ACOs and CINs — Where Did They Start, How Have They Evolved, and Where Are They Going Next?

Amy Flaster:                 Thanks everyone for joining us today. Jonas and I are very excited to be with you talking about pop health and many of you are using your lunch hour for this, depending on where you are in the country, so thanks so much for being with us.

Next slide.

This is our agenda for today, and when Jonas and I were thinking about what to cover on our webinar and our time with you, we knew we wanted to dive into alternative payment models and that field is changing and evolving so much and there’s so much ground to cover that we’ve decided to scope this somewhat. So, the structure of today’s Webinar is going to begin with some context. Where have things been historically in terms of value based care? What is the evolution of payment reform looked like that has gotten us to today? We’ll then get into a quick overview of ACO’s and CIN’s. So, really the 101 of the ACO and CIN models. We’ll then move into the evolution of ACO’s and CIN’s. How have the model’s changed and why have they changed? We’ll then talk a little bit about lessons learned, things that have remained pillars for success in the ACO and CIN models over time. And then finally, we’ll take a look at what’s coming down the pike. What do we anticipate will change in the world of alternative payment models and what are some new competencies to develop to meet those changing demands.

Next slide.


Starting with background and context. Before we talk about payment reform, it’s imperative to discuss how we got here. As I’m sure basically everyone listening to this webinar knows, the growth of healthcare is un-sustainable. It’s the largest driver of American federal debt. It puts a huge amount of pressure on state budgets through Medicaid. And in the private sector, cost to self-insure employee groups places immense pressure on companies and their balance sheets.

I was recently reading an article that a colleague and friend of mine published a couple of years ago, and I think he summed it up nicely. He said, if Medicare spending continues to exceed GDP by two and a half percentage points a year, and that’s the traditional gap over the past 40 or so years, a greater than 160 percent increase in individual income taxes would be needed to pay for it. And even if the gap were just one percentage point, so half of the gap it is now, the increase in income taxes needed would still be over 70 percent, with a consequent contraction, or reduction, in GDP of three to 16 percent. That’s all to say, the rise of healthcare spend is unsustainable.

Next slide.

And so, in response to really massive financial and massive political pressure around the cost of healthcare, we’ve been on a journey. All of us on the phone have been on a journey of health care and payment reform that has spanned decades. Starting on the left, I think many would say a pivotal starting point in this journey took place almost 100 years ago, in 1932, when the report of the committee on the cost of medical care came out. This is actually a screenshot of the title page. And that report recommended what they described as the integrated practice of medicine rather than autonomous individual sets of practices. We were grappling with this almost a hundred years ago. How do achieve true integration?

Fast forward to the seventies, 1973, with the passage of the Federal HMO Act, which encouraged the growth of prepaid medical groups, HMO’s, IPA’s, and what are really the predecessors to today’s ACO structures. I think the language from that time is interesting because at the time these were constructed as “an alternative to existing fee for service medical care, bringing together a comprehensive range of medical or healthcare services in a single organization and noting that HMO’s would provide these services as needed to subscribers in return for a fixed monthly or annual payment periodically determined and paid in advance.” In many ways these were the first example of early ACO’s and were the first example of value based payments.

Fast forward on this journey a little bit to the nineties. We get to the managed care era, the managed care movement, and, as we know, insurance companies contracted with providers at the time with payment that was often capitated. And, history has told the story, but with little quality assurance, little protection in terms of insurance risk being borne, there was a lot of very public backlash on the part of patients, on the part of providers.

But interestingly, around this very same time, in 1996, CIN’s were actually being born as well. The first organizations to integrate themselves clinically actually prompted the DOJ and the FTC to create standards around this integration. They issued the statements of antitrust enforcement policy in healthcare, which basically allowed these joint ventures to exist and also provided guidance around clinical integration. So that was the nineties.

Fast forward to 2007, and that is the year when Elliott Fisher published this health fair paper. You can see here in the middle. Really introducing the term accountable care organizations, ACO’s, to the world, which ultimately, of course, made its way into the affordable care act years later.

That very same year, 2007, I think something that really is a landmark change happened, which is the IHI developed the triple aim framework, with a triple focus, as we know, on trying to improve patient experience, improve the health of a population, so improve the quality and reduce healthcare costs. This triple aim, and ultimately the quadruple aim, which was born out of it, I included here not because they are landmark legislative movements or acts in the development of value based care, but I really think, when this went on the map in 2007, the triple aim has since provided a real moral compass for healthcare reform. So, thought it was worth noting.

So, fast forward to our current decade, 2012 is when ACO programs first launched, and just six years after CMS’s first program, we’re now sitting at over a thousand ACO’s covering $30 million lives. One in 10 Americans. And as we’ll discuss today, this journey is continuing. It’s continuing to evolve. Definitely a dynamic process.

Next slide.


That was a little bit of context and now I’ll shift into talking about ACO’s and Jonas about CIS’s, and really cover the basics. So, ACO and CIN 101.

Next slide.

I’ll start off with ACO’s. I do think it’s worth beginning by just defining what an ACO … what an accountable care organization is, before we get into the greater depths of what’s in the contracts and what some lessons learned are.

