My name is Andy Allison. I am Arkansas's Medicaid Director, and I run the Division of Medical Services at the Department of Human Services in Little Rock. I have been in this position for a year, and before that Directed Kansas' health care programs, including Medicaid, for 6 years. I'm also the current and outgoing President of the National Association of Medicaid Directors, which I helped establish in 2010 as an independent organization in order to strengthen the voice of program directors in national policy discussions. I want to thank the Committee for inviting me and the other states represented here today to discuss these critical policy issues.
In my testimony I would like to describe for you the decision process Arkansas is going through to determine whether to expand the Medicaid program under the authority granted by the Patient Protection and Affordable Care Act of 2010, and also provide comments on the future of the Medicaid program. The seismic economic shift that began in this country in 2008, along with passage of the ACA in the Spring of 2010, have in my view combined to make the present period arguably the most important in Medicaid's history. The risks and opportunities associated with the twin challenges of expansion and fiscal duress are compounded by what has, at times, been nearly overwhelming uncertainty regarding the future of the program. Unresolved debates in the courts and in Washington regarding the program's size, shape, and funding have made it more difficult for states to plan and improve the program even though the need to plan and improve is greater than ever, as is the opportunity to serve more Americans who need Medicaid's services.
State consideration of the Medicaid expansion: Arkansas faces many challenges. It is one of the poorest states in the U.S. More than three out of four (78%) of Arkansans earn at or below 400% of the federal poverty level. The median household income for Arkansas is the third lowest in the country at $38,413.
Lack of insurance is a significant problem for many Arkansans. One-quarter (25%) of 19-64 year olds are uninsured. This leads to problems in accessing and affording needed health care: 16.5% of Arkansans recently reported being unable to see a doctor due to cost. Arkansas's high rates of uninsurance are associated with growing uncompensated care costs. In 2010, uncompensated care costs to Arkansas hospitals were estimated at more than $338 million. In addition to the costs borne by hospitals for uncompensated care, Arkansas families with health insurance also pay for uncompensated care through increasing premiums. Premiums for insured Arkansans have risen an estimated $1,500 a year to cover the cost of uncompensated care. Arkansas families' average health insurance premiums nearly doubled over ten years to reach $11,816 in 2010.
DHS estimates that expanding Medicaid would result in an additional 250,000 Arkansans receiving coverage through Medicaid. In other states that have already expanded their Medicaid programs, the expansions have led to reduced adult mortality as well as higher levels of health insurance coverage, financial stability, access to health care and health status. [Sources: Benjamin D. Sommers, Katherine Baicker and Arnold M. Epstein, "Mortality and Access to Care among Adults after State Medicaid Expansions" New England Journal of Medicine; 367:1025-1034; September 13, 2012. Katherine Baicker and Amy Finkelstein, "The Effects of Medicaid Coverage -- Learning from the Oregon Experiment," New England Journal of Medicine; 365:683-685; August 25, 2011.]
Medicaid Expansion in Arkansas - status of state's decision: Pursuant to the U.S. Supreme Court's decision in NFIB v. Sebelius, states now have the option to choose whether or not to expand their Medicaid programs to cover low-income adults. After receiving written confirmation this summer from the Centers for Medicare and Medicaid Services that states that cover the expansion group could later drop the coverage, Governor Mike Beebe offered his full support of expansion. His public support followed an already expressed inclination to moving forward after receiving detailed estimates from the Department of Human Services showing a positive net fiscal impact of Medicaid expansion on the state budget, particularly in light of the financial duress that Arkansas hospitals would experience without expansion, and recognizing the tremendous good that health insurance coverage would do for the more than 200,000 low-income adults who will gain coverage as a result of the expansion. A small number of staff in my agency are developing plans for a Medicaid expansion and are working closely with counterparts in the Arkansas Insurance Department to identify ways to enhance continuity of coverage and coordination of care for families and individuals who experience changes in income or family status and need to switch between Medicaid and private sources of coverage in the health insurance exchange.
