VIEWS: 4 PAGES: 5 POSTED ON: 6/21/2011
JUN -- 6 2011 COMMITTEE ON WAYS AND MEANS u.s. HOUSE OF REPRESENTATIVES WASHINGTON, DC 20515 June 6, 2011' Donald Berwick, MD, Administrator Centers for Medicare & Medicaid Services Department of Health and Human Services Room 445-0, Hubert H. Humphrey Building 200 Independence Avenue, SW Washington, DC 20201 Subject: Medicare Program; Medicare Shared Savings Program; Proposed Rule, Federal Register 76, No. 67 (April, 7, 2011) [CMS-1345-P] Dear Dr. Berwick: I write in response to the Centers for Medicare & Medicaid Services' (CMS's) proposed rule for the Medicare Shared Savings Program, also known as Accountable Care Organizations (ACOs), implementing Sec. 1899 of the Social Security Act as added by Sections 3022 and 10307 of the Patient Protection and Affordable Care Act (PP ACA) (Pub. L. 111-148). First and foremost, I congratulate your staff on this immense and complex undertaking. Their work under a tight deadline and in collaboration with other federal agencies is truly commendable. ACOs have the potential to address two of the most troubling problems in our delivery system fragmented care and the perverse incentives inherent in the fee-for-service (FFS) payment model - by encouraging providers to work together to furnish coordinated, high quality, efficient health care. They can also facilitate relationships between patients and their providers and have the potential to achieve better care for individuals, better health for populations, and reduced per capita costs. However, I recognize that while ACOs have great promise they may not be appropriate for all providers, and that is why the Affordable Care Act included many new tools to encourage better coordination of care and improved outcomes for beneficiaries. My comments reflect support for the beneficiary assignment proposal, which encourages pursuit of improved care for all beneficiaries. I also support the move to two-sided risk, which is important in ensuring good stewardship of Medicare trust fund dollars. I do have concerns with other provisions of the proposed rule as I believe it is important to mitigate the potential for ACOs to avoid at-risk populations, preserve beneficiary-access to safety net and teaching hospitals, and access to care for medically under served populations. As discussed in more detail below, CMS .. may be able to address these concerns through the flexibility afforded in Section 1899(i) of the ":i"'Social Security Act. - Beneficiary Assignment I support CMS's decision to assign Medicare beneficiaries retrospectively. I understand that many in the provider community would prefer prospective assignment, but fear it could create a two-tier system where assigned beneficiaries receive a heightened level of care and attention while the remainder of the patient popUlation receives a lower level of care. These provider concerns should be ameliorated by the fact that CMS will provide ACOs with information on their historical patient populations, so they will know what patients are likely to be assigned to them at the end of each year. If ACOs are truly committed to providing coordinated, efficient care, they should do so for all of their patients, not just those patients that provide the best financial incentive. Our intent in creating ACOs was to once again use Medicare to drive systemic, positive change in the delivery system. Retrospective assignment helps accomplish this goal by ensuring the best care for all. Risk-Sharing I would also like to emphasize the importance of ultimately achieving two-sided risk in the Shared Savings Program. As noted by MedPAC in their November 22,2010 letter to you, a one sided, or bonus-only, model places all of the risk on the Medicare program, while offering little incentive for providers to control spending within the ACO. 1 I understand that providers are at different stages of preparedness to become ACOs and that such ventures will require extensive and costly investments, but believe it is unwise to assume that start-up costs alone will provide sufficient cost-control incentives in a bonus-only model. Therefore, I commend CMS for moving providers toward a shared-responsibility model and echo MedPAC's recommendation that"[ 0 ]ver time, the two-sided risk model should become the dominant or the only model available.,,2 Again, Congressional intent was to improve care and find greater efficiencies. This can only be achieved if risks are shared by both Medicare and the providers. Monitoring Provider Behavior for Avoidance of At-Risk Patients ACOs that decrease expenditures relative to spending in the original FFS or Medicare Advantage programs share in a portion of the savings. This goal should be accomplished through better care coordination resulting in the provision of fewer duplicative or unnecessary services. However, an ACO can also lower its expenditures via patient selection recruiting and retaining less costly patients and avoiding more costly patients. This is especially problematic for beneficiaries eligible for both Medicare and Medicaid, as they account for a large proportion of Medicare spending. Section 1899(d)(3) of the SSA authorizes the Secretary to impose sanctions on an ACO, including termination from the Shared Savings Program, if the "Secretary determines that an ACO has taken steps to avoid patients at risk in order to reduce the likelihood of increasing costs to the ACO." In the proposed rule, CMS states that it will look for "trends and patterns suggestive of avoidance of at-risk beneficiaries") by conducting medical record audits and '4~!,eneficiary satisfaction surveys, and analyzing claims data. ACOs found to be engaging in o patient selection will be placed on a corrective action plan and face possible termination from the ) I MedPAC letter to Dr. Donald Berwick, November 22,2010, pgs. 2-3. 2 ' " ld., p. 4. 3 76 Fed. Reg. 19625. , = program. Patient selection can often be quite subtle, such as steering patients to providers outside of the ACO or structuring provider hours so they are not convenient for certain beneficiaries. With this in mind, I impress upon CMS the need for vigilance in its monitoring activities and swift action should it .find any improprieties. Beneficiary Access to Safety Net and Teaching Hospitals The proposed regulation includes direct graduate medical education payments (DGME), indirect medical education (IME), and disproportionate share hospital (DSH) payments in both the benchmark and the expenditures. Doing so creates an incentive for ACOs to reduce their .expenditures by steering patients away from safety net and teaching hospitals and toward hospitals that do not receive these payments. Such a policy could adversely affect patient care and beneficiary choice - if a patient is sent to a hospital that is inappropriately equipped to care for his needs because it is financially advantageous for the ACO participant. The resulting loss of support for teaching hospitals could