seniors since 1996.
See Mercer Government Human Services Consulting, Pharmacy Plus 1115 Waiver
Feasibility Study, Commonwealth of Pennsylvania (Phoenix: October 2002).

43See GAO-02-817, 23-24.

States Have Taken Few Steps to Evaluate Demonstrations, and HHS Has Not
Ensured Sufficient or Timely Progress Reporting

unpredictable changes in costs and enrollment growth. One study has raised
concerns about the potential effects on Medicaid seniors, noting that as
state spending approaches the limit of what the federal government will
match, states may feel pressed to reduce optional expansions of
eligibility or optional benefits.44 States could also try to control
spending without reducing eligibility or services by lowering provider
reimbursements-a step already taken in Illinois, although not in response
to pharmacy demonstration enrollment or spending-or by implementing
preferred drug lists.45

As of February 2004, efforts by the states and HHS to evaluate and monitor
the four approved demonstrations, and to address some of the research
questions the Pharmacy Plus initiative raises, were in their early stages.
The four states with approved demonstrations had taken few steps toward
implementing the evaluation plans required as a condition of approval, and
an independent evaluation of two of the demonstrations, contracted by HHS
and started in October 2002, was not scheduled to report until September
2005. In the interim, HHS has not ensured that the states' required
progress reports contain sufficient information for monitoring whether the
demonstrations are functioning as intended or that these reports are
submitted in a timely manner.

States Have Taken Few As a condition of Pharmacy Plus approval, HHS
requires states to design Steps to Implement and carry out an evaluation
and to report their results after the Evaluation Plans demonstration
ends.46 States are required to submit a plan for this

evaluation in their proposals and in the operational protocols that HHS
approves before states begin the demonstrations. Although the four states
with approved Pharmacy Plus demonstrations submitted the required
evaluation plans-containing research hypotheses, possible outcome

44Jocelyn Guyer, The Financing of Pharmacy Plus Waivers: Trade-offs
between Expanding Rx Coverage and Global Caps in Medicaid (Washington,
D.C.: Kaiser Commission on Medicaid and the Uninsured, May 2003).

45According to an Illinois official, the reimbursement rate for all
providers, including pharmacists, was cut by 6 percent in state fiscal
year 2003, and that step helped the pharmacy demonstration operate well
below its targeted first-year spending limit.

46States are required to provide HHS with drafts of their final evaluation
reports within 180 days of the end of the 5-year demonstrations; Pharmacy
Plus terms and conditions do not require interim evaluation reports from
the states. HHS also requires the states to cooperate with federal
evaluators and contractors in the independent evaluation.

measures, and data needs-as of February 2004, they had taken few steps to
put their evaluation plans into practice.

As HHS requires, the four states' initial proposals and operational
protocols included plans for how they would evaluate whether their
demonstrations were working as intended. With some variations, all the
plans proposed to address the overall research question of how providing a
pharmacy benefit to non-Medicaid-covered seniors would affect Medicaid
costs, service use, and future eligibility trends, including whether
savings achieved by diverting individuals from Medicaid eligibility would
offset the benefit's cost. The first demonstration proposal, from
Illinois, initially contained an extensive plan to assess demonstration
outcomes; the plan later changed significantly. The initial plan proposed
that the state collect data from sources such as Medicaid and Medicare
claims systems, surveys of participants or case-study interviews, and
demonstrationspecific claims. In terms of outcome measures, Illinois' plan
proposed comparing seniors who do have the drug benefit with seniors who
do not on such measures as hospitalization rates, health care service
costs, use of emergency room services, and rates and length of nursing
home stays. A later version of Illinois' plan (as described in the state's
operational protocol), however, calls for using existing Medicaid claims
data for only one outcome measure, Medicaid spending for seniors.

Both South Carolina and Wisconsin adopted Illinois' relatively extensive
initial evaluation plan in their demonstration proposals, and as of
February 2004, neither South Carolina nor Wisconsin had changed its
proposed plan. Florida, which did not submit an evaluation plan in its
demonstration proposal, provided a two-paragraph discussion in its
operational protocol. This discussion listed several hypotheses and
indicators to be monitored, noted that data would be collected using the
state's current Medicaid system, and gave no details about how or when the
plan would be implemented.

As of February 2004, the states had taken few steps to implement their
demonstration evaluation plans or to determine how they would collect or
analyze data to support their evaluations. States' evaluation activities
were generally limited to collecting and reporting to HHS data from their
existing Medicaid data systems. Although plans for Illinois, South
Carolina, and Wisconsin call for starting their evaluations at the start
of their demonstrations to draw on data about services used before and

throughout beneficiaries' enrollment, these states and Florida indicated
they were just beginning to collect and report data to implement their
evaluation plans:

o  	Florida and South Carolina officials told us that they had not decided
whether their evaluations would be designed and conducted by the state
Medicaid agency or by an outside entity such as a university. Neither
state had developed an evaluation implementation schedule.

o  	Illinois and Wisconsin reported providing extensive state data for
HHS's independent evaluation of their demonstrations but, at the time of
our review, had not begun their own evaluations. State officials told us
they understood that participating in the independent evaluation would
exempt them from conducting their own evaluations. But HHS officials told
us that state evaluations were still required.

HHS Has Contracted for an Independent Pharmacy Plus Evaluation

HHS has contracted with independent university researchers for an
extensive evaluation of the Pharmacy Plus demonstrations in Illinois and
Wisconsin. The evaluation's goal is to document achievements and
difficulties in implementing a Pharmacy Plus demonstration, as well as to
identify impacts on entry into Medicaid and on costs to Medicare.
According to HHS, the evaluation aims to address whether providing
prescription drug benefits to non-Medicaid seniors will keep individuals
relatively healthy, divert them from full Medicaid eligibility, and thus
lower Medicaid and Medicare costs. To address these issues, the evaluation
contract calls for four components of work, including (1) site visits to
Illinois and Wisconsin to describe the demonstrations and their
implementation; (2) telephone surveys of demonstration beneficiaries in
those states about their health status, access to health care, and prior
drug coverage; (3) analysis of Medicaid, Medicare, and demonstration
claims data to assess patterns of drug use and effects on Medicaid and
Medicare costs; and (4) an analysis of enrollment trends in each state's
Medicaid program to determine if diversion assumptions are met. In
addition, the evaluation aims to compare the experiences of demonstration
beneficiaries with a similar population in another state that does not
offer a prescription drug benefit.47

47The evaluators chose Ohio because of its proximity to both Illinois and
Wisconsin; the characteristics of its senior population; and likely
similarities in the attitudes of that state's policymakers, physicians,
and public about the use of long-term care services.

