Download as PDF

September 16, 2019 

The Honorable Seema Verma
Centers for Medicare & Medicaid Services
Department of Health and Human Services
200 Independence Avenue, S.W.
Washington, D.C. 20201  

Submitted electronically via 

RE: (CMS–5527–P) Medicare Program; Specialty Care Models to Improve Quality of Care and Reduce Expenditures 

Dear Administrator Verma: 

The National Association of ACOs (NAACOS) is pleased to submit comments in response to the Notice of Proposed Rulemaking, Medicare Program; Specialty Care Models to Improve Quality of Care and Reduce Expenditures, as published in the July 18, 2019 Federal Register.[1]

NAACOS is the largest association of accountable care organizations (ACOs), representing over 6.1 million beneficiary lives through 350 Medicare Shared Savings Program (MSSP), Next Generation, and commercial ACOs. Our members, more than many other healthcare organizations, want to see an effective, coordinated, patient-centric health system. Our recommendations reflect our desire to enhance care coordination and health outcomes for Medicare beneficiaries, reduce healthcare costs, and improve quality in the Medicare program. NAACOS shares with the Centers for Medicare & Medicaid Services (CMS) the goal of moving toward a value-based payment system and the cost reductions and quality improvements it brings. ACOs have for years invested in resources, such as data analytics, information technology, and care coordinators, and worked to change institutional culture to focus on prevention and care coordination so that they can succeed in alternative payment models ACOs have been instrumental in Medicare’s shift to value-based care. While the origin of Medicare ACOs dates back to the George W. Bush administration, ACOs have grown considerably in recent years and now includes nearly 560 ACOs, covering more than 13 million beneficiaries. The model is a market-based solution to fragmented and costly care that empowers local physicians, hospitals, and other providers to work together and take responsibility for improving quality, enhancing patient experience, and reducing waste. Importantly, the ACO model also maintains patient choice of clinicians. 

While NAACOS supports the efforts of CMS and the CMS Innovation Center to introduce value-based payment models, various models overlap with each other and – from time to time – conflict with each other, unfortunately. This has been the case with CMS Innovation Center models on bundled payment programs where patients cared for by providers participating in a bundled payment are also assigned to ACOs. We have long called on CMS to resolve the unintended consequences of program overlap by instituting a transparent, agency-wide policy, rather than handling situations on a program-by-program basis.[2]

NAACOS is concerned that CMS has chosen not to prioritize total-cost-of-care models over other episodic-based payment programs, including bundles like the recently proposed Radiation Oncology (RO) Model and ESRD (End-Stage Renal Disease) Treatment Choices (ETC) Model. Most recently, CMS opted to not exclude MSSP Track 3 ACO patients from Bundled Payments for Care Improvement Initiative Advanced (BPCI-A) bundles.[3] This policy has the potential to harm ACOs when produced savings are attributed to the bundled payment and not the ACO. NAACOS is concerned this problem will worsen if the proposed RO and ETC Models are implemented without changes in a final rule. Furthermore, patient confusion can result from their care being handled by multiple programs, and health systems would duplicate resources – meaning waste resources – devoted to multiple coordination efforts.  

NAACOS has a simple solution; Test new models on non-ACO patients and let total-cost-of-care models, like ACOs, operate as expected. Specifically, NAACOS urges CMS to exclude all ACO patients from attribution to any other payment models to reduce duplicative care coordination efforts and create a clear, transparent and understandable policy across all Innovation Center models. 

ACOs have grown tremendously in recent years and now provide coordinated care for almost a fifth of Medicare beneficiaries. With Medicare Advantage caring for roughly a third of Medicare patients, that leaves roughly 26 million seniors in uncoordinated, unmanaged fee-for-service care on which to test new Medicare models. This is a sufficient pool of patients on which to test various demonstrations, such as the proposed RO and ETC Models, without undermining ACO’s long-term, successful efforts. In many cases, ACOs have spent years and millions of dollars investing in tools necessary to thrive as ACOs, including in data analytics, information technology, and care coordinators. That work shouldn’t be hamstrung by competing, mandatory CMS Innovation Center projects. We urge CMS to finalize a policy to exclude ACO patients from assignment to RO and ETC models. 

