MedPAC Reviews Blending Medicare and Medicaid

Introduction

In its June 2012 Report to the Congress, the Medicare Payment Advisory Commission (MedPAC) included an examination of current options and activity with respect to programs that integrate – or have the potential to integrate – Medicare and Medicaid services and financing for those individuals with coverage from both programs, often referred to as dual eligibles.[1]  While the term "integration" does not have a single meaning in health policy discussions, in this context it generally refers to efforts to bring both Medicare and Medicaid dollars and Medicare and Medicaid services into a single system of care, so that the individuals using the services do not have to pay attention to whether they are from Medicare or Medicaid.  It is believed that "integrating" the programs can both improve the quality of health care services people receive and lower the cost of providing that care.

The Center for Medicare Advocacy has been working on issues affecting those who are dually eligible since its creation in 1986.  We have often joined forces with national and state-based advocates working to affect the design of programs to improve care for dual eligibles – a vulnerable population whose voice and needs should be heard in discussions about how best to provide their care.

The MedPAC Report

The MedPAC Report looks at cost and quality experience (as well as some other elements of experience) of Programs of All Inclusive Care for the Elderly (PACE) and Medicare Advantage Special Needs Plans, then examines the emerging state proposals, currently before the Medicare and Medicaid Coordination Office (MMCO) for review, to create new delivery systems and financing structures for their dual eligibles.

The Report notes, first, the great heterogeneity of the dual eligible population, which includes individuals with multiple chronic conditions, difficulties with activities of daily living, cognitive impairments such as dementia, physical disabilities, developmental disabilities and severe mental illness.  It also includes healthy individuals who are dually eligible mostly because they are 65 or older and are poor.

The Report then notes the high costs, on average, of providing care for this population. While comprising about 18% of traditional Medicare enrollment, they account for about 31% of traditional Medicare spending.  While comprising about 15% of Medicaid enrollment, they account for about 40% of Medicaid spending.  Because Medicaid spending is shared between the state and federal government and all Medicare spending is federal, estimates suggest that about 80% of all spending on dual eligibles is federal.

Programs of All Inclusive Care for the Elderly (PACE)

PACE is a program in Medicare and an optional program in Medicaid (meaning that a state can choose to have a PACE program or not) and generally operates as an integrated system through which individuals receive both Medicare and Medicaid services.  It is available to individuals age 55 and older who need a nursing home level of care.  PACE programs generally operate through day-care centers.  Individuals enrolling in PACE must use a PACE physician and thus generally need to give up their existing physician. The program is designed to keep people out of nursing homes. PACE programs are paid a monthly fee per enrollee (capitated rate) and accept the risk of providing nearly all services for that fee.

PACE programs have a statutory waiver that allows them to use Medicare dollars for non-health care supplies or services, if those services are identified as necessary by the enrollee's interdisciplinary team in the enrollee's care plan. 

With respect to quality, MedPAC notes that PACE programs report on measures such as rate of routine immunizations, grievances and appeals, disenrollment, hospital readmissions, emergency care, unusual incidents, deaths, falls or traumatic injuries resulting in death or hospitalization, infectious disease outbreaks and acquisition of pressure ulcers.  But, it notes, these quality measures are not publicly reported by the Centers for Medicare & Medicaid Services (CMS), which receives the reports from the programs.

Without access to the quality measures themselves, MedPAC reviewed the literature and found that PACE programs generally performed better on measures regarding hospitalizations, nursing home use and mortality compared with the experience of comparable beneficiaries in traditional Medicare.  However, differences between the two populations decreased, with respect to hospitalizations and nursing home use, when looked at after 18 months and after 24 months.

Compared with another program that also integrates Medicare and Medicaid, the Wisconsin Partnership Program (WPP), PACE programs did better on reducing hospital and Emergency Room utilization.  Differences in the two programs include that WPP does not operate through day care centers, enrollees keep their original physician and the interdisciplinary team in WPP is smaller than that in PACE and does not include the enrollee's physician.

Another evaluation compared PACE programs in one state with that state's program of offering long-term care services in the community (generally referred to as home and community based services, or HCBS), rather than in institutions.  It found that PACE enrollees had a lower risk of dying and greater stability in physical functioning.  In this situation, the comparison is not exact because the HCBS program is only paying for the Medicaid services; there is no integration with Medicare.  MedPAC notes that the state paid the PACE program more than it paid the HCBS program.

With respect to Medicare spending on PACE (the report includes little information about Medicaid spending), MedPAC notes that rates paid to PACE programs continue to be based on a system that pre-dates the Affordable Care Act, which changed the payment system for Medicare Advantage (MA) plans.  Moreover, PACE rates include an adjustment related to dementia and an adjustment for frailty, neither of which is factored into payments for Medicare Advantage plans.  These adjustments are to compensate for the relatively higher costs associated with individuals who need a nursing home level of care. 

MedPAC believes that changes to the payment system will save Medicare money and result in more accurate payments to PACE providers.  It recommends that PACE programs be paid on the same basis as Medicare Advantage plans, with more refined adjusters to take into account multiple conditions and functional status.  PACE programs would also participate in the MA bonus payments system for plans with a certain number of stars in the quality rating program.

MedPAC also recommended, with respect to PACE, that enrollment be open to individuals under the age of 55 who require a nursing home level of care  MedPAC also recommended that payment be available for less than a full month, to allow programs to enroll individuals as the need for long-term care services arises.  It also recommended that Congress establish outlier protection for the first three years of a program's operation, to account for exceptionally high cost beneficiaries.