undermine their ability to provide important community services such as uncompensated care, trauma care, bum units; and neonatal ICUs. For these reasons, we strongly urge CMS to exclude DHS, IME, and DGME payments from the ACO benchmark and expenditures in the final rule. Selecting a different methodology for determining . the benchmark and expenditures is comparable to the alternative models contemplated by CMS in the proposed regulation and is clearly within the Agency's authority under Section 1899(i). Section 1899(i) grants the Secretary the authority to use "any payment model that the Secretary determines will improve the quality and efficiency of the items and services furnished under this title" as long as the model does not result in additional spending..In the proposed rule, CMS notes that this authority allows the Agency to: 1) create a hybrid shared savings model that "combine[s] many of the elements of the one-sided model under section 1899(d) of the Act with a risk-based approach under section 1899(i) of the Act,,;4 2) update the benchmark by a local/State factor as opposed to the national factor outlined in Section 1899(d)(1 )(B)(ii);5 and 3) select a net sharing rate methodology for one-sided ACOs that differs from the one outlined in Section 1899(d)(2). 6 . Based on the above assertions in the proposed regulation, the Secretary can utilize the authority. in Section 1899(i) to exclude DSH, IME, and DGME payments from the benchmark and expenditures. In addition to Section 1899(i), principles of statutory construction require that a statute must be read as a whole and in keeping with congressional intent. As Members of the Ways and Means Committee who worked closely on the development of health reform legislation, we are able to provide insight into Congressional intent with regard to the calculation of the benchmark and expenditures. Section 1899(d)(1)(B)(ii) states that the benchmark "shall be adjusted for beneficiary characteristics and such other factors as the Secretary deems appropriate," while Section 1899(d)(1)(B)(i), which deals with the expenditures, does not contain the "other factors" language. Removing DSH, IME, and DGME payments from the benchmark, but not the expenditures, would result in a skewed comparison between the two metrics, and we ~.are at a loss to point to a situation in which doing so would be appropriate. Reading the statute ~o prevent the Secretary from adjusting the expenditures for "other factors" effectively negates 476 Fed. Reg. 19603. s 76 Fed. Reg. 19610. 676 Fed. Reg. 19613. ) ~. / that language with regard to the benchmark. This was not the intent of Congress. Adjustments made to the benchmark should also be made to the expenditures in order to provider a meaningful comparison and I encourage CMS to remove DSH, IME, and DGME payments from both the benchmark and the expenditures in the final rule. Failing to do so could limit beneficiary access to high quality, innovative care, and even destabilize broader community services for areas in which ACOs are dominant. Updating the Benchmark I am concerned that the methodology selected to update the benchmark in the proposed regulation disadvantages high cost areas while bestowing an unearned benefit on low cost areas. Each ACO will have a benchmark, or baseline, against which its expenditures will be measured. , " The benchmark will be set at the beginning of each agreement period and will be updated yearly to account for normal growth in Medicare spending. Section I 899(d)(I)(B)(ii) states that the benchmark shall be "updated by the projected absolute amount of growth in national per capita expenditures," meaning the national growth rate will be converted to a flat dollar amount and that amount will be added to each ACO's baseline. CMS also contemplated using its authority under Section I 899(i) to create an alternative update factor that would take into account the local/State rate of growth. Ultimately, CMS proposed to use the update factor outlined in the Act, but sought comments on alternative proposals. There a number of valid reasons why some areas have higher costs, including higher wage and real estate expenses, and the underlying health and socio-economic status of the patient population. The current methodology penalizes certain providers for factors outside of their control. To address this disparity, CMS could use its authority under Section I 899(i) to implement an update factor that provides strong motivation for savings in both low cost and high cost areas. This might include something akin to the hybrid methodology proposed for the Pioneer ACO model, a local/State growth factor, or a blend of national and local/State growth, \ any of which would provide a more appropriate growth factor for high cost areas and increase the incentives for savings in low cost areas. Access to Care for Medically Underserved Populations The proposed regulation is in danger of effectively excluding FQHCs and RHCs, and their patient populations from meaningful participation in ACOs. While CMS considered using its authority to include FQHCs and RHCs, the agency ultimately determined that it was required to base beneficiary assignment on the utilization of primary care services provided by a physician. Because FQHCs and RHCs frequently utilize non-physician practitioners and do not currently collect the specific data necessary to determine the exact service furnished and the type and specialty of the provider rendering that service, CMS concluded that they cannot assign beneficiaries to an ACO based on care they receive from an FQHC or RHC. While CMS tried to encourage inclusion of these providers in ACOs by offering a slightly higher amount of shared _ "savings, this provision is unlikely to have the intended outcome because patients that receive ~bltheir primary care services from FQHCs and RHCs still will not be assigned to the ACO. In order to ensure the inclusion oflow-income patients most in need of an ACO's care coordination and quality improvement efforts, I encourage CMS to include FQHCs and RHCs as eligible providers, which can be done by invoking Section'1899(i). This new model would recognize that FQHCs and RHCs "predominately provide primary care services to their populations,,7 and assume that any visits to these institutions are for primary care. It would also waive the physician only assigilment provision contained in Section 1899(c) and allow patients to be assigned based on the provision of services by both physicians and non-physician practitioners. I appreciate the opportunity to provide my recommendations to CMS and hope that you will take them under advisement when developing the final rule. If you have any questions or would like further infonnation, please contact Jennifer Friedman of my Committee Staff at (202) 225-4021. Pete Stark Ranking Member Subcommittee on Health 7 76 Fed. Reg. 19538.
Pages to are hidden for
"Pete Stark Letter on ACOs"Please download to view full document