This is the accessible text file for GAO report number GAO-12-864 
entitled 'Medicare Special Needs Plans: CMS Should Improve Information 
Available about Dual-Eligible Plans' Performance' which was released 
on September 19, 2012. 

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

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

United States Government Accountability Office: 
GAO: 

Report to Congressional Requesters:  

September 2012:  

Medicare Special Needs Plans: 

CMS Should Improve Information Available about Dual-Eligible Plans' 
Performance: 

GAO-12-864: 

GAO Highlights: 

Highlights of GAO-12-864, a report to congressional requesters.  

Why GAO Did This Study: 

About 9 million of Medicare’s over 48 million beneficiaries are also 
eligible for Medicaid because they meet income and other criteria. 
These dual-eligible beneficiaries have greater health care challenges 
than other Medicare beneficiaries, increasing their need for care 
coordination across the two programs. In addition to meeting all the 
requirements of other MA plans, D-SNPs are required by CMS to provide 
specialized services targeted to the needs of dual-eligible 
beneficiaries as well as integrate benefits or coordinate care with 
Medicaid services. GAO was asked to examine D-SNPs’ specialized 
services to dual-eligible beneficiaries. GAO (1) analyzed the 
characteristics of dual-eligible beneficiaries in D-SNPs and other MA 
plans, (2) reviewed differences in specialized services between D-SNPs 
and other MA plans, and (3) reviewed how D-SNPs work with state 
Medicaid agencies to enhance benefit integration and care 
coordination. GAO analyzed CMS enrollment, plan benefit package, 
projected revenue, and beneficiary health status data; reviewed 15 D-
SNP models of care and 2012 contracts with states; and interviewed 
representatives from 15 D-SNPs and Medicaid agency officials in 5 
states.  

What GAO Found: 

About 9 percent of the dual-eligible population is enrolled in 322 
Medicare dual-eligible special needs plans (D-SNP), a type of Medicare 
Advantage (MA) plan. All dual-eligible beneficiaries are low income, 
but those in D-SNPs tended to have somewhat different demographic 
characteristics relative to those dual-eligible beneficiaries in other 
MA plans. On the basis of the most current data available (2010-2011), 
compared to those in other MA plans, dual-eligible beneficiaries in D-
SNPs were more frequently under age 65 and disabled, more likely to be 
eligible for full Medicaid benefits, and more frequently diagnosed 
with a chronic or disabling mental health condition. In spite of these 
differences, the health status of D-SNP enrollees as measured by their 
expected cost to Medicare was similar to the health status of dual-
eligible enrollees in other MA plans in 2010.  

D-SNPs provide fewer supplemental benefits-—benefits not covered by 
Medicare fee-for-service (FFS)-—on average, than other MA plans. Of 
the 10 supplemental benefits offered by more than half of D-SNPs, 7 
were offered more frequently by other MA plans and 3 were offered more 
frequently by D-SNPs. Yet D-SNPs spent proportionately more of their 
rebate—-additional Medicare payments received by many plans—-to fund 
supplemental benefits compared to other MA plans, and less to reduce 
Medicare cost-sharing, which is generally covered by Medicaid. The 
models of care GAO reviewed, of 107 submitted for 2012, described in 
varying detail how the D-SNP planned to provide specialized services, 
such as health risk assessments, and meet other requirements, such as 
measuring performance. However, the Centers for Medicare & Medicaid 
Services (CMS), which administers Medicare and oversees Medicaid, did 
not require D-SNPs to use standardized measures in the models of care, 
which would make it possible to compare the performance of D-SNPs. 
While D-SNPs are not required to report that information to CMS, such 
information would be useful for future evaluations of whether D-SNPs 
met their intended results, as well as for comparing D-SNPs.  

CMS stated that contracts between D-SNPs and state Medicaid agencies 
are an opportunity to increase benefit integration and care 
coordination. Our review of the contracts indicated only about one-
third of the 2012 contracts contained any provisions for benefit 
integration, and only about one-fifth provided for active care 
coordination between D-SNPs and Medicaid agencies, which indicates 
that most care coordination was done exclusively by D-SNPs, without 
any involvement of state Medicaid agencies. However, some D-SNP 
contracts with state Medicaid agencies specified that the agencies 
would pay the D-SNPs to provide all or some Medicaid benefits. 
Representatives from the D-SNPs and Medicaid officials from the states 
GAO interviewed expressed concerns about the contracting process, such 
as limited state resources for developing and overseeing contracts, as 
well as uncertainty about whether Congress will extend D-SNPs as a 
type of MA plan after 2013, and the implementation of other 
initiatives to coordinate Medicare and Medicaid benefits for dual-
eligible beneficiaries that could replace D-SNPs.  

What GAO Recommends: 

To increase D-SNPs’ accountability, GAO recommends improving D-SNP 
reporting of services provided to dual-eligible beneficiaries and 
making this information available to the public. In its comments on a 
draft of GAO’s report, CMS generally agreed with our recommendations.  

View [hyperlink, http://www.gao.gov/products/GAO-12-864]. For more 
information, contact James C. Cosgrove at (202) 512-7114 or 
CosgroveJ@gao.gov.  

[End of section]  

Contents: 

Letter: 

Background: 

Dual-Eligible Beneficiaries in D-SNPs Had Similar Health Status to 
Other Dual-Eligible Beneficiaries: 

D-SNPs Typically Offer Fewer Supplemental Benefits Than Other MA 
Plans, and Information on Planned Services Is Not Comparable: 

Although Some D-SNP Contracts with State Medicaid Agencies Expressly 
Provide for Benefit Integration or Care Coordination, Most Do Not: 

Conclusions: 

Recommendations for Executive Action: 

Agency and Other External Comments and Our Evaluation: 

Appendix I: Comments from the Department of Health & Human Services: 

Appendix II: GAO Contact and Staff Acknowledgments: 

Tables: 

Table 1: Legislation Establishing and Modifying Authority for Special 
Needs Plans (SNP) to Operate under the Medicare Advantage (MA) Program: 

Table 2: Percentage of Rebate Allocated to Supplemental Benefits, 
Reducing Cost Sharing, and Premium Reduction, 2012: 

Figures: 

Figure 1: Enrollment of Dual-Eligible Beneficiaries by Plan Type, 2011: 

Figure 2: Characteristics of Dual-Eligible Beneficiaries in D-SNPs 
Compared with Dual-Eligible Beneficiaries in Other MA Plans and FFS: 

Figure 3: Average Risk Scores of Dual-Eligible Beneficiaries by 
Characteristic and Plan Type, 2010: 

Figure 4: Supplemental Benefits Offered by More Than Half of Dual-
Eligible Special Needs Plans (D-SNP) Compared with Other Medicare 
Advantage (MA) Plans, 2012: 

Figure 5: Dual-Eligible Special Needs Plans (D-SNP) Integration of 
Medicaid Benefits, 2012: 

Abbreviations: 

CMS: Centers for Medicare & Medicaid Services:  

C-SNP: chronic condition special needs plan:  

D-SNP: dual-eligible special needs plan:  

ESRD: end-stage renal disease:  

FFS: fee-for-service:  

FIDESNP: fully integrated dual-eligible special needs plan:  

HCPP: health care prepayment plan:  

HEDIS: Healthcare Effectiveness Data and Information Set:  

HHS: Department of Health and Human Services:  

I-SNP: institutional special needs plan:  

MA: Medicare Advantage:  

MedPAC: Medicare Payment Advisory Commission:  

MIPPA: Medicare Improvements for Patients and Providers Act of 2008:  

MMA: Medicare Prescription Drug, Improvement, and Modernization Act of 
2003:  

NCQA: National Committee for Quality Assurance:  

PACE: Program of All-Inclusive Care for the Elderly:  

PPACA: Patient Protection and Affordable Care Act:  

SNP: special needs plan:  

SSA: Social Security Act: 

[End of section] 

United States Government Accountability Office: 
Washington, DC 20548:  

September 13, 2012: 

The Honorable Sander Levin: 
Ranking Member: 
Committee on Ways and Means: 
House of Representatives:  

The Honorable Henry A. Waxman: 
Ranking Member: 
Committee on Energy and Commerce: 
House of Representatives:  

The Honorable Pete Stark: 
Ranking Member: 
Subcommittee on Health: 
Committee on Ways and Means: 
House of Representatives:  

About 9 million of Medicare's over 48 million beneficiaries are also 
eligible for Medicaid, a joint federal-state program that finances 
health insurance coverage for certain categories of low-income adults 
and children.[Footnote 1] In 2007, these individuals, referred to as 
dual-eligible beneficiaries, made up 18 percent of all Medicare 
beneficiaries but accounted for 31 percent of Medicare spending. 
[Footnote 2] In the same year, dual-eligible beneficiaries were about 
15 percent of Medicaid enrollees but accounted for nearly 40 percent 
of Medicaid spending.[Footnote 3] Disproportionate spending for these 
individuals is largely because dual-eligible beneficiaries are more 
likely than other Medicare beneficiaries to be disabled; report poor 
health status and limitations in their activities of daily living, 
such as bathing and toileting; and have cognitive impairments, mental 
disorders, and certain chronic conditions such as diabetes and 
pulmonary disease. In addition, dual-eligible beneficiaries' health 
care services must be coordinated across Medicare and Medicaid, and 
each program has its own set of covered services, provider networks, 
regulations, and payment policies. 

The Medicare Prescription Drug, Improvement, and Modernization Act of 
2003 (MMA) authorized the creation of a type of Medicare Advantage 
(MA) plan referred to as a special needs plan (SNP), to address the 
unique needs of certain Medicare populations.[Footnote 4] About 1.2 
million of the dual-eligible population is enrolled in dual-eligible 
SNPs (D-SNP),[Footnote 5] which are SNPs exclusively for dual-eligible 
beneficiaries. Dual-eligible beneficiaries may also choose to enroll 
in other types of SNPs for which they are eligible, including 
institutional SNPs (I-SNP) for individuals residing in nursing 
facilities or institutions, chronic condition SNPs (C-SNP) for 
individuals with severe or disabling chronic conditions, or other MA 
plans, or remain in Medicare fee-for-service (FFS).[Footnote 6] In 
addition to meeting all the requirements of other MA plans, all SNPs, 
including D-SNPs, are required by the Centers for Medicare & Medicaid 
Services (CMS)--the agency in the Department of Health and Human 
Services (HHS) that administers Medicare and oversees Medicaid--to 
provide specialized services targeted to the needs of their 
beneficiaries, including a health risk assessment and an 
interdisciplinary care team for each beneficiary enrolled.[Footnote 7] 

You asked us to examine the extent to which D-SNPs provide unique 
services for dual-eligible beneficiaries and how D-SNPs and states 
work together to serve these beneficiaries. In this report, we (1) 
describe the characteristics and health status of dual-eligible 
beneficiaries in D-SNPs and how they compare to those of dual-eligible 
beneficiaries in other MA plans and Medicare FFS, (2) determine the 
extent to which D-SNPs' specialized services differ from those offered 
by other MA plans, and (3) describe how D-SNPs work with state 
Medicaid agencies to enhance benefit integration and care coordination 
for dual-eligible beneficiaries. 

To describe the demographic characteristics of dual-eligible 
beneficiaries in D-SNPs and how they compare to those of dual-eligible 
beneficiaries in other MA plans and Medicare FFS, we analyzed July 
2011 enrollment data from CMS. To compare the mental health 
characteristics and health status of dual-eligible beneficiaries in D-
SNPs with dual-eligible beneficiaries enrolled in other MA plans and 
FFS, we analyzed CMS data on 2010 beneficiary risk scores, which 
measure expected Medicare costs for each beneficiary on the basis of 
demographic and diagnosis data.[Footnote 8] Specifically, we 
calculated the average risk scores for dual-eligible beneficiaries in 
D-SNPs, other MA plans, and FFS in 2010.[Footnote 9] In all of these 
analyses, we used enrollment and plan benefit data to identify the 
type of plan (D-SNP, other MA plan, or FFS) in which each beneficiary 
was enrolled. 

