Dear Chairman Baucus and Ranking Member Hatch:
Thank you for your continued bipartisan efforts to reform the Medicare physician payment system. Physicians are the linchpin of our health care system, and Medicare physician payment consequently has system-wide consequences for the well-being of Medicare beneficiaries and health care costs. We appreciate your ongoing work on this critical policy issue and the opportunity to comment on the SGR Repeal and Medicare Physician Payment Reform Discussion Draft. Our comments reflect our experience in care improvement and payment reform, and the work conducted in the Richard Merkin Initiative on Payment Reform and Clinical Leadership and the “Bending the Curve” project at the Brookings Institution.
The Committees’ discussion draft provides a path forward to permanently reform the Sustainable Growth Rate (SGR) formula and gradually shift the physician payment system to one that rewards physicians for providing better quality, higher-value care for patients. Permanent reform of Medicare’s physician payment system is urgently needed, and the Senate Finance and House Ways & Means proposal builds on previous legislation from the House Energy and Commerce Committee to create major progress toward effective reform.
The framework outlined in the discussion draft will create much-needed payment stability for physicians. In addition, the proposal provides a clear, coherent pathway for physicians to move away from Medicare reimbursements based on the volume and intensity of specific services – payments that even with a permanent “fix” are likely to become increasingly tight and out of step with the personalized, coordinated care that individual patients need. Instead of FFS payments, physicians will receive additional payments for implementing changes in practice to improve care, and to adopt alternative payment mechanisms that provide better financial support for care delivery reforms that improve quality of care and in so doing may reduce overall health care costs. We agree with the basic framework of the physician payment reforms put forth in the discussion draft, and believe it takes an important step toward person-centered payments for health care. We describe some ways to make further progress toward achieving this goal.
Our comments focus on the Value-Based Performance (VBP) Payment Program, the Alternative Payment Models (APMs), and the quality measures and data infrastructure that are foundational to the success of these payment reforms. While our comments focus primarily on physician payment reform, we also address Medicare reforms that would improve support for other clinicians that are part of effective clinical care teams.
Our specific recommendations to support physician payment reform that facilitates person-centered care include taking additional steps in the following areas:
- Strengthen incentives to provide high-value care by decoupling the bonus payments in the consolidated Value-Based Performance (VBP) Payment Program from a physician’s recent fee-for-service payments, and provide clarity in advance to physicians about how their VBP scores translate into payments.
- Ensure that the APMs represent an increasingly significant shift away from FFS, not just in revenue share but in care delivery, by requiring APMs to be actuarially neutral in terms of overall Medicare spending (not counting initial bonus payments).
- Accelerate the timely availability of relevant APMs to a broad range of physicians to increase physician participation.
- Encourage the development and use of meaningful outcome and care experience measures.
- Assure that physicians have timely access to actionable Medicare data on their beneficiaries to enable physicians to monitor their performance and identify opportunities for improvement.
- Consider financing permanent physician payment reform through reforms in other Medicare payment systems, particularly post-acute care, that reinforce the same movement away from fee-for-service and into more person-centered payments.
- Make progress toward the payment reforms outlined here even if permanent reform is not feasible as part of the end-of-year legislative process.
Strengthen the incentives to provide better, higher-value care for each patient in the Value-Based Performance (VBP) Payment Program.
Currently, physicians receive small increases in their payment rates for reporting on quality through the Physician Quality Reporting System (PQRS), and small increases in rates for meeting the “meaningful use” standards for adoption of health information technology. In addition, starting in 2014, Medicare is scheduled to phase in a “value-based modifier” (VBM) that will also adjust payment rates based on measures of the quality and efficiency of care. Each program is also scheduled to phase in penalties for non-participation or (in the case of VBMs) poor performance in the next several years. An understandable concern of physicians is that multiple, small payment adjustments make participation in each program relatively burdensome and complex given the small payoffs. To address this, the Centers for Medicare & Medicaid Services (CMS), with support from outside organizations such as the National Quality Forum and physician advocacy groups, has been taking steps to encourage consistency in the measures and reporting requirements across the programs.
