 Sarah, are you doing this together or separately? I think we're doing it separately. Okay. Okay. So Sarah, it's a job with Blue Cross and Blue Shield and we're on. And I believe you have some questions about the impact hospital price cuts would have on our numbers, and so that's what I was prepared to talk about as well as some of the examples. Yeah, basically, we've heard review of the hospitals that the non-consumers may not see the old benefit of a cut that would replace, unless you have to be in the contract here, which I understand is January 1. So I thought I heard clearly from you that the last time you were here that the monitors would see the benefits of those cuts and in fact it gets compounded by knowing the date of the cut. So I was hoping to address all those issues again. Sure. Okay. So I wanted to talk first about who benefits from a hospital price cut. And I was going to start with members. For monitors, all members who have cost sharing, deductibles, and co-insurance would immediately see the impacts of a hospital rate. So just to give you a little clarification on high deductible health plans component of that. And that's about almost 34% of our members are in those types of plans. The next thing would be self-insured employers or businesses would also immediately see the benefits of a hospital rate cut. So they would see it in the payments that they're making because they're self-insured. That's about 58.2% of our members. And then fully insured members, and so this is the piece. And to give you an example of self-insured, that could be the high plans. So the schools would immediately see the benefit. For fully insured members, and I think we've often talked about the qualified health plans with those, they would see the reduction in two ways. So in 2019, if it was done soon enough that it could be built into our rates, it would be included in the estimates of the qualified health plan rates for 2019. In addition, any amounts that are reduced in 2018, and so this is where the piece is with the more months that you see, would also be included in the 2019 rates. So we would see that in our presentations there. Can you remind us how many you have at UHPs now? It's about 70,000. And how much in large group? About 15,000. And how much in large group? I had about 58%, which is 165,000 members roughly. So I was anticipating maybe some of your questions. So one of the questions I anticipated is why a hospital price reduction sooner is more impactful than a month later. And that's because of three reasons. One, it lowers healthcare costs to the entire health system immediately. Another is that there is the immediate benefit that I talked about earlier, to members and to employers. And the third is the longer term benefit that would be realized by fully insured members in 2019. So to say that more clearly, any 2018 rate cut will mitigate the 2019 premiums. Is there any shrinkage of funds or anything that we should be concerned about? There seems to be a common concern that not all the money will actually end up to the consumer. So the only thing that I would say there is that not all the money is through the qualifying health plans. It's across all of our numbers. So if you think- There are no consumers too. Sure. Yeah. So it would go across all the different types of members that we have. And I think because you- You know, the Green Mountain County Board reviews the qualified health plans. That's where you're going to see it. But you won't see it all in that one segment of our population. Well, we review more than qualified health plans. We also have the large group. True. And you'll see it in those places as well. Yep. Okay. So then I was going to jump to the estimates that we prepared for you. And I know you have your own staff that does that for you as well. But to give you an idea for the qualified health plans in 2017, about, let's see, one-third of those members received their charges through a UVM-MC facility or a UVM-MC provider. So just to give you an idea of who's impacted in that group. If you looked- We took a look at our 2017 calendar of your charges and trended them forward and asked to make that each 1% reduction is worth about $280,000 for each month that is reduced. Just to give you an order of 8,000. Per month, per 1%. So the impact, if you started on July 1st, just as an example, with 1% is about $1.6 million. And the impact, if you started on October 1st, is $870,000. You know, those are rough numbers. We do see more of our claims in the end of the year. So that across your entire book of business? It is. Let me check my notes. It's four. Qualified health plans, large group, insured, and ASO. I think the federal plans are excluded from that. And then the last thing I was going to talk a little bit about how we would operationalize a reduction, and I know you've heard some of this before, but just to repeat a little bit, we can accommodate a flat percentage rate reduction across the board. And we would probably apply it to the DRG case rates. DRGs are diagnostic-related groupings in these schedules for inpatient, outpatient, and physician charges. And that's how we would be able to handle it in the four-week timeframe that we talked