agilon health Responds to the Centers for Medicare and Medicaid Services Request for Information on Proposed Direct Contracting Models
Newly announced models may provide a unique opportunity for high performing physician practices to expand access to care for traditional Medicare beneficiaries
Director, Center for Medicare & Medicaid Innovation
Centers for Medicare & Medicaid Services
7500 Security Boulevard
Baltimore, MD 21244
Re: Request for Information on Direct Contracting—Geographic Population-Based Payment Model Option
Dear Director Boehler:
On behalf of agilon health, we appreciate the opportunity to respond to the Request for Information (RFI) on Direct Contracting—Geographic Population-Based Payment (PBP) Model Option. agilon health partners with primary care physicians to define a new standard of quality, efficiency and patient experience. We bring the people, solutions, capital and technology necessary to ensure long-term success and to bring back the joy of practicing medicine. We work with physician practices in California, Hawaii, Texas, Ohio and Pennsylvania, serving 1,500 primary care physicians caring for more than 360,000 patients. We will be expanding to four additional states in 2020.
We applaud the Innovation Center on its new direction for developing the global risk payment model in traditional Medicare. Global risk delegated service models are effective in integrating financial and clinical accountability and have demonstrated success in improving population health and individual patient care in the Medicare Advantage program, Medicaid programs and commercial plans. The Direct Contracting models are a step toward achieving the benefits of this model for physicians and patients in traditional Medicare.
Global risk delegated service models like Direct Contracting are among the most successful payment models to move physicians away from a broken fee-for-service system. If structured appropriately, Direct Contracting models restructure and align incentives toward high-quality patient care. These models start with the premise that the primary care physician is central to the patient’s long-term engagement with the health care delivery system and consequently to life-long high quality, satisfying and cost-effective care. The models enable investment in resources that identify patient risk factors, allowing the provider to more effectively organize individualized care. The payment model supports organization and infrastructure that allows physicians to invest time and resources to better streamline workflows, using health information and analytics, chronic condition and episodic care management, care coordination and social support services for patients.
As a result, physicians participating in these models recover their sense of mission, mastery and autonomy, prerequisites to restoring joy in the practice of medicine. Physician burnout has been on the rise and administrative burdens have taken a severe toll on the medical profession, and as a result, on access to high-quality care and systemic stability. We are committed to advancing models that allow physicians to focus on what matters most – providing high-quality care to their patients. Our experience is demonstrating the positive impact on global risk models like Direct Contracting have on primary care physicians’ professional satisfaction. The average Net Promoter Score (NPS) for each of our primary care partnerships or networks has shown year-over-year improvement since the implementation of delegated full risk payment models in partnership with agilon health.
Direct Contracting models in traditional Medicare offer the opportunity to expand on the success of delegated risk models. We welcome the opportunity to provide feedback on the Direct Contracting professional, global and geographic models and to offer specific responses to the questions set forth in the RFI. We recognize the key elements for the Direct Contracting model portfolio are yet to be released and we hope to serve as a resource as you finalize the model design. As you further refine the Direct Contracting model options, we recommend the following guiding principles:
- Fostering predictability, stability and fairness for physicians and patients. New models should be implemented in a manner that creates stability, predictability and fairness for providers and patients. We agree that reforms to the payment and delivery system can be bold. But, the success of these transitions will rely on stable and predictable payments. For example, global payment models in Medicare Advantage have been so successful in part because of the relative stability and predictability of a payment model based on market benchmarks which allow physicians and health plans to invest in the long-term care of beneficiaries. We encourage CMMI to focus on replicating the long-term predictability of revenue based on Medicare Advantage-like market benchmarks in order to encourage early investment and long-term participation by high performing medical groups.
- Creating consistency across Medicare programs and aligning model design to Medicare Advantage. New models should be designed and implemented with an eye toward creating consistency across payers. Today, physicians face conflicting and overlapping requirements across different payer types, including across traditional Medicare and Medicare Advantage. For example, physicians typically report different sets of quality metrics across Medicare fee-for-service alternative payment models (APMs) and Medicare Advantage. Streamlining these and other program requirements will reduce the burden on physician practices – a shared goal of the Administration and physicians across the country. Where conflicting requirements exist, we urge CMMI to use the Medicare Advantage program as the standard. Our experience with physician practices shows that delegated capitation in Medicare Advantage is the gold standard for a model that incentivizes a team-based approach that is focused on patient outcomes and that standardization of goals across Medicare Advantage plans results in significant process improvement, with gains in quality, efficiency and physician and patient satisfaction.
