Fair warning these are the comments on a regulation straight up. We will be doing a blog style version on Monday.
February 6, 2015
Marilyn Tavenner, Administrator
Centers for Medicare & Medicaid Services
Department of Health and Human Services
Hubert H. Humphrey Building
200 Independence Avenue, SW
Washington, D.C. 20201
Re: CMS-1461-P Medicare Program; Proposed Rulemaking for the Medicare Shared Savings Program
Dear Administrator Tavenner,
Aledade (aledade.com) helps independent physicians make the transition to value based purchasing. Value based purchasing is unquestionably the future payment mechanism for health care. The new Medicare is leading the way in the move to value through the Medicare Shared Savings Program. We appreciate the true partnership of CMS with physicians to create a better payment model and ensure safety for all.
Shared savings is a powerful model that aligns incentives for the provider, the beneficiary and society in a way never before achieved on such a scale. However, shared savings may be a bridge to other models that align incentives even more and/or reduce the massive administration effort that fee for service requires. We look forward to participating in the Health Care Payment Learning and Action Network as that future is explored.
As part of that learning, we applaud CMS for outlining proposals to continue to improve and adopt the shared savings program making it more equitable, strengthening the alignment of incentives between providers, beneficiaries and society and potentially providing much needed certainty to the businesses who want to learn how to take risk.
Aledade was created to put primary care doctors at the center of this move to value. It’s good for patients who will find that their trusted primary care doctors are more available and better informed than ever before. It’s good for doctors who want to practice the best medicine possible, the way they have always wanted to. It’s good for businesses and health plans looking for health care partners that deliver the highest possible value and outcomes. It’s good for the country as higher quality; lower cost care will lessen the strain on our budget and our economy.
To this end, Aledade proudly represents small, independent physician practices as part of the Health Care Transformation Task Force (http://www.hcttf.org/) and is a signatory to its comments to CMS. We offer our brief framing for areas covered by the taskforce and comment in depth on several other issues in the NPRM not covered by the Task Force. Thank you for your partnership in this unprecedented effort in health care.
ACOs come in all sizes and forms. Some are set up by organizations with a history of taking on risk, others are created by small physician practices banding together. Risk management is not taught in medical school and physician practices do not have actuaries on staff. Acknowledging these differences with different policies when seeking to encourage providers to two-sided risk will ensure a needed diversity of ACOs.
Recent results in the ACO program highlight the need to differentiate particularly when it comes to move to two-sided risk. Many organizations have merely hedged their bets with low-regret moves like buying up practices and forming organizations that are Accountable Care Organizations (ACOs) in name only. These actions consolidate a health system’s referral base, but administrators do not employ strategies to reduce costs, which are their revenues. Put differently, these “ACO squatters” say there are embracing new payment models, but remain stuck in the mentality of the do-more, get-paid-more system.
Unfortunately, this strategy is already too widespread, and likely to grow as long as large organizations are allowed to continue in “one-sided” (upside only) shared savings models. Defensive moves by hospital systems provide a veneer of action, while consolidating regulator-blessed market dominance that can increase costs without improving quality at all.
We suggest three policies to encourage ACOs with large institutional partners (hospitals or health plans) to move to two-sided risk. First, require that ACOs with large institutional partners have achieved shared savings in at least one of the three years of the first contract. Second, require that ACOs with large institutional partners explain in their application for their second contract why it is not financially feasible for them to move to two-sided risk. Second, evaluate ACOs for market consolidation and require ACOs that dominate their local health care market to move to two-sided risk. These policies will prevent large institutions and dominate market players from “squatting” in one-sided risk.
With nearly all ACOs choosing one-sided risk and CMS projections of 90% continuing to do so, small tweaks like floating the MSR/MLR rate in track 2 (something that would have only affected 13 2012/2013 ACOs) or reducing the savings rate for ACOs in their second contract year simply will not change ACO behavior. Only bold actions like the increase in shared savings rate in Track 3 and rewarding quality with higher savings will truly shift the business equation all ACOs must compute when considering two-sided risk. We encourage CMS to finalize these policies with the tweaks presented in the Task Force’s comments.
