 Good afternoon and welcome to the Green Mountain Care Board meeting. My name is Kevin Mullen, Chair of the Board, and I will call this meeting to order. The first item on the agenda is the Executive Director's Report, Susan Barrett. Thank you, Mr. Chair. A few announcements. First are ongoing public comment, we're currently accepting any public comments regarding the next potential model with CMMI, next potential all-payer model that has been on our website for well over a year now. Please submit those comments to us and we will share those with our colleagues at AHS and the Governor's Office as they are leading the negotiations on the next model. I also wanted to announce that tonight we have our primary care advisory group that starts at 5 p.m. and the login link is on our website. And then last but certainly not least, today we're going to be hearing from Sarah Lindbergh and I wanted to announce to the public that she has just been, she's just moved into a new role at the Board and she is currently now the Director of Health Systems Finance. She joined the Board in 2017 where she has been holding the role as the Director of Health Systems Data and Analytics. And before that, before joining us, she worked with the Department of Financial Regulation in Insurance Regulation and has a very broad background that I know that will serve us all well in her new role leading the finance team. So I just wanted to welcome Sarah Lindbergh to that role and announce that to the public and with that I will turn it back over to you, Chair Mullen. Thank you, Susan. The next item on the agenda are the minutes of Wednesday, May 11th. Is there a motion? So moved. Second. It's been moved and seconded to approve the minutes of Wednesday, May 11th without any additions, deletions or corrections. Is there any discussion? Hearing none, all those in favor of the motion please signify by saying aye. Aye. Aye. Aye. Any opposed, please signify by saying nay. Let the record show that the motion carried unanimously. At this point in the meeting I'm going to turn it over to the previously mentioned Sarah Lindbergh and congratulations Sarah along with Michelle Degree to talk to us about the all-payer model federal reporting updates. So I'm not sure who's going first but one of you take it away. I have the pleasure of kicking us off so let me share my screen. Are you all able to see that? We are. Yes. Okay. So as Chair Mullen mentioned we're here to just talk about the all-payer model federal reporting updates. So the reports that we've submitted so far this calendar year. There we go. So today we're going to cover the 2020 statewide health outcomes and quality of care results, the 2020 total cost of care results and the preliminary look at 2021 scale targets and alignment results. The first two reports have already been submitted to CMMI. The last report, the scale report is due at the end of June but since we have a somewhat good idea we decided we'd use this opportunity to present some high level insights for you today. One major consideration that we wanted to flag sort of that goes is a thread through all of our reporting. Of course everyone here is very well aware of is the public health emergency. You'll see this language throughout all of our public reports that we've submitted for 2020. It's likely to continue through 2021 and we'll continue to assess beyond that. But it's just highlighting that the changes are necessary and are likely to have impacted preventive care and health promotion activities and that the full effects of the public health emergency or the COVID-19 pandemic are still developing and we kind of expect them to distort some of our trends for many years. We're going to start by reviewing the 2020 health outcomes and quality of care results. So again, considerations in this report, public health emergency which we just talked over, scale growth, specification changes and population risk. So, you know, as this is the third annual report, I know a lot of folks are eager to draw year over year comparisons based on the three years of available data. However, we continue to emphasize that comparisons should be made with extreme caution, particularly in light of the public health emergency and its impact on care pattern. Changes in risk status and growth of the attributed lives over this time period should also be considered. For example, in some cases, payer-specific attribution has doubled since 2018. And again, we're talking about 2020 performance results like nowhere in 2022. So we have to think back a little bit here. Additionally, and we'll talk through these a little in a little more detail, but there have been measurement changes that will allow for very limited comparability from year over year. Specifically, there are two heating specifications that changed and there is one measure that we are proposing to change to better align with current state reporting from the health department. So we're going to go through the results starting with our population level health outcomes. From this, we can see that we are on track to meet three of the five reportable population level health outcomes targets. For the agreement, we need to be on track for it's four of six before we trigger corrective action. The deaths due to drug overdose measure, the first one here, is no longer reported in that manner by our health department. So we are suggesting a change there. The change that we're suggesting goes to specifically deaths related to opioid related fatalities. So the health department has sort of moved away from the all drug reporting and more specifically into the space of opioid related reporting. And so we want to follow suit there and a couple of things to point out on this this page and I know some folks had questions about this. So while Vermont's percent of adults with a personal care doctor or care provider decreased from the 2018 and 2019 results, it's important to note that the national average for this response is 77 percent when you combine the yes only one provider and yes more than one provider