To determine the extent to which D-SNPs' specialized services differed 
from services offered by other MA plans, we used CMS's 2012 plan 
benefit data to compare D-SNPs' supplemental benefits, such as dental 
and vision coverage, that are outside the original Medicare FFS 
benefit package, with the supplemental benefits offered by other MA 
plans.[Footnote 10] We also used CMS's 2012 bid pricing tool data, 
which contain information MA plans submitted to CMS on their projected 
revenue requirements for providing Medicare-covered services to 
enrolled beneficiaries, to understand how D-SNPs and other MA plans 
fund the supplemental benefits they offer. Additionally, we 
interviewed officials from 15 D-SNPs that had enrolled at least 100 
dual-eligible beneficiaries in both 2010 and 2011 to discuss what 
services are available to dual-eligible beneficiaries in D-SNPs and, 
when applicable, whether differences exist between the care 
coordination services they offer in their D-SNPs and their MA 
organization's other MA plans. We judgmentally selected these D-SNPs 
to cover a range of geographic regions and plan sizes. We also 
reviewed the 2012 models of care submitted to CMS by these 15 D-SNPs--
representing 14 percent of the 107 models of care we received from 
CMS.[Footnote 11] The model of care provides a narrative description 
of how the D-SNP will address certain clinical and nonclinical 
elements, such as a health risk assessment and an adequate provider 
network. We focused in particular on how D-SNPs identified the most-
vulnerable subpopulations--beneficiaries that need the most-intensive 
care--and how they planned to meet these care needs.[Footnote 12] 

To describe how D-SNPs worked with state Medicaid agencies, we 
analyzed all of the 124 contracts between D-SNPs and state Medicaid 
agencies that were submitted to CMS for 2012 to determine whether the 
contracts contained provisions expressly addressing benefit 
integration and care coordination for dual-eligible beneficiaries 
between the D-SNP and the state Medicaid agency.[Footnote 13] We 
supplemented this analysis with interviews of officials from five 
state Medicaid agencies judgmentally selected to cover a range of 
geographic areas,[Footnote 14] and with officials from our sample of D-
SNPs. 

We assessed the reliability of the data we received from CMS by 
performing appropriate electronic data checks and by interviewing 
agency officials who were knowledgeable about the data. This allowed 
us to determine that the data were suitable for our purposes. We did 
not independently verify the statements of interview respondents or 
the statements in documents, such as models of care that were 
submitted by D-SNPs. 

Our analysis has several limitations. We limited our analysis to the 
50 states and the District of Columbia.[Footnote 15] Although most of 
the analysis is based on 2012, the demographic data are from 2011, and 
the health status data are from 2010, the most recent years available. 
The contracts between D-SNPs and state Medicaid agencies we reviewed 
are limited to contracts submitted to CMS for 2012. Because only 40 
percent of all D-SNPs were required to submit contracts in that year, 
these contracts may not represent the full range of possible D-SNP 
arrangements with state Medicaid agencies. Additionally, because we 
judgmentally selected the state Medicaid agencies and D-SNPs for our 
interviews, we cannot generalize the findings from these interviews to 
all states and all D-SNPs. We did not assess the quality of care 
provided by D-SNPs. 

We conducted this performance audit from September 2011 through 
September 2012 in accordance with generally accepted government 
auditing standards. Those standards require that we plan and perform 
our work to obtain sufficient, appropriate evidence to provide a 
reasonable basis for our findings and conclusions based on our 
research objectives. We believe that the evidence obtained provides a 
reasonable basis for our findings and conclusions based on our audit 
objectives. 

Background: 

Special Needs Plans: 

SNPs, including D-SNPs, have been reauthorized several times since 
their establishment was first authorized in 2003. For example, the 
Medicare Improvements for Patients and Providers Act of 2008 (MIPPA) 
and the Patient Protection and Affordable Care Act (PPACA) both 
contained provisions reauthorizing and modifying SNPs. See table 1 for 
a summary of legislation establishing and modifying SNPs. 

Table 1: Legislation Establishing and Modifying Authority for Special 
Needs Plans (SNP) to Operate under the Medicare Advantage (MA) Program: 

Legislation: The Medicare Prescription Drug, Improvement, and 
Modernization Act of 2003[A]; 
Specific provisions: Authorized the establishment of MA coordinated 
care plans that exclusively serve individuals in three classes of 
special needs: (1) beneficiaries entitled to Medicare and Medicaid 
(dual-eligible beneficiaries), (2) beneficiaries with severe or 
disabling chronic conditions, or (3) institutionalized beneficiaries. 
Authorized HHS to designate other MA plans that disproportionately 
serve special needs individuals as new SNPs. 
Authorized SNPs to restrict enrollment to individuals in the classes 
of special needs through December 31, 2008.  

Legislation: The Medicare, Medicaid and SCHIP Extension Act of 2007[B]; 
Specific provisions: Beginning on January 1, 2008, and extending 
through December 31, 2009, established a moratorium on the authority 
of HHS to designate other MA plans disproportionately serving special 
needs individuals as SNPs and to permit enrollment of individuals in a 
new or expanding SNP (which was not subsequently extended). 
Extended authority for SNPs to restrict enrollment to individuals in 
the classes of special need through December 31, 2009.  

Legislation: The Medicare Improvements for Patients and Providers Act 
of 2008[C] (MIPPA); 
Specific provisions: Effective January 1, 2010, requires all D-SNPs to 
have contracts with state Medicaid agencies to provide--or arrange to 
provide--benefits to eligible individuals under Medicaid. Made an 
exception through December 31, 2010, for existing plans not expanding 
their service areas. 
Extended the moratorium on the authority of HHS to designate other MA 
plans that disproportionately serve special needs individuals as SNPs 
through December 31, 2010, but lifted the moratorium on new SNPs and 
existing SNPs at the end of 2009. 
Extended authority for SNPs to restrict enrollment to individuals in 
the classes of special need through December 31, 2010.  

Legislation: The Patient Protection and Affordable Care Act (2010)[D] 
(PPACA); 
Specific provisions: Effective January 1, 2012, and subsequent years, 
required all SNPs to be approved by the National Committee for Quality 
Assurance (NCQA) based on standards established by HHS. 
Extends the exception for existing D-SNPs that do not expand their 
current service areas to continue operating without contracts with 
state Medicaid agencies through December 31, 2012. 
Authorized special payment rules for fully integrated SNPs for dual-
eligible beneficiaries. 
Extended authority for SNPs to restrict enrollment to individuals in 
the classes of special need through December 31, 2013.  

Source: GAO analysis. 

[A] Pub. L. No. 108-173, § 231, 117 Stat. 2066, 2207. 

[B] Pub. L. No. 110-173, § 108, 121 Stat. 2492, 2496. The acronym 
SCHIP stands for the State Children's Health Insurance Program. 

[C] Pub. L. No. 110-275, § 164, 122 Stat. 2494, 2571. 

[D] Pub. L. No. 111-148, § 3205, 124 Stat. 119, 457.  

[End of table]  

In 2012, 322 D-SNPs are operating in 38 states and the District of 
Columbia.[Footnote 16] CMS pays D-SNPs the same way that it pays other 
MA plans; that is, a monthly amount determined by the plan bid--the 
plan's estimated cost of providing Medicare Part A and Part B 
benefits--in relation to a benchmark, which is the maximum amount the 
Medicare program will pay MA plans in a given locality. CMS then 
adjusts the monthly payments to MA plans on the basis of 
beneficiaries' risk scores.[Footnote 17] If an MA plan's bid exceeds 
the benchmark, the plan must charge each of its beneficiaries an 
additional premium to make up the difference. If a plan's bid is less 
than the benchmark, a proportion of the difference is returned to the 
plan as additional Medicare payments called rebates, which must be 
used to reduce premiums, reduce cost sharing, or provide mandatory 
supplemental benefits, such as vision and dental care.[Footnote 18] 
Beginning in 2012, CMS has begun to phase in PPACA-mandated 
modifications in the rebate amount and introduced varied rebate 
amounts based on CMS's assessments of plan quality.[Footnote 19] For 
2012, rebates ranged from 66.67 percent of the difference between a 
plan's bid and benchmark for plans with the lowest quality ratings to 
73.33 percent of the difference for plans with the highest quality 
ratings. 

D-SNPs must meet the same requirements as other MA plans, such as 
submitting an application to CMS. And like other MA plans, D-SNPs that 
meet minimum enrollment requirements are also required to submit data, 
such as the Health Effectiveness Data and Information Set (HEDIS) 
quality measures.[Footnote 20] In addition, they must conduct quality 
improvement activities, which include the reporting of certain 
structure and process measures, such as describing how they manage 
medication reconciliation associated with patient transitions between 
care settings.[Footnote 21] 

CMS requires D-SNPs to develop a model of care that describes their 
approach to caring for their target population.[Footnote 22] The model 
of care must describe how the plan will address 11 clinical and 
nonclinical elements established in CMS guidance: (1) describing the 
specific target population, (2) tracking measurable goals, (3) 
describing the staff structure and care management goals, (4) 
providing an interdisciplinary care team, (5) establishing a provider 
network that has specialized expertise and describing the use of 
clinical practice guidance and protocols, (6) training plan employees 
and the provider network on the model of care, (7) performing health 
risk assessment, (8) creating individualized care plans, (9) 
establishing a communications network, (10) providing care management 
for the most vulnerable subpopulations, and (11) measuring plan 
performance and health outcomes. These models of care are reviewed and 
approved by NCQA--a private health care quality organization--on the 
basis of scoring criteria developed with CMS that emphasized the 
inclusion of in-depth descriptions or case studies. In their MA 
applications, D-SNPs must also "attest" that they meet a total of 251 
subelements related to the 11 elements in their model of care. 

New and expanding D-SNPs are required to contract with state Medicaid 
agencies in 2012, and beginning in 2013, all D-SNPs will be required 
to have contracts with state Medicaid agencies.[Footnote 23] According 
to CMS, the contracts are an opportunity to improve the integration of 
Medicare and Medicaid benefits, and the agency has implemented this 
requirement with the goal of "increased integration and coordination" 
for dual-eligible beneficiaries.[Footnote 24] 

D-SNPs do not cover the same categories of dual-eligible 
beneficiaries, but their chosen category(ies) must correspond to those 
under the Medicaid program in the state in which the D-SNP is being 
offered. Dual-eligible beneficiaries fall into two main categories. 
One group, termed full-benefit dual-eligible beneficiaries, may 
receive the entire range of Medicaid benefits, including long-term 
care. The other group, partial-benefit dual-eligible beneficiaries, 
does not receive Medicaid-covered health care services, but Medicaid 
covers Medicare premiums or cost-sharing, or both, for these 
beneficiaries. Some D-SNPs limit enrollment to full-benefit dual-
eligible beneficiaries, while others are open to all dual-eligible 
beneficiaries. Additionally, some D-SNPs are open only to disabled 
beneficiaries under age 65, whereas others are open only to those aged 
65 and over. 

Other Initiatives for Care Coordination for Dual-Eligible 
Beneficiaries: 

Recently, the federal government, states, researchers, and advocates 
have focused increased attention on care coordination for dual-
eligible beneficiaries. PPACA required HHS to establish the Federal 
Coordinated Health Care Office (generally known as the Medicare-
Medicaid Coordination Office) within CMS to more-effectively integrate 
Medicare and Medicaid benefits and to improve federal-state 
coordination for dual-eligible beneficiaries to ensure that they 
receive full access to the items and services to which they are 
entitled.[Footnote 25] Experts believe that, in addition to benefiting 
dual-eligible beneficiaries, more-effective benefit integration and 
care coordination can generate significant savings by, for example, 
lowering emergency room use. 

The Medicare-Medicaid Coordination Office, working with the Center for 
Medicare & Medicaid Innovation (called the Innovation Center), is 
beginning a financial alignment initiative that is expected to enroll 
up to 2 million beneficiaries in 26 states and is intended to align 
Medicare and Medicaid services and funding so as to reduce costs while 
improving beneficiaries' care.[Footnote 26] Through these 
demonstrations, the federal and state governments expect to realize 
savings from aligning the payments and integrating care. Under 
existing coordination efforts, integrating benefits requires an 
investment of resources from states to work with D-SNPs, or other 
stakeholders, but according to experts most of the financial savings 
accrue to Medicare, because most savings result from services that are 
largely paid for by Medicare, such as reductions in the number or 
length of hospital stays. Under the financial alignment initiative, 
savings will be shared by Medicare and Medicaid without reference to 
whether the savings are achieved in Medicare or Medicaid services, 
although the allocation of these savings between Medicare and Medicaid 
has not yet been finalized. Two models will be tested: a managed FFS 
model, under which payments are adjusted retrospectively, and a 
capitated model under which one payment is made to an MA plan under a 
three-way contract among Medicare, the state Medicaid agency, and the 
plan. 

