CBO Issues Cost Estimate for Creating High-Quality Results and Outcomes Necessary to Improve Chronic Care Act

WASHINGTON, Aug. 1 -- The Congressional Budget Office issued the following cost estimate:

S. 870

Creating High-Quality Results and Outcomes Necessary to Improve Chronic (CHRONIC) Care Act of 2017

As ordered reported by the Senate Committee on Finance on May 18, 2017

SUMMARY

S. 870 would affect the Medicare and Medicaid programs in several ways. Specifically, the bill would:

* Modify and extend programs that provide services to beneficiaries with chronic

conditions or other special needs,

* Expand use of remote (telehealth) services, and

* Rescind funding dedicated to improving the Medicare fee-for-service program and the management of the Medicaid program.

CBO estimates that enacting S. 870 would not affect direct spending in fiscal year 2018; would reduce direct spending for the Medicare and Medicaid programs by $217 million over the 2018-2022 period; and would have no significant effect on total direct spending over the 2018-2027 period. Pay-as-you-go procedures apply because enacting S. 870 would affect direct spending. Enacting the bill would not affect revenues.

CBO estimates that enacting the legislation would not increase net direct spending or on-budget deficits by more than $5 billion in any of the four consecutive 10-year periods beginning in 2028.

The bill contains no intergovernmental or private-sector mandates as defined in the Unfunded Mandates Reform Act (UMRA).

ESTIMATED COST TO THE FEDERAL GOVERNMENT

The estimated budgetary effect of S. 870 is shown in the following table. The effects of this legislation fall within budget functions 550 (health) and 570 (Medicare).

By Fiscal Year, in Millions of Dollars

2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2017-2022 2017-2027

INCREASES OR DECREASES (-) IN DIRECT SPENDING OUTLAYS

Independence at Home Demonstration 0 0 2 7 7 0 0 0 0 0 0 16 16

Special Needs Plans 0 0 6 13 13 14 14 15 15 16 17 46 123

Value-based Insurance Design Demonstration 0 0 0 40 50 0 0 0 0 0 0 90 90

Telehealth Costs in Medicare Advantage Bids 0 0 0 -10 -10 -10 -10 -10 -10 -10 -10 -30 -80

Telehealth in ACOs 0 0 0 5 5 5 5 5 5 10 10 15 50

Telehealth Services for Stroke Patients 0 0 0 0 10 15 20 25 30 35 45 25 180

Assignment of Beneficiaries to ACOs 0 0 0 5 5 5 5 5 5 10 10 15 50

Use of In-network Providers by ACO Beneficiaries 0 0 0 -5 -7 -7 -7 -7 -7 -7 -7 -19 -54

Rescissions:

Medicare Program 0 0 0 0 -235 -135 0 0 0 0 0 -370 -370

Medicaid Program 0 0 0 0 -5 0 0 0 0 0 0 -5 -5

Total Changes 0 0 8 55 -167 -113 27 33 38 54 65 -217 0

Notes: Budget authority is equal to outlays.

ACO = accountable care organization.

BASIS OF ESTIMATE

For this estimate, CBO assumes that S. 870 will be enacted near the end of fiscal year 2017.

CBO estimates that enacting S. 870 would affect direct spending in each year, beginning in 2019, by a significant amount, but would have no significant net effect on total direct spending over the 2018-2027 period. The provisions that would affect direct spending are discussed below.

Independence at Home Demonstration. The bill would extend the Independence at

Home (IAH) program for two years, through late fiscal year 2019, and would increase the aggregate cap on the number of Medicare beneficiaries served by participating providers from 10,000 to 15,000.

Primary care services provided in a number of settings, including a patient's home, are covered by the Medicare program. The IAH program was established to test whether providing a financial incentive--bonus payments--for providers to deliver primary care services in a patient's home would reduce Medicare spending and improve the quality of care. Providers participating in the IAH program receive a bonus payment if their practice meets quality standards and the average cost of Medicare benefits for its patients is less than 95 percent of the average cost of such benefits for similar patients in the community.1

Those bonus payments would add to federal costs. The ultimate budgetary effect would depend on whether they resulted in offsetting reductions in Medicare spending. However, determining that the patients served by participating providers have Medicare costs that, on average, are below that 95 percent level does not necessarily indicate that the IAH program reduces Medicare spending, because it does not indicate that the program has changed Medicare's costs for beneficiaries served by participating providers. Expanding the use of home-based services through the IAH program would probably increase the use of certain services, but would ultimately reduce Medicare spending if the resulting change in practice patterns lowered health care costs or if the IAH program shifted market share from higher-cost to lower-cost providers, as long as the resulting savings amounted to more than the bonuses paid through the program. To date, interim evaluations of the IAH program have not assessed whether such changes have occurred. In the absence of such information, CBO has no basis for concluding whether the bonus payments offered through the IAH program have spurred participating providers to make changes affecting Medicare spending.

