Download as PDF

Summary of Next Generation Model Program Methodology Changes for 2019 and 2020

Background: The Next Generation Model will undergo a number of changes to calculating expenditures and benchmarks for performance years 2019 and 2020, which are the final two years of the program. Operated by the Center for Medicare and Medicaid Innovation (Innovation Center) and described here, this model has seen other modifications since its inception in 2016. The Innovation Center is following through on its plans to release updated methodologies for 2019 and 2020 and announced these changes in March of 2018. The final two program years represent an extension of the model and are available to ACOs that began the program in 2016, 2017 or 2018. Below is a summary of the changes, and the full changes are detailed in this Innovation Center document. NAACOS encourages ACOs to share feedback on these changes by emailing us at [email protected]

Overview: The revised methodologies and calculations described below include expenditure data from two benchmark years instead of one year, which is currently used. ACOs are increasingly compared to providers in their region and the Innovation Center takes a two-pronged approach which treats ACOs differently depending on whether their expenditures are higher or lower than those in their region, thus adding or subtracting amounts to/from the baseline. 

Base Period: The Innovation Center will utilize expenditure data from two calendar years (CY) to set the financial benchmark. They will use a blend of two years with the simple average of the standardized, trended baseline expenditures for these two years. Baseline year one is three years prior to the performance year and baseline year two is two years prior to the performance year. For example, 2016 and 2017 are the baseline years for performance year (PY) 2019. 

Expenditure data: The average expenditure PBPM for each base year will be risk-standardized, standardized for the impact of the geographic adjustment factors that were used in the base year(s) to calculate fee-for-service (FFS) provider payments, and trended to the performance year. The benchmark will then be risk-adjusted and GAF-adjusted to reflect the geographic payment adjustment factors used to calculate fee-for-service payments in the performance year. 

Discount: The discount is applied to the benchmark depending on the shared savings rate selected by the ACO. The performance-adjusted benchmark will be discounted by 0.5 percent if an ACO chooses a shared savings rate of 80 percent and will be discounted by 1.25 percent if an ACO chooses a shared savings rate of 100 percent.  Under the new methodology, the discount will be based on the savings rate and not on quality, and the quality bonus will be handled as a withhold independent of the discount. 

Risk Adjustment: The PY4 Benchmark will be risk adjusted to reflect the difference between the baseline risk score and the average risk score of the performance year aligned beneficiaries. A prospective coding adjustment will be applied to performance year risk scores. The benchmark risk score for the performance year will be not less than 100 percent and not more than 103 percent of the baseline risk score. The PY4 method will continue to include an explicit coding adjustment. 

Regional Trend and Adjustment: The PY4 Benchmark will continue to use a prospective regional trend based on the adjusted United States Per Capita Cost. The benchmark will include an adjustment that reflects the attained performance of the ACO during a two-year baseline period. Conceptually, the attained performance adjustment will be similar to blending the ACO’s baseline expenditure with the average baseline expenditure in the region served by the ACO. An ACO with a baseline expenditure that is less than the average expenditure incurred by beneficiaries in its region will have an amount added to its baseline, while an ACO with a baseline expenditure that is higher than the average expenditure in its region will have an amount deducted from its baseline. The maximum upward adjustment to the baseline will be 10 percent, and the maximum reduction to the baseline will be 2 percent of the baseline. 

Varying the baseline adjustment based on the ratio of the regional standardized baseline to the national standardized baseline is intended to increase the attained performance adjustment for a “low cost” ACO in a historically “low cost” region and to reduce the attained performance adjustment for a “low cost” ACO in a historically “high cost” region. Conversely, it is intended to require slightly greater improvements in performance for a “high cost” ACO in a historically “high cost” region and to require slightly smaller improvements in performance for a “high cost” ACO in a historically “low cost” region.The adjustment also factors in whether the ACO’s baseline is above or below the Next Generation Model standardized national baseline, which is the average standardized operating costs for all NGACO alignment-eligible beneficiaries. The Innovation Center explains that by using this approach an ACO that has achieved substantial savings will continue to retain a portion of those savings and will have an incentive for continuous improvements in operational efficiency, while also creating an incentive for ACOs whose aligned beneficiaries have historically low costs to continue in the program. 