The initiative is being conducted under the demonstration authority of 
the Innovation Center, under which the Secretary of HHS may conduct 
evaluations that analyze both quality of care and changes in spending. 
[Footnote 27] For purposes of testing models under this authority, 
budget neutrality--which would require that no more be spent under the 
demonstration than is currently being spent on care for dual-eligible 
beneficiaries--does not apply. The Secretary can expand the 
demonstrations nationwide if the demonstrations are determined to 
reduce spending without reducing the quality of care or improve the 
quality of care without increasing spending.[Footnote 28] 

The state demonstrations under the financial alignment initiative do 
not necessarily include D-SNPs, and in some cases may replace D-SNPs. 
As of June 2012, all 26 states had submitted their proposals for the 
demonstrations. Decisions about implementation of these designs had 
not been announced as of July 2012 yet implementation of these models 
is expected to begin by January 2013 and continue into 2014. 

Dual-Eligible Beneficiaries in D-SNPs Had Similar Health Status to 
Other Dual-Eligible Beneficiaries: 

The demographic and mental health characteristics of dual-eligible 
beneficiaries enrolled in D-SNPs in 2011 differed from those of dual-
eligible beneficiaries in other MA plans and, to a lesser extent, from 
those of dual-eligible beneficiaries in FFS.[Footnote 29] Despite 
these differences, dual-eligible beneficiaries in D-SNPs and dual-
eligible beneficiaries in FFS and other MA plans had very similar 
health status in 2010, the year for which the most recent data were 
available, as measured by Medicare risk scores. 

Dual-Eligible Beneficiaries in D-SNPs Had Characteristics That Were 
Generally Similar to Their FFS Counterparts but Differed in Key 
Respects from Such Beneficiaries in Other MA Plans: 

Dual-eligible beneficiaries in D-SNPs (9 percent of all dual-eligible 
beneficiaries in 2011, as shown in figure 1) were most similar to dual-
eligible beneficiaries in FFS, but differed substantially from dual-
eligible beneficiaries in other MA plans on certain demographic and 
mental health measures.[Footnote 30] 

Figure 1: Enrollment of Dual-Eligible Beneficiaries by Plan Type, 2011: 

[Refer to PDF for image: pie-chart]  

FFS: 80%; 
Other MA[A]: 10%; 
D-SNP: 9%; 
Other: 1%: including: 
PACE[B]: 15%; 
Coast plan/HCPP[C]: 16%; 
I-SNP[D]: 27%; 
C-SNP[E]: 42%.  

Source: GAO analysis of CMS data.  

Notes: Percentages are based on 9,188,828 dual-eligible beneficiaries 
in Medicare and Medicaid as of July 2011. Data sources include CMS's 
2011 enrollment data, 2011 plan benefit package data, and 2010 risk 
score data (used only to determine residence). Data exclude dual-
eligible beneficiaries in Puerto Rico and those living outside of the 
50 states and the District of Columbia in 2010. 

[A] Other MA plans include non-SNP health maintenance organizations 
(including point of service options), local and regional preferred 
provider organizations, provider-sponsored organizations, private fee-
for-service plans, and medical savings account plans. 

[B] Program of All-Inclusive Care for the Elderly (PACE), which 
provides a range of integrated preventative, acute care, and long-term 
care services for the frail elderly. 

[C] Cost plans provide original Medicare services, but enrollees can 
receive covered services outside of the health maintenance 
organization or competitive medical plan network. Health care 
prepayment plans (HCPP) are union-or employer-sponsored plans that 
provide or arrange for some or all of Part B Medicare benefits on a 
prepayment basis, while payments for Part A services are made on a FFS 
basis. 

[D] Institutional special needs plan (I-SNP). 

[E] Chronic condition special needs plan (C-SNP).  

[End of figure]  

A larger proportion of dual-eligible beneficiaries in D-SNPs, as well 
as dual-eligible beneficiaries in FFS, were under age 65 and disabled 
in 2011 compared with those in other MA plans (see figure 2). 
Additionally, similar proportions of dual-eligible beneficiaries in D-
SNPs and dual-eligible beneficiaries in FFS (15 and 16 percent, 
respectively) were diagnosed with a chronic or disabling mental health 
condition such as major depressive disorder or schizophrenia, compared 
with just 10 percent of dual-eligible beneficiaries in other MA plans. 
Among the characteristics in our analysis, the largest difference 
between D-SNPs and other MA plans was the proportion of full-benefit 
beneficiaries in each plan type: 80 percent of dual-eligible 
beneficiaries in D-SNPs and 75 percent of dual-eligible beneficiaries 
in FFS were eligible for full Medicaid benefits, compared with just 34 
percent of dual-eligible beneficiaries in other MA plans. 

Figure 2: Characteristics of Dual-Eligible Beneficiaries in D-SNPs 
Compared with Dual-Eligible Beneficiaries in Other MA Plans and FFS: 

[Refer to PDF for image: horizontal bare graph]  

Percentage of dual-eligible beneficiaries:  

Beneficiary characteristics: Under 65 and disabled[A]; 
Dual-eligible special needs plans: 38%; 
Other MA plans: 26%; 
Medicare fee-for-service: 44%.  

Beneficiary characteristics: Chronic or disabling mental|health 
condition[B]; 
Dual-eligible special needs plans: 15%; 
Other MA plans: 10%; 
Medicare fee-for-service: 16%.  

Beneficiary characteristics: Eligible for full Medicaid benefits; 
Dual-eligible special needs plans: 80%; 
Other MA plans: 34%; 
Medicare fee-for-service: 75 

Beneficiary characteristics: Lived in institution[C]; 
Dual-eligible special needs plans: 4%; 
Other MA plans: 7%; 
Medicare fee-for-service: 12%.  

Beneficiary characteristics: 85 and older; 
Dual-eligible special needs plans: 8%; 
Other MA plans: 13%; 
Medicare fee-for-service: 12%.  

Beneficiary characteristics: Racial/ethnic minority[D]; 
Dual-eligible special needs plans: 60%; 
Other MA plans: 40%; 
Medicare fee-for-service: 41%.  

Beneficiary characteristics: Female; 
Dual-eligible special needs plans: 63%; 
Other MA plans: 66%; 
Medicare fee-for-service: 61%.  

Source: GAO analysis of CMS data.  

Notes: Sources include CMS's 2011 enrollment data, 2011 plan benefit 
package data, and 2010 risk score data. Percentages for demographic 
data are based on about 851,000 dual-eligible beneficiaries in D-SNPs, 
882,000 in other MA, and 7.3 million in FFS as of July 2011. 
Percentages for chronic or disabling mental health conditions are 
based on about 755,000 dual-eligible beneficiaries in D-SNPs, 757,000 
in other MA, and 7.1 million in FFS for whom risk score data were 
available as of July 2010, the most recent year available, and are 
based on 2009 diagnosis data. Actual totals vary for each 
characteristic due to missing data. Data exclude beneficiaries in 
Puerto Rico and those living outside of the 50 states and the District 
of Columbia in 2010. 

[A] Under 65 and disabled does not include beneficiaries with ESRD. 

[B] A chronic and disabling mental health condition is defined as a 
diagnosis of schizophrenia; schizoaffective disorder; or major 
depressive, bipolar, or paranoid disorders. 

[C] Refers to beneficiaries who lived in an institution during July 
2011. 

[D] Racial or ethnic minorities include Black, Hispanic, Asian, and 
North American Native beneficiaries, and beneficiaries who reported 
"Other" race; however, CMS's enrollment data have limitations in 
accurately identifying beneficiary race and ethnicity, resulting in an 
underreporting of Hispanics, Asian/Pacific Islanders, and American 
Indian/Alaskan Natives.  

[End of figure]  

While dual-eligible beneficiaries in D-SNPs were generally similar to 
those in FFS, there were several demographic measures on which dual-
eligible beneficiaries in D-SNPs differed from both those in FFS and 
other MA plans. A smaller proportion of dual-eligible beneficiaries in 
D-SNPs lived in institutions (e.g., nursing facilities, intermediate 
care facilities, or inpatient psychiatric hospitals) in July 2011 
compared with dual-eligible beneficiaries in FFS, and, to a lesser 
extent, other MA plans. D-SNPs also enrolled a smaller proportion of 
dual-eligible beneficiaries who were 85 or older compared with the 
other plan types, as well as a larger proportion of beneficiaries who 
were racial or ethnic minorities.[Footnote 31] 

Health Status of Dual-Eligible Beneficiaries in D-SNPs Was Similar to 
That of Dual-Eligible Beneficiaries in FFS and Other MA Plans: 

Dual-eligible beneficiaries in D-SNPs had very similar health status 
as measured by their 2010 risk scores, the year for which the most 
recent data were available, when compared with dual-eligible 
beneficiaries in FFS and other MA plans.[Footnote 32] As shown in 
figure 3, the average risk score--which predicts Medicare costs--of 
dual-eligible beneficiaries in D-SNPs (1.29) was similar to the 
average scores for dual-eligible beneficiaries in FFS (1.35) and other 
MA plans (1.34).[Footnote 33] Just under 20 percent of dual-eligible 
beneficiaries in D-SNPs and dual-eligible beneficiaries in FFS and 
other MA plans were expected to cost Medicare at least twice as much 
as the average Medicare FFS beneficiary, and less than 10 percent in 
each plan type were expected to cost at least three times the average. 

Figure 3: Average Risk Scores of Dual-Eligible Beneficiaries by 
Characteristic and Plan Type, 2010: 

[Refer to PDF for image: vertical bar graph]  

Beneficiary characteristic: All dual-eligible beneficiaries; 
Average risk score: 
Dual-eligible special needs plans: 1.29; 
Other MA plans: 1.34; 
Medicare fee-for-service: 1.35.  

Beneficiary characteristic: Under 65 and disabled; 
Average risk score: 
Dual-eligible special needs plans: 1.10; 
Other MA plans: 1.13; 
Medicare fee-for-service: 1.11.  

Beneficiary characteristic: 85 and older; 
Average risk score: 
Dual-eligible special needs plans: 1.64; 
Other MA plans: 1.63; 
Medicare fee-for-service: 1.68.  

Beneficiary characteristic: Eligible for full Medicaid benefits; 
Average risk score: 
Dual-eligible special needs plans: 1.28; 
Other MA plans: 1.50; 
Medicare fee-for-service: 1.39.  

Beneficiary characteristic: Eligible for Medicaid premium and cost-
sharing assistance; 
Average risk score: 
Dual-eligible special needs plans: 1.34; 
Other MA plans: 1.23; 
Medicare fee-for-service: 1.24.  

Beneficiary characteristic: Chronic or disabling mental health 
condition[A]; 
Average risk score: 
Dual-eligible special needs plans: 1.64; 
Other MA plans: 1.78; 
Medicare fee-for-service: 1.72.  

Beneficiary characteristic: Racial/ethnic minority[B]; 
Average risk score: 
Dual-eligible special needs plans: 1.21; 
Other MA plans: 1.24; 
Medicare fee-for-service: 1.29.  

Beneficiary characteristic: Female; 
Average risk score: 
Dual-eligible special needs plans: 1.31; 
Other MA plans: 1.32; 
Medicare fee-for-service: 1.38.  

Beneficiary characteristic: Living long-term in institution[C]; 
Average risk score: 
Dual-eligible special needs plans: 1.95; 
Other MA plans: 1.98; 
Medicare fee-for-service: 2.00.  