Final results for all components of this planned 3-year evaluation, which
began in October 2002, are scheduled to be reported to HHS by September
2005.48 Specifically, a final report to HHS on the patterns of drug use is
due in September 2004; final reports on the demonstrations' cost effects
on Medicaid and Medicare are due in September 2005. The evaluation
contract does not indicate when results from the work may be available to
other researchers or the public.

According to the HHS evaluation project officer, the independent
evaluators completed state site visits to Illinois and Wisconsin in July
2003 for the descriptive work component and submitted draft reports to HHS
in December.49 These reports were in review as of March 2004, and the
project officer expected them to be approved and posted on HHS's Web site,
although he did not know when posting would occur. A report containing
results from the second evaluation component, the telephone surveys of
beneficiaries, was expected later in 2004.50

HHS Has Not Ensured That States Meet Progress-Reporting Requirements

HHS's monitoring and reporting requirements, which the states agree to
carry out under HHS oversight, are set forth in the terms and conditions
attached to each demonstration's approval letter. Although HHS and the
states participated in required telephone conference calls to monitor the
demonstrations' start-up, HHS has not ensured that all states submit the
required quarterly and annual progress reports. The lack of sufficient and
timely information from progress reports may impair the department's
ability to monitor demonstration operations and accomplishments.

48HHS selected Illinois and Wisconsin, the first two demonstrations to be
implemented (in June and July 2002), for the independent evaluation. In
March 2004, HHS's evaluation project officer indicated that the new
Medicare prescription drug program and its potential effects on Pharmacy
Plus demonstrations may lead the department to reassess plans and
schedules for the independent evaluation.

49The descriptions include such information as how each demonstration was
established, demonstration features, each state's economic and political
environment, obstacles encountered during program implementation, and the
state's progress toward implementation and ensuring beneficiary access to
prescription drugs.

50The evaluators have also developed a database, including data provided
from Illinois' and Wisconsin's Medicaid data systems, allowing them to
track and analyze demonstration enrollment, costs, drug use, and
demographics, along with HHS Medicare claims data for Pharmacy Plus
beneficiaries from 1 year before enrollment in the demonstration through
their first year.

Monitoring and reporting requirements are not as clearly established for
the Pharmacy Plus initiative as for the Health Insurance Flexibility and
Accountability (HIFA) initiative:

o  	The HIFA and Pharmacy Plus initiatives both require states to
participate with HHS in monthly telephone monitoring calls. For pharmacy
demonstrations, however, monthly calls are required for 6 months after
implementation and only as needed thereafter; for most approved HIFA
demonstrations, monthly calls are unlimited.

o  	States with approved HIFA demonstrations are required to submit
quarterly progress reports in a format agreed upon with HHS, and
demonstration terms and conditions describe the required content of these
reports. The terms and conditions for Pharmacy Plus demonstrations are
less specific regarding progress report format and content.

o  	HIFA demonstrations are expected to submit separate annual reports
that discuss progress in evaluating the demonstrations, including results
of data collection and analysis to test research hypotheses. Pharmacy Plus
annual reports, in contrast, may be combined with or include the fourth
quarterly progress report, may follow the same broad content guidelines as
quarterly reports, and are not required to report progress in evaluation.

As of March 2004, HHS and the four Pharmacy Plus states had participated
in the initial monitoring phone calls and begun to gather data on how
their demonstrations were working. HHS and the states confirmed
participating in monthly telephone calls for the first 6 months and then
agreeing to maintain contact as needed. An HHS official told us the
department did not set agendas or document these informal contacts, which
focused on demonstration operations as states tracked enrollment and began
to gather information about drug use and expenditures for new
beneficiaries. States reported taking some steps to develop the capacity
to report on their demonstrations. Florida, Illinois, and Wisconsin, for
example, reported having or developing data management systems containing
state Medicaid and other data that are capable of generating
demonstrationspecific reports. South Carolina expected to rely on existing
Medicaid data systems. None of the states, however, were tracking the
number of demonstration enrollees who had become eligible for Medicaid,
although officials in three states reported the ability to do so. Further,
the states had not provided information to HHS to assess whether diversion
savings were occurring.

The information that HHS requires states to report has been insufficient
for determining whether the demonstrations are operating as intended.
According to one HHS official, HHS has not prescribed a standard format
for, or specific information to be provided in, either the quarterly or
annual progress reports; rather, the department works with the states to
obtain needed information. The Pharmacy Plus terms and conditions
stipulate that written quarterly and annual progress reports contain, at
minimum, (1) a discussion of events during the quarter, including
"enrollment numbers, lessons learned, and a summary of expenditures";51
(2) notable accomplishments; and (3) problems and questions that arose and
how they were resolved. The same HHS official told us that in response to
these general requirements, states' progress reports did not always
include all information considered useful for monitoring purposes. For
example, HHS reported that officials were working with Illinois to obtain
additional information to complete its draft annual progress report.
Illinois' six-page annual report, submitted in September 2003, reported
only on new demonstration beneficiaries and did not include first-year
starting or ending enrollment or cost information for the state's Medicaid
senior program as a whole-the services and population affected by the
Pharmacy Plus spending limit. One HHS official told us that after review
of Illinois' report, these cost and enrollment data were specifically
requested to assess whether the new drug benefit was keeping seniors from
becoming eligible for full Medicaid benefits. As of February 2004,
Illinois had not provided this information.

Finally, HHS has not insisted on timely submission of the required
quarterly and annual reports. Although Pharmacy Plus terms and conditions
specify that quarterly reports are due 60 days after the end of the
quarter, and annual reports are due 60 days after the end of the fourth
quarter, HHS has not ensured that states submit the reports on time.
Again, the department's policy is to work with the states toward
compliance. As of January 2004, Florida and Wisconsin had submitted all
required written quarterly reports, mostly on time, while South Carolina
had submitted only one of three required progress reports.52 Illinois,
whose

51Centers for Medicare & Medicaid Services, Pharmacy Plus: A Demonstration
Program under Section 1115: Model Special Terms and Conditions of Approval
(Washington, D.C.: U.S. Department of Health and Human Services, n.d.), 4,
www.cms.hhs.gov/medicaid/1115/pharmplustemptc.pdf (downloaded Apr. 6,
2004).