rate. So while we have started to fall, we are still above that national average. And I just wanted to slide that for folks. Additionally, diabetes prevalence nationally is at nearly 11 percent and COPD nationally is right around the 6 percent range as well. For deaths related to suicide, something that we piloted a few times. Small numbers cause a lot of noise and volatility in this rate in Vermont, specifically looking at our health care delivery system quality target. The agreement states that we need to be meeting four out of seven of these in the current language, however, that changed to the denominator of nine because of some disaggregation here. If you look, we're on track here to meet five of the eight reportable health care delivery system quality targets. Medicare cap data was not collected in 2020 due to the public health emergency, but has since resumed. So we will have this information for 2021 and beyond. I think a big thing to highlight here is that changing utilization patterns are quite evident, especially with the significant drop in the growth rate of substance abuse related ED visits from 2019 to 2020. You see that we're at a negative 16 percent there. And again, on this page specifically, we see a lot of measures that are ACO specific. So here, just remembering that we saw relatively significant increases in scale, which should be a consideration when we begin to interpret these results. For a process milestone, we're making progress towards five of the five reportable milestones, though we are proposing to change the Vermont Prescription Drug Monitoring System measure. And this page reflects those two HEDIS measure changes that we saw. The BPMS measure that we're substituting to change with, again, is just due to the way that BDH is currently reporting. And we want to follow suit and not be asking for too much additional data runs from them or through special measure specifications. So we are proposing to substitute this with a measure that's reported regularly, which is morphine milligram equivalence. There's a lot more detail in the actual written report, which is on our website. No Medicare benchmark here was reported for clinical depression and follow up, so we don't have a comparison there. And then as I mentioned earlier, there are measurement changes at the HEDIS level, which are shown here for asthma medication and Medicaid adolescents with well care visits. These are really out of our control. You know, I think folks know that part of the discussion when we were developing the agreement was to have measures that were already reported to really reduce that administrative burden at a state or national level. And when we when we use things like that, they're out of our control, right? So we have a harder time being able to ask for changes to those. And we are no longer able to report them the way that they're specified in the agreement, so I just wanted to flag that. And again, there's some more detail in the written report about how those changes impact this measure. What we've put and proposed to do is to just work with CMI to update the target rate that we would need to see here. And propose a new baseline, which will have to be the 2020 year, just given that that's the first year of the measurement change itself. So it'll take a little time, but we will start having those conversations with CMI. And with that, I am going to turn it over to Sarah to discuss total cost of care results. Sarah, just let me know when you want me to advance. Yeah, thanks. If you would please advance the slide 10. So the all pair model agreement sets two separate financial targets. One is the all pair target, and that's for all Vermont residents for whom we have claims in our all pair claims database, whether or not they're attributed to the ACO. And the target is a cumulative target through the life of the agreement, the growth rate should be between 3.5 and 4.3 percent and averaged out annually. That target was set based on different growth state product growth in Vermont, depending on whether you use a five or 10 year period. So we are trying to constrain growth within that range. For Medicare, our target is tied to projections that are made for Medicare fee for service nationally. And Vermont's growth on a per person basis is supposed to fall within the range of 0.2 percentage points below to up to 1.1 percentage point above those national projected targets. For Medicare, we are limited to those attributed to the ACO in performance years 17, 18 and 19. However, the last two years of the agreement will be the full Vermont Medicare population. So on slide 11, we see always good to put this in perspective. We see that the total cost of care really is only a proportion of the spending made on behalf of Vermont residents. So the 2020 expenditure analysis, which was recently released, showed that the resident spend was estimated to be $6.4 billion and 44 percent of that is within the total cost of care as defined by the all pair model. The expenses that are excluded have to do with pharmaceuticals, which are excluded. About half of Medicaid spending is excluded from this target as our other non-claims based payments outside of the system in things like the administrative cost of health insurance. So of that, 44 percent of our total spend, about 35 percent of that in 2020, flowed in and around one care Vermont. They were accountable for that spend, even if they weren't directly making those payments. And the other 65 percent is kind of the outstanding scale that we have to try to bring into this model. And on slide 12, we see that the all pair model agreement is increasing the state's influence on some populations. So the top line is showing the proportion of expenditures in the all pair models total cost of care that has some influence by the state's regulatory processes. And that is defined as folks that are attributed to a ACO regulated by Vermont. Those who participate or are beneficiaries to Medicaid. Those with fully insured plans in the commercial market, as well as expenditures