Source: GAO analysis of CMS data, 

Notes: Sources include CMS's 2010 enrollment data, plan benefit 
package data, and risk score data. Data exclude new enrollees, 
beneficiaries with ESRD, and those living in Puerto Rico and outside 
of the 50 states and the District of Columbia. We adjusted all average 
risk scores to account for CMS's normalization factor of 1.041 in 
2010; additionally, we adjusted average risk scores of D-SNPs and 
other MA plans downward by 3.41 percent to account for diagnostic 
coding differences between MA plans and FFS. 

[A] A chronic or disabling mental health condition is defined as a 
diagnosis of schizophrenia; schizoaffective disorder; or major 
depressive, bipolar, or paranoid disorders. 

[B] Racial or ethnic minorities include Black, Hispanic, Asian, and 
North American Native beneficiaries, and beneficiaries who reported 
"Other" race; however, CMS's enrollment data have limitations in 
accurately identifying beneficiary race and ethnicity, resulting in an 
underreporting of Hispanics, Asian/Pacific Islanders, and American 
Indian/Alaskan Natives. 

[C] Refers to beneficiaries who were considered to be "long-term 
institutional" (i.e., living in an institution for at least 3 months) 
as of July 2010.  

[End of figure]  

The average risk scores of dual-eligible beneficiaries in each plan 
type also generally were similar across demographic groups. However, 
there were some differences between plan types. Dual-eligible 
beneficiaries in D-SNPs who were eligible for full Medicaid benefits 
had lower average risk scores than those of similar beneficiaries in 
other MA plans and FFS.[Footnote 34] Dual-eligible beneficiaries in D-
SNPs with a chronic or disabling mental health condition also had 
lower average risk scores than those in other MA plans. Additionally, 
partial-benefit dual-eligible beneficiaries in D-SNPs had higher 
average risk scores than their counterparts in other MA plans and FFS. 

D-SNPs Typically Offer Fewer Supplemental Benefits Than Other MA 
Plans, and Information on Planned Services Is Not Comparable: 

Dual-eligible beneficiaries in D-SNPs do not necessarily get more 
benefits than those in other MA plans, although D-SNP representatives 
told us their care coordination services are more comprehensive than 
other MA plans. D-SNPs and other MA plans varied in how frequently 
they offered supplemental benefits--benefits not covered by FFS--and 
MA plans offered more of these supplemental benefits than D-SNPs. 
While the models of care we reviewed described in varying detail how 
the D-SNPs plan to provide other services, such as health risk 
assessments, to beneficiaries, most D-SNPs did not provide--and are 
not required to provide--estimates of the number of dual-eligible 
beneficiaries that would receive the services. 

D-SNPs Offered Fewer of Their 10 Most-Common Supplemental Benefits 
Than Other MA Plans: 

D-SNPs provide fewer supplemental benefits, on average, than other MA 
plans. Of the 10 supplemental benefits offered by more than half of D-
SNPs, 7 were offered more frequently by other MA plans and 3 were 
offered more frequently by D-SNPs.[Footnote 35] (See figure 4.) These 
3 supplemental benefits were offered much more frequently by D-SNPs 
compared to other MA plans: they offered dental benefits one-and-a-
half times more often, over-the-counter drugs nearly twice as often, 
and transportation benefits almost three times more often. However, a 
smaller proportion of D-SNPs compared to other MA plans offered 
hearing benefits, as well as benefits for certain inpatient settings 
and outpatient services. For some of the services D-SNPs offered less 
frequently, dual-eligible beneficiaries may receive some coverage 
through Medicaid. In addition, according to CMS, some of the benefits 
offered more frequently by other MA plans (e.g. international 
outpatient emergency) are not necessarily as useful a benefit for D-
SNPs.[Footnote 36] 

Figure 4: Supplemental Benefits Offered by More Than Half of Dual-
Eligible Special Needs Plans (D-SNP) Compared with Other Medicare 
Advantage (MA) Plans, 2012: 

[Refer to PDF for image: vertical bar graph]  

Percentage of plans:  

Type of supplemental benefit: Vision[A]; 
Dual-Eligible Special Needs Plans: 89%; 
Other MA Plans: 92%.  

Type of supplemental benefit: Preventive health care[B]; 
Dual-Eligible Special Needs Plans: 87%; 
Other MA Plans: 97%.  

Type of supplemental benefit: Dental[C]; 
Dual-Eligible Special Needs Plans: 77%; 
Other MA Plans: 46%.  

Type of supplemental benefit: Transportation[D]; 
Dual-Eligible Special Needs Plans: 68%; 
Other MA Plans: 24%.  

Type of supplemental benefit: International outpatient emergency[E]; 
Dual-Eligible Special Needs Plans: 67%; 
Other MA Plans: 96%.  

Type of supplemental benefit: Over-the-counter drugs[F]; 
Dual-Eligible Special Needs Plans: 65%; 
Other MA Plans: 33%.  

Type of supplemental benefit: Outpatient hospital blood[G]; 
Dual-Eligible Special Needs Plans: 57%; 
Other MA Plans: 84%.  

Type of supplemental benefit: Hearing[H]; 
Dual-Eligible Special Needs Plans: 55%; 
Other MA Plans: 64%.  

Type of supplemental benefit: Skilled nursing facility[I]; 
Dual-Eligible Special Needs Plans: 52%; 
Other MA Plans: 67%.  

Type of supplemental benefit: Inpatient hospital acute[J]; 
Dual-Eligible Special Needs Plans: 52%; 
Other MA Plans: 86%.  

Notes: Source data are for the 2012 contract year. The following types 
of MA plans are included: health maintenance organizations, local 
preferred provider organizations, regional preferred provider 
organizations, private fee-for-service plans, and provider-sponsored 
organizations. We excluded plans that charged a Part C premium. 

[A] Vision benefits can include coverage for routine eye exams, 
contact lenses, or eyeglasses (lenses and frames). 

[B] Preventive health care benefits can include screenings and 
immunizations beyond what Medicare fee-for-service (FFS) covers, as 
well as health education and fitness club membership. 

[C] Dental benefits can include oral exams, teeth cleanings, fluoride 
treatments, dental X-rays, or emergency dental services. 

[D] Transportation benefits can include travel from a beneficiary's 
home to medical appointments. 

[E] International outpatient emergency benefits can include additional 
services beyond what Medicare FFS covers. 

[F] Over-the-counter drug coverage can include nonprescription 
medicines not covered under Medicare Part D. 

[G] Outpatient blood benefits can include units of blood received as 
an outpatient or as part of a Part B-covered service for the first 
three units not covered by Medicare FFS. 

[H] Hearing benefits can include coverage for hearing tests, hearing 
aid fittings, and hearing aid evaluations. 

[I] Skilled nursing facility benefits can include waiving the 3-day 
inpatient hospital stay requirement in Medicare FFS. 

[J] Inpatient hospital acute benefits can include additional days 
beyond what Medicare FFS covers.  

[End of figure]  

For the three most-common D-SNP supplemental benefits--vision, 
prevention, and dental--we analyzed the individual services covered 
under these benefits and found that D-SNPs' vision and dental benefits 
were generally more comprehensive than those offered by other MA 
plans. For example, in their vision benefit, a larger proportion of D-
SNPs compared to other MA plans covered contact lenses and eyeglasses. 
In addition, a larger proportion of D-SNPs compared to other MA plans 
included in their dental benefit coverage of oral surgery, 
extractions, and restorative services. However, D-SNPs were less 
likely than other MA plans to include membership in health clubs as 
part of their preventive health care benefits. 

Despite offering these supplemental benefits somewhat less often than 
other MA plans, D-SNPs allocated a larger percentage of their rebates 
to supplemental benefits than other MA plans. (See table 2.) They were 
able to do so largely because they allocated a smaller percentage of 
rebates to reducing cost-sharing. Most dual-eligible beneficiaries 
will have their cost-sharing covered by Medicaid, so D-SNPs have less 
need than other MA plans to cover cost-sharing. We also found that D-
SNPs tended to receive smaller rebates than other MA plans ($70 per 
member per month on average compared to $108). 

Table 2: Percentage of Rebate Allocated to Supplemental Benefits, 
Reducing Cost Sharing, and Premium Reduction, 2012: 

Supplemental benefits[A]; 
D-SNPs: 51%; 
Other MA: 14%. 

Reduce A/B cost share[B]; 
D-SNPs: 19%; 
Other MA: 53%. 

Part B premium buy down[C]; 
D-SNPs: 1%; 
Other MA: 6%. 

Part D basic premium buy down[D]; 
D-SNPs: 19%; 
Other MA: 11%. 

Part D supplemental premium buy down[E]; 
D-SNPs: 10%; 
Other MA: 15%. 

Source: GAO analysis of CMS data. 

Notes: Data are from CMS's 2012 bid pricing tool. The following types 
of MA plans are included: health maintenance organizations, local 
preferred provider organizations, regional preferred provider 
organizations, private fee-for-service plans, and provider-sponsored 
organizations. We excluded plans that charged a Part C premium. 
Numbers may not add up to 100 percent due to rounding. 

[A] MA plans may use their rebates to offer supplemental benefits, 
which are benefits beyond the Medicare FFS package. 

[B] MA plans may use their rebates to reduce cost-sharing for Parts A 
and B. 

[C] Beneficiaries in MA plans continue to pay the Part B premium; 
however, MA plans may use their rebates to reduce beneficiaries' Part 
B premiums. 

[D] MA plans may use their rebates to reduce the premium for the basic 
Part D drug benefit. 

[E] MA plans may use their rebates to reduce supplemental premiums for 
drug benefits beyond the basic Part D package.  

[End of table]  

Models of Care Describe How D-SNPs Plan to Provide Specialized Health 
Care Services, but CMS Lacks Standard Measures Needed for Evaluation: 

Although the 15 models of care we reviewed described the types of 
services D-SNPs intended to provide, D-SNPs generally did not state in 
their models of care how many of their enrolled beneficiaries were 
expected to receive these services. The criteria D-SNPs are evaluated 
on in the approval process emphasize the inclusion of in-depth 
descriptions and case studies rather than details about how many 
beneficiaries would likely receive these services, for example, the 
number of beneficiaries that will use additional services targeted to 
the most vulnerable. CMS does not require D-SNPs to report that 
information in the models of care, although such information could be 
useful for future evaluations of whether D-SNPs met their intended 
goals,[Footnote 37] as well as for comparisons among D-SNPs.[Footnote 
38] Three D-SNPs we interviewed told us that a lack of specificity in 
the model-of-care scoring criteria confused some D-SNPs; having more 
specific scoring criteria may also eliminate some uncertainty in the 
approval process. 

Knowing the extent of the special services D-SNPs expect to provide 
would assist future evaluations of whether they met their goals, but 
most models of care did not include this information. For example, all 
15 D-SNPs stated in their models of care that they planned to conduct 
health risk assessments for beneficiaries within 90 days of enrollment 
and an annual reassessment, as they are required to do by CMS. 
However, only 4 provided information on how many members had actually 
completed a health risk assessment or reassessment in prior years, 
with cited completion rates for 2010 ranging from 52 to 98 percent. In 
addition, none of the D-SNPs we reviewed indicated in their models of 
care how many beneficiaries were expected to receive add-on services 
such as social support services that were intended for the most-
vulnerable beneficiaries. The models of care we reviewed did include, 
as required in the model-of-care scoring criteria, information about 
how the D-SNP identifies the most vulnerable beneficiaries in the plan 
and the add-on services and benefits that would be delivered to these 
beneficiaries. D-SNPs' models of care described a variety of methods 
used to identify these beneficiaries: health risk assessments (10 D-
SNPs); provider referrals (9); and hospital admissions or discharges 
(7). However, the models of care generally did not indicate how many 
or what proportion of beneficiaries were expected to be among the most 
vulnerable, although one D-SNP's model of care stated that complex-
care patients constituted over one-third of its membership. 
Furthermore, it was sometimes unclear whether the services described 
as targeted to these beneficiaries were in addition to those available 
to all dual-eligible beneficiaries in the D-SNP. D-SNPs also described 
the services that they plan to offer to the most-vulnerable 
beneficiaries, the most frequent being complex/intensive case 
management (6 D-SNPs). Other services D-SNPs planned to offer the most-
vulnerable beneficiaries included 24-hour hotlines, social support 
services, and supplemental benefits beyond what was planned to be 
offered to all dual-eligible beneficiaries in the D-SNP. 