52According to an HHS official, quarterly reports for South Carolina's
demonstration were delayed because the state had difficulty generating
data on demonstration expenses, requiring department officials to work
with the state on the problem.

Conclusions

demonstration was the first to be implemented, did not submit any of the
three required quarterly reports before submitting its combined fourth
quarterly and first annual report early in September 2003.

HHS's approval and monitoring of state demonstrations under the Pharmacy
Plus initiative raise cost and oversight concerns and, ultimately, program
concerns. The department's approval of four states' demonstrations raises
questions about HHS's basis for its decisions. Because HHS based the
spending limits it approved on higher-thanjustified growth rates, these
spending limits do not, in our view, represent reasonable estimates of
demonstration costs over the 5-year trial periods and are not budget
neutral. It was difficult to assess the reasonableness of the spending
limits themselves, given that they were decided upon through an
undocumented negotiation process, and neither public nor HHS internal
documents stated the rationale for approving higher growth rates. We found
that if HHS's benchmarks had been used to establish the spending limits,
the federal government's liability for the four demonstrations could have
been $1.6 billion lower over 5 years. Moreover, the approved
demonstrations rely on highly questionable assumptions about the extent to
which savings would accrue to Medicaid from improved health of people
receiving the new pharmacy benefit, particularly since many of them
already had pharmacy benefits through existing state-funded programs.

In addition, the Pharmacy Plus initiative raises important evaluation
questions about how improved access to prescription drugs may affect
seniors' health and Medicaid and Medicare costs. Although some of these
questions will likely be addressed by the independent evaluation of two
states' demonstrations, in the interim HHS does not appear to be ensuring
that states provide sufficient, consistent, and timely information for
demonstration monitoring or that states begin implementing their own
evaluation plans. The limited available information on how these
demonstrations are operating makes it difficult to assess whether they are
operating as intended.

The concerns about HHS's approved Pharmacy Plus demonstrations parallel
those we have raised about other section 1115 waiver demonstration
approvals over the past decade. These include the extent to which the
department is protecting the Medicaid program's fiscal integrity and the
need for clear criteria and a public process when HHS reviews and approves
demonstrations. Along with the authority to waive Medicaid requirements,
and the flexibility given states to test new approaches for

delivering services more efficiently and effectively, comes the
responsibility for making decisions based on clear criteria and for
monitoring the demonstrations and learning from them. More can and should
be done to fulfill this responsibility.

Recommendations for Executive Action

In light of our findings that the four HHS-approved Pharmacy Plus
demonstrations are likely to substantially increase federal Medicaid
spending, as previously approved Medicaid section 1115 demonstrations have
done; that HHS's review process and basis for these approvals have not
been clearly set forth; and that approved demonstrations are not all
meeting evaluation and monitoring requirements, we are making seven
recommendations to the Secretary of HHS related to the section 1115
demonstration process.

To improve HHS's process for reviewing and approving states' budget
neutrality proposals for Pharmacy Plus and other Medicaid section 1115
demonstrations, we recommend that the Secretary take three actions:

o  	For future demonstrations, clarify criteria for reviewing and
approving states' proposed spending limits.

o  	Consider applying these criteria to the four approved Pharmacy Plus
demonstrations and reconsider the approval decisions, as appropriate.

o  	Document and make public the basis for any section 1115 demonstration
approvals, including the basis for the cost and enrollment growth rates
used to arrive at the spending limits.

To ensure that approved Pharmacy Plus and other Medicaid section 1115
demonstrations fulfill the objectives stated in their evaluation plans, we
recommend that the Secretary take two actions:

o  	Ensure that states are taking appropriate steps to develop evaluation
designs and to implement them by collecting and reporting the specific
information needed for a full evaluation of the demonstration objectives.

o  	On acceptance, make public the interim and final results of HHS's
independent Pharmacy Plus evaluation.

To ensure that the Secretary and other stakeholders have the information
needed to monitor approved Pharmacy Plus and other Medicaid section 1115
demonstrations to determine if they are functioning as intended, we
recommend that the Secretary take two actions:

o  	Ensure that states provide sufficient information in their
demonstration progress reports, in a consistent format, to facilitate the
department's monitoring.

o  Ensure that states submit required demonstration progress reports in a

Agency and State Comments and Our Evaluation

timely manner.

We provided a draft of this report for comment to HHS and the states of
Florida, Illinois, South Carolina, and Wisconsin. HHS and Florida,
Illinois, and Wisconsin responded with written comments, which are
reproduced in appendixes III through VI, respectively. South Carolina
provided technical comments, which we incorporated in our report as
appropriate.

HHS's Comments and Our Evaluation

HHS concurred with five of our recommendations to strengthen the processes
for approving and overseeing Pharmacy Plus and other Medicaid section 1115
waivers and disagreed with two. It concurred with our recommendations to
make public the basis for section 1115 demonstration approvals and to
ensure that Pharmacy Plus and other Medicaid section 1115 demonstrations
fulfill the objectives of their evaluation plans by working with the
states toward useful program evaluations and making results of the
independent Pharmacy Plus evaluation publicly available. HHS also
concurred with our recommendation to ensure that adequate information is
available to monitor the demonstrations to determine if they are
functioning as intended. In this regard, HHS stated that it has provided
each state that has implemented a Pharmacy Plus demonstration with an
example of an outline and content to be used as a guide for progress
reports and that it will make concerted efforts to ensure that states
submit the reports in a timely manner.

HHS did not concur with our recommendation that the Secretary of HHS
clarify criteria for reviewing and approving states' proposed
demonstration spending limits, indicating that although the department
recognizes the importance of using criteria for reviewing budget
neutrality, strict criteria cannot be determined in advance because
states' circumstances differ. HHS also strongly disagreed with our
recommendation that the Secretary consider applying clarified criteria to
the four approved Pharmacy Plus demonstrations and reconsider the approval
decisions as appropriate. HHS stated that it used criteria to review each
of the approved, disapproved, and pending demonstration proposals;
believes the four approved demonstrations were based on wellsupported
budget estimates of future state spending; and does not believe

it appropriate to reconsider approved demonstrations before the end of the
approval periods.