- Protecting beneficiary choice. New models should protect the beneficiary’s choice of provider and should not disrupt existing physician-patient relationships. As new models are designed and implemented, they should continue to empower consumers to make informed choices about their healthcare. New models should focus on providing consumers with complete information about the choices that are available, whether they are enrolled in traditional Medicare or Medicare Advantage. CMS should empower consumers with information about quality and cost sharing in each option.
- Encouraging early adoption of financial risk models. As new models are introduced, we ask that the agency use caution so as not to disadvantage early adopters of financial risk arrangements. The early investments physician practices have made in setting up value-based programs should be taken into consideration as new models are designed and implemented. Specifically, physicians participating in Comprehensive Primary Care Plus (CPC+) or Accountable Care Organizations (ACOs) should be able to immediately transition to the new DCE payment model without loss of surplus sharing from their prior participation in value-based programs. Recognizing their early commitment to this new direction will also foster greater participation in new models as they are released, rather than creating incentives for providers to sit on the sidelines waiting for the next new model to be introduced.
- Incentivizing uptake of higher levels of financial risk. We agree with the agency that incentives should be put in place to encourage uptake of higher levels of financial risk. These incentives can take a variety of forms, including a greater share of financial savings and waivers of burdensome administrative requirements and rules that are intended to protect against fee-for-service incentives for overutilization and self-referral.
- Being mindful of unintended consequences such as provider consolidation and hospital acquisition of physician practices. Evidence shows that hospital acquisition of physician practices leads to increased healthcare costs, including higher out-of-pocket costs for patients. We encourage the agency to adopt rules that do not tip the balance in favor of health system participation, leading to additional acquisition of physician practices, but instead equally allow physicians to maintain their independence while embracing the move to global payment methodologies.
Specific Considerations for Direct Contracting Models
As you continue to develop Direct Contracting professional, global and geographic options, we offer the following considerations that are not specifically raised in the scope of the RFI.
Financial Model Considerations
We understand that key features of the financial model for all tracks of Direct Contracting are still under development. We have several recommendations rooted in our experience with managed care and risk-based arrangements.
For the global Direct Contracting model, the Innovation Center should offer a benchmarking model that will be attractive to providers who are performing below their regional benchmarks, creating a model that is more attractive to high-performing groups. To achieve this goal, the benchmark should use a higher weighted regional component. This approach will attract efficient providers to participate in the model—the very practices that CMS and CMMI should be encouraging to participate in a global model that involves significant financial and clinical accountability. These practices have the core competencies to manage the risk and improve the health of the patient population.
Testing a benchmark with a higher regional blend would create a distinct model design element as compared to the other offerings in the Medicare ACO space. Both the Medicare Shared Savings Program and the Next Generation ACO program have relatively small regional adjustments. Adding a model with a benchmark consisting of a higher regional blend offers an alternative model design option for efficient providers to engage in CMMI programs.
By designing the financial model this way, the Innovation Center can create incentives for participation that change market dynamics for traditional Medicare beneficiaries and physicians. Offering a model that encourages participation by high performers will allow those organizations to begin to streamline their practices and improve efficiencies and to better compete for traditional Medicare patients. This model will give high performing providers an additional tool to attract less efficient physicians in their local market to the delegated capitated model. Letting these high performers drive the model expansion is another approach to spreading risk models in local communities—an approach that has not been explicitly tested by CMMI before. Rather than designing the benchmark that facilitates participation by providers with spending higher than their region, this model should be designed to encourage efficient providers to participate and to grow their practice with the available resources to compete for traditional Medicare beneficiaries. To achieve this, the model should include a more Medicare Advantage-like blend of regional factors. We are conducting extensive analysis on benchmarking options and will look forward to sharing our findings upon completion of that work.
agilon health works with providers who are currently participating in MSSP ENHANCED, CPC+ and other models. We recognize that some of these participants may want to transition to Direct Contracting professional or global models in 2020 or 2021. We respectfully request that the agency quickly identify the options and pathways for these organizations to make this transition.
We also ask that agency keep in mind an important consideration for entities participating in advanced Alternative Payment Models (APMs) in 2020 with regards to their Quality Payment Program (QPP) status. Our understanding is that the alignment year, or performance year (PY) zero, of Direct Contracting would not qualify as an advanced APM for QPP purposes. Therefore, we request that entities that are currently participating in advanced APMs in 2020 be permitted to continue with that participation and to make a smooth transition into new models in 2021, should they desire to do so.
To the extent that the agency is considering additional flexibility or tools for DCEs to use voluntary alignment, organizations participating in advanced APMs in 2020 but preparing to make the move to Direct Contracting in 2021 should be afforded access to these same tools in the 2020 performance year. This could be achieved either by allowing these entities to participate in both their existing advanced APM and PY zero in 2020, or by extending the PY zero tools and flexibilities to participants in other models. This approach will ensure that early adopters of advanced APMs are not penalized and are permitted to operate on a level playing field with new DCEs.
