Earlier this summer, the Centers for Medicare & Medicaid Services released a final rule that updates provisions of the Medicare Shared Savings Program, the nation's largest accountable care organization experiment involving more than 400 ACOs serving some 7 million Medicare beneficiaries.

At 592 pages long, the final rule attempts to accomplish two basic goals: increase participation in the still nascent MSSP while gently nudging ACOs to take on financial risk for improved quality and patient outcomes. It's a step?a leap, really?that few ACOs are willing to take.

In fact, of the 404 ACOs currently participating in the MSSP, only four are following Track 2, a two-sided model that allows ACOs to share in any savings generated by the ACO, but also requires them to pay back any losses to CMS. The vast majority of MSSP ACOs are following Track 1, a shared savings-only model.

The final rule introduces Track 3, a new model that includes higher rates of shared savings, prospective assignment of beneficiaries and the opportunity to use new care coordination tools. Perhaps more importantly, the final rule also allows ACOs to stay in Track 1 at the same shared-savings rates, delaying the pain of taking on performance-based risk for another three years. Without the changes, some ACOs had threatened to leave the MSSP. With the changes, which take effect with agreements with a 2016 start date, CMS expects 90 percent of ACOs to stay.

But it's what the final rule does not address?at least not definitively?that stands to make or break the MSSP over the next three years. That issue is benchmarking, the methodology that CMS uses to establish the baseline for ACO performance and, by extension, the threshold at which an ACO is eligible to share in savings or risk.

When setting an ACO's benchmark, CMS looks at the most recent available three years of beneficiary expenditures for parts A and B services for Medicare fee-for-service beneficiaries assigned to the ACO. If an ACO's beneficiary expenditures come in at a certain level below the benchmark, it gets a check from CMS for a portion of the difference. If the ACO is in Track 2 or Track 3 and its beneficiary expenditures come in at a certain level above the benchmark, it writes a check to CMS.

Benchmarks are different for every ACO and can be adjusted based on a myriad of factors, including unique characteristics of the ACO's assigned patient population. They?re set at the beginning of each three-year agreement period and then updated each year of the agreement based, again, on a number of factors. One of these is the projected amount of growth in the national per-capita expenditures for parts A and B under traditional Medicare fee-for-service.

Benchmarking is a sore spot for MSSP ACOs, which argue the current methodology is unfair and effectively punishes high-performing ACOs, forcing them to chase after diminishing returns in subsequent agreement periods when the benchmark is reset. In comments to the proposed rule released in December 2014, some ACO leaders argued they have to continually beat their own best performance in order to achieve share in savings, while higher-cost ACOs are rewarded for their historical inefficiency with higher benchmarks.

The numbers would seem to bear that out. Of the 220 MSSP ACOs that have reported results thus far (those with agreements starting in 2012 or 2013), 118 achieved savings over traditional Medicare FFS; yet in large part because of the way benchmarks are set, only 52 were eligible to share in those savings. Moreover, of the 52 that did share in savings, the average payout was $6.1 million. When you consider the large investment required to launch an ACO and estimated operational costs of $1 million to $2 million annually, the fact is most MSSP ACOs are losing money while even the best among them are breaking even.

CMS knows its benchmarking methodology is a problem, and the final rule released in June 2015 makes a few key improvements requested by the ACO community. For example, when resetting the historical benchmark for ACOs in their second or subsequent MSSP agreement, CMS will weight each year equally instead of the progressive weighting used in the original rule. The new rule also accounts for any shared savings earned by an ACO when resetting its historical benchmark at the start of a second agreement period.

But even CMS admits the new provisions don?t go far enough. Moreover, the agency declined altogether to tackle one of the biggest complaints against benchmarking: using national cost trends when establishing, updating and resetting benchmarks. Like many in the ACO community, CMS would prefer to use regional or local market cost trends, but exactly how to do that remains a huge question mark. 

CMS has said it plans to propose and seek comment on a new benchmarking methodology later this summer. Even if a final rule is issued by the end of the year, changes to the benchmarking methodology would take effect with ACO agreements starting in 2017. That's too late for the 220 MSSP ACOs whose current agreements expire Dec. 31, 2015. Those ACOs had to decide whether to stay in the MSSP under the current rules by June 1, 2015, when applications for agreements with Jan. 1, 2016, start dates were due. CMS is expected to accept or reject those applications this week, and new contracts will be signed this fall.

That means Track 1 ACOs, those considered to be one-sided risk only, that have struggled to achieve shared savings thus far but do not want to walk away from their ACO investment could lose even more money over the next three years. It will be late 2018 before they have another opportunity to sign new MSSP agreements that likely will use revised benchmarking methodology. Sure sounds like a risk to me.
Follow April Wortham Collins on Twitter @aprilworthamDRG

A novel approach to treating neurodegenerative disease

View Now