Stop loss: There are a number of changes to the stop-loss arrangement. The new policy uses four attachment points with the percentage of expenditures accruing to an ACO’s baseline and performance year decreasing at each attachment point. Attachment points will be calculated on a per-capita basis rather than per-beneficiary-per-month basis. The attachment points will be set prospectively and will be geographically adjusted. The stop-loss charge is based on a revised rolling 2-year baseline period in a manner similar to other benchmarking calculations. Aged & Disabled and ESRD experience is combined, and the attachment point is increased by each month that a beneficiary has ESRD status. NGACOs still have the option to forego CMS stop-loss and seek external reinsurance coverage in 2019 and 2020. 

Partial year beneficiary eligibility: Alignment-eligibility and service-area exclusion procedures will be modified to allow an aligned beneficiary to accrue a partial-year of experience. In these instances, a beneficiary will remain aligned and will accrue experience through the last month of continuous eligibility. For example, suppose a beneficiary was in Medicare FFS and aligned to the NGACO from January through the end of July and then she enrolls in a Medicare Advantage plan August 1. That beneficiary would accrue to the NGACO for January through July and that experience would be included in the financial settlement. The baseline expenditure will similarly reflect the experience of beneficiaries with a partial year of eligibility.Partial year eligibility will also apply to beneficiaries excluded by the service area requirements. Once a beneficiary becomes ineligible, he or she will be removed prospectively from the aligned population. The Innovation Center notes that this change will increase the predictive value of the quarterly benchmark reports by reducing the impact of exclusions, particularly the year-end exclusions, on financial performance. This change means a beneficiary may have a partial year of experience based on the death of the beneficiary, the occurrence of an event that terminates the beneficiary’s eligibility and exclusion based on the service area requirements. 

Benefit enhancements: NGACOs will be offered three new benefit enhancements starting in 2019 and further detailed in this CMS memo. 

Cost Sharing Support for Part B Services
CMS will allow NGACOs to cover the cost sharing payments of patients assigned to participants and preferred providers starting in 2019. The objective is to promote high-value services and remove financial barriers for certain patients to obtain needed care, comply with treatment plans and improve their health. To participate, NGACOs must enter into a “Cost Sharing Support Agreement” with participants and preferred providers and identify subsets of patients and services for which cost sharing payments will be covered. Agreements can cover part or all of cost-sharing payments. More information can be found in this document

Care Management Home Visits
NGACOs can be paid for some limited home visits to patients. The visits are meant to stave off a potential hospitalization and build on existing waivers for post-discharge home visits. The Care Management Home Visits Benefit Enhancement will allow up to two home visits for patients with a care plan within 90 days of seeing a NGACO provider. CMS provides more details in this resource. CMS clarifies that these are not, however, home health visits, but rather a supplement to in-office primary care visits. 

Chronic Disease Management Reward Program
NGACOs can provide gift cards to assigned patients up to $75 to incentivize use of a chronic disease management program. The ACO will pay for the gift card out of their own funds. As with other benefit enhancements, NGACOs who offer this service must submit a plan detailing how they intend to structure their chronic disease management reward program. 

Quality: CMS will withhold a percentage of a NGACO’s Performance Year Benchmark that can be earned back by hitting quality scores. The quality withhold will replace the standard discounts in the first three years of the NGACO program as the mechanism to recognize providers’ quality. In 2019, 2 percent of the ACO’s benchmark will be held back with all of it earned back with a full quality score. In 2020, 3 percent will be withheld and adjusted back on quality performance. If a NGACO receives a quality score of 95 percent, it will receive 95 percent of the withheld amount back. CMS explains the change in this document. 

TABLE A: Comparison of the Current Next Generation Model Methodology and the Revised Methodology

Change

Current Methodology

(2016 – 2018)

Revised Methodology

(2019 – 2020)

Baseline Period

Baseline period is one year (CY 2014) for PY1, PY2 and PY3.

Baseline period is a rolling 2-years:

  • The first baseline year is the CY three years prior to PY
  • The second baseline year is the CY two years prior to PY
  • There is a 50%/50% equal weighting of trended baseline years one and two expenditures

Regional expenditures

The benchmark was based on historical ACO expenses only in CY 2014. (Note: a regional efficiency adjustment was applied to the discount rate.)

Regional adjustment is applied to benchmark expenditures according to the regional blend percentage.  The adjustment is:

  • If historical benchmark > regional benchmark then up to 2% reduction will be applied to benchmark using a regional blend percentage between 10% - 15%
  • If historical benchmark < regional benchmark then up to 10% upward adjustment applied to benchmark using a regional blend percentage between 30% and 40%

Discount rate

The discount rate is determined for each ACO and is adjusted for efficiency and quality. The standard discount is 2.25%, then is adjusted based on regional and national efficiency and quality. The minimum potential discount is 0% and the maximum is 3.75%.

