Special Needs Plans and the Coordination of Benefits and Services for Dual Eligibles

Abstract

Special Needs Plans (SNPs) are a new type of Medicare Advantage plan with the potential to coordinate Medicare and Medicaid benefits and services for dually-eligible beneficiaries. However, experience to date suggests that SNPs have not greatly expanded the number of dual eligibles enrolled in joint Medicare-Medicaid products. Importantly, SNPs need to have some contractual relationship with state Medicaid plans to add value to dual eligible beneficiaries beyond traditional Medicare Advantage plans. Although SNP enrollment is higher in states with such a relationship, several steps are identified to further expand this market and improve coordination between SNPs and state Medicaid programs.

Individuals who are dually eligible for both Medicare and Medicaid have received considerable policy attention in recent years due to their high cost and complex health needs. Although this population is relatively small in number, consisting of about 7.5 million individuals, spending on dual eligibles accounts for roughly 24% of Medicare’s total spending and 42% of Medicaid’s spending.¹ Mainly because of their poor health status and continual health care needs, the Medicare costs of dually eligible beneficiaries are 1.5 times that of other Medicare beneficiaries.²

Quality of care is also an important concern in the care of the dual eligible population.³ Relative to beneficiaries covered only by Medicare, dual eligibles are much less likely to receive specific types of preventive care, follow-up care or testing.⁴ For example, roughly 25% of dual eligible women received a mammography every two years, relative to 40% of Medicare only beneficiaries. In terms of long-term care needs, over half (58%) of the elderly community-based dual eligibles requiring assistance with activities of daily living have some unmet need.⁵

The poor coordination of Medicare and Medicaid benefits and services has been a long-standing problem in the care of the dual eligible population.⁶ The coordination of benefits refers to the bundling of Medicare and Medicaid coverage such that there is an incentive to deliver services efficiently. For example, under a joint Medicare-Medicaid product, there will be an incentive to minimize transitions from Medicaid-financed nursing home care to higher cost Medicare-financed hospital care. The coordination of services refers to the explicit management of acute and chronic care across settings. For example, case managers can be used to coordinate the acute and chronic care of a dual eligible beneficiary in order to ensure appropriate placement and care across settings.⁷ In terms of linking the two concepts, the coordination of acute and chronic care benefits can be thought of as a necessary, but not sufficient, condition for the coordination of acute and chronic care services. There may be gains associated with simply coordinating benefits across the two programs, but without such an alignment, there is little reason to suspect that there will be any coordination in the delivery of services.

One recent policy innovation towards potentially coordinating Medicare and Medicaid benefits and services has been the establishment of Medicare Advantage special needs plans (SNPs). SNPs were authorized under the Medicare Modernization Act (MMA) of 2003 with the idea of attracting a different type of beneficiary into Medicare Advantage. This paper provides a brief overview of SNPs, outlines the opportunities and challenges for key stakeholders in terms of care coordination, documents SNP growth and coordination of benefits across states, and discusses how SNPs might be improved as a mechanism towards better aligning Medicare and Medicaid coordination of care.

Special Needs Plans (SNPs)

SNPs can target one of three types of beneficiaries: (1) dual eligible enrollees, (2) residents of nursing facilities or similar institutions, and (3) residents with severe or disabling chronic conditions. Dual eligible SNPs almost exclusively serve those eligible jointly for both Medicare and Medicaid, while approximately half of enrollees in institutional and chronic care SNPs are dual eligibles.⁸ As such, this paper will focus on dual eligible SNPs, which constitute 57% of plan and 72% of enrollees. Similar to other Medicare Advantage plans, SNPs receive a capitated payment to provide Medicare benefits.

From the perspective of program coordination, dual eligible SNPs provide the opportunity to combine Medicare and Medicaid managed care contracting for dually eligible beneficiaries. In theory, the use of a capitated payment may encourage a more efficient production of health care services across the two programs. Typically, this type of coordination would require special Medicare demonstration authority from CMS. Under the original MMA rules, there was no mandate that dual eligible SNPs coordinate benefits with Medicaid. As such, the majority of dual eligible SNPs offered only Medicare-covered benefits and did not provide a joint Medicare-Medicaid product. Although these Medicare-only SNPs may still provide value to beneficiaries by coordinating various acute Medicare services, it is unclear whether they provide any additional benefit above-and-beyond traditional Medicare Advantage plans.

