What’s Next for Medicare Provider Payment?

(Please note this is an unedited transcript, for exact quotes please reference the video.)

SARAH DASH:  Alright, well good afternoon, everyone. Thank you so much for joining us on a busy day in Washington for today’s briefing on what’s next for Medicare provider payments.


My name is Sarah Dash, and I am President and CEO of the Alliance for Health Policy. For those who are not familiar with the Alliance, we are a non-partisan organization dedicated to advancing knowledge and understanding of health policy issues. So, we are really glad you’re here, and hello as well to those who might be on Twitter at the hashtag #allhealthlive. You all can feel free to tweet while you’re here too, and feel free to submit questions that way as well.


Before we get started, I would like to thank the Commonwealth Fund for making today’s briefing possible, and for their partnership and support.


Let’s dive into the topic. After a deliberative process lasting since the passage of the Medicare Access and CHIP Reauthorization Act in 2015, earlier in November, CMS released its final rule pertaining to MACRA quality payment program. And as we know, efforts are well underway around the country to encourage a shift to value-based payment models for Medicare providers. Meanwhile, the Medicare Payment Advisory Commission — MedPAC, recently recommended a move away from the key component of MACRA — the Merit Based Incentive Payment System, towards a different approach to value-based payment, and there are many nuances and complexities to the implementation of many of these measures. So, we are here today to examine the issues that are on the table, as healthcare providers around the country work to understand and implement these measures, and as policymakers consider additional shifts to the way providers might be paid. And what all of this really means for improving health outcomes and quality, and for healthcare costs.


You are going to hear today from five really excellent speakers, each of whom brings a different perspective to the discussion, and we are really grateful to have them here today, to shed some light on the discussion.


Joining us today, and I will just introduce them in speaking order. Len Nichols, who is Director of the Center for Health Policy, Research and Ethics at George Mason University. Len was also appointed to the physician-focused Payment Model Technical Advisory Committee, or PTAC, which was created in the MACRA legislation. Len holds a PhD in Economics from University of Illinois at Urbana-Champagne.


Next, Aucha Prachanronarong, who is Director of the Electronic and Clinician Quality Division at the Centers for Medicare and Medicaid Services. And Gregory Woods, who is the Director of the Division of Alternative Payment Model Infrastructure, at the CMS Innovation Center. And they are going to give a joint presentation about efforts at CMS and CMMI. Aucha has a Masters in Health Sciences and Health Policy from Johns Hopkins University, and Gregory holds a Masters in Public Affairs from Princeton’s Woodrow Wilson School.


To my left is Elizabeth Mitchell. She is President and CEO of the Network for Regional Healthcare Improvement, and serves as Vice Chair of the PTAC. She’s also a Guiding Committee Member of the Healthcare Payment Learning and Action Network, or the LAN, and she’s on the Quality Improvement Strategy Technical Expert Panel. So, we are really delighted to have her today. In addition to all of those accomplishments, she’s also served two terms in the Maine State Legislature, and chaired their Health and Human Services Committee.


Finally, last, but certainly not least, we will hear from Mark Miller. Mark is the Executive Director of MedPAC, a non-partisan federal agency that advises Congress on Medicare payment, quality, and access issues. Prior to joining MedPAC in 2002, Mark held leadership positions at the CVO, at CMS, and at OMB. He holds a PhD in Public Policy Analysis from State University of New York at Binghamton, and I have to give a special note, which is, we are really honored to have Mark here today, because this is his last day at MedPAC, and I hope I’m not breaking any news here. Mark has really done extraordinary work with the Commission over the past 15 years, and has really been on a number of Alliance panels as well, helping to eliminate complicated Medicare issues, so we are really, really thrilled that you joined us here today, and wish you well on your next move.


With that, I’m going to turn it over to Len Nichols, to kick things off.


LEN NICHOLS:  Okay. Can you hear me okay in the back? Good.


So, I will just say, basically, very briefly, I’m going to talk about payment reform, but it’s also true, since I have the grayest beard in the office, they wanted me to start with just a little bit of history. All of you of course went to good schools, and learned everything you need to know in kindergarten. It took me till 7th grade, “Two roads diverged in a yellow wood.” You may remember that poem. And so, what I want to show you, is essentially, believe it or not, I know it’s hearsay to say it in this building, but believe it or not, it doesn’t matter who wins these next set of elections, everybody is going to want to cut healthcare cost. Some people would like to spend less on healthcare and Medicare in particular, so that they can spend it on coverage expansion. Some people like to use it to pay for tax cuts, but everybody is going to be pretty seriously focused on getting healthcare costs down, because they have ulterior motives to spend it on.


Now, here comes the first history lesson:  This is the single most important variable that connects the budget debate with the health policy debate, and you are going to hear more and more and more about that, after they vote on the tax cut. And what you have here is the ratio of GDP claimed by debt. That is, what we owe each other, and the Chinese, as a ratio to GDP; which is of course our total National Product. I found this incredible graph, which comes from, I guess, CBO. Yeah, it goes all the way back to the revolution. So, it starts with 1789, and all of that. And we borrowed 25% of our GDP from the French to win the war, and then we had to pay them back, and we did. And we had basically no government there for a while. And then we had the Civil War, and we borrowed some more, and paid it back again. All of the widows finally died off, and so then we didn’t spend any money again. And then World War I, keep going, okay. Then you see the Depression, and it kind of went up a lot. Then you see World War II went up a whole lot. It went up over 100% of GDP, which is proof that you can live that way for a very short amount of time, but it’s also true, if you look very carefully, you will see after World War II, it started coming down. What that is — I know you have never seen this in your lifetime, but this is called a bi-partisan consensus to get your fiscal house in order. And what we did was, we did what you are supposed to do. You borrow when you have to, because you had to borrow to build those battleships, to defeat tyranny around the world. We face an existential threat to our way of life. But then we paid it off when we could. And that’s what happened. And we got it down to 20% of GDP, and there were some hiccups. All those little hiccups on the way down are recessions. Big recession in ’74 when OPEC happened, and oil prices jumped up. But still, we were in high 20s, low 30s. Then it started going up in 1981. Anybody remember what happened in 1981? Cardinals did not win the World Series. Ronald Reagan. And Ronald Reagan, sort of like today, told us that we could cut taxes and balance the budget. It turns out, you can’t. so, what happened was, debt started going back up, until Ross Perot. Anybody remember Ross Perot? I personally love Ross Perot. He’s the only public speaker that has bigger ears than me. And Ross Perot bought TV time with his one money, held up one little Power Point, and said, we’ve got to do something about this debt. And he did that enough so that Clinton and Gingrich — notice, bipartisan. Notice, don’t agree on much, but did agree on a balanced budget act, boom, and then the debt started to fall in a serious way. When Clinton left office, he had a two hundred-billion-dollar surplus. And that rocked along until we had Medicare drug benefit we didn’t pay for, we had a couple of wars we didn’t pay for, and then recent economic Great Recession, and the stimulus package was the last third of that.


So, again, we face the next single threat to our way of life — it’s okay to borrow when you are in a big bind. But then you need a bi-partisan consensus to pay it off when you can. That is what we lack, which is why it heads up and up and up, and it’s going now — it’s headed toward a direction way above 100% of GDP, and there will come a point when we have no choice, because guess what? The Chinese want their money back. So, we do have to deal with this, and let me tell you, there is no way to deal with this unless we deal with healthcare cost growth, which is why everybody cares.


Okay, so what you have next, is another version of history lesson, focused on Medicare. And what I’m going to show you here, all the different big policy changes that occurred through time. What you have here, is Medicare spending as a shared GDP. It starts out, 65, when the program was created very low. About a half a percent of GDP until 1970. Then it starts going up, and you see TEFRA. How many of you know what TEFRA is? Very good. Historians in the bunch. TEFRA invented what we now call Medicare Advantage. Private sector health plans to organize Medicare services. DRGs — have you heard of DRGs? DRGs did not come from God, they started in 1983; and what they did was they changed the way we paid hospitals, and look what happened. Share of Medicare, claim on GDP, was pretty flat after DRGs. They worked. Fundamentally changed incentives for hospitals, length of stay fell a day and a half within a year. And nobody died. It was pretty amazing. And once they figured out they could get the Medicare population out the door quicker, they started getting everybody out the door quicker. It actually did change incentives in a serious way.