We recommend that CMS finalize the new track 3 with its enhanced shared savings/losses.
We recommend that CMS account for ACO characteristics of institutional partners and market dominance when considered applications for a second one-sided risk contract.
An accurate benchmark is the single most important number for an ACO. Working against a benchmark that does not reflect the true expected costs of the population makes savings either too hard to achieve or too easy to achieve. Savings should be achieved through successful ACO interventions to improve care, improve health and remove waste. The accuracy of the benchmark determines whether that is true.
Risk adjustment is currently used to predict whether last year’s actual costs or regional costs are a good indicator for the expected costs for a specific ACO population. CMS should itself and with partners make the accuracy of risk adjustment methods a subject of continual study and improvement. One programmatic improvement to MSSP is to allow risk adjustment to adjust the benchmark up or down for all beneficiaries.
We acknowledge that the current model relies on coding that can be subject to manipulation. As with Medicare Advantage (MA), we believe the best solution is to manage the manipulation (through an adjustment or possibly capping upward risk adjustment at the national growth rate) rather than taking upward risk adjustment out of the benchmark and losing the corresponding accuracy.
CMS should also consider when things outside of an ACO’s control might reduce the accuracy of the benchmark. For example, if there is massive consolidation in a market or within a given specialty in the market that could significantly reduce the accuracy of the benchmark from the historical benchmark. CMS should monitor for these effects and explore possible policy solutions.
We believe that a move to regional benchmarks makes the program more sustainable and appropriately shifts the incentives away from being better than last year to being better than a peer group over time. We recommend the following transition schedule for all ACOs:
Transitioning from ACO historical costs to regional benchmark: Two paths
- Below regional benchmark at the end of the first contract
- 2nd contract: 50% (Historical Benchmark) / 50% (Regional Benchmark)
- 3rd contract: 20% (Historical Benchmark) / 80% (Regional Benchmark)
- 4th contract: 100% (Region)
- Above the regional benchmark at the end of the first contract
- 2nd contract: 80% (Historical Benchmark) / 20% (Regional Benchmark)
- 3rd contract: 50% (Historical Benchmark) / 50% (Regional Benchmark)
- 4th contract: 20% (Historical Benchmark) / 80% (Regional Benchmark)
- 5th contract: 100% (Regional Benchmark)
We recommend that CMS allow for risk scores to go up as well as down for continuously enrolled beneficiaries. To protect against coding related fluctuations, CMS should cap how much a risk score can go up for an ACO at the national rate of growth in risk scores.
We recommend that CMS finalize the transition to a regional benchmark over the course of several contracts (option 5) for all ACOs.
We encourage CMS to look at each waiver for its appropriateness for all ACOs individually. While some waivers may only be appropriate for two-sided risk, others (such as tele-medicine) may be more applicable to all.
We recommend that CMS finalize all proposed waivers for all ACOs in two-sided risk.
We recommend that CMS evaluate each waiver individually for appropriateness for some or all ACOs in one-sided risk.
We recommend that CMS explore the possibility of allowing waivers in benefit design (i.e. waiving the co-pay for Chronic Care Management) and beneficiary assistance (air-conditioning, gym/wellness, transportation, home accessibility) for two-sided ACOs.
Currently, CMS only acknowledges primary care physicians when beneficiaries vote with their feet by walking through the practice’s door. However, there are many times where for a particular year that may not be an accurate reflection of the beneficiary’s wishes and normal care pattern. Simple and common examples, such as dealing with an acute illness or condition requiring specialized evaluation and management services, extended time away from primary residence, low health care utilizers where a single service plays a big role in determining the plurality of primary care services, or primary care physician (PCP) switching for a patient when they enter a skilled nursing facility (SNF), etc. all could lead to inaccurate attribution. Beneficiaries should be able to also vote with their voice and declare that despite the data from a single, peculiar year, “this physician, this nurse practitioner, this physician assistant is whom I have a special relationship with, this is who I want to coordinate my care.” Thus we promote patient engagement and make known an active patient and physician relationship.
