 Okay, it's 930 and I'll call the Green Mountain care boards meeting of December 19th, 2022 to order today. We have the Medicare benchmark proposal presentation by our staff. Sarah Lindberg, our director of health systems finances and Lindsay kill our data analytics information chief here at the care board with that. I'll turn it over to the executive director. Thank you, Mr. Chair. I have a few scheduling announcements. First, as a reminder, we have a board meeting at 1 p.m. on Wednesday. And then later that day, we have our primary care advisory group meeting starting at 5 p.m. And in terms of public comments, we have 1 ongoing public comment that is regarding the 1 care Vermont. FY 23 budget and as I said, our board meeting on this Wednesday, the 21st, the topic is the 1 care Vermont FY 23 budget with a potential vote. We also have an ongoing public comment regarding a next all pair model with our federal partners. Please share any of those comments with us and we will share those with the agency of human services and the governor's office as they are leading the next, the negotiations on the next model. And with that, I will turn it back to you, Mr. Chair. Okay, we have the meeting minutes from last Wednesday, December 14, 2022. Is there a motion to approve the minutes from December 12th? I'm sorry, December 14th. So moved. Oh, second. And is there any more discussion? All those in favor, please say, hi. Hi. Hi. Hi. Hi. Hi, I'm in the vote unanimously carries and we'll move to the next agenda item, which is Medicare benchmark proposal and I'll turn it over to our staff, Ms. Lindberg. Good morning. My name is Sarah Lindberg. I had the finance team here at the Green Mountain Care Board. This is 1 of my older duties. So you won't hear a whole lot from me. Instead, Lindsay kill will be our primary presenter, but I'm also here. Good morning, everyone. My name is Lindsay kill and I'm 1 of the analysts on the data and analytical team here at the Green Mountain Care Board. I'm going to pull up and share my slides. So today we are talking about the 2023 Medicare benchmark recommendation. Here's an overview of the agenda. We're going to start today by discussing the recommendation that we as staff are making. Then we're going to move into the background. We're going to talk a little bit about the all payer model and how the Medicare benchmark fits into our all payer model. And then we're going to go over experience to date. We're going to talk about settlements over time. 1 care Vermont results through 2021 prior year benchmarks and outcomes. So this year, our staff is recommending using the maximum allowable trend for 1 care for months, Medicare benchmarks. There are 2 benchmarks, which we'll explain a little bit more about later, but there's 5.2% trend for non ESR D. That's end stage renal disease and a 3.9% benchmark trend for ESR D. Our estimate, our estimates suggest that these trends will allow Vermont to stay on track with its financial accountability targets and also accounts for Medicare reimbursement increases. And we'll help bolster our fragile delivery system. So with our recommendation, here's just a high level view of the pros and the cons. And we're going to go into this more in detail. The pros for not hospitals are financially fragile. This is a trend that we've observed nationally. And so we feel that this will help support them. The maximum trend will increase the amount of federal dollars available through the current all payer model agreement, and we're going to show more detail on those numbers in a moment. And also the maximum trend is estimated, again, estimated to keep the state on track for its financial targets. And the only con that staff could really think of is that the maximum trend may endanger the ability of the state to fulfill its financial targets from the all payer model agreement. But again, we're going to talk more about why we are confident that that's probably not going to happen. So first, we have some comments that we just wanted to capture that we've heard on maximizing the Medicare benchmark. These comments, most recently we've seen from the UVM health network budget narrative and also from OneCare Vermont's budget narrative. And we have citations to that here. And what we want to point out is that over time within this model from 2018 through 2022, the year that we really that we did not maximize the trend the benchmark that was available was really 2022. So you can see in the third column trend limits to date. We have non-ESRD and ESRD. The non-ESRD trend was 10.4 percent. The ESRD trend was 7.6 percent. And the board voted on the same trend for both 7.3 percent. In 2018 and 2019, the board did elect for the maximum allowable trends. And in 2020 and 2021, due to COVID-19, we actually used retrospective trends. So it's really just 2022. And what this slide is showing, I'm going to break it down for everyone. What this slide is really showing is the relationship between two really important components to the equation of money on the table for Vermont. One is the trend rate, and that is a GMCB decision. The trend rate I'm showing here on the x-axis, we're comparing 3.5 percent and 5.2 percent. Again, we are not recommending a 3.5 percent trend. It's just for illustrative purposes. We are recommending the max trend, which for non-ESRD is 5.2 percent. That's what we're showing here. The y-axis is maximum potential savings in millions. The gray line is the advance on shared savings. This year it's about 9.5 million. And what we're comparing here is the difference between the GMCB decision, so the 3.5 percent trend and the 5.2 percent trend, versus one care of Vermont's decision for their risk corridor. And so that is these two groups you see in the red, the 3 percent risk corridor and the 4 percent risk corridor. And we can see the difference between the trend, so 3.5 percent trend versus a 5.2 percent trend in both scenarios. And then we can see the difference overall between the two elected risk corridors. And the difference we're interested in is from the top of these bars in this bar chart to this advance line, because that difference is what would potentially be received at settlement. So why are we looking at the scenario between risk corridor and trend decision? And it's about leveraging federal funds. To date, the GMCB decisions, their votes on the trend, have added federal dollars for Vermont providers. You can see the benchmark decisions, advance shared savings, that's 31 million, the imputed claims for 2020 experience for that 2021 benchmark, that's 457,000, and the floor in 2018 196,000. So when we think about that 2022 trend that was set, the 7.3 percent versus the optional 10.4 percent, what we had chosen to do at that time was using actual, excuse me, we had chosen to use imputed claims for 2020's experience. Had we used actual experience and the maximum trend for the 2021 benchmark, that would have reduced ACO's benchmark by $14 million. At the time, they had elected a 2 percent risk corridor. So that would have reduced the settlement amount by 457,000. As illustrated on the previous slide, GMCB's trend decisions are constrained by the ACO's risk corridor. And we want to point out that the reduction in risk corridor has a larger impact on the opportunities for increasing federal dollars. That, again, was that difference between the top of the bar in the chart and that gray line. Just pop back there for a second. I'm a visual person. So again, the tops of these bars to this line are, there's a larger difference here with a higher risk corridor. So the ACO's, for example, the ACO's gross savings in performance year 2021 were $22.3 million. With