As part of the initial Medicare Part D enrollment process (effective January 1, 2006), Medicaid managed care plans were given a one-time opportunity to “passively” (unless the beneficiary notified the plan otherwise) enroll dually eligible members into their companion Medicare plans. For the 13 states (42 SNPs) with mature Medicaid managed care programs that included dual eligibles, passive enrollment provided an opportunity to greatly expand the number of dually eligible beneficiaries in joint Medicare-Medicaid products. For example, an estimated 47,000 Medicaid beneficiaries in managed long-term care plans became dually enrolled through passive enrollment in Arizona (via the Arizona Long Term Care System), Minnesota (via the Minnesota Senior Health Options [MSHO] program) and Texas (via the STAR+PLUS program).⁹

The Medicare Improvements for Patients and Providers Act (MIPPA) of 2008 requires new dual eligible SNPs—starting in 2010 when the current moratorium on new plans expires—to have contracts with state Medicaid agencies. Existing dual eligible SNPs that do not have a contract with the state may continue to operate but cannot expand into new service areas. Importantly, the legislation does not mandate that states contract with SNPs.

Opportunities and Challenges for Care Coordination

This section reviews the primary opportunities and challenges for the coordination of Medicare and Medicaid benefits and services for the key stakeholders: dual eligible enrollees, SNPs, and Medicare and Medicaid.

Enrollees

The incentive for dual eligible beneficiaries to enroll in a dual eligible SNP—as compared to a traditional Medicare Advantage plan—relates to whether the plan can add value to the beneficiary. Disease management and coordination of Medicare benefits is already common in traditional Medicare Advantage plans. Thus, dual eligible enrollees would have to be given some benefit they were not already receiving from a traditional Medicare Advantage plan in order to incentivize them to enroll in a dual eligible SNP. Clearly, better coordination across Medicare and Medicaid could be the feature that attracts dual eligibles into SNPs and keeps them in these plans over time, but to date, it is unclear how well benefits and services are in fact being coordinated for most dual eligible SNP enrollees. The recent MIPPA legislation mandating that all new dual eligible SNPs have contracts with state Medicaid agencies will be a step towards ensuring greater coordination of Medicare-Medicaid benefits with dual eligible SNPs. Whether this new mandate of coordinated benefits will translate into a better coordination of care is still unclear.

As a more general issue, many elderly individuals have a negative impression of managed care, likely related to concerns that that they will have to change doctors, go to new locations for care, and have fewer choices.¹⁰ The Program of All-Inclusive Care for the Elderly (PACE), a voluntary program that combined Medicaid and Medicare financing in a managed care framework, was found to attract disproportionately healthy enrollees.¹¹ It is still unclear whether large numbers of dual eligible enrollees will voluntarily enroll in managed care plans, especially when the potential “value added” of better Medicare-Medicaid coordination may not be apparent.

Plans

Because SNPs are paid similarly to other Medicare Advantage programs, higher capitated payments do not appear to be a significant rationale for establishing a SNP relative to a traditional Medicare Advantage plan.¹² For companies that were already providing Medicare Advantage plans however, it has been relatively straightforward to establish new SNPs. Other possible factors explaining the decision to offer SNPs include the potential for plans to add new markets, the opportunity to build on existing areas of expertise (e.g., disease management), and the opportunity to tailor and market benefit packages for specific subpopulations.¹³ In order for these plans to be profitable, they must modify care patterns such that unnecessary use of costly Medicare services is minimized. Dual eligible SNPs also have the potential to address the care coordination challenges stemming from beneficiaries’ reliance on Medicaid for those services not covered by Medicare.