We present specific suggestions on how to accomplish this in the Task Force Comments.
ACOs are divided over whether prospective attribution or retrospective attribution is better. Some ACOs crave the certainty of the prospective attribution. Others ACOs balk at the idea that a beneficiary who didn’t receive care from them would be in the ACO or that a beneficiary who did receive care from them would not be. Both are concerned with uncertainty. Few things are more valuable than certainty. In a prospective model, the ACO is certain that the resources they expend will be on beneficiaries for whom they are accountable. In a retrospective model, CMS is certain that the beneficiaries on which they pay out savings received services from the ACO and it is on the basis of those services that the savings were generated and not due to beneficiary selection by the ACO. Similarly, the ACO is certain that the beneficiaries on whom savings are based received services from the ACO.
However, this is not an either/or situation. The closer the prospective attribution matches the retrospective attribution, the more certainty everyone has.
Therefore, we suggest that CMS not only institute a prospective option, but also seek to improve prospective attribution. Currently, CMS reports a 24 percent difference between the two. However, that decreases to 17 percent after accounting for eligibility. It is in that 17 percent where the uncertainty lies.
We suggest that the path forward is to refine prospective and retrospective attribution so that the difference between the two disappears. For example, CMS could exclude from the prospective attribution list beneficiaries who move away during the year and those who begin a long-term care arrangement that causes a change in their primary care physician.
We urge CMS to implement a mechanism that will allow beneficiaries to choose their primary care physician and thus their ACO participation.
We recommend that CMS aggressively evaluate refinements to attribution that bring prospective and retrospective attribution closer together.
We recommend that CMS establishment prospective attribution for ACOs updated annually even as it works to bring the two methods together.
In summary more of it, delivered more timely and more transparently. Data powers the interventions that make a difference in care, health and costs in a timely fashion.
We urge CMS to make open source its actual code for all formulations in the ACO program.
CMS should work with the Office of the National Coordinator for Health Information Technology (ONC) and industry to reduce the cost and burden for providers to access, aggregate, and exchange clinical information from their certified electronic health record. In particular, the certification criteria 45 CFR 170.314(b)(7) – Data Portability seems to hold a lot of promise for population health, but in reality has not proven to be a reliable gateway to the information contained in the certified electronic health record.
The other very promising technology is reliable admission, discharge and transfer feeds. Supporting technology for ADT feeds is established and the purpose they serve should not be underestimated. By serving as a trigger for transition of care management, a simple alert from an ADT triggers a cascade of action that can prevent readmissions, coordinate care and prevent future inappropriate emergency department use. This cascade requires a trigger and that trigger is ADT feeds. CMS should work with ONC to encourage ADT feeds in every state. CMS should also explore other opportunities such as the State Innovation Models to support this simple, yet very powerful tool.
Attachment 1: Table of Comments on CMS Proposals
||Proposed Change in the Regulations
||Corrections to ACO participant, professional, provider/supplier, assignment, and hospital definitions. Other definitions have their changes in their own sections.
||Agreement Requirements: List of required elements, emphasis on the one year term
||5000 Bene Min: Allowing for the corrective action plan to be in a year specified by CMS rather than it having to be the next performance year. Not necessarily requiring a CAP if the ACO has already submitted participant additions
||We applaud CMS’s efforts to work with smaller ACOs to maintain their participation in the MSSP program.
||This puts into regulation text the previous sub-regulatory guidance that CMS will not remove a terminated ACO participant from the calculations and reporting requirements. This provides an opportunity to make the case that such a participant should be excluded especially if they terminate early in the year.
||We understand the desire of CMS to not allow ACOs to use participate removal as a financial tool. However, we see no reason why a participant that has left the ACO should be included in quality reporting. The ACO has lost its influence and access to the practice and its information. Including them in quality reporting distorts the true quality performance of the ACO.
||Considering delaying the addition of ACO providers/suppliers to conduct program integrity screening which only happens twice a year.
||Physicians change positions just like other professionals. Delaying the addition of ACO providers/suppliers would hamper an ACO in terms of data and practice transformation.