Agency Commitment to Payment Model Stability
The Direct Contracting model portfolio offers a higher level of potential risk and reward for participating entities. Given that these models incorporate some level of capitation, we request that the agency use a more transparent, Medicare Advantage rate notice-like process when announcing the model to afford DCEs an opportunity to review key model design elements and to provide feedback before those elements are finalized. We also ask that the agency commit to maintaining key design elements for a full performance year before making changes. In the current risk models at the Innovation Center, the agency has, at times, made mid-year changes to key program design elements such as risk adjustment. These types of changes can be disruptive in a capitated environment. Providing an opportunity for comment and then adhering to model terms for a defined period of time will increase stability and predictability in the program and will better align the program operations to Medicare Advantage.
RESPONSE TO RFI QUESTIONS
Questions Related to General Model Design
- How might DCEs in the geographic model address beneficiary needs related to social determinants of health?
DCEs across all models of Direct Contracting (professional, global and geographic) should have the opportunity and flexibility to address the social determinants of health for their aligned beneficiary populations. In a recent speech, Department of Health and Human Services Secretary Azar commented that social determinants are the root cause of much of our health care spending. We agree that managing population health and controlling costs includes identifying and assisting patients in managing social determinants of health.
CMS recently finalized additional flexibility for Medicare Advantage plans and providers to address the social determinants of health. We encourage the agency to continue to identify opportunities to align the flexibilities in Medicare Advantage with those in traditional Medicare performance-based risk models. For example, in Medicare Advantage, there is greater flexibility and clarity around the rules for providing transportation, and other services like meals, to patients without potentially implicating the fraud and abuse prohibitions. To achieve the goal of alignment, we would welcome the opportunity to work with the Innovation Center to further determine what waivers would allow risk-bearing provider organizations to better address patients’ total care needs, including the social determinants of health.
Questions Related to Selection of Target Regions
- What criteria should be considered for selecting target regions where the geographic model would be implemented?
As we mentioned above, a key principle for implementing new models should be to promote stability for the delivery system and for patients. In order to achieve this goal, agilon health recommends that the agency select target areas for geographic DCEs where there has not been significant adoption of advanced APMs. Where advanced APMs and care coordination efforts are already underway, a geographic DCE could disrupt care delivery redesign that is already underway and two-sided risk contracts that are midstream. We recommend instead that the agency focus on areas that have not voluntarily adopted and implemented advanced APMs where there may be a more significant opportunity for a geographic DCE to enter a market and adopt new innovations to control costs and improve care coordination where existing models have not been successful.
Questions Related to DCE Eligibility
- What selection criteria and core competencies should CMS consider requiring of applicants?
The geographic Direct Contracting model as described by the agency contemplates a high level of financial risk across a potentially broad geographic area and for populations that could be randomly assigned to the DCE. Given all of these factors, the selection criteria should be sufficiently stringent to ensure that beneficiaries are protected in the new model and that entities in the model have the capability to manage the risk.
Geographic DCEs should be required to meet the following criteria (in addition to the criteria set forth in the RFI):
- Demonstrated experience and success with managing two-sided risk;
- Evidence of the ability to deliver high-quality care in a risk-bearing contract;
- The ability to pay claims downstream and experience in structuring innovative downstream reimbursement models that encourage member engagement and provider coordination; and
- Ability to manage Part D (discussed below).
Questions Related to Beneficiary Alignment
- How should CMS think about attribution for geographic DCEs?
In general, we caution against the use of random assignment in this model, as it may be confusing for beneficiaries and providers in geographic model regions. Instead, attribution for geographic DCEs should prioritize the beneficiary’s choice of provider. Protecting this choice should be the first principle. The attribution methodologies should not advantage geographic DCEs over other types of performance-based risk model participants.
Consistent with our principle around not disadvantaging early adopters of performance-based risk, geographic DCEs should not be attributed beneficiaries that have already been aligned to another performance-based risk model. We recommend that the Innovation Center adopt a more formal approach to addressing potential overlaps in attribution between models. Models with the greatest amount of financial risk should take precedence over lower risk models. We recommend a model precedence “waterfall” that would operate as follows:
- Direct Contracting Global
- Direct Contracting Professional
- Accountable Care Organization
- Primary Care First or CPC+
- Geographic model takes the remainder of unassigned beneficiaries
agilon health also recommends that Direct Contracting Model professional and global participants and ACOs be able to nest primary care medical home, bundles and other specialty models within the total cost of care model, as is currently done in Medicare Advantage arrangements.
- How might DCEs inform beneficiaries of payment model options and engage them in their care? What barriers would DCEs face in engaging with beneficiaries in their target region?