CMS defines an ACO as a legal entity recognized and authorized under applicable federal or state laws, comprised of eligible groups of providers that work together to manage and coordinate care for a payer specific population. And while that is the legal CMS definition, I’ve included here some other definitions, just to say, there isn’t one standard industry definition of an ACO. We see the AFP’s definition, the ACP’s definition. But, what I do see, and what I’m sure you all see, is that there are some commonalities between all of these disparate definitions.

ACO’s are, across the board, comprised of groups of doctors and hospitals. They’re voluntary entities, as well as legal entities, and across the board they evolve shared responsibility for both quality and cost for a population. We can see those threads really throughout these definitions. And then, one final feature to mention is that unlike the traditional fee for service reimbursement model, which is really as we know an eat what you kill model, paying providers for the services delivered, an ACO directly ties payments to the outcomes of care and the cost of care provided.

Next slide.

So again, covering some of the basics. I thought I would dive a little deeper into what some of the commonalities are across ACO’s. And this applies to both Medicare ACO’s, which is really what we’ll mostly focus on today, but also commercial ACO’s and Medicaid ACO’s.

Clinically, we know that ACO’s share a number of features. They’re oriented around standardized clinical protocols aiming to reduce variation. They coordinate care amongst providers in the ACO, so amongst PCP’s, hospitals, post-acute facilities and specialists. They’re committed to meeting quality targets, both to defining these, measuring performance towards these, as well as supporting the care and the programs needed to achieve these targets. ACO’s clinically stand up and maintain a broad spectrum of care management efforts that really varies based on the population being managed in the ACO. This could be chronic care management, Longitudinal Care Management for Medicare patients. This could be transitional and shorter term care management across the board. This may be care management with the behavioral health focus for certain populations. But care management is a core clinical pillar.

We also know, clinically, ACO’s promote a site of care strategy, as well as an access strategy, really aiming to keep patients connected to primary care and in the lowest acuity, but still appropriate site of care. And lastly, ACO’s promote a patient engagement strategy.

Administratively, ACO’s also share a number of features. They support the creation of governance, as well as an administrative infrastructure to support the work being done. They provide for enhanced IT data analytics and other support functions. They develop a system, as well as a staff, to monitor quality and payment targets and performance towards these targets. They support contracting, in all of its forms, including contract negotiations, contract management, as well as relationships with external partners. And they establish both relationships with participating providers in the ACO, as well as supporting the flow of funds, so the procedures to distribute financial payments to participants.

Next slide.

Those are some common clinical and administrative aspects of ACO’s. But I thought I would take a minute to dive a little deeper into ACO contracts. ACO contracts differ so much based on the kind of ACO, that it can be hard to make generalizations, but I do think one common theme amongst, really, all ACO’s is a contract goal that is shared, which ultimately is to reduce a population specific total cost of care, to reduce TME. Otherwise, the contracts vary greatly in the specifics.

Some things to look for in a contract, and some questions to ask are, what benchmark is being used? Is it a regional cost trend? Is it based on the historical costs of attributed patients? Does it account for gain deficiencies in the market? What is the minimum and maximum savings rate? And how do these vary over time? Are they steady, or do they ramp up? How is quality being measured in the contract? How negotiable is that? And, what is the quality reporting program in the contract?

How is attribution being managed? I think there are reasonable arguments in favor of both prospective and retrospective, but an important feature to consider. What is the performance period of the contract? And, does it give you, as an organization, enough time to allow for some stability in your strategy? How much data do you get access to in the ACO contract? How is it served up? And how often do you get it? And lastly, what is the contract review process like? And specifically, what events trigger a positive review?

Those are some of the basics of the ACO model. And I’m going to turn it over to Jonas to talk a little bit about the CIN model, before we go a level deeper into some of the recent changes.

Jonas Varnum:              Thank you Amy. Amy talked a lot about the ACO 101. I’ll talk a little bit about the CIN 101, and we’ll get to evolution here. But, if the ACO ties payment to outcomes, and there’s some legal requirements as how ACO’s have to hit, or how they’re built, and what they’re going to be held accountable to it, it’s one form of collaboration. The CIN, the clinically integrated network, it’s also a collaborative entity. It has some clear guidelines of what organizations have to achieve to meet some of those federal definitions of integration and other kind of legal requirements that Amy hit on earlier. This structured collaboration, it’s aligning different providers, it’s aligning your community providers, different groups of physicians in hospitals, different groups of non-provider organizations that want to join the CIN.

It’s really designed to achieve efficiencies of healthcare services, also to better quality, but its goal is on this efficiency and quality similar to the ACO, but across healthcare services. It allows organizations, really, to jointly agree on terms or conditions of a contract with one single signature, and the FDC has said that’s pro-competitive. That’s really what allows the CIN to grant the legal opportunity to organize in the fashion that it does with its … across its continuum of participants.

One kind of key component here is really that CIN’s, and all of those participants, are interdependent and they’re cooperating in order to drive those efficiencies. So I’m going to touch a little bit now on some core principles, or four key principles, that every CIN has, and it starts with these providers that are across the continuum and they’re coming together, they’re signing participation agreements, and there’s active participation and engagement of these provider groups. They are becoming interconnected. They have to be to meet the standards here. And they’re held accountable within the CIN group. There’s recourse if they’re noncompliant, or if they’re struggling to maintain a level of excellence that the CIN itself has defined. And, I think that ability to say, this is our network and we believe in it as a provider group, it really does allow physicians and other members of clinical teams to have a really good say of what the standards of care are within that network. That’s another component of really having clinical quality standardization.