that happen in Vermont's hospitals to Vermont residents. So in 2020, that was 74 percent of that all pair model total cost of care and 63 percent of the people. I will note these are likely a little bit underestimated as we are not pulling out the state employees or the teachers plans in this analysis. And we also can't pull out the professional expenditures associated with Vermont's hospitals very cleanly. We're working on those enhancements, but so this is kind of a lower estimate than probably is true in fact. And on slide 13, we're going to talk about the results. It's hard to call them that. The things we observed about the total cost of care in 2020. So in slide 14, so again, there are no annual targets in the all pair model agreement. I do show the annual growth by the pair type and at the end of the day, the accountability metric is the compounding growth to date for all pairs. And that is point four percentage points. So you can see that between 19 and 20, we experienced a significant decline in the total cost of care. So more than a 7 percent decline, most dramatically seen in the Medicare population as they not only are more expensive, but are the most likely to have encounters with the health care system. And I believe that it was somewhat less dramatic for the Medicaid population in part due to the fact that they had prospective payments flowing during this time period. So on slide 15, it looks like some of our squiggly lines got a little misaligned. But this is just showing that the growth rate from 2014 to the baseline year of 2017 had been 5.6 percent of that CAGR or compound annual growth rate. Through from 17 to 19, we had already seen a little bit of a decline. It had been 4.6 percent average growth to that point. However, 2020, again, not really anything about models, just the pandemic and its associated response means that that compounding growth rate has dropped to 0.4 percent. So that also telegraphs the fact that we will expect some trend lines correcting that in the next few years. So they'll be going right back up. So that shouldn't be a surprise to anyone. And then on slide 16, we're just trying to show how the expenditures on a monthly basis differ between those attributed to the ACO and those who don't. Largely driven by the Medicare population, we do see higher expenditures on a monthly basis for the ACO population. And that is because they are more likely to have contact with the health care system. They're more likely to utilize care, which is part of the reason they tend to have higher average expenditures. We also see that for similar reasons that the decline in that population was much more significant from 19 to 20. So those are people getting the types of care that were deferred in 2020. While those not attributed were less likely to have those patterns of care seeking. And on page slide 17, you can see how, again, that's the most dramatic in the Medicare population. They not only have the highest monthly expenditures but have the greatest delta between the ACO attributed population and those not attributed to the ACO much closer for the commercial and Medicaid populations. We see that the Medicaid population, in fact, looks a little bit less expensive on a monthly basis. That's largely due to the people who aren't attributed likely are seeking different types of care and therefore aren't being attributed through primary care. And then another thing to note here is that, again, harkening back to that slide where we show the proportion of spending that's in the total cost of care. We're only capturing half of the Medicaid's expenditures. So this is just the payments flowing through diva. So there's quite a bit of Medicaid spending that is not represented by this. All right. And on slide 18, we're going to transition over to the Medicare total cost of care. Again, those are tied to national projections. It's a guess about how spending will change in the upcoming year for the national Medicare beneficiary population. So we are far... We're doing a... We're quite below target for both of these populations. So ESRD stands for End-Stage Renal Disease. This is a very small population but has extremely high expenditures as they need quite intensive care to manage renal disease. So the target to date was a growth rate of 3.1% from 2017 to 2020. And Vermont's performance is 10.4 percentage points below that. For the non-ESRD population, the target is 3.8%. And we are at 4.5 percentage points below. As you can see, we've consistently been below these targets over time. So just another testament to how efficient Vermont is in its Medicare population. On slide 20, we can really hammer some of these points home. So when we look at the per beneficiary per year or the PBPY expenditures from Vermont Medicaid beneficiaries, it's consistently below national expenditures. We also see that Vermont had a bigger nose dive in expenditures from 19 to 20 than was observed nationally. The PBPY went down by one and a half percent in Vermont. It was seven and a half percent. And that is a testament to Vermont's swift and effective response to the virus. So people really stopped seeking care. And also we had really low prevalence of COVID comparatively. So the treatment expenditures are not in there. And finally, we also had a material attack to our large health network, a cyber attack that enforced operations to pause for about a month. So that also is contributing to that trend line that we see. So on slide 21, just not only is Vermont a lower cost compared to national, but if you look at it by state, it's among the lowest costs or lowest expenditures of any states for these Medicare beneficiaries. And we're not only comparing ourselves to national here, but also the other states that are engaged in all pair of models with the federal government. So you can see Maryland is among the most costly and Pennsylvania is right near national. So I just think it's helpful to remember that Vermont it's not reasonable to expect a lot of cost savings in the Medicare population and that thinking about