It was a fascinating exercise to read through the news announcement that came out late Tuesday from the federal Centers for Medicare & Medicaid Services (CMS) around the latest statistics coming out of Medicare’s two main accountable care organization (ACO) programs, the Medicare Shared Savings Program (MSSP) for ACOs, and the Pioneer ACO Program. It was fascinating because of the way in which senior CMS officials apparently decided to try to “spin” the numbers in the most positive way possible, to the point where the agency seems to have purposely obfuscated some of the numbers. But let’s talk about the results first.
As Senior Editor Rajiv Leventhal reported in his news article this morning, CMS reported that, “In total, 353 Medicare accountable care organizations (ACOs) generated more than $411 million in net savings in 2014, although many of those ACO organizations did not generate enough savings to receive bonuses. According to the CMS data,” Leventhal reported, “92 of the 333 Medicare Shared Savings Program (MSSP) ACOs held spending $806 million below their targets and earned performance payments of more than $341 million as their share of program savings. In the Pioneer ACO program, which began with 32 ACOs in 2012, but is now down to 20 after several organizations dropped out, 11 organizations generated savings outside a minimum savings rate and earned shared savings payments of $82 million. In total, 103 Medicare ACOs, or 29 percent, received bonuses in 2014.”
Leventhal’s report went on to note that, “What’s more, bonuses aside, the data revealed that 15 out of the 20 Pioneer ACOs (75 percent) and 181 of the 333 (55 percent) MSSP ACOs generated some savings in 2014, meaning that 25 percent of those in the Pioneer program and 45 percent of MSSP ACOs generated no savings last year. As a whole, Medicare ACOs generated over $417 million in savings in 2013, a number slightly higher than what 2014 savings delivered.”
And, his report noted, “Furthermore, in their third performance year, Pioneer ACOs showed improvements in 28 of 33 quality measures. The results show that ACOs with more experience in the program tend to perform better over time, CMS said. Of the 333 Shared Savings Program ACOs, 119 are in their first performance year in Track 1, which involves standing up the program without the financial risk associated with later tracks. Medicare Shared Savings Program ACOs that reported quality measures in 2013 and 2014 improved on 27 of 33 quality measures, according to CMS. “
So here’s the interesting part of this: CMS itself, in its announcement, completely avoided stating the core numbers and percentages of ACOs, both in the MSSP and the Pioneer Program, that failed to generate any savings at all (though fortunately, apparently, none of the ACOs in either program performed so poorly that they actually went into a category in which they cost CMS beyond the normal cost parameters of the Medicare program). As a result, the agency avoided stating the following: in 2014, 181 of the 333 MSSP ACOs generated some level of savings, while 152 ACOs in that program generated no savings; and 15 of the 20 ACOs in the Pioneer ACO Program generated some level of savings, while five generated none. Expressed in terms of percentages, 55 percent of MSSP ACOs generated some level of savings, while 45 percent generated none; meanwhile, 75 percent of Pioneer ACOs generated some level of savings, while 25 percent generated none.
On one level, I can empathize with senior CMS officials, because these results, particularly in the MSSP program, are not exactly spectacular. To state clearly that 45 percent of the ACOs participating in that program generated no savings for the program, would sound discouraging, of course. But the reality is that this kind of work is very difficult, and everyone knows that, particularly everyone involved in the two programs.
What’s more, there are lots of different ways to look at this: yes, 45 percent of ACOs failed to generate savings, but at least a majority, 55 percent, did so. And the news was better in the Pioneer Program, where 75 percent generated at least some savings. The reality that only 92 of 333 MSSP ACOs received shared savings payments, representing only 27.6 percent of all of the ACOs in that program, must probably have been greeted with discouragement among senior CMS officials as well. And of course, these lukewarm results will inevitably give ammunition to the many Negative Nancys out there who, wrong-headedly, are determined to believe that the great accountable care experiment will necessarily end in failure (and really, who needs friends like them, anyway??).
But truly honestly, being a determinedly glass-half-full person myself, I would say this: both of these programs are relatively young ones, and everyone involved in them is working through some very big obstacles and challenges. Indeed, just yesterday, I interviewed two senior leaders at a large medical group that joined the MSSP program in January, who noted that they continue to struggle, several months into their participation in that program, with core patient attribution issues and the data issues related to core attribution issues.
So this is hard stuff, and everyone involved knows the challenges. In that regard, I honestly think that CMS officials could have approached their release of these new numbers in an entirely different way. Now, what they actually did in their press release of yesterday was to provide a very boilerplate-type statement from Acting Administrator Andy Slavitt. Here’s what that looked like: “’These results show that accountable care organizations as a group are on the path towards transforming how care is provided,’ said CMS Acting Administrator Andy Slavitt. ‘Many of these ACOs are demonstrating that they can deliver a higher level of coordinated care that leads to healthier people and smarter spending.’”
Of course, there’s nothing wrong in itself with that statement, and I suppose it’s likely that CMS officials are under pressure to slap a yellow happy-face sticker on everything. But wouldn’t it be better to approach this in a frank, straightforward manner, and in fact congratulate those ACOs that have made significant progress, while noting the difficulty of doing so?
Here’s another example of how CMS officials handled things in their press release. The press release goes on to state that “The results shared today demonstrate significant improvements in the quality of care ACOs are offering to Medicare beneficiaries. ACOs are judged on their performance on an array of meaningful metrics that assess the care they provide – including how highly patients rated their doctor, how well clinicians communicated, whether they screened for high blood pressure and tobacco use and cessation, and their use of Electronic Health Records. In the third performance year, Pioneer ACOs showed improvements in 28 of 33 quality measures and experienced average improvements of 3.6 percent across all quality measures. Shared Savings Program ACOs that reported quality measures in 2013 and 2014 improved on 27 of 33 quality measures.”
These data points could be interpreted in a number of different ways, too. For one thing, they mean that Pioneer ACOs failed to make improvements in six of the 33 core quality measures in that program, and MSSP ACOs failed to make improvements in seven of the 33 core quality measures in that program. Those results are far from tragic. But CMS officials could have been more straightforward, and could easily have noted that they’ve ensured a degree of rigor in all those quality measures that is demonstrated by the very fact that not all of them were improved upon last year.
And honestly, isn’t important to establish and maintain some degree of rigor in all of this? Stated another way, if 100 percent of the ACOs participating in both programs had improved on all 33 quality measures, wouldn’t that mean that the bar had been set too low on performance across those measures? This reminds me a bit of how the meaningful use program is unfolding, since if all hospitals and eligible providers in that program were hit successfully fulfill 100 percent of MU requirements, wouldn’t that imply that that program lacked programmatic rigor?
It just seems to me that CMS officials could do themselves an ongoing favor by being more straightforward about their regular reporting tasks. The reality is that all the ACO programs out there—MSSP, Pioneer, the new Next-Generation ACO Program, and the dozens and dozens of ACO programs being sponsored by private, commercial health insurers—are going to be challenging programs to participate in, since what participants in all such programs are doing is changing healthcare delivery, under the aegis of reimbursement incentives. Really, this stuff shouldn’t be easy—if it were easy, it would already have been accomplished in some form, right?
So I’m find it both fascinating and a bit amusing, and perhaps even a bit irritating, that CMS officials would feel the need to spin the numbers this week, on the latest results coming out of their two main ACO programs. Because, honestly, there really isn’t any need to spin, and indeed, it would be best, in my view, to be bluntly honest about the difficulties and challenges inherent in this work, and therefore actually make the leaders of those organizations that are performing the best in those programs feel even better. Because a glass-half-full approach means recognizing how full a half-full glass really is—am I right?