Thank you all for coming. This is the Value-Based Care in Addictions Treatment Workshop. So if this is not where you intended to be, I'll give you a chance to exit. But we're very happy to see you all. And I'm happy to be here along with my co-presenters. So we'll just go down the line and introduce ourselves. My name's Sabrina Poon. I am at Vanderbilt University Medical Center. I'm actually in emergency medicine. And I work clinically in the ER there. But I also am in the Department of Population Health in an office called Episodes of Care, where we do a lot of value-based work for the institution. And so that is what I'm doing here. And I will pass it along. Hey, good morning, everyone. My name's Dave Markovits. I'm an addiction psychiatrist at Vanderbilt University Medical Center, where I direct our division of addiction psychiatry. It's great to see everyone. Thanks. Good morning. I'm Robin. I'm an assistant professor at Columbia Addiction Psychiatrist, also the chief medical officer at Ophelia Health, which is a telehealth company for treating patients with OUD that I'll use as a case later in the presentation. And I've also been doing work with NCQA and NQF, some of the alphabet soup we'll get into over the course of the talk on quality measure development for the opioid crisis. Hi, everybody. My name's Jeff DeVito. And I am also an addiction psychiatrist. And I'm out in California, where I wear a number of different hats, one of which is working with Marin County, California, in the Department of Health and Human Services, where I'm the medical director for our, I'm going to add some alphabet soups, and our DMCODS program. For those of you from California, you might know what that is. But otherwise, I will explain that later. And I'm also the behavioral health clinical director for a managed Medi-Cal plan, a partnership health plan, which covers from, basically, San Francisco up to the Oregon border. And I'll explain more about that as well. Because California, as in many things, is unique and has its own acronyms and whatnot. But great to see everyone. And we will try to make this as riveting as possible. All right. These are our disclosures. I will leave them up there for 10 seconds. All right, so we have learning objectives. Here they are. Our first one is to just describe what's going on right now in terms of reimbursement for addictions treatment and how value-based care can help to address some of the challenges. Our second objective will outline three different value-based care and addictions programs, just to give you a flavor of what's out there. And then the third objective, we're going to break out into three or four small groups, I think, and really try to have an interactive discussion about how you might think about some challenges in value-based care, potentially at your own institution. So the first one, again, understanding the landscape of current reimbursement and how value-based care can address those gaps. Our goal for this objective will be to give you an overview of what value-based care is, for those of you who may not be as familiar. But I guess I'll just preface this by saying this is no way a comprehensive, deep dive into value-based care. But we do hope that you'll walk away with some conceptual understanding, and then a little bit more about how that can be applied to addictions more specifically. OK, so probably by now, the majority of you have seen this, value equals outcomes over spend. And that's value-based care in a nutshell. Certainly there are more nuances and more complexity, as you can imagine. But this is how, certainly, I think about value in health care when it really comes down to it, is how do we improve outcomes and or reduce spend to improve value to the patient? It's slightly more formal definition is paying for patient health outcomes. So how do we, again, pay for quality instead of volume? So right now, in our fee-for-service world, we are reimbursed and incentivized based on volume. You do more, and you get more as a provider paid. And so value-based care is trying to align those incentives in a different way. So CMS has a definition, and that is paying for health care services in a way that links performance on cost, quality, and the patient's experience of care. So there are a lot of measures there, and we'll, I think, get to some of those a little bit later. This is from the Health Care Learning in Action Network. It's a value-based care consortium, a private-public partnership. And they've got, essentially, a continuum of value-based care. So category 1 is fee-for-service, which is to say not value-based care. It's our current world, the vast majority of time. And then you go all the way down to category 4C, which is an integrated finance and delivery system. And so category 1, again, our current world. And they include things like DRGs, which some may think are a first step towards value-based care. But for the most part, they're not linked to quality and value. And so they really consider that part of category 1. If we jump over to category 3, these are payments based on cost and utilization performance against a target in general. So kind of trying to determine what appropriate care is and reimbursing for that. I would say 3A is most accountable care organizations. That's sort of where they sit today. 3B, some of you who may be familiar with the Medicare BPCIA program, that sort of sits there. And then some bundled arrangements you could see sitting in category 3 as well. And then category 4 is really the most prospective, sort of total cost of care concept. And they pay for these comprehensive services versus maybe a particular episode of care, like in category 3. But you could see some condition-based bundles moving towards category 4, sort of taking care of an entire population for a certain condition, not just an isolated episode. And all these things, this slide is really to illustrate that we're trying to align incentives more appropriately for all the stakeholders involved in the health care system. So the benefits for aligning payments with quality instead of volume would be lower costs and better outcomes for patients, higher patient satisfaction, more efficient care for providers. Again, lower cost, higher cost control for payers and suppliers in society, so on and so forth. So I think that's all I'm going to opine about on the benefits of value-based care. And I'm going to turn it over to Robin to talk a little bit more specifically about addictions. So this is a little bit of background context to complement what Sabrina was just reviewing. So there's been a real shift, not just in recent years, but really over the last 10, 20 years of moving from more traditional fee-for-service payment structures, where you're able to reimburse for an E&M code, for instance, for a billing code, a service, and moving toward value, as she was just showing in the figure. And part of our old model that we're trying to move away from, if you think about the incentives for fee-for-service care, the incentive is to over-diagnose, over-test, over-treat, in order to bill as much as possible to drive revenue. Not just more services, but also more expensive services. And I think it's still, I'm just perpetually surprised at how plans will eagerly pay for very expensive so-called detox admissions for three to five days, and then not be willing to pay for buprenorphine afterwards, or extended release injectable that's a fraction of the cost, even though that's where the evidence base is for long-term care. So there's a long shadow, I think, for the fee-for-service approach, in terms of what's covered and how it's covered. The ACA, in particular, was trying to address some of these problems. Sabrina had alternative payment models, so the APMs for value-based care, tons of abbreviations, as you can imagine. So this could be a partial capitated rate, fully capitated rate, not necessarily just a per-patient, per-month rate. So OTPs, methadone programs are a good example, where they may have weekly rates for the patients, for instance. And then collaborative care models I'll touch on, and then come back to with the case study with Ophelia, where we're pulling in collaborative care, which is really the use of a behavioral health care manager and a remote psychiatric specialist, either a psychiatrist or a psych NP, who can provide expertise on the periphery. But it's really more of a care management arrangement. And these are ways to have services that traditionally wouldn't be reimbursed by a plan to be covered under these bundled rates. So we'll talk about that in more detail. So what are the incentives with value-based care? It's really this quadruple aim from the IHI. So improving patient health, also, of course, thinking about population health, but enhancing the patient experience, and also enhancing the provider experience while trying to contain costs or decrease unnecessary costs. I'm not going to go into all of the detail with these initiatives and acts of Congress. But you can see that this really has been about 15 years now with major steps at the federal level. So this is much more like turning a ship than flipping a switch in terms of being able to realign provider and payer activity. So many reasons to be optimistic. And yet, in the behavioral health space, I think that we do have some headwinds and some structural problems. This is a slide largely borrowed from our vice chair at Columbia, Harold Pincus, who's done a lot of work with the measure developers over the years and the