innovation and trying to coordinate care better should be kind of maybe emphasizing other outcomes or thinking those through. And then slide 22 is going right back to Michelle. So as promised, we'll do a quick preview of the 2021 scale information. So in performance year four, Vermont achieved 54% Medicare scale performance and 46% all-pair scale performance, which means we did not achieve the scale targets that were set forth in the agreement. This is not a surprise. We have anticipated this beginning with last year's preliminary budget, just like we can see here for performance year five with the preliminary budget, we will still fail to meet those targets. As we've discussed in, oh, let me show you one more thing. Here's just another quick look of scale results. I know some folks like to see it this way a little better. So looking at scale relative to those actual targets showing the trajectory over time. And we're gonna take a quick look at attribution here. So again, we've got attribution from performance year zero in the Medicaid population all the way through performance year five, preliminary information here, just to sort of show you that growth over time. I think a couple of the bigger things in Medicare for 2021, we had a hospital join and there is a slight drop noted in the Medicaid population which we were aware of. There was a loss of two provider practices who to my knowledge have since rejoined in the ACO. So you see the population come back up in the 2022 preliminary numbers. A couple of factors that just sort of influence scale here, provider network. So again, we saw some increased participation in the commercial and the Medicaid space, payer participation, something that we noted in the report is that we're pretty near saturation in the Medicaid space and for attribution methodology, refinements and improvements through diva. So we're seeing the traditional and the expanded population. And just noting that since attribution is provider driven, there can be a slight disconnect between where people live and where they're seeking care. As noted in prior years, starting with last year's report, we did put forth a few alternative, all payer scale measures. And I'm gonna go through a couple of those with you now or Sarah and I will sort of tag team this part. The first one is adjusted scale. So again, you'll recognize this from last year. We've updated it. Adjusted scale removes populations over which the state has no data or no authority to regulate. So Sarah sort of mentioned earlier in this conversation. So in particular here, we're talking about self-funded employer plans and Medicare Advantage plans. So the table models, removing these groups from the scale denominator in an effort to show progress towards scale where the state has data or some regulatory influence. So this change in perspective, as you can see, results in a 9% increase in scale performance bringing the achieved rate up to 55%. Even still the suggested scale performance demonstrates that the scale targets are very ambitious goals and that the state would have still fallen short for all reported performance years despite these adjustments. We still think that this adjustment will continue to be a better reflection of the leverage available to the state and maybe more material as more beneficiaries start to join Medicare Advantage plans. Another look that we've started to provide is the proportion of hospital revenue. So the all pair model, as you all are well aware, it's premised on the idea that providers will alter behavior based on payment incentives. In Vermont's model, the hospitals operating in the state are taking on a lot of the risk. One method to measure how much the model is penetrating the delivery system is to measure how much hospital revenue is shifting from traditional fee for service reimbursement to value-based arrangements. An important note here is that the APM alone does not include any mandates associated with having fixed perspective payment. So any prospective payments made to hospitals at this time represent provider-based risk and therefore represent material changes to the way hospitals are reimbursed for care. The proportion of hospital revenue measure estimates how this is changing over time and different by hospitals. It's calculated through actual financial data submitted to the GMCB. The calculation here is using hospital discharge data to estimate the proportion of hospitals and that patient revenue that comes from Vermont residents as Vermont residents are the population eligible for attribution in most ACO programs. And just to note here, you can see that the proportion of revenue is growing. There is a decline in prospective payments in 2021 and this is due to the influx of COVID relief funding and not necessarily due to a decline in provider election of prospective payments. With this one, I am going to turn it over to Sarah. Sure. So the issue that we have for the Medicare scale targets is there are people who aren't eligible to be attributed to the Medicare program because they don't meet certain criteria such as having both part A and B, not having Medicare as a second payer and not having Medicare advantage. So once we make those eligibility adjustments, we see that in 2019, nearly 20,000 people in 2020, 22,000 people in 2021, 26,000 people drop off before we even get a chance to attribute them essentially. So that substantially changes our scale performance. We're still below the very ambitious targets set forth in the all-pair model agreement. But this would also be even further adjusted because some Medicare beneficiaries have primary care relationships outside of the state of Vermont. So that performance would likely improve even further if we were to kind of do mock attribution for potentially eligible or beneficiaries potentially eligible for attribution. So just things to be thinking through for future targeting and important details for the math. Thanks, Sarah. And I'll just wrap it up quickly with a few conclusions. So performance year four results show