And then lastly, it’s these resources, again, becoming interdependent. How do you do that? How do you put in resources and processes that promote this cooperation, that promote this aligned value across your network. Examples include just being able to have additional insights into a broader service area, having the ability to opt into contracts, to purchase services, or to have group service, group purchasing services, those types of arrangements. That’s really what the CIN model can be effective for.

Similarly, as ACO kind of got … or, as Amy got a little bit deeper into the ACO, this is a little bit deeper into the components of the CIN. We talked about participation agreements. I’m going to start out here in the quadrant that’s at about 2:00 here, with the legal structure and governance, and what participant agreements and participation criteria are really important, engaging and constantly having educated physicians, also vitally important. So there’s, I don’t want to say you can just put a legal structure and a governance on the shelf after the CIN is built, you need to build this constant engaged group of physicians and clinical leaders to really make sure that the CIN has staying power. Performance improvement, we talked a lot about efficiency, already. Creating additional efficiencies across the network is the core component of a CIN.

I think I’ve seen CIN’s that started out with a focus on the centers of excellence model, really across a network. I’ve seen CIN’s start with your traditional hospital quality and efficiency programs. But really, these core groups that are saying, here’s our standard of care that we want to hit and we’re going to reduce the clinical variation and we’re going to achieve additional best practices across that specific target. That’s where CIN’s have evolved significantly. And I think one of the components there has been the patient experience. Patient reported outcome measures is a component here. We’ve talked … I mean, patient satisfactions been part of healthcare for a long time now, but, really bringing in new data and new information of how patients, and patients across the continuum, are experiencing their care. Where are they experiencing it across a network? That’s kind of a new area of performance improvement for the CIN.

Information technology. Anything that you’re trying to improve, you need to measure. I think the first CIN’s, they started with just having initial amounts of data that could see across the continuum, but now we’re really aggregating a ton of data inside these entities, we’re creating business intelligence that spans both a broader brush, plus a much deeper depth, into a whole bunch of data sources that really give insights into our patient population. Information technology is not all about just bringing in data and measurement reporting, it’s also about fundamental components of just access.

Telemedicine has been a huge component of allowing different CIN’s to have access to service, or having the members of CIN’s get access to services that they might not have ever been able to afford, or come together and say we need this. And some of this innovative transformation effort is also inside the IT space.

Lastly, financial and operational administrative components. You generally start out with one or two contracts, you understand the funds flow. I think staffing is extremely important where there’s always this conversation of, “Do we build? Do we buy? Do we create our own? Do we partner?” And as you put together the staff of the CIN, you’re also evaluating different relationships that are the backbone of the CIN. So, I’m going to actually I think turn it over to Sarah for our first poll question as we get onto some of the evolutions of these programs.

Sarah Stokes:                Yes! Thanks Jonas. Okay everyone, we have our first poll question here for you that I’m going to launch. So we’d like to know, which value-based contracts does your organization currently participate in? And your options today are Upside risk Medicare ACO/product, Downside risk Medicare ACO/product, Medicaid ACO/risk product, Commercial ACO/risk product, and Payer/provider partnership. And while we’re letting those votes come in, we’ve had a couple questions, Jonas, and maybe Amy too, if you could just briefly explain sort of the key differentiation between an ACO and a CIN. What is that difference between the two?

Jonas Varnum:              Sure. And Amy, I’m happy to take this. I think there’s plenty of components that are very much aligned. I definitely think of ACO’s as a little bit more contract oriented. They can definitely have similar structures, services, but they’re contract oriented, and really they’re payment is tied to outcomes that sometimes CIN’s, they have a little bit more flexibility of some of the contracts that they can accept, some of the services they can offer. There’s a lot of both federal and state level definitions that really define each market’s distinct differences. But in general, I think that’s a good way to think about the difference here.

Sarah Stokes:                Perfect. Okay.

Amy Flaster:                 Yeah. I would just add to that Jonas, exactly as you said, but put simply ACO’s are tied to the contract with the payer. So an ACO is an organization, a population of patients put into a risk contract with a given payer. You can have Medicare ACO, an MMSP, a Next Gen, which we’ll talk about. You can have a Medicaid ACO. A CIN is the organizing body amongst providers that can support multiple contracts. So exactly as Jonas said. ACO is a contract based term.

Sarah Stokes:                Okay. Perfect. Well, I’m going to go ahead and close this poll and share the results. So 53% reported a Payer/provider partnership, so that was our highest area. Then we had 40% reported Upside risk Medicare ACO/products. 18% reported Downside risk Medicare ACO/product. 22% reported Medicaid ACO/risk product. And 34% said Commercial ACO/risk product. Does that align with what you would have envisioned Jonas or Amy?

Jonas Varnum:              Sure. I’ll take that one and we’ll get into the evolution here. I think that does align. I know Amy will get into a little bit some of the Downside risk Medicare ACO/products coming up. But the 50% plus for the Payer/provider partnership, that’s great. I think we’re seeing some evolution just in that particular market as well, and I’ll talk a little bit about that where even the very nitty gritty components of what that partnership means are getting transferred a little bit over to the provider groups, so we’re talking about too. So I think that can mean a lot of different things.