The decision of whether to expand Medicaid now rests with the Arkansas General Assembly. A supermajority (75%) vote is required to appropriate funds in Arkansas, regardless of source. In order for the Department to extend coverage to the expansion group of poverty-level adults, the Legislature will need to increase my agency's budget by approximately $500 million to cover expenses between January and June 2014. Arkansas's Legislature convenes its 2013 session in January and is scheduled to meet for approximately three months. Arkansas's Legislature will experience significant turnover as a result of the November 2013 elections, and it is too soon to extend any prediction regarding the new Legislature's choice.
Note: DHS' estimates do not include inflation in either the costs or the savings items. Since inflation would tend to increase virtually all cost estimates as well as macroeconomic (tax) effects, its inclusion in the estimates would tend to make all of the dollar amounts in the estimates larger, thereby inflating the nominal value of the projected savings in the out years. To better communicate the impact of expansion to policy makers in Arkansas, inflation was ignored. Dollar costs in each year are intended to reflect "real" 2012 dollars.
Estimated (gross) costs of the Medicaid expansion in Arkansas: Arkansas DHS first released comprehensive estimates of the impact of Medicaid expansion in July 2012, and recently updated those estimates. The newer estimates will form the basis for legislative requests for appropriations that would be necessary as a result of the expansion. DHS estimates that the ACA Medicaid expansion would, if implemented in Arkansas, generate net savings or increased tax receipts totaling in excess of $700 million dollars for the time period from federal fiscal year 2014 through 2025. The gross costs of the Medicaid expansion total approximately $900 million per year by 2017 and include estimated reimbursements and other payment to providers for services covered under the expansion (for the expansion group) or under the regular Medicaid program (for current eligibles who enroll as a result of the "woodwork" effect of the ACA expansion) as well as added administrative costs. DHS estimates that the net state savings of expansion persist in the long run, even after the percentage of expansion costs funded by the federal government fall to 90% in 2020. Along with this testimony, I am sharing the more detailed year by year estimates that we have recently updated, and will now explain the assumptions and analysis that have gone into those estimates.
Enrollment: Arkansas DHS's estimates of the size of the Medicaid expansion use as a starting point the Urban Institute'sMarch 2011 state-level projections of the expansion, which includes an estimate of additional enrollment due to "crowd-out." Crowd-out is when a government expansion of coverage displaces private insurance by inducing individuals to sign up for Medicaid instead of private coverage, or by inducing business to not offer insurance to their employees because some or all could enroll in Medicaid instead. Arkansas'sMedicaid expansion estimates also include an additional 29,000 Medicaid enrollees (over and above the Urban Institute estimates) to allow for additional crowd-out, great program participation due to more effective outreach and enrollment, or other factors that might increase Medicaid enrollment in Arkansas. Overall, DHS's estimates assume a participation rate in Medicaid that is significantly higher than the Urban Institute'sMarch 2011 state-level estimates. This estimate of additional participation helps protect Arkansas from the unexpected costs of enrollment that might occur if crowd-out is unusually high in Arkansas. In DHS's updated November 2012 estimate of the Medicaid expansion, a three-year ramp-up period is included to more realistically project the length of time it would take to reach full participation in the expansion. This three-year ramp-up coincides with the period of highest federal contributions to the costs of the expansion population, and its inclusion has very little (net) fiscal impact on the state.
Costs per person: DHS's estimates of costs per new enrollee in Arkansas'sMedicaid expansion estimates use as a starting point two separate but coinciding estimates of the cost of the adult expansion population. The first source is an estimate for 2011 from the Urban Institute, which pegged per-person costs at around $295 per person using a variety of sources, and which was adjusted for the expected health status and health care needs of the expansion population. The second source is Arkansas's direct experience with working adults in approximately the same income range of the expansion population through ARHealthNetworks - also under $300 per person. ARHealthNetworks is a state-initiated, Medicaid-financed limited expansion of coverage to low-income workers in small businesses. Like the Urban Institute, DHS's estimates assume that new eligibles will be less expensive, on average, than adults currently participating in Arkansas Medicaid due to their higher income, work status, the fact that the expansion population will - by definition - not be categorically disabled, nor will they qualify for the expansion because they (unlike some parents in Medicaid now) have incurred very large health care bills.