progress, but again, we ultimately fail to meet scale targets for a couple of other points here. The IIP was created in 2020 to address issues of scale participation. And then we've provided a few alternate measures of scale to sort of offer a fuller picture of that statewide scope. And we've provided a link at the end just where you can find all of our federal reports in case folks are curious. They are all listed there. And Chair Mullen, I will turn it back to you for questions. Super, thank you so much, Michelle. Thank you, Sarah. Board members, questions or comments? I have a couple of questions. Great, thank you, Tom. My first question is why were there no other questions? No. So I'm kind of looking at slides 14, 15 and 17, which were the total cost of care over the years. If we could go back to one of those at least. So this is 14, let's try 15 here. Okay, we can try this. So maybe we should go to 16 and 17. Sorry, we'll get there in a minute, right, 17. The next one. Okay, we can use this one. So looking at the kind of flat line there for Medicaid, I'm just wondering how, let's assume there's a cost shift. People, some will argue there is no cost shift, but let's assume there's a cost shift. And so the Medicaid payments are regulated by government. So when one looks at this line from 2014 to say 2019, it looks pretty flat. How would one distinguish that or assess that flatness relative to what might be a cost shift versus what might be improved performance because of the all-paramodal and the things that we're doing within the Medicaid population? Like for example, fixed prospective payments and Medicaid, a big portion of their expenditures are fixed prospective payments. But how could one tell or differentiate what's driving that relative slow growth, cost shift versus other causes? Yeah, if I were gonna tackle that question, I would definitely want to look at a bit different data. I would want to be able, this is limited in two ways. One is we only have again, half of the Medicaid expenditures are in scope for the all-paramodal. And two, we're also limited for things that we see in claims for a lot of the data we have while quite a bit of Medicaid spending flows outside of claims by design. And so doing that analysis, I think what you would need to think about the whole picture. And sometimes it's hard to compare because Medicaid covers certain services that other payers don't. So I do think untangling, we have that problem with Vermont's long history of reform as well, it's hard to determine causal effects of any one variable in those equations. So yeah, I would say this is really just a descriptive look at the difference, but only a really small piece of the puzzle I think you're talking about. So if we could go to slide 20, I think it is. So this one struck me as, and I think when we were talking about this before, it was kind of a wow, I mean, wow. And so you can see the differences and you explained a little bit the differences between the national number and the Vermont number for 2020. But this trend goes back clearly to 2012, 2017. And what, what is, I mean, and this is, let's see the difference in 2019 between those two numbers is 18%, which is a significant number. So what, why is that? What why is, especially with Vermont having, you know, the rapidly aging population with the second, I think we're the second oldest in the state. What, why are we so much less costly, total cost of care for Medicare than the rest of the nation? Yeah, so this is again kind of a descriptive look at that difference. It's not more of a inferential look to help just explain the differences, but some potential things that can explain it are efficiency in the healthcare system. That certainly seems to be in line with results from the Dartmouth Atlas that compares Medicaid expenditures by hospital referral regions where Vermont stands out as extremely efficient. It also could point to people getting a different amount of care. So the utilization differences could explain some of the differences in this among, you know, many other potential things going on. But I do think that when the model was negotiated, there was concern that while Vermont was low cost, it was growing at a higher rate than other states. And that's why the all pair model was in part negotiated to help protect against that. But we are definitely not seeing evidence of above average growth as you can see from these results. So, but we haven't done at any point in time the deeper dive into what explains this, that this is kind of just kind of a top side graphic. But there's still a lot of effort to really understand those differences that would seem to me. Yeah, and Medicare is in part based upon a regional adjustment. So, you know, part of it is probably the reimbursements are lower in Vermont than in other places. So, you know, there's a lot of a done pack. Moving parts, slide 21. And I just, what this one says to me is that, you know, Vermont is at the bottom of each one of these scales for almost every year. I mean, there's some movement here, but what this would say to me maybe is there's not much room for improvement in Vermont. That we're already, when it comes to Medicare, we're already there and people's expectations about a dramatic move would be unfounded. Yeah, I think financial savings not only due to the low per person expenditures, but also the size of the state are unlikely to be realized. And then finally, slide 29. I just always kind of want to make the point that, you know, looking at that 15.4% number or is that is a number which is a blend of Medicare, Medicaid and commercial. But they are close to each other and it's really important to understand that, I think in terms of looking at the future success of the model that 15.4% number is comprised of a Medicare number that's at 33, 34% Medicaid at 43% fixed prospective payments and commercial at three tenths of 1% based on the 2022 hospital budget presentations. And so, you know, what we don't see here but the reality is that the commercial payers aren't really at the