But I will continue on now with a little bit of the evolution of, and we’re going to start with CIN’s and kind of just keep the CIN flow going. I think that the Clinically Integrated Infrastructure, this is probably the starting place where you’ve aligned, you’re independent physicians, different groups of physician entities, and some of those groups are pretty mature by themselves. You’re aligning hospitals, health systems and post-acute providers, other service providers. I’ve mentioned, they can all join this clinically integrated entity. And then they’re delivering services and how the entity’s structured is through these very specific committee structures. This is an example. This is again, you’ve got some legal requirements that you have to hit, but you want to focus on finance. You want to focus on clinical quality. Some of those level deeper components, and you’re just putting in some staff, and some initial tasks, and initial charter responsibilities for those members of those committees.

The CIN then has really taken into, we got to get to speed quickly. Let’s get into some of these contracts, and it becomes kind of a multi payer effort as the next step. So early on, some of the self-insured products of CIN’s program, now I’d say even maybe CIN’s are kind of focused in on Medicaid, or I’ve seen the first contract has been really on a Medicaid focused scenario. But the interventions are kind of just built upon and building blocks, the network value. You want to be quick and provide low hanging fruit, but then it really needs to continue as you take on additional contracts and providing value to your participants.

And so that’s where I think we’re seeing CIN’s be today. I think they’ve scaled some of their opportunities across additional provider groups. And so Amy’s been talking a little bit about the ACO world. I’ve seen organizations, I’ve seen one ACO where physicians join an ACO contract and be involved even in additional clinically integrated network. And then I’ve seen those organizations really align with more regional or what’s been called a Super-CIN Model. And their goal is just sharing additional services and additional contracts. And when you even … Sometimes these are getting pretty darn complex here on the right where providers are seeking contracts by themselves, in the blue box. And then in the gray box, they’re seeking their network options. They’re looking for how we can scale some of the operational success across all of these different CIN’s entities that are in their market. Super CIN’s are coming in and offering very specific services to different providers. So it’s becoming a pretty interconnected group of providers. And again, what that goal of achieving efficiencies and understanding how to scale their operations.

Amy Flaster:                 Great. Thank you Jonas. So again, now that we’ve covered the evolution of the CIN Model, we thought we would flip to take a look at how ACO’s have evolved over time. And while this presentation is focusing on CIN’s and ACO’s, I did want to take a minute to remind ourselves that these are just two of many, many innovative models and contracts coming out of Medicare. So on the left we see 2017 and 2018 participation in Medicare APM’s, alternative payment models. And while we see that the MMSP and Next Gen ACO participation is growing most significantly in two of the only models with increased participation, there is still real participation in the other models, the oncology, or end stage renal disease care models, the BPC model, et cetera. So a reminder that these are just two of many.

And on the right we see that while ACO’s are growing overall in number, this actually has historically been driven mostly by commercial ACO’s. Though that’s flattened off in the past year, that is been the historic driver. We do see Medicare ACO contracts continuing to increase over time and though trailing far behind, we also see Medicaid ACO’s with a slow and steady rise. But none the less, mostly commercially driven.

And so that’s all to say our focus today has been and will be primarily on Medicare ACO’s but not because they’re the only Medicare APM model, nor because they’re the only ACO Model, but rather because they’re a national ACO Model and with such a diverse group of folks on the phone from different regions and different markets, we thought it made sense to focus on models that we can all participate in and share in, despite our region and our market.

Next slide. And so again, admittedly focusing on Medicare, I thought I would walk us through some of the Medicare ACO tracks. So going in a little deeper than we covered before. And here I have MMSP Track 1 on the left, which is the least risk bearing of the MMSP tracks, an upside only model, MMSP 1+ on the right with the most downside risk of the MMSP tracks, to Next Gen on the right of that, which is the most sophisticated and risk bearing of the ACO Models. And so just quickly to cover some of the facts, MMSP Track 1 has a minimum of five thousand beneficiaries. It’s a one sided risk model with sharing and up to 50% of savings, no losses in this model, not to exceed 10% of the benchmark. This uses a retrospective attribution model, so a look back based on where patients received their care. And an organization has to reduce costs by 2 to 3.9% to share in the savings.

Track 1+ with some increased risk bearing. Again, has a five thousand Medicare beneficiary minimum, is a two sided risk model with up to 50% sharing of savings and 30% sharing of losses. This model shifts to prospective attribution. One must reduce costs by .5 to 2% to share in the savings. And finally the Next Gen Model, which is what we are at here at Partners, requires ten thousand Medicare beneficiaries, to create is a two-sided risk model with two tracks or approaches, with either 80% or 100% risk. It’s a perspective attribution, has a first dollar savings, so above or below the benchmark for first-dollar savings, and has four options in terms of flow of funds. So payment ranging from fee for service all the way to full capitation in terms of how the payments take place.

And then looking on the right, we can see how enrollment has tracked by year. So all the way back to the pioneer ACO’s in 2012, we see these models have really achieved significant adoption with over 10 million enrolled in the MMSP, and rising enrolling in Next Gen as well. So really significant adoption of these Medicare ACO Models. And interestingly, a fact that’s not reflected in this slide, but 21 organizations actually opted into downside risk CMS ACO’s in 2018 without prior CMS ACO experience. So folks aren’t necessarily starting with Track 1.