CMS guidance also requires D-SNPs to describe how they intend to 
evaluate their performance and measure outcomes in achieving goals 
identified in their models of care, but CMS does not stipulate the use 
of standard outcome or performance measures in the model of care, such 
as measures of patient health status and cognitive functioning. As a 
result, it would be difficult for CMS to use any data it might collect 
on these measures to compare D-SNPs' effectiveness or evaluate how 
well they have done in meeting their goals. Furthermore, without 
standard measures, it would not be possible for CMS to fully evaluate 
the relative performance of D-SNP models of care. While it is not 
required, an evaluation of D-SNPs could both help to improve the D-SNP 
program and inform other initiatives to better coordinate care for 
dual-eligible beneficiaries. The models of care we reviewed had little 
uniformity in the measures plans selected. Four D-SNPs discussed their 
approach to performance and health outcome measurement largely in 
general terms, such as describing which datasets they would use or the 
categories of outcomes that would be measured. The other 11 D-SNPs 
provided specific measurements, which included measuring items such as 
readmissions, emergency room utilization, and receipt of follow-up 
calls after inpatient stays. Were CMS to move to a standard set of 
performance and outcome measures, it could be less burdensome and no 
more costly than what some D-SNPs currently collect. Using standard 
measures could also streamline the models-of-care review process. 

D-SNPs Reported More-Comprehensive Care Coordination Services and 
Greater Beneficiary Interaction Than Other MA Plans: 

Of the 15 D-SNPs we interviewed, 9 were in organizations that offered 
both D-SNPs and other MA plans, and representatives from 7 of those D-
SNPs told us that their care coordination services are different from 
those in their organization's other MA plan offerings. For example, a 
representative from one D-SNP told us that while care coordination and 
case management were available in both types of plans offered by that 
organization, dual-eligible beneficiaries in the D-SNP are 
continuously enrolled in case management, whereas dual-eligible 
beneficiaries in other MA plans who need these services receive them 
for only a limited time. A representative from another D-SNP said that 
the plan provides care coordination services similar to those of other 
MA plans offered by its organization but that dual-eligible 
beneficiaries in the D-SNP who need these services are identified 
faster than are dual-eligible beneficiaries in the other MA plans. A 
representative of a third D-SNP said it has a community resource unit 
that is not available in other MA plans offered by its organization, 
which works with local agencies such as long-term care providers and 
adult protective services. 

Multiple representatives of the 15 D-SNPs we interviewed described 
their care coordination services as being "high touch"--meaning that 
the plans, particularly the case managers, have frequent interaction 
with dual-eligible beneficiaries in the D-SNP. For example, 
representatives from one D-SNP told us that its plan includes in-
person meetings with case managers. Representatives from another D-SNP 
described several specific examples of care coordination successes, 
such as when a case manager followed up on a beneficiary's Medicaid 
reenrollment application to ensure that the beneficiary did not lose 
eligibility, and another situation in which a case manager worked with 
the complex social and housing needs of a beneficiary who had both 
physical and mental health issues. Representatives from a third D-SNP 
noted that they have providers who conduct home visits to help prevent 
hospitalization. 

Although Some D-SNP Contracts with State Medicaid Agencies Expressly 
Provide for Benefit Integration or Care Coordination, Most Do Not: 

CMS stated that contracts between D-SNPs and state Medicaid agencies 
are an opportunity to increase benefit integration and care 
coordination.[Footnote 39] However, only about one-third of the 2012 
contracts we reviewed contained any provisions expressly providing for 
D-SNPs to deliver Medicaid benefits, thereby achieving benefit 
integration. Only about one-fifth of the contracts expressly provided 
for active care coordination between D-SNPs and Medicaid agencies, 
which indicates that most care coordination was done exclusively by D-
SNPs, without any involvement of state Medicaid agencies. Further, D-
SNP representatives and state Medicaid officials expressed concerns 
about resources needed to contract with D-SNPs, and uncertainty about 
the future of D-SNPs. 

Some Contracts Include Provisions to Address the Integration of 
Benefits or Meaningful Coordination of Services between D-SNPs and 
Medicaid Agencies, but Most Do Not: 

Benefit Integration: 

The 2012 D-SNP contracts with state Medicaid agencies we reviewed 
varied considerably in their provisions for integration of benefits 
and state payments to D-SNPs for covering specific services. According 
to CMS, "[t]his variability is to be expected, as States and MA 
organizations can develop agreements for [D-SNPs] to assume 
responsibility for providing or arranging for a wide range of Medicaid 
services based on each State's ability and interest in integrating its 
Medicaid program with Medicare via a SNP."[Footnote 40] Thirty-three 
percent of the 124 D-SNP contracts with state Medicaid agencies for 
2012 that we reviewed expressly provided for the delivery of at least 
some portion of Medicaid benefits, thereby integrating Medicare and 
Medicaid benefits. The contracts varied in the extent of the Medicaid 
benefits for which a plan was responsible. About 10 percent provided a 
limited number of Medicaid services, such as dental or vision 
benefits. In contracts where there was some integration of Medicare 
and Medicaid benefits, states contracted for the different services, 
making comparisons among the contracts difficult. Of the 23 percent 
that integrated most or all Medicaid benefits, 64 percent of D-SNPs 
provided all Medicaid benefits, including long-term care support 
services in community settings and institutional care; 25 percent 
provided most Medicaid benefits, including long-term support services 
in community settings but not institutional care; and 11 percent 
provided most Medicaid benefits but did not provide any long-term 
support services or institutional care. (See figure 5.)  

Figure 5: Dual-Eligible Special Needs Plans (D-SNP) Integration of 
Medicaid Benefits, 2012: 

[Refer to PDF for image: pie-chart]  

No benefits: 67%; 
Some benefits: 10%; 
Most benefits: 23%; of that: 
No long-term support services: 11%; 
Community long-term support services: 25%; 
Community long-term support services and institutional care:
64%.  

Source: GAO analysis of D-SNP/state contracts.  

Note: Percentages are based on 124 D-SNP/state Medicaid agency 
contracts for 2012 that were reviewed by GAO.  

[End of figure]  

Sixty-seven percent of contracts between D-SNPs and state Medicaid 
agencies did not expressly provide for D-SNPs to cover Medicaid 
benefits. To carry out MIPPA's requirement that each D-SNP contract 
provide or arrange for Medicaid benefits to be provided, CMS guidance 
has required that contracts list the Medicaid benefits that dual-
eligible beneficiaries could receive directly from the state Medicaid 
agency or the state's Medicaid managed care contractor(s).[Footnote 41] 

For D-SNPs contracting with state Medicaid agencies to provide all or 
some Medicaid benefits, the capitated payment reflected variation in 
coverage and conditions. One state that contracts for all Medicaid 
benefits except a limited number of services including long-term care 
services paid the D-SNP at a rate of $423 per member per month. 
Another state, which contracted for a limited number of benefits, 
including Medicare-excluded drugs, expanded dental coverage, and case-
management services, paid the D-SNP $132 per member per month. This 
state's Medicaid agency retained responsibility for inpatient hospital 
services and long-term care coverage. Some contracts, rather than 
stating a single capitation rate, gave payment rates for different 
categories, including risk or acuity level, beneficiary age, and 
service location, as well as whether the beneficiary was designated as 
nursing home eligible and whether services for these beneficiaries 
were provided in the community or facility setting. Within one state, 
payment rates ranged from just under $170 per month for dual-eligible 
beneficiaries who were neither nursing home eligible nor had a chronic 
mental health condition and were living in the community to over 
$8,600 per month for dual-eligible beneficiaries residing in a nursing 
facility and requiring the highest level of care. 

Some of the 2012 contracts providing for payments from state Medicaid 
agencies to D-SNPs did not address the direct provision of benefits, 
often providing for payments to the D-SNP for assuming the state's 
responsibility for paying dual-eligible beneficiaries' Medicare 
copayments, coinsurance, and deductibles. These payments ranged from 
$10 to $60 per member per month. 

Coordination of Services: 

While all contracts between D-SNPs and state Medicaid agencies for 
2012 provided for some level of care coordination to beneficiaries, 
approximately 19 percent expressly provided for active coordination of 
beneficiary services between the D-SNP and the state Medicaid agency. 
[Footnote 42] Most active coordination occurs when dual-eligible 
beneficiaries transition between care settings or between Medicare and 
Medicaid. Thirteen percent of all contracts contained provisions 
requiring D-SNPs and the state Medicaid agency to coordinate the 
transition of beneficiaries between care settings (such as hospital to 
nursing home) within a given time frame. For example, one state's D-
SNP contracts directed the plans to notify the Medicaid service 
coordinators or agency caseworker, as applicable, no later than 5 
business days after a dual-eligible beneficiary had been admitted to a 
nursing facility. The other 6 percent of contracts included provisions 
for providing different coordination activities such as requiring the 
plan to work with Medicaid staff to coordinate delivery of wrap-around 
Medicaid benefits. 

The remaining 81 percent of all contracts did not specifically address 
D-SNPs' coordination with state Medicaid staff, such as case managers. 
Rather, these contracts indicated that the D-SNP would coordinate 
Medicaid and Medicare services but did not specify the role of the 
state Medicaid agency in coordinating those services. Because D-SNPs 
are required by Medicare to provide care coordination services to dual-
eligible beneficiaries, these services are often provided without 
reimbursement or payment from the state Medicaid agency. 

D-SNP and State Officials Expressed Some Concerns about Contracting: 

Concerns about Resource Investment: 

D-SNP representatives and state Medicaid officials we spoke with 
reported that contract development and submission to CMS are resource-
intensive. State officials reported that because they had limited 
resources, they needed to balance the benefits of the contract with 
the time and resources needed to develop and oversee it. As one state 
Medicaid official said, the state "bandwidth"--resources--was a 
challenge, and she was concerned about contracting with the large 
number of D-SNPs in her state. This official added that the state did 
not want to be in the position of making contractual commitments that 
could not be honored because of limited funds or other resources. In 
contrast, the plan representatives we interviewed expressed interest 
in continuing to operate D-SNPs and were therefore eager to contract 
with states despite any challenges that might exist. Beginning in 
2013, D-SNPs will not be permitted to operate without state contracts. 

Representatives from 12 of the 15 plans and officials from 3 of the 5 
state Medicaid agencies we spoke with pointed out that establishing a 
contract between Medicaid and a Medicare plan highlights conflicts 
between federal and state requirements. A representative from one D-
SNP told us that it was challenging for plans and state Medicaid 
agencies to agree about the characterization of dual-eligible 
beneficiaries because Medicare and some states have different 
definitions. Officials from one state Medicaid agency and D-SNP 
representatives reported difficulty reconciling the difference between 
the Medicare contracting cycle, which is based on the calendar year, 
and the fiscal year contracting cycle for their states. They reported 
that, if a contract would not cover the entire calendar year, CMS 
would not approve it. In one case, a state Medicaid official reported 
that CMS's deadline of July 1, 2012, for 2013 contracts would occur 
before the state signed contracts for 2013. Sometimes non-Medicaid 
structures conflict with CMS's contracting requirements for D-SNPs. A 
representative of one D-SNP told us that Medicaid benefits for 
individuals with developmental disabilities were managed through a 
contract with the state's family services agency, not the state 
Medicaid agency. Therefore, to provide services to this population, 
the D-SNP had to become a subcontractor to the family services agency. 
The official said that the D-SNP and the state need to work with CMS 
to develop a subcontracting relationship that is acceptable.[Footnote 
43] State Medicaid officials and D-SNP representatives reported that 
they did not always have the resources or the administrative ability 
to resolve these types of issues before entering into a contract. 

Beginning in 2013, D-SNPs must secure a contract with the state 
Medicaid agency in each state in their service area. To do this, D-
SNPs may need to establish new relationships with state officials who, 
according to the D-SNP representatives we interviewed, sometimes have 
very limited knowledge of Medicare and its requirements. However, some 
states have experience with Medicaid managed care and in some cases, D-
SNP representatives had previously worked with the state on Medicaid 
contracts, thereby somewhat easing the transition to working with D-
SNPs. 