table here as much as Medicaid and Medicare are. And I just think that we need to be giving that lack of participation, special attention because commercial payers are the biggest piece of the pie. And so I just wanted to make that point that that 15.4% is comprised of a set of numbers that diverge greatly. Yeah, it's just gonna say it's also for the whole system. There's certainly variation by hospital as well that is demonstrated. Yeah, absolutely. Thank you. Thank you, our unmute chair Mullen. Well, thank you, Sarah. I wondered why nobody was answering me. Other board members, questions or comments? They asked a couple. Thank you so much for the presentation, both of you. Really appreciate it. One quick question actually following up on one of Tom's questions. Sarah, from 2019 to 2020, obviously that chart showed that there was a pretty significant decline in total cost of care for the Medicare population. I think it was the side you just had up relative to the national average decline. If we think about that as more deferred care in Vermont relative to the rest of the nation, that slide, yeah, thank you. The 7.5 compared to the 1.6. I'm just wondering if you're seeing or you're anticipating seeing a higher growth rate than from 2020 to 2021 for the Medicare population relative to the rest of the country. If it was more, if it really was deferred care, pent up demand is gonna make up for that. Expenditures everywhere increased quite a bit from 2020 to 2021. I don't have comparative information yet, but at the end of the year, it looked like Vermont was about on par with what we saw nationally, but I'll certainly update this once we have that. Okay, and I guess with respect to the scale targets, I think we learned a lot in the first agreement about how we set those scale targets and who's reasonably included and who should not be included in that denominator. And I'm just wondering if you were to kind of summarize some of those key takeaways for us as we're thinking about another agreement and what we imagine next go around for who should be in the denominator. Given a lot of thought to that, not enough thought yet. Well, I do think that thinking about participation targets should probably follow the payment model. So if there were gonna be a payment model centered around providers, participation likely would make more sense to be measured from that lens instead of from a Vermont resident lens. So, but I do think that there was hearty negotiation on appropriate people for the denominator that was not necessarily successful in conversations. And I think that over time, both sides have kind of learned and that's part of the reason the targets aren't being enforced, that they probably aren't the fairest way to look at it. Most fair, maybe. Got it, yeah, I know I think it's important as we think about going forward, what do we learn from this? And I guess my last question really is for Michelle and I'm not sure you have the answer to it, but it just struck me as interesting why the state, the Department of Health is moving away from monitoring overdose deaths related to all drugs and moving only towards opioid related overdose. And I know you may or may not even know the answer to this, but struck me as I think it's pretty important to understand overdose deaths across all drug categories, not just opioids. So I was curious as to the motivation behind that. Sure, and I may have misspoke. So what is published now is no longer that a data sheet that used to be all drugs. I believe I could ask for a special data run and get that, but given the public health emergency and the health department's sort of involvement in that, we were not trying to ask them for additional data runs at this time. And what they publicly report now is that opioid specific sort of data run. Within that report, there are other drugs, right? So it's opioids and XYZ and there's quite a large table there of other kind of included drugs within that population. I'd be happy to reach out to them once things subside a little bit and see if they will continue to report on all drugs, but at last check, that was not something that they were planning on doing at the time. Okay, great. Thank you very much. That's all my questions. Thanks, Jess. Other board members, questions or comments? I don't have questions, but Jess's question about takeaways from scale got my wheels turning. So I thought I would just chime in with a couple other thoughts on that. In addition to like how to think about scale, I think the other pieces that we need to think about which you had alluded to, Sarah, are folks who seek care out of state and how that impacts on any new payment model or at the provider community and remoders in general. So that was the other component, not specific necessarily to scale, although it does impact the Medicare scale. And then the only... Robyn's computer is sacking up for the day. Robyn, I lost you, I think you're frozen. Let's give her a second as she tries to unfreeze. I'll text her. Did any other board members have any questions or comments while we're trying to get Robyn back on? Apologies, Mike, in turn. Yeah, this isn't much better, Robyn. Can others see this? The only thing I was gonna... What's that, Michelle? No, it's Robyn. Hi, is this better? It is, thank you. Sorry about that. Not sure what's going on with the internet here, but the last thing I was just comment I was gonna make was that in the negotiation around the Medicare total cost of care target, I think part of the reason why we got such a low target of point one to point two below national was because in part of our low total cost of care. So sorry about all that disruption from the technology. No worries. Other questions or comments from board members? Sure, this is Tom Walsh. Thanks for the presentation and thanks for the other board members' comments and questions, I guess reflecting on it, this past hour just raises more questions in my mind and things that I feel like I need to think longer and harder about and benefit from these type of conversations. The