Then RBRS. That’s how we pay docs, right? RBRS. Look what happened. That might be the single biggest disaster in Mark Miller’s lifetime. RBRS just didn’t work. We started paying ‘em more systematically, but guess what? The people who control the payment system like to create new codes, and now we have 9000 codes, and off we go.


So, then we got the BBA. Do you remember the BBA? The BBA was the Clinton/Gingrich solution to — and look it worked — Medicare share, GDP fell. It fell so much, people got nervous, and of course they started to try to modify it. And then back up in Part D, and so forth.


And then the ACA, and look, the only other period where Medicare as a shared GDP is relatively constant, only other period other than the five or ten years after DRG is the ACA. Now, we could say that’s because of the brilliant activities that have been going on in the innovation center. And lord knows, there are brilliant activities going on in the innovation center, but that ain’t why this happened. This happened because of market [inaudible] reduction. It happened because of Medicare payment for hospital changes. Right? And that’s really what drove those savings. Therefore, what we have is a bunch of ideas that are being tested. I’m going to talk about them in just a minute. They are trying to set in motion a set of incentive changes that will at least try to hold the rate of growth of healthcare down to the rate of growth of the GDP, preferably below overtime, and it leaves — it tries to get Medicare under control. MACRA, of course, the last little piece of the puzzle so far, you may get the theme here, we keep trying to do this because we don’t really succeed. MACRA is designed to turbo-charge the way we change, the way we pay physicians, and we will talk about that the rest of the time.


What do we know so far? Well, kind of interestingly, a whole bunch of evaluations have come out in the last year of the very clever — I’m not being factitious at all — very clever innovations that were tested as a result of the Affordable Care Act, and ACL has got the most attention. Shared savings pioneer now, Next Gen. Primary care — comprehensive primary care has now morphed into CPC Plus. And of course, the bundled payment models; all of which held great promise. All of which were good ideas; we had to try them, right? The bad news is, shared savings cost us money. The good news is, Pioneer actually satisfied the conditions to be expanded, but only 12 Pioneers remain. Next Gen sort of learned from that, I think 45 of them. You get the point. It’s really hard to save money, even though it looks like incentives are aligned. And those who are ready to bear risk can all fit in my Prius. So, it’s really true, we don’t have enough people who are quite ready to go full speed. Then let’s look at the biggest innovation that was tried. Biggest in terms of number of docs. The Comprehensive Primary Care Initiative. It’s Christmas, so everything is red and green. Red is bad, green is good. Okay, green means you save money, red means you didn’t. You notice first of all, not a whole lot of red or green. Second, you may notice — I don’t know if you can read that — in year three, that is to say the cumulative over the three years, we did not save money in a comprehensive primary car initiative. Oklahoma did, God bless them. We can explain that why, in a bar, later. But it really did work in Oklahoma. No place else sustained it over time. And no one saved money compared to the PMPM we paid them. This is disappointing. Then you go to the private sector, and I think it’s fair to say, there is a heck of a lot of innovation in the private sector in this patient-centered medical home space in particular. A couple of surveys have been done of recent, sort of meta-analysis, and some of them are disappointing, except for those examples that focused on particurly high cost patients. Multiple chronic conditions, et cetera. And I would say, the bottom line here is that a lot of people are concluding, even if some of these private sector ones — and I think the Care First one around here did save about 2% off trend. Even if they did save that, it’s not enough. We’ve got to move beyond what care coordination win-win kinds of utilization reductions can give us. And so, this is the bundled payment, I will just say basically, we did save money in the bundled payment initiative in model two, which is a combination of post-acute, and acute. By the way, that was Mark Miller’s first major contribution to health economics, showing the post-acute bundle is a good idea. So, we did save money, but we didn’t save money in in-patient setting, only by reducing utilization in post-acute. And by the way, the other bundled payment models didn’t save money.


So, to sum up, here are the big lessons that I think we can learn from where we are, and I would say, at some level, the MACRA bill and all this stuff that’s coming out about what you are going to hear, that’s a result of the reality that we haven’t solved the problem. We do need to work harder on changing incentives. The Physician-Focused Payment Model Technical Advisory Committee, which is surely the longest name of a committee in the history of civilization, on which I and Elizabeth Mitchell sit, was created to make a window into CMMI to try to bring ideas from the private sector — hey, what a concept — so that people can say how they would like to reorganize the way they are paid in incentives. They think it would make sense for their particular sets of patients. And you will hear more about that going on. But anyway, this is where I would think the literature kind of [unintelligible] the egghead stink we have learned from what we’ve done so far. Focus on identifying the right patients. Those patient-centered medical homes that worked, all have to do with picking the right patients and focusing on them. And that may be more important than building quality improvement capacity across the entire patient population or the practice. Per member, per month, payment up front, doesn’t necessarily perform better than other kinds of things. Maybe risk is not as important as we thought, but what does seem to really matter is targeted in-kind assistance. A lot of these practices, you give them PMPM, they don’t know how to spend the money. You can tell them how to spend the money, that’s what CPC did, but it didn’t work. So, maybe what you want is a collaboration between the payer and the provider, and some of that is really creative, going on in Medicare. A lot of it is creative going on in the private sector. Third, don’t give up on total cost of care. My personal disappointment in the way CPC Plus evolved from CPCI, is that CPCI, you may have noticed didn’t save money on that. So, they dropped total cost of care as part of the objective function in CPC Plus. That was a wrong decision in my opinions. It’s true, primary care can’t control all spending, but they darn well can influence what choices are made, and you don’t want to take that off the goal. Second, or finally, I would say the other big lesson is: We’ve got to focus on prices. Great paper written by the late, great [unintelligible] Jerry Anderson a few years ago, It’s the Price That’s Stupid. We pay more than everybody in the world for everything. And we are probably going to have to stop doing that, if we really want to save money.


Finally, I would say, you probably want to think about focusing on social determinants of health. It turns out, some interventions in those areas really can save money. Some interventions in those areas like housing first for the seriously mentally ill, and people with substance use disorder, transportation of non-emergency, as well as some forms of food, actually reduce healthcare cost and criminal justice cost enough to pay for the stuff you are doing. Really cool stuff.


Finally, I will say, our HIT systems, you may have heard this before, are not quite ready for primetime. There is a tremendous amount of frustration in the physician community, a lot of it has to do with the fact that these HIT systems were oversold. They are not the public — the population health tools that they were told they would be. And it’s incredibly frustrating to report this stuff on their own, at night, after they have seen patients all day. We’ve got to figure out how to bill back-office HIT, which exists, which is why Tulsa did what they did, so that we suck the data out and don’t require docs to push it.


I don’t think I’ve depressed you completely, but it looks like I’ve done a lot, so I will stop now, and we’ll move on. Thank you.


SARAH DASH:   Thank you, Len, for setting that context and that overview, and now we will hear from Aucha and Gregory.


AUCHA PRACHANRONARONG:  Thank you. So, we are going to do a quick overview of the MACRA Year 2 Rule, since it just came out about a month ago. As a reminder, MACRA will replace the sustainable growth rate formula for updating physician fee scheduled payments with a program that provides for two participation tracks.


The first track is the merit-based incentive payment system, or MIPS, which consolidates and replaces three legacy programs — the Physician Quality Reporting System, the Medicare Electronic Health Record Incentive Program for Eligible Professionals, and the Value-Based Modifier. The second participation tract is the Advanced Alternative Payments Models track, and when we sought to implement what we call the Quality Payment Program, we sought to improve beneficiary outcomes through patient-centered MIPS and APM policy design and development, enhance the clinician experience to increase the adoption of alternative payment models, or APMs. To promote program understanding and participation, to improve data and information sharing, to deliver IT system capabilities that meet the needs of users, and to ensure operational excellence. We want to support patients and their clinicians in making their own decisions, and about healthcare, using data driven insights, increasing aligned and meaningful quality measures, and innovative technology. The quality payment program is really designed to be flexible, transparent, and is structured to improve over time, and will continue to evolve as we hear more from clinicians, patients, and other stakeholders.