Uncertainty about D-SNPs and Other Initiatives for Dual-Eligible 
Beneficiaries: 

Plan representatives and state Medicaid officials told us that 
uncertainty about the future made them cautious in contracting. 
Authority for SNPs to restrict enrollment to special needs populations 
(such as dual-eligible beneficiaries) currently expires at the end of 
2013; SNPs may not continue as a unique type of MA plan if Congress 
does not extend this authority. Were this to occur, states would lose 
any advantages they might have gained from investing their resources 
to work with D-SNPs to integrate benefits and coordinate care. 
Furthermore, uncertainty regarding the future of D-SNPs creates 
uncertainty for the states about how to continue to serve dual-
eligible beneficiaries currently enrolled in D-SNPs. 

Uncertainty about the implementation of state demonstrations under the 
CMS initiative to align Medicare and Medicaid services--financial 
alignment initiative--has made some states hesitant to enter into 
contracts with D-SNPs. As of June 2012, all proposals had been made 
available for public comment but CMS had not finalized agreements with 
the states. Medicaid officials from two states told us that if their 
proposed financial alignment demonstrations were implemented, D-SNPs 
in their states would cease to exist.[Footnote 44] Some states were 
moving forward with D-SNP contracts while concurrently preparing to 
shift D-SNPs to a different type of managed care plan if their 
demonstration proposal is implemented. However, officials from one 
state told us that they did not have sufficient clarity about the 
direction of the state Medicaid program in relation to its proposed 
demonstration to enter into contracts. Even in those states where 
demonstrations would not eliminate D-SNPs, contracting challenges as 
well as potential financial incentives associated with the 
demonstrations from the financial alignment initiatives create 
disincentives for states to work with D-SNPs outside of the financial 
alignment initiatives and, therefore, leaves the future of D-SNPs in 
question in these states as well. 

Conclusions: 

D-SNPs have the potential to help beneficiaries who are eligible for 
both Medicare and Medicaid navigate these two different systems and 
receive the health services that meet their individual needs. However, 
CMS has not required D-SNPs to report information that is critical to 
better holding plans accountable and determining whether they have 
realized their potential. Although the models of care D-SNPs must 
submit to CMS generally state what these plans intend to do, they do 
not all report the number of services they intend to provide. For 
example, plans are not required to report the number of enrollees they 
expect to designate as most vulnerable, or how many and which 
additional services they will provide to these enrollees. Although D-
SNPs are required to collect performance and outcome measures, they 
are not required to use standard measures such as existing measures of 
hospital readmission or patient health status and cognitive 
functioning. Further they are not required to report these measures to 
CMS, and, lacking standard measures, it would in any case be difficult 
to compare D-SNPs' effectiveness. Standardizing these measures should 
have a minimal effect on D-SNPs' administrative efforts, because 
additional measures could replace some or all of the measures 
currently used as well as much of the narrative in models of care. 
Standardizing measures could also reduce CMS's administrative efforts 
by streamlining review of D-SNPs. Additional standardized information 
would allow CMS to meet its goals for accountability for effective and 
efficient use of resources. 

Further, CMS has neither evaluated the sufficiency and appropriateness 
of the care that D-SNPs provide nor assessed their effectiveness in 
integrating benefits and coordinating care for dual-eligible 
beneficiaries. Nonetheless, CMS is embarking on a new demonstration in 
up to 26 states with as many as 2 million beneficiaries to financially 
realign Medicare and Medicaid services so as to serve dual-eligible 
beneficiaries more effectively. If CMS systematically evaluates D-SNP 
performance, it can use information from the evaluation to inform the 
implementation and reporting requirements of this major new initiative. 

Recommendations for Executive Action: 

To increase D-SNPs' accountability and ensure that CMS has the 
information it needs to determine whether D-SNPs are providing the 
services needed by dual-eligible beneficiaries, especially those who 
are most vulnerable, the Administrator of CMS should take the 
following four actions: 

* require D-SNPs to state explicitly in their models of care the 
extent of services they expect to provide, to increase accountability 
and to facilitate evaluation; 

* require D-SNPs to collect and report to CMS standard performance and 
outcome measures to be outlined in their models of care that are 
relevant to the population they serve, including measures of 
beneficiary health risk, beneficiary vulnerability, and plan 
performance; 

* systematically analyze these data and make the results routinely 
available to the public; and: 

* conduct an evaluation of the extent to which D-SNPs have provided 
sufficient and appropriate care to the population they serve, and 
report the results in a timely manner. 

Agency and Other External Comments and Our Evaluation: 

We obtained comments on a draft of this report from HHS and the SNP 
Alliance, which represents 32 companies that offer more than 200 SNPs. 
CMS provided written comments, which are reprinted in appendix I, and 
technical comments that we incorporated where appropriate. 
Representatives from the SNP Alliance provided us with oral comments. 

HHS Comments: 

HHS concurred with our recommendation that CMS should require plans to 
explicitly state in their models of care the extent of services they 
expect to provide, and agrees that information about the extent to 
which D-SNPs provide certain services would increase accountability 
and facilitate evaluation. HHS also stated that CMS recently began to 
collect information on the completion of health risk assessments, but 
has not made it public because the information is relatively new. HHS 
did question the usefulness of quantifying the number of members 
expected to receive services described in the documents, stating that 
the model of care is a framework for indicating how the SNP proposes 
to coordinate the care of SNP enrollees. However, as we noted in the 
draft report, we believe such information could be useful in later 
evaluating whether D-SNPs met their intended goals. 

HHS also concurred with our recommendation that CMS should require D-
SNPs to collect and report standard measures relevant to the 
populations they serve, and stated that CMS is working to create new 
measures that will be relevant to dual-eligible beneficiaries in D-
SNPs. HHS also stated that CMS currently collects a broad range of 
standard quality measures, including HEDIS, as well as structure and 
process measures. HHS included in its response a recent Health Plan 
System Management memorandum that CMS sent to MA organizations, 
including D-SNPs, which outlined updated reporting requirements for 
2013. HHS also noted that in addition to the data it currently 
collects, CMS requires D-SNPs to conduct both a Quality Improvement 
Project and a Chronic Care Improvement Project, and asked GAO to note 
this in the final report. We did not include these because, as we 
noted in the draft report, quality and quality measures were not in 
the scope of our work. 

HHS also concurred with our other two recommendations. 

SNP Alliance Comments: 

First, SNP Alliance representatives stated that the benefits D-SNPs 
provide most frequently are more meaningful to dual-eligible 
beneficiaries than some of the supplemental benefits provided more 
frequently by other MA plans. We note in the report that some of the 
supplemental benefits offered at lower rates by D-SNPs may be covered 
by Medicaid, thereby reducing the need for them to be covered by D-
SNPs. Second, SNP Alliance representatives were concerned with our 
definition of FIDESNPs. They explained that CMS's definition, which we 
used, limits FIDESNPs to those that integrate all Medicare and 
Medicaid benefits without any limits, such as the number of nursing 
home days covered. They contended that some D-SNPs may be considered 
fully integrated even though they do not include all benefits, such as 
nursing home care, and may have some limits. However, in reporting on 
CMS activities we have no basis for using different definitions than 
those formally applied by the agency. Third, SNP Alliance 
representatives stated that the ability of their D-SNP members to 
fully integrate benefits through contracting is limited by the 
capacity and interest of state Medicaid agencies. We note in the 
report that state Medicaid agencies we interviewed acknowledged 
limitations in their capacity for contracting. Fourth, SNP Alliance 
representatives had some concern with our emphasis on estimating how 
many beneficiaries are expected to receive the services described in 
the model of care, stating that all dual-eligible beneficiaries would 
have access to the services described based on need. However, as we 
stated in the draft report, information is not generally available on 
the number of beneficiaries that use these benefits. Finally, SNP 
Alliance representatives were supportive of the state demonstrations 
under the financial alignment initiative. They noted that D-SNPs are 
being used as a platform for half of the state demonstrations, with 
the remainder being based on a Medicaid model. They considered the 
adoption of the D-SNP model by many of the state demonstrations as 
evidence of D-SNPs' success. 

SNP Alliance representatives generally agreed with our 
recommendations. They said that they support better aligning reporting 
requirements with the models of care, and stated that D-SNPs need a 
set of core measures that are most relevant to the dual-eligible 
population they serve. 

As agreed with your offices, unless you publicly announce the contents 
of this report earlier, we plan no further distribution until 30 days 
from the report date. At that time, we will send copies to the 
Secretary of HHS and to interested congressional committees. In 
addition, the report will be available at no charge on the GAO website 
at [hyperlink, http://www.gao.gov]. 

If you or your staff have any questions about this report, please 
contact me at (202) 512-7114 or CosgroveJ@gao.gov. Contact points for 
our Offices of Congressional Relations and Public Affairs may be found 
on the last page of this report. GAO staff who made major 
contributions to this report are listed in appendix II.  

Signed by:  

James C. Cosgrove: 
Director, Health Care: 

[End of section] 

Appendix I: Comments from the Department of Health & Human Services: 

Department of Health & Human Services: 
Office of The Secretary: 
Assistant Secretary for Legislation: 
Washington, DC 20201:  

August 27, 2012:  

James Cosgrove: 
Director, Health Care: 
U.S. Government Accountability Office: 
441 G Street, NW: 
Washington, DC 20548:  

Dear Mr. Cosgrove:  

Attached are comments on the U.S. Government Accountability Office's 
(GAO) report entitled, "Medicare Special Needs Plans: CMS Should 
Improve Information Available about Dual-Eligible Plans' Performance" 
(GAO-12-864).  

The Department appreciates the opportunity to review this report prior 
to publication.  

Sincerely,  

Signed by:  

Jim R. Esquea: 
Assistant Secretary for Legislation:  

Attachment:  

General Comments Of The Department Of Health And Human Services (HHS) 
On The Government Accountability Office's (GAO) Draft Report Entitled: 
"Medicare Special Needs Plans: CMS Should Improve Information 
Available About Dual Eligible Plans' Performance" (GAO-12-864):  

The Department appreciates the opportunity to review comment on this 
draft report.  

GAO was asked to analyze the characteristics of dual eligible 
beneficiaries in Special Needs Plans that serve dual eligibles (D-
SNPs) and other Medicare Advantage (MA) plans, review the differences 
in specialized services between D-SNPs and other MA plans, and review 
how DSNPs work with state Medicaid agencies to enhance benefit 
integration and care coordination. GAO found that, in order to 
increase D-SNP accountability, the Centers for Medicare & Medicaid 
Services (CMS) should improve D-SNP reporting of services provided to 
dual eligible beneficiaries and make this information available to the 
public.  

HHS concurs with GAO's recommendations, and the agency is either 
implementing or currently taking steps toward implementing three of 
the four recommended actions, as described further below. We 
respectfully request that the report be revised to include the 
information summarized below, so as to provide a more complete and 
accurate picture of CMS's oversight of D-SNPs and the agency's efforts 
to evaluate the extent to which these plans have provided appropriate 
care to dual eligibles.  

GAO Recommendation:  

CMS should require D-SNPs to state explicitly in their models of care 
the extent of services they expect to provide to increase 
accountability and to facilitate evaluation.  

HHS Response:  

HHS concurs with GAO's recommendation and agrees that information 
about the extent to which D-SNPs provide certain services would 
increase accountability and facilitate evaluation. Currently, CMS 
collects some of the information referenced in the report. For 
example, the draft report states (on page 21) that, in its review of 
the models of care (MOCs) of 15 D-SNPs, GAO found that the D-SNPs 
generally did not state how many of their enrolled beneficiaries were 
expected to receive certain services. As an example, while all of the 
15 D-SNPs stated that they planned to conduct health risk assessments 
(HRAs) within 90 days of enrollment, and an annual reassessment, the D-
SNPs were not required, and thus, in most cases, did not report 
information on how many of their members had actually completed an 1-
1RA or reassessment in prior years. However, CMS requires D-SNPs to 
report this information separately, via the annual Part C reporting 
process. As this data collection is relatively new, CMS has not yet 
made information regarding SNPs' HRA completion rates public, but we 
have reviewed this data to identify areas of improvement. To that end, 
CMS has engaged a contractor to develop a standardized HRA tool (HRAT) 
and is testing the feasibility of using this tool once it has been 
tested and validated. CMS will also identify approaches for ensuring 
that D-SNPs have acceptable HRAT completion rates. HHS suggests that 
the report be revised to include this information, so as to provide a 
more complete picture of CMS's requirements.  