The Ways & Means/Finance reform framework takes a major step toward consolidating quality measures and incentives by combining these programs into a single “value-based performance” (VBP) program starting in 2017. Under the VBP program, provider payments would be adjusted at the end of each year to reflect performance on a composite index that takes into account quality measures, resource use, participation in clinical practice improvement activities, and meaningful use of electronic health records (including use of electronic records or registries to report on quality measures). The clinical practice improvement activities are also supported by assistance to small practices, and special attention to rural and underserved areas.
While consolidating the three existing quality programs into a single VBP program will substantially increase the impact of the various value-based programs, the simplified VBP program still appears to be a payment rate adjustment that amplifies the current FFS payment system. That is, the value adjustment remains a multiplier for a physician’s FFS payments. The effect mathematically is similar to the current FFS adjustments related to quality: physicians can increase their payments by getting better quality scores, or by billing more services; and the more they bill, the bigger the effect of the VBP multiplier. As a result, physicians will still be pulled in different directions. On the one hand, higher payments for better quality and value will provide important new support for their activities to improve quality and avoid unnecessary costs. On the other hand, because these payments apparently remain multipliers to FFS payment rates, they will also yield a higher pay off when providers bill a greater volume or intensity of services, rather than when they take steps to improve care that are often uncompensated under the FFS system or to avoid unnecessary services. This conflict could be avoided by decoupling the VBP payment from recent fee schedule billing – for example, by providing the payment as a flat bonus to a physician or group based on their VBP score, by tying the VBP bonus to the number of beneficiary cases or episodes treated (not the intensity of actual billing), by tying it to FFS billing in a base year, or by tying it to some other index independent of the actual intensity of services billed.
We also support the Committee’s proposal to direct a substantial share (8 percent increasing to 10 percent or more) of Medicare physician payments into a pool that would be used for VBP payments. However, this approach might imply that physicians will be paid based on VBP scores relative to each other, rather than on achieving a given level or improvement in VBP scores. We believe that it should be possible, through the usual rulemaking process, to use past performance to inform physicians in advance about the level of performance and/or improvement in VBP scores needed to achieve payment bonuses or penalties. More important than the share of payments involved may be the magnitude of differences in payment for different VBP scores or improvements in scores. As in previous Medicare payment reforms, we believe that rewards and penalties should depend at least in part on improvements in scores to account for providers who may have more challenging patient populations or more limited capabilities at the start. By reducing uncertainty about what the VBP means for their practices, setting specific performance goals in advance should increase physician acceptance of the new program and result in greater performance improvement activities.
Recommendation 1: Decouple the VBP bonus payments from a physician’s recent fee schedule billing, so that the VBP bonus does not automatically reward greater intensity, and provide clarity in advance to physicians about how their VBP scores translate into payments.
Ensure APMs are effective in shifting toward payments that reward physicians for better quality care and efficient resource use, and encourage greater physician participation in APMs through significant startup bonus payments.
The most important long-term feature of the payment reform is a pathway for physicians to transition from pure FFS payments in Medicare to payments in Alternative Payment Models (APMs) that promote quality care appropriate to the patient, not just more services. We and others have been working with physician groups on a range of such proposals, including medical homes (including in specialty areas like oncology) and other case-based payments for physicians, “mixed” systems with partial case-based payments combined with FFS, and bundled payments across providers. In primary care and a growing range of specialty care areas, such reforms are not only in development but are in increasingly widespread use.
In the new framework, providers who participate in APMs that meet minimum standards could receive payments through these programs instead of the VBM program, and receive additional financial support. Eligibility for the bonus payments requires providers to receive at least 25 percent of their overall Medicare revenues in APMs (for example, through case- or episode-based payments) in 2016-2017, then at least 50 percent of Medicare revenues in 2018-2019 and 75 percent beginning in 2020. Alternatively, after 2017, providers can also achieve the threshold based on total practice revenue, including revenue from other payers (at least 50 percent total and 25 percent Medicare in 2018-19, and 75 percent total and 25 percent Medicare in 2020-21). While this may seem like an aggressive transition time frame, it seems feasible based on pilots in Medicare and private plans today, and the growing activity around APMs in many clinical areas.