We agree with HHS that some flexibility is appropriate in considering the
unique Medicaid section 1115 demonstrations proposed by different states.
Consistent with our analyses of other section 1115 demonstration waivers
over the past decade, however, we believe HHS's review process and
decision criteria should be clear, and the results-particularly when
approved spending limits deviate significantly from limits developed using
benchmarks that HHS said it uses as a starting point-should be publicly
explained and documented in the demonstrations' approval letters and terms
and conditions. Even though HHS has developed a standard application form
for Pharmacy Plus demonstrations, that form and other guidance does not
provide written criteria for how HHS reviews and approves the growth rates
that states propose. HHS's rationale for significantly deviating from
benchmarks for projecting future program growth in establishing different
states' spending limits has not been documented or made clear to us or to
others, including other states that may be seeking approval of
demonstration proposals. Without such clarity, questions arise as to how
consistently states have been or will be treated in applying for
demonstrations. Further, in our view, Pharmacy Plus demonstration
approvals were based on questionable savings assumptions. We believe that
HHS should establish clear criteria on which to base the spending limits
and should reconsider its spending limit decisions for the approved
Pharmacy Plus demonstrations in light of such criteria.

HHS also commented that it was premature to evaluate the Pharmacy Plus
demonstrations given the limited experience from 12 to 18 months of
operation. HHS said that were the outcome predetermined, a demonstration
would serve no purpose. The agency believes the Pharmacy Plus initiative
provides states an opportunity to use a Medicaid demonstration to test if
providing drug coverage will prevent the aged and disabled low-income
population from becoming Medicaid eligible. HHS noted that the four
approved demonstrations together are providing drug coverage to 346,000
seniors who would otherwise be without this important benefit.

We agree that it is too early to evaluate the outcomes of the 5-year
demonstrations and that section 1115 demonstrations are intended to test
new propositions. More needs to be done, however, to ensure that states'
evaluations collect the information needed to determine whether those new
propositions are functioning as intended. Four states have Pharmacy

Plus demonstrations in place to test such propositions, and substantial
federal funding is involved, including costs that were previously paid for
by the states themselves. For these reasons, HHS has a responsibility to
(1) make fiscally prudent decisions in its approvals, (2) ensure that
savings hypotheses have some grounding in experience or research, and (3)
ensure that the evaluations are planned and conducted in a way that will
produce adequate information regarding the demonstrations' research
hypotheses.

We also agree that the demonstrations can provide a valuable benefit to
low-income seniors and disabled individuals who might otherwise be without
drug coverage. But three of the four states with approved demonstrations
had state-funded drug coverage programs in place before implementing their
Pharmacy Plus demonstrations, and these state-funded programs became
eligible for federal matching funds when the demonstrations were approved.
We therefore find HHS's statement that the demonstrations are providing
drug coverage to seniors who would otherwise be without it to be an
overstatement.

HHS also commented on how MMA may affect the operation of approved
Pharmacy Plus demonstrations and the review of pending and new
demonstration proposals. HHS stated that seniors covered by the four
Pharmacy Plus demonstrations will be able to begin receiving drug coverage
under the Medicare Part D program in January 2006, and states will be able
to use their own funds to "wrap around" the Medicare benefit to assist
other Medicare beneficiaries whose incomes exceed the level for low-income
subsidies. At that time, HHS believes there will be less need for Pharmacy
Plus demonstrations, given expanded Medicare coverage for prescription
drugs, and states operating the demonstrations will need to decide if they
want to continue doing so and if their demonstrations can continue to be
budget neutral. We have reviewed and incorporated this new information as
appropriate.

HHS's written comments appear in appendix III, along with our response to
additional comments that HHS provided on the findings in our draft report.
The department also provided technical comments, which we considered and
incorporated as appropriate.

Illinois' and Wisconsin's Illinois and Wisconsin officials commented that
our draft report overstated Comments and Our the demonstrations' financial
risk to the federal government and was Evaluation unnecessarily alarming
in light of data showing that the demonstrations

are operating well within their spending limits. In its comments, Illinois

said that it stood by the growth rates it used to develop the spending
limit for its Pharmacy Plus demonstration; it further argued for the
soundness of its demonstration's premise-that providing a drug benefit to
seniors will keep them healthier than if they had no drug coverage. In its
comments, Wisconsin said that the draft report failed to consider the
significant benefits its demonstration offers to the federal government
and to seniors.

We agree that providing a drug benefit to seniors could keep them
healthier, and we do not dispute the benefit to seniors of the states'
drug programs started or expanded through Pharmacy Plus demonstrations.
The demonstrations were approved, however, on the presumption that the
cost of each state's prescription drug program would be paid for in
savings from keeping seniors with little or no previous drug coverage
healthy enough that they would not become eligible for full Medicaid
benefits. Illinois' demonstration was approved on this presumption even
though most of the beneficiaries were already receiving some prescription
drug coverage through the state's existing state-funded program. We remain
concerned that HHS is not maintaining its policy to ensure demonstrations
are budget neutral.

Illinois also commented that it had taken all necessary steps to conduct
its own evaluation and that it had cooperated fully with federal
evaluators and HHS officials. Illinois said that although it officially
filed its quarterly reports late, it submitted all the detailed data
contained in those reports to CMS monthly. We are principally concerned
with the extent to which the information that Illinois provided could be
used to monitor whether the demonstration was operating as intended. Its
one- to two-page quarterly reports, filed late, tallied the number of
beneficiaries enrolled in the demonstration and drug expenditures to date
and provided a narrative paragraph on accomplishments, problems, or
issues. The information itself, however, furnishes little insight as to
whether the demonstration is operating as intended or whether the benefit
is reducing Medicaid costs.

Wisconsin commented that the draft report failed to ascribe any value to
the government of Wisconsin's agreement to cap its federal Medicaid
funding for seniors as a condition of Pharmacy Plus demonstration
approval. We believe the draft report accurately captured HHS's approach
to limiting the federal liability for the Pharmacy Plus demonstrations by
establishing a "cap," or spending limit, as a condition of approval. We
remain neutral on the "value" of this cap for several reasons. Requiring
states to abide by a spending limit is a departure from the open-ended
entitlement nature of the Medicaid program. We also recognize that under

the Medicaid program, states have considerable discretion to alter
spending by increasing-or decreasing-coverage for certain populations and
services. In addition, we recognize that HHS's budget neutrality practices
provide for flexibility in approach and that HHS has established such a
limit on other section 1115 demonstrations before the Pharmacy Plus
initiative.

Wisconsin also commented that the draft report failed to mention that the
demonstrations were reviewed, determined reasonable, and approved by OMB.
We recognize that OMB is involved in assessing budget neutrality and other
aspects of Pharmacy Plus and mentioned that agency's role in our draft
report. Nevertheless, as OMB officials told us, the authority for section
1115 waiver approval rests with the Secretary of HHS, and responsibility
for final Pharmacy Plus approval decisions rests with the Secretary and
his designees.