Next slide. And so now to look to the future. We’ve covered what ACO’s are, some of the details of the current Medicare offerings, but where is this all headed? And so I thought I would include some direct quotes from CMS and CMMI just to give everyone a sense of the message they’re sending and where their themes are. So Seema Verma, the CMS administrator, made a speech in May and this is a quote from him. She says, “We are working for competition and better value by moving away from a fee for service approach, to a system that is value based and that rewards value over volume. We also want to think about models that create a true competitive market where providers compete for patients on the basis of price and quality, and moves the government out of the business of setting prices. And in all of our models, we will also make sure that our beneficiaries have incentives to seek value when they obtain care.”

And so similar theme on the right, actually, from CMMI director Adam Boehler. He says, “It’s also CMMI’s job to say to folks, ‘If you’re not cutting it, get out of the way, because there are others that will come that will cut it.’ People will come in and take that risk and do something with it.” So some early signaling that there is an emphasis on having some skin in the game and creating a true competitive market and letting market forces work.

Next slide. And so, then we transition from some of the signaling that we’ve received from CMS and CMMI to the proposed ruling that has been unveiled with some non-final changes to the MMSP program called Pathways for Success. And again, these are non-final changes. I believe comments were being accepted as recently as last week. I think October 16th is when they closed. But final changes are expected to occur in the next year. And so there are some key changes to the MMSP program stemming from this proposed ruling. And so as you can see here, there are two tracks being proposed in the new revamped MMSP program. And those are the BASIC Tracks, formally MMSP tracks one and two. And the ENHANCED Track, which folks are likening most closely to MMSP track three. And the BASIC Track would have, as you can see, a maximum of two years of upside only risk, and then would transition for the remaining years to increasing levels of risk, downside risk year over year.

And the exception to that, or the nuance to that, is that for organizations that are currently in upside only risk contracts, so currently in MMSP Track 1, there would only be one additional year of outside only risk available. One benefit of this BASIC Track is that the risk of this track by year five would qualify by macro rule as an APM. The second track, the ENHANCED Track is a little bit simpler. Again, most often compared to MMSP Track 3, it would offer a constant level of risk for each of the five contracting years, so no ramp up there, and qualifies by macro ruling right off the bat.

Some other notable changes to this model, the contract agreement period is five years, up from three years, so a longer stretch. With this model something great is that operational support mechanisms are becoming more standardized, including establishing a beneficiary incentive program. The SNF 3-day Waiver program, et cetera. But as I’m sure many on the phone are aware and have discussed in your own organizations, this model has faced some criticism for a number of reasons. One is that it does represent ultimately a reduction in the percentage of maximum savings. So for example, in the current MMSP Track 1, there’s a maximum of 50% shared savings, and as you can see here, that really is quite a bit lower in the BASIC Track with 25% shared savings available in years one and two, up to 40% in years four, only reaching that same level in years five. So really does mean from years one through four, the amount of available shared savings is reduced.

There have been a lot of reactions to this, positive and negative. NAACOS initially did publish some survey data suggesting righty off the bat that 70% of responding ACO’s were likely to leave the program if forced to assume financial downside risk, though those statistics have gone down with time as folks have learned more. I think most recently it was 36%, down from 70%.

Next slide. And so then that brings us to the data. With so much discussion of the ACO Model, how we got to this place, and what division for updating the ACO Model and specifically the MMSP model in the next year is, it begs the question, do ACO’s work? Do one-sided ACO’s work? Do two-sided ACO’s work? Do they improve care? Do they lower the total cost of care? And the short answer is that the data is mixed, and the jury is very much still out.

So on the left we can see CMS’s latest publication, and this data would suggest that upside only ACO’s are losing money, and that this is driven really by hospital based ACO’s as opposed to physician led ones that did save money. And we can also see here from the item labeled Exhibit 4, that two-sided ACO’s are thought to be cost saving. But what I think is potentially most interesting as opposed to which ACO’s are losing money and which are gaining, is that the actual net impact is not huge. It’s 49 million, 33 million. But relative to the total budget of CMS, these are really sort of a drop in this bucket.

With that said, on the right we have recent papers in the form white papers, and in the medical literature, that are actually suggesting more significant savings from MMSP ACO’s. So the first paper on the right, from the New England Journal of Medicine, published in September of this year by McWilliams and some of his colleagues, did suggest that the MMSP program in fact resulted in a net savings of 256 million dollars to Medicare in 2015. Similarly though, he did see that spending reductions stemmed from physician group ACO’s. So those are the ones that were most effective with actually no net savings from spending reductions in hospital integrated ACO’s, those really being offset by the bonus payments.

And then likewise, on the far right, this paper in green is a white paper published by NAACOS this past August, and it also suggested that ACO’s in the MMSP did generate savings, a gross savings of actually 1.84 billion, a net savings of 541.7 million, spanning 2013 to 2015. Which is all to say the data is mixed.

Next slide.

Jonas Varnum:              Yeah. Thanks Amy. I have been following the ACO movement for a while, as you know, but I love seeing these slides. I think that one of the subtle messages that I know we’ve been communicating throughout this webinar has really been that looking back even at the evolution of, as you’ve kind of pointed out, that the 1930’s work all the way through HMO’s, people are taking on risks, some various form of provider group, or provider structures going to take on risk. And it’s just … What I like about the ACO movement, the CIN movement, is we’re getting a lot of muscle memory in some of our provider organizations, so that those organizations in time really are going to be key to that risk bearing group. And that really leads into what are these key components of success that we’ve seen have not really changed too much?