Further, the bonus payments, as designed, are not targeted exclusively at inducing changes to reduce spending. Instead, providers with relatively low costs would qualify for bonuses whether they make any changes in the way they provide care or not. Similarly, providers who do make changes, but do not lower spending by enough to qualify for a bonus would not receive one. On the basis of the bonus payments made to date, CBO estimates that Medicare would make annual bonus payments to participating providers that average about $5 million per 10,000 beneficiaries for each additional year of the demonstration. Taking into account both the 5,000 increase in the cap on the number of participating beneficiaries and the effect of interactions between changes in spending in the fee-for-service sector and payment rates in the Medicare Advantage (MA) program, CBO estimates that the bill's changes to the IAH program would increase Medicare spending by $16 million over the 2018-2027 period.

Special Needs Plans. Special needs plans (SNPs) are private health insurance plans in the Medicare Advantage program that limit enrollment to beneficiaries who require an institutional level of care, have certain chronic conditions, or are enrolled in both Medicare and Medicaid (dual eligibles). Under current law, the authority for an MA plan to operate as a SNP will expire at the end of calendar year 2018.

S. 870 would permanently authorize SNPs if certain requirements are met. In particular, SNPs that limit enrollment to dual eligibles (D-SNPs) would be required to establish formal agreements with state Medicaid programs by January 1, 2021, to coordinate the provision of Medicaid-covered long-term services and supports (LTSS) or behavioral health services. Feedback from stakeholders indicates that state Medicaid programs find that D-SNPs offer an attractive option for identifying and contracting with private insurers to provide LTSS. Therefore, CBO expects that authorizing D-SNPs beyond 2018 would increase the number and the scope of managed LTSS programs covered by state Medicaid programs.

Based on analysis of information from stakeholders, CBO concludes that managed LTSS plans enroll a small number of individuals who otherwise would receive informal, non-federally financed care in the community. Once those individuals are enrolled in a managed LTSS plan, they would receive Medicaid-financed LTSS for the first time.

Compared to current law, CBO estimates that the number of people who would receive Medicaid-financed LTSS under S. 870 would grow over time. That increase would rise to about 1,300 by 2027. CBO estimates that expansion of participation in Medicaid-financed LTSS would increase federal Medicaid outlays by $123 million over the 2018-2027 period. CBO further estimates that permanently authorizing SNPs would not have a significant effect on Medicare spending because CBO estimates that Medicare payments to SNPs, on average, are comparable to Medicare's payments to other MA plans or to providers in the fee-for-service sector.

Value-based Insurance Design (VBID) demonstration. The Center for Medicare and Medicaid Innovation (CMMI) began conducting a demonstration program in January 2017 to test the effectiveness of permitting private health insurance plans participating in the MA program to vary cost-sharing and benefits for Medicare beneficiaries with certain conditions in order to encourage the use of certain services and providers. As with other models tested through the CMMI, the Secretary will be permitted to expand the program if, after evaluating the results of the demonstration program, the Chief Actuary of the Centers for Medicare and Medicaid Services certifies that expansion would not increase Medicare spending and the Secretary determines that the expansion would not reduce quality of care. S. 870 would modify that demonstration project to make VBID available in all 50 states in 2020 and 2021.

Expanding to all 50 states during testing would limit the Secretary's flexibility to design and modify the demonstration. For example, it would be more difficult to focus on elements of the experiment that an initial evaluation suggests might be most promising or to ensure that the demonstration involves a control group that is adequate for the evaluation to produce meaningful conclusions. CBO expects that limiting that flexibility would be unlikely to result in greater savings than a similar model designed and refined under the existing CMMI program but could result in greater costs. Based on that one-sided effect on potential savings, CBO estimates that this provision would increase Medicare spending by a total of $90 million 2020 and 2021. That estimate is in the middle of the range of possible outcomes.

Telehealth costs in Medicare Advantage bids. Under current law, MA plans may

provide some telehealth services as part of the standard benefit, mirroring what is covered for beneficiaries enrolled in Medicare's fee-for-service (FFS) program. However, if an MA plan wants to provide telehealth services that go beyond what is covered in the FFS program, the plan must receive approval to provide those services as supplemental benefits and use its "rebate" to pay for those services.2 S. 870 would allow MA plans to include the cost of additional telehealth services in their bids for contracts that cover 2020 or subsequent years. The costs included in the bid would not include capital or infrastructure expenses. Telehealth services would not count toward meeting network-adequacy requirements, and plans could not use the availability of telehealth services to limit access to in-person services.