Prior to MIPPA, there was no mandate that plans coordinate benefits with the states. As such, the majority of dual eligible SNPs, to date, have not provided a coordinated product but rather simply provided Medicare services under a capitated payment arrangement. However, there have been a few examples of coordinated Medicare-Medicaid service delivery models within SNPs across different states.¹⁴ For example, MSHO uses a single nurse or social worker to coordinate services, the Wisconsin Partnership Program employs a multi-disciplinary team that included a nurse practitioner, and the Massachusetts Senior Care Options (SCO) program utilizes a team of nurses and social workers. In Wisconsin, nurse practitioners accompany patients on physician visits as a means of more closely connecting community and medical care.

Medicare and Medicaid

In theory, a better coordination of Medicare-financed acute and post-acute services and Medicaid-financed long-term care benefits and services should improve quality for dual eligible enrollees while lowering (or at least not raising) program costs. However, there are several potential barriers that may impede the coordination of benefits and services under dual eligible SNPs.

First, a potential issue for the states is the alignment of incentives under a combined Medicare-Medicaid product. For example, the SNPs (and potentially CMS) share in savings from lower Medicare hospital costs, but states do not necessarily directly benefit.¹⁵ In order to increase interest and participation from the states in SNPs, there will need to be greater transparency in payments and—to the extent they are not already in place—the use of mechanisms that allow states to share in any potential decrease in Medicare expenditures. Although a direct payment to Medicaid might be unlikely to gain approval from Congress, this could also be realized in the negotiation of Medicaid capitated payments upfront.

Second, the government’s ability to coordinate care within a single SNP has suffered from the underlying conflict between federal and state approaches for managed care. The federal approach with Medicare Advantage plans is based around consumer choice and values variation across plans in the marketplace. Medicare freedom of choice is an important beneficiary right, but it can confound coordination if dual eligibles enroll in different Medicare and Medicaid plans. In contrast to Medicare, state Medicaid plans typically emphasize long-term investments with a limited number of plans and uniform benefits to promote equity in a publicly-funded program.

Finally, the lack of data sharing across the Medicare and Medicaid programs has hampered the success of the SNPs. Part of the problem is that many key data elements are not currently collected by Medicare and Medicaid. Where available, the sharing of clinical and claims data on utilization, encounters, diagnostics and key events would be useful to ensure quality, monitor access to services, inform rate-setting, monitor the adequacy of provider networks, and coordinate benefits.¹⁶ States could make this data sharing a condition of their Medicaid contract with the SNPs, although smaller states may lack the “market power” to enforce such an arrangement.

SNP Expansion and State Coordination

The number of dual eligible plans has expanded from 42 plans in 2005 to 439 plans as of June 2008 (see Exhibit 1). In spite of the number of plans offering services, enrollment to date has been relatively modest. Currently, there are 854,877 dual eligible beneficiaries enrolled in SNPs, which constitutes roughly 11% of the dual eligible population nationwide.¹⁷ Given passive enrollment, even fewer dual eligible beneficiaries have actively chosen to enroll in a SNP.¹⁸

Exhibit 1.

Enrollment in Dual Eligible Special Needs Plans

Year	Plans	Enrollees
2005	42	n/a
2006 (September)	256	491,877
2007 (September)	320	722,286
2008 (June)	439	854,877

Sources: Milligan and Woodcock, 2008 and Centers for Medicare and Medicaid Services, Special Needs Plan Comprehensive Report, June 2008. Available at: http:/www.cms.hhs.gov/MCRAdvPartDEnrolData/SNP/list.asp.

SNP enrollment is fairly concentrated within a minority of states (see Exhibit 2). Puerto Rico and nine states account for 60.4% of SNPs and 85.4% of enrollees. There are currently eight states in which a SNP is not offered. Florida has the highest number of dual eligible SNPs (n=65) followed by New York (n=43), California (n=41), Texas (n=28), Puerto Rico (n=21), Minnesota (n=15) and Arizona (n=15). Puerto Rico has the largest number of dual eligible SNP enrollees (n=201,324).¹⁹ Other states with high numbers of enrollees include California (n=120,803), Pennsylvania (n=100,605), New York (n=66,097), Florida (n=64,526), and Texas (n=46,821).

Exhibit 2.