 

 

The discount rate depends on the shared savings rate selected by the ACO

  • 0.5% discount rate when an 80% savings rate is chosen
  • 1.25% discount rate when a 100% savings rate is chosen

The discount rate will not be adjusted by the quality score.

Risk Adjustment

Baseline expenditures are risk adjusted to performance year risk by using the CMS-HCC model.  For PY1, the risk score is re-normalized relative to the entire NGACO reference population so that the average score is 1.0.  The change in PY risk score is no less than 97% of the 2014 risk score and no more than 103% of the 2014 risk score.  In PY3, and optionally for PY2, risk scores are no longer re-normalized.  The coding adjustment method is used to remove the coding-based increase in risk score from 2014 to PY3. The coding adjustment method caps the risk score so that it is no less than the 2014 risk score and no more than 103% of the 2014 risk score.

The coding adjustment approach continues from PY3.  The coding adjustment is no longer calculated relative to the 2014 baseline year but is relative to the 2-year rolling baseline period.  The coding adjusted risk score can be no less than the rolling baseline period risk score and no more than 1.03% of the rolling baseline period risk score.   

Stop-loss

Optional stop-loss is elected.  The difference between the stop-loss charge and PY excess claims is used to adjust savings/losses. The stop-loss charge is baseline claims in excess of the 2014 attachment point adjusted by membership, risk score, regional trend and discount rate.  If PY expenditures in excess of the attachment point including IBNR exceed the stop-loss charge then the savings are increased by the difference.  Otherwise, savings are decreased by the difference.

There are a number of changes including:

  • Four attachment points with the percentage of expenditures accruing to an ACO’s baseline and performance year decreasing at each attachment point
  • Attachment points will be calculated on a per-capita basis rather than a per-beneficiary-per-month basis
  • Elimination of the adjustment for beneficiary partial year enrollment
  • A prospectively determined attachment point based on 2017 experience and trended to 2019 for PY4 and based on 2018 experience trended to 2020 for PY5.
  • The attachment point is geographically adjusted at the end of the performance year
  • Aged & Disabled and ESRD experience is combined, and the attachment point is increased by each month that a beneficiary has ESRD status
  • The stop-loss charge is based on a revised rolling 2-year baseline period in a manner similar to other benchmarking calculations

 

Partial-year enrollment

Beneficiaries are fully excluded from baseline and performance year expenditures if they become ineligible for alignment to the ACO during the performance year. This can occur as a result of the following:

  • Loss of Part A or Part B coverage
  • Enrollment in Medicare Advantage
  • Medicare not being the primary payer for the beneficiary
  • Residence outside the U.S.

Note, aside from ineligibility there are other reasons for exclusion from financial settlement such as moving outside the ACO’s service area1

Beneficiary baseline and performance year expenditures will be adjusted for partial year enrollment. This can occur based on the death of the beneficiary, the occurrence of an event that terminates the beneficiary’s eligibility and exclusion based on the service area requirements.

 

Quality Measure Set

Mirrors MSSP quality measure set

No change

Quality withhold

A quality withhold was not used in initial program years when CMMI used a discount which was affected by quality performance and efficiency.

In PY 4, 2% of the Performance Year Benchmark will be withheld and adjusted based upon the NGACO’s performance. As such, NGACO’s can earn up to 2% of the quality withhold through a full quality score. In PY 5, 3% of the Performance Year Benchmark will be withheld and adjusted based upon the ACO’s performance on the quality measure set.

Benefit Enhancements

3-Day SNF Rule Waiver

Telehealth

Post-Discharge Home Visits

New starting in 2019:

  • Cost Sharing Support for Part B Services
  • Care Management Home Visits
  • Chronic Disease Management Reward Program

 

 

[1] Per existing program requirements, prior to financial settlement a beneficiary will be excluded if the Next Generation Beneficiary was a resident of a county that was part of the ACO’s service area in the last month of the 2-year alignment period but was a resident of a county that was not part of the ACO’s service area in the PY. During the base or performance year (respectively, for BY and PY aligned beneficiaries) at least 50 percent of Qualified Evaluation and Management (QEM) services used by the Next Generation Beneficiary were from providers practicing outside the ACO’s service area.