It will be important to ensure that the APMs represent an increasingly significant shift away from FFS, not just in revenue share but in incentives to avoid unnecessary costs while improving quality. To accomplish this, the legislation could include additional clarity about the types of meaningful payment reforms that would qualify as an APM. Simply “paying for performance” by adding in new payments for additional services or a new case-based payment to existing fee-for-service payments is not likely to be enough to support substantial changes in practice that improve quality and reduce costs. To address this, APMs could require an “actuarial offset” for additional physician payments in the APM. For example, a new case-based payment could be financed by an incremental reduction in fee-for-service payments, or could come from demonstrated savings in non-physician costs, but it would not simply be an add-on to the existing payment system. Physicians could still get significantly greater net practice revenues because of the changes that they implement in care – practice revenues would not go down because the case payments would offset the fee-for-service payment reduction, and savings in other parts of Medicare would provide additional revenues. To cover the significant costs and efforts to implement these practice reforms, the initial bonus payments for adopting an APM should also be significant – five percent each year for 2016-2021. Evidence of savings in non-physician costs (e.g., radiology, physician-administered drugs, emergency department and hospital use, etc.) could also be considered for the actuarial offset. APMs that share overall savings with physicians should be encouraged.
Recommendation 2: Include clearer requirements to assure that, when APMs are implemented, they will achieve improvements in both quality of care and spending – for example, through requiring that any new physician payments in the APM have an “actuarial offset” in reduced fee-for-service payments, and through tying any additional payments to physicians to achieving shared savings. Provide a substantial bonus for APM adoption to address the costs of improving care with the shift to the APM.
Support accelerated development of alternative payment models.
Former Brookings Expert
Mark McClellan is a senior fellow and director of the Health Care Innovation and Value Initiative at Brookings. He has written a number of publications on care delivery reform, physician payment reform, and accountable care implementation in the U.S. and abroad.
Former Brookings Expert
Visiting Scholar - Engelberg Center for Health Care Reform
CMS is piloting many APMs today, a process that can be extremely helpful in turning good payment reform concepts into broadly viable payment arrangements in Medicare. Congress could take additional steps beyond encouraging CMS to test APMs to accelerate their timely availability and relevance, and thus physician participation. Congress could direct CMS to collaborate with physician groups and other experts to develop and implement a specified number of APMs within the traditional Medicare program that meet the Congressional standards within the next several years. This initiative, reinforced by physician group efforts and potentially with funding support through the Center for Medicare and Medicaid Innovation, would increase and speed the availability of APMs to a broader range of physicians.
Recommendation 3: Accelerate the timely availability and relevance of APMs, and thus physician participation, by encouraging CMS to develop and implement a specified number of APMs within the next few years.
Support physician efforts to redesign care delivery and improve quality by advancing the development and widespread use of meaningful performance measures.
Better quality measurement is foundational to the success of these payment reforms. The Congressional proposals take important and necessary steps to advance the use of meaningful quality measures by aligning Medicare’s various quality programs, strongly encouraging the use of electronic health records, promoting transparency, and providing funding for the development and stakeholder endorsement of quality measures. However, as the Senate Finance Committee noted in its roundtable on quality measurement last summer, there are still important gaps in the availability of meaningful quality and resource use measures, particularly in terms of outcomes that matter to patients and key aspects of patients’ experience with care. Further, to have more confidence in such outcome measures, physician specialties, regional collaboratives, health plans and other private organizations have developed and are expanding an increasing array of structural quality measures, process of care measures (evidence-based where possible), and additional, proximate outcome measures.
To enhance the impact of these measurement and improvement efforts, and to make more consistent and meaningful performance measures both within and outside of Medicare, the SGR reform legislation could include a requirement for Medicare to implement a core set of key quality and cost measures that are practically relevant and outcomes-oriented. These measures would gradually replace the many process-oriented measures currently used in PQRS, and should become publicly available for use by providers and consumers. The core set of measures should cover the bulk of the spectrum of clinical care, reflect the six priority areas outlined by the National Quality Strategy, and be produced in consultation with the National Quality Forum (NQF), provider organizations, and additional experts. This might mean, for example, that additional funding for NQF would focus on the endorsement and updating of a limited number of performance measures reflecting key quality and cost outcomes and patient care experiences to support the goal of person-centered care.