Commonwealth Fund. And this is really approaching a laundry list of problems in terms of what makes it so difficult to have value-based care work well in the BH setting. So I'll just touch on these. It can become very complicated, the whole mix of what is a full versus partial risk in the contract, but then also thinking about risk adjustment. There was just a study out recently, I think it was JAMA this past week. And in short, as you can imagine, you can have providers or care settings where you have the most complex patients presenting. And then it's going to be more difficult to satisfy quality measures for those higher need patients. The providers may look as though they're performing worse, but they're actually outperforming. They just have an incredibly complex patient population. It's also been difficult to have BH quality measures. In particular, I think that I've seen in the last five years a ton of interest in having clinical outcome measures, and yet it's been very hard to have a scale with 10 items or something that you can give to diverse populations and reflect how well people are doing in recovery, for instance, from substance use disorders. Historically, there's been much less investment in behavioral health quality measure development than other areas of health care. I think we all know there's a ton of silo problems, interoperability problems. When I was in training, which wasn't that long ago, but 10, 15 years ago, we were still handwriting notes at a lot of the addiction treatment programs that we would rotate through in fellowship. 6Ps, which is connected to that, is the patient providers, the payers and plans, the policy makers, the public health officials, and everyone really works in different domains that are not necessarily synergistic. I had a straw poll at AAAP a couple of years ago when we were in person still, and it was very few of us who accept Medicaid, who are in network with Medicaid. And that's true of psychiatrists throughout the country, usually in the single or low double digits percentage-wise who participate in Medicaid plans. And finally, number 10 here, just to emphasize about cost shifting. Often, we can get patients connected to care, bring in a site consult service in the hospital. The CEOs of the hospitals think that this is very expensive and there's not much to show for it. It really takes a lot of work to show, well, actually, we're saving days on the surgery services or patients are less likely to be readmitted. That's even more true when you're thinking about acute care service settings, where the savings may be more to society or to other payers, even outside of the health care space. But it's hard to show that to demonstrate value. So final slide, this is very busy. The point, if you look on the left, is the pie chart for insurance coverage for individuals in the US. And as you could guess, the sort of royal bluish-purple, the big chunk is commercial. And then a significant percentage is Medicare and green. And on the left, the lighter blue, Medicaid. And that proportion is more or less in parallel with who pays for health care in the middle in the United States. And on the right, who pays for prescription drugs. And the point, really, to take home here is that, if you look at the bottom pie chart, who pays for substance or addiction treatment in the US is just completely different. And a ton of it is public discretionary dollars, not even Medicaid or Medicare. So it means that a lot of our activity, in terms of what, realistically, what patients who are being treated for addiction, the services they're receiving in the US aren't even billable to Medicaid or to a plan because they aren't medical services. You think about all the sober houses and group programs and 12-step programs. People in recovery themselves who are working with others, and it may be very altruistically motivated, and yet it's not really clinical care because there are not any licensed clinical professionals. So there's a lot of work to do in the DH space. Thank you, Robin. We're coming toward the end of our portion on background and introduction to value-based care. And I sought to focus in even more on the addiction landscape in talking about some of the opportunities and challenges with value-based care. And so I'll just run through these quickly. So many of you have heard the statistic that 10% to 20% of general hospital admissions include an SUD-related diagnosis. That's often in addition to tobacco use disorder. So it's a significant contributor to the medical spend. And as Robin was saying, we think, therefore, that we can bend the cost curve. Oftentimes, we find that it's harder than we think. But we're still trying. We do think there's a 2.5 to 4x increase in the medical spend when substance use disorders are added. And so I think there's a lot more to be learned about whether we can bend that cost curve. Of course, like other illnesses, the chronic episodic illness model applies, and we'll talk about that in some of our case examples, including the bundle that we're doing at Vanderbilt. And then, of course, there's also a huge societal morbidity opportunity with over 100,000 overdoses in the year that ended in April, mostly young people. And finally, the effect sizes for medications for substance use disorder dwarf many interventions in behavioral health. So that's a unique opportunity, especially for opioids, nicotine, and alcohol. Lends itself to clear quality measures, which I think we've seen emerging in the OUD space with continuous MOUD prescribing being endorsed by the NQF. And I know Robin has educated me a lot on that. And at the same time, again, it's largely theoretical. There's some evidence now about utilization coming out. Robin has published himself on that, the benefits of MOUD. But I think we have a lot more to see. Some of the challenges, of course, the siloization between general medicine, behavioral health, and addiction, and social systems. The community-based organizations that are helping treat patients with high social determinative health burdens are considerably diverse by region. And I think Robin's slide with the pie charts really highlights that this is an even bigger problem in this area because of how much of the spend is outside of our other payers, and also how important these community-based organizations are to recovery. And finally, this is arguably the most stigmatized of mental illnesses, many patients avoiding the treatment system. So we know that 90-plus percent are not seeking treatment. We have a lot of work to do on the engagement side. And I don't know if we'll have time to talk about it, but with our bundle at Vanderbilt, which goes live in January, we're paid a fixed bundle even if people don't stay engaged. And that could be problematic for the payer. Of course, it's baked into some of the assumptions on pricing, but these things can be a bit unpredictable. So one other thing I wanted to share is that there's actually a group that has done a lot with value-based care in addiction treatment. And they've published a white paper, which is quite comprehensive. And I think it's something to really measure things you're reading about against, because it's quite impressive in how much it anticipates all of the different facets of value-based care in addiction. So just to go through the five core elements, this is the Alliance for Addiction Payment Reform. You can check out their website. So they first describe a payment model. And it has opportunities for shared risk and also shared rewards, bonuses for quality and performance. It includes a process for creating quality measures and provider standards. It describes a network of the addiction, medicine, psychiatry continuum that serves as the backdrop for the episodes of care. And then it prioritizes a central team across that continuum that can focus on long-term recovery by measuring data at intake and then making an individualized plan across the continuum. And then finally, it links standard assessment tools to patient-centered planning through that recovery plan that's developed by the team. So keep thinking about this as we go through some of the case examples. And with that, I will cede the floor for our second learning objective. Who am I ceding it to? I think we're probably going to start with this case, so. OK. Great. Oh, sorry. Playing the part of Robin will be Jeff DeVito today. So I want to start by setting a little bit of the context here for why I think this is important and what I think can be interesting about this, which is that oftentimes, we're confronted with problems that we see clinically. And oftentimes, the problem is not necessarily a clinical problem. It's a system problem. And again, I'm kind of speaking to people's experiences around this, I'm sure. And the idea here being that when we look at systems issues, there's often money that's involved. And understanding how the money flows and why the money flows in a particular way can be important for then understanding, number one, where we can advocate for change, and then number two, where we can potentially change it. So that being said, I'm here to describe some of my experience and my work in California in the Medicaid system, because that's where I've been for the past six years, and that's kind of what I know best. Now, if you had told me 15, 20 years ago, you're going to be in the Medicaid system in California, I would've been like, you are crazy. I'm going to be playing guitar and living on a beach somewhere. But no, here we go. So let's