the 2 percent risk corridor as selected, the maximum savings were 10 million thereabouts. And the missed opportunity was $12.3 million. And then we show further down the chart 3, 4, and 5 percent risk and how much money is the difference there. So we're calling the missed opportunity. The 25 million, had they elected a 5 percent risk corridor, 25 million would have been the savings and that would have been the maximum savings realized for that year. So again, it is our recommendation this year that we use the maximum allowable trend. That is 5.2 percent for non ESRD and 3.9 percent for ESRD. The request, we also request an advanced shared savings of $9,545,916 to fund the blueprint for health programs and SASH. So now I'm going to go into background and context that support those recommendations. So first, we want to talk about the all pair model agreement and the many agreements and contracts that are kind of happening within that umbrella. On the left hand side, we have the Vermont all pair ACO model agreement or the agreement, which is, and that this is the one that I'll be referring to when I say the agreement throughout the rest of this presentation. So that is an agreement between CMS and the state of Vermont. And from state of Vermont, we have the GMCB, us, the agency of human services, AHS and our governor who all signed that agreement. And then there is separately contracts with the ACO. And within those contracts are between payers, Medicare, Medicaid, commercial insurers, and providers, hospitals, independent providers, designated agencies. The agreement requires GMCB to set benchmarks for ACO's Medicare program. These benchmarks then have to be approved by CMS prior to the performance year. So I think it's helpful to remember how we measure the all pair model. There are three areas of performance that we at the GMCB monitor and report on for our federal partners. The total cost of care or TCOC is one, what I call yardstick. So is the financial yardstick by which we measure performance. There are many ways to measure financial performance. We use total cost of care. Scale or the proportion of the population in Vermont that's aligned to an ACO is the second yardstick. So that's just trying to think about how is the ACO network growing. And quality is our third yardstick. And that's really measuring the state's trajectory toward improving patients and providers' outcomes. So we have financial performance, scale performance, and quality performance, all as part of our all pair model. Where we are today deciding on the Medicare benchmark trend is sort of encouched in this financial yardstick. We want to look at this Medicare-aligned population because they're really our state's entry point to regulating healthcare for Medicare beneficiaries. And also, although the state has tried other mechanisms in the past, the Medicare benchmark is currently how we fund the Medicare piece of the blueprint for health and support services at home, SASH. And this is a pictorial representation of what was on the previous slide. So underneath all pair model financial targets, we have these three areas. We have the all-payer total cost of care per capita growth. That is 3.5% to 4.3% average from 2017 to the end of the agreement. The Medicare total cost of care per capita growth is separate from that. And Medicare is one of several payers in this model, so separate but kind of underneath that. So we want the average from 2017 to the end of the agreement to be negative 0.2 percentage points to plus 1 percentage points of the national projections. And we'll talk about those national projections momentarily. So again, what we're talking about today is the GMCB's duty to set the ACO Medicare benchmarks. And those are annual growth targets for Medicare beneficiaries attributed to the ACO. The ACO's financial targets for these other payers, Medicaid and commercial payers, those are negotiated annually with the ACO. And the GMCB addresses those by using its ACO oversight to monitor how those targets relate to that all-payer model total cost of care. Medicare relies on the GMCB to propose annual financial targets for the ACO on its behalf. So the agreement includes certain criteria the proposal must meet. And then as mentioned before, CMMI approves or may request modification of the proposal. So here's what the benchmark looks like, at least part of it. So the leftmost square, we have historical experience. We use estimated medical claim spending in the current year. So right now it's calendar year 2022 for beneficiaries who would have been attributed to the ACO in 2022 based on that 2023 ACO network. And we call that the benchmark reference population. In our current agreement, we need to use the prior year to estimate the next year. And that's the way the agreement was written. And so obviously we're still in 2022. So we do use some estimating of that historical experience to round out that year. We may think differently about that in future agreements because it does make for yet another piece of this equation that has to be estimated. But that's what we do. The next piece, this blue box here, the ACO aligned beneficiaries. This is the number of beneficiaries in the ACO that the ACO will be accounted accountable for in 2023. And then this next piece, the percent trend, this is what our decision is at the GMCB. This is really the estimated change from the 2022 benchmark population to the 2023 actual population for ACO attributed beneficiaries. And when we multiply all those things together, we get a total dollar amount, which is the benchmark. And that's the financial target for the ACO for Medicare. So per the agreement, the trend set by the GMCB must meet the certain criteria. The trend that is set must be at least .2 percentage points lower than the projected growth for Medicare fee for service nationally. So in April, around April of the preceding performance year, there's what's called the Medicare Advantage Call Letter that is released. We use that number to then get our maximum allowable trend for consideration. So for example, the 2023 trend for the non-ESRD fee for service Medicare expenditures was 5.4 percent. Therefore, Vermont's maximum allowable trend per the agreement could be 5.2 percent. So we have the trend piece covered. I'm going to talk briefly about baseline experience and we wanted to include this one care Vermont comment that we received. This was in a letter to the GMCB on December 16th. It says, we appreciate and support the GMCB staff recommendation to utilize the maximum allowable trend rate when computing the 2023 Medicare benchmark. However, we are very concerned with the calculation of the 2022 base spend. So that's that historical experience upon which that trend is applied and believe much more work and analysis should be undertaken to ensure it's sufficient. We believe there are alternatives to ensure the base is accurately set to reflect actual actual cost of care and would want to see that memorialized in writing. If the 2022 base spend numbers shared by the GMCB on December 16th 2022 are utilized, the 2023 target will be 0.1 percent higher than the 2022 target, which falls well short of the inflationary trends at a point when our network providers are facing extreme financial stress. Part of the value of the all-payer model agreement was to ensure adequate Medicare payments to Vermont providers. Undershooting on our Medicare target negates this benefit. So again, that is discussing the historical experience piece of the overall equation. So we wanted to share some more about the baseline experience. Each year the GMCB reviews the baseline estimate and works with OneCare and CMMI to