cost shift conversation that keeps recurring, if we were seeing a cost shift, we should see things like hospitals that have a higher proportion of their patients being covered through Medicare or Medicaid, they should be charging higher prices to commercial payers. And that relationship we should see across the board. The RAND study that recently that's come out the four part one that people have been talking about and others show that that relationship is just not there. There's no correlation between the proportion of payer mix and the prices that are set. Our own price variation reports show no correlation between those two things. So we really, I think need to pause and think about how we think about the pricing variation that we see in our state as well as across the country. The graph that showed total cost of care, we need to remember that the total cost was made up of the amount paid and utilization. When Medicare or Medicaid are paying, their payments are the same to hospitals, so to similar hospitals. So when we see differences in the total cost of care in those groups, that's due to utilization differences because the prices are held steady. So our utilization is lower than other similar states, which on one hand is great, but it confuses me greatly because we have these tremendous wait times. We have a sense that there's a shortage, a workforce shortage. We also have data from the American Hospital Association, the American Academy of Medical Centers that show things like the average discharge coming out of UVMMC, that's just the one that I looked up. We're 54% more hours per discharge compared to the national average. So these things are confusing to me because it looks like we utilize less care than similar hospitals when we hold the prices steady, but our system seems internally gunked up somehow. I think using the word efficient because we're lower the total cost of care could be misleading, I'm not sure yet, but I don't see signs of efficiency when I look at data on hospital performance with our hospitals. So more questions than answers, but those are kind of my observations and things I look forward to continuing to think about with people. So thanks for the presentation and the good work. It's been a very difficult time with the pandemic and there's just, these are, they're hard. So thanks for all the work you're doing on it. Thank you, Tom. And any follow up from any other board members? If not, I'm gonna turn it over to public comment. Does any member of the public wish to offer comment on the presentation? And I see Robert Hoffman. Thank you, Chair Mullen. I wanted to dovetail with what member Walsh addressed and that is if you actually in earnest make some effort to drill down somewhat more than this very summary level presentation has and expand your aperture to look at a broader view of the system what we see is not efficiency at all but actually very, in fact, extremely troubling trends. We see that primary care utilization and this is based on blueprint for health, published data in their 2021 annual report. Primary care utilization declined to staggering 16.3% in Burlington where reform efforts were most focused over the same period. We see breathtaking decline of 28%. ED visits increased 9% statewide and in Burlington, 10%. Substance abuse treatment initiation decreased 5% statewide in Burlington. Again, breathtaking 18% decrease. Chronic condition inpatient admission rates increased 18% statewide in Burlington, also 18%. Heart failure admissions increased 18% statewide, a 31% increase in Burlington. Uncontrolled diabetes increases 6% statewide. When we look at the most important and sadly a look that nobody in the state is bothering to take, when we look at age adjusted mortality rates for 45 to 64 year old Vermonters, these are the prime candidates that should be benefiting from the purported interventions of preventative care on the all-pair model. As compared to the national average in 2014, Vermont mortality grew across key leading disease such as diabetes, heart disease, cancer, et cetera. 10.2 times the national average and 15, 11.8 times the national average and 16, 9.2 times the national average and 17, 11.1 time, 2018, 43.39, 2019, 16.4 times and in 2020, 3.74 times the national average. There are only 10 COVID deaths in 2020 in this age group so it's not material. So the overall aggregate growth rate of mortality in this key demographic on average over that same period was 15.1 times the national average. Concurrently hospital net revenue declined by 90% from roughly 130 million in 2015 to 1.7 million in 2020 while concurrently, as I said, mortality grew 1,500% in this age group. Based on extrapolation from this data, we can infer that potentially 320 Vermont lives in the 45 to 64 year old population were lost due to these mortality growth rates. I just wanna strongly encourage this board. It's time to put our adult hats on and really start looking at this experiment seriously because others are and people are frightened by what they're seeing. Here in my state of Pennsylvania, I'm lobbying legislators and physician associations to look strongly at Vermont's example that we are trying to implement here in our rural health model. You have two and hopefully three members leaving this fall. We would love, I think those of us interested to see some real subject matter and please join us in the form of health care administrator and physicians. I think the era of technocrats is long done. It's time for qualified professionals to join this board and provide strong oversight and leadership of their staff. Thank you. Next, I'm gonna turn to Walter Carpenter. Walter? Thanks, Kevin. Just a couple, a comment and then a couple of questions. The comment is, I do agree with Rob Hoffman. So I want to rappelate on that. I also wanted to thank Tom Walsh for his comments because I backed Tom up wholeheartedly on those. I won't agree any more about that. The question is, I came on late so I might have missed something but Vermont seems to have missed the targets for the ACO