Wisconsin further commented that in criticizing CMS for not obtaining
better evidence to support projected savings, our report fails to consider
that the reason for demonstration projects is precisely to test such
propositions. We maintain, however, that when HHS establishes a new
initiative to encourage states to apply for Pharmacy Plus demonstrations,
it is the agency's responsibility to ensure that each demonstration's
evaluation objectives are reasonable, each demonstration's savings
assumptions are realistic and grounded in some evidence, and the
evaluations are well planned and data monitoring is established early
enough to assure that the questions can be answered.

Other States' Comments and Our Evaluation

Florida commented that its demonstration was predicated upon savings to be
achieved over the 5-year life of the program and that its proposed
spending limit was close to-less than 1 percent above-the conservative
benchmark spending level we calculated. We agree that Florida's spending
limit was relatively close to a limit based on the benchmarks and included
that information in the draft report.

South Carolina provided technical comments that we incorporated as
appropriate.

As arranged with your offices, unless you release its contents earlier, we
plan no further distribution of this report until 30 days after its date.
At
that time, we will send copies of this report to the Secretary of Health
and
Human Services, the Administrator of the Centers for Medicare &
Medicaid Services, and others who are interested. 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 listed in
appendix VII.

Kathryn G. Allen
Director, Health Care-Medicaid

and Private Health Insurance Issues

Appendix I: Calculating a State's Pharmacy Plus Spending Limit for All
Medicaid Seniors

To achieve budget neutrality, a state's projected 5-year spending with its
Pharmacy Plus demonstration cannot exceed 5-year projected costs without
the demonstration. As a result, the projected costs of a state's existing
Medicaid program for seniors effectively sets the spending limit while the
demonstration is under way. Calculating this withoutdemonstration limit
(steps 1-5 in fig. 1) starts with a base year, generally the most recent
full year for which data are available; calculations for each subsequent
year are based on numbers from the previous year. The result limits a
state's Medicaid spending for all services provided to all Medicaid
seniors in the state.

Appendix I: Calculating a State's Pharmacy Plus Spending Limit for All
Medicaid Seniors

Figure 1: Steps to Calculate Projected 5-Year Without-Demonstration Costs
Step 1

Estimate Medicaid costs per senior for each year

Estimate Medicaid senior enrollment for each year

Determine 1-year costs for Medicaid seniors

Determine costs for each of 4 more years

           Set Pharmacy Plus spending limit for all Medicaid seniors

Source: GAO.

Appendix I: Calculating a State's Pharmacy Plus Spending Limit for All
Medicaid Seniors

Calculating projected 5-year with-demonstration costs follows the same
steps but, in addition, factors in the estimated number of new
beneficiaries receiving only the prescription drug benefit; the costs of
providing them the benefit; and the expected savings, mainly from keeping
these beneficiaries healthy enough to avoid eligibility for full Medicaid.

Appendix II: Denied, Withdrawn, and Pending Pharmacy Plus Demonstration
Proposals as of May 2004

                          State and status Description

Denied

Hawaii Projected enrollment: Individuals with incomes at or below 300
percent of the federal Submitted January 2003 poverty level (FPL).

Denied April 2003 	Coverage and cost sharing: Sought federal assistance
only for administrative costs for a demonstration to make prescription
drugs available at the discounted Medicaid rate plus a dispensing fee.
State was to contribute $1 toward the cost of each prescription in the
first year, increasing to $8 by the fifth year.

Reasons for denial: Exceeded the Pharmacy Plus income limit at or below
200 percent of FPL, provided for only minimal state financial
contributions to pharmacists, and did not include the necessary budget
neutrality analysis.

Delaware Projected enrollment: Seniors and adults with disabilities with
incomes at or below Submitted December 2002 200 percent of FPL or, if
income is above 200 percent of FPL, with prescription drug Denied July
2003 expenses exceeding 40 percent of their incomes.

Coverage and cost sharing: All prescriptions covered by the Medicaid state
plan, up to an annual benefit limit of $2,500. Participants to pay
co-payments of $5 or 25 percent of the cost per prescription, whichever is
greater.

Reason for denial: State already provided drug benefits to the people to
be covered under the demonstration.

Withdrawn

Massachusettsa Submitted July 2002 Withdrawn March 2003 Projected
enrollment: Seniors with incomes at or below 188 percent of FPL.

Coverage and cost sharing: Same broad prescription drug coverage as state
Medicaid plan. State proposed three levels of co-payments (exact amounts
not specified): generic drugs, designated brand-name drugs, and all other
brand-name drugs. Full cost of prescriptions to be covered after
participants reached annual out-of-pocket spending limits: for example, a
single person would pay the lesser of $2,000 or 10 percent of gross annual
income.

Reason for withdrawal: State's existing pharmacy assistance program for
seniors already covered the populations to be included in the
demonstration, and without an expansion the state and the Department of
Health and Human Services (HHS) could not reach agreement on budget
neutrality.

Pending

Connecticut Projected enrollment: Seniors and adults with disabilities
with incomes up to 300 Submitted December 2001 percent of FPL.

Coverage and cost sharing: All prescription drugs and insulin and syringes
with specified exceptions, such as cosmetics and antihistamines. Annual
registration fee of $25 and co-payments of $12 for those with incomes up
to approximately 233 percent of FPL and $20 for those above.

State program: Covers low-income seniors and people with disabilities with
incomes up to approximately 233 percent of FPL. Demonstration would expand
eligibility up to 300 percent of FPL.

Appendix II: Denied, Withdrawn, and Pending Pharmacy Plus Demonstration
Proposals as of May 2004

                          State and status Description

New Jersey Projected enrollment: Seniors and adults with disabilities with
incomes at or below Submitted March 2002 200 percent of FPL.

Coverage and cost sharing: All prescription drugs covered by state
Medicaid plan, with $5 co-payment for each prescription. For brand-name
drug when generic is available, $5 co-payment plus cost difference between
the two.

State program: Covers seniors and adults with disabilities with incomes up
to 222 percent of FPL if single and 202 percent if married. Demonstration
would cover individuals with incomes at or below 200 percent of FPL.

Arkansas Projected enrollment: Qualified Medicare beneficiaries age 65 or
older with incomes at Submitted September 2002 or below 85 percent of FPL.

Coverage and cost sharing: Would cover two prescriptions per beneficiary
per month. Annual $25 enrollment fee and co-payments of $10 for each
generic prescription and $20 for each brand-name drug.