talk about California Medicaid. One thing to understand before I go through the alphabet soup here is that in California, Medicaid services are basically, and I'm oversimplifying here for those of you that are in California, so bear with me. In California, Medicaid services are divided in terms of, on one hand, you've got behavioral health services, and on the other hand, you've got medical services. The funding flows differently for these two different services. The funding for substance use services primarily flows through, and I say specialty substance use services. This funding primarily flows through the state to the counties. The counties administer the substance use benefit for the Medicaid beneficiaries in that county. Now, in California, we've got a lot of counties. We've got big counties. This would be harder to pull off in a place like Massachusetts where the county might be two blocks. But in California, this is how it's set up. The medical benefit is provided through a managed care organization, or what we call the MCPs, the managed care plans that receive money from the state, and then they provide the medical benefit, they pay the providers who are providing the medical services in various counties. It might be a single county, it might be multiple counties. In my roles, I work both with a county in their delivery of substance use services, and I work with a managed care plan that provides the medical benefit, because as you can imagine, there are substance use services that are provided through the medical setting, like the FQHCs, the Federally Qualified Health Centers, and other places where medical services are provided. That's where people might be getting buprenorphine, that's where they might be getting some modicum of counseling. That's getting paid for, if it happens at the FQHC, it's being paid for through a different channel of money than if they received it through a specialty addiction treatment service in the county. With that as backdrop, let me describe what California undertook in 2015 to expand the specialty addiction treatment offerings in the state. In 2015, California launched what's called the Drug Medi-Cal Organized Delivery System. This is the DMCODS mouthful acronym. This is a first-of-its-kind Medicaid SUD treatment pilot. This was started as a pilot in California, with the basic premise being, if we allow counties to draw down more money from the Medicaid coffers for the provision of additional addiction treatment services, will this benefit and yield cost savings across the entire system? This is the premise that they started with in 2015. Counties were able to opt in and build these systems out and offer expanded services. I'm going to talk a little bit later about specifically what got expanded, but I want to set the context first of why this is and what it looks like. The overall idea, again, is that if you invest in treatment here, you will see benefits here. Already, based on what I told you a couple minutes ago, there's a problem, because I'm paying you here to provide more services so that this person over here, who's not connected to you, gets the benefit, like the decreased ED utilizations, for example. You're asking the county to invest more, pull down more Medicaid dollars, build out a more robust system, and then you're going to try to track the benefit of that by having to look at a completely different system. This gets challenging, so I'm setting the stage here with this. In order to do this, I want to take a step back. I didn't know anything about Medicaid when I started this work. I still know very little about Medicaid. There are a few people in this world that understand this stuff really well. Maybe somebody's in the audience and can help us out. Let me give you my overview of Medicaid. This is a slide that has a guaranteed probability of seeing some people run for the door or fall asleep. Let's talk about the history of Medicaid. Wow, really exciting stuff here. I'm not going to go through all this in detail, but I want to highlight a couple different things. The first of all is back in 1935, when the Social Security Act was passed, a conceptual leap was made at that time that had previously not been the case. I wasn't around in 1935, but this was a big deal because it set the precedent for the concept of states and the federal government sharing costs because previously it was one or the other. With the Social Security Act, it established a framework for sharing costs. That's going to become important because that's basically the crux of Medicaid payment, is a shared cost between the feds and the state. Fast forward through this, you can see when Medicaid and Medicare were developed. You can see that generally the trend is that in time both Medicare and Medicaid have expanded beyond what they were originally designed to serve or who they were originally designed to serve. You can see moving up through the years here. In 1981, this is when we see the formation of the first managed care plans through Medicaid. In this model, as I mentioned, one of my jobs is with a managed care plan in California. The basic financial model for the managed care plan is that the managed care plan receives a fixed monthly premium per individual that is enrolled in their program and then are responsible for providing all of those services. The way they provide those services is that they contract out for the medical services. They contract with the FQHCs to provide the medical services. They contract out with different specialists for specialty services. That's kind of how it looks. But anyway, in 1981, that's kind of when that model expanded into the system. And then culminating in 2010 with the Affordable Care Act, and I'm going to talk a little bit more specifically about dimensions of the Affordable Care Act, but the Affordable Care Act really expanded a lot of potential offerings within Medicaid for individuals and expanded the eligible population of people that could benefit from Medicaid enrollment. So here's my very complicated slide. It took me a little while to find the smiley face, but here's the basic financial model in Medicaid, which is that you have a federal financial participation. This is the FFP. You also have the state dollars. Those are combined. If the state spends money, the feds match that money, that ends up as a Medicaid expenditure. That's why people are happy. That's the money. This is interesting to think about because there are arrangements that have been made through the years that are complicated around this matching piece. It may be one-to-one in some instances, and for some populations, it may be 90-to-one in other populations, and I'll explain this as we move forward, but with the ACA in California, because they were an expansion state, all newly eligible people in California for Medicaid, the state pays 10 percent. The feds pay 90 percent of those expenditures. This is huge federal money on the table for the states. This was to incentivize states to enroll in the Medicaid expansion. Here are some of the dimensions of the ACA that the expansion allowed for. The ACA said, okay, we now are going to list out the essential health benefits, those services that we feel are essential and that you need to offer if you're going to participate in this program, and those include ambulatory patient services, emergency services, so on and so forth. I'm not going to go through all of them, but included in this were mental health and substance use disorder services, which was a big deal, kind of expanding that and codifying that into Medicaid. We eliminated preexisting conditions. There are annual limits on out-of-pocket expenses. This is applied to all individuals under 65, and there was work on parity enforcement as part of this. These are just some numbers in terms of what the expansion has looked like in terms of just people. We've got 77.9 million people in 2020 that were covered under Medicaid. As of September of 21, this represented 23 percent of Americans. California has the most number of people that are enrolled in Medicaid. However, New Mexico has the highest percentage of their population that's enrolled in Medicaid. Medicaid is the nation's largest payer for addiction treatment services. They're the largest payer for addiction treatment services. I just repeated that. An important thing to note, though, is that because you have this nuanced relationship between the states and the feds about the money, each of these programs in each state look different. It's a state plus fed equals what you see on the ground in terms of the expenditures. Each state's different, unfortunately, because it makes it harder, because what I tell you about California might not apply to Nebraska. This is California. Just a look at sort of what the expenditures look like. This was in 2020. I'm sorry I didn't put this up here, but 9 percent of $105.2 billion was spent on behavioral health. However, behavioral health, by analysis, is a leading driver of cost in the Medicaid system. You can see it covers a lot of Californians. This is what, and this is hard, I know this is small and it's hard to read for folks, but I wanted to give a flavor of before the DMC ODS program in California, what services could you get under Medicaid? This matches pretty closely with federal Medicaid guidelines before Medicaid expansion. It's not much. There's basically no residential treatment, for example, that was covered prior to this drug Medi-Cal organized delivery system. You've got a couple other things. You've