Penetration of Dual Eligible Special Needs Plans (SNPs) by State

State	Number of SNPs (June 2008)	SNP Enrollment (June 2008)	Total Number of Full and Partial Dual Eligibles (June 2006)	“SNP Penetration” (SNP enrollment/ Dual Eligibles)
Puerto Rico	21	201,324	194,763	1.03
Minnesota	15	36,368	108,806	0.33
Pennsylvania	14	100,605	305,949	0.33
Arizona	15	39,032	119,872	0.33
Oregon	8	17,224	63,612	0.27
California	41	120,803	843,121	0.14
Florida	65	64,526	476,829	0.14
Alabama	5	23,778	185,526	0.13
Colorado	4	7,971	62,898	0.13
Tennessee	12	30,967	248,508	0.12
New York	43	66,097	585,237	0.11
Utah	2	2,696	24,497	0.11
Rhode Island	2	3,599	34,433	0.10
Texas	28	46,821	506,841	0.09
Kentucky	1	9,288	142,518	0.07
Hawaii	3	1,606	26,119	0.06
Maryland	5	5,417	88,285	0.06
Connecticut	5	3,855	64,510	0.06
Massachusetts	7	12,621	216,932	0.06
Idaho	2	1,387	24,782	0.06
New Mexico	5	2,625	51,494	0.05
North Carolina	2	8,601	270,543	0.03
Wisconsin	7	6,320	200,761	0.03
Louisiana	7	4,481	157,327	0.03
Illinois	13	7,694	271,997	0.03
Georgia	6	5,526	223,211	0.02
New Jersey	7	3,600	147,184	0.02
Ohio	7	5,529	231,710	0.02
Mississippi	1	2,899	134,208	0.02
Nevada	1	675	31,599	0.02
Arkansas	4	2,088	100,237	0.02
District of Columbia	2	262	13,910	0.02
Washington	4	2,023	121,462	0.02
Maine	2	1,058	90,991	0.01
Delaware	2	195	20,459	0.01
Missouri	2	1,417	161,462	0.01
Oklahoma	3	693	79,236	0.01
Michigan	1	1,384	195,115	0.01
Nebraska	1	199	33,422	0.01
South Carolina	1	721	123,800	0.01
Virginia	3	622	150,019	<0.01
Iowa	1	74	67,570	<0.01
Kansas	2	52	48,320	<0.01
West Virginia	1	57	63,326	<0.01
Alaska	0	0	11,622	0
Indiana	0	0	99,944	0
Montana	0	0	17,728	0
New Hampshire	0	0	22,280	0
North Dakota	0	0	12,972	0
South Dakota	0	0	17,137	0
Vermont	0	0	27,166	0
Wyoming	0	0	8,404	0
National	439	854,877	7,530,654	0.11

Sources: Centers for Medicare and Medicaid Services, Special Needs Plan Comprehensive Report, June 2008. Available at: http://www.cms.hhs.gov/MCRAdvPartDEnrolData/SNP/list.asp; and, Centers for Medicare and Medicaid Services, 2006 Medicaid Managed Care Enrollment Repoort, June 2006. Available at: http://www.cms.hhs.gov/MedicaidDataSourcesGenInfo/Downloads/mmcer06.pdf.

Notes: The state SNP counts exclude the 56 plans and 97 beneficiaries in SNPs with an enrollment of less than 10 beneficiaries. As such, the state counts of SNPs and enrollees will not sum to the national totals.

As shown in the final column of Exhibit 2, states also vary in the degree of SNP “penetration” into the dual eligible beneficiary population. The penetration rate was calculated by dividing total dual eligible SNP enrollees by the total number of dual eligibles living in the state.²⁰ Puerto Rico (103%), Minnesota (33%), Pennsylvania (33%), Arizona (33%) and Oregon (27%) have the highest dual eligible SNP penetration rates. Not surprisingly, states that were able to passively enroll dual eligibles into a SNP such as Minnesota and Arizona have relatively high penetration rates.

States differ based on the degree of coordination with the SNPs and their capacity to contract with Medicaid in the future. If SNPs are unable to coordinate benefits and services across Medicare and Medicaid, it is unclear whether they add much value to dual eligible beneficiaries relative to a traditional Medicare Advantage plan. There are 13 states that currently contract with SNPs; 20 states that cover some dual eligibles in comprehensive Medicaid managed care plans; and 9 states that have Medicaid managed care systems that include some long-term care benefits.²¹ There is strong overlap across these groups, with Arizona, California, Massachusetts, Minnesota, New York, Texas and Wisconsin meeting all three criteria.