Based on a review of the literature and discussions with experts, CBO concluded that coverage of telehealth services by private payers sometimes results in higher spending and sometimes results in savings; in either case, the effects on spending tend to be small. For MA plans that offer telehealth services as supplemental benefits, this provision would increase spending, because Medicare's payment would reflect the full cost of those benefits instead of the 50 percent to 70 percent of the cost that is covered by the rebate. (The other 30 percent to 50 percent is covered by displacing other supplemental benefits that would be attractive to potential enrollees.)

In general, CBO expects that an MA plan that begins or expands coverage of telehealth benefits under S. 870 would do so based on the plan's expectation that it could manage telehealth services in a manner that would enable it to lower its bid. Because coverage of telehealth benefits as a supplemental benefit is very limited, CBO estimates that the savings from plans that begin or expand telehealth services would slightly exceed the increased cost for plans that already offer telehealth services as a supplemental benefit. On net, CBO estimates that enactment of this provision would reduce direct spending by $80 million over the 2018-2027 period.

Telehealth in Accountable Care Organizations. The bill would expand the ability of certain ACOs to receive Medicare payment for telehealth services beginning

January 1, 2020. Under current law, Medicare only pays for telehealth services delivered in rural locations, with the remote provider paid under the physician fee schedule and the originating site receiving a facility fee. Nevertheless, an ACO has an incentive to provide noncovered telehealth services if it expects those services to reduce the total cost of care for the ACO's beneficiaries and to result in larger bonus payments from Medicare.

S. 870 would eliminate the geographic component of the originating site requirement for ACOs and allow those programs to receive Medicare payment for certain telehealth services furnished to the ACO's beneficiaries in their homes. No facility fee would be paid for services provided in the home of a beneficiary. CBO estimates that change would increase direct spending for Medicare by $50 million over the 2018-2027 period.

Telehealth Services for Stroke Patients. Under current law, coverage of telehealth services is restricted to Medicare beneficiaries in rural areas. Beginning on January 1, 2021, S. 870 would remove that geographic restriction for telestroke services (a subset of telehealth services that involves consultation with a neurologist for a patient suspected of having had a stroke).

There are two types of stroke: bleeding in the brain (hemorrhagic) and clotting in the brain (ischemic). Use of a clot-dissolving drug to treat clotting strokes within three to four-and-a-half hours of the onset of symptoms substantially improves outcomes, both by increasing survival rates and by reducing the likelihood that a stroke patient will be moderately or severely disabled. However, administering the clot-dissolving drug to a patient with a bleeding stroke is likely to cause death. Therefore, a timely neurological evaluation is essential to determine whether to administer the clot-dissolving drug to a patient with stroke symptoms.

Emergency medical services in most urban and suburban areas have protocols to identify patients with stroke symptoms and transport those patients directly to a hospital that is a "stroke center." As a result, a large majority of stroke patients in those areas are taken directly to a stroke center. Such a facility always has a neurologist available--either onsite or via telehealth--to determine which drugs to administer to a stroke patient, so enacting S. 870 would not affect outcomes for such patients.

On the basis of an analysis of Medicare claims data, a review of the relevant literature, and discussions with experts, CBO estimates that about 550,000 strokes occur in the Medicare population in nonrural settings each year. Under S. 870, by CBO's estimates, the proportion of those cases that is handled using telestroke services would increase from about 6 percent in 2021 to 14 percent in 2027.

To develop spending estimates for the bill's extension of telestroke services, CBO focused on cohorts of Medicare patients who receive a telestroke consultation in a given year. That approach, which tracks groups of patients over a span, is particularly appropriate when spending is changeable over time. On the basis of a review of the relevant literature and discussions with experts, CBO concluded that spending--by the federal government and nonfederal providers combined--for a cohort would increase in the year in which the telestroke consultation occurs and then decline in subsequent years.

Higher spending in the first year would be the result of: additional consultations, more medications, additional treatment, and--for patients who otherwise would not have survived--more spending for post-acute-care services during the 90 days after a hospital stay. Annual spending would be lower in subsequent years largely because the number of patients who are discharged from the hospital with moderate or severe disability would decline significantly as would spending for long-term care.

Because Medicare does not cover long-term care services such as nursing home care, much of the savings from avoided long-term-care services would accrue to beneficiaries, other private payers, and state Medicaid programs--and not to the federal government. The federal government would share in the savings that accrue to state Medicaid programs.

For a given cohort, CBO estimates that cumulative spending--including spending by nonfederal payers--would be reduced beginning in the fourth year after the telestroke consultation. Federal spending would follow the same basic pattern but with a lag because much of the savings would accrue to nonfederal payers. CBO estimates that federal spending would be reduced beginning in the sixth year after the telestroke consultation.

Taking into account that pattern of an initial increase in spending and a reduction over time for each cohort of patients each year, CBO expects that expanding Medicare coverage of telestroke services ultimately would reduce Medicare spending. Over the 2018-2027 period, however, CBO estimates the expansion of telestroke services would increase direct spending by $180 million.