ensure it's accurate. Adjustments have been made in prior years. The GMCB is coordinating a meeting between us and them to help identify the reasons for the differences between CMMI and OneCare's estimates. And historically OneCare Vermont has overestimated the final per member per month. That's the PMPM per member per month at this time of year. So in performance year 2021, the OneCare Vermont estimate was $838 to $858. And the actual ended up being $821, again per member per month. And in 2022, the estimate is $867 per member per month from OneCare Vermont. And CMMI's current estimate is around $820 per member per month. And the 2023 benchmark estimate is $829 per member per month. So while the targets between 2022 and 2023 are close, that is largely due to the trend rate, the baseline experience estimate between these years actually increased by 2.5%, which is what the table below is showing. CMMI's most recent estimate for performance year 2022 experience is around $820 per member per month, which would result in OneCare maximizing their potential savings. So we see the difference between the baselines is 2.5%. The trend, it's actually a negative 2% difference. And the target is differs by about 0.5%. So jumping back into our discussion about the different trends that we set, we do set two for one for end stage renal disease and one for non-end stage renal disease patients. So we set them separately in order to mitigate risk. There are very few beneficiaries eligible due to ESRD. However, their average expenditures are much, much higher than the remaining population. So by setting two different trends, we're able to see that bear out and again mitigate the risk that were the final impact to the benchmark. We also wanted to address what is a common point of confusion with the benchmark and the AIPBP. So Medicare offers perspective payments. Those are called all-inclusive population-based payments or the AIPBP. Very fun acronym to say fast. These payments are designed as a cash flow mechanism to provide more stability to providers during the year. Ultimately, the AIPBP is reconciled to the would have been paid amount and that's on behalf of the those attributed beneficiaries. So when we think about the total Medicare AC, the Medicare ACO total cost of care, it's a sum of those traditional fee-for-service payments and the AIPBP claims. That AIPBP amount is calculated separately and reconciled independently from this benchmark. So for example, in 2019, we have an AIPBP of $0, a little asterisk there about there was a hospital that signed up but was not eligible and that represented around $12.2 million. It was identified early, one care held onto the funds, but that's why it looks like $0 there. In 2020, it was a negative 36 million and 2021 negative 7 million. So that's the adjustment. The full settlement amount related to the benchmark was 11 million in 2019, 16 million in 2020, 10 million in 2021. And the settlement, the final settlement minus that advance from advanced shared savings was around $5 million in 2019, $8 million in 2020, and $1 million in 2021. The AIPBP was disrupted in 2020 due to the pandemic that has been relatively close since. Again, this AIPBP reconciliation is separate from the benchmark settlement. So we want to think AIPBP is cash flow and benchmark is performance. We also want to talk about the advanced shared savings. Medicare's investments in the blueprint for health programs ended in 2016 and those programs are the PCMH program, community health teams and SAASH support services at home. So the agreement included provisions to allow for the continued funding of these blueprint for health programs by Medicare. And that funding is attached to the Medicare benchmark, but does not represent performance risk. The advance that's provided with the benchmark is reconciled at settlement. So that's that amount that's always subtracted from the total settlement. Again, just looking at the performance year with the approved benchmark trend and the trend limits to date and some notes on each of those benchmark decisions. So experience to date. On this page, we wanted to show the difference between the prospectively aligned population and the population included for settlement over time. So the Vermont Medicare ACO program limits which beneficiaries are included in the final financial settlement. In order to be included, you must maintain eligibility for the entire performance year or up until you pass away or receive 50% or more of your primary care services in the ACO's service area. So what we're seeing here where in the beginning of the model 2018, there wasn't a ton of beneficiaries who were prospectively aligned and then not included for settlement. And over time that difference is getting larger. We're seeing more beneficiaries drop out of the prospectively aligned population to those who are ultimately included at the time of settlement and that is due to more Vermonters opting for Medicare Advantage plans. So they opt for these plans at some point during the year and then they're losing their eligibility for to be in the model. Here we're looking at the experience to date with those settlements. What we really want to highlight here, I'll go through all the rows and then I'll talk about the talking points. We have gross savings and losses, the cap on the savings and losses. So that's considering the ACO's elected risk corridor. What would that maximum be? The capped savings and losses. So how much was earned? Any quality adjustment where there was one? The ACO risk arrangement. So within their risk corridor, they have a percent of that risk corridor that they would take on risk. We have the adjusted cap savings and losses. And these numbers, there's an asterisk here because this here also includes a deduction for sequestration, which is part of our agreement. And the advanced shared savings amount. So this row here is the amount of money that was provided in advance for funding those blueprint programs. And then you can see the adjusted capped savings and losses minus the advanced shared savings. That's what we mean by it's attached to the benchmark, but is not a part of performance risk. It's just subtracted off at the end. We get these net settlement adjusted for advanced shared savings. So this bottom row here in bold. So the takeaway from this slide is that over the last four years of the all-payer model, Vermont providers have received $31.1 million and the ACO has netted $20 million. And our concern staff is that future models may not be as beneficial for Vermont. On this slide, we're looking at the financial target compared to the ACO's expenditures and the difference between the two. So this is the result of the risk arrangement at a very high level. And what we're trying to illustrate here with this slide is how close it's been over time, which is a good thing. That means that the estimates are accurate and the benchmark is appropriate. So we see the fee-for-service amount plus the AIPBP amount is that total Medicare ACO total cost of care. In 2021, the financial target was $492 million. The ACO expenditures in total were $479 million and that difference was $13.5 million. Here's a picture representation again of just thinking about the Medicare benchmark, performance risk, historical experience times the number of beneficiaries times the trend, and then the advanced shared savings that gets tacked on the end there. And that's the addition of funds for those programs. And the sum of them is the total benchmark. So we mentioned at the beginning of the presentation a pro to maximizing the trend. We believe is that it's an alignment with our agreement with CMMI