“[T]his is tough. I don’t know how to proceed…. Lord help the staff who must bring all this together.”
That was how Dr. Francis Crosson, chairman of the Medicare Payment Advisory Commission (MedPAC), reacted to the commission’s baffling discussion at its January 11 meeting moments before it voted 14-2 to replace the Merit-based Incentive Payment System (MIPS) with something called the “voluntary value program” (VVP) (pp. 167-169 of the transcript ). MedPAC’s staff must now summarize the January 11 discussion and prepare a report for inclusion in MedPAC’s March 2018 report to Congress.
MIPS is a pay-for-performance (P4P) scheme imposed on the traditional fee-for-service Medicare program by an act of Congress known as MACRA. MIPS requires that CMS measure performance on cost and quality at the level of the individual doctor, something MedPAC recently acknowledged can’t be done after spending 13 years claiming it could be done.
The portion of the commission’s January 11 discussion that focused on the repeal of MIPS was not hard to understand. The commissioners agreed that MIPS cannot work for multiple reasons, the most important being that the pools of patients treated by individual doctors are too small to permit accurate measurement of cost and quality. “MIPS will not succeed in helping beneficiaries choose clinicians, helping clinicians … improve value, or helping the Medicare program to reward clinicians based on value,” explained MedPAC staffer Kate Bloniarz. (pp. 116-117) Only one of the 16 commissioners present (Dr. Alice Coombs) disagreed with that statement.
Zen and the art of summarizing doubt
It was the commissioners’ discussion about what to replace MIPS with that will be very difficult to summarize. That’s because the discussion consisted largely of expressions of doubt about the VVP, which is essentially a proposal that all doctors who treat Medicare patients either join a “group” (aka ACO) or lose 2 percent of their Medicare payments. The discussion, which followed a vague opening presentation by two MedPAC staff members, consisted of numerous questions posed to the staff that neither the staff nor Dr. Crosson could answer. Because so many issues remained unresolved, ten of the 16 commissioners (one was absent) expressed reservations about voting for the VVP. How does the staff or anyone else summarize a discussion like that? How does the staff explain why the commission voted to recommend the VVP to Congress when a majority of commissioners have multiple concerns about it?
The two questions that got the most discussion were: Will functioning ACOs (as opposed to Potemkin village ACOs that have no internal cohesion) be available for all doctors to join, and; won’t CMS’s crude method of measuring “performance” punish doctors who treat poorer and sicker patients? These and other issues were raised not only at the January 11, 2018 meeting, but at five previous meetings in 2017. That these questions were still unanswered after six meetings is compelling evidence Dr. Crosson and the staff cannot answer them and never will.
Yet despite the obvious discomfort around the table at the last meeting, not one commissioner proposed voting only on MIPS and putting off a vote on the VVP until more research could be done. Commissioners David Grabowski and Paul Ginsburg made it clear they would be willing just to repeal MIPS and leave discussion of a replacement for later, but they were ignored. That’s as close as the commission came to discussing a stand-alone “repeal” recommendation.
Just before the final vote, Dr. Crosson, sensing he had a mutiny on his hands, promised the commission he would bring the VVP back for further discussion. For some reason, that satisfied the doubters. Dr. Crosson then put the entire motion to a vote, and by a margin of 14-2 the commission voted to recommend that Congress repeal MIPS and enact the VVP. 
How will the staff characterize this vote in the upcoming March report? Does this vote mean MedPAC has committed itself to the VVP despite all the concerns expressed by commissioners, and despite chairman Crosson’s promise to continue the discussion about it? MedPAC’s record indicates the answer to the second question is yes. MedPAC’s modus operandi is to recommend evidence-free proposals – P4P, ACOs, “medical homes” are examples – with no details and to leave to Congress and CMS the impossible task of making MedPAC’s lead balloons fly. But the persistent questioning of the VVP proposal by so many commissioners over the course of six meetings, and their recent bad experience with MIPS, suggests MedPAC might rethink its hasty endorsement of the VVP.
If that doesn’t happen – if MedPAC ultimately fails to retract its endorsement of the VVP, and Congress does repeal MIPS and authorize CMS to set up the VVP – the VVP will fail. In that event, MedPAC will have repeated the same cycle it just went through with MIPS. They will have voted for a vaguely worded, evidence-free recommendation to Congress; Congress will have enacted the recommendation into law; and the recommended nostrum will have failed. In that event, MedPAC will have to visit the scene of yet another crashed lead balloon and figure out what to do next.
I will devote the rest of this article to demonstrating that the VVP in its current form is a lead balloon. I’ll do that by discussing two of the more important questions commissioners raised about the VVP that the staff failed to address over the course of an entire year.
(I’ll stop here for a moment to report a very strange development at MedPAC. Sometime around December of 2016 or early in 2017, MedPAC’s staff removed all transcripts prior to September 2010 from their website, and made it impossible to right-click on and copy recent transcripts. On January 23, 2018, I asked an employee of the Geek Squad if he could copy the transcript of the last meeting. He could not.)