||Defining significant change as change in ACO ownership, turnover of 50 percent or more of ACO participants and any change that prevents the ACO from meeting the eligibility requirements of MSSP.
||Seeking comment on whether to require 45 to 60 days of advance notice.
||Impossible to know whether a given situation will be known in advance. For example, should an ACO provide advance notice if they are getting close to the 50 percent turnover. How close is needed to trigger notification. Ownership changes might be subject to non-disclosure requirements. Request, but don’t require advance notice. 30 days after the fact can be a requirement
||Merged and Acquired TINs: While the proposed solution works for one instance (all providers in a TIN are acquired), it does not address situations where some a TIN splits.
||Physicians change jobs/opportunities just like anyone else. Having no mechanism for physicians to keep their patient history if they don’t also keep their group tax accounting history is an inaccurate way to group care. CMS should implement a process where for attribution and historical benchmarking can account for NPI/TIN combinations so that all situations can be addressed as opposed to the singular situation CMS covers in the proposed rule.
||Legal Structure and Governance: Board of the ACO must be the same as the board of the legal entity that is the ACO. The board of a multi-participant ACO must not be identical to the board of one of the participants
||Remove flexibility for ACOs to deviate from the requirement that 75% of the ACO’s governing body must be held by ACO participants
||Beneficiary representative can not be an ACO provider/supplier
||Leadership and Management Structure: Revising the regulation to not require the medical director to be an ACO supplier/provider. Three reqs are Board-certified, licensed in a state in which the ACO operates, physically present on a regular basis of the location of the ACO or ACO provider.
||Concur. This provides the flexibility for ACOs to select the best medical director possible, as a medical director’s skill set is not completely analogous to that of a full time practicing physician.
||Required Process to Coordinate Care: Additional application requirement to describe how it will encourage and promote the use of enabling technologies for improving care coordination for beneficiaries.
||Concur. In addition we urge CMS to continue to work with ONC to ensure that the data available in health it (particularly certified EHR technologies) is available to ACO participants and providers/suppliers on a population level. ACOs are uniquely positioned to combine information from health it and the information from CMS to create a complete picture of health. The biggest barrier to this is not analytics or tools to use the analytical results, but access to data itself whether it is locked at CMS, another payer, an EHR or a practice management system. Everything CMS and ONC can do to free that data will contribute to ACO success.
||App requirement to describe partnerships with long-term and post acute care providers to improve care coordination, milestone based timeline for all capabilities and use of tele health.
||Coordination with all health care providers in critical for the long-term success of any ACO. We do agree that every ACO should have a plan to form these partnerships. CMS should allow submission of these plans as-is. It is an unnecessary administrative burden to re-write the ACO’s plan into CMS application format for this one particular health care segment. CMS should consider that re-formatting especially of a graphical timeline is an exercise measured in hours and not minutes. Multiplied by all applicants this could add up to a thousand hours or more.
||Condensed application for a Pioneer switching to MSSP
||Former Pioneer ACOs would not be permitted to enter MSSP under Track 1.
||Concur. The eligibility requirements for the Pioneer program are indicative of the type of institutional capabilities that support two-sided risk.
||Renewal of Participation Agreements: Whether the ACO satisfies the criteria for operating under the selected risk model.
The ACO’s history of compliance with the requirements of the Shared Savings Program.
Whether the ACO has established that it is in compliance with the eligibility and other requirements of the Shared Savings Program, including the ability to repay losses, if applicable.
Whether the ACO met the quality performance standards during at least 1 of the first 2 years of the previous agreement period.
Whether an ACO under a two-sided model has repaid losses owed to the program that it generated during the first 2 years of the previous agreement period.