One concern with prioritizing the limited funds available for measure development and endorsement to key outcome measures may be that such measures alone are not likely to be sufficient to improve care. As many physician groups and others have noted, achieving improvements in outcomes, care experience, and other meaningful measures is difficult. Because such measures are influenced by many factors and are not the only important aspects of quality, reform should also support a more efficient way to handle the many clinically plausible quality measures that currently make up the bulk of PQRS and other physician performance programs, and the further development of the diverse promising systems to help physicians improve care on these key measures and other measures. Detailed and extensive measures to help physicians identify specific opportunities for performance improvement make up the bulk of a growing array of clinical quality improvement systems today, including some clinical registry programs, some board maintenance of certification programs, some private-sector quality improvement tools, some regional health information exchanges, and a growing number of other resources. However, trying to develop a detailed certification or endorsement standards for all of these diverse programs and measures would be very costly, difficult to keep up to date, and might restrict innovative approaches to care improvement. Instead, CMS, with assistance from outside groups, could identify some basic, core elements of effective support systems for performance measurement and improvement. Physicians could then be encouraged to use the support system or systems that best reflects their needs and the opportunities for care improvement in their community. The APMs would thus focus on reporting and improvement on the key performance measures, supported by a rich and potentially diverse set of quality improvement programs.
Recommendation 4: Include a requirement for Medicare to implement a core set of practically relevant and outcomes-oriented quality and cost measures in the VBP program and APMs. Also encourage CMS, with assistance from outside groups, to identify some basic, core elements of supporting systems for performance measurement and improvement.
Provide physicians with timely Medicare data on their beneficiaries that enables them to track their performance and identify opportunities for improvement.
Also critical for physicians to succeed under these payment reforms is timely access to actionable data on their beneficiaries. CMS has taken important steps in this direction through new data sharing programs in CMMI payment reform pilots and in the Medicare Shared Savings Program for Medicare accountable care organizations. The approach generally consists of sharing Medicare claims data for beneficiaries who do not opt out, and providing periodic summaries of performance measures calculated from those files, for all beneficiaries treated by the physicians involved in the payment reform. However, Medicare claims files can be hard for physician practices to use (especially smaller practices) and may not be quickly available. In addition, providers may have trouble reconciling the claims data with CMS summaries. As a result, despite important progress by CMS in providing physicians access to data on their beneficiaries, physicians participating in innovative payment reforms still have difficulties evaluating how they are doing and how they can improve on performance measures related to both quality and spending that CMS calculates.
To address this issue, the SGR reform proposal could include the establishment of and funding to support a specific program, either within CMS or through an independent contracted entity, to provide timely, relevant Medicare claims data and the capacity for physician groups to calculate performance measures accurately based on those data. This program would have specific performance standards for data reliability, usability, and timeliness, and for the accuracy and actionability of interim performance measures produced for the physician groups. It would also continue to meet Medicare privacy and confidentiality standards. In addition, this program should be implemented in a way that promotes consistent claims data sharing and measurement standards not just in Medicare but from private insurers and Medicaid plans as well. Such supports should provide special assistance for smaller physician practices, especially those in rural or underserved areas. Medicare’s Quality Improvement Organization program or initiatives like the ONC Regional Extension Centers could provide these kinds of supports.
Recommendation 5: Establish and fund a program to support provider access to timely relevant Medicare claims data on their beneficiaries and enable physicians to use the data for performance improvement.
Consider financing permanent physician payment reform through reforms in other Medicare payment systems, particularly post-acute care, that reinforce the same movement away from fee-for-service and into more person-centered payments.
The estimated budgetary cost of the reform framework was not released, and additional details are likely to be needed for an official score by the Congressional Budget Office – including on issues we have addressed above, such as how likely APMs are to be “budget neutral” versus provide ways for certain providers to get higher payments without improving care, and how the VBP system would be financed and operate. The score is likely to be in the range of $150 billion. This includes the cost of replacing the SGR with a zero percent update over the next ten years, plus some additional costs related to bonus payments for transitioning to alternative payment systems, and the cost of providing the quality and cost measures and the support needed for the physician payment reforms to succeed.
Failing to enact these reforms will not avoid these costs; indeed, Congress has spent approximately $150 billion on short-term SGR fixes over the past ten years. However, financing the short-term fixes generally involved squeezes on other provider payment rates in Medicare, reducing the predictability of payments for other providers and giving them little support for improving quality as well. Implementing a permanent reform that would reduce overall Medicare costs through a person-centered focus could enable physician payment reform legislation to do much more for improving quality and value across the Medicare program.