Chronology of MedPAC’s latest lead balloon
MACRA (Medicare Access and CHIP Reauthorization Act), which authorized MIPS, was signed by President Obama in April 2015. Although the fatal defects in MIPS were obvious the day Obama signed MACRA , MedPAC waited nearly two years to begin a discussion about whether MIPS could work. Instead of sounding the alarm immediately, MedPAC wasted the rest of 2015 discussing some vague principles that should guide CMS in developing the “alternative payment model” portion of MACRA. 
MedPAC did not buckle down and start discussing the repeal of MIPS until its January 2017 meeting meeting. By then, CMS was already implementing MIPS. The commission’s discussion at that meeting made it clear the staff and most commissioners were ready to recommend the repeal of MIPS. But despite the fact that CMS was already implementing MIPS, the commission waited yet another year before formally recommending the repeal of MIPS. At the January 2018 meeting, Commissioner Alice Coombs criticized this delay. “In 2015, MedPAC did not say anything about eliminating MIPS,” she said. “In our report last year [the June 2017 report to Congress], I don’t remember a … recommendation ever saying anything [about repeal]. [W]e’ve shifted from maybe tweaking it to getting rid of it in twelve months’ time.” (p. 134)
I have no idea how to explain the delay between April 2015, when Obama signed MACRA, and January 2017 when MedPAC finally started talking about repealing MIPS. I do know what caused the failure to act between January 2017 and January 2018: That was Dr. Crosson’s and the staff’s insistence that MedPAC should propose a replacement for MIPS at the same time they recommended its repeal. This insistence manifested the staff’s near-religious devotion to P4P despite a large body of evidence indicating P4P does not work at any level – at the individual physician level, the clinic or hospital level, or even at the level of large pools of providers.
Because of this devotion, staff and Dr. Crosson could not bear the thought of repealing MIPS and thereby leaving 80 percent of the doctors and nurses who treat Medicare patients unmolested by P4P. (Eighty percent appears to be the commission’s latest estimate of the percent of Medicare “clinicians” who will not be in ACOs or other entities that qualify as “advanced alternative payment models” [APMs] under MACRA by 2019, the first year MIPS penalties and rewards kick in.) It was this blind insistence on forcing some form of P4P upon that 80 percent that caused the staff to link repeal with replacement right from the start of the repeal discussion in January 2017.
The staff presented its replacement proposal – what they would eventually call the “voluntary value program” – at the January 2017 and March 2017 meetings. (For more details on these two meetings, see my report here) At both meetings, and at four subsequent meetings (October, November, and December 2017 and January 2018), commissioners peppered the staff with questions about the VVP – to no avail. Dr. Crosson and the staff resolutely refused to add any details to the vague VVP proposal first presented in January 2017. For example, all the staff would say about the “groups” that doctors would have to join in order not to lose 2 percent of their payments is that they must be “sufficiently large to have statistically detectable performance on population-based measures” such as use of “low-value” services and “healthy days at home” (see p. 12 of the transcript of the October 7 meeting). That was their story, and they stuck to it. The skeleton-like proposal the commission voted on at the January 11, 2018 meeting was identical to the skeleton-like proposal they first heard in January 2017.
Two questions staff refused to answer
To get some sense of how often commissioners sought to get answers from staff and how futile their efforts were, I review in this section the efforts commissioners made over the course of January 2017 to January 2018 to get answers to two important questions raised by the VVP proposal:
- Will ACOs, “alternative payment models,” or other “groups” be available for doctors to join if they want to escape the 2 percent penalty?
- Will “groups” that treat sicker and poorer patients suffer under the VVP proposal?
The first question was the one most frequently raised by commissioners, and the one most frequently dodged by Dr. Crosson and the staff. The importance of this question is obvious: If doctors are going to be penalized for not joining a “group,” MedPAC has an obligation to determine first that groups will be available for doctors to join. At the January 2017 meeting, commissioner Craig Sammit began the discussion of the VVP proposal with this question: “[D]o we have a sense of how accessible and feasible it will be for clinicians to join APMs? … [W]ill it be feasible for every clinician who wants to be in an APM to get into an APM?” (p. 248, transcript) David Glass’s reply on behalf of the staff indicated how little thought the staff had given to its proposal: “I guess it depends how it eventually works out;” and “I would say that’s a little hard to say right now.” (p. 249)
One year later, it was still “a little hard to say.” When the January 2018 meeting rolled around, staff still had no answers to Dr. Sammit’s question. At the January 2018 meeting:
- Commissioner Paul Ginsberg noted “there’s a lot of uncertainty about what proportion of physicians will get into groups.” (p. 121)
- Commissioner Kathy Buto stated, “I don’t know that we know that much about APMs yet.” (p. 134)
- Commissioner David Nerenz stated: “[T]here is a significant amount of social engineering going on here with no real evidence.… We’re talking about pushing physicians into groups that have to be a certain large size in order to have denominators big enough to do the least sensitive of measures. I’ve looked and looked, and I see no empirical evidence that that structure is better than other structures.” (p. 140)
- Commissioner Warner Thomas observed that doctors “in some areas … have trouble getting into APMs.” (p. 144)
The staff was equally disinterested in addressing the commissioners’ concern that the VVP will harm the sickest and poorest Medicare beneficiaries. Numerous commissioners made it crystal clear, repeatedly, over the course of the six meetings held between January 2017 and 2018 that they expect the staff’s VVP proposal will worsen disparities via two mechanisms: Providers who treat the sick and the poor will be disproportionately exposed to the 2 percent penalty because they can’t form or join groups; and for those who do make it inside a group, CMS’s crude method of adjusting cost and quality scores for factors outside physician control will punish their group because the group treats sicker and poorer patients.