I’m going to start with giving an overview of the MIPS portion of the Year 2 Rule, and then I will turn it over to my colleague, Greg Woods, who will provide the overview of the APM portion of the rule. Just as a quick overview, the Merit-Based Incentive Payment System is based on performance in four different performance categories: Quality, cost, improvement activities, and advancing care information. The points from each performance category are added together to give you the MIPS final score. Then what we do, is we compare the final score to what’s called a MIPS performance threshold. If your final score is equal to the MIPS performance threshold, then the clinician gets no payment adjustment. If the final score is above the MIPS performance threshold, then the clinician or group will get a positive payment adjustment. And if the MIPS final score is below the performance threshold, the clinician or group will get a negative payment adjustment. The payment adjustments are budget neutral and the amount at risk increases over time. So, for example, for the first payment year, there is a maximum of 4% at risk; for the second payment year, there is a maximum of 5% at risk. In the calendar year, 2018, final rule with comment period, we continued to build and improve upon our transition year policies, as well as address elements of MACRA that were not included in the transition year policies, including virtual groups, and beginning in 2019, the ability to be assessed based on facility performance. In this final rule with comment period, we continue the slow ramp-up of the program, by establishing policies that are aimed at encouraging successful participation while reducing burden, reducing the number of the clinicians that are required to participate in the program, and preparing clinicians for full implementation by Year 3, which is what is required by law.


In your packet, you should have a more comprehensive summary of what’s in the Year 2 rule, but a few key highlights that I wanted to share with you today, are: The cost performance category is weighted at 10% in 2018. For the first performance period, as you will recall, cost was weighted at 0%. And then of course, by law, in Year 3, cost has to be weighted at 30%. Second, we increase the MIPS performance threshold to 15 points from 3 points during the transition year. Previously we had stated that for the Advanced and Care information, Performance category, clinicians and groups were required to use the 2015 certified electronic health record technology for Year 2, but in this final rule with comment period, we are giving clinicians the flexibility to use either 2014 addition certified electronic health record technology, or 2015 certified electronic health record technology, or a combination. And of course, those that are fully using the 2015 edition technology would get bonus points under the advancing care information performance category. We are also awarding up to five bonus points for clinicians that afford the treatment of complex patients, and we’ve also added policies to reweight the quality advancing care information improvement activities based on natural disasters that may prevent or hinder clinicians’ ability to submit data. We are awarding five bonus points to the MIPS final score for small practices, which are defined as practices with 15 or fewer clinicians. And then practices that have 10 or fewer clinicians, or solo practices, can join together to form what is called a virtual group, and participate together as a virtual group. And then lastly, we are decreasing the number of clinicians that are required to participate in MIPS by increasing the low volume threshold to exclude clinicians or groups with less than or equal to $90,000 in Part B allowed charges, or less than or equal to 200 Medicare Part B beneficiaries.


And then although this was finalized in the transition year rule, I just wanted to highlight that the minimum performance period for a cost and quality for Year 2, is 12 months, compared to 90 days in Year 1, there is no change in the minimum performance category between Year 2 and Year 1 for the improvement activities and the advancing care information performance category.


Lastly, just a quick slide — this depicts a quick overview of the timeline for Year 2 clinicians, it should be collecting and capturing data between January 1st, 2018 and December 31st 2018. And then the submission of data to CMS needs to occur by March 31st, 2019. Then CMS will take the data, analyze it, calculate scores, and calculate payment adjustments, and apply payment adjustments beginning January 1st, 2020. So, I will turn it over to Greg now.


GREGORY WOODS:  And according to the clock in front of me, I have seconds to tell you everything about APMs and MACRA, so I think this should go well.


On a high level, what are we talking about, when we talk about alternative payment models, or APMs? What we mean is innovative approaches to paying for healthcare that [inaudible] quality and value. There is a specific definition in MACRA, which defines APMs as anything tested by the CMS Innovation Center, where I work. Also, the Medicare Shared Savings Program, which was a separate program set up by the Affordable Care Act. And certain other demonstrations required by federal law. What this means in practice, is that includes things like ACO’s, bundled payments, medical home models, the kinds of things that Len discussed. Also, it includes everything else that the Innovation Center does, so that includes things like models focused on the social determinants of health, models focused on health plans in Medicare Part C or D. Models focused on the dual-eligible populations. So, there is a broader universe of what we talk about when we talk about APMs. In the context of MACRA, most frequently when we talk about APMs, what we are talking about is what we call in the rule, advanced APMs. And this is a specific subset of all of the alternative payment models that are out there, that meet certain criteria. So, specifically, they must require participants to use certified VHR technology. They must base payment on performance on quality measures comparable to those used in MIPS. And lastly, and this is the one that is the tough one:  It must either be a medical home model expanded under Innovation Center authority, and at the moment, that’s a null set. Or, require participants to bear more than a nominal amount of financial risk. So, that’s what we mean when we say, an advanced APM. In 2017, which is the first year of MACRA, we have a list of advanced APMs, they include the downside risk-bearing tracks of the Medicare Shared Savings program, or Next Generation ACL model. It includes our comprehensive drug replacement episode based payment model. It includes our ESRD total cost of care model. It also includes our Comprehensive Primary Care Plus initiative.


What does it mean to be part of an Advanced APM? Four eligible clinicians — so, this is physicians and other clinicians covered by MACRA who are part of an advanced APM. If they are participating in that advance APM to a sufficient degree, and there are certain thresholds that are expressed in terms of payments or patients that are set forth in the statute. They get a couple of benefits. One, they are exempted from MIPS, so they don’t have to worry about reporting all of the things that Aucha just talked about. There is no payment adjustment there. Two, at least for the first several years under MACRA, they are eligible for a five percent bonus payment. And this, I should make clear, is separate from whatever incentives are embedded in the models in which they participate. So, for instance, if you are an eligible clinician, you are participating in an ACO. You are eligible for this five percent incentive payment. You also would be eligible to receive shared savings payments, should your ACO qualify. So, it’s a separate incentive layer on top.


In this year’s rule that we just finalized, let’s just briefly, we largely left our definition of advanced APMs unchanged. I think the most significant policy, which is described here, is we extended the definition of nominal risk for an additional — so what constitutes downside financial risk, which as I said, was the third criteria to be an advanced APM. We extended that for an additional couple of years. In particularly, that will allow the Medicare shared savings program Track One Plus, which is beginning in 2018, to continue to qualify as an advanced APM.


Okay, so that’s advanced APMs. Then there are also what we call MIP APMs, or the APM squaring standard. Briefly, what this means is there are certain APMs that for whatever reason may not meet the criteria to be an advanced APM, or there may be clinicians in advanced APMs who don’t meet the criteria to become QP’s, which is to say, to receive the five percent incentive payment and be exempt from MIPs. However — and I will say, the most prominent example of this, is the Medicare Shared Savings program, Track One, which is a one-sided ACO model. So, there is no downside risk. It’s not an advanced APM, however, clinicians who are participating in that model, are required to report significant quality data, and also are generally required to do certain improvement activities. So, the theory here is:  It would be burdensome to require clinicians to report quality twice. Once as part of their APM, and another time for MIPS. And so, what we are doing here is saying, we can take the qualities out of your report for your ACO, or for your APM, and we are going to use that to give you a MIPS score. We are going to exempt you from cost, and we are going to automatically give you improvement activity points based on your participation in APM.


Again, in this year’s rule, it’s largely continuity from what we did in previous years, although there is a slight change here where some of our MIPS APMs or APMs under the scoring standard, for the first year, because of some technical challenges, we exempted them from being scored on quality for Year 2; for 2018, we expect all APMs under the APMs scoring standard to receive 50% for quality under MIPs APMs. And then very briefly, something that we focused on, and this year’s rule, is the all-payer combination option, and this is something that was built into the MACRA statute, as I said, in order to become a QP, in general, you need to have Medicare participation in advanced APM beyond a certain payment or patient threshold. Starting in Year 3 of the program, so starting in performance Year 2019, there is an additional pathway that is laid out in the statute for becoming a QP, for being exempted from MIPS, and for earning that bonus payment. And what that says is, if you are in a Medicare Advanced APM, you are participating to a significant extent, but not sufficiently to actually become a QP, then you have the option to request that we also look at all of your other payers. So, your participation in advanced APMs that might be offered by Medicaid, by commercial payers, by Medicare Advantage. And we will look at that, and if you hit relevant thresholds based on the combination of participation in a Medicare Advanced APM, and an Advanced APM to cross hair other payers, then that is another alternative for you to be a QP, earn the five percent bonus, be exempted from MIPS. There is a lot of detail in this year’s rule, which I certainly won’t try and summarize here, about the [unintelligible] combination option. It’s a slightly new thing for CMS, since it requires us to look at the details of payment arrangements that we don’t actually have ownership of, that are offered by a Medicaid or private payers. So, there is a lot of detail about how we will payers and clinicians to request that information, to analyze it, and to give clinicians credit where they meet the requirements.