and also is not threatening our total cost of care per member per month goals for this model. And so that is what we are trying to illustrate here. We have actuals per member per month. This is the all payer model PMPM. So this is everyone folded in Medicaid, Medicare and commercial payers. So we see the dip in 2020 and we have really good estimates for 2021 as that data is nearly complete. And we see the corridor here, the low end being the low end of our prediction being $569 per member per month in 2021 and 2022 $585 per member per month and on the high end $592 in 2021 and $599 in 2022. And those are all still well below the total target, which is $612 to $641 per member per month. So again, it is our recommendation to use this maximum allowable trend for one care Vermont Medicare benchmarks 5.2% for non ESRD and 3.9% for ESRD. And we're requesting an advanced shared savings of $9,545,916 to fund the blueprint for health programs and sash. Great. Thank you very much, Ms. Kill. I appreciate that. At this time, I'll turn it over to board members for any questions or comment. Does anyone have any? Please just go ahead. I can go ahead and jump in first if that works for folks. Please. Thank you, Lindsay. And thank you, Sarah. The Medicare ACO program is not simple. So I appreciate that you really walked us through it again this year. So I just want to make, I just want to confirm some understanding to make sure that I'm thinking about this the right way. Would you mind pulling the slides back up so we could look at the visuals? Thank you. So if we go to slide, it's either six or seven. I think let's look at seven. I think, yeah. So my takeaway from slide six through eight or nine is that basically our decision about the trend rate has between a 300,000 to, well, pretty much a $300,000 impact versus the decision about the risk quarter has closer to a $5 million impact. Am I, is my math correct? I see you said shaking your head, yes. And so if we go back to the six with the 2022, we had quite a discussion is my recollection about whether or not to go with the max versus a slightly reduced trend last year. And my recollection of the why here was that I recall looking at data that reflected Vermont's experience versus national experience to kind of look at how our experience compared to this trend rate, which was based on national experiences that, Sarah, can you, I know you did the presentation last year, but can, am I remembering that right? Yeah. Yeah. And it was a bit of an either or proposition. We could either adjust the base experience to try to address COVID or use the maximum trend. And so it turned out to be much more beneficial to kind of impute or use a counterfactual for 2020, then maximizing the trend on actuals. So that was given that we right, you know, tried to address the COVID dip in the base experience. We picked a growth rate that we thought was appropriate based on, you know, Part B and Part A premium growth, which appears to be pretty, pretty on target for this year. Great. Thank you. Thank you for refreshing my recollection there because I had remembered part of it, but not the whole story. So I appreciate that you refreshing for me. Can we go to slide 15, please? So thank you, Lindsay, for going through the calculations again. And so the one piece that I wanted to just confirm from my prior memories around this is that the, while we have these two different targets, but of course they interact, right? Because when we are calculating the all-pair total cost of care per capita growth, the Medicare experience is included in that. So the choice that we make on the, related to the Medicare total cost of care also has an impact on the total cost of care per capita. Am I remembering that right? That's correct. Yeah. Okay. Thanks. And so, and we're bound to both of these components of the agreement in the state of Vermont in terms of targets, although there's quite frankly not the risk of blowing the targets is really a corrective action plan. Is that, what? Yeah. Yeah. The other thing that happens is if we have an event that triggers CMS to intervene, they would take over setting the benchmark from the state of Vermont. So that's one consequence if we don't meet our target. Thank you. Yeah. It's just good to remember kind of what, you know, how these things all interact. Okay. And then I did want to talk a little bit about the letter that we received from One Care and that you've alluded to and quoted from in the presentation. And so, I think starting with the benchmark, first of all, we don't decide the benchmark. That is something that CMS has retained authority to decide, right? Yeah. We propose a benchmark for CMMI to approve, but we are at their mercy to set the historical experience due to clean light and all that stuff. Yeah. Yeah. Thank you. I misspoke. I meant the base experience component, but thank you. Yeah. Okay. So in terms of the base experience, CMS does that part. Our role is to decide the trend. Yeah. And we do certainly work to make sure that the base experience is correct. And, you know, it hasn't always been accurate. So we have a role to play there. Great. So it sounds like in terms of the benchmark experience concern that One Care raised in their letter, that's you're working to get CMMI and One Care together to continue that conversation. That's not a decided component yet. So it sounds like there's still conversations happening. Yeah. We're working to get a meeting on the books this week to figure out why we're not tying out. Okay. Great. And has that happened in the past, that sort of a meeting? We have had those in the past, yes. Okay. Great. And have we successfully figured out the mathematical glitches? I would say that in my experience, what is this, year five? Pretty much most years we go in with One Care thinking it's not high enough in CMMI thinking it's too high. And that's pretty standard this time of year. Okay. All right. Thanks. In terms of the other component of the letter, which is related to the blueprint and SASH dollars. So this has also been, I think, an area of confusion in the past with the hospitals feeling like they're getting delegated risk for blueprint components. And so I think, you know, from my way I've always thought about this, we're adding the money in at the beginning so that they can get the cash, which I believe comes quarterly. Am I remembering that right? Yes. That's correct. And so that brings these additional federal dollars into the state. But then because we've added it on top of the benchmark, we pull it out of the benchmark so that the assessment around settlement is really on the performance risk, which is the other components of, and maybe it would make sense to go to that slide that you made, Lindsay, that has the different components of the benchmark. And I don't know if that's an overly simplistic way of thinking about it, but that's how I've always kind of thought about it, as we're adding it in, we're taking it out, what their shared savings is then based on the performance risk. And so if we didn't have this in the benchmark, they'd still be assessed on their performance risk. Correct. So I guess I'll just make a comment, which is I'm a little confused about why one care is so concerned about the blueprint and sash component, because to me, it seems like it adds clarifications to the risk bearing entities, which are the hospitals primarily, although we do know there's some risk on primary care providers, and would sort of clarify what they're actually working towards in terms of a benchmark. And this isn't really a question for you. It's more a comment is I had been hoping when there was in the public