I’m going to go through these pretty quickly so we can get into what’s coming down the pipe. But as we look at organizations that really have been successful, we see some elements that are pretty consistent. And I think first of all I’ll start by saying organizations really need to know their own goal and that has to be drilled into them. I can’t think of how many times where I’ve had conversations, and I’ve heard stakeholders and key doctors, or groups from siloed departments say, “I really struggle or I’ve become misaligned. Our organizations kind of really been become misaligned in what we’re trying to do in pop health.”

This is marathon work and keeping and distributing one vision, one definition of success, really focuses the efforts over time. And leads right into having and educating and engaging physicians and stakeholders. I think if defining success is orientation, then it’s also key to have really an awareness and appreciation across all the different stakeholders of who’s adopting and where they are in the framework of population health over time. So if we’re doing a lot up front, we’re building awareness, we’re building desires, and as we onboard this new payment structure, some new operations, we need to provide people regularly with knowledge and the ability, and I think ability is pretty tactical in this area to sustain some of that adoption success.

All that to say is we need a true north, and the patients should be the true north. The patients should be the guide. So I went over some internal stakeholder keys to success on the last few slides. This is really who the external stakeholder is. And whatever you’re building, that supports your population health effort, that gray box down there at the bottom of this pyramid, maybe you’re building a care management team, maybe you’re building a data operating system that really brings in and aggregates data. I talked about little bit about telemedicine, you’re bringing in different technology tools. These are how’s. These are some of the what and the strength of helping your patient population. Maybe you have a longitudinal view here and longitudinal strategy around opting into various delivery models or really building your own health plan. I’d still consider that on operational enabler. And I think the patient specific, the market specific patient population, that needs to be the front and center. That needs to be the true north. It should drive what contracts you’re opting into. Some of those components that Amy went over inside your ACO’s of how you look at your benchmarks, all of those should actually be centered around the types of patients. So knowing who your healthiest populations are, really knowing the breakdown of your patient group, that’s the organizing key element here.

I think next becomes, if we’ve got key elements, we’ve got some of the internal and external stakeholders really consistent across the CINs over time, having an actual cyclical philosophy, creating a framework of how you’re going to be successful is the other real big sticking point that has held true over time. This is just one example. This is one framework for transformation. It starts a little bit with a data-centric model, and I think that’s really important in having organizations build the amount of information that are inside these CINs and ACOs.

I’ll tell you what, if this framework actually speaks to you as an organization, I have definitely seen a much bigger appreciation over time for, inside this kind of data transformation phase, of transparency in your data, of actually understanding what’s under the hood, being able to know exactly how a metric is calculated, having clinical individuals that say, “You know what, that should have moved the needle. What was it that we actually did?” And being able to look into that. I think that’s really key here. But, really, this framework, it should be cyclical. It should really promote a continual process of operational success. Aggregate your data, provide that data in a meaningful way to stakeholders, and have those insights drive what payment models you need to be participating in, what opportunities you have to improve your clinical populations over time, and what care transformation activities you should be entering into.

So, what’s coming down the pipe? What’s next? What has changed? Where are some of the more evolved organizations, both CINs and ACOs, in the space today? I think the first thing we’ve seen is we’ve seen organizations get really purposeful and aggressive around their contracting strategies. So, we are thinking very specifically about mixing various types of contract and revenue opportunities, so that we have a really healthy portfolio on a constant interval basis. We can use some of those revenues to increase some of the different tactics and operational enablers for your populations.

Having additional depth in some of the contract operating benchmarks and attribution, I think, Amy touched about how some markets, we’ve seen a regional cost trend approach in some of these benchmarks, or, well, what we’ve seen a regional cost trend just overall included, but some markets are even looking at total growth in claims costs, and total growth in medical costs, so benchmarks are becoming a little bit more conversational, you’re negotiating here. Having really transparent historical data to drive those conversations is key.

I think a lot of organizations are starting to see beyond just the CMS claims data source, and realizing wow, one payer upstream has a lot of different opportunities to think about claims data, and they’re changing them left and right. How do we normalize that? Can we define that with them upfront? That’s a really good conversation to have. Same with quality measures. So, really work to create a simple slate of quality metrics across multiple payers, especially in your commercial markets, where you’re deciding on … And don’t make 10 to 15 max, even less than that hopefully, and internally at the same level be very purposeful on your quality measures.

We had 50 plus percent of us say they’re in a payer-provider partnership. There’s a lot of payer-provider integration, and the question is, on what level of the continuum is there integration? So, CIMs and provider systems are starting health plans. They’re building a lot of their internal infrastructure to be able to really come to a payer and say, “Let’s negotiate, let’s work together, let’s build something ourselves, let’s build something in tandem.” Some components I think the provider and the CIMs are definitely able to take are really understanding some of the network utilization, understanding the continuum of the care management conversation and you see the organizations taking this on, both to bring some additional revenues to their operating structure to create wellness in their market, but also, it’s a competitive market right now. United has, what, the most employed primary care docs as ACOs constantly being started. I think they’re building their own EHR now. There’s been a lot of interest in the employer market, and so CINs have the ability to either attack that, or as Adam Boehler was saying, somebody’s going to come in and take that risk.