So, that I think concludes our presentation. This slide gives some detail about where you can find more information about MACRA, because there is a lot more information. I will stop there.


SARAH DASH:  Thank you. And we will move to Elizabeth Mitchell.


ELIZABETH MITCHELL:  Thank you very much. I am going to go really quickly through some contact slides, because Len told me this morning, everybody knows all that. And his slides were cooler.


But never the less, I think it is very important to sort of remember why we are here, because we do have a problem. We have got to deal with healthcare costs. And since most people don’t think in GDP, at the community level, this is where they would rather be spending their money. They money is coming from housing, transportation, food, clothing — you might be thinking about tax cuts here, they are thinking about their household needs. And if you are an employer, you are actually paying for healthcare. You have been seeing all of these investments and payment changes and new payment models, and quality improvement programs, and the value is actually getting lower. So, it is increasingly urgent when you get out there in the world. So, MACRA is supposed to change the trajectory for all of this. I think payment reform, and care delivery reform, are sort of the most palatable approaches to actually reducing healthcare cost and improving its value. And as you heard, I sit on the guiding committee of the Healthcare Payment Learning and Action Network, and they have come up with this neat framework for how we will get from here to there, starting with fee-for-service, where we are now, in category one, moving up to population based payment, category four. So, if we migrate in that direction, we are anticipating sort of more accountability, more efficiencies, and just better value for spending. And they have these nifty bubbles where we will go from everybody in fee-for-service, up to bundles and population payments in a few years. Anyone who has tried to do this will tell you, as nice as that looks on a slide, we do not have the infrastructure out there to make it happen. We entirely new measures of quality and cost — patient experience, patient reported outcomes, total cost of care. We need a shared data infrastructure. If we are not sharing data, this will not work. And we appreciate the push for health IT from CMS, but using certified EHR technology is not equivalent of interoperability or data sharing. So, we’ve got to figure that out. We need new incentives. We’ve got to have transparency. As David Lansky would say, we are about to shift a trillion dollars around and we can’t even define value. So, how are we defining healthcare quality and cost, and value? We’ve got to have an alignment across payers. If you are sitting in a practice, you get a report from Aetna, you get a report from CMS, you get a report from Cigna, they cancel each other out. It is noise. We’ve got to have common payment, common measurement, across payers. We need new care models, and we’ve got to have new relationships and community partners. So, really new governance for this type of work. So, that’s a heavy lift on the ground, when you are trying to adapt to what we — I think everyone agrees — is the right direction.


I represent National Network of Regional Health Improvement Collaboratives. These are multi-stakeholder, non-profit entities working in regions. Mostly statewide. To improve the quality and affordability of healthcare. One of our local heroes, Curt Bramer, is here in the back, he runs our member in Cincinnati, or as he calls it, America. And they are doing this work in real life. Sharing data across providers, across communities, they are doing the improvement work, they are a CPC site, they’ve aligned plans and incentives, and what I’m here to tell you is that, if you can suspend disbelief, it actually is working in communities. Our members are embedded in their regions, they have close, trusted relationships with their local stakeholders, and so, I was tasked with bringing back some feedback from some of the practitioners in the field. What are they experiencing out there when they are thinking about implementing MIPS and APMs? This is obviously an incredibly brief snapshot, but the key barriers they continue to raise are access to actual quality data; you can’t manage a population if you don’t have information on that population. EMR vendor support — they are not getting it. They are not getting information on measurement, or what’s happening when their patients are getting treatment in other sites, or other systems. And there is a real sense of frustration. These are some feedback from our folks who are working directly with practices. They just wish this would all go away, some of them. It’s bookkeeping, it isn’t contributing to care. However, the good news is there are things that are working. There is a receptivity and a desire for this technical support at the local level, with actionable data. When it’s local and trusted, they are using this information, they are making some of these needed changes. And there is a readiness to do this. So, there is opportunity here if we do it right.


So, to boil it down, what we understand that folks need to actually be successful under MACRA, goes in these three buckets: They need data and information, they need alignment across payers, both with incentives and measures, and they need technical assistance and support to get there. Now, the good news is, as Len says, we have spent a lot of quality time together on PTAC. We’ve had about 20 proposals form the field for new payment models since receiving them literally a year ago today. It’s our anniversary of receiving models. And we stepped back as a committee and said, what have we heard over the last year? And we sent a set of recommendations to the Secretary based on that experience, what we were hearing from physicians in the field who were trying to change care and payment. They need individualized technical assistance and model design. Most physicians understand clinical change, or practice change, but they aren’t trained to do payment model designs. And then to implement those payment model designs. So, there is a need for that support. They have got to have data and analytics. How do they measure their patient population? How do they identify them? How do they get timely information to change something if it isn’t working? So, that access to meaningful data and information and analytics, is actually essential to success. They need data sharing across platforms. Again, not just the technology, but actually data sharing and interoperability. We’ve had this lovely submitter come in, he had worked on a proposal for about a year, and in his spare time he came up with this really great idea, but it required coordination across sites. And we said, well, how would you do that? He said, well, everyone would just have to buy Epic, that’s the only way we know how to do this. And that really was not the aim of what we were trying to do. It’s really about getting these systems to work together effectively so that they can identify what’s happening in different care settings.


And then we need to do limited scale testing. We have to have an opportunity to try these in different markets and understand, what are some of the unintended consequences of testing these models? So, before we go to scale, it would be like, releasing self-driving cars without testing them somewhere. How do we make sure that we learn what works and what needs adjusting before we go to scale? So, there is a readiness out there, we are having a lot of interest from specialist groups, from other physicians who want to change care models, and have that supported by payment. But we need a few changes.


So, very briefly, I wanted to bring just a few success stories from the field, where they are addressing affordability through community approaches and I wanted to highlight one project that we have been leading, which is to test total cost of care and resource use to report it in a standardized way across seven states, across all commercial payers. Think of it as the Dartmouth Atlas for commercial payers. It was 2017, and this hadn’t been done before, which is really surprising to me. But several of our members combined this data, produced total cost and resource information across their payers, and they were developing and sharing performance feedback to providers. The demand for this information is off the charts. The providers need this if they are going to be successful under these new payment models. They understand how they fare compared to their peers in terms of utilization, ER use, other things like that, where they can make the improvement. And we did it not just in Oregon; we did it in Maine, so it can be done in multiple markets. Another success story; lest you think they are all in blue states, in Louisiana, they use their HIE, and their relationships with State Medicaid program and hospitals, to actually give notifications when people were admitted to the Emergency Room unnecessarily. And they saw a reduction in over 20% of ED visits for some of their members. This is inappropriate utilization, or care that could have been provided in a different setting. This is better care at lower cost through using information and community stakeholder engagement. So, it can be done, it is possible, and it achieves the aims that I think we’re all after.


And then finally, we heard that not all of the CPC sites were as successful as we had hoped, but our members in Cincinnati and in Oklahoma saw significant improvements. They saw reduced utilization of the ED, in-patient days, they saw improvements in quality, and this was done, again, through that shared data, through those close, trusted relationships for improvement, and they saw savings for Medicare based on better care for patients. So, the good news is that this can be done, and we can capture these in these multi-payer models, and change payment across Medicare and the commercial sector, we think there is significant hope for actually achieving the aims of MACRA. I will stop right there, and give it to Mark.


MARK MILLER:  One thing I will say about Len’s presentation, is just to emphasize, we should at some point be talking very much about prices. We aren’t doing it today, but it’s incredibly important. And if you ever want me to come back and rant on that point, I will be happy to do it.


But first of all, I would like to thank you for asking MedPAC here to talk. I was asked to comment on a couple of things:  Our views on APMs, and our views on MIPS, and we have a proposal around on MIPS that is causing a lot of consternation, and I’ll talk you through that. I’m going to try to do this all very quickly, and try and stay on time here.