The results of a program integrity screening of the ACO, its ACO participants, and its ACO providers/suppliers (conducted in accordance with § 425.304(b)).
|As discussed in our narrative, we think special consideration should be given to institutional partnerships and market consolidation in renewals of one-sided risk.We agree with CMS that compliance is a given and quality performance is a good standard for renewal.
||Changes to regulation changes that affect beneficiary assignment (they now would) in the current contract year
||Expand beneficiary identification to any beneficiary that has a primary care service in the most recent 12-month period
||We agree that this would make ACOs more aware of patients who may see an ACO participant as their primary care provider, but through oddities of that year not be assigned to the ACO. We recommend that CMS expand the look back period to 2 years for beneficiaries without an address change to further ensure that beneficiaries do not miss out on the opportunities available from ACO providers
||Replacing the letter process with 1-800-Medicare
||Concur. This is major administrative burden on the ACO over which the ACO has no control. With CMS providing the current letter verbatim, it only makes sense that CMS communicate directly to the beneficiaries.
||Continuation of the requirement for the ACO participants to notify the beneficiaries in writing at the point of care utilizing the poster notification instead of the letter.
||We agree with CMS’s proposed focus on the poster as opposed to giving every patient a written form. ACO providers have indicated a very strong preference to not provide the form to every beneficiary at the first office visit of the year. A practice some ACOs were doing out of an abundance of compliance caution. The language in the proposed rule is quite clear. We recommend that it be included in the finalization of this proposal.
||Opt-out data consequences
||When a beneficiary opts out of data sharing, the beneficiary should no longer be included in ACO assignment. We simply do not foresee a situation where a beneficiary would say I want a physician to coordinate my care, but I don’t want that same physician to have access to the data needed to do that.
||Suppression of information related to the diagnosis and treatment of alcohol or substance abuse.
||We refer CMS to the Task Force comments.
||Definition of Primary Care Services: Inclusion of TCM and CCM codes in assignment methodology
||We agree with the inclusion of these codes and with moving this determination to the annual PFS rulemaking process. We recommend to CMS that they monitor CCM as this code can be used up to 12 times a year. If assignment changes for many providers based on CCM (i.e. absent CCM inclusion the beneficiary would have been assigned to another provider), that raises questions for both ACO assignment and for the use of CCM that should be explored by CMS.
||Physician Specialties and Non-Physician Practitioners in Assignment: Inclusion of NP, PA and CNS in step one of assignment.
||We applaud CMS’s proposal as an innovative way to both honor the statute while supporting the critical role these providers play in primary care. We suggest that CMS finalize as proposed. We would encourage CMS to work with NP, PA and CNS community on ways to begin to recognize that these providers also specialize. For example, if they have a supervising physician they could inherit the specialty of that physician. Alternatively, CMS could begin to collect specialization information from NP, PA and CNS themselves.
||Physician Specialties and Non-Physician Practitioners in Assignment: Limitation of some physician specialties from being included in the assignment process.
||We agree that some specialties do not provide primary care, but due to an acute event in a given year (such as a surgery) may achieve plurality of services. We would like to see CMS give further thought to the following specialties to hospice. We suggest CMS look at the distribution of this specialty. For example, it could be a positive to have beneficiaries move out of ACOs when they start hospice, as most savings opportunities from keeping the patient healthier are no longer applicable. For the same reason it would be strange for an ACO to pick up beneficiaries as they enter hospice. The inclusion or exclusion of this specialty should not serve as a proxy for inclusion/exclusion of hospice services from the ACO model. That should be done directly.
||Physician Specialties and Non-Physician Practitioners in Assignment: Combining step 1 and step 2
||We strongly oppose the combination of step 1 and step 2. We believe this would create a decided shift away from primary care in the ACO model. Assignment would be based on the plurality of evaluation and management services not the plurality of primary care services if the steps were to be combined. Table 2 is still a wide variety of specialists. An allergist is not well positioned to coordinate care for beneficiary recently diagnosed with diabetes for example.
||Assignment of Beneficiaries to ACOs That Include FQHCs, RHCs, CAHs, or ETA Hospitals: Any physician for initial population and any provider for attribution.
||Effective Date for Finalization of Proposals Affecting Beneficiary Assignment: Changes not applicable until the next performance year
||While we believe that many of the changes to assignment are needed enhancements to the assignment process, we agree that it would be tumultuous to apply these changes in the middle of a performance year.
