January 16, 2026

Dear Co-Chair Murphy, Co-Chair Joyce, and Chair Schrier:

On behalf of Accountable for Health (A4H), we appreciate the opportunity to provide feedback on the modernization of programs related to the Medicare Access and CHIP Reauthorization Act (MACRA).

A4H is a non-partisan, national advocacy and policy organization accelerating the adoption of effective accountable care. We aim to inform policy that advances the movement in the health care system toward accountable care that achieves better outcomes, improved care experiences, increased access and lower costs. We have a shared goal of ensuring that federal payment policy meaningfully improves care quality and affordability while reducing unnecessary administrative burden on clinicians.

Medicare’s accountable care models have demonstrated that, when properly structured and supported, they can improve care experiences, expand access, and lower costs. For example, most of Medicare’s Accountable Care Organization (ACO) models and several specialty models have produced savings for Medicare while improving care through strategies such as improving timely access to care, making it easier to find high quality specialty care, improving communication, and coordinating care across settings.¹ CMMIs work must continue without undue interference and any legislative strategies should focus on expanding its role in transforming care delivery rather than contracting it.

In 2015, MACRA laid out a framework to advance the adoption of accountable care in traditional Medicare. MACRA included “u-shaped” incentives for advancing alternative payment model participation. This included a 5% bonus for a period of time that would eventually drop to 0% and be replaced with a favorable conversion factor update for participants in advanced APMs beginning in 2026. The conversion factor differential, which starts as 0.5% (a 0.75% increase for those in advanced APMs as compared to a 0.25% increase for those that are not) compounds over time.

MACRA’s incentives worked to grow participation in advanced APMs and more can be done to accelerate the transformation. Importantly, MACRA showed success in transitioning providers to accountable care.

These APM incentives and growing participation in the models have transformed care. The availability of the incentive payments has supported investments in new staff and expanded approaches to population health management. Broader participation has added benefits to patient populations in terms of better health outcomes, better care coordination, and care teams that meet people’s total health needs. Additional detail on the importance of these incentives to invest in the infrastructure for accountable care is available in BRG’s Transitioning to Value-Based Care: Financial Implications for Providers and Policymakers.

Although MACRA has been successful in driving early adopters to enter accountable care models, new tools are needed to encourage late adopters to move into these models and to encourage ongoing participation in two-sided risk. Unfortunately, at the time we need more tools to attract participants, the existing bonus structure works against the goal of getting providers to pursue delivery system transformation.

Below are our comments on your specific questions.

What legislative reforms are most needed to ensure future CMMI models deliver real improvements in cost and quality, while also ensuring successful scaling of innovations?

The CMS Innovation Center has played a critical role in advancing accountable care. Several structural and operational issues would improve its ability to deliver consistent, durable improvements in cost and quality or to scale successful models. Legislative reforms should focus on predictability, transparency, and sustainability, particularly for physician-led and rural participants.

In addition to the CMMI-specific recommendations below, we encourage you to expand this legislation to include a statutory requirement for a full-risk ACO option incorporating beneficiary engagement tools (e.g., ability to offer additional benefits and improved options for voluntary alignment for beneficiaries to elect accountable care options). Such legislation would build on CMMI’s successes over the last 15 years with the accountable care portfolio. We have attached as an appendix our legislative recommendations specific to a full-risk accountable care option.

Congress should strengthen and clarify CMMI’s evaluation framework.

Model evaluations drive significant decision-making in Congress and at CMMI about what has worked and what has not in health care payment and delivery system reform. Yet the current model evaluations present limitations that can be addressed. Note that while there may be a role for legislation in this area, we also believe there is a fair amount of interpretation within existing legislative language that should also be examined. Current model evaluations are often slow (released years after the model performance year) and disconnected from the reality of practice transformation. Methodological rigidity, limited stakeholder engagement, and delayed findings reduce confidence in evaluation results and complicate decisions about continuation or scaling. Yet, evaluations are routinely used to inform changes to models, which have significant financial implications for participants. Legislation and/or instruction to improve evaluation methodologies would materially improve CMMI’s ability to achieve savings and improve health outcomes in the long term.