State program: No state-funded pharmacy assistance program for seniors at
the time of demonstration proposal submission.

Indiana

Submitted June 2002
Revised proposal submitted May 2003

Projected enrollment: Seniors with incomes at or below 135 percent of FPL.

Coverage and cost sharing: Same prescription drugs as the state's Medicaid
program, plus insulin, up to annual benefit caps set on a sliding scale:
$1,000 for people with incomes up to 100 percent of FPL; $750 for those
with incomes up to 120 percent of FPL; and $500 for those with incomes at
or below 135 percent of FPL. Participants would pay 50 percent of the
discounted program price, which is the same as the Medicaid price, for
each prescription.

State program: Existing state-funded pharmacy program for low-income
seniors to be covered under the demonstration with no change in
eligibility or drug coverage. State indicated that increased enrollment
was expected in the demonstration following a change from a mail-in rebate
system to a point-of-sale system using a discount card.

Maine Projected enrollment: Seniors age 62 or older and adults with
disabilities with incomes Submitted August 2002 at or below 185 percent of
FPL.

Coverage and cost sharing: Prescription drugs for specified conditions
with 20 percent co-payment for each prescription, or 10 percent if from
mail-order sources. Broader range of drugs available for coverage with 20
percent co-payment after $1,000 out-ofpocket expenses.

State program: Demonstration would cover state-funded pharmacy program,
expand conditions covered, and add voluntary mail-order purchase.

Rhode Island

Submitted October 2002

Projected enrollment: Seniors and adults with disabilities or chronic
illness, including chronic mental illness, with incomes at or below 200
percent of FPL.

Coverage and cost sharing: All prescription drugs covered by state
Medicaid plan. Annual $25 enrollment fee (waived for first year of
program) and co-payments that increase after participants have incurred
$1,800 of drug expenses per year under the program, from $2 to $4 for
generics and from $8 to $12 for brand-name drugs with no generic
equivalent; other brand-name drugs have a $25 co-payment.

State program: The demonstration would cover individuals with incomes at
or below 200 percent of FPL from three state-funded pharmacy programs,
while individuals in those programs with higher incomes would continue to
be state funded. The scope of drugs covered by state programs would be
expanded under the demonstration.

Appendix II: Denied, Withdrawn, and Pending Pharmacy Plus Demonstration
Proposals as of May 2004

                          State and status Description

North Carolina Projected enrollment: Seniors with incomes at or below 200
percent of FPL.

Submitted January 2003 	Coverage and cost sharing: All prescription drugs
and insulin. Co-payments of $5 for generic and $15 for brand-name drugs;
annual benefit limit of $1,000 per participant.

State program: Demonstration would cover and expand existing state-funded
program by broadening prescription drugs covered from drugs for three
specific conditions to those for all conditions, reducing cost sharing,
and increasing annual benefit limit from $600 to $1,000.

Michigan Projected enrollment: Seniors with incomes at or below 200
percent of FPL.

Submitted February 2003 	Coverage and cost sharing: Most prescription
drugs covered by state Medicaid plan, plus insulin and syringes. Annual
$25 enrollment fee and coinsurance of 20 percent of cost of each
prescription up to a monthly cap on a sliding scale determined by
household income. An additional co-payment would be charged for brand-name
drugs with generic equivalents.

State program: Demonstration would cover existing state-funded pharmacy
assistance program with the same eligibility and coverage and expand
enrollment.

Source: GAO analysis of state and HHS documents.

Notes: These descriptions of states' Pharmacy Plus demonstration proposals
are based on the proposals as submitted for HHS review. Changes to
proposals that may be made during the review process and before approval
are not available in documents.

aIn March 2003, Massachusetts withdrew two separate section 1115
demonstration proposals from review: a Pharmacy Plus demonstration for
seniors (the proposal described in this appendix) and a prescription drug
benefit for individuals with disabilities as an amendment to the state's
section 1115 Medicaid managed care demonstration. At the same time,
Massachusetts submitted a new proposal-not a Pharmacy Plus proposal-to add
a drug benefit for certain seniors and disabled individuals as an
amendment to its existing managed care demonstration. In August 2003, that
proposal was also withdrawn.

Appendix III: Comments from the Department of Health and Human Services

Appendix III: Comments from the Department of Health and Human Services

Appendix III: Comments from the Department of Health and Human Services

Appendix III: Comments from the Department of Health and Human Services

Appendix III: Comments from the Department of Health and Human Services

Appendix III: Comments from the Department of Health and Human Services

Appendix III: Comments from the Department of Health and Human Services

Appendix III: Comments from the Department of Health and Human Services

                              Now pages 19 and 20.

Appendix III: Comments from the Department of Health and Human Services

Appendix III: Comments from the Department of Health and Human Services

Appendix III: Comments from the Department of Health and Human Services

GAO's Response to In addition to indicating whether it concurred with our
seven recommendations, HHS commented on the report draft's findings in
threethe Department of areas.

Health and Human

Services' Specific

Comments on GAO's

Findings

Effect of HHS-Approved Pharmacy Plus Demonstrations on Federal Medicaid
Spending

HHS disagreed with our conclusion that the four approved Pharmacy Plus
demonstrations will not prove to be budget neutral to the Medicaid program
and will possibly result in increased federal Medicaid spending. HHS
stated that the department takes seriously its responsibility to ensure
budget neutrality in the Medicaid demonstrations it approves, noting that
it approved four Pharmacy Plus demonstrations while denying two and
reviewing but not approving nine other proposals whose budget estimates
were not well supported.

HHS was concerned that we missed the fundamental purpose of budget
neutrality, which HHS says is not to hold states to a formula-driven cap
but to estimate the amount of future Medicaid spending. HHS believes that
the four approved demonstrations' spending limits were based on
wellsupported budget estimates of future state spending,1 and said its
policy has never been to hold states to benchmark levels of growth. Those
benchmarks are, in HHS's view, a starting point in projecting how the
program will grow, because HHS typically permits states to present
rationales for higher growth rates.

We agree that there may be state-specific circumstances that justify
departures from benchmarks HHS considers as starting points. Nevertheless,
we believe that, given the potential impact on federal Medicaid spending,
HHS and states should justify and document any significant departures from
those starting points. In conducting our assessment, we interviewed HHS
officials and Illinois and Wisconsin state

1HHS also stated that the 0.7 percent per year state historical average
enrollment growth rate we cite for South Carolina (table 2) is in error,
because its records showed that South Carolina's historical average for
enrollment growth was 1.0 percent. In verifying the state's historical
enrollment rate, we noted that the rate had been "rounded up" to the next
full percentage from the 0.7 percent actual historical rate. For
consistency with other rates in the table, we did not round it.

