On June 4, the Centers for Medicare & Medicaid Services (CMS) released its anxiously anticipated 592-page final rule making many changes to the Medicare Shared Savings Program (MSSP). In our white paper, A Dangerous Balancing Act—Proposed Changes to the Medicare Shared Savings Program, published in January, we highlighted our “Top Ten to Watch,” an analysis of CMS’ proposed changes that we believed would have the greatest impact on the future of the MSSP, as well as other payment and delivery system reforms.
With some notable exceptions, CMS now has finalized these changes. While our detailed analysis of the Final Rule is forthcoming, the following summarizes CMS’ action on our previously identified Top Ten.
1. CMS-provided data on assigned beneficiaries
Near the beginning of each performance year, each accountable care organization (ACO) receives from CMS information on each of its prospectively assigned beneficiaries, including name, sex, date of birth, and health insurance claim number. Additionally, CMS provides the ACO with aggregated expenditure and utilization data for its entire beneficiary population.
CMS now has significantly expanded the breadth and depth of the information to be provided. First, these reports will include all beneficiaries for whom any ACO participant has provided a primary care service in the last 12-month period, not just those prospectively assigned to the ACO.
Second, CMS will provide additional data points for each beneficiary, including: (1) additional demographic information (e.g., enrollment status); (2) health status information (e.g., chronic conditions); (3) utilization rates of Medicare services; and (4) expenditure information related to utilization of services.
But wait, there’s more! Currently, if an ACO wants to receive from CMS individually identifiable claims data for its prospectively assigned beneficiaries on a monthly basis, it must (among other things) mail to each such beneficiary a CMS-approved notice regarding his or her opportunity to opt-out of such disclosure of his or her information. Thirty days after sending the notices, the ACO can request from CMS individually identifiable claims data. It’s proven to be an expensive, time-consuming process which often creates confusion for beneficiaries.
CMS now has eliminated this process. ACO participants still will be required to provide written notice at the point of care, but CMS’ disclosure of claims data would not be dependent on any prior notification.
2. Beneficiary assignment, part 1
Beneficiaries are not assigned to an ACO in a traditional sense, as a beneficiary still may receive services from any Medicare provider he or she chooses, regardless of whether that provider is associated with the ACO. Instead, CMS assigns beneficiaries to an ACO for the purposes of setting spending benchmarks and calculating whether the ACO has been successful in reducing total costs of care.
Under the current regulations, CMS uses a two-step process to assign beneficiaries to an ACO for this purpose:
Step one: Assign to an ACO any beneficiary who received any primary care service (as defined in the regulation) from one of the ACO’s primary care physicians (PCPs) during the most recent 12-month period, but only if the total allowed charges for primary care services furnished by the ACO’s PCPs during that time period are greater than the total allowed charges for primary care services furnished by PCPs outside the ACO.
Step two: Attribute to an ACO any beneficiary who did not receive primary care services furnished by any PCP (inside or outside the ACO) during the most recent 12-month period but did receive primary care services furnished by one of the ACO’s specialist physicians during that period, but only if the total allowed charges for primary care services furnished by all ACO physicians and non-physician practitioners during that time period is greater than the allowed charges for primary care services furnished by all physicians and non-physician practitioners outside the ACO.
CMS now will implement three key improvements to the assignment process. First, the definition of primary care services has been expanded to include transitional care management and chronic care management services.
Second, CMS has revised step one to include services furnished by non-physician practitioners (i.e., nurse practitioners, physician assistants, and clinical nurse specialists.) Under the new step one, a beneficiary would be assigned to an ACO if any PCP in the ACO furnished a primary care service to that beneficiary during the most recent 12-month period, but only if the total allowed charges for primary care services furnished by the ACO’s PCPs and the ACO’s non-physician practitioners during that time period are greater than the total allowed charges for primary care services furnished by the same types of providers outside the ACO.
Third, step two of the current beneficiary assignment process now has been revised to limit the types of specialist physicians whose services would be considered for assignment purposes.
These changes in the beneficiary assignment process will become effective at the beginning of 2016, and all benchmarks would be adjusted to reflect the new population of assigned beneficiaries.
3. Extension of Track 1
To date, nearly all ACOs have opted for the one-sided risk model (Track 1) with the opportunity to receive 50% of any savings. Under current regulations, these ACOs would be required to move to Track 2 – with its downside risk – at the end of their first three-year participation agreement – or leave the program.
Rather than running the risk of a mass exodus by ACOs unable or unwilling to accept downside risk at the present time, CMS now will permit Track 1 ACOs to continue under the one-sided shared savings model. This option, however, is limited to those ACOs that (1) satisfied quality performance requirements in at least one of its first two performance years, and (2) did not generate losses in both of its first two performance years.
CMS did decide not to finalize its proposal that second-term Track 1 ACOs be eligible to receive only 40% of any savings; instead, that amount will remain at 50% during the second term.
4. Modification to Track 2
Under Track 2, an ACO is eligible to receive up to 60% of savings. However, a Track 2 ACO bears the risk of having to repay up to 60% of any loss (i.e., actual total cost of care in excess of the ACO’s benchmark). For the handful of ACOs currently participating in Track 2, the minimum loss rate protects them from having to repay a portion of any loss of less than 2%.