comments, some concern expressed about this, that we would get some sort of analysis that explained the concern. And I feel like the letter that we received more asserted a concern without really providing me with enough facts to understand the basis for the concern. So what I'm hoping for is that before Wednesday, there might be some additional clarification from one care about the facts. For example, if there's some audit concern or some other accounting concern that has not been expressed, that would be helpful information to understand. And the other piece that I'll just say out loud is that we know from the hospital budget process that many, although not all hospitals, don't necessarily reserve for the delegated risk. They sort of assume, since it's up and downside, that over time, they're just going to break even. And so I'm just putting that out there because I think it's, this is sort of a different, one care is looking at it in a different way that I'm used to hearing hospitals talking about it as the risk bearing entity. And so I think some additional explanation would be helpful, including any analysis or facts that support the concern. Because my understanding is the money will still be paid quarterly under the staff recommendation. And that the, this component would be just like it is today, part of the settlement, which would for 2023 would incur would happen in mid 2024, I believe. Is that right? Usually around the fall, yeah. Okay. So the financial issue would hit in the fall of 2024 potentially, and that the, if there is one based on the performance, and the risk would still be delegated to hospitals for the performance components. And so it seems to me, like I'm just, I'm not understanding the cash flow issue that they raised because I don't see where that comes into play in 2023. I suppose there could potentially, if things really are much different than they've historically been, be an issue that hit in 2024. But the way that one care's administrative budget is typically funded is through hospital dues. And so it seems like there's a number of issues that could be considered in 2024, if things really go south. So maybe I'm not understanding the concern. But that's how I've been thinking about how this works. So if I'm wrong, I would love to understand that better. And that's really more a comment for one care, I think, than you guys, although obviously you can respond if you want to. Yeah, I think that the pressure is that another way to think about this is if they don't save 9.5 million, that they would have to pay that back. And so that is in fact added on here. But if they even don't meet the target by $1, they have to start paying it back. Right. So the mechanisms I think that we had talked about before, should that come into play in the fall of 2024, is that there are some reserves. So there's $3.9 million of reserves that the board had required. There's the $10 million line of credit, which could be accessed and guarded against any cash flow issues. Obviously then paying back the line of credit would have to be considered if it was greater than $3.9 million. But I think for what I had been thinking was that clarifying the delegation of actual risk to the hospital network made sense given quite frankly the issues that they've raised with us about it. And that if things, it seems to me quite unlikely that it would, given the past historical experience, that as long as the Medicare benchmark stuff is sorted out in an appropriate way so that the benchmark is appropriately set, that it was an unlikely occurrence that the losses would go beyond, certainly beyond the $3.9 million that we ordered and that if there was something really extraordinary that happened, that's certainly something that they could come in at that time and talk about and we could discuss how to remedy. So that's how I've been thinking about it. So I just wanted to put that out there. If there's corrections or a misunderstanding, love to hear it. I would love to hear it with data and analysis from one care if they should like, you know, it's up to them whether they provide that or not. So sorry for taking up so much time, but I was just trying to clarify all these issues in my own head. So thank you to my fellow board members for your indulgence. So Thomas Borys has his hand up, but we'll go through and finish the board questions and comments in the HCA. And then Mr. Borys, just hold your thought and if you have others that come out of this, we can, you can add them. So I'll call on you once we're done with the board member questions, okay? But thank you. Any other board member questions or comments at this time? I ask a clarifying question. Lindsay, would you be able to walk through slide eight again for me and just explain what these numbers represent? I can take a shot. Sarah, do you, do you maybe want to jump in as well or do you want me to go ahead and? Sure, I'm happy to. So over the years we've gotten a lot of feedback that the board is failing to maximize the trend on the benchmark. So I'm just, this is an attempt to lay out how board decisions have tried to leverage federal dollars to their maximum extent. And so one is the advance, which truth be told was only encoded in the agreement for 2017, but we've been able to negotiate to keep those dollars flowing since. Also again, we are the ones that worked with CMMI to impute the experience for 2020 instead of using the actual claims for 2020. And so that ended up adding an extra half a million to the settlement that went to one care. And then in the first year 2018, due to where the national projections landed, we had the ability to use the floor, which added a percentage point to the trend. And again, because it's constrained on the risk corridor that added just $100,000 to the settlement that eventually went to the ACO. So we're really trying to do two things. One is that the board, yeah, yeah. So these, yeah, I was going to say so two things. One, I don't think that we've really undermined any of the potentials. The risk corridor is just a much more powerful way to leverage federal funds than the trend decision when it comes to checks actually being sent to formal providers. Thank you. And then on slide 29, I'm just trying to figure out how to think about this net settlement adjusted for advanced shared savings line. So if we do the adjusted capped savings slash losses, the third from the bottom line, that is the savings that we through the all pair model slash model slash one care as a state have received annually on Medicare patients adjusted attributed to the model, correct? Yep, that's the actual financial settlement. Yep. Okay. And then the advanced shared savings money is negotiated with CMMI but is in advance on potential savings. Yep. That's why it's reconciled at settlement. Yep. So if you were to remove the advanced shared savings would one care have actually been able to receive more adjusted capped savings in a year like last year because it's not part of the risk. Right. The only yeah, there's a minor mathematical. So because we're adding, you know, 9 million, the corridor increases by, you know, whatever percentage. So if it's a 2% corridor, they that adds whatever 2% of 9 million is a little so yeah, they would get the savings. But yeah, but the blueprint would not be funded. Okay. So it's not like if they held the risk for the advanced shared savings that then last year they could have potentially gotten 18.7 million, they would still get 10 million in that line. Yep. It wouldn't be added to those two. Okay. Yeah, exactly. Yep. And then the last line is essentially. Okay. What it is the difference between the two above it. Got it. Yep. So