Appendix III: Comments from the Department of Health and Human Services

officials and requested all documents that were considered in their budget
neutrality negotiations. Those interviews and documents, which we
discussed in the draft report, did not fully support the higher growth
rates that were approved. We note that enrollment growth rates, in
particular, can have a significant multiplier effect on future spending
estimates. Further, we note that HHS allowed at least one state to argue
for a higher growth rate using broad justifications-such as the effect of
the Social Security Administration's nationwide outreach program for
low-income Medicare beneficiaries-that other states could also have used
but did not, raising questions of clarity and consistency in both the
process and the final decisions.

Documentation of HHS's approval decisions and the basis for approved
spending limits could provide a rationale for higher cost and enrollment
growth rates and offer guidance and assurance of consistent treatment to
other states applying for Pharmacy Plus demonstrations. Absent such
documentation, neither HHS nor the states have adequately justified the
departures from states' historical growth rates or the CMS Actuary's
growth projections in establishing states' spending limits.

HHS's Review Process and Basis for Approvals

In its comments, HHS stated that the federal review process for Pharmacy
Plus demonstration proposals is similar to the review process for other
Medicaid section 1115 demonstrations, indicating that the process is
necessarily interactive and involves numerous meetings within the federal
team and with states. We acknowledge that the review process for Medicaid
section 1115 demonstration proposals benefits from being inclusive and
interactive, and we are not suggesting that HHS should establish a new or
different review process specifically for the Pharmacy Plus
demonstrations. Our concern is that the basis for its decisions and any
agreed-upon spending limit be clear and justified, not only for Pharmacy
Plus demonstrations but for all section 1115 approvals. As noted in the
draft report, the concerns raised by HHS's approved Pharmacy Plus
demonstrations parallel those we have raised about other section 1115
waiver demonstration approvals over the past decade, including concerns
about the extent to which the department is protecting the Medicaid
program's fiscal integrity and the need for clear criteria and a public
process in reviewing and approving demonstrations.

Approved Demonstrations' HHS commented that the department plans to
continue working with Evaluation and Monitoring states toward developing
useful program evaluations based on consistent Requirements data
collection as well as sufficient, consistent, and timely monitoring

information. HHS also plans to make results of the independent Pharmacy

Appendix III: Comments from the Department of Health and Human Services

Plus evaluation available on the CMS Web site. With regard to states' own
evaluations, HHS emphasized practical limitations, such as constraints on
state financial and staff resources, indicating that while states ideally
would develop evaluation plans before implementing demonstrations, in
practice such plans often change. HHS commented that it obtains sufficient
information for monitoring the demonstrations through telephone contacts
and progress reports that respond to an example outline the department
provided to each demonstration state.

We recognize that state resources are limited, demonstration
implementation tends to be a higher priority than evaluation, and the
independent contractor evaluation of Pharmacy Plus will provide
substantial information. Nonetheless, the lack of action to monitor key
information-such as whether demonstration enrollees are being diverted
from Medicaid-to plan how their evaluations will be conducted, or to
collect data needed for such evaluations suggests a low priority for
ensuring that evaluations can and will be done. HHS needs to ensure that
states provide sufficient, consistent, and timely information for both
demonstration monitoring and for determining whether the demonstrations
are functioning as intended and to ensure that evaluation plans are put
into place.

Appendix IV: Comments from the State of Florida

            Page 63 GAO-04-480 Medicaid Pharmacy Plus Demonstrations

            Page 64 GAO-04-480 Medicaid Pharmacy Plus Demonstrations

Appendix VI: Comments from the State of Wisconsin

Appendix VI: Comments from the State of Wisconsin

Appendix VI: Comments from the State of Wisconsin

Appendix VI: Comments from the State of Wisconsin

Appendix VI: Comments from the State of Wisconsin

               Appendix VI: Comments from the State of Wisconsin

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

In addition to overall comments on our draft report contained in its
letter and discussed in the body of this report, Wisconsin provided 11
specific comments in an attachment to its letter, which is reproduced on
pages 67 through 69. Our responses to Wisconsin's specific comments are
numbered below to correspond with each of the state's numbered comments.

1. 	Wisconsin commented that our $416 million figure (the estimated
federal share of the difference between HHS-approved and benchmark 5-year
spending limits in table 3) exaggerates the federal fiscal effect, because
the actual costs of the demonstration's first years have come in under the
projected costs. The state currently projects federal costs for the new
drug benefit under the demonstration totaling $250 million over 5 years
instead of roughly $537 million, which is the federal share of $919
million approved for the new benefit (see table 1). Although we recognize
that the actual costs of Wisconsin's demonstration to date are less than
the costs projected at the time the waiver was approved, our analysis
examined the extent to which HHS ensured that the demonstrations-in the
form they were approved-maintained spending limits that were budget
neutral to the federal government. Because Wisconsin's approved spending
limit represents the total amount the state is authorized to spend over
the demonstration's 5year life-span, the federal government could be
liable for as much as $416 million more than what it would have been
liable for had HHS held the state to a spending limit based on benchmark
rates (see table 3).

2. 	Wisconsin commented that it is unreasonable to hold HHS to applying
the lower of two benchmark growth rates in calculating budget neutrality:
state experience or projections by the Centers for Medicare & Medicaid
Services' (CMS) Actuary for Medicaid costs. The state also expressed
concern that our analysis did not incorporate factors other than the
benchmarks that affect program growth. We believe that it is reasonable to
expect HHS to use objective benchmark growth rates in projecting the
Medicaid costs on which it bases spending limits and to document its
reasons for deviating from those benchmarks-even if the department regards
them as starting points. Otherwise, the department's rationale for setting
higher spending limits (based on higher growth rates) for some states than
for others is not apparent to other states involved in waiver negotiations
and reviews. As noted in the draft report, HHS responded to Wisconsin's
request for higher growth rates but did not, in our view, adequately
document the basis for approving higher rates. In our own analysis of the
spending limits, we did not include the additional factors that Wisconsin
asserted

Appendix VI: Comments from the State of Wisconsin

should raise its spending limits because neither the state nor HHS
provided adequate support to justify doing so.