There are likely to be some relatively high cost newly-eligible expansion enrollees who have significant physical or behavioral health needs, but are not (yet) disabled. However, there are a limited number of these individuals - and they are only in the new eligible category until they obtain a federal disability status. They will be dramatically outnumbered by low-income workers who are expected to have lower costs (as described above). Costs for these individuals should be incorporated into the Urban Institute estimates we used as a starting point given their comprehensive methodology. Even so, the possibility that there could be more high-cost individuals than Urban already accounts for, or that their costs could be higher than expected, helps explain the use of a per-capita cost figure somewhat higher than Urban's March 2011 estimate.
DHS's estimates or participation and costs per person interact. The most likely new eligibles to enroll are those with the highest costs, i.e., those who most frequently seek services, incur costs, and come into contact with providers who are motivated to help the individual enroll in Medicaid. Arkansas's estimate of participation after the three-year ramp-up period is aggressive, and likely includes individuals with better health and lower health care need than those included in Urban Institute'sMarch 2011 estimates, which served as the starting point for our cost assumptions. However, to be conservative, DHS's estimates do not raise the per-person cost estimates for the expanded Medicaid population during the period in which the federal government is paying the full costs of the expansion (thereby minimizing the expected gains from the expansion's macroeconomic impact on tax revenue), and also do not lower the per-person cost estimate when full participation is reached (and the state begins to pay for a percentage of these costs).
Woodwork effect: Health needs and medical bills are assumed to increase the likelihood of enrollment. Those currently enrolled in Medicaid are presumed to be more likely than those (eligibles) that have not enrolled to have incurred significant health bills, or otherwise have the greatest health needs. Providers, for example, often help facilitate or otherwise encourage patient enrollment in Medicaid to help assure a source of payment for the services provided. As a result, currently eligible non-participants are less likely than participants to have seen a provider. Therefore, those current eligibles that enroll because of the ACA's "woodwork" effect, e.g., from increased outreach, publicity, and streamlined enrollment procedures, are assumed to have fewer and less costly health needs than those already participating in Medicaid.
Administrative costs: Administrative costs: Costs also include administrative expenses associated with both groups of new enrollees (new eligibles and woodwork enrollees). Arkansas's administrative costs are low (approximately 4% of service costs), and include some fixed costs that would not increase with additional enrollees. Arkansas estimates an additional administrative cost of approximately $14 million per year at full implementation associated with the expansion. DHS' estimates include a predicted amount of new administrative spending for the expansion for added costs related to claims processing, provider and customer support, oversight and engagement. The total new administrative costs are estimated at $7.4 million all-funds in 2014 and $14 million all-funds in 2015 and each year thereafter.
The state general revenue impact for administrative costs would be $2.1 million in 2014 and $4 million per year in following years. Costs of the expansion are born primarily by the federal government in each year, but especially in the 2014-2016 period when the federal matching rate for new eligibles is 100%.
Estimates of Savings and increases in State General Revenue due to Medicaid Expansion: There are also expected savings for the State of Arkansas associated with the Medicaid expansion. Projected savings come from three general areas: (1) savings from the natural migration or explicit transition of select Medicaid populations to the newly eligible expansion group; (2) savings from reductions in uncompensated care costs provided by state agencies outside of the Medicaid program; and (3) savings from additional tax revenue associated with new federal spending that is contingent on the state's decision to expand Medicaid.