Generally, the commission supported the direction of the MACRA legislation, getting away from the SGR, and generally setting mechanisms in place to try and move people from fragmented fee-for-service, to more organized alternative payment models. I’m going to be saying APMs, even when I mean AAPMs, because I’m not going to say that a million times. But the support of the move from fee-for-service to kind of alternative payment models. With respect to APMs, the commission has given a lot of guidance to both the Congress and to CMS, and it comes from the following set of principles. Quickly, to go through them are:  The commission would reward successful APMs, not just give a bonus for being an APM, but give a bonus if the APM is actually successful. We think the APM should be risk-based. We think you should tend towards judging on the basis of the total patient experience in terms of cost and quality. So, we tend to bias towards looking across providers, and looking across time when we measure performance. We think that there should be — they should be large enough that you can detect population-based cost and quality measures. By that, I mean, things like avoidable hospitalizations, avoidable Emergency Room — I think that some of that was just mentioned. Survey-based beneficiary experience, and spending, per capita. That type of thing. Now, large doesn’t mean you have to be large. That’s one way you can do it. A different way you can do it, is you can take individual entities and say, well, are you willing to be measured as an aggregate or alternative, take a small entity and use multiple years of data so that you have enough information to measure change. But the idea is to get to population based measures; and I’m going to say a little bit more about that. Keep that in your mind, if you will. We think that the beneficiaries should also get some savings out of this. Large dollar transfers between providers and the government, maybe the beneficiaries should get a benefit. I don’t have time to talk about it in eight minutes, but if anybody wants to ask what that means, I will be happy to tell them.


Then two other things to keep in mind:  We think that the legislated revenue threshold should be removed and you should pay the bonus on any dollar that travels through the APM, and I can talk more about that. And then also, we have some ideas for how you get small practices to participate in an APM such as asymmetric risk corridor, so that you have more upside than downside. And again, if anybody is interested in that, I will take that on question afterward.


I think they want me to focus on MIPS because we’ve made this proposal and everybody is upset about it, and so now that’s where we are going to go to next. I would say, “predictably” upset about it. So, before I get into it, I want to tell you the principles of where we are coming from. For many years now, the Commission has been saying that the quality measurement process is overbuilt, it’s burdensome, it’s collecting lots of measures that don’t show a relationship to quality outcomes. Consistent with that, we think measuring at the individual physician, individual measure level, is ultimately going to be an unworkable process, and has been an unworkable process for the last several years. We also think that fee-for-service measurements shouldn’t drive APM measurement, it should be the reverse. APMs should be able to operate in the population based measure space, and that is what you want to happen in fee-for-service, not the reverse. Although there is a role for more individual measurement, but I will say something about that towards the end. We think that this process has resulted in a very burdensome process for physicians. It’s built off of, in addition to other things, it’s built off three old fee-for-service systems. For those of you who know these, PQRS, the Value Modifier, and the HER meaningful use. These systems have generally been confusing. Imposed data collection, burdens on practices, have not led to higher quality, which I think Elizabeth was just making that point. And here is a couple of annoying facts: We’ve been collecting this information for something like eight years, and none of it is up on Physician Compare. And CMS’s regulatory burden estimates are about that this is a billion dollars for practices, whether you are talking about the old system or the new systems here. And I want to be really clear on some of the things that I’m about to say: I think CMS is acting in good faith, and doing the best that they can to implement this, but if you think about what just went through on the MIPS side, CMS is making exceptions. They are pulling people out of the MIPS measurement process, and allowing people not to have to report. They are creating multiple ways to report in order to be more flexible, which is a good thing on the one hand, but on the other hand, it creates more confusion about how somebody reports. And the industry continues — the physician groups continue to call for more delays. There is two years that we’ve been involved in delays, and they are asking for three years more in delays. We think the measurement is flawed. As mentioned, there is a focus on the process and attestation measures that are often tapped out, as I’ve said. I don’t think they have strong relationships with outcomes. Physicians pick their own quality measures, so we are allocating billions of dollars on the basis of people using different measures and choosing which measures that they use. Generally, all of the rules allow maximization of measurement. So, you pick topped up measures, you pick your own measures, you pick the highest values on those measures, and that results in two types of equities. And again, as CMS just said, using the regulatory flexibility, they’ve set the benchmark really low — three out of a hundred. So, right now, lots of people exceed it, but there is no reward because very few people are under it, and that’s where you’re supposed to get the dollars to give the reward. Arguably, if you are above it, you are not really getting a reward. In a couple of years, by law, it has to go to the average, and there is going to be so much compression in those scores, that at the top end, that small changes in scores will produce big rewards and penalties, and that will be a different type of equity issue that will arise. And the physician groups at this point are asking the law not to go into place.


Anyway, I think there are a couple of different kinds of equity issues there. If any of this confused you, I will summarize it for you with this: The AMA has a website, and the point of this website is, one patient, one measure, no penalty. So, if you report one measure for one patient, there is no MIPS penalty. Basically, that’s the state-of-the-art in MIPS right now. We have a different idea here, and this the idea: We would stop collecting data at the physician level. We would begin to have CMS compute population based measures such as avoidable hospitalization, avoidable ED, patient experience using surveys, that type of thing. We would take a two percent withhold, which is not dissimilar to the way the law works. We would use that pool of data to reward and penalize physicians who stay in fee-for-service. And we would give providers three choices: Like under current law, if you go into an APM, get your withhold back, get your five percent bonus, and if the APM is successful, you get shared savings. So, we would do that. The second choice, which is also contemplated by laws, and this is the idea that is apparently upsetting everybody, is you say to the physician:  You can voluntarily create a virtual group. You don’t have to make a delivery system. You have a group of physicians that you are willing to be measured at the population level. So, that’s the second idea. The third idea is:  You don’t do anything. You take the two percent withhold, you lose it, and you walk away because you don’t want to participate, and that money goes to physicians that are willing to participate.


In sum, we think this reduces burden, we think it creates greater comparability and equity. We think it moves population more towards population based measures instead of tethering it to the old fee-for-service systems, and then the thing I want to say is:  There may still be a role, and I think there is still a role for more granular data collection and information for physicians, not inconsistent with what Elizabeth is saying. But, maybe that is not where the government needs to be. Maybe the government says: We’ll move dollars around on large population based directions, and organizations like specialty societies and Elizabeth and people like that work with the providers at the more micro level to look at data that helps meet the larger goals. With that, I will stop.


SARAH DASH:  Thank you all for your presentations; I’m looking forward to a great discussion. We have about a half an hour for discussion and question and answer, so I hope you will all stick around.


I’m going to kick it off with a couple of questions, but as you are thinking through this, you can write a question on a green card that should be on your table. Or, you can go to one of the mics on either side of the room to ask a question of our panelists. I want to ask this question, because there has been a lot of different points that have been raised, and there have been some tensions raised with regard to individual level measurement versus population measurement. Measuring quality versus measuring actual outcomes. The pace of change and for whom this notion of flexibility versus simplicity. And I wonder if the panel could comment on maybe any or all of those, and where do you see — is there a Goldilocks sort of middle ground here? Or is it really that you need sort of different approaches for different kinds of providers, different circumstances of providers, and where are we along that trajectory?


ELIZABETH MITCHELL:  Just one comment is:  I think while the pace of change may feel fast for some, I would say the pace of change for employers and purchasers and patients and families and communities, feels incredibly slow. We heard about the quality chasm issues, what? Fifteen, twenty years ago now? We haven’t seen measurable improvements. Identified 30% waste, decades ago. We haven’t seen measurable improvement. So, I think the perspective on pace needs to just be reflective of all the stakeholders.


Also, I really want to underscore, I think, Mark’s proposal about measuring — measurement has different purposes. So, it can be as granular as you want for improvement at the physician level, but for accountability, where I think we are talking about, it’s the population performance. Total cost of care. Patient recorded outcomes. Are people functioning happily in their lives? So, I think there is just a different level of measurement at this for payment and accountability than there is for improvement, and I think they can co-exist.


SARAH DASH:  Thanks.  If I can ask a quick follow-up question with that. I mean, you mentioned very briefly in your presentation that for the employers, particularly value, has been decreasing. I wonder if you could just say a couple more words on that, and what is it with regard to particular measures or is it health outcomes? If you could just explain that a little more; that would be great.


ELIZABETH MITCHELL:  Sure. That was sort of an aggregate statement, but frankly it really goes to the conversation about pricing, which Mark promises we can have. Because they are spending more, but they are not seeing corresponding improvements in quality. So, quality is essentially flat more or less, but the spending is increasing, and I think we need to start talking about pricing if we really want to get into that.