got methadone only covered in narcotic treatment programs. Drug Medi-Cal organized delivery system capitalized on what we call an 1115 waiver, which allowed California to pilot a new program to see, like I mentioned before, whether it has an impact on overall costs in the Medicaid system. This is now showing all the things that we expanded or that were allowed to expand under the DMC ODS. The additional services that people could provide for the treatment of substance use disorders and addictions. You can see it's a lot more than what we had before. In particular, the residential substance use now was open to everybody. It wasn't just open to a very small number of individuals that were, in particular, perinatal folks that were eligible for it before. We also have additional, like we can bill for case management. We can bill for recovery services, which is an ACM level of care. We can pay for physician consultations. Offered up a whole slew of new things that we can now pay for by drawing down Medicaid dollars. This is exciting. It's also complicated. It's also hard, right? Now you're saying to the counties, you can do all these things. You can bill for all these things, but you have to build a system and you have to find people that can do this. You can't pick and choose either. You have to offer everything. Net result, not every county opted into this. Initially, some of the smaller counties, like one of my favorites is Modoc County, has 9,000 people in it. Probably also the size of Rhode Island. It's large geographically, but small population. Their health and human services department has three people. They can't do this on their own. They can't develop all these things, but nonetheless, 37 of 58 counties have adopted it, including these different components that they now had to offer, like NTP services, which was also a challenge because there isn't an NTP in every county. How are you going to offer NTP counties? In some instances, we're transporting people on a daily basis over 200 miles to get methadone services in order to meet the requirements that the state has asked us to do. Again, cost, but on balance, is it better to drive 200 miles a day and get someone to methadone from a cost standpoint versus having them in the ER every other day? Again, tough to calculate because it's difficult to do that, but nonetheless, important to think about. I'm not going to go into great detail here, but this is more details about how the matching of funding has happened in the state. Like I mentioned before, not all the matching between the feds and the state or the counties and the state is one-to-one. The state, to incentivize counties, basically said, for your expanded services, we're going to give you a lot more money. You're not going to have to put much out of your own money in the county to float this program. The state will float the program for you. We want to see you do this. We want to incentivize you to do this, and again, consequently, the state also is drawing down more money from the feds to be able to do this. So let's fast forward a little bit here, and I want to talk, my last couple slides here, talk a little bit about impacts because I'm going to speak to two different ODS, we even shorten our acronyms from DMC ODS to just ODS. I'm going to speak to two different ODS systems that have implemented this program and some of the outcomes that we're seeing. So for Marin County, what we're seeing, the first major thing that Marin County was interested in was not necessarily the cost savings that the system was experiencing, but first a proof of concept of if we build it, will they come? Will we be successful in getting patients and getting them into treatment if we offer these expanded services? This slide is highlighting that, yes, we have seen significant expansions in numbers of people who are accessing services and the types of services that they're accessing, and consequently, as more transactions are happening, we're seeing more money. Now, this isn't money like profit. This is money that we're just seeing the payment for the services provided is actually coming through. We've seen increases in access to medications for addiction treatment, and we have also seen increases in lengths of stay that people remain in treatment. And remember, this is kind of an important metric when we think about addiction research in general, is we want to see, do people stay in treatment? Because the presumption is that the longer you stay in treatment, the better you do. So we've seen that. We've seen that our individuals, once they get engaged in our system, are actually staying longer, which presumably is better. That's one model ODS system. The other model ODS system is, and I didn't talk about this too much before because I didn't want to, you know, have too many people fall asleep, our managed care plan took these counties like Modoc that I mentioned before that only have 3,000 people or 9,000 people living in the county, and they banded a bunch of those counties together and said, we'll help you do this through collective resources between these different smaller counties. So we developed our own ODS program for multiple small counties, seven small counties to be specific. Again, it's too early for us to look at and analyze the cost savings, but in terms of engagement and sort of like who's engaging, you can see some of the breakdowns here of what this, you know, what this looks like. So we've had, you know, over five, this was in just fiscal year 21, 22, you know, over, you know, close to 6,000 folks that were engaged in services. You can see the ethnicity breakdown. This shouldn't be, if you're familiar with the area, this isn't that surprising from the standpoint of the population from, you know, north of San Francisco up to Oregon is predominantly white. But nonetheless, still, you know, opportunities that we're tracking to figure out, like, even within our predominantly white catchment area, are we equitably reaching the different populations that are in those regions? We're seeing a lot of people with housing instability, unemployment rate is high. This is a Medicaid population, so that shouldn't be all that surprising. Sixty-six percent of clients are self-referred, that they referred themselves in, meaning that we've gotten the word out, people are aware of the services and are self-referring. We're seeing 34 percent of clients left treatment AMA. This is difficult. We don't have a metric to look at in terms of, like, how does this compare? Is this lower or higher? Because there's nothing really comparable to a regional ODS model in sort of a predominantly rural northern California region. So these are the impacts. The next steps for us are going to be trying to figure out and answering this question, does this impact the cost expenditures, in particular around some of our high-utilizing, you know, complex care patients, which has been, you know, is a challenge for every system. But what's unique about this is that we do have the opportunity to get that data and see it, if we can get proper data-sharing agreements between the different organizations, to see if, does this drive costs down? If we take a really complex individual, for example, and we're able to get them in a residential treatment program for, you know, two months, will that impact their ED utilization, their ICU stays? And while it is an expense to keep them in residential treatment, is it on balance more cost-effective to do that than to have this person kind of continually in and out of ICU? So I am sure that I went way over my time, which is what I always do. But that, I believe, is my last slide for that. Now it's Robin's turn. Okay. Thanks, everyone, for bearing with me. Medicaid. So I'm going to talk about Ophelia Health. This is a private telehealth company that Adam and I helped set up. Adam Asaga is our senior medical director, also on faculty at Columbia, at the back of the room. He can also field questions about Ophelia afterwards. Clearly there's a huge treatment gap in the United States in terms of patients who need access to MOUD for treating OUD. More recent estimates, more rigorous methodology suggests that we have closer to 7, 7.6 million people with OUD in the country rather than the 2 million, 2.1 million that is often quoted from NSDA, from SAMHSA. So there's just a ton of people, especially in more rural areas, you know, this is not the first time we've heard this, but it's always astonishing. People are traveling hundreds of miles a day on a daily basis to access methadone because the OTPs are typically only in more urban areas. Easily 40, 50% of our counties in the United States don't even have a buprenorphine-waivered provider 20 years after buprenorphine came to market. So these really intractable problems in the United States. So are there care models that can help patients connect with evidence-based treatment that are more convenient and just easier for patients to access and to retain in care? Early retention, especially 180-day retention, is a quality measure that NQF has now endorsed to CMS and is now widely used by insurance plans. So with Ophelia, what we wanted to do was basically have the same high-quality care that a patient would have in a clinic, like in an OBOT clinical setting. You know, we both work at Columbia, think about treatment out of academic medical centers, high levels of supervision, rigor in terms of the treatment. The workforce that we put together is a prescriber, usually an advanced practitioner nurse, but