Transition populations: A number of populations currently served through traditional Medicaid will migrate or will otherwise transition into the new eligible group, resulting in savings to the state. Medicaid transition populations include those currently participating in Medicaid or CHIP who will end up participating in the Medicaid expansion instead. They also include the ARKids B population, since the Federal government essentially picks up the full tab for them beginning in October 2014. Transition populations include ARHealth Networks, Arkansas'sMedicaid waiver expansion group, since the Medicaid expansion (and to a lesser extent new tax subsidies) provides more complete coverage. Transition populations also include some of the state's "medically needy," or "spenddown" Medicaid enrollees. The medically needy group that is assumed to transition to full coverage under the expansion represents those who currently have to medically impoverish themselves in order to reach Arkansas sub-poverty income thresholds. Under the expansion, these individuals will already be insured by Medicaid (and to a lesser extent new tax subsidies) due solely to their low income when they incur large health bills, and will not need to qualify under the older and more restrictive eligibility rules.
ARHealthNetworks is a healthcare benefits program designed for small businesses and self-employed individuals without medical coverage. The population currently on the program will be able to receive coverage in the future via Medicaid expansion or with subsidies through the health insurance exchange, depending on their income level. DHS estimates that transitioning this program will save approximately $12 million a year in state general revenue.
ARKidsB - enhanced FMAP - Arkansas's current FMAP = + additional 32% FMAP increase =100% federal funding for this program starting October 1, 2015 will generate a projected $23 million a year savings in SGR -
AFDC Medically needy spend-down: Arkansas provides temporary Medicaid coverage to parents with low- but excessive incomes who experience a significant health care expense and, after netting out these health care bills, meet the income criteria for the program. These individuals may have suffered a catastrophic acute care cost and have spent nearly all their income on associated bills. Counting these bills against their income as a "spend-down," these individuals are subsequently considered to be "medically needy" for sox month intervals. Following an expansion of Medicaid, parents will be highly likely to have coverage before they incur the high-dollar health care costs, and if so are unlikely to be able to spend down enough income to become Medicaid eligible under the old rules. Although Arkansas would be obligated to maintain this eligibility category even after a Medicaid expansion, DHS estimates that the category will largely de-populate.
Pregnant women: In Arkansas, Medicaid currently pays for nearly 66% of all births. Under current pregnant women eligibility categories, women become eligible due to pregnancy status and lose coverage shortly after the birth of their child. After 2014, with the implementation of the health insurance exchanges and through Medicaid's new eligible group, a large percentage of these women will already be covered before becoming pregnant and their coverage will not be tied to pregnancy status. This will improve continuity of care and coordination of coverage. A single streamlined application will ask about pregnancy status at the time of application. If not pregnant at the time of application, the women will be enrolled in the new eligible group. Arkansas estimates a savings of $21 million in state general revenue from transitioning a large percentage (75%) of pregnant women to the new adult group. We have just learned about the positive impact on coverage of low-income pregnant women. Under 2012 regulations published by CMS, it is now clear that pregnant women will no longer need to wait until they are pregnant to have access to affordable care. Many will be covered through private insurance, and others will already be covered through the Medicaid expansion before they become pregnant. When they become pregnant, they will not need to switch back to Medicaid - nor transition from the new expansion Medicaid group back to the old (existing) Medicaid pregnant women eligibility group. They can simply remain in the affordable health plan they already have. This will promote continuity of coverage, better preventive and prenatal care, and will save the state millions of dollars. Also note that pregnant women above 138% of poverty will not have access to Medicaid after January 2014, and this will help promote private (and continuous) coverage for them and their families. Many more women will already be covered.