MARK MILLER:  Again, as somebody else who has a gray beard — maybe not as gray as Len’s but, Len has more hair than me. So, I feel like as old as I am, I feel like this tension in the physician world has played out time and time again. It’s like, the physician community demands — and you can understand it — you have to measure what I do, you know, I’m a left-handed surgeon, and I only do these types of things, and it’s not fair if you don’t measure that. And there is some real truth to that. I’m not being dismissive about that. But then it can’t be burdensome, and it has to be accurate. That’s where I think it begins to fail. And then if you go to the aggregate, the physician community’s counter argument, which is not illegitimate, is: But it isn’t me, those are other people, and you can’t hold me responsible. So, there is sort of an impossible policy environment null set that comes out of that. One thing I would ask you to think about, because I know our idea causes consternation among different groups of people, is ask yourself, and Elizabeth might have said this, although I don’t want to attribute it if she didn’t, is where should the government be, and where should other actors be in terms of collecting and measuring data? I think one thing that we’re trying to say, is maybe at the more population levels, the role of government set a general direction, and also, I think those measures, a small set of population measures, also allow you to coordinate with Medicaid and commercial payers much more easily if you can come to an agreement on a smaller set of measures.


SARAH DASH:  I’m going to get to your question in just a second, but I want to ask this question on a card that came in, which was that MACRA created pathways for adopting new payment models, but not really for narrowing options with demonstrated effectiveness. So, the question has to do with, in the long term — is there a long-term equilibrium? Are we going to have dozens of payment systems co-existing? Or is there going to be — do you envision a consolidated number of fewer best practices once we get a chance to test things out and see what works? And how do we get there?


LEN NICHOLS:   I think it’s fair to say that the people who are most excited about the window that MACRA created for this physician-focused payment model thing, were the specialists, because as Mark said, they felt left out of the patient-centered — the traditional APMs and even Advanced APMs that CMMI had worked out. And so, they wanted their own little way to get to the promised land, which was the five percent bump up in the year to payment. And so, they felt like this was an opportunity. In fact, I think as Mark said, it is sincere in the sense that their patient populations are specific, they do focus on particular conditions, and there is a pretty good argument to be made that a lot of these models that have come to us — I will just speak for myself — are fairly creative in the spirit of what the bundle population based stuff is all headed toward, but specific to these clinical populations that these specialists deal with. So, I don’t think there is anything wrong with kind of letting — I wouldn’t say a thousand flowers bloom, but it would be okay to have 13 out there. But it’s interesting, when we recommend something to the Secretary, essentially it then goes to my colleagues here on the panel, and they get to decide essentially — or somebody over there gets to decide what actually happens. I would say that the Gestalt is we don’t want to do a payment model for every physician group in the United States. That would clearly be infeasible, and arguably unwise. So, they are taking what comes, if you will, over our transom, that we try to screen for quality of idea, then give it to them, and they sort of look at it, and I think the basic approach is, okay, we’d like to write an RFP for this type of model, and get more people involved, and maybe set some parameters, and maybe have a little bit of flexibility here, there and yon. But in this general direction, to expand the scope of conditions, of patients, or providers that are currently able to access this alternative payment world. The problem with what I just said, if you followed half of it, is it takes a long time. So, they’ve got a hard job to do their existing day job, and then take these new ideas as they come in over the transom and try to fit ‘em into a framework that actually makes sense. I think that’s where we all are kinda struggling — how do we speed this up with overburdening? It’s a moving target, and you know, I would just say, bear with us all, because we are all kind of learning what is the right balance.


GREGORY WOODS:  I would just add — and I think in this question and the last question, the correct access of tension has been identified, which is between sort of flexibility and personalization and I don’t have any blanket statement to make, but you could sort of — on the spectrum between the one size fits all model, and the model for every kind of specialty, probably the answer lies somewhere in the middle. The one thing I would want to note from the perspective of the Innovation Center, is our portfolio of models is not static and it’s not intended to be static. We are a research organization and we are testing new models, and I described some of the results of them earlier. I would probably put a somewhat more optimistic gloss on that, but I would say it’s fair to say that we are not necessarily going to bat a thousand; as we test these models, we are going to learn and modify. Our statute is constructed to allow us to expand models that work, to modify models, to terminate models. I do think it’s appropriate in the — or, I think it’s reasonable within the context of the MACRA statute that eligible clinicians who raise their hands and say that they are willing to participate in those experiments, and recognizing that it’s not going to be a static landscape, can receive credit for that, even if the version of payment reform that they are operating under at a given moment, is maybe not where we ultimately will land.


SARAH DASH:  Let me, if I could, just ask a quick follow-up question, and follow the line of the discussion, because somebody on a card asked whether CMMI is the only pathway for new advanced APM model? So, perhaps it’s a little bit more of a process question, but can you clarify? You know, CMMI’s role versus perhaps other aspects of CMS?


GREGORY WOODS:  So, as I described in my presentation, there is a specific definition within MACRA of what an alternative payment model is, that includes all of the models tested by CMMI. It also includes the Medicare Shared Savings Program, and some other demonstrations authorized by other statutes. I guess, what I would say in general is that under existing law, CMMI is the primary pathway for CMS to introduce new alternative payment models.


SARAH DASH:   Thank you. Alright, you’ve been very patient. Thanks.


AUDIENCE MEMBER:  I’m Carl Polzer, I’m a health policy analyst and consultant, and I’m currently working with both provider and consumer groups. Len identified the ACA payment cuts as a major event in Medicare payment history, and a probable cause of contributing to the leveling off of relative cost growth. I think you are referring to the one percentage point cut in the update that nursing homes and hospitals get?






LEN NICHOLS:   It’s aggregate GDP minus two.




LEN NICHOLS:  So, it’s big.


AUDIENCE MEMBER: It was one, I think, at the beginning. So, I’m not up-to-date. Since this is a huge natural experiment, what do we know about it? Do we know how these institutions adapted? Did they really do productivity cuts? Did they do other things? How much can they do in the future? How did it impact consumers?


LEN NICHOLS:   I don’t know about consumers. Maybe Mark does. But I think it’s fair to say, we know it was the single biggest financing component of the ACA. I can guarantee you that. I think it’s also roughly true that approximately four thousand hospitals did the math and hired approximately a million consultants to spend approximately two million dollars on trying to fix it. And what it meant was though, seriously, they had to become at least as much more efficient each year, as the economy on average, in order to break even. And that did push them to investigate strategies that could achieve that. It’s also true, a lot of them still lost money. I mean, you probably know, Carl, when the ACA passed — thanks to Mark’s data, we know this — roughly 75% of hospitals had a negative Medicare margin. Today, it’s 85. I give you the impact of the market basket update reduction. Even though we’ve expanded a whole lot of other stuff, right? So, I think it definitely reduced the trajectory of revenue flowing to hospitals. I think it definitely instituted an imperative to become more efficient. Another strategy you may have noticed, they adapted, was to consolidate and to make up for those negative Medicare margins with positive commercial margins, which is how they end up with a positive net margin overall. I will stop and let Mark kick in.


MARK MILLER:  Yeah, so a couple of things. At least in terms — we check access every year. Hospital, all of the post-acute care sectors, physicians, all of that stuff. And there has been no change in access, and in some places service utilization has slowed down. But that slow down was actually around 2008, and was pretty much economy wide. It wasn’t just Medicare. There are a couple of things I would say about the pain level, because you know, the industry loves to point to that and say, you know, all of the wheels in the republic are going to fall. The wheels are going to come off and the republic are going to fall. Let’s be clear in a post-acute care sector, even with those changes, the average post-acute care margin in Medicare, I don’t know, 12-14% in that neighborhood. So, we’ve been overpaying on the post-acute care side for a century, and apparently, we still are. On the hospital side, you guys are both correct. The margins were negative; they were largely negative before that, and some of the issue on the hospital side — and we’ve made this argument, and I will make it very briefly, but this is the rant that I was going to do on prices, so you are getting it free of charge. But I will keep it a short rant, which is this:  There has been huge amounts of consolidation, even before the ACA. There were huge amounts of consolidation. Consolidation means that you drive prices in the private sector, in an industry where cost to drive prices, you are basically jacking up cost. So, Medicare has always looked like a poor payer, because it’s chasing higher costs that are driven by the consolidation and higher prices. And what we’ve said in MedPAC, is despite the fact that these margins are negative, we still need to be cautious and give stingy updates, because otherwise we are just going to fuel that cost growth as well. That’s a short version of that.