Bundling payments for post-acute care (PAC) services
Well-developed reforms that could pay for SGR reform include reforms in other Medicare payment systems that reinforce the same movement away from FFS and into more person-centered payments. In particular, current Medicare payments for post-acute care (PAC) vary based on where patients are treated and the intensity of service, rather than the needs of the beneficiary or the value of services provided. CMS has worked extensively on methods for standardizing the assessment of patients at discharge, as well as tracking their functional status and complications in a consistent way across care settings and over time, enabling the development of a person-centered PAC payment system. A transition to partial or more complete case-based payments based on beneficiary needs could do more to promote appropriate and well-coordinated care for beneficiaries than other recent proposals to simply reduce payments to various PAC providers, while still achieving significant budgetary savings.
Analogous reforms to make payment rates more equal for certain ambulatory and outpatient procedures that are currently reimbursed at much higher rates in hospital outpatient departments compared to physician offices and ambulatory surgical centers could also provide savings while encouraging higher-value care.
Reforms in Medicare benefits and Medigap to support higher value
Reforms in Medicare benefits and Medigap that could reduce Medicare costs without increasing payments by beneficiaries have also been widely proposed, including by us. These reforms include a single deductible, modernized copayment, and an out-of-pocket limit on Medicare spending, along with reforms limiting or imposing fees on “first-dollar” Medigap coverage to reflect their impact on program costs. They can be implemented in ways that give beneficiaries better protection against high medical costs and that reduce beneficiary out-of-pocket spending on average, along with reducing overall Medicare spending. These reforms would also reinforce the physician payment reforms, because beneficiaries would have more opportunities to save money when they engage in getting better care – for example, by enrolling in a lower-cost Medigap plan with lower copays for choosing providers who achieve better outcomes at a lower cost as reflected in the VBP system.
Other potential Medicare reforms with bipartisan support
Finally, other Medicare reform proposals have garnered some bipartisan support as well, such as greater means testing of Medicare benefits. Putting together such a set of financing reforms to offset the cost of physician payment reform will be difficult in the limited time ahead. But if the bipartisan momentum for reform can extend to the financing side as well, the proposal could provide needed political momentum for at least a “small bargain” on the overall Federal budget at the end of the year.
Progress in a Short-Term Reform Alternative
The Committees are appropriately focusing on a permanent reform of the SGR, for all the reasons we have described. However, in the event that permanent reform is not feasible as part of the end-of-year legislative process, it is possible to move forward on the initial steps of the payment reform described in the Discussion Draft. In particular, a zero update in Medicare physician fees for one or two years could be implemented at a cost of only about $18-36 billion. The transition of the various quality-related payment incentives in Medicare to a more comprehensive Value-Based Performance Payment program in 2017 could be accomplished with little net budgetary cost. While a long-term bonus payment for moving to APMs would not be feasible, Congress could provide a more limited funding pool for significant bonuses for providers who move to early APMs that become available between now and 2015. These bonus payments could be front-loaded to fit primarily within a two-year window, and would have limited budgetary cost because only a more limited range of physicians would be ready to participate and have APMs available. For example, physicians switching to APMs before 2015 could get a bonus of 5 percent for two years, and/or those switching before 2016 could get a one-year, 8 percent bonus. This would provide important momentum for APM development, and this initial program would provide valuable experience as well as a foundation for later, more comprehensive Medicare physician payment reform proposals. Some funds should also be included in short-term legislation to move forward with the reformed financing for quality measure development and use (focusing on key measures that involve outcomes that are meaningful to patients, and backed by supporting quality improvement systems and programs) and the support for better data for physicians to help them improve care and lower cost. These additional programs would likely add about $5-10 billion to a two-year “fix,” could be financed by the same kinds of reinforcing provider payment reforms that we have described above, and would represent real progress toward the permanent solution proposed by the Committees.
We thank you again for your leadership and the bipartisan effort to address this longstanding issue of repealing the SGR and implementing meaningful reforms to the Medicare physician payment system. We appreciate the opportunity to offer our comments on this promising, yet challenging endeavor. We look forward to continuing to work with you on these critical issues, and hope you will consider us a resource as these efforts move forward.