Achieving fiscal sustainability in Medicaid. I would like to conclude my remarks by addressing the fiscal duress the Medicaid program is facing. With two negligible exceptions, Medicaid spending in this country has grown as a percentage of the value of the nation's economic output every year since it was created. In Arkansas that growth has been even more pronounced in the last decade: Medicaid spending grew by 1.5 percentage points of the state's economic output over the 2001-2011 period. Obviously that trend cannot continue forever, a point that observers of Medicaid and other entitlement programs have been making for quite some time. Arkansas and other states have recognized that the time to address the long-run imbalance between growth in Medicaid and growth in the tax base that supports it. As it has become clearer in the economic aftershocks of the near-meltdown of our financial markets in 2008 that we now live in a somewhat different economy - and are not simply experiencing an especially long recession -- state interest in reducing growth in Medicaid spending has taken on a new and, in some cases, unprecedented urgency. Since 2010, even with historically high levels of persistent unemployment, Medicaid spending growth has abated as states' imperative to manage costs has grown and the level of activity directed towards wise stewardship of public resources has intensified. In Kansas, where I helped lead state health care programs from 2006 through 2011, the Medicaid program is being transformed by a comprehensive, state-of-the-art implementation of managed care. In Arkansas, we are engaged in comprehensive payment reforms designed to move almost completely away from a fee-for-service system that rewards utilization, and creating in almost every corner of the program, a focus on paying for quality outcomes and efficiency. Our goals are to transform and improve health care, and to slow the rate of growth in costs by as much as 2 percentage points per year. In other markets and with other types of goods, quality and value both improve over time. We are deciding in Arkansas that it is time to expect, support, and incentivize that kind of improvement in health care as well.
State innovation in Medicaid -- Arkansas' Payment Improvement Initiative: Arkansas is in the midst of creating a sustainable patient-centered health system that embraces our Triple Aim: (1) improving the health of the population; (2) enhancing the patient experience of care, including quality, access, and reliability; and (3) reducing, or at least controlling, the cost of health care. Achieving this Triple Aim will require transforming our care delivery system from fragmented and encounter-based care to coordinated, patient-centered and cost-effective care, organized around consumers' comprehensive health needs across providers and over time. It also requires shifting away from pure fee-for-service payment mechanisms that lead to fragmented care with incentives to over-utilize services, to value-based payment mechanisms that reward effective care coordination and superior outcomes with respect to both quality and cost containment. The description I offer of Arkansas' plans is taken from the state's application to the Center for Medicare and Medicaid Innovation's State Innovations Model (SIM) grant program. Arkansas is one of the many states seeking to take advantage of CMMI's offer of financial support, technical assistance, and programmatic flexibility to aid in the transformation of their health care systems through the Medicaid program. Arkansas is requesting support to offset the hundreds of millions of dollars of investments needed in its health system over the next few years to generate billions in savings in the years that follow. This level of federal support is critical in Medicaid, and appropriately reflects the fact that the proceeds of innovation achieved by any one state Medicaid program, reflected in both program outcomes and fiscal relief, will accrue disproportionately to other states and the federal government. Without substantial federal aid and assistance for innovation, states will be faced with the choice to essentially donate most of benefits of its investments in change to other states and the federal government.
Our goal is to fully develop this system within the next 3-5 years by adopting a model that integrates two complementary strategies for promoting clinical innovation on a multi-payer basis across the entire state: population-based care and episode-based care.
Population-based care delivery. Within 3-5 years, most Arkansans will have access to a medical home that offers a local point of access to care and proactively looks after his or her health on a "24-7" basis. Special needs populations with developmental disabilities (DD), those requiring long-term services and support (LTSS), and those with serious behavioral health (BH) needs will also have access to health homes.
- The medical home will support patients to connect with the full constellation of providers who together form their health services team, customized for their personal care needs and with a focus on prevention and management of chronic disease. For patients with chronic conditions, the medical home will assist with monitoring their progress and coordinating care among what will often be a multi- disciplinary provider team. The medical home will bear responsibility for coordinating care to address the complete health needs of a population.
- The health home will be accountable for the full experience of individuals with special needs--the frail elderly, those with developmental disabilities, those with severe and persistent mental illness, and other high needs behavioral health clients. Accountability will include health outcomes, streamlining care planning, and ensuring each person has a single integrated plan across all types of care. To accomplish this, health home providers will work closely with consumers, their families, and other direct service providers, offering support and coaching in a community setting. The health home complements the medical home: the medical home will continue to retain responsibility for diagnosis, treatment, and referral, while the health home will help to ensure proper follow-up, treatment adherence, and communication between providers, individuals receiving services, and their families.
