Skip to main content

Pathways to Success – MSSP Continues to Evolve

Written by Travis Broome, VP of Health Policy

Just before the holidays, the Centers of Medicare & Medicaid Services (CMS), led by Administrator Seema Verma, released the latest evolution of the in Medicare Shared Savings Program – Pathways to Success in the continuing effort to measure the value of value-based care. What follows is a high-level summary of the key changes in the final rule. With the uncertainty behind us, we can now all take the next step towards an outcome-based future.

Basic and Enhanced Tracks

Since the beginning of the program, CMS has recognized that one size does not fit all and has used “tracks” in MSSP. In Pathways to Success, CMS replaced the current Track 1, Track 2, Track 3 and Track 1+ models with two tracks: Basic and Enhanced. Basic is a 5-year glide path from the old Track 1 to the old Track 1+. Enhanced is essentially Track 3. Every ACO will move from current tracks to this Basic/Enhanced framework over time. Basic E and Enhanced qualify as advanced alternative payment models.

A Note About 2019

The July 1st start date for 2019 creates unique concerns around how to calculate performance in 2019. All financial and quality metrics are based on the whole of 2019. There are no allowances in reporting, in reconciliation or any other financial or quality measures for the fact that a practice may only be in the ACO 6 months.

For ACOs that switch to the new Tracks on July 1, CMS will evaluate costs for the calendar year 2019 twice: once under the existing rules and once under the new rules.

First Six Months – $4,000,000 in Savings Under Old Rules and a 90% Quality Score
Total Savings x Share Rate x Quality Score x Portion of the Year
$4,000,000 x 50% x 90% x 50% = $900,000

Last Six Months – $5,000,000 in Savings Under New Rules and a 90% Quality Score
Total Savings x Share Rate x Quality Score x Portion of the Year
$5,000,000 x 75% x 90% x 50% = $1,687,500

Total 2019 Payments to ACO: $2,587,500 ($900,000 + $1,687,500)

Benchmark Methodology

The hardest thing in value-based care is continually improving every year. The second hardest thing in value-based care is measuring the value being created. Programs have to predict what the world would have been had the accountable care organization (ACO) not existed. A mix of policy, economics, and actuarial science must come together to set cost targets or benchmarks for ACOs. The new changes in benchmarking methodology place more weight on how an ACO performs compared to other providers in their local markets and less weight on how the ACO performs compared to the national average. There are many changes in the rule. We talk about our top three.

Risk Adjustment

Risk adjustment2 can go up (subject to a 3% from benchmark cap) and down (no cap) in any year now, but it is the difference between the ACO’s risk experience and the region’s experience that matters. If an ACO’s risk score goes up by 4% from the benchmark and the region’s risk does not change, the ACO’s costs will be adjusted by 3%. However, if an ACO’s risk score goes up 5%, but so does the risk score in the ACO’s region, then the impact on the ACO’s benchmark is mostly negated.

This was a much-needed change from CMS’s old policy of pretending that established patients of physician practices never got sicker. This policy falls solidly in the not perfect, but better than before bucket. In the rule, CMS said that nearly 30% of ACOs will hit the 3% cap. We think that is far too many and will continue to encourage CMS to raise the cap so that it only captures outliers.

Regional Benchmarking

The most impactful change of the new rule was the incorporation of regional benchmarking in all years of the new tracks. We are strong proponents of regional benchmarking going back to 2014 when we were talking about “blended” benchmarks based on historical costs and regional cost efficiency.

As with the current rules, the percent of the ACO benchmark made up of regional efficiency increases over time and depends on whether the ACO is regionally efficient or not.

CMS is also now capping the bonus that comes from regional benchmarking at the category (ESRD, Disabled, Dual, Aged) level. The cap is 5 percent of the national average for that category. For example, if the national average for aged beneficiaries is $10,000, then the regional bonus for the aged category can be no more than $500.

Regional Inflation Blended with National Inflation

Just as regional benchmarking is moving to the first agreement period for an ACO, so is regional inflation. Unlike the blended benchmark, we support 100% use of regional inflation as a better measure of ACO performance. National inflation creates advantages and disadvantages to the ACO that are not due to the work of the ACO, but are really the result of luck.

Regional inflation is defined as the year-over-year change in risk-adjusted costs in the counties in which the ACO has patients.3 CMS includes the ACO’s assigned beneficiaries in the county cost and risk. So when an ACO reduces costs, it also reduce the county’s costs, lowering regional inflation. If an ACO is in New York City and only accounts for 2% of the region’s beneficiaries, this is not a big deal; however, many rural ACOs account for over 20% of their region and 13 ACOs nationally account for more than 50% of their respective regions. The greater an ACO’s market share, the less it benefits from its own cost reduction accomplishments. We detailed the problem in an article early in 2018.

CMS acknowledged the problem, but their “solution” doesn’t solve the problem. They will blend regional inflation and national inflation together for ACO based on market share. If the ACO accounts for 20% of the beneficiaries in their region than their inflation update is 80% regional and 20% national. This reintroduces the luck factor into the inflation update. If an ACO is in a region with lower than national inflation, it benefits from this policy. If the ACO is in a region with higher than national inflation, it loses under this policy. We continue to encourage CMS to directly solve this “rural glitch” and remove the ACO’s beneficiaries from regional comparison.

Final takeaways:
1.ACOs have to move to risk faster in new tracks with longer contracts

  • Understand the difference between revenue-based risk and total cost of care risk
  • Consider moving to risk as soon as the ACO is ready to get higher savings

2.ACOs are measured against their local markets more than ever before

  • Know whether your ACO is regionally efficient or not
  • Monitor your region to understand how changes in it will affect your performance
  • Be aware of your patients’ risk scores – the more accurate the risk score the more accurate the ACO’s benchmark

3.ACOs have more choices to make than before and more than we could cover here

  • Choose annually between prospective and retrospective assignment
  • More waivers than before: SNF 3-Day waiver, telehealth and beneficiary incentives

The evolution of the Medicare Shared Savings Program to Pathways to Success increases our ability to measure value. Just as those creating the value must continually improve so must the measurement of value continuously improve. We look forward to the bright future in Pathways Success and the ACO movement.



  1. A low-revenue ACO is where the Medicare revenue received by the ACO providers for all patients is less than 35% of the total cost of care of the ACO assigned patients. Analysis of MSSP results through 2016 by CMS showed that low revenue ACOs, which include fewer components of the health care system in the ACO itself, performed better as a cohort than high revenue ACOs. While physician-only ACOs tend to be low revenue this is not always the case. Each ACO should conduct an analysis of its Medicare revenue to determine its status.
  2. CMS used the Hierarchical Category Condition Model (CMS-HCC) to reflect the health status of the ACO’s assigned population. This model is prospective meaning it attempts to predict next year’s cost using information about a person’s health condition this year.
  3. The change from year to year in each county is weighted by the percent of ACO assigned patients who live in that county. If 20% of the ACO assigned patients live in Montgomery County then Montgomery County’s 2019 inflation will make up 20% of the ACO’s regional inflation for 2019.