With regard to the Direct Contracting geographic model, CMS will have to carefully consider the types of communications that beneficiaries receive, especially with regards to designing the model. CMS should build on the experience of the Medicare ACO programs and Medicare Advantage when determining how to communicate the new program to beneficiaries, including emphasizing beneficiaries’ options when it comes to freedom of choice of provider and maintaining their physician-patient relationship. Communications to beneficiaries should be clear and transparent and should be tested to ensure that they do not create confusion for seniors.
For the Direct Contracting professional and global models, CMS should allow DCEs waiver flexibility to engage beneficiaries in the model. We understand the agency’s commitment to the open network concept of traditional Medicare and maintaining beneficiary freedom of choice of provider. But we also understand that capitated models work best when paired with an engaged beneficiary. Therefore, we recommend that the Innovation Center work with DCEs to determine what types of tools and incentives would be most valuable to educate beneficiaries about their choices and high-value health care options. This flexibility could build on the flexibility that was offered in the Next Generation ACO program, including the coordinated care reward payment to certain beneficiaries who received primary care services from NextGen ACO providers.
Furthermore, agilon health recommends that CMS incorporate flexibility and any necessary waivers to allow providers in all Direct Contracting models (geographic, professional and global) to talk to patients about the Medicare model that is best for the patient, including an assessment of their Medicare Advantage options. In some cases, the beneficiary’s health and financial status may indicate that a different Medicare coverage option provides the beneficiary better care coordination or lower out-of-pocket spending. In that case, providers should be able to communicate with the patient about which option is best for them. The agency should provide information to providers to help them understand which patients might benefit from Medicare Advantage coverage. We ask that the agency simultaneously continue to evolve the Medicare Marketing Guidelines to allow providers to proactively engage with patients regarding the most appropriate type of Medicare coverage for their needs.
Questions Related to Payment
- CMS envisions applicants proposing a discount to the benchmark for the aligned population. What is the range of discounts CMS can expect applicants to propose and why? How should the agency think about structuring discounts over the life of the model?
For the geographic Direct Contracting model, CMS should expect applicants to propose a five percent discount. This discount is consistent with the Medicare Advantage program. In addition to considerations about discounting related to cost savings, we underscore the importance of maintaining quality performance standards in capitated risk arrangements to ensure that cost savings are not achieved at the expense of patient care. External (e.g., Medicare Advantage Star Ratings) and internal facing (e.g., dashboards and scorecards) quality performance measurement and reporting continues to be a critical feature of the success of Medicare Advantage in general and delegated capitation in particular. Therefore, we recommend that CMS consider a way to incorporate quality performance in the payment model for DCEs. Again, the Medicare Advantage experience could serve as a guide.
- Should geographic DCEs benchmarks include accountability for Part D drug costs?
The geographic Direct Contracting model is an appropriate vehicle for CMS to begin smaller scale testing of incorporating accountability for Part D drug costs. In the recent Pathways to Success rulemaking cycle, CMS solicited and received comments on the desirability of incorporating Part D drug risk into accountable care models. Commenters suggested that CMS develop a voluntary demonstration to test accountable care model accountability for Part D costs. Commenters stated that integrating pharmacy care for fee-for-service beneficiaries could have benefits, such as reducing the risk of adverse events, improving medication adherence and facilitating counseling services. However, some commenters also expressed concerns about barriers to information sharing and variation across plans. Given both the potential opportunities and challenges, a narrower test of this integration seems appropriate, and given that CMS intends to select a limited number of regions for the Direct Contracting geographic model, it seems to be a good candidate for such testing. Furthermore, incorporating accountability for prescription drug spending could serve as another prong in the Administration’s strategy to control prescription drug costs.
- If DCEs enter downstream payment arrangements with providers, how should cost-sharing amounts be determined and collected from beneficiaries?
Geographic DCEs will have to create implementation plans for their downstream providers to collect cost-sharing amounts. The Innovation Center should develop a process for reviewing and approving such plans. In addition, the Innovation Center should ensure that cost-sharing amounts collected from beneficiaries remain consistent across Direct Contracting model options. The geographic model beneficiaries should not have advantages or flexibilities that the Direct Contracting global risk beneficiaries do not also have.
We appreciate the opportunity to comment on this RFI. Please do not hesitate to contact Lisa Dombro, Senior Vice President & Chief of Provider Network Strategies & Engagement, email@example.com with additional questions.
Chief Executive Officer
 Shanafelt, T. et al, Changes in Burnout and Satisfaction with Work-Life Integration in Physicians and the General US Working Population between 2011 and 2017, Mayo Clin. Proc (2019), available at https://www.mayoclinicproceedings.org/article/S0025-6196(18)30938-8/fulltext (accessed May 17, 2019).