sometimes a physician, who's a primary prescriber, coupled with both a care manager and a care coordinator. So this team of three people works with our patients longitudinally throughout the care journey for a given patient, and it's real-time, face-to-face clinical visits. Video is allowable if there's a technological issue with video, but the great majority of the sessions are going to be face-to-face sessions the same way you would have in a clinical setting. By having care coordinators involved, though, we're able to troubleshoot, especially early in treatment, to help patients stabilize, help them with their induction, help troubleshoot issues with pharmacies and insurers. You probably won't be surprised, over half of the inbound that our care coordinators get for patients has to do with pharmacies. And there are increasing studies coming out, interviews with pharmacists all over the country, different kinds of survey studies, and easily 20-plus percent of pharmacies in rural areas don't even cover, don't carry buprenorphine, can't dispense it. There's just a study out recently showing some hospitals in Texas don't even have buprenorphine, wouldn't be able to get it within a day, even if they wanted to. So this is really a difficult problem, especially in rural areas. Just to flesh out some other things that we provide, we focus on a three-pronged approach to patient care for optimizing outcomes. So one is retention. Secondly, we think about adherence. There was a question yesterday for the personalized psychiatry and the risk scoring, risk stratification for patients in OUD care. And we know that adherence does matter, and there have been a couple studies showing that if you can keep your medication adherence, or the proportion of days covered, in terms of buprenorphine prescription activity, keep it above 80%, so in a given 30-day month interval, if the patients are taking buprenorphine at least 24, 25 of those 30 days or more, you're going to have lower risk of overdose and lower billed charges to the insurance plan. So we think about retention, we think about adherence, and then we also think about comprehensive care for treating commonly occurring psychiatric comorbidities, so depression, anxiety, PTSD, insomnia, easily 40%, 60% of our patients have active and untreated or undertreated psychiatric comorbidities when they come into care. In order to do this, we have both the real time face-to-face visits, but also have built out these other ways of communicating with patients, clearly texting, phone calls, things like this. Some of the things are easily billable to insurance plans. We're trying to get all of our patients, they're probably 80% Medicaid patients, get in network with their plans to minimize the out-of-pocket costs to the patients. But clearly some of these things on the lists are not traditional services that can be billed to insurance plans. So when we think about value-based care and alternative payment models, how could we pay for this? So CMS pre-COVID, Ophelia was also conceived pre-COVID, so we were anticipating needing to have the first visit in person. But thereafter, having a remote-only virtual first treatment platform. Before COVID, CMS, and I don't think this got a lot of attention at the time, came out with a bundled rate for Medicare plans. So this would be along the lines of $300 in the first month, the G-2086, in terms of specifically a bundle for telehealth delivery of MOUD to patients. Medicare tends to be, you can think of it like the pace car for other insurance plans. So Medicaid and commercial plans tend to structure their benefits off of Medicare. The commercial plans will usually pay at a premium to the Medicare reimbursement rate. Medicaid usually pays a little bit of a discount to the Medicare reimbursement rate. But if CMS comes out with billing codes for Medicare, usually those then become adopted by all the commercial plans, and then it's much more mixed but by the Medicaid plans as well. So this bundle's very appealing, because this says, we're gonna give your provider network the $300 or so a month to provide services. There is, you can see here, the list of things that they want you to have available for patients. But there's much more flexibility, so that you're not just billing a 908, a 99214, for instance, for a 20-minute visit with a prescriber. You're also able to build in all these other supportive services. When you look at the evidence base for what really helps patients with OUD in terms of treatment, it's medication. I think Dave made the point that the medication effect size for methadone or buprenorphine is so large, it then gets really hard to move the needle further with additional services. Along those lines, though, and I've performed some of these studies with colleagues at Columbia, NYU, and Rutgers, when you look at certainly requiring, but even offering psychotherapy with a licensed psychotherapist, the outcomes don't necessarily improve beyond how they've already improved. And we think yesterday they showed very convincingly a synopsis of the literature about 80% reduction in mortality, 80% reduction in overdose while patients are on medication. It's a lot harder to do even better than that. We feel very strongly, though, that the evidence is not there for requiring psychotherapy or even having psychotherapy available beyond having care managers and care coordinators and having that rapport and alliance. Our patients love coming in. We're moving in the direction of becoming a health home for individuals with OUD because we've now been treating patients continuously for two and a half years. Some of them were care transfers, so it was just easier for them. Rather than driving back and forth across the county to a clinic every week, they can see our clinicians remotely from the parking lot on a lunch break or in the evenings. We have a lot of evening and weekend hours. They wanna stay with us, even if they've been in long-term recovery, haven't used opioids in years or decades, taper off of buprenorphine, they have this attachment to the care team. And I think that speaks to the importance of the human connection, which clearly is feasible in a remote care environment. That being said, just because these bundles exist doesn't mean that all plans will pay for them, especially Medicaid plans. So an alternative is the collaborative care model. And if you're not familiar with the AIMS Center at Washington University, I would encourage you to go to their website, the AIMS, A-I-M-S. Collaborative care basically was designed for primary care settings. So it intentionally was not designed for BH specialty care settings. But when you think about who's prescribing buprenorphine in the United States, just like SSRIs for depression, about 80% of it is gonna be in primary care settings or in general care practice settings. So the idea is that you have these primary care physicians or nurse practitioners who are seeing patients for a whole array of conditions. They don't have the BH expertise. And the idea with collaborative care is that you have the primary care prescriber who's working with a behavioral care manager who's in communication with a psychiatric specialist, either a psychiatrist or a psych NP. And that specialist is a remote sort of a peripheral role who doesn't engage directly with the patient but can provide counseling to the providers to try to optimize the patient care and the response to treatment. Part of what's baked into this is measurement-based care and then having a registry to track the patients over time. Satisfying collaborative care model billing codes allows for programs to bill for this additional monthly fee on top of the E&M codes. So it's a workaround, which is not always allowable. It depends on the plan and the state in terms of what's required. But a workaround for having services covered that otherwise wouldn't necessarily be billable to plans. So with Ophelia, this is just an example. And on the way out, I do have handouts. This is a journal article in JGIM that just came out last week that's reflected here. But we looked at our retention for buprenorphine patients compared to patients receiving buprenorphine with commercial insurance or Medicaid in the US in the figure on the left. And we looked at our geographic. These are the states of New York and Pennsylvania. It's hard to see the state borders. But we're showing where our patients are located, which are disproportionately in rural areas compared to where ex-waivered providers are in the SAMHSA treatment locator. And just visually, it's very compelling how there's a lot of geographic reach through telehealth, which we just can't have with brick-and-mortar treatment settings. And our retention is about double what you see for usual care multi-site populations. So we think that this is a scalable, effective model. We're now in about 14 states after starting in these two. But there are a lot of headwinds. And Jeff was pointing out Medicaid plans, just like commercial plans, tend to have a BH payer office separate from a medical services payer office. And I'll tell you from first-hand experience, often the plans aren't even sure which of their offices is supposed to be writing the contract to pay for your care services. So it just takes a lot of time and effort and persistence. But we're very slowly turning this ship towards more of a value-based care approach. Thank you. Thank you. All right, I'm gonna spend the next six and a half minutes telling you about our SUD bundle at Vanderbilt. So