Total-cost-of-care models have proven to be a superior lever to reducing Medicare spending compared to episodic payment programs. Findings from the first three years of the Bundled Payments for Care Improvement showed lackluster results. Formal evaluations revealed Medicare experienced net losses after taking into account reconciliation payments to participants.[4] The formal evaluation for the Comprehensive Care for Joint Replacement (CJR) model resulted in a mere 0.5 percent savings to Medicare, but those savings couldn’t be concluded with statistical certainty.[5] Comprehensive Primary Care Plus had few effects on cost, service use, and quality for Medicare FFS beneficiaries in the first year.[6]

Meanwhile, a growing body of “counterfactual” data, which compares Medicare spending to what spending would be like in the absence of ACOs, shows ACOs are lowering Medicare spending by 1 percent to 2 percent, which translates into tens of billions of dollars of savings when compounded annually. Researchers at Harvard University, the Medicare Payment Advisory Commission and Dobson DaVanzo & Associates have all done such work. Medicare ACOs in 2017 showed a continuing trend of ACOs saving money compared to the CMS-set benchmarks. CMS acknowledged “spillover” savings in last August’s proposed Pathways rule, pointing out that ACOs lowered Medicare spending outside the ACO program by $1.8–$4.2 billion in 2016 alone in such ways as lower Medicare Advantage payments. 

Furthermore, the Next Generation ACO Model has been praised by the CMS administrator as a shining example of the Innovation Center’s work in value-based care for its year-after-year savings.[7] The CMS actuary certified the Pioneer ACO model for expansion in 2015.[8] Most recently, the ACO Investment Model decreased outpatient hospital spending by 4.4 percent and skilled nursing facility spending by 6.6 percent in its first two years without decreasing quality. Overall, spending dropped by 3.5 percent.[9]

NAACOS welcomes CMS’s recognition in the proposed rule that the new RO and ETC models would overlap with ACO initiatives, when providers operate in both an ACO and one of the new models or when ACO patients trigger an episode under one of the new models. The agency’s willingness to resolve these overlaps is appreciated. Too often Innovation Center models work in conflict with each other, rather than complimenting efforts. For example, coordination payments under CPC+ are counted as ACO expenditures, hurting ACO’s efforts to lower spending relative to their CMS-set benchmark. However, the agency needs to draw a more definitive line that won’t harm total-cost-of-care efforts like ACOs, and we believe excluding ACO patients from assignment to bundles is the appropriate step. 

We are also concerned about the mandatory nature of participation. According to CMS, mandatory models are needed to build an appropriate evidence base on which to test new programs. However, forcing participation onto unwilling providers isn’t healthy in the long-term. CMS should instead focus on creating programs that incent voluntary participation, while also creating potential savings for the broader health system. This is what’s been achieved with the Medicare Shared Savings Program, which has seen a healthy growth in participation since 2012.  

NAACOS strongly opposes mandatory models while supporting the movement to value-based care and encourages CMS to examine alternatives to mandatory participation. If CMS is successful in creating policies that address the concerns of doctors and hospitals, the agency will create a program that’s attractive enough for robust voluntary participation. After which, there should be ample opportunity to find comparable practices by which to measure the success of the proposed RO and ETC models.  

NAACOS has in the past called for appropriately scale testing of Innovation Center models before expansion and providing sufficient safeguards for beneficiaries.[10] Neither the RO or ETC models adhere to those principles and lack evidence on the impact on the healthcare system. The potential negative unintended consequences on patients and providers must be fully considered before moving ahead with any proposed Innovation Center model.  

We believe that’s best done by prioritizing total-cost-of-care models and excluding ACO patients from bundles. NAACOS appreciates CMS’s work to introduce new payment models and hopes a final rule will help all value-based care models, including ACOs, to further reach their goals, which align with CMS’s.  


Clif Gaus
President and CEO