Value-Based Tools

With CMS Demanding More Accountability in the MSSP, Will It Be ‘ACOs Overboard’?

Three quarters of the original ACOs that remain in the MSSP are still in Track 1. The Trump administration is tired of waiting for them to take on two-sided risk.

That splash you hear may be dozens of ACOs jumping off the deck of the USS Seema Verma.

For 169 ACOs, time is up to take on downside risk. This group—30% of all of those in the Medicare Shared Savings Program (MSSP)—are Track 1, upside risk–only ACOs and are nearing the end of their contracts with CMS. Many have said they’re not ready to leap to two-sided risk, but in August CMS Administrator Verma rejected pleas for an extension in Track 1. In essence she said, it’s time to move up or walk the plank.

CMS’s proposed rule about changes to the MSSP, released August 9, was widely expected to place stronger limitations on the length of time an ACO may take upside-only risk. It did indeed, but the agency’s unequivocal end of indulgence caught many by surprise. The proposed rule shortens the upside-only risk period from six years to two. The likely outcome, said President and CEO of the National Association of ACOs (NAACOS) Clif Gaus in a statement released to the media, “will be that many ACOs quit the program, divest their care-coordination resources, and return to payment models that emphasize volume over value.”

Armed with an analysis to show that Track 1 ACOs are costing rather than saving Medicare money, Verma summarized the proposal in a Health Affairs blog. CMS plans to scrap the MSSP’s four tracks in favor of a two-track program to accelerate the migration of providers into advanced alternative payment models (APMs) under the Quality Payment Program. CMS would consolidate the MSSP’s existing Track 1 (upside–risk only) with Tracks 1+ and 2 (two-sided risk) into a single “basic” track, in which ACOs start with two years of upside risk–only and progress to higher levels of twosided risk in successive years. Track 3—a higherrisk, higher-reward proposition—is essentially unchanged and renamed the “enhanced” track.

Currently, ACOs sign up for three-year stints. Under CMS’s new scheme, enrollment in either track would last five years, with the basic track being a one-time-only affair. ACOs in the basic track would qualify for advanced APM status— and the 5% bonus that comes with it—by Year 5, while those in the enhanced track would automatically be an advanced APM.

Warning signals

While the degree of change to the MSSP may have surprised some, all indications were that the administration was in no mood for leniency. In March, HHS Secretary Alex Azar said that value-based care “needs to accelerate dramatically” and that results of the ACO program had been “lackluster.”

In May, Verma put a chill on attendees at the American Hospital Association annual meeting, saying that Track 1 ACOs are increasing Medicare spending. “The presence of these upsideonly tracks may be encouraging consolidation in the marketplace”—a point she reiterated in her Health Affairs blog. “While we understand that systems need time to adjust, our system cannot afford to continue with models that are not producing results,” she warned at the AHA meeting.

A month later, Joe Grogan, director of health programs at the Office of Management and Budget, told the National ACO, Bundled Payment, and MACRA Summit that ACOs “need to accept risk sooner than later.” That Grogan delivered his remarks to this group was “a pretty clear indication that he [was] trying to send a signal to the community that changes are coming,” says Larry Kocot, Washington-based principal and national leader of the Center for Healthcare Regulatory Insight at KPMG, and a former senior adviser to the administrator of CMS.

Forcing Track 1 ACOs to take on risk at the end of their current agreement periods may well chase some of them right out of the MSSP. In May, NAACOS surveyed the 82 charter ACOs that are still in Track 1. Among respondents, 71% said they would be likely to leave if they had to assume risk. Of those, 39% said the level of risk they would have to take on was too great.

“There’s a reluctance, especially for ACOs that have not had numerous years of success with this program,” says Allison Brennan, NAACOS’s vice president of policy. “It’s a very long learning curve, and it’s a challenging program. We still only see about a third of ACOs each year earn shared savings.”

Evidence is mounting that the longer an ACO works with risk, the better it becomes at managing it. An analysis by Robert Saunders, David Muhlestein, and Mark McClellan on the Health Affairs blog found that the earliest class of ACOs—those that joined the MSSP in April 2012—produced a net savings per beneficiary of $140 in 2016, the latest year available. For ACOs that joined in January 2016, the net at the end of the year was minus $25.

And these figures don’t account for the cost to operate an ACO. NAACOS estimates that the investments ACOs make in information technology and population analytics average $1.6 million annually. An ACO receives no credit against its CMS-established cost-reduction benchmark for start-up and operating costs— investments that take time to recoup even as the ACO is trying to reduce spending.

Almost 40% of the ACOs in the NAACOS survey that were inclined to walk away said they were concerned about unpredictable changes to the MSSP program. That’s not hard to understand, given that they were the early adopters. In its first three years, Kocot points out, the MSSP underwent a lot of transition, and ACOs struggled to make sense of the changes.