Very tactically, very low inside the granularity here of what we’re seeing inside some of these organizations, is there is an enhanced speed in truly knowing who you can target and how, and why. Having patients that I think everybody starts with a patient population. The pyramid of knowing who your patient population is. You start in any of these contracts with an initial patient population and you’ve probably looked at one source of data, maybe you’ve combined a couple different components of data to understand that population and then you start doing a PDSA type cycle where you’re intervening with that population, but now, I think we actually have a lot more good granular data tools that are tracking the very specific components inside those patients or that are part of that patient population that allow us to quickly and rapidly build upon our interventions.

So, look at claims data that’s about six months old and intervening and watching that population overtime, that’s great, but I guess this is one way to say there’s a little bit of machine learning. There. I used that term because in reality, what we’re seeing is there are discrete elements of data that can really tell us what part of that patient we need to attack and why and we have that ability to really see that quickly now, much faster than I think we’ve had as the organization started.

Amy, I’m going to turn this over to you.

Amy Flaster:                 Great. I’m just going to build on some of the points that Jonas mentioned with a couple other trends that we’re seeing in ACO’s and value based care coming down the pike and the next reasons I’m going to speak to are ones that very much have affected the work we’re doing here at Partner’s Healthcare and my experience as a primary care doc.

One trend that I’m very optimistic about is ongoing expansion of valued based care within Medicaid. We know that in recent years since the ACA’s been enacted, Medicaid enrollment has grown by about 30% and while many states have been in the Medicaid managed care business for a long time and many states have had early success with the Medicaid ACO models, Maine and Colorado come to mind, others are just beginning to explore this work, Massachusetts is one of them and I think that this is something we can expect to see trending overtime and also is impacting the kind of investments that health systems are making as they manage patients in value based care model.

So, I think investments in populations that historically may have no had their needs met adequately and as we think of investments and social determinants of health, behavioral health, really positive investments that our health systems are making to better serve these populations. So, an optimistic trend. Next slide.

Very much related to what I just mentioned was a trend that we’re seeing around data, which is leveraging increasingly sophisticated data sources and data components and as Jonas was saying, we view data as a key pillar to success in value based care. It was step one on that transformation framework that Jonas reviewed and I believe that the better a provider system is at accessing data and interpreting data about patients and it’s risk contracts, the more likely it is to succeed in alternative payment models. That’s a basic fact about knowing who your patients are and that’s what’s really making some of these novel data sources so compelling.

We know that systems are increasingly recognizing, as I was mentioning, the importance of social determinant of health data, finding new ways to collect this data, new ways to act on this data often coupling it with geo mapping work and I think this is a really compelling space for us to provide better care to vulnerable populations.

On the other end of the spectrum, we see genomic data becoming more accessible, more affordable, more of a commodity and increasingly impacting patients health plans and patients healthcare and finally, we know that patient engagement is a key part of the equation. I think we listed it as a key component of an ACO and as such, patients are providing us with novel data sources that we can use to affect the way that we manage the care of a population.

So, data from patient wearable technology, as well as data reported outcome measures, data that patients are impacting themselves, can provide really impactful data and can influence the way that a population of patients is medically managed within a risk contract. Next slide.

So, lastly, to touch upon care management models for a minute and the way that we’re seeing them evolve, I think that in the work that I do within care management and the models that I see around the country, we can expect to see some pretty novel changes within care management over the coming years.

One thing that we’re seeing more and more is using different types of care managers to target care coordination and care management to specific patient populations. So, really leveraging diverse professionals, operating to the top of their licenses to target care management to certain populations. This may mean social workers doing care management for populations of patients with primarily psychosocial issues. This may mean community health workers from within a specific community targeting care management to those communities, as well as the traditional nurse side model. There’s room for all of them and I think it’s an opportunity to really target the care being delivered to the population.

We’re seeing increased reach of vendors. Vendors increasing their span within care management from providing, initially, technology and billing support to really now providing fully outsourced solutions. I think that’s something that we’re seeing in provider systems really across the country and finally, something that’s very exciting is the digitization of care management.

So, ways that care coordination and care of management is being targeted to patients, leveraging technology. Examples, more sophisticated patient cohort identification and patient stratification using machine learning. Risk identification or rising risk population for targeted preventive interventions and lastly, using telemedicine. So, using new technology to actually structure the encounter between the care manager and the patient.

So, much to come in terms of care management to support some of these evolving ACO models. Next slide.

So, I’m going to pause here. I think we’ve covered a lot today. Covered some of the evolution and history around payment reform. The 101 in basics of CIN’s and ACO’s as well as how we’re seeing them evolve, how those models are changing. Some key lessons learned from ACO and CIN’s over recent years, as well as some trends that we’re seeing coming down the pike.

So, I’m going to turn it over to Sarah now for this poll question and then I think we’ll take some questions from the audience.

Sarah Stokes:                Yeah. Thank you Amy.

So, audience, I’ll go ahead and launch this last poll for you before we jump into that Q&A. So, while today’s topic was an educational webinar focused on the history and future of ACO’s and CIN’s, some of you may want to learn more about how to get started or make progress with your ACO or CIN. If you’d like someone from Health Catalyst to follow up with you, please answer this poll question and we’ll give you just a moment there as we prep to kick off the Q&A session here and Amy, Jonas, are you both okay to run a little bit over the top of the hour if we have to? To get all of these questions in?