The staff’s response to these concerns was brief and superficial. They simply threw out two phrases – “risk adjustment” and “peer grouping” – and alleged that these phrases represented adequate solutions, and said no more.  Understandably, by the time the January 2018 meeting occurred, many commissioners were still worried. “I get chills … thinking about the doctors who are in the trenches taking care of vulnerable patients [in] Watts or Compton,” said commissioner Coombs at that meeting. (p. 136) “[P]oor people will be hurt by this” said commissioner Nerenz. (p. 140) “The idea that we could actually be creating or encouraging a greater disparity in physicians who treat low-income populations, I think that’s something to worry about,” said commissioner Kathy Buto. (p. 148)
Chairman Crosson was unfazed. “I do think this has been a thorough process,” he declared near the end of the January 11 meeting. (p. 147) That statement was so inaccurate it may accurately be called unethical.
Lessons from another “repeal and replace” campaign
All sentient readers will recall the sad ending of another “repeal and replace” campaign. It had a sad ending because it was driven by ideology, not evidence. MedPAC appears hellbent on repeating the same mistake Republicans made in declaring their intention to “repeal and replace” the Affordable Care Act. MedPAC linked repeal of MIPS with its replacement before they had time to research the replacement. MedPAC might as well have said, “We will replace MIPS with something terrific!” MedPAC thereby set itself up to repeat the MIPS fiasco and the Republicans’ “repeal and replace” fiasco. 
As it became clear near the end of the January 2018 discussion of the repeal and replacement of MIPS that chairman Crosson was going to put the entire repeal-and-replace motion to a vote despite all the unanswered questions, commissioner Thomas raised the possibility of postponing the vote. “[P]erhaps we ought to take a step back and just say, ‘Are we headed in the right direction?’” he suggested. (p. 153) He went on to warn his fellow commissioners they were about to repeat their MIPS mistake. “I don’t want to replicate … MIPS,” he said, “where essentially you had something that was probably not as well thought out as it could have been, and yet it was put together and … made law and implemented.” (p. 164)
It didn’t work. Dr. Crosson did promise to bring the VVP back for more discussion. Whether that was a victory for the ten doubting commissioners remains to be seen. Dr. Crosson then put the entire repeal-and-replace motion before the commission. Fourteen voted yes and only two voted no. It’s hard to see how this ends well. Either the VVP, possibly with a few tweaks, becomes law and fails, or MedPAC will retract its endorsement, which will raise questions about why they rushed to judgement.
 Here is the wording of the recommendation that so concerned the commissioners which nevertheless passed by a 14-2 vote: “The Congress should eliminate the current merit-based incentive payment systems; and establish a new voluntary value program in fee-for-service Medicare in which clinicians can elect to be measured as part of a voluntary group; and clinicians in voluntary groups can qualify for a value payment based on their group’s performance on a set of population-based measures.” (p. 118) According to Chapter 5 of the commission’s June 2017 report to Congress, examples of the “population-based measures” the commission has in mind are: “potentially preventable admissions and emergency department visits; mortality and readmission rates after inpatient hospital stays; healthy days at home; patient experience; rates of low-value care; and relative resource use.” (p. 167) These measures, allegedly can be accurately calculated by CMS from claims data.
 On April 12, 2015, three days before Obama signed MACRA, I posted a blog comment predicting MIPS would fail. If I could see that so clearly at that time, why couldn’t MedPAC?
 MedPAC staffer Kate Bloniarz identified the defects in MIPS for the first time at the commission’s October 2015 meeting, but she and the rest of the staff insisted the commission devote what time they had for MACRA to the “alternative payment model” portion of MACRA. (See her brief remarks on MIPS at pages 92 and 93 of the transcript )
 Readers who wish to see for themselves how superficial the staff’s response to the issues raised by commissioners were should read pages 5-8 of the transcript of the November 2017 meeting.
 For a discussion of MedPAC’s history of endorsing vaguely articulated, evidence-free managed care notions, see my discussion of MedPAC’s endorsement of the “three pillars of MACRA” over the period 2003-2008.