As expected, there are significantly higher SNP penetration rates in states with the capacity to coordinate with SNPs (see Exhibit 3). Specifically, the SNP penetration rate is 12.2% in the 13 states that coordinate benefits with SNPs and 6.6% in the remaining states. In the 20 states that cover some dual eligibles in a comprehensive Medicaid managed care plan, the penetration rate is 12.6% relative to 2.8% in states that do not offer such coverage. Finally, the 9 states with Medicaid managed care including some long-term care benefits have a 12.6% penetration rates relative to 6.1% in the remaining states.

Exhibit 3.

Special Need Plans (SNP) coordination with Medicaid and dual eligible SNP market peneration (SNP enrollees/dual eligibles)

Sources: Centers for Medicare and Medicaid Services, Special Needs Plan Comprehensive Report, June 2008. Available at: http://www.cms.hhs.gov/MCRAdvPartDEnrolData/SNP/list.asp; Centers for Medicare and Medicaid Services, 2006 Medicaid Managed Care Enrollment Repoort, June 2006. Available at: http://www.cms.hhs.gov/MedicaidDataSourcesGenInfo/Downloads/mmcer06.pdf; and Verdier et al 2008.

Although the coordination of Medicare-Medicaid benefits via SNP-state contracts cannot be equated with the coordination of services, it would seem to be a necessary component. That is, without some coordination of benefits, it is hard to envision any meaningful efforts by either the SNPs or the state to coordinate services. Thus, a key issue moving forward is increasing the number of states contracting with SNPs.

Where Do We Go From Here?

The recent MIPPA legislation extended the authorization for SNPs to December 31, 2010. Prior to MIPPA, three of the key stakeholder groups—states, dual eligible SNPs and beneficiaries—all had the right not to participate in a coordinated approach to care. MIPPA now requires that new dual eligible SNPs have contracts with state Medicaid agencies to provide or arrange for the provision of Medicaid benefits. As shown in Exhibit 3, dual eligible SNP penetration rates have has been highest in those states that contract with SNPs. However, as of 2006, less than half of all states contracted with state Medicaid programs, suggesting that new dual eligible SNPs will not be able to enter the majority of U.S. markets. Much of the onus now falls to state Medicaid programs to engage SNPs in new partnerships to increase the number of dual eligible beneficiaries enrolled in joint Medicare-Medicaid products.

To date, states have used two primary models to coordinate services with SNPs.²² The first option that states have pursued is a voluntary Medicaid managed care model in which dual eligible beneficiaries who choose to enroll in a SNP for their Medicare benefits can also enroll in the same plan for their Medicaid benefits. Examples of these programs include the MSHO and Massachusetts SCO. Another model used by states is mandating enrollment in a Medicaid managed long-term care plan with the idea that the beneficiary will also choose to enroll in a SNP for Medicare benefits. Ideally, the beneficiary will choose to enroll in a SNP that also serves his or her Medicaid plan. Examples of states implementing mandatory programs include Arizona and Texas (in certain counties).

Clearly, there are advantages and disadvantages for the states to these two models in terms of Medicaid expenditures, administrative costs, ease of implementation, the need for oversight, the role of side agreements, and most importantly for this paper, the coordination of benefits and services. The model that likely works the best in terms of coordination is the voluntary one in which the SNP receives capitation payments from both Medicare and Medicaid. Because the SNP bears the full financial risk for the dual eligible’s care, it has a strong incentive to provide services cost-effectively. The mandatory program option can result in a similar incentive structure, but only if the Medicaid managed care plan and the SNP are one in the same.