“The thing is, do you really want to throw the baby out with the bathwater? They’ve really learned a lot, and by all rights these early adopters are performing well and have all the promise to move to risk. They big question is when and under what conditions?”

And the number is…

Of course, it was never CMS’s intention to let them stay in Track 1 forever, and CMS has already given the early adopters a number of breaks. Originally, an ACO could take on upside-only risk for just three years, but the final MACRA rule allowed Track 1 ACOs to stay put for a second, three-year hitch—much to the liking of NAACOS.

In its proposed rule, CMS acknowledged that ACOs’ extended stays in Track 1 may have had a couple of unintended consequences. One is provider consolidation, which drives up costs. An ACO cannot tell beneficiaries where to get care, so to reduce “leakage”—the percentage of attributed beneficiaries who get care outside of the ACO—hospital systems buy medical practices or merge, in essence expanding their own hospital and physician networks. The other is the possibility that some ACOs are in it for the wrong reasons. ACOs are shielded from federal self-referral and anti-kickback laws, and CMS is concerned that some hospital systems may be using these protections to game the system.

Perhaps more importantly, yet another extension in Track 1 would have contradicted CMS’s goal of moving providers into advanced APMs, wrote Kocot and Ross White, a manager in the Center for Healthcare Regulatory Insight at KPMG, in a pair of Health Affairs blog posts in July. Ask Kocot and White, and they will tell you that an underlying question that often surfaces in discussions about when ACOs should be required to accept two-sided risk is: Is the MSSP actually saving money? The answer depends on whom you ask and how the question is framed.

By CMS’s reckoning, Track 1 ACOs cost Medicare $49 million in 2016 after accounting for shared savings payments. The Health Affairs analysis by Saunders and colleagues showed a Year 2016 $39 million net loss across the MSSP after CMS paid out bonuses, the lion’s share of which went to Track 1 ACOs. An Avalere analysis found that upside risk–only ACOs collectively have increased federal spending by $444 million while two-sided risk ACOs have reduced spending by $60 million.

In June, MedPAC weighed in, saying that the program may have saved more money than CMS believes. MedPAC compared ACOs’ performance to what it called the “counterfactual”— that is, if the ACOs did not exist. In this light, said MedPAC, total savings to Medicare may amount to one to two percentage points more than CMS believes, but MedPAC didn’t indicate whether that meant the program is actually saving money in the aggregate.

“They could be saying, ‘If the benchmarks were more equitable, ACOs may have been saving one to two percent more,’” says White. “Or they could be saying the benchmark methodology in general is not the most accurate measure of actual savings, and that it should be counterfactual to similar providers in that region who are not participating [in the MSSP].”

The essence of MedPAC’s remarks, believes Brennan, at NAACOS, is that whether an ACO beats its benchmark doesn’t tell the whole story. “The way to do that is through more sophisticated research comparing ACO providers and beneficiaries to those outside of ACOs.”

CMS didn’t do this in its proposed rule, though it did align its benchmarking methodology more closely with that of Medicare Advantage. CMS says this will help to avoid pitfalls that penalized ACOs that managed to reduce spending.

The right policy decision?

CMS’s proposal defied the wishes of NAACOS and a number of other groups, including the AMA and Premier Hospital Alliance, that had asked CMS to allow any of the original ACOs that are still in Track 1 to remain there for a third, three-year period if they have demonstrated either superior quality or net savings for four years. “Some of these ACOs have said ‘we’re on the right path, but we’re not there yet and our performance hasn’t been strong enough to make that leap,’” says Brennan.

CMS’s response raises the question of what it means for the MSSP if dozens of ACOs simply jump ship. “It depends on who’s leaving,” says Kocot, at KPMG. “If it’s those who have done reasonably well but are fearful of risk, it would be a loss.”

Kocot and White, his KPMG colleague, had recommended a middle path: Rather than hold all Track 1 ACOs to the same timetable for accepting two-sided risk, CMS should institute a waiver process that would allow those that can demonstrate that they are making progress to stay in Track 1 a little longer. These ACOs would have to commit to a plan showing how and when they will be ready to transition to two-sided risk. They would have to follow the plan or ship out.

“How do you achieve the program goals without compromising the program? It seemed to us that letting people go at their own pace wasn’t a sufficient answer,” says Kocot, but “making them jump to risk before they are ready is not a very good policy decision, either.”

We’ll know what ACOs think very soon.

Two-sided risk is hard, says Allison Brennan of NAACOS. “It’s a very long learning curve, and it’s a challenging program.”

Many ACOs will leave the program if CMS proceeds with its proposed rule change, warns Clif Gaus, head of the National Association of ACOs.

A middle path on upside-only risk might have been the most prudent policy decision, says Larry Kocot at KPMG. In its proposed rule, CMS disagreed

Our most popular topics on