the SUD bundle is part of a suite of bundles that we've been offering since 2020, again, based under our population health umbrella. And what we do with all of our bundles is we look at three main aspects, clinical, financial, and service aspects, and really try to think how we can improve upon each one. So in terms of the clinical arena, we, again, focus on an episode of care. That'll make sense in just a bit once I tell you more about the SUD bundle. But we work with the clinical teams directly to try to figure out what's working well for them clinically, what they wish they could do, they can't do given their current reimbursement practices. And we really try to build all of those components into our bundle as much as we can. Financially, the bundle is a fixed price for a duration of time for an episode of care, and we cover everything related to that episode. And we, again, also focus on the service experience. So how is the patient experiencing their care, and what can we do to make it better? We have a team of patient navigators to help patients through that. And I will say that all of our episodes, they are offered at zero out-of-pocket cost for the patient, and that is part of our recruitment mechanism, I'll say. So that's kind of the 20,000-foot view of what an episode is, what a bundle is. I mentioned there are several benefits to the patient, including no surprise billing. They shouldn't be getting a bill for any of their services. Our team of patient navigators, they help schedule, et cetera. And again, this is just 20,000-foot view. I'm gonna go into a little bit more detail here. I mentioned that we have a suite of bundles now. We started in 2020 with a maternity bundle. So that covers everything from the first prenatal visit through three months after delivery. Again, that's a fixed cost to the employer. We cover all the related services in that episode. We have several other episodes that are more traditional based around a surgery. And then we have some condition-based bundles. So we have an osteoarthritis bundle, a shoulder pain bundle. Those last for a year or nine months, and they include all the non-operative services related to osteoarthritis or shoulder pain, but they also include surgery if you need it during that period of time. And so our bundle price is fixed at a percentage of surgery. And then starting in the new year, we'll be offering the SUD bundle together with a kidney stone bundle. So you might be wondering, who are these bundles for? Right now, I'll preface that. You saw this pie chart in Robin's talk. And who we're offering it to is the big chunk, the employer-sponsored health plans. The light blue chunk are commercial plans, and two-thirds of commercial insurance is employer-sponsored. And these are, sorry, two-thirds are self-funded, meaning the employer pools their own money to pay for their employees' services. And the difference between that and a fully-insured plan is that the employer in a fully-insured plan would pay a premium to an insurer to cover their employees. In a self-funded plan, you pool your own money together. And the big part of that is that the employer assumes the risk. So if at the end of the year, your costs are lower than expected, you keep that money versus if you're paying premiums to an insurance company, your premiums are your premiums, and you have little control over cost savings at the end of the day. So because they have more risk, they're also more invested in that spend. And why are they so interested? Well, it's actually, in addition to cost savings, it's also a benefit to their company and their company's productivity. There's turnover and replacement costs that they save when people get treated for substance use disorders, lost time from days off work, and again, healthcare costs. So employers are very interested in helping their employees recover from their substance use disorders. So a little bit about the substance use disorder specifically. It's really designed to incentivize in-network high-value SUD treatment. So in our relationship with some of the employers that we're working with, they've really emphasized to us that they spend a good portion of their out-of-network spend on residential treatment programs of varying effectiveness. And so in this way, we're really trying to incentivize high-value, high-quality treatment. And again, as we've all mentioned, there's a growing appreciation that overall cost of care to the health plan is being driven by behavioral health factors. We wanna, again, align incentives and align risk as best we can. And so ED visits, inpatient admissions, et cetera, are all included in this year-long bundle. And the facilitated access piece is that, again, our bundle's available for zero out-of-pocket cost to the patient, and the navigation and the access are tied in because the employees, the patients, are now going to be able to call a number to get into the bundle and to get into treatment, which has not been the access model so far for this population of patients. I know you can't read all the detail here. This is sort of our patient journey flow. But really, it's here to kind of illustrate that there are different journeys that a patient may take in this bundle in this year-long period. And the way that we've been able to mitigate our risk is to assign different tiers, which correspond to different prices of the bundle. And so we don't exclude anybody from the bundle. You can have comorbidities and come in. You can start as an inpatient and come in. And we assign the tier based on what level of care you start in. And this was designed together with an actuarial firm who analyzes benchmark commercial claims. And so we discovered that based on where you start your treatment journey dictates a lot of the downstream costs. And so that's how we deal with kind of the high-risk, high-utilization population. I will say, so at Vanderbilt, there's a comprehensive SUD care. However, there's this chunk in the middle on the continuum that we just don't provide. So we don't have residential treatment at Vanderbilt, for example. And so these services, if they're deemed necessary, would be provided elsewhere. And those costs are not included in how we priced the bundle. So this may change someday, but right now you maintain the relationship with the treatment team and then have those services elsewhere and then sort of come back into the fold for care at Vanderbilt. So again, a lot of detail here, but I just wanted to share this slide to say that we include pretty much all services related to SUD care, including inpatient visits, ED visits, et cetera. And the services that we don't provide are medical complications related to SUD use. For example, if you were admitted to the ICU for delirium tremens, for example, that would not be included. But we really seek to include all behavioral health related services, but we do also include labs and certain injection, in-clinic injection medications, for example. And this slide, again, a lot of detail, but just wanna go into a little bit of the operations of how the bundle works. Essentially, a patient would enroll, their clinical journey would begin. All of the claims are submitted to our own administrator who separates the claims into either being in the bundle or out of the bundle. If they're out of the bundle, they go back to the regular insurance coverage. All the bundled claims get bundled together and then the employers are invoiced that one single price. At the end of the 12 months, the bundle ends and the patient continues on their treatment journey. So that was a lot, but I'm gonna pass it over to Dave to just talk a little bit about the care provided. Thank you, Sabrina. So this is hard work and there's a lot of different parts to it as you're hearing. So Sabrina just talked about what it's like when you have a population health department in a large medical center health system that can work with a clinical delivery team to build a new payment model, to improve quality, to define metrics. And one of the assumptions there is that a given episode of care has comprehensive services available. And we know that available services vary quite a bit from medical center to medical center. We spent the three or four years prior to being approached by population health expanding our continuum of care and addiction. And that's just what this slide is showing. If you think back to the Alliance for Addiction Payment Reform, they talk about the importance of having a network, but also having a comprehensive care plan at the outset. And so by having these services linked across Vanderbilt University Medical Center, we're able to develop a care plan with patients and follow them longitudinally in an individualized manner. And I think in a lot of ways that laid the foundation for us to be able to partner with the bundles team. So that's what I wanted to share here. And of course, as Sabrina shared, we don't currently have residential care or partial, but we're able to do a lot between ED-based integrated providers, inpatient services that are focused on co-occurring disorders, a transitional bridge clinic from the hospital in ED and then longitudinal clinics that are team-based. Of course, the longitudinal clinic is the final common pathway for the patients in the bundle and really ideally for most of the patients entering our services who want to stay at Vanderbilt. And we've done a lot to work with not only different payers, but also SOR funding to be able to include uninsured patients, which are considerable