||Modifications to the Existing Payment Tracks
||Our comments on changes to payment tracks are covered in detail in our narrative and in the Task Force comments. Top tier ACOs in quality should be rewarding with better sharing rates than currently available in any track. Performance Year 2 and 3 should not see such a dramatic fall (analysis indicates to an average of 37%) in the shared savings rate.
||Creating Options for ACOs That Participate in Risk-Based Arrangements: New Track 3
||Our comments on and endorsement of the proposed Track 3 are covered in detail in our narrative and in the Task Force comments.
||Seeking Comment on Ways To Encourage ACO Participation in Performance-Based Risk Arrangements: General
||As discussed in our narrative, the best ways to encourage ACO participation in risk are:
– Increase the shared savings rate
– Reward top tier quality with better rates
– Finalize waivers for two-sided risk
– Acknowledge differences in ACOs ability to take on risk
||Seeking Comment on Ways To Encourage ACO Participation in Performance-Based Risk Arrangements: Waivers
||Our comments on and endorsement of the proposed waivers are covered in our narrative and in the Task Force comments.
||Seeking Comment on Ways To Encourage ACO Participation in Performance-Based Risk Arrangements: Beneficiary Attestation
||We do not believe there is any reason to limit beneficiary attestation to ACOs in two-sided risk. We urge CMS to finalize beneficiary attestation. More comments on the details of such a mechanism can be found in our narrative and the Task Force comments.
||Seeking Comment on Ways To Encourage ACO Participation in Performance-Based Risk Arrangements: Beneficiary Attestation: Step-Wise Progression
||As stated at the beginning of our review of this section, we believe strongly that it is the lack of a compelling case for ACOs to take on two-sided risk that is holding them back not the lack of an “on-ramp” or other such bridge mechanism between one-sided and two-sided risk.
||Modifications to Repayment Mechanism Requirements: Streamlining the Repayment Mechanism
||While we understand CMS desire to protect the Trust Fund, the setting aside of 1 percent could essentially double the cost of setting up an ACO. Many ACOs struggle to develop the capital to fund the operations of the ACOs. To have to come up with twice as much capital up front is a major deterrent to two-sided risk. By CMS tying up twice as much capital it essentially halves the possible rate of return on that capital. CMS should expand the principles of the ACO Investment Model while not funding the ACO directly for operations, but subsiding or otherwise assisting targeted ACOs with the repayment mechanism.
||Seeking Comment on Methodology for Establishing, Updating, and Resetting the Benchmark: Resetting
||Our comments on and endorsement of the proposed option 5 (and downstream effects if option 5 is not adopted) are covered in detail in our narrative and in the Task Force comments.
||Seeking Comment on Methodology for Establishing, Updating, and Resetting the Benchmark: Updating
||Our comments on and endorsement over using a regional update factor are covered in detail in the Task Force comments.
||Public Reporting and Transparency: Format
||We agree that 30 days is a reasonable update timeframe. The CMS policy of requiring ACOs to submit screen shots to CMS has generated near uniformity in design for public reporting. Use of a template would just formalize what has occurred due to the submission requirements.
||Public Reporting and Transparency: Additional Information
||We do not have any concerns conceptually with the greater disclosure of data. In particular, we wholeheartedly agree that disclosing the makeup of ACO participants would add a lot of value to beneficiaries and the public at large. In combining this additional information with CMS’s desire to create a standard format puts a lot of pressure on the design of the template. The utility of the information particularly to beneficiaries is going to hinge not just on disclosure, but design of the template so we urge CMS to be very selective in how they get to design the template. Possibly even opening it up for outside submissions.
||Public Reporting and Transparency: Reporting on CMS Websites
||We have no objection to posting detailed information about ACOs on CMS’s website. We applaud the recent data releases on the characteristics of 2012/2013 starts and eagerly await such information on 2014 and 2015 starts.
||Terminating Program Participation: Reconsideration Review Process
||We do not disagree with the proposed changes in this area. However, we urge CMS to begin the timeliness clock once informal work has ended. We believe that alternative is clearer to all parties and encourages the great partnership between CMS and ACOs.