Evaluators must seriously consider changes to the comparison groups used for model evaluations. Two important shifts have occurred since the original design of the CMMI evaluation process – over half of Medicare beneficiaries are enrolled in Medicare Advantage and over half of the remaining beneficiaries in traditional Medicare are in an accountable care model. Yet, model evaluations compare model participants’ performance against those in other alternative payment models in traditional Medicare. This means all new models must outperform other models in traditional Medicare. In addition, the evaluators do not consider Medicare Advantage in model comparisons, even though MA makes up more than half of the Medicare population. Congress should instruct model evaluators to develop fair comparison groups and to produce a range of analyses of CMMI models, as opposed to producing one unbalanced analysis presented as a pass/fail test.

In addition to evaluating overall model performance, Congress should encourage CMMI to assess common elements that consistently contribute to success across models—such as care coordination, navigation support, and patient engagement. These components are known to enhance quality, access, and patient experience and may reduce costs, even if they are not evaluated as standalone models. To date, CMMI evaluations have emphasized model-level outcomes and rarely undertake systematic, cross-model analyses of these embedded components or patient-centered measures. Including component-level assessments and patient-reported outcomes in the evaluation framework would enable policymakers to identify and scale high-value practices, strengthen future model design, and accelerate system transformation.

Congress should require greater transparency in CMMI operations and data access.

Today, much of the financial methodology behind CMMI models, such as risk adjustment, attribution, quality performance mechanics, are available only to model participants. Often, key documents and details are provided to model participants through model-specific portals that only specific individuals can access.

A key function of CMMI is to test models that save money and improve care and to share that information with the broader health care industry so that other payers can replicate CMMI’s successes. This sharing is interlocked with the goals around achieving savings, not just for traditional Medicare but will spillover improvements across the entire health system. Therefore, Congress should require that CMMI model materials be made public, provided that those materials do not contain protected beneficiary information. Information such as risk adjustment methodology and the underlying data sets used to construct them, should be made publicly available for replication, validation, and analysis. In addition, we have long believed that making more of these tools available publicly would facilitate participation from a wider range of entities, and particularly smaller, rural entities who may not be able to invest in vendor support to replicate these calculations on their own.

Congress should establish clear glidepaths for models to evolve and provide sufficient notice ahead of model terminations.

Clinicians and model participants invest substantial resources to participate in CMMI models. In recent years, abrupt model terminations or major design changes have undermined long-term investments in care transformation. CMMI models should include required notice periods, defined off-ramps, and structured transitions to successor models to ensure continuity of care and preserve provider investments. Currently, the practice of wind down and termination is left to CMMI’s discretion, and while the most recent model terminations announcements afforded sufficient notice, there is no guarantee of that timeline in the future.

While we understand that the purpose of CMMI is to test models, some of which will end, we also believe that more can be done to communicate the transitions from model to model as model end dates approach. Winding down model participation requires eliminating staffing, rolling back additional benefits to patients and other disruptions, all of which harm beneficiaries and providers. CMMI should be required to engage in early dialogue with stakeholders about future plans for models and provide more transparent processes around model termination or transition strategies.

Congress should create clear pathways to permanence for successful models.

The CMS Innovation Center’s authority to test models is most valuable when success leads to predictable adoption. Legislative reforms should streamline the process for making effective models permanent or broadly available, particularly those demonstrating improved outcomes at a lower cost. Today, some of the CMMI experiments have demonstrated success for over 10 years, including accountable care organizations, accountable care models for complex populations, and kidney care models. Congress should clearly delineate pathways for successful models and model elements to become permanent parts of the Medicare program once they are successful without exclusively relying on narrow evaluation methodologies.

Question 2: If MIPS were to be reformed or replaced entirely, what would a new physician-led quality program look like? How can we ensure a new program reduces administrative burdens and is applicable to all types of clinicians in all settings, while focusing meaningfully on real outcomes?

After more than three decades of efforts to design a universally “fair” physician quality reporting program, it has become increasingly clear that the Merit-based Incentive Payment System (MIPS) has failed to achieve its intended goals. Despite substantial investments in time and resources by clinicians and the federal government, MIPS has contributed little to measurable quality improvement while imposing significant administrative burden. The fact is, we are not confident that meaningful clinical quality measures can be developed to measure the performance of individual clinicians because of the small sample sizes.