when I'm thinking of yep, I'll leave this for Wednesday's hearing then. Thank you. I brought and I really actually appreciate your comments and talking through how you understand that was quite helpful to me. So thanks. That's all for now. Actually just a quick question of all this slide is up that second line, the cap on savings and losses that's determined by the ACO's decision about the risk corridor, right? Correct. And everything flows down from there. So that number could be quite a bit higher if they would choose a wider risk corridor. And then it would flow all the way down the spreadsheet to the bottom and be a much larger number. Correct. Okay. And one more question. Could you just put up slide 20 for a second? I just want to do we know what one care Vermont's estimate for 2023 is? No, I mean, that's not as easy for them to predict because they have a new organization in St. J participating and they don't get to see those claims until they sign the participation agreement. So, so they're probably going to guess that it's not a lot of lives, but they might be a little bit more expensive. We also have the ongoing Medicare Advantage penetration rate. So yeah, my guess is they would think it'd be a little bit above the 867 they're predicting for 2022. Okay. And the other question I had is when it says CMMI and it says say for example, 821 and that's actual is that was that 821 also would seem I guess I'm trying to understand what did one care Vermont estimate at that at this time this, you know, in 2021 for performance year 2021, what did CMMI estimate at this time? And then what was the actual that second column that CMMI is that what they estimated or is that what the actual wound up being? That is what the actual winded up being. I can check my notes about what they were predicting. Well, they wouldn't, they would be predicting that. Yeah, I can go look that up. Okay, I'm just trying to understand that, you know, I understand there's a discrepancy this time of year. Oh, yeah, they didn't for 21 because it was retro. That's why they didn't. Oh, that's right. It was going to be retro. Yeah, that's why. Okay. That makes sense to me. Thank you. Okay. Okay. Thank you very much. I appreciate all of this. And I also just want to reiterate the point earlier, I think that Robin made that we don't actually set the baseline experience. I recognize that we have to do a, so to speak, a sniff test on that. But the reality is that we don't set the baseline that is done by CMMI and their consultants. What we actually have leverage and decision over is the trend rate and and you're proposing the maximum trend rate this year. Okay. Thank you very much. Just a quick couple of comments and like, well, thank you, Robin, for helping deepen my understanding. And thank you, Lindsay and Sarah. This has been very helpful. This is my first go around with ACO review as part of the board. This is the most complicated ACO agreement that I've come across having experienced a handful in different parts of the country. What I want to share to check my thinking in other conversations with other organizations, when we have thought about maximizing the federal dollars that would flow to the organization, how wide we set the risk corridor, I have come to see as a bet on our, meaning an ACO, a bet on ourselves that if we are sure we can meet the quality standards and we are sure that we can meet the dollar amount, the higher, the wider our corridor, the more money we can earn. So it's basically a bet whether we can perform or not. Because if we set a wider corridor, we get more money in return. And so the past couple of months learning about these things here with this particular ACO has caused some self-doubt for me. Because they're not looking at it that way, but at least that I can tell. It appears they either don't believe they can achieve their goals, or they don't understand how to maximize the return for federal dollars. Either is concerning. So if I'm wrong about that, I'm happy to have anybody tell me. But that's how I've come to understand how ACOs work. That I'll turn it back to the chair or whoever wants to go next. Thank you. And I told Mr. Boris to hold a second. So I'll just do HCA, Mr. Boris, and then I'll turn to you and then we'll go to public comment. Before we do that, could I just jump in for one minute? So, Tom, I would say the third possibility is that they have a risk averse provider network that is not confident taking on the risk. Thank you, Robin. Great. Health care advocate, do you have any questions or comments? Thank you. Just a brief comment. First off, I appreciate this is complicated stuff, and I appreciate the board for its work and its presentation. I thought it was presented carefully and well this morning that made it easier to follow along. I also want to appreciate Board Member Lunge's question at the beginning. We landed at very much the same place in our discussions this morning. What is it that one care is pushing back on? And if they have data to provide, why have we, you know, we'd be interested in seeing it? Beyond the answers to those questions, of course, we're interested in the answer to those questions, but beyond that, just a clear statement that the HCA supports the board staff recommendation. Thank you, Mr. Chair. I thank you very much, Mr. Fisher. Mr. Borey, if you'd still like to speak, the floor is yours, and then I'll turn to public comment if you'd still have something to contribute. Thank you. I'd love to say a few things. Sorry for raising my hand at the inappropriate moment. Sorry to cut you off. I just wanted to set it out, but go ahead. No problem at all. All right. Hey, everyone. So a couple of different points I'd just like to make throughout the presentation. Chairperson Lunge, what you said was exactly right. If I now analyze a target, and even I feel like it's a very sufficient target or one that we can beat, we are up against the risk tolerance of providers out there in Vermont. They make these decisions relative to their own financial health, their own balance sheets, and often what I hear is if we have a rough year, we essentially lose our margin for that particular performance here, and with a number of critical access hospitals in the network, I think that's a tough position to put them in. So, sure, we could, we have one care could push, and we have for expanded risk corridors in the past, but we do run up against that conceptual barrier with our network of providers. Just a quick point of clarity. Chairperson Merman, you asked about the shared savings dynamic in the advance. One of the dynamics that's come into play in the past is that we do well relative to the risk corridor. It does have the pad for the advanced shared savings to fund the blueprint on it, but when that shared savings comes in, a significant percentage of those savings are obligated to fund those blueprint expenses, which leaves less for the providers. And this is also compounded by the narrow risk corridors, but just to explain that dynamic a little bit on the upside, it does limit how much shared savings potential there is because a big chunk that nine and a half million has to come off to fund the blueprint upfront. I also appreciate the staff's recommendation for the maximum trend. I think that makes a good sense at this particular time. I want to say that publicly. Risk asymmetry. So, Chairperson Lons, you asked about the blueprint dynamic. I think the critical, and I'll try to explain this in some submissions or more of a visual representation, perhaps, but the dynamic that's really important is that the refund of advanced shared savings, if this does happen, is before any