Accountable Care Organizations: Shared Savings Calculation – Part II – Expenditure Benchmark

May 6, 2011

A key component for each ACO participating in the Medicare Shared Savings Program (MSSP) will be Centers for Medicare and Medicaid Services’ (CMS) determination of the ACO’s expenditure benchmark. The following article discusses how the benchmark is established and how certain adjustments and updates are made.

A.    Establishing the Benchmark

In order to determine whether or not an ACO saves the Medicare program money, CMS must set a per capita expenditure benchmark (Benchmark) by which to judge any savings or losses the ACO generates for the program. In essence, the Benchmark is a “surrogate measure of what the Medicare FFS Parts A and B expenditures would otherwise have been in the absence of the ACO.” CMS has proposed the following methodology to establish the Benchmark for each ACO.

CMS will begin the process by estimating a baseline for Medicare expenditures using information CMS has for the ACO participants at the start of the ACO’s three-year agreement period. CMS will use the claim records of those ACO participants to determine a list of beneficiaries who receive the plurality of their primary care services from primary care physicians participating in the ACO in the prior three-year period (Beneficiaries). This is the Beneficiary attribution model adopted by CMS in the proposed rule.

CMS will base the Benchmark on actual per capita Medicare Parts A and B fee for service (FFS) expenditures for the Beneficiaries. To minimize variation from catastrophically large claims, CMS will truncate an assigned Beneficiary’s total annual per capita expenditure at the 99th percentile.

CMS will then determine an appropriate growth index for each of the Benchmark years and trend them to Benchmark year three (BY3). This will be done by using health status measures for the Beneficiary population from the data in each of the three years leading up to the establishment of the Benchmark. The health status indices for each year will be adjusted and restated to reflect BY3 risk.

By combining the actual initial per capita expenditures for each year with the respective growth and health status indices, CMS will create risk-adjusted per capita expenditures for Beneficiaries historically assigned to the ACO in each of the three years used to establish the Benchmark and then restate those amounts in BY3 risk and expenditure amounts. CMS will then create a single Benchmark weighing each year of the estimate Benchmark as follows: BY3 at 60%; BY2 at 30%; and BY1 at 10%.

Lastly, for years 2 and 3 of the three-year agreement period, CMS will update the Benchmark by the projected absolute amount of growth in national per capita expenditures for Parts A and B services under the original Medicare FFS program. This update will be the same for all ACOs. All other Benchmark computations will only be rebased at the start of a new agreement (once every three years).

B.    Adjustments – Beneficiary Characteristics

The statute creating the MSSP states that the Benchmark “. . . shall be adjusted for beneficiary characteristics and such other factors as the Secretary determines appropriate . . .” In creating the adjustment factors, CMS stated that “Our goal is to maintain improvements in care delivery of an ACO and to make appropriate adjustments to reflect the health status of assigned patients as well as changes in the ACO’s organizational structure that would affect the case mix of assigned patients rather than apparent changes arising from the manner in which ACO providers/suppliers code diagnosis.”

With this goal in mind, CMS proposes adjusting the Benchmark by using the CMS hierarchical condition category prospective risk adjustment model used under the Medicare Advantage program (MA CMS-HCC). The MA CMS-HCC model covers patient demographic factors (e.g., sex, age, basis for Medicare entitlement and Medicaid status), as well as diagnostic information to create a risk score for each Beneficiary.

Using the MA CMS-HCC methodology, CMS will create a risk adjusted score for the ACO’s historically assigned Beneficiary population and a single Benchmark risk score for each ACO. This risk score will then be applied throughout the three-year agreement period to the annual per capita expenditures for Beneficiaries attributed to the ACO. Changes in the risk score for Beneficiaries attributed to the ACO for the three-year Benchmark period during the performance year will not be incorporated.

C.    Adjustments – Technical

In addition to the adjustments described above, CMS is considering some technical adjustments, but declining others. These include:

  1. IME and DSH Payments. No adjustments will be made to account for the impact on teaching hospitals providing individual medical education (IME) and hospitals serving a disproportionate share of low-income Beneficiaries (a Medicare disproportionate share hospital (DSH)). The proposal is to maintain IME and DSH payments in the per capita costs included in the ACO Benchmark calculation.
  2. Geographic Payments. The proposal will not remove geographic payment adjustments from the Benchmark calculation.
  3. Bonus Payments and Penalties. The proposal is to exclude Medicare expenditures or savings for incentive payments and penalties under value-based policy initiatives such as Physician Quality Reporting Systems, eRx and the EHR incentives under the HITECH Act from the Benchmark calculation.

D.    Comments

CMS is seeking comments on its proposals to establish and adjust the Benchmark especially with respect to how various approaches might affect ACOs or particular types of ACOs to participate in the MSSP.