In addition, CMS requires that each SNP describe its target 
population, and characteristics of that population in its MOC. D-SNPs 
are expected to target plan benefits and services to their specific 
population of dual eligibles. The report recommends that D-SNPs also 
he required to quantify the number of members who would receive such 
services, but it is not clear what purpose that information would 
serve, as the purpose of the MOC is to provide a framework for 
appropriately coordinating the care of SNP enrollees. CMS is willing 
to consider modifications to the MOCs, but would appreciate further 
clarification from GAO as to what information would he useful (beyond 
the example described above).  

GAO Recommendation:  

CMS should require D-SNPs to collect and report to CMS standard 
measures relevant to the population they serve, including measures of 
beneficiary health risk, beneficiary vulnerability, and plan 
performance.  

HHS Response:  

HHS concurs with GAO's recommendation. CMS currently requires D-SNPs 
to report a set of standardized measures and is developing additional 
standardized measures. CMS collects a broad variety of measures 
through the Health Plan Employer Data and Information Set (FIFDISg), 
Health Outcomes Survey (FIGS), and Consumer Assessment of Health Plans 
Survey (CAHPS®) and via CMS contractors or administrative avenues. 
These measures, including readmission rates and other state-of-the-art 
outcome measures, are used in evaluating the quality and performance 
of contracts that include D-SNPs. In evaluating these MA plans, CMS 
relies on consensus building organizations, such as the National 
Committee on Quality Assurance (NCQA) and Pharmacy Quality Alliance 
(PQA), to identify and specify the appropriate measures for these and 
other special populations. Thus, CMS already requires a standard set 
of measures for reporting. As additional measures become available 
through these organizations, CMS will implement them as appropriate 
for these plans. Further, CMS is in the process of determining, as 
part of Medicare's star rating system, which measures are 
appropriately collected at the plan level (e.g., for a given SNP) 
versus at the broader contract level (e.g., across all plans within a 
contract). These determinations will need to consider the effect of 
small sample sizes (a particularly critical issue for SNPs), as well 
as the substantially higher cost for achieving appropriate sample 
sizes for a large number of SNPs.  

We have attached our recent Health Plan Management System (HEMS) 
memorandum, dated August 3, 2012, which outlines updated requirements 
for reporting calendar year (CY) 2013 HEMS*, HOS, and CAHPS® measures. 
(See Table 2 of the attached memo, which lists required SNP-specific 
HMIS measures, including plan all cause readmissions. In addition, 
public use tiles containing SNP-specific HEMS measures, from 2008-
2012, arc available on the CMS website at: [hyperlink, 
http://www.cms.cov/Research-Statisties-Data-rand-Systems/Statistics-
Trends-and-itcports/MCRAtivPariDEnrolData/SNP-HED1S-Public-Use-
Filcs.html.] GAO acknowledges (footnote 19) that D-SNPs are required 
to submit a subset of the HEWS measures hut, given that the title of 
the report and its recommendations focus on CMS's publication of a
standard set of quality measures for D-SNPs, HHS recommends that the 
report properly reflect CMS's efforts in these areas.  

The report also fails to mention that all Special Needs Plans (SNPs), 
including D-SNPs, are required to submit both a Quality Improvement 
Project (QIP) and a Chronic Condition Improvement Project (CCIP), 
which are unique quality improvement initiatives that must be 
conducted as part of the plan's Quality Improvement Program Plan in 
accordance with regulations at 42 CFR §422.152. For CY 2012, we are 
aligning the requirements of these initiatives with broadly-focused 
Department of Health and Human Services (I-11-1S) activities. As such, 
all plans, including D-SNPs, must conduct a 3-year QIP that is focused 
on reducing all-cause hospital readmissions, in accordance with the 
Partnership for Patients initiative, an HHSwide initiative that 
includes a variety of initiatives focused on improving care and 
reducing costs. Plans must also conduct a CCIP that is focused on 
reducing risks for cardiovascular disease, in support of the national 
"Million Hearts" campaign. Plans are currently submitting their "plan" 
section of the "Plan, Do, Study, Act" cycle, and CMS staff, with 
contractor support, will review and score this section of the QIP and 
provide feedback and technical assistance to plans over the next 
several months.  

In addition to the SNP-specific HEDIS measures described above, D-SNPs 
are required to report their scores on six specific structure and 
process measures to NCQA, which then prepares summary reports for CMS. 
While these reports are not yet publicly available, CMS is willing to 
discuss these results with GAO, and is considering when and how to 
make them publicly available. We would also note that CMS continues 
its efforts to develop standardized outcome measures to assess 
healthcare quality for all Medicare Advantage Organizations {MAO) 
products, including SNPs, through a contract with RAND. These outcome 
measures are expected to be grounded in evidence-based practices and 
feasible to apply at the plan level. A number of potentially-viable 
outcomes measures have been identified as a result of a recently-
completed pilot study, and RAND is currently preparing for the 
implementation of a validation study. Once standardized measurements 
are developed and tested, CMS expects to incorporate the measures 
within the MA program.  

Finally, we also note that HHS engaged the Measure Applications 
Partnership (MAP), a multi-stakeholder group of public and private-
sector organizations and experts convened by the National Quality 
Forum (NQF) in May 2011. The MAP's latest report, "Measuring Healthcare
Quality for the Dual Eligible Beneficiary Population," released in 
June 2012, recognizes the fragmented and episodic nature of the care 
the dual eligible population receives and stated that measurement can 
set expectations and provide powerful incentives for change. (The full 
text of the report is available at: [hyperlink, 
http://www.qualityforum.org/Setting_Priorities/Partnership/Measure_Appli
cations_Partnership.aspx]. The report highlights a recommended starter 
set of specific measures for immediate use that are sensitive to the 
unique needs of dual eligible beneficiaries, including measures of 
patient/caregiver experience, hospital readmissions, care transitions, 
detecting and treating depression, and screening older adults for fall 
risk. In addition, the report also highlighted measures that the 
workgroup fell should be expanded or broadened and provided guidance 
for further measurement development work as well as measure gaps for 
dual eligible populations. HI-IS intends to move forward in assessing 
the suggestions within the report, including the modification of these 
measures as appropriate for application at the plan level.  

GAO Recommendation:  

The CMS should systematically analyze these data and make h results 
routinely available to the public.  

HHS Response:  

HHS concurs with the GAO's recommendation, as CMS currently makes 
certain D-SNP performance data publicly available. Please refer to 
HHS's response to GAO's second recommendation above for more 
information.  

GAO Recommendation:  

CMS should conduct an evaluation of the extent to which D-SNPs have 
provided sufficient and appropriate care to the population they serve 
and report the results in a timely manner.  

HHS Response:  

HHS concurs with GAO's recommendation and intends to continue to 
evaluate the information described above to examine D-SNPs' overall 
performance. HHS is contemplating a more formal, comprehensive 
evaluation, but such a review is likely dependent on the program's 
continued operation, as the SNP authority currently expires at the end 
of 2013.  

[End of section]  

Appendix II: GAO Contact and Staff Acknowledgments: 

GAO Contact: 

James C. Cosgrove, (202) 512-7114 or CosgroveJ@gao.gov: 

Staff Acknowledgments: 

In addition to the contact named above, Phyllis Thorburn, Assistant 
Director; Ramsey Asaly; George Bogart; Melanie Anne Egorin; Linda 
Galib; Giselle Hicks; Corissa Kiyan; Elizabeth T. Morrison; and 
Kristal Vardaman made key contributions to this report. 

[End of section]  

Footnotes:  

[1] Medicare is the federally financed health insurance program for 
persons 65 years of age or over, certain individuals with 
disabilities, and individuals with end-stage renal disease (ESRD). 
Medicare fee-for-service (FFS), also known as original Medicare, 
includes Medicare Parts A and B. Medicare Part A covers hospital and 
other inpatient stays. Medicare Part B is optional insurance, which 
covers hospital outpatient, physician, and other services and requires 
a monthly premium. Medicare Part B beneficiaries have the option of 
enrolling in a Medicare Advantage (MA) plan--a private plan 
alternative to Medicare FFS that operates under Medicare Part C--to 
receive their Parts A and B benefits. In addition, all Medicare 
beneficiaries may opt to receive prescription drug coverage under 
Medicare Part D either through a separate Part D plan or through an MA 
plan.  

[2] Medicare Payment Advisory Commission (MedPAC), A Data Book: Health 
Care Spending and the Medicare Program (Washington, D.C.: June 2011). 

[3] Kaiser Family Foundation, Dual Eligibles: Medicaid Enrollment and 
Spending for Medicare Beneficiaries in 2007 (Washington, D.C.: 
December 2010). 

[4] Pub. L. No. 108-173, § 231, 117 Stat. 2066, 2207 (codified at 42 
U.S.C. §§ 1395w-21(a)(2)(A) and 1395w-28(b)(6)). 

[5] This number includes 233,902 dual-eligible beneficiaries in Puerto 
Rico. 

[6] Throughout this report, "other MA plans" refers to MA plans that 
are not SNPs. Enrollment in MA plans, including D-SNPs, is voluntary, 
but dual-eligible beneficiaries are allowed to change plans each 
month, whereas most other Medicare beneficiaries may change plans only 
during the annual open enrollment period.  

[7] Throughout this report, "specialized services" refers to services 
CMS requires SNPs to provide, such as health risk assessments, as well 
as supplemental benefits, which are benefits not provided under 
Medicare FFS that may be offered by MA plans. 

[8] Risk scores are a relative measure of projected Medicare costs for 
each beneficiary--with lower scores indicating lower expected costs--
and are expected to be the same for beneficiaries with the same health 
conditions and demographic characteristics. In this report we use 
lower risk scores as a proxy for better health status. We report 
average risk scores in 2010 because 2010 was the year for which the 
most recent risk score data were available.  

[9] We excluded enrollees with ESRD, those living outside of the 
United States, and new enrollees from our calculations of average risk 
scores for each plan type. We adjusted all average risk scores to 
account for CMS's normalization factor of 1.041 in 2010; additionally, 
we adjusted the risk scores for D SNPs and other MA plans downward by 
3.41 percent to account for CMS's estimate of the diagnostic coding 
differences between MA and FFS. (This is likely a conservative 
estimate. See GAO, Medicare Advantage: CMS Should Improve the Accuracy 
of Risk Score Adjustments for Diagnostic Coding Practices, [hyperlink, 
http://www.gao.gov/products/GAO-12-51] [Washington, D.C.: Jan. 12, 
2012].)  

[10] We focused our analysis on the following types of MA plans: 
health maintenance organizations, local preferred provider 
organizations, regional preferred provider organizations, private fee-
for-service plans, and provider-sponsored organizations. We excluded 
one type of MA plan--medical savings accounts--because they are not 
allowed to offer mandatory supplemental benefits. Mandatory benefits 
must be provided for every person enrolled in the plan, whereas 
optional supplemental benefits are available to those enrollees who 
elect and pay for them. In addition, we excluded plans that only 
provided Medicare Part B benefits and plans that restricted enrollment 
to members of an employer group or religious fraternal benefit society. 

[11] An individual model of care may cover multiple D-SNPs offered by 
a single MA organization. 

[12] According to CMS, the most-vulnerable beneficiaries include, but 
are not limited to, those beneficiaries who are frail, disabled, or 
near the end of life. 

[13] For 2012, only new SNPs and those expanding their plan service 
areas were required to contract with the state Medicaid program. See 
42 C.F.R. § 422.107(d) (2011). Our review may not account for relevant 
terms that may have been incorporated by reference into the contracts. 
CMS reviewed and approved these contracts for compliance with federal 
law and CMS regulations. 

[14] The states in our sample were Alabama, California, Massachusetts, 
Minnesota, and Oregon. 

[15] We excluded D-SNPs and dual-eligible beneficiaries in Puerto Rico 
from our analyses because Medicare enrollment, cost, and use in Puerto 
Rico are different than in the states, including a far greater 
proportion of Medicare beneficiaries enrolling in MA plans. In 
addition, the CMS enrollment data we received did not include dual-
eligible beneficiaries in Puerto Rico (they were coded as "low-income 
territory beneficiaries" and were not included in the data extracts). 