SARAH DASH:  I want to follow-up quickly on the consolidation question, but also want to encourage folks, don’t be shy to go ahead and wave if you have a question on a card. I think in the reading materials, there was a point made about — back to the individual physician level, and some being perhaps more ready than others. The small practice issue. Is this driving consolidation? Is — I don’t want to particularly point a finger at MACRA or MIPS, but the general drive towards quality measurement and delivery system reform, driving consolidation. Or not.


ELIZABETH MITCHELL:  It is. And the need for data analytic systems, for managing complex models, there are plenty of big systems who have the resources to do that. But the smaller practices that don’t, are being acquired. And that is contributing to all sorts of trends, particularly the pricing.


LEN NICHOLS:   And even if they aren’t being acquired, they are joining into affiliations, which may make a lot of sense. Some for ACOs, both commercial and Medicare. For reporting purposes, because I have to tell you, these little, small practices, they bought EHRs that are not fully functional. And I’m sorry, but it’s just true.  A lot of those practices cannot generate data in the way they need to, or without help. And the big systems can do it, and they have an advantage if they go into a big system and report as a group. A, from the common scale of reporting, but B, then they are grouped with the group, and they have a better shot at coming out ahead. So, fundamentally, it’s in their interest. I also would observe — and again, I’m working with 200 small practices in Virginia, God bless them. This is smaller than 10 clinicians. A lot of these guys are just exhausted. They are working their tails off, they are making $100,000 maybe. And they’re just pushing a square boulder up a hill. And they sort of give up, and they hate it. They do not want to be employees, but on the other hand, it’s so much more financially safe, and it makes it easier on a day-to-day basis. And you’ve probably seen the burn-out statistics, they are stunning at the primary care level, and part of this is driven by this quality reporting [unintelligible] we’ve created.


MARK MILLER:  I believe all of that. I believe it’s a factor. And I know you guys know, there is a tremendous amount of consolidation occurring before the ACA, before MACRA, and money drives that. You can make a lot more out of your commercial payment if you are consolidating. Then things that Medicare and private payers were doing, but Medicare was doing, is you buy a practice on Friday. On Monday, you don’t bill it through the physician fee schedule, you bill it through the out-patient fee schedule, and you increase your payments by 100%. That doesn’t have anything to do with quality, that has to do with money. I think the quality reporting matters, it’s yet another factor, but there were a lot of reasons to do it, even before that. That is said with complete respect to what you guys just said.


AUDIENCE MEMBER:  Hi, I’m Shannon Bromley, I’m Senior Vice President of the Laon Institute, which is a small think-tank based on Boston. I want to reiterate something that Elizabeth said first, and Len echoed, which is that primary care in particular is drowning in the documentation. And even if they have great electronic medical records, they are still drowning in documentation. To connect that to something that Mark talked about, about the disconnect between quality metrics and outcomes — does MedPAC have any recommendations on how to make those quality metrics actually be metrics that matter?


MARK MILLER: Can you give me another pass at the question?


AUDIENCE MEMBER:   Yes, we know that a lot of the quality metrics — this is one of the biggest complaints that primary care physicians have, is that the quality metrics that they are supposed to hit, often are not evidence based, and certainly get in the way of really good and personalized quality patient care, and they are all based on under treatment as far as I know. I’m really interested in the problem of over treatment. In MedPACs recommendations for how to change some of these MACRA approaches, is there any recommendation about how to get to the metrics that really matter? Reduce the number of metrics and get to the ones that really matter?


MARK MILLER: Well, all we’ve done — and I think this is the best answer I can give you, if I follow your question. All we’ve done is gone through and said, here are the kinds of measures we are talking about — which I rattled off a few times in my presentation, I won’t bore you again with those. In saying, you know, you move two things like this, and actually in my slide presentation, but also in our report, the last slide is a list of the kinds of measures that we are talking about. So, that’s my best answer.


ELIZABETH MITCHELL:  I would like to pile on a little bit, and cite a GAO report that came out last year. It’s found that five percent of measures used by commercial plans were common. That means that 95% are not common. And that might be 60 different diabetes measures that are slightly tweaked. But it makes it incredibly challenging for a practice, and they estimated that it’s $40,000 per position, per year, that they are spending on quality measurement, and about 78 hours per physician, per year. So, this is an area that is ripe for improvement. So, we have members in California, Integrated Healthcare Association, has a common measure set — 34 measures. They might not be the perfect measures, but they are common measures that really evaluate cost and quality. And Blue Shield of California just agreed to adopt those measures. That will reduce the burden for those providers, and that will actually move towards common signals. Other health plans can do that. There’s been a lot of conversation about the need for measure alignment. It is a business decision, and I think it is a big opportunity to reduce cost and burden.


LEN NICHOLS:  So, I would just add a corollary, and that is that it’s not just about the measures per se, although I take all of the points that have been made, quite well, including your original one, Shannon. It’s also about how they are reported, and how the data are gathered, and what Tulsa, Oklahoma does, and what Cincinnati does, is they build HIEs that essentially reach into the EHRs and pull the data out. So, the doc doesn’t have to spend his nights producing the report. That’s the key fundamental transformation that is going to make this work or not. Those places, what they do, is they pull out the CCD, they pull out the raw relational database, and they generate their own measures that they think makes sense for them and their multi payer context. That’s what you want. Actionable data. Because as you know quite well, hemoglobin AIC should be different depending on the co-morbidity, but right now, the meaningful use metric has a single threshold either above or below, right? So, that’s technical work for this economics call “stupid”. We should have a more nuanced version of that. And you can only do that if you have a system that’s supple. But you cannot, and this is the only thing that I hope you remember:  We cannot depend upon doctors to push those data out of a very clunky EHR system, we have to go get the data for them. It’s not hard, it’s can be done, it is being done in these places, we can do it.


AUCHA PRACHANRONARONG:   And then from the CMS perspective, I just wanted to add that over the past several years, we’ve really been pushing to adopt more and more outcome measures and replace the process measures that we have, with more outcome measures. And while we’ve been aggressively trying to develop our own outcome measures, you know, the vast majority of clinician measures are developed by the specialty societies and the clinicians themselves. So, we really need their help also, in getting to more outcome measurements that are relevant.


AUDIENCE MEMBER:  My name is Don [name] and I’m a Research Analysts at [unintelligible] and we are a start-up to help PCPs organize into ACOs. I just want to hear the panel’s thoughts on risk adjustment, and whether that’s sort of the formula for mitigating these complaints about population level payment reform systems. They don’t apply to individual level practices, because they think that they are different. So, if we can get risk adjustment correct, I think we can mitigate that. I just wanted to hear the panel’s thoughts.


MARK MILLER: I think risk adjustment can always be improved, but I also think it’s important to keep in mind, we are moving billions of dollars right now on a risk adjusted basis, and people are happily taking the checks. So, I think there is improvement always that needs to be in place, but I also think that there is a fair amount of risk adjustment that is already within reach. Now, places where risk adjustment, I think, matters, and needs to be improved specifically, just to say that there are improvements — that’s you, right? I’m trying to find the questioner. We’ve done a bunch of work on SES, and think that it is an important variable. Our position is that you don’t take the measure, the SES factor, and build it right into the measure, because then in a sense, you are sort of hiding the actual effect. But when you deal with payment, you make payments and adjust on the basis of SES for that purpose. So, you might say to a hospital:  Your readmission rate is high, and part of that is because you have a more difficult population. We are going to adjust the money, not only what we pay, but if there is a penalty or a reward, that type of thing, to recognize that. But keep the measure out there so that the hospital continues to work on improving its performance for poor people too. You know, you don’t want to forget the poor people. Then the last part of our idea is we would put people into categories. I have a lot of poor people — I don’t have a lot of poor people — and kind of have those groups work against — not against — but relative to each other, so that even if you are poor, a lot of poor people in your hospital, but you are doing well, other people in that category should be trying to keep up with you, so that you don’t lose the incentive to improve for poor people. But we do think the money needs to get adjusted to recognize it.


SARAH DASH:  Okay, so we have less than ten minutes left, so I want to get to a couple more really good questions that are on cards, and then kind of ask a wrap-up question. Now is a good time to start thinking about filling out your blue evaluation form, if you have one. There are a few questions for Mark, and you get to answer them, since it’s your last panel with us. One question is, kind of going back to your point about beneficiaries. How do you suggest the beneficiaries share in savings with the new payment models? And maybe you can point folks to some resources on that.