3. 	Wisconsin stated that our interpretation was unreasonably narrow in
not accounting for potential savings accruing to Medicare, as well as to
Medicaid, from expanding prescription drug coverage for seniors. We
considered savings to Medicaid alone because HHS allows states to include
savings only to Medicaid, not Medicare, in determining whether their
Medicaid demonstrations are budget neutral.

4. 	Wisconsin commented that because its historical cost growth rate has
been rising, it was appropriate for HHS to calculate the state's spending
limit using a rate higher than its historical 5-year average. We believe
that whenever HHS allows growth rate projections that exceed its
benchmarks, it should document the basis for this deviation.

5. 	Wisconsin mentioned two state programs that it believes will, like the
Social Security Administration's outreach program for low-income Medicare
beneficiaries, help increase senior enrollment in Wisconsin's Medicaid
program because they are likely to identify individuals who qualify for
full Medicaid benefits. But the state did not quantify or provide any data
or other evidence to show the potential effects of these programs or of
Social Security Administration outreach. We did not include these effects
in our benchmark analysis for the same reason we did not include other
factors that Wisconsin believed should raise is spending limit (see our
response to comment 2).

6. 	Wisconsin noted that we found no firm evidence to support the idea
that expanding drug coverage would produce significant savings in Medicaid
by diverting or delaying Medicaid enrollment. The state asserts that this
criticism ignores the central purpose of these demonstrations: to
determine if an important health care benefit can be delivered cost
effectively. We acknowledge the value of demonstrations to test health
care alternatives, but we believe that the case for substantial savings to
Medicaid due to expanded prescription drug coverage is not well supported.
We also believe that HHS has not done enough to ensure that states develop
and implement demonstration evaluation designs. Although we do not dispute
Wisconsin's comment that research suggests coverage of prescription drugs
benefits seniors, we believe that demonstrating the effects of drug
coverage on avoiding Medicaid enrollment is a separate issue.

7. 	Wisconsin has interpreted our mention of congressional concern about
the extent to which HHS has ensured that section 1115 demonstration
waivers promote the goals of Medicaid as implying that the state's

Appendix VI: Comments from the State of Wisconsin

demonstration is not providing critical prescription drugs to a vulnerable
elderly, low-income, uninsured population. We did not intend to suggest
that Wisconsin's demonstration is not a valuable benefit to these
individuals. We were referring to our earlier work on section 1115
Medicaid and State Children's Health Insurance Program (SCHIP)
demonstrations, which, in addition to raising concerns about HHS's use of
section 1115 waiver authority to approve demonstration spending limits
that were not budget neutral, also found that HHS was allowing states to
use unspent SCHIP funding to cover childless adults, despite SCHIP's
statutory objective of expanding health coverage to low-income children.1

8. 	Wisconsin commented that we mischaracterized the growth rates approved
by HHS for the state's demonstration as too high. See our response to
comment 2.

9. 	Wisconsin objected to our conclusion that the lack of available
information on how these demonstrations are operating compromises attempts
to assess whether they are operating as intended. This statement does not
apply to any one state alone but synthesizes our findings for the four
approved demonstrations taken together. We acknowledge that Wisconsin has
been responsive to HHS's requirements for informative and timely progress
reports and have revised our report as appropriate.

10. Wisconsin stated that its data reporting system allows its staff to
monitor that the demonstration is operating as intended. In the draft
report, we noted that Wisconsin officials reported having the capability
for monitoring. We have not assessed Wisconsin's monitoring system.

11. Wisconsin commented on the importance of, and Wisconsin's full
participation in, CMS's contracted independent evaluation as an effective
approach to reviewing the agency's assumptions relating to budget
neutrality and program effectiveness. We believe the draft report captured
the plans for this independent evaluation, as well as the apparent
confusion over each state's responsibility for conducting

1See GAO-02-817 and U.S. General Accounting Office, SCHIP: HHS Continues
to Approve Waivers That Are Inconsistent with Program Goals, GAO-04-166R
(Washington, D.C.: Jan. 5, 2004).

Appendix VI: Comments from the State of Wisconsin

its own evaluation. We have revised our report to reflect that Wisconsin
officials believe the state is not required to conduct an evaluation,
whereas HHS officials told us the state would be required to do so.

Appendix VII: GAO Contact and Staff Acknowledgments

GAO Contact Katherine Iritani, (206) 287-4820

Acknowledgments 	In addition, Tim Bushfield, Ellen W. Chu, Helen
Desaulniers, Behn Kelly, Suzanne Rubins, Ellen M. Smith, and Stan
Stenersen made key contributions to this report.

Related GAO Products

SCHIP: HHS Continues to Approve Waivers That Are Inconsistent with Program
Goals. GAO-04-166R. Washington, D.C.: January 5, 2004.

Medicaid and SCHIP: Recent HHS Approvals of Demonstration Waiver Projects
Raise Concerns. GAO-02-817. Washington, D.C.: July 12, 2002.

Medicare and Medicaid: Implementing State Demonstrations for Dual
Eligibles Has Proven Challenging. GAO/HEHS-00-94. Washington, D.C.: August
18, 2000.

Medicaid Section 1115 Waivers: Flexible Approach to Approving
Demonstrations Could Increase Federal Costs. GAO/HEHS-96-44. Washington,
D.C.: November 8, 1995.

Medicaid: State Flexibility in Implementing Managed Care Programs Requires
Appropriate Oversight. GAO/T-HEHS-95-206. Washington, D.C.: July 12, 1995.

Medicaid: Statewide Section 1115 Demonstrations' Impact on Eligibility,
Service Delivery, and Program Cost. GAO/T-HEHS-95-182. Washington, D.C.:
June 21, 1995.

Medicaid: Spending Pressures Drive States toward Program Reinvention.
GAO/T-HEHS-95-129. Washington, D.C.: April 4, 1995.

Medicaid: Spending Pressures Drive States toward Program Reinvention.
GAO/HEHS-95-122. Washington, D.C.: April 4, 1995.

Medicaid: Experience with State Waivers to Promote Cost Control and Access
to Care. GAO/T-HEHS-95-115. Washington, D.C.: March 23, 1995.

GAO's Mission

Obtaining Copies of GAO Reports and Testimony

The General Accounting Office, the audit, evaluation and 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.

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 fulltext 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,	Contact: Web site: www.gao.gov/fraudnet/fraudnet.htm

Waste, and Abuse in E-mail: [email protected]

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

Jeff Nelligan, Managing Director, [email protected] (202) 512-4800

Public Affairs 	U.S. General Accounting Office, 441 G Street NW, Room 7149
Washington, D.C. 20548
*** End of document. ***