Even if SNPs are able to better coordinate acute and long-term care services for dual eligible enrollees, it is unclear whether this coordination will ultimately result in lower expenditures. The PACE and the MSHO programs are the two most rigorously evaluated models coordinating Medicaid and Medicare benefits and services for dual eligibles. Although quality of care and enrollees’ access to services were found to improve under PACE and remain relatively stable under the MSHO, the evaluations of PACE and MSHO both found higher program costs relative to comparison groups.²³ For example, the total capitated payment to PACE enrollees was 9.7% higher in the first year of enrollment than the projected Medicare and Medicaid cost if the enrollees had continued to receive care in a fee-for-service setting.²⁴ Thus, the best evidence to date from the PACE and MSHO programs does not suggest that joint Medicare-Medicaid capitation models are associated with lower costs relative to fee-for-service comparison groups. If this also proves to be the case for SNPs, the question for CMS is whether the benefits of increased coordination are worth the additional costs.

Concluding Thoughts

SNPs have the potential to add value by coordinating Medicare and Medicaid benefits and services for dual eligible beneficiaries. However, there has been a relatively modest enrollment into these plans thus far in spite of relatively strong entry into this market on the part of the SNPs. A primary concern is that—without some contract with state Medicaid programs—dual eligible SNPs provide little potential value to beneficiaries and to the Medicare program. The encouraging news is that dual eligible SNP penetration rates are significantly higher in states in which SNPs have the potential to contract with state Medicaid agencies. These contracts have contributed to the expansion of coordinated service delivery models in states such as Massachusetts, Minnesota and Wisconsin. It will be important to continue to track outcomes in these states to determine the impact of this coordination on program costs and beneficiary outcomes.

Moving forward, several different policies may further encourage the viability of SNPs as a mechanism towards coordinating services. First, a robust Medicaid managed care market is essential towards encouraging a fully capitated Medicare/Medicaid model. Historically, states have implemented these programs by securing CMS waivers. The easing of Federal requirements such that states can contract with SNPs for Medicaid-financed services without having to obtain a Medicaid waiver will be an important step.

Second, the MIPPA mandate that dual eligible SNPs contract with state Medicaid programs may be a double-edged sword for the SNP marketplace. On the positive side, this mandate ensures that new dual eligible SNPs do not offer “Medicare only” plans, but rather engage Medicaid in offering a joint Medicare-Medicaid product. The lack of Medicare-Medicaid coordination was an important concern regarding the early SNPs.²⁵ On the negative side however, over half of the states do not currently contract with SNPs, suggesting that dual eligible SNPs will no longer be able to enter many U.S. markets. Moreover, over 95% of dual eligible beneficiaries continue to be enrolled in Medicaid fee-for-service. In the absence of Medicaid managed care, Medicaid could contract with a vendor (ideally the SNP) to administer fee-for-service benefits, allowing the SNP to coordinate these benefits with the Medicare benefits it was already providing.²⁶ Although this model will likely not lead to the care coordination present in SNPs in Massachusetts, Minnesota and Wisconsin, the coordination of benefits could be a ‘second best’ solution.

Finally, there has been little evaluation of SNP performance in terms of beneficiary costs and outcomes. The lack of research is partly attributable to the fact that Medicare Advantage plans do not report claims or encounter data under Parts A or B. However, with some further data development, there are a series of different measures that might be used to evaluate dual eligible SNP performance including Part D claims data, Medicaid claims data (where accessible), beneficiary surveys (CAHPS, HEDIS), SNP enrollment/disenrollment rates, complaints and grievances against the SNPs, and the use of savings (or rebate) dollars by dual eligible SNPs.²⁷ Although one might posit that better care coordination should improve outcomes and lower costs, evaluations of previous initiatives that have coordinated Medicare and Medicaid benefits and services for dual eligibles in a managed care framework such as the PACE and MSHO programs have indicated higher spending.

In sum, dual eligible SNPs have great promise to address the poor coordination of Medicare and Medicaid for dual eligible enrollees. However, SNP enrollment has been relatively modest to date, likely reflecting the fact that—unless SNPs contract with state Medicaid programs to offer a coordinated product—these plans offer little additional value to dual eligible beneficiaries relative to traditional Medicare Advantage plans. Moving forward, the MIPPA legislation will ensure that new dual eligible SNPs contract with states in coordinating benefits and services. Ultimately, it will be important to track whether improved care coordination via dual eligible SNPs translates into better beneficiary outcomes and lower expenditures.