Assignment of beneficiaries to accountable care organizations. In general, Medicare beneficiaries are assigned to an ACO when they receive much of their primary care from providers affiliated with a particular ACO. For most ACOs, assignment of their beneficiaries is retrospective--that is, final assignment of a beneficiary occurs after analysis of the beneficiary's claims for a year. S. 870 would allow ACOs to have beneficiaries assigned to them prospectively beginning in 2020. CBO estimates that this provision would increase federal spending by $50 million over the 2018-2027 period. That conclusion is based on two factors: First, prospective assignment would result in some beneficiaries being assigned to an ACO who would not be assigned to any ACO under current law. That increase in assignment rates would result in an increase in the number of beneficiaries for whom Medicare makes a "shared-savings" payment to an ACO. Second, it would also weaken the incentive for an ACO to lower costs for beneficiaries who are not assigned to it.

Use of in-network providers by AOC beneficiaries. Under current law, health care providers generally are prohibited from offering financial incentives to Medicare beneficiaries to patronize the provider. S. 870 would waive that prohibition with respect to financial incentives offered by certain ACOs to beneficiaries who receive primary care services from a provider within the ACO's network. CBO estimates that enacting that provision of S. 870 would reduce direct spending for Medicare by $54 million over the 2018-2027 period.

Eligible ACOs would have two potential reasons to offer financial incentives for patients to use primary care providers in their networks. The first would be to increase volume at the expense of providers that are not part of the ACO's network, which would affect providers' incomes, but would not have a significant effect on Medicare spending. The second reason is because the ACO would expect that, on average, a dollar spent on financial incentives would be more than offset by higher "shared-savings" payments. Because shared-savings payments would increase only if Medicare spending was lower, CBO expects that eligible ACOs would use those incentive payments in ways that would result in lower Medicare spending relative to current law.

Rescissions. S. 870 would rescind amounts earmarked for making improvements to the Medicare fee-for-service program and to managing the Medicaid program. CBO estimates those rescissions would reduce direct spending for Medicare by $370 million and would reduce federal spending for Medicaid by $5 million.

PAY-AS-YOU-GO CONSIDERATIONS

The Statutory Pay-As-You-Go Act of 2010 establishes budget-reporting and enforcement procedures for legislation affecting direct spending or revenues. The net changes in outlays that are subject to those pay-as-you-go procedures are shown in the following table.

CBO Estimate of Pay-As-You-Go Effects for S. 870, as ordered reported by the Senate Committee on Finance on May 18, 2017

By Fiscal Year, in Millions of Dollars

2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2017-2022 2017-2027

NET INCREASE OR DECREASE (-) IN THE DEFICIT

Statutory Pay-As-You-Go Impact 0 0 8 55 -167 -113 27 33 38 54 65 -217 0

INCREASE IN LONG-TERM DIRECT SPENDING AND DEFICITS

CBO estimates that enacting the legislation would not increase net direct spending or on-budget deficits by more than $5 billion in any of the four consecutive 10-year periods beginning in 2028.

INTERGOVERNMENTAL AND PRIVATE-SECTOR IMPACT

S. 870 contains no intergovernmental or private-sector mandates as defined in UMRA. CBO estimates that the state share of increased Medicaid spending for higher enrollment in LTSS plans would total $93 million over the 2018-2027 period. Because states have significant flexibility to adjust their financial and programmatic responsibilities, such additional expenditures would not result from an intergovernmental mandate as defined in UMRA.

ESTIMATE PREPARED BY:

Federal Costs: Alice Burns, Lori Housman, Jamease Kowalczyk, Kevin McNellis, Andrea

Noda, Lisa Ramirez-Branum, Lara Robillard, Colin Yee, Rebecca Yip

Impact on State, Local, and Tribal Governments: Zachary Byrum

Impact on the Private Sector: Amy Petz

ESTIMATE APPROVED BY:

Holly Harvey

Deputy Assistant Director for Budget Analysis

1. Measuring the cost of similar patients in the community has proved to be a very difficult technical challenge. As a result, each time the evaluators have analyzed the data for a performance year, they have recommended making substantial changes to how those costs will be estimated for a subsequent performance year. Participating providers have been given the choice of continuing to use the existing method or switching to the newly developed method.

2. The rebate is a portion of the amount by which the "benchmark" amount for the geographic area covered by the plan exceeds the MA plan's bid for services it is required to cover. The benchmark is based on estimated spending per beneficiary in the fee-for-service sector in that geographic area. The rebate portion is between 50 percent and 70 percent, based on the plan's score on certain measures of quality of care. MA plans are required to use the rebate to pay for benefits not covered in the fee-for-service sector.