Episode-based care delivery. Within 3-5 years, substantially all acute care and complex chronic conditions (50-70% of total health care spending) will be proactively managed by a principal accountable provider (PAP), who will embrace their role as the "quarterback" responsible for quality, access, and efficiency of all services delivered in response to a patient's immediate needs. PAPs will be evaluated on their performance over entire episodes of care, with an expectation of coordinated, team-based management of services. Better data will help PAPs to understand and improve their performance over time, thus enhancing quality and outcomes and increasing cost-effectiveness of care.
Arkansas has developed and successfully implemented the first wave of episodic payment reforms, which focused on pregnancy and birth, attention deficit hyperactivity disorder, and upper respiratory infections. Arkansas worked closely with providers at the state level to develop the new payment incentives, and achieved regulatory approval at both the state and federal level this past summer. CMS' support for Arkansas' initiative is noteworthy. Arkansas was able to obtain approval from the Centers for Medicaid, CHIP and Survey and Certification (CMCS) for the new payment mechanisms through the standard Medicaid state plan amendment process in less than 90 days despite the ground-breaking nature of the changes, which introduce both positive and negative financial incentives associated with high-quality and efficient use of services. The nature of the amendments that were ultimately approved could serve as a model for other states, and greatly streamlines future additions to Arkansas' payment reforms (and there could be dozens if not a hundred or more to come). To help states transform their Medicaid programs and set Medicaid on a sustainable path, the pace and scale of such innovation will need to increase significantly -- potentially stretching the boundaries of current federal law, regulations, policies, and approval process.
Reduced state spending on uncompensated care for the uninsured: If Medicaid expansion is approved, approximately 250,000 additional Arkansans will have a payer for their care; consequently, uncompensated care provided by state agencies outside of Medicaid should substantially decline. Program areas affected may include health costs for state prisoners, state subsidies to community health centers and community mental health centers, and (unmatched) state funding of the University of Arkansas for Medical Sciences hospital that helps them close the gap on unfunded care. DHS estimates that offset savings for state spending on the uninsured will total $22 million in savings starting in 2014 and will rise to $58 million by 2019 and years thereafter.
Additional tax revenue: Arkansas' decision to expand Medicaid under the ACA would carry with it significant macroeconomic consequences for the state. Without the expansion, federal Medicaid payments to the state (e.g., as reimbursement for added state reimbursements to providers) will be approximately $800 million less than they would be if the state did choose to expand Medicaid. Conversely, if the state Legislature approves the expansion, federal Medicaid payments to the state are expected to grow by around $800 million. Given Arkansas' small size versus the Federal tax base -- approximately 1% of all federal revenues associated with the added federal Medicaid payments would come from Arkansas taxpayers -- Federal Medicaid payments are treated in DHS' estimates as if they came from taxpayers in other states. Put simply, Arkansas' economy will be hundreds of millions of dollars larger if it chooses to expand Medicaid, and this economic expansion will have some impact on state tax revenue. To estimate the impact of additional federal Medicaid payments on Arkansas' economy and tax receipts, the Department made the simplifying assumption that each new federal dollar entering the state's economy through the Medicaid program during a fiscal year would be taxed once at an average rate of 4%. A more sophisticated analysis would have more carefully identified the proportion of federal dollars that would, through increased Medicaid reimbursements and administrative costs, accrue to Arkansas businesses and individual taxpayers as income, the rate of income taxation applied, the consumption rate of spending for those businesses and individuals, and the subsequent rate of state taxation of that consumption spending, including a reasonable assumption about the number of times a new federal dollar might cycle through to new tax-paying entities (i.e., the "multiplier" effect). In lieu of such a sophisticated analysis, the Department instead made the very conservative assumption that the new federal payments would not cycle, and would be taxed just once as income at an average rate of 4%. The 4% tax revenue assumption generates a savings to the state of over $13 million in 2014 and in excess of $30 million in state general savings in years thereafter.
Read this original document at: http://energycommerce.house.gov/sites/republicans.energycommerce.house.gov/files/Hearings/Health/20121213/HHRG-112-IF14-WState-AllisonA-20121213.pdf