Medicare Advantage Special Needs Plans (MA SNPs)

MedPAC reports that about 500 SNPs are currently serving 1.4 million Medicare beneficiaries.  While not all of those individuals are dually eligible for Medicare and Medicaid, a very high proportion of them are.  Dual-Eligible SNPs (D-SNPs) enroll about 1.16 million dual eligibles (they cannot enroll beneficiaries who are not dually eligible).  SNPs for people with chronic conditions (C-SNPs) and for people needing a nursing home level of care (Institutional SNPs or I-SNPs) include dual eligibles in their enrollments, but these numbers are not broken out from overall enrollment numbers.  Because these are all Medicare plans, even if they serve dual eligibles, they are generally only offering Medicare services.

A subset of D-SNPs are called FIDE-SNPs, or Fully Integrated Dual Eligible SNPs.  These plans, as their names suggests, integrate Medicare and Medicaid services together.  MedPAC notes that, using a FIDE-SNP definition that includes coverage of all Medicaid primary, acute and long-term care services, fewer than 20 such plans, enrolling about 23,000 beneficiaries, existed in February 2012.

Using existing data sources (and noting their limitations), MedPAC concluded that D-SNPs performance was mixed.  On five measures, the D-SNPs outperformed non-SNP MA plans; on 11 there was no difference and on 29, D-SNPs performed worse than non-SNP plans.  FIDE-SNPs performed better than both regular D-SNPs and non-D-SNP plans on the specific measures that SNPs are required to report.

With respect of spending, MedPAC notes that SNPs are paid, on average, four percent higher than the cost for comparable beneficiaries in traditional Medicare and FIDE-SNPs are paid eight percent more.

MedPAC raises the question of whether SNPs should be permitted to use Medicare dollars for non-Medicare covered services, as PACE programs can, to promote keeping individuals out of institutions.  It makes no recommendations on either payment levels or use of Medicare dollars for non-Medicare services.

CMS Demonstrations on Integrated Care

MedPAC devotes five pages of its report to comments on the demonstration initiatives currently underway under the auspices of the Medicare and Medicaid Coordination Office (MMCO) of the Centers for Medicare & Medicaid Services (CMS) to integrate service delivery and financing of Medicare and Medicaid services for dual eligibles.  After describing the initiatives and the authority for them, MedPAC raises issues and concerns as falling into three areas: the large proposed scope of the demonstrations, the standards for the plans that participate in the capitated (per-member-per-month fee for covered services) models and passive enrollment.  It notes that these characteristics could have negative effects on dually eligible beneficiaries' access to and quality of care.

Scope of Demonstrations

Noting that many states propose to enroll their entire dual eligible population into the demonstrations, MedPAC notes that this makes them appear to be large scale program changes, rather than demonstrations, though their effectiveness and quality has yet to be proven.  It raises the question of whether plans will have the capacity to serve such large numbers of individuals.  It notes that such a large scale would make it difficult to move people out of the program if the program proves ineffective.  Moreover, MedPAC notes, the scale makes evaluation complicated.  If all dually eligible beneficiaries are enrolled in the demonstration, no comparable group outside exists against which to compare the demonstration's performance.

Plan Experience

Referring to its examination of FIDE-SNPs, it notes that only about 20 health plans have experience being capitated and at risk for all Medicare and Medicaid benefits. Those plans do not operate in all the states that have submitted proposals for the demonstrations, they do not even operate state-wide where they exist, and they do not serve all of the sub-populations within the disparate world of dual eligibles.  MedPAC notes that few standards by which plans will be evaluated are publicly known and that those that are "preferred" by CMS may be changed in its negotiations with each state.

MedPAC states that plan participation standards should be transparent and should consider quality rankings, provider networks, plan capacity and experience with Medicare and Medicaid services for dually eligible enrollees.  It also raises the question of whether states will have the resources required for the very necessary monitoring of access to, and quality of, care.

Passive Enrollment

MedPAC describes the enrollment process that CMS and the states propose as "passive enrollment with an opt-out provision…Under this enrollment strategy, states will assign beneficiaries to a health plan through "intelligent assignment" unless the beneficiaries opt-out of the demonstration or select a health plan." (p. 88)

Noting that this strategy could be effective as a way to increase enrollment in integrated care plans with proven experience in serving the population with high quality care, MedPAC questions whether states have the resources to make effective assignments that will truly meet individuals' needs and whether, in fact, every plan in the demonstration will offer high quality care.  It notes that beneficiaries will need good education about their choices and continuity of care.  It questions whether these features exist as well as whether plans will have the capacity of undertake an assessment of each beneficiary's needs shortly after enrollment.

Additional issues

Finally, MedPAC notes additional issues it wishes to consider with respect to the demonstration projects: whether savings should be taken out upfront, by paying plans less than the current cost of serving the population (it raises the possibility that they should not); how risk adjustment should be made to the payments to account for the specific needs of the population being served; and how and what data will be collected and how demonstrations will be evaluated.

Conclusion

The MedPAC Report raises many issues about demonstration efforts to integrate Medicare and Medicaid.  These issues have been raised for the past year in the advocacy community.  A future CMA Alert will focus more specifically on these demonstrations.

For more information, contact Patricia Nemore (pnemore@medicareadvocacy.org) in the Center for Medicare Advocacy's Washington, DC office at (202) 293-5760.