Under the new regulations, a Track 2 ACO will have a choice among several options for establishing its maximum savings/loss rate (MSR/MLR): (1) 0% MSR/MLR; (2) symmetrical MSR/MLR in a 0.5% increment between 0.5 – 2.0%; and (3) symmetrical MSR/MLR that varies based on the ACO's number of assigned beneficiaries according to the methodology established under the one-sided model.
5. New Track 3
Rather than making changes to Track 2 to make it more attractive to potential risk-takers, CMS has established a new Track 3. These ACOs would be eligible to receive up to 75% of savings, but also would be at risk for up to 75% of losses. Track 3 ACOs will have the same MSR/MLR options as Track 2 ACOs.
A Track 3 ACO’s performance payment limit will be 20% of its benchmark and its upper loss limit would be 15% of its benchmark. For example, if an ACO’s benchmark was $10,000, it cannot receive more than $2,000 in shared savings and will not be at risk for more than $1,500 of losses. By contrast, a Track 2 ACO’s performance payment limit is 15% of its benchmark, and its upper loss limit is 10%.
6. Beneficiary assignment, part 2
CMS also will use a different beneficiary assignment methodology for Track 3 ACOs, borrowing from the Pioneer ACO Model. As discussed above, CMS currently uses a two-step process to identify those beneficiaries to be assigned to an ACO. CMS provides an ACO with a list of prospectively assigned beneficiaries at the beginning of the year based on the primary care services received during the preceding 12 months.
Each quarter, CMS updates that list based on a rolling 12-month period. A beneficiary prospectively assigned to an ACO may roll off its ranks if he or she receives primary care services from a provider outside the ACO.
Three months after the end of the year (to allow sufficient time for all claims to be filed and paid), CMS makes a final, retrospective assignment of beneficiaries who received the plurality of their primary care services from the ACO during that year. CMS then calculates the total cost of care for these beneficiaries, compares that amount to the benchmark, and determines whether the ACO is eligible for shared savings (or is liable for shared losses).
As a result of this retrospective assignment, an ACO does not know for which beneficiaries it will be accountable during the performance year. CMS reports that ACOs experience an average “churn” rate of 24%. That means nearly a quarter of the names on the first prospective assignment list are different than the names on the end-of-the-year list.
By contrast, Track 3 ACOs will be accountable for the cost of care for those beneficiaries identified at the beginning of the year, with no end-of-the-year adjustments based on where these beneficiaries actually receive primary care services.
7. Risk rewards
Probably the most disappointing news in the final rule was CMS’ decision to finalize only one of its four waivers of Medicare reimbursement rules for ACO participants. CMS will waive the rule that requires an inpatient hospital stay of no less than three consecutive dates for a beneficiary to be eligible for Medicare coverage of inpatient skilled nursing facility (SNF) care. CMS has experimented with this waiver in the Pioneer ACO Model, and now is extending this opportunity to Track 2 and Track 3 MSSP ACOs.
For the present, CMS decided against the proposed payment rules relating to telehealth, the homebound requirement for home health services coverage, and the prohibition against hospitals steering patients to specific, high-quality Medicare providers of post-hospital care services. Instead, CMS intends to further study the impact of these and other payment waivers.
8. Split track ACOs
As CMS noted in the proposed rule, many ACOs have expressed a desire to split their participants into two tracks, allowing a subset of the ACO to move into a risk arrangement. These ACOs emphasize the advantages of providers continuing to work together through the ACO infrastructure, even though some providers remain unwilling to accept risk.
CMS, however, was convinced there are too many unanswered questions regarding the impact of split track ACOs, and thus deferred any action until a later date.
9. Repayment mechanisms
Under the original MSSP regulations, a Track 2 ACO must establish a repayment mechanism equal to at least 1% of its total per capita Medicare Parts A and B expenditures for its assigned beneficiaries, as determined based on expenditures used to establish the ACO’s benchmark at the beginning of a performance period. An ACO may demonstrate its ability to repay losses by obtaining reinsurance, placing funds in escrow, obtaining surety bonds, establishing a line of credit, or establishing another appropriate repayment mechanism that will ensure its ability to repay the Medicare program.
In publishing the final revised regulations, CMS declined the opportunity to require increases to the value of an ACO’s repayment mechanism based on changes to its benchmark. The agency, however, has now limited repayment mechanisms to escrow accounts, lines of credit, and surety bonds, noting other mechanisms have proven impractical.
10. Benchmark adjustments
CMS has finalized its proposed modifications to the benchmark rebasing methodology, to include equally weighting the ACO's historical benchmark years, and accounting for savings generated in the ACO's first agreement period when setting the ACO's benchmark for its second agreement period.
CMS also announced its intention to publish later this summer a proposed rule on a benchmark rebasing approach that accounts for regional FFS costs and trends in addition to the ACO's historical costs and trends. CMS will seek comment on the components of and procedures for calculating a regionally trended rebased benchmark through this soon-to-be-published proposed rule.
This is only the first of the regulatory changes. It will require much more in-depth analysis to ensure long-term success of the MSSP. PYA has worked with dozens of organizations and ACOs and believes that regardless of specific provisions, long term-success will be based on the hard work and commitment of the providers versus the structure of the program. In light of CMS’ provisions to the MSSP, what remains is for providers to remain diligent and hardworking as they move forward into the future.