despite the billions that we've already spent on it. And I don't know if the targets mean anything or not but in any case to say that we missed them for the million billions we've spent on this is something to think about with awe. And then with all of the data that's been going on in these reports, I was wondering if the quality for care for Vermont patients has improved any? Is there any improvement in the quality of care for Vermonters? And that was the one thing that was not measured in all of this data. In other words, can a patient go in to see a primary care physician and not be kicked out in 10 minutes? Stuff like that. That's all. Thanks, Walter. Is there other public comment? Is there other public comment? Hearing none. Sarah and Michelle, I wanna thank you for a very good presentation. I will just touch on one point that Walter questioned and that was the failure to meet targets. Some of the targets, Walter, are never attainable and aren't even realistic targets. And the people who negotiated the agreement have since acknowledged the fact that they were going on data that they had available at the time but that it just the numerator denominator doesn't work and Sarah Lindberg could expound on that but I think that's probably a topic for another day and unless you want to, Sarah. Yeah, I think that the targets we're not meeting are partially unachievable due to the way people come into this model, some of which definitely might have to do with the ACO but other, they can't make people participate in this voluntary model today. So there also needs to be participation from payers and it needs to be people who are actually seeing primary care providers. So that's just kind of the way that a lot of these programs are set up but the idea that 90% of Medicare beneficiaries could ever be attributed through this methodology is literally impossible. So yeah, I think it's, yeah. I think having half the state in this model to me is pretty impressive. I know it's maybe underwhelming just to those out in the wild but that's a, that's a... I think underwhelming is probably a good word and might be an inaccurate word and when you consider that we've spent so much money on this and it's coming from Vermonters fees, premiums and taxes. Which is why I think board member Walsh is correct on his idea that we've really got to go down and I guess look at this, you know, he's right. This stuff doesn't add up. And even if you lowered the goal post, you still wouldn't meet them. It's the ACO I don't think was designed to meet them in the first place. Yeah, well, I think, yeah, like the idea is you wanna know who you're accountable for as a provider but in truth, it's hard to predict in advance where people are gonna get their care for an upcoming year. So usually, if you look at a retrospective model, you tend to have better kind of alignment in who people provided care for but there's trade-offs there. So I think it's, it's a... Well, it is, while I appreciate the kind words from Walter and Rob, but these are difficult things and the work, Sarah, that you've done and Michelle and others helping us think about this and openly sharing things that may not really look that great but help inform our thinking, I think that's really important. And there are, like I said earlier, I think there are some fundamental things. Another just in wrapping it up, across the country, people felt like it was a good idea to try to tie the growth rate of healthcare toward the non-healthcare sector that they should grow kind of evenly. But if we step back and reflect on Vermont, we are an aging state, we have more people not working and becoming less healthy as they age. So why those two things should come together, right? When the growth of the overall economy, people are leaving that as they retire and age and they're going to a place where they're gonna be consuming more of healthcare. Lining those things up, it may be wise to try to do that in parts of the country, but not here in our state or other states where people are aging rapidly in place. So it's not that anybody was malicious in the design in the past, they're using the information they had being as thoughtful as they could. And as we've done these experiments across the country, some have succeeded and some haven't. So what do we learn from all of that and try to do it better with the next one? We've got to do something, right? The way that it is right now is not working. So we try things and get better. I think one of the problems with all the models and go through a model is what the phrase Tom Walsh uttered and he said, consuming more healthcare, we keep treating healthcare as a commercial product and that these models do the same bloody thing. You have to stop that idea. Yep, let's, I appreciate that you saying that and I should say utilize because I agree with you. I'm happy to talk more with you about it. Yeah, anytime Tom. Cool. Okay, other public comment? Hearing none. Sarah, Michelle, thank you very much. A lot of information to chew on and to digest and as Tom said to learn from and that's the key thing. We have to continually learn and evolve and the key point also that Tom made was that the status quo isn't working and so to do nothing is not a plan either. So we have to continue to try to move in the right direction and we'll continue to do that here in Vermont. Okay, is there any old business to come before the board? Is there any new business to come before the board? Hearing none before I take a motion to adjourn I just want to remind people that this is the final week to get applications in if you are interested in being a board member. It's not an easy application. So if you have an interest, you better start now because there's only a couple of days left and it's an extensive process. So with that, is there a motion to adjourn? So moved. Is there a second? Second. It's been moved and seconded to adjourn. All those in favor, please signify by saying aye. Aye. Any opposed, please signify by saying nay. Thank you everyone. Have a great rest of the day.