risk corridor limitations. It's an order of operations factor in settlement. So in the worst case scenario, we have our target, we have the add-on, recognizing that's there. If the healthcare expenditures are such that we reach the bottom of the risk order, the worst case outcome, that's about a 4.6% deviation from target. You have to eat in the whole blueprint, nine million, plus the 3% corridor. In that scenario, we first refund the blueprint advanced shared savings, the 9.5 million, and then have a 3% risk corridor on the whole program. So effectively, the downside exposure is the 3% corridor plus the advanced shared savings. That's the asymmetry that comes into play. When you compound it with the dynamic up top with some of the shared savings being consumed for the blueprint, it becomes very asymmetrical graphically and even on the maximum risk limit contracts amendments that we have with our providers. It shows that asymmetry in place. Happy to discuss that more. Last is around the 2022 base PMPM upon which that trend is applied. It's actually this slide on the screen here. First, the 2021 isn't exactly an apples to apples comparison. The one care estimate was pre-COVID exclusion. So it's not exactly the same as what was in that rightmost column for CMMI. But really, the issue I think at play here isn't who's right, who's wrong, good data or bad data. It's being able to reconcile between data sources. We receive a claims feed from Medicare and we receive summary reports from Lewin that corroborate the 867. We have some adjustments in there for QEMS, other estimates factors that come into play. But the data that we use ties and aligns with our regular reporting throughout the 2022 performance year. It is possible that the actual 2023 population is much healthier and therefore we would expect or their actual 2022 experience was $820 PMPM. But we don't have any information to tie those together. In other words, there's no data that we have that suggests 2023's population is significantly less cost. And in 2022, we're running very close to target. So we don't think that we're in a situation where the actual spend coming in this year, this calendar year, is materially less than the target we have in 2022, which for reference is about 870 PMPM. So that that's where the challenge lies. It's not so much in again, who's right or wrong, but it's being able to connect the data sources so that I have something I can say to our network that shows here's our actual 2022 experience. Here's the 2023 population. Here's their morbidity, risk scores, all those factors and can support the target. So I look forward to continuing to work with Sarah and Lindsay and the rest of the staff and NCMMI directly if possible to help understand why these two data sources are so different. Because without that, I'm stuck in a position where I can't recommend the target to our network. Thank you so much. Thank you very much for your comment and appreciate it. Next we'll turn to public comment in the first hand raised is Ham Davis. Mr. Davis, please go ahead. Thank you, Mr. Chairman. Just two or three comments here. One is, I think it's in my experience, this is by far the best performance that I've seen of the financial staff of the Green Mountain Care Board. I think this, in my judgment, this is the first time that we've been able to really see in detail the whole landscape. I don't mean to suggest that I have fully grasped it completely, but I can see I think that it is all there. And that's a major accomplishment and it augurs something for the future as far as the ability of everybody to figure out how to manage this situation. Number two, I think that it's very, it's critical to remember when we look at this whole situation, the whole landscape, is that we really have two businesses. We have really two systems here, not one. We have the UVM network, which has a completely different business model from the other 11 hospitals. In effect, the UVMMC has such superior cost performance and the only reason for that is that they have, that the incentives for their doctors is completely different given the kinds of contracts they have, whereas the non-network, the 40% of the system, the other 11 hospitals, has a essentially totally fee-for-service. So they have completely different, they have different business models and I think it's, so what you make out of it, what you, once you, Sarah's people have lay out the entire landscape, then the question really comes, that is going to devolve to not only what one care does with this, but what the Green Mountain Care Board does with the whole situation. The final thing I would say is that a huge, what I think is one of the biggest difficulties that I've seen watching this system really since 1980, is that it's very hard, very hard to connect the kind of abstruse calculations that you go through, the really talking about how to manage big databases and the ability in big databases to distinguish signals from noise is one of the most complicated problems that we have in public policy, not only in healthcare, but in all public policy. But what I would assert is, and I would just bet, that given how complicated this is and how it just is just difficult, that the real judgment about what's going on in the system, do we have confidence, somebody said, do we have confidence, is there confidence in the system it can perform. The people, the real judgments about performance in this system are going to be made by doctors. It's not that the big, big foot administrators don't have any effect of the CEO or the CFO, it's not they don't have any effect, but what the effect they do have is the pressure they put on their own doctors. All the medical care, okay, it's decided basically by doctors, anybody that doesn't get that doesn't get it. And I can, I just, I've been in every one of those hospitals, I've talked to doctors in every one of them, and I can just tell there's no way, no way that a given doctor coming into work on a Tuesday morning or what is today, Monday morning, okay, and they've got a room full of patients and they just start working and they hope they get through it by the end of the day. And there's no way anybody can connect these kinds of targets and this and that and percentage of this and percentage of that. Those, those, those, there's no, there's just simply no connection there, so that what you're really doing is trying to manipulate big complex databases without any effect to get out what the real drivers of the performance are. Thank you. Thank you very much for those insights, Mr. Davis. The next hand raised, sorry, Member Walsh, did you have your hand raised? I did. I just want to point out briefly that the evidence is pretty clear that the highest performing ACOs across the country are those run by healthcare providers. So they are able to do their job taking care of patients and understand these contracts and fulfill them successfully. That's pretty clear. Great. Thank you. And for Robert Hoffman, Mr. Hoffman, please go ahead. Thank you, Chair. Can you hear me? Loud and clear. Yes. How are you? Good. Thank you. Good. I think Member Walsh's final comment is the important one here. And this distinction between risk aversion and confidence in ability to execute is really just semantics. Providers, given the right tools, we've been shown clearly in other models, provider led ACOs can make an impact. They need the tools. They deferred to the state's preference for a large centrally controlled ACO and provided them tens of millions of dollars for the infrastructure rather than supporting the individual participants. So the ACO as structured in this state has the horsepower