in volume in Tennessee, given our lack of Medicaid expansion. And so a little, one slide about the longitudinal clinic, which is that we do a psychiatric evaluation with all patients upon entry, even if they've already been evaluated and seen in our co-occurring unit. And then we offer a number of different services like Robin talked about with Ophelia, including comprehensive case management and recovery coaching. But our goal is to really be able to stabilize the most complex patients during the first year of care. And that's sort of our mission as a group. We know that in Middle Tennessee and really throughout the region, if a patient wants co-occurring treatment for psychiatric disorders along with MOUD, there are very few options. And so we really want to pride ourselves on taking care of more complex patients. So this bundle will actually allow us to, along with some of our other payments, we have a Medicaid bundled payment as well. It's actually more of a per diem like what Robin talked about with Ophelia. But all in all, it's going to allow us hopefully to start to reward teamwork as opposed to individual provider volume that was largely directed at physicians and nurse practitioners in the past. And in fact, we actually just launched our first team-based incentive payment this past academic year where we had a team-based quality goal and incentive payment and then individual provider goals based on hitting minimum encounter targets. So I want to take a step back now because this is my last slide. Before we go into the final learning objective, is just to encourage everybody to sort of step back and think about what you've seen. Because we've had three very different case examples. And keep in mind, again, those five points around the circle of the Alliance for Addiction Payment Reform. You have the network that you need to establish. You have the quality metrics. You have the care plan and the care team. And of course, you also have the payment model. So those are kind of the five points. And each of our speakers and case examples has come at this from a very different constituency. So if we think about Dr. DeVito's example at the beginning, this is really showing how the payer can drive expansion of an addiction care network. And in the case of Medicaid, very intentionally work with the counties to do that. And then the work cut out for them, of course, is they also get to define the quality metrics. But it's a lot of work just to build the care system because they're not necessarily the deliverer of care. So one constituency that can drive on this is the public payers. In each case, it requires a lot of collaboration. And so they have their work cut out for them. And in the case example that Robin gave with Ophelia, they have the ability as the care deliverer to really define the quality metrics that they want to measure internally, but they don't get to define the quality metrics that the payers are looking at. They get to do a lot with the care team and the care process. And at the same time, they're limited on how they can sort of set the standard for what is being paid for. Would you agree with that, Robin? I'm trying to give sort of an on-the-fly summary because I think it's a lot to digest. And then finally, in the case example that Sabrina gave at Vanderbilt University Medical Center, we've actually had the opportunity as an employer health plan to be the guinea pig for our own delivery systems, our own care delivery in terms of actually trying to build a bundled payment system. And so Population Health is a department that wants to market our own bundles, our own delivery system to other payers, but they have the good fortune to be able to start with our own employer health plan, which is self-funded. And so they know that they automatically have a buyer for the product and that if they do a good job, there are gonna be other employer health plans that wanna buy it. And in fact, how many employer health plans have bought some of the early bundles? Maybe half a dozen now. So it's a really exciting model for hitting a lot of the bubbles around the Alliance for Alternative Payment Reforms concept model. And it's just, it's a lot to think about. So let's take a pause there and see if we wanna, if we have time to break into small groups. We have about 15 minutes. Do you guys wanna do that or do we wanna take a question sort of as a panel from the group? See if there are questions, good idea. I can just talk really loud. There it is, there it is. Now I had a question about your Vanderbilt model because I don't see MAT listed anywhere other than your treatment for pregnant women. And so if you're putting out a bundle, that is an SUD bundle, why isn't the focus more aligned with something like Ophelia? Aligned with something more like? Like the Ophelia product, which is saying we're going to try to enhance MOUD treatment because that's the treatment for OUD. Yeah. I guess you're trying to do all SUDs. No, no. I'm going to interrupt you there because there's some things that are assumed that we didn't put on the slide. So MOUD really is at the core of what we're doing. Okay, it didn't, your slides didn't show it. That's right. I just was curious. You're exactly right. But no, it is at the heart of what we're doing with the bundle. Yeah. Hey, thanks so much for the talk. This was a great overview of a lot of the work that you're doing. My name's John. I'm a medical student. I've heard talk from Don Berwick, who is formerly of the CMS and IHI. And one of the concerns that he mentioned about value-based care is, especially in private cases, when you provide this lump sum for care teams, that some of the profit incentives will lead to taking away of services, especially when it's harder to demonstrate outcomes from those interventions. So I'm wondering how you've been able to make sure we're still providing interventions that we find are helpful but may be a little bit more difficult to show reflection and outcomes. Yeah, great question. I think one concern with bundled payments always is that the incentives are going to kind of go the other way, right, that providers are going to start to capitate their care because of this lump sum payment. I think that's what you're getting at. So part of the bundle design, and part of it is just lack of time to go into it, but we set quality and cost metrics for provider groups to hit in order to access this additional, beyond their fixed payment, an additional sum of money that we call an incentive payment. And so we're able to set what those quality measures are and what those cost measures are. And so that's part of how we try to get around that. I will say that it's been a process to even establish a bundled payment in the institution, and the trickle-down of the philosophical design of a bundle has been slow, I think, to reach trickle-down from chair to department, division, individual provider. Our hope is that that day will come when the provider really is like, oh, my incentive in providing care to this patient, according to these five quality metrics is tied to my payment, right? I mean, we're all trying to do the right thing. But at the end of the day, that's what these alternative payment models are really designed to do. If I could just add one thing to that, the last example is really complex that Sabrina gave, and we went through it pretty quickly. But one of the things you're not hearing is that in order to establish the price of the bundle, and Sabrina mentioned this in passing, that their actuarial firm looks at claims that are based on traditional fee-for-service and volume, and then they say, if we took control over all of that and paid it as a bundle rate, how much money would we expect to pay? And so essentially they're kind of herding cats and saying, we're going to take over the entire payment structure, and then we're going to be in control to be able to define quality metrics later on. And if you're seeing that we haven't totally defined them yet, that's because we're in the first stage of it where maybe for the second year of the bundle, like Sabrina's saying, we'll be able to say, okay, 180 days of MOUD, we want to reward the providers for that, and we actually have the ability to do that because we're both the payer and the deliverer. Does that sound right? Awesome. Thank you. Thank you. Thank you so much for such a comprehensive overview of three distinct systems. One thing that I work for the VA, so I wanted to compare the federal funding and how you have looked, if you have looked into that. The second curiosity I have is how do you deal with the complex cases, and particularly when you're talking about the value-based care and the complex cases, the time required, the, you know, inductions that, you know, require some monitoring, and I don't know how all of that factors into this. You know, one thing I'll say, one way to frame my response to start is that I think, obviously, in the past, especially pre-COVID health emergency, we were used to providing care in person, and the more rural and frontier areas of the country have been way ahead of a place like New York City in terms of realizing that they're not going to have specialists and subspecialists in all of their EDs or in all of their hospitals. And so they've been remoting in specialists. You think about echo models for Hep C or things like this, remoting in specialists to complement the in-person care. That's been shown, especially for particular conditions, to work really well. And I think that we can