In addition, contrary to the Congressional intent of MACRA, MIPS requirements have increasingly been incorporated into advanced Alternative Payment Model (APM) requirements. This is particularly evident in the Medicare Shared Savings Program advanced APM tracks, as demonstrated by the transition to electronic clinical quality measures (eCQMs), the information blocking proposal, and the Promoting Interoperability (PI) category. In addition to creating unnecessary costs, these requirements weaken the incentives for providers to participate in advanced APMs.

Congress should eliminate MIPS rather than attempt to redesign or replace it.

Future policies should emphasize moving clinicians into models where population health and shared savings reduce costs, rather than relying on MIPS’ process-heavy quality reporting. Incremental reforms or rebranding efforts are unlikely to address the program’s core limitations, including imposing high administrative burden with little evidence of improved quality or cost control, creating unpredictable and inequitable payment adjustments, disadvantaging small practices, and delaying the transition to meaningful value-based care. Further, even without MIPS, Congress can continue to support quality measurement, technology use and other useful activities that well-position fee-for-service providers to participate in accountable care arrangements. For example, health information exchanges and easily accessible registries can support real quality improvement and claims/administrative data can be used to generate additional insights into cost and quality results.

Continued efforts to retrofit MIPS divert attention and resources from more effective approaches while clinician burnout and disengagement continue to rise. If MIPS cannot be eliminated entirely, legislative proposals should clearly indicate that advanced APM participation is preferred to fee-for-service MIPS participation by ensuring the MIPS bonus cannot be larger than any advanced APM incentive, whether that is a bonus payment, an increased conversion factor, or any other payment incentive. In addition, future legislation should direct CMS to disentangle APM reporting requirements from MIPS requirements to the greatest degree possible, consistent with Congress’s intent when passing MACRA.

Congress should adopt policies that incentivize participation in Advanced Alternative Payment Models (APMs) over fee-for-service arrangements.

MACRA’s incentives have provided a powerful driver to move clinicians into advanced APMs. The availability of the incentive payments has supported investments in new staff and expanded approaches to population health management. However, while the total number of clinicians participating in advanced APMs has continued to grow over time, and the number of MIPS-eligible clinicians has gone down, there is still substantial opportunity to move more fee-for-service providers to advanced APMs. In 2023, there were still 541,421 clinicians participating in MIPS.²

Rather than constructing a new MIPS-like program, CMS and Congress should prioritize improving advanced APM incentives, expanding model eligibility, and lowering barriers to participation so that clinicians can transition into value-based arrangements that better reflect how care is delivered. Congress should establish an expectation that all primary care providers and most specialists will continue to join APMs and bend all relevant payment incentives and requirements in that direction through the following:

• Retain a positive payment differential for advanced APM participation with lower payment for remaining in fee-for-service. This can be done in a way that is targeted and clear to drive participation in advanced APMs that save money and improve care. For example, to transition to incentives that are value-based not volume-based, Congress could create a per-aligned beneficiary incentive rather than an incentive based on claims. To strengthen the incentive for APM participation, Congress should require that bonuses are paid in the APM performance year, rather than two years after performance has occurred.
• Focus quality measurement on a limited set of meaningful outcome measures. Like ACO REACH, quality accountability should rely on a limited set of meaningful outcome measures, patient experience indicators, and population health metrics that are relevant across specialties and care settings. Measures should be embedded within payment models rather than layered on top as separate reporting requirements.
• Ensure that reducing administrative burden is a central design principle. Any future quality framework should minimize manual reporting, leverage existing clinical data, and avoid duplicative or low-value measures. The goal should be to allow clinicians to focus on care delivery, not compliance.

A4H appreciates the opportunity to provide comments on these important issues. Transforming our nation’s healthcare system from its reliance on fee-for-service to one that is accountable for health outcomes is critically important. We look forward to continuing to work with both the GOP Doc Caucus and the Democratic Doctors Caucus to advance legislation that strengthens Medicare payment policy and improves care for patients with chronic and complex needs.

Sincerely,

Mara McDermott
CEO
Accountable for Health

View Appendix A: SEC. Full Risk ACO Program on page 7 of the pdf.