and should be providing the horsepower down to their providers to execute. So confidence is the same as risk appetite. They don't have the risk appetite because as the NORC report showed, less than 10% of them are even using the analytics provided because it's not helpful data. Secondly, as the NORC report suggested, there is a non-existent capacity of clinical information for them to actually execute on the population health efforts that they would normally in another otherwise constituted ACO be able to perform. So it's unsurprising that the risk aversion is there because they're reporting back to their ACO. We don't really feel the ability to affect the change that normally we would. So the great part for the executives today who have basically threatened the state to nuke this entire system if they don't get it the way they want it, their salaries and your irrespective of what the risk is for the network. They take full bonuses in 22 because they satisfy their own internally structured guidelines for how they get their bonuses and their system can continue to defer or decline to take risk with no impact to their bottom line. This is not in any way a line the way that an ACO should be and I think the public is really looking to this administration to this board right now. This is an inflection point. Are you going to negotiate with hostage takers who enumerate every last benefit that they provide to the system that unfortunately because it's too big to fail now and you can't find another way to disperse these payments out to the communities that desperately need them. If you don't give them what they want what they've written this morning is they'll threaten to take it all away and not function as the conduit the state needs to disperse those funds. We need to see real leadership here and see pushback on that because you cannot be allowed to become a passive captive agency because something has due to architectural error become too big to fail in the state. Thank you. Thank you for your comment. The next hand raised was Michael Del Treco. Mr. Del Treco, how are you? Please go ahead. Good. Hey, thank you for allowing me to speak today. My slide numbers might be off. I think I was using an earlier slide deck so apologize if I get some slide numbers wrong but I wanted to go back to the benchmark component slide for a second. I don't know what slide number that is and then move to I think what is slide 32. I think it was 17, Lindsay, Mr. Del Treco. Yeah, that's great. I think I'll steal maybe a couple hand Davis's words and signals and noise. It's so important to get this calculation right. And when I when I heard the discussion around targets, historical experience, the beneficiaries, the trend and where we are and how we've performed, I landed on slide 32. So if we could go there quickly and no matter whose projection is accurate or not, it seems to me that we have about somewhere in the neighborhood of $40 per member per month and this total cost of care that we could that we could bring into the state of Vermont to help alleviate some of our financial challenges. And that $40 on me, they're taking, you know, if you use $569 or $592 to the min or the max of the total cost of care target, you get to about $40. So interestingly enough, on the slide where we had the OCV projection versus the actual, that's about $20 to $40 off too. So again, I don't know who's right or wrong. And if our goal is to maximize the federal dollar and this total cost of care methodology in the all pair model, wouldn't we be wanting to hit the 612, 641 range as opposed to, you know, 60570 and 60585? And what are the opportunities there, forgetting about who's doing the calculation? How do we maximize this thing? And it's more than just the max trend. And I do appreciate the recommendation to use them, the max trend. So I'll stop there. But I think maybe you get what I'm trying to get at. And if you don't, please let me know. Yes, thank you very much for your comment. Mr. Davis, did you raise your hand again? Do you have another thing? I couldn't tell if it was left off a few additional comments. Yes, thank you. I'd like to comment about what Mr. Walsh has to say about other ACOs. Here's why I think that it's important to remember, at least in my judgment, is that most of the really, there's a lot of ACOs out there that are just doctors, not hospitals. And so it's hard to make a judgment about that. Number two, okay, is that number two is that if you want to find out, if you look at the, if you look at the actual data that exists in your database, which never gets mentioned here, I would just mention these numbers. The PQI, for the, for UVMMC is 15, is 5.9. The PQI, for the lowest average is 13, okay. There are at least three hospitals in Vermont that are higher than 13, okay. And the average, okay, is at least twice UVMMC. The reason for that is that the UVMMC has a completely different business model and a completely different relationship towards doctors. If you take the UVMMC and, which is 60% of the system and you apply that 60% of the system, if you extended it to the whole system, then the Vermont ACO, which is a broad ACO, it has designated agencies, it has primary care, it has all, it has 13 out of the 14 hospitals, it has a lot of stuff, that that, that the UVMM, that the, that the Vermont performance, if it was based on the 60%, would be the best in the United States. If he doesn't agree with that, show me your numbers. Thank you. Thank you very much. And Dr. Merman, please go ahead. And we'll, this will be the last word as we have an 11 o'clock meeting. So please, Dr., go ahead. Oh, sorry. I was just wondering if, if Lindsey or Sarah could comment on Mike D'Altrecco's comments regarding slide 32 and, you know, being able to get closer to that all-pair model to a total cost of care target. Yeah. So this is what was spent. So I hear that he's saying that there's headroom. But I guess maybe what he's saying is that providers should be doing more to add to the base. That's the only reason, way I could see this going up. I think what he's suggesting is, is could we increase our Medicare base rates such that we could get closer to that total cost of care target and bring more federal money into the state? Yeah. Yeah. So the reason it's so hard to predict with the bases is a risk protection mechanism that, that Medicare provides to the ACO where anyone who gets the majority of their primary care outside of the network is pulled out. And so that qualified evaluation and management has actually been really beneficial to the ACO. We can talk about changing that so that there's more stability in the benchmark, but it would also add more risk. And so these conversations about risk aversion and doing that have been kind of made my head spin a little bit. So that is why we don't know what the base is. But unfortunately, we can't just add more money to the base. That's based on historical experience. So I don't see a practical way to do that. Great. Thank you. And I see there's additional hands raised, Mr. Davis and Mr. Diltreco. If you could submit it in writing, that'd be appreciated. We have a hard stop for an 11 o'clock meeting we have to get to. And so I'll turn it to the board. If there's any old business to come before the board, please speak up. Any new business? And is there a motion to adjourn? So moved. Seconded. We're not going to be able to vote today. I think we're hoping to get a vote. I think we'll vote on Wednesday, Sarah. Okay. Okay. Cool. Sorry about that. Thank you. All those in favor, please say aye. Aye. Aye. And the vote carries. Thank you, everyone, for your assistance. And have a great day. Thank you.