[16] Of these 322 D-SNPs, 17 have been designated by CMS as fully 
integrated dual-eligible special needs plans (FIDESNP). A FIDESNP is a 
CMS-approved D-SNP that (1) enrolls special-needs individuals entitled 
to medical assistance under a Medicaid state plan; (2) provides dual-
eligible beneficiaries access to Medicare and Medicaid benefits under 
a single managed care organization; (3) has a capitated contract with 
a state Medicaid agency that includes coverage of specified primary, 
acute, and long-term care benefits and services, consistent with state 
policy; (4) coordinates the delivery of covered Medicare and Medicaid 
health and long-term care services using aligned care management and 
specialty care network methods for high-risk beneficiaries; and (5) 
employs policies and procedures approved by CMS and the state to 
coordinate or integrate member materials, enrollment, communications, 
grievance and appeals, and quality improvement. 42 C.F.R. § 422.2 
(2011). In addition, there are 8 D-SNPs operating in Puerto Rico in 
2012. 

[17] In 2012, FIDESNPs will be eligible for increased payments when 
certain requirements, such as having a similar average level of 
frailty as the Program of All-Inclusive Care for the Elderly (PACE) 
program, are met. 

[18] Under each category of supplemental benefits, plans can provide 
coverage for a variety of individual services, such as eye exams, 
eyeglasses, or contact lenses under a vision benefit; however, plans 
do not have to provide coverage for all individual services under a 
supplemental benefit category. 

[19] CMS assesses plan quality using a five-star rating scale based on 
measures of clinical quality, patients' reported care experience, and 
contract performance. Once fully phased in after 2014, the revised 
rebates will range from 50 percent of the difference between a plan's 
bid and benchmark for plans with the lowest quality ratings to 70 
percent of the difference for plans with the highest quality ratings. 
See Pub. L. No. 111-148, Title III, Subtitle C, 124 Stat. 442, as 
amended by the Health Care and Education Reconciliation Act of 2010, 
Pub. L. No. 111-152, § 1102, 124 Stat. 1029, 1040. 

[20] MedPAC attempted to analyze the quality of SNPs, but found 
isolating results for SNPs is difficult because of the way 
organizations that sponsor SNPs report HEDIS quality measures. Most 
HEDIS data are reported across all of an organization's plans (i.e., 
both SNPs and other MA plans) rather than reported separately for 
SNPs. SNPs do report separately on a subset of 12 of 45 HEDIS 
measures. When comparing these measures, MedPAC found in general that 
SNP performance was poorer than other MA performance, but there was 
wide variation across plans. See MedPAC, Report to the Congress: 
Medicare Payment Policy (Washington, D.C.: March 2012), 328-329. 

[21] See 42 C.F.R. § 422.152(g)(2)(iii) (2011). 

[22] 42 C.F.R. § 422.101(f) (2011). 

[23] Social Security Act (SSA), § 1859(f) (requirement for contract); 
Pub. L. No.110-275, § 164(c)(2), 122 Stat. 2573 (temporary exemption 
for certain SNPs) as amended by Pub. L. No. 111-148, § 3205(d), 124 
Stat. 458 (extending temporary exemption through 2012). CMS 
regulations require that contracts between D-SNPs and states include 
documentation of (1) the MA organization's responsibility, including 
financial obligations, to provide or arrange for Medicaid benefits; 
(2) the category(ies) of eligibility for dual-eligible beneficiaries 
to be enrolled; (3) the Medicaid benefits covered; (4) the cost-
sharing protections covered; (5) the identification and sharing of 
information on Medicaid provider participation; (6) the verification 
of enrollees' eligibility for both Medicare and Medicaid; (7) the 
service area covered; and (8) the contract period. 42 C.F.R. § 
422.107(c) (2011). Although such contracts will be required for D-SNPs 
to operate within a state, there is no requirement that a state enter 
into such a contract. 

[24] CMS, Common Contracting Issues and Discussion (Baltimore, Md.: 
revised September 2011). 

[25] Pub. L. No. 111-148, § 2602, 124 Stat. 315. 

[26] In 2011, the Medicare-Medicaid Coordination Office, in 
partnership with the Innovation Center, entered into contracts with 15 
states for up to $1 million each to design state demonstrations. 
Furthermore, in July 2011 CMS issued a letter calling for additional 
state Medicaid agencies to submit letters of intent to participate in 
the demonstrations to better align Medicare and Medicaid funding. 

[27] SSA, § 1115A. 

[28] The demonstration may be expanded through rulemaking if (1) the 
Secretary determines that the expansion would reduce spending without 
reducing the quality of care or improve the quality of care without 
increasing spending; (2) the CMS Chief Actuary certifies that the 
expansion would reduce or not increase net program spending; and (3) 
the Secretary determines that the expansion would not deny or limit 
the coverage or the provision of benefits for applicable individuals. 
SSA, § 1115A(c). 

[29] Data on mental health characteristics are based on the 2010 risk 
score data. 

[30] Demographic data refer to dual-eligible beneficiaries in July 
2011. Mental health data refer to dual-eligible beneficiaries in July 
2010 and are based on 2009 diagnosis data. Both sources exclude 
beneficiaries in Puerto Rico and those living outside of the 50 states 
and the District of Columbia in 2010. 

[31] Racial or ethnic minorities include Black, Hispanic, Asian, and 
North American Native beneficiaries, and beneficiaries who reported 
"Other" race. CMS's enrollment data have limitations in accurately 
identifying beneficiary race and ethnicity, resulting in an 
underreporting of Hispanics, Asian/Pacific Islanders, and American 
Indian/Alaskan Natives. 

[32] We report data on health status based on dual-eligible 
beneficiaries enrolled in Medicare and Medicaid as of July 2010. These 
data are not linked to beneficiaries' demographic characteristics in 
2011. 

[33] Average risk scores for D-SNP enrollees and dual-eligible 
beneficiaries in other MA plans and FFS were even-more similar when 
comparing only the risk scores for beneficiaries who lived in the 
community in July 2010 (D-SNP=1.27, other MA=1.29, FFS=1.28). This is 
true in part because a larger proportion of dual-eligible 
beneficiaries who were considered to be long-term institutional (i.e., 
those who resided in an institution for at least 3 months) enrolled in 
FFS than in D-SNPs in 2010. CMS calculates the risk scores of these 
institutional beneficiaries using a different model than it does for 
community beneficiaries, and the resulting risk scores for the two 
groups are different. Institutionalized dual-eligible beneficiaries in 
D-SNPs, other MA plans, and FFS had similar risk scores in 2010 (D-
SNP=1.95, other MA=1.98, FFS=2.00). 

[34] However, when taking into consideration the large proportion of 
FFS beneficiaries who lived in institutions compared to D-SNP 
beneficiaries, full-benefit dual-eligible beneficiaries in D-SNPs and 
FFS had similar average risk scores (1.25 and 1.29, respectively). 

[35] We examined a total of 20 supplemental benefits offered by D-SNPs 
and other MA plans. In addition to the categories listed in figure 4, 
we also examined the following 10 supplemental benefits categories 
offered by fewer than half of D-SNPs: (1) podiatry, (2) meal benefits, 
(3) chiropractic, (4) acupuncture, (5) point-of-service or out-of-
network option, (6) home infusion, (7) inpatient psychiatric 
hospital/facility, (8) U.S. Visitor/Travel program, (9) 
cardiac/pulmonary rehabilitation, and (10) outpatient drug benefit. Of 
these additional 10 benefits, 5 were offered less frequently by D-SNPs 
compared to other MA plans. We limited our analysis to plans with no 
premium because 99 percent of D-SNPs have no Part C premium, and in 
our interviews with D-SNPs, representatives emphasized the importance 
of zero-premium plans for the dual-eligible population. 

[36] For 2013, CMS will allow certain D-SNPs that meet integration and 
performance standards to offer additional supplemental benefits beyond 
those CMS currently allows all MA plans to offer, where CMS finds that 
the offering of such benefits could better integrate care for dual-
eligible beneficiaries. Such benefits are subject to CMS approval, but 
may include nonskilled nursing services, personal care services, and 
other long-term care services and supports designed to enable 
beneficiaries to remain in the community. D-SNPs must offer these 
additional supplemental benefits at no additional cost to the 
beneficiary. 

[37] CMS officials told us they plan to conduct a review of how 150 
SNP models of care have been implemented in 2012. CMS also asks plans 
to collect information on certain measures related to the model of 
care as a part of their quality improvement program; however, this 
information is only required to be made available upon request and is 
not systematically reported to CMS or the public. CMS officials also 
told us about several SNP quality initiatives that are in operation or 
in planning stages; however, these activities were not equivalent to a 
full evaluation of the SNP program. 

[38] Improved collection of information allows an agency to meet its 
goals for accountability for effective and efficient use of resources, 
and is consistent with standards for internal control. GAO, Standards 
for Internal Control in the Federal Government, [hyperlink, 
http://www.gao.gov/products/GAO/AIMD-00-21.3.1] (Washington, D.C.: 
November 1999). 

[39] The contracts were not required to address these goals. However, 
the contracts must document the Medicaid benefits, if any, to be 
provided by the D-SNPs. See, 42 C.F.R. § 422.107(c) (2011). 

[40] CMS, State Resource Center: State Options for Designing Dual SNP 
Contracts with Medicare Advantage Organizations that Adhere to MIPPA 
Requirement (Baltimore, Md.: 2009), 5. 

[41] CMS stated in its state contract training materials for 2013 D-
SNP-state Medicaid agency contracting that contracts for 2013 must 
specify how Medicare and Medicaid benefits are integrated and 
coordinated. 

[42] For D-SNPs that provided all Medicaid benefits as part of their 
plan, coordination with the state Medicaid agency was likely not 
needed as the plan provided all benefits. 

[43] For 2013, CMS issued guidance specifying criteria under which D-
SNPs and states may enter into such subcontracting arrangements. 

[44] As of July 2012, one additional state from our sample had 
proposed as part of its Financial Alignment Initiative proposals 
shifting D-SNPs into demonstration plans and, therefore, end D-SNPs in 
their state. An additional five states not included in our sample 
propose shifting D-SNPs into demonstration plans or otherwise ending D-
SNPs in their states. 

[End of section]  

GAO’s Mission:  

The Government Accountability 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.  

Obtaining Copies of GAO Reports and Testimony:  

The fastest and easiest way to obtain copies of GAO documents at no 
cost is through GAO’s website [hyperlink, http://www.gao.gov]. Each 
weekday afternoon, GAO posts on its website newly released reports, 
testimony, and correspondence. To have GAO e-mail you a list of newly 
posted products, go to [hyperlink, http://www.gao.gov] and select 
“E-mail Updates.”  

Order by Phone:  

The price of each GAO publication reflects GAO’s actual cost of 
production and distribution and depends on the number of pages in the 
publication and whether the publication is printed in color or black 
and white. Pricing and ordering information is posted on GAO’s 
website, [hyperlink, http://www.gao.gov/ordering.htm].  

Place orders by calling (202) 512-6000, toll free (866) 801-7077, or 
TDD (202) 512-2537.  

Orders may be paid for using American Express, Discover Card, 
MasterCard, Visa, check, or money order. Call for additional 
information.  

Connect with GAO:  

Connect with GAO on facebook, flickr, twitter, and YouTube.
Subscribe to our RSS Feeds or E mail Updates. Listen to our Podcasts.
Visit GAO on the web at [hyperlink, http://www.gao.gov].  

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

Contact: 
Website: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]; 
E-mail: fraudnet@gao.gov; 
Automated answering system: (800) 424-5454 or (202) 512-7470.  

[End of document]  

Congressional Relations:  

Katherine Siggerud, Managing Director, siggerudk@gao.gov, (202) 512-4400
U.S. Government Accountability Office, 441 G Street NW, Room 7125
Washington, DC 20548.  

Public Affairs: 
Chuck Young, Managing Director, youngc1@gao.gov, (202) 512-4800
U.S. Government Accountability Office, 441 G Street NW, Room 7149 
Washington, DC 20548.