flip that and also think about care that's primarily virtual-based, but then there are also services that the patient will need that need to be provided in person. And so in the past, while their health home may have been a brick-and-mortar setting and then you virtually remote in specialists to complement those services, I think it's just as feasible that you have patients who are primarily able to be cared for sustainably in a virtual first model, but then clearly they're going to have some services where they really need to go in person. Maybe these are groups for group counseling. Maybe it's in-person drug testing, at least early in care for patients who are having a harder time stabilizing their opioid use or something like that. So we work virtually, but our care coordinators in particular are often in communication with outside prescribers, outside treatment programs. We know that IOPs are often skeptical or suspicious of medication-based treatment platforms, and I think that level of suspicion is highest when those programs don't have their own ex-waivered providers to actually help their patients get on to buprenorphine. But when you look at patients who are in an IOP treatment setting, if you add in medication, they stay in the IOP twice as long. So medication is really additive to helping patients stabilize and be able to take advantage of other services. So I think that this is sort of bidirectional or synergistic in a way. We can be more creative now. I think we have a license to be more innovative now. I can add a little bit to this. I know I'm not going to be able to answer the question about the VA part of this. I'm just not familiar with that. Others might be a little bit more familiar with it. But complex patients are complex, and they're complex for different reasons. And I think that kind of acknowledging that at the outset and saying that, like, you know, there's not going to be one single solution that's going to solve all complex patients' issues. Or if you kind of split it down and you take a look at certain populations of complex patients, there are some themes that arise. And one of the themes that arises, at least in the Medicaid system, with individuals that have serious mental illness, homelessness, significant trauma, complex medical issues, complex substance use issues, one of the things that we noticed and one of the – I mean, I don't want to toot our own horn, but I really am proud of the program. The innovation that we've done in our county system, which we're now trying to expand throughout the whole health plan, is the introduction of what we call recovery coaches. We've hired nine certified drug and alcohol counselors. The certified drug and alcohol counselors in California have to complete – there are three certifying agencies. It's up to 3,000 hours' worth of training that they have to undergo and document, many of whom are in recovery, many of whom have lived experience. So it's a little different than saying it's a peer model because it's not purely peers because they are certified drug and alcohol counselors, but many of them do have peer experience. We hired them in the county as independent contractors, and this is an important point because – and I was talking with one of our other participants here the other day about this. These recovery coaches have become the case managers that you always wanted but never had because they can independently move anywhere they want because they're not tied as an employee to a specific agency. We pay them for their services as independent contractors. The benefit of this for the individuals who have these kinds of complex social issues, the homelessness, the substance use, the trauma, all these things, these are individuals that need a lot of hand-holding to get them from A to B. And the recovery coaches, for example, will say we've got Bob. Bob is well known to the community. He's been to the emergency room 1,000 times. Everyone knows all of his problems, but yet everyone keeps saying the same thing, which is like, well, Bob just needs to do this, do this, do this, do this, but Bob's not doing this to this to this. We can send out a recovery coach, and the recovery coach, it might be – it's not an instant fix. It might be a slow movement of that individual through engagement, but the recovery coaches can come at it from a different angle. They can say, hey, Bob, let me go take you out for coffee. Take Bob to coffee. I can't go out. My malpractice would jump over itself if I tried to do that. But for them, that's within their scope of work. That's what a drug and alcohol counselor can do. We've seen benefits from the recovery coach model for some of these complex clients where the connections, in particular because they might be so significantly impaired because of their SMI, can't make it from A to B, and they really need somebody that can move them physically from A to B and understand the individuals in the system that they're going to interact with. We have never been able to replicate that with a case manager that is tied to a specific organization because the case manager can coordinate care within that organization, but they can't get into the other one, and our recovery coaches are able to do that. So in terms of a model for thinking about it, and the recovery coaches, by the way, they fit into what I was talking about because they're providing what we call either case management or recovery services for people that are lower needs. We can bill for that with Medicaid, which we previously couldn't. So there's the assumption there that if you pay for it, that investment in those services will provide benefit with some of the complex clients. Is it going to help everybody? No, because people are complex for different reasons. Yeah. I just want to appreciate Jeff sharing that because I think he has the vantage point of somebody looking at large Medicaid populations, and I think that addressing high utilizers really is a particular type of value-based care that is really hard to do well because it involves collaboration with state agencies, with Medicaid. In the case of the VA, you may have patients who go outside of the VA system for care, and so you really need those collaborations. And for some of the other things we've talked about, I think it's hard for the delivery systems to fully address it. We've tried. It's hard to move the needle because of the disintegration in the system, but we can do a lot of good through building alternative payment models, even if we don't address the high utilizers. So that's just a little bit of a synthesis. Do we have one other question? Yes. I was thinking, how high should we aim in quality? So is our goal to get patients to adhere to treatment and stay in treatment, or is our goal to get from crisis stabilization to a flourishing life? And so when you get to value-based care, are we going to put a ceiling on quality, or what's our aim? Great question. I'd love to hear Robin weigh in on that because he thinks so much about quality measures. I know from talking with Robin that we're on a journey towards integrating quality of life measures, but we're early in the process. What would you say? I have a couple points of reference that I always bring up for this kind of question. I was on a Pew, Mathematica had sponsored an expert panel on developing outcome measures, patient-reported outcome measures, survey instruments through the APA, and the long story short was that they disbanded the effort because when they went out to stakeholders in the field, the providers told them this isn't feasible, I wouldn't use it, and we don't see this ever being impactful. And yet there's growing interest in being able to better characterize in a standardized way how patients do well in treatment. And I always think about the WHO definition of health, it's not just the absence of disease, and there's been this fixation on drug testing and the overuse of drug testing in the addiction treatment field, I think largely because it drives revenue, if you can bill for it. But ultimately, we want our patients to be doing much better than just having certain kinds of urine results. And at the same time, I think it's been really difficult, I think I mentioned this earlier, to scale with 10 items or 20 items of self-report. You think about 7.6 million people just with OUD in the United States, I think you're going to have 7.6 million different recovery journeys and ways that people can do really well that look totally different, because a lot of it's in relation to their baseline and the counterfactual of how much better could they be doing, we don't really know. So I think it's an important area, and we just need a ton of research. It's going to be hard to figure out. Another thing that came to my mind when you think of complex cases, I think of modularized approaches to complex cases, so divide the problems into modules and find the right person for that module or the right team for that module to help in that aspect of the patient's care. Yeah, thank you for that comment. We are at time, and so I just want to thank everybody and say that there's clearly a lot of ways to influence value-based care from different constituencies, so I hope we'll all go forth and maybe reconvene next year and talk more about it. Thank you.

value-based care

substance use disorder

SUD

New York State Medicaid program

network of providers

Ophelia Project

medication-assisted treatment

MAT

primary care settings

patient outcomes

SUD bundle

bundled payment models