Jonas Varnum:              Yes.

Amy Flaster:                 Sure.

Sarah Stokes:                Perfect.

Jonas Varnum:              Happily.

Sarah Stokes:                Okay. All right. The votes are just tapering off now. Okay.

We’re just going to start with one of our questions here. So, our first question comes from James and he asks, “Any examples of community organizations outside of a medical system joining an ACO, not just as a planning partner, but actually taking on risk in some way?”

Jonas Varnum:              Yeah. You know, I think it would depend upon some of the definition there, the community partner, community provider. I definitely have seen organizations hold community participants to specific metrics and they’re basically just aligned inside the risk barring agreement. So, it’s a pseudo risk model. I’ve seen that happen where you’re going to get definitely some bonus payments with organizations and by joining the CIN if you achieve your own quality metrics, you’re definitely going to provide plenty of value and I think one of the key components there is if you’re providing a lot of services or some of a lot of the customers basically, then it’s only a matter of time if the contract isn’t there yet where you also have some of that down side risk shifted to you.

Amy Flaster:                 I would agree with what you just said Jonas, the only thing I would add is that we see, especially in the Medicaid world, community partners being leveraged into risk barring models, not necessarily a part of the legal ACO structure, but barring risk themselves in a two sided model with Medicaid or the state government acting as the intermediary.

So, connecting a number of now risk barring community partners to a number of risk barring ACO’s.

Sarah Stokes:                Okay. Excellent. Our next question comes from Mary Ann who asks, “How does a practice determine which model, ACO, or CIN is right for them?”

Amy Flaster:                 That is a complicated question that probably will take more time to answer than we have right now. Again, those are CIN’s and ACO’s are different. So, I will start off by answering that in terms of which ACO contract to enter and I think that has to do with a lot of factors. I think, as practices and physician organizations and hospitals are taking a look at their infrastructure, their existing governance, their existing access to data, their existing networks and partnerships, they have a sense and can develop a sense for how much risk they really are positioned to take on.

It really does require understanding what your population is, where your population is getting care, what the access to care in your population and in your community is, what the competitive market force are and an aggregate, I think that can inform some financial modeling and some actuarial work to decide which model makes the most sense. Is it a model with one sided or two sided risk? Are you an organization that would benefit most from perspective or retrospective attribution based on knowledge of who your patients are and how much you anticipate they may get care outside of your network?

So, that was a long answer to say it’s a complex planning process that takes into account lots of levels of readiness, financial modeling and some actuarial work.

Sarah Stokes:                Okay. Our next question comes from Janet and she asks, “Have you seen much happening with Medicaid ACO’s or CIN’s for safety net providers?”

Jonas Varnum:              Yes. I think that there’s … First of all, some of the state run contracts and Medicaid organizations there that are definitely including paths that are specific to the safety net organization.

So, as reform efforts come in and new legislation hits, I continually see the safety nets having a tiered approach where they’re going to be placed inside one of the tiers and it aligns to value based goals, it aligns to some of these contract types that we’ve taken and definitely has a place for them to both grow and sustain.

Amy Flaster:                 I would add, I think some of the coolest work that I’m seeing in Medicaid ACO’s is taking place with ACO structures comprised primarily of safety net organizations. You can imagine if you’re a large, diverse healthcare system serving Medicare, serving commercial patients, serving Medicaid patients, you’re certainly going to create programmatic support and strategy and initiatives to support your Medicaid risk contracts, but it’s not the only thing you’re going to do, but when we see some of these new Medicaid ACO’s being set up that are comprised of a number of safety organizations and health centers binding together, they’re doing really initiative work to support their Medicaid populations because it is their primary area of focus.

Sarah Stokes:                Okay. Excellent. Our next question and I think it will be our last one today is, “How do you engage with health systems and physician groups to improve ACO’s and CIN’s?”

Amy Flaster:                 I’m happy to start this one off as a physician that is often charged with doing some of this work. I think there are a number of key tactics to succeed in physician and clinician engagement. Jonas mentioned one of them, which I will focus on, which is, I think Jonas, the way you phrased it was keeping the patient as your north star and I think physicians certainly respond to information that is data driven. They certainly are interested in revenue and what their financial model will look like, but above all, most docs went into medicine because they want to take really great care of their patients, not because they’re particularly interested in one value based care model or another and so, I think a way to engage clinicians is to lead with the ways that these models will improve the care that their patients are receiving and their ability to do medicine, why they went into medicine in the first place.

So, telling a primary care doc that by entering into some of these value based care models, there will be an investment made in care coordination and their most complex patients who they often feel helpless when treating are going to have a whole infrastructure, a whole support network to coordinate and support their care, is really compelling. Telling a doctor going to have integrated behavioral health and social work and psychiatry support in the office is something that really gets people’s attention. Telling docs, “Yes, you’ve upheld the quality metrics. You have been for a while, but we’re going to have pop health managers or coordinators that help you do this work and help you understand which of your patients are behind on certain quality metrics, reach out to them and support this kind of work,” is very compelling because ultimately, what makes these models so exciting is they do enable us to deliver better care to patients.

To all patients, but in particular, to our most complex patients and that is something that is very compelling and interesting when you’re trying to engage your clinicians and your providers.