MARK MILLER:   Well, one thing — and by the way, I think CMS has also tried to focus on this too, so it’s not just — this clever idea is not just from us. So, one thing that we said, was in an ACO model, forgive the beneficiaries cost sharing when they go to an ACO primary care physician. And that’s a way to send the signal to the beneficiary, to bind them to the ACO primary care, and then if that person is doing their job, they’ll manage their care. So, we said, forgive the cost sharing, so that the [unintelligible] says, oh, well if I go to this doc, I don’t have a cost sharing. And CMS tried to implement that in their Next Gen. They couldn’t get around it by law, so they gave, I think a gift card or a payment to the beneficiary, which was — and hey, I think they are out there trying to make it happen. And so, I thought that that was good. Also, in one of the bundling demonstrations, the ACE demonstration, if the beneficiary chose — I think I have this right — one of the actors who was involved in it, they also got a dollar payment for doing that. So, it’s co-payment signals, dollar signals, if that’s what they have to do.


LEN NICHOLS: And also, just know, we have the VBID model in Medicare Advantage, which is similar, but in the Medicare Advantage context,


SARAH DASH:  And VBID is Value-Based Insurance Design? Thank you. There is a quick clarifying question that someone had, which is kind of clarifying a little more of the relationship between the PTAC and CMMI, and so I don’t know how short or long of an answer that is, but can someone outline that process?


LEN NICHOLS:  So, God made PTAC over here, and CMMI lives in HHS, and PTAC statutorily was created — we are appointed by GAO, and we recommend to the Secretary — the Secretary is on top, CMS, HRQ, FDA, CMMI is under CMS. The Secretary then decides what to do with our recommendation. Rumor has it, he consults the people in CMMI to help him decide what to do, but after that, it really becomes a CMS process.


GREGORY WOODS:   I think that’s right. And just to be clear, as Len alluded to, the PTAC makes recommendations. CMMI doesn’t have any statutory responsibility to do anything with those recommendations, but we take them very seriously. And I am not the lead on this within CMMI, but I think the PTAC has made recommendations on four proposals so far. We’ve given some formal responses to, I think, three of them. And it does — as was alluded to earlier, it does take time from getting a proposal to actually getting to the stage where you are ready to stand up a model. But I would say all of the recommendations that have been made by the PTAC, and that will be made, are serious drivers as we look at our portfolio, of what we consider. And we take them very seriously.


ELIZABETH MITCHELL: I would just add that I think PTAC has really created an important venue. We are seeing so much interest from the clinical community. As I’ve said, we’ve got 20 full proposals in the first year, so there are real innovators out there, ready to redesign care. And I think the opportunity to have a transparent, open process to really vet and evaluate those proposals, has been very welcome. So, as we work closely with CMS, CMMI, we really hope to accelerate some of that progress, because we think the field really needs these opportunities.


SARAH DASH:  Thanks. Okay, a couple more technical questions, and then we are going to wrap it up. These are questions more related to MedPACs proposal. I’m going to try to combine them. The question is: We are not even through year one of MACRA, so isn’t it too soon to throw in the towel on MIPS? I will just say that the other person starts with: Thank you for thinking outside the box. But the questions are kind of inter-related. One notes that 64% of clinicians are already exempt, and the “pick your pace” approach provides an easy on-ramp for participation, and asks if the two percent withhold that’s part of your proposal, is really enough to incent change. Then kind of on a related note, and maybe you can clarify if this is a change that Congress would need to make, and if so, what are — are there unintended consequences that might result?


MARK MILLER:  The first question — it is really frustrating to get this comment, because it’s — and I have been in rooms with a lot of different organizations who are saying, yes, we know it’s a mess, but don’t back up. I don’t understand that line of reasoning. I mean, it just seems like, don’t make any changes. If you had that line of reasoning, we’d never change anything, we’d still be doing the SGR, which all of these groups wanted to change. And so, the point we are making is, the deeper you get into it, the more entrenched you get people who win under it, and the less likely you are going to be able to change it. If you feel like you are going to run into a problem in an equitable situation, change it now. Much of this change that I’ve articulated — I mean — I will stop there. Much of this change I’ve articulated involves changes in law. I think CMS has some flexibility, but I really think a lot of the structural stuff I’m talking here is not CMS’s issues, I think it’s law. And then you said, “Two percent, is that enough?” Basically, I should have been clear when I roared through this thing in eight minutes or a little more. Two percent is — we are just trying to block through the policy. If you really want a lot of change, then make the percentage higher. If you are sort of thinking the fee-for-service is really not where the action is, then keep it low. But the two percent is definitely not a hard number, and some of my own commissioners have said, this number should be bigger.


SARAH DASH:   Thanks. Alright, let me ask you all kind of a lightening round final question.


MARK MILLER:  Can I say one thing? Olivia Bursey, who is sitting right there, she is responsible for all of this. So, any of the problems that you have, you take to her. But get a card from her, she can give you information on anything that I said.


SARAH DASH:  Great, thank you. Well, let me ask you in the last minute or two that we have left, if you can quickly summarize as best you can, what is success? So, if we look even five years down the line, what is success when we are looking at Medicare physician payment? And if you feel like it; what is the biggest change in your view, that we need to achieve those results?


MARK MILLER:  You keep working that. I think Medicare and the government should never tell a physician how to practice medicine. Blow by blow. The government’s role is just to set up a payment system that directs them towards good outcomes. I think something that was said, is that we should be looking both at under and over — or, we call it low value care. So, to my mind, at a payment system government level, it would be those kinds of broad signals, and not being in the physician shorts to collect the data, or to micro-manage the care.


ELIZABETH MITCHELL:   I guess I would say that all stakeholders — patients, employers, providers, fell that they are getting better care at lower cost. There is lower burden. So, physicians actually want to stay in practice. We have mechanisms to coordinate across communities. The models that we are seeing at PTAC and elsewhere, they require coordination of transportation and the social determinants. All of the things that contribute to health, we are going to have to find a way to enable that connection through better data systems and different relationships, but it should be experienced by all the stakeholders as an improvement.


SARAH DASH:  Thank you.


GREGORY WOODS:  I wouldn’t want to speak for all of CMS, but I will say, on the APM side, one of our goals is to increase the opportunities for clinicians to participate in APMs, and in advanced APMs and to increase participation over time. I would say for the first year of MACRA, which is coming to an end, I don’t have final numbers, but I think in the rule that we issued last year, we projected that something like 70,000 to 120,000 clinicians would be QPs. I have no reason to think that we are that far off in that estimate. I think to sort of respond to some of the points that Mark made earlier, the vast majority of those are going to be from total cost of care models, with downside financial risk. So, I think Len said at the beginning, that the number of people who are willing to assume risk and take accountability would fit in his Prius. I don’t know how big your Prius is, but I think that’s a slight underestimate, and I think there are really positive signs, and I don’t have a specific target, but I think more opportunities for eligible clinicians to participate in advanced APMs and more clinicians actually participating.


AUCHA PRACHANRONARONG:  Mine would be pretty similar to Greg’s, is I think we would want to see more eligible clinicians participating in APMs than in MIPS over time, and that those that are required to participate in MIPS have the buy-in, and are engaged, and see the value of their participation.


LEN NICHOLS:   So, I would define success as sort of good public policy is channeling self-interest to pursue the social interest. I tried to make a case that social interest is about reducing cost, so you’ve got to realign incentives, so that they help us and not fight us. And the only way to do that is to take their input into the design of what these evolving incentive structures are going to be. And I think there is a lot of promising stuff across the board. What we need to make it happen, in my view, the single greatest hole is the data infrastructure. We do not have a data infrastructure that enables clinicians to engage easily in the process of engaging us, engaging the professionals, engaging the people who can help them redesign their incentive structures. And the only way to build that, in my view, is to have a public/private partnership to make it happen. It’s feasible. It exists in some places. In my view, it could be done fairly quickly. It’s not a technical problem, it’s really a property rights and political problem.


SARAH DASH:  Great. Well, thank you all. And thank you all for your participation today. For those of you who are interested in more discussion on this, the Alliance will be holding a briefing in just a week from today, on December 8th, to look a little more closely at the workforce issues, some of which were raised today. So, December 8th at noon, watch our website for that. I want to thank the Commonwealth Fund again for making this briefing possible, and please join me in thanking our panel.