about before. I think I just want to say it again so that you know we could do it if it was implemented any time of the month that doesn't need to be on the first. And I think that was a question that was asked before. And we would open our contracts with a Green Mountain Care Board order. And then, typically, we adjust the fee schedule on January 1st. In the typical year when you make your hospital budget orders in October, it's not until January 1st when you actually see the change in the schedules. And so... Plus one, two, zero, three, nine, eight, four, nine, five is one, five is now joined. I think that might be everything. I think that's on the line in New York. I encourage that. So, this is Sarah D. Jobs from Blue Cross Blue Shield. It's still testifying, and Susan will be up here in a minute. And then just the last thing I wanted to touch upon is that there has been some precedent for reducing rates in the past. We looked at Marklin Regional Medical Center spray reductions that were done in 2006... They were done in 2016 on the after-wide 15 budget. So the... And you're probably aware of this already, they did the first part of the decrease, affected May 1st, 2016, and that was for 3.7%, and then they did the additional reduction on October 1st, 2016, for a total of 5.1%. So it has been done in the past. I can try to answer questions if you have them, but those were the main points that I was trying to make. So are there questions for Sarah? I have one question. If you had indicated if you were to turn around the ringtime in four weeks, you'd do across the board, across all related DRG groups. If you had additional time, would you be able to do something more targeted? So for example, one of the areas that last year in our hospital budgets we were looking at was the pay parity issue? I think if you ordered us to look at that, we could. And certainly it takes more time to go through and analyze each of the different types of services. Other questions? So we're going to, I guess we'll hear from MBP. Sarah, if you could stick around just in case there's public comments or questions. Yeah, I'll be right here. Thank you. So welcome, Susan. Hello, this is Susan Krakowski for MBP. Could you introduce who's on the phone? Yes, I will. So on the phone we have Kathleen Fish, who's our Chief Actuary, and George Thompson, who's Vice President of Network Management. So he handles the provider contract inside of things. So I think Kathleen is going to start us off. And the question that we were asked to address is because you've made the decision that you will implement any rate cut in October. So in October 1 with the next year's hospital budget orders we were asked to, how would that show up in rates? So that's the question that we're prepared to talk about today. So Kathleen, I'll turn it over to you. Can you hear? I can't. Thank you, Susan. Okay. So we would be filing our exchange rates well before that. And what we have typically done when we are asked to file rates before we know what the approved rate increases will be is we assume the most recent approved rate increases for that time period. And then we update them through the course of the rate review to reflect the most current information prior to rates being approved. So in the cycle last year we had been asked by Elny to adjust our trend assumptions based on the requested rates from the hospitals. And then we were finally asked to adjust them from the Green Mountain Care Board based on what they ultimately approved. And that was done prior to the final rate approval. So if a similar cycle happens this year we would reflect any rate reduction for that hospital and all the other approved hospital rate increases prior to finalizing and getting, you know, final rate approval from Green Mountain Care Board. And it would all benefit the 2019 rates in the same way Sarah explained. Any other questions of Kathleen? No? Okay. Kathleen and George, it looks like there are no questions. This is probably my shortest time ever. And it's here in front of anybody. Simple story. Simple story. Okay, well thank you very much. And I will be here if there are any questions. And thank you, George and Kathleen. Before you get out of the chair, I will ask a couple of questions because we got some numbers of lives from who crossed the shield. But could you update us on your numbers of lives now? Sure. Kathleen, I have a rough idea of our numbers. Do you have perhaps more accurate numbers? Do you think? I don't know. That's no matter what it does. Yeah, we have roughly 20 to shy of 29,000 members in our fully insured products. And that would be large group and exchange products. Correct. Somebody, could you just break that down for us? I'd be super. Okay, yes. I will do that and get that to you. Okay. Thank you, Susan. Okay, thank you. Okay, this time I'll open it up to the public. If you have any comments or questions. Plus 1-8025-27307 is now joining. Most people are now entities. Wait for the exit. I'm Carmen Austin and I am the vice president of commercial payer contracting at the University of Vermont Health Network. And in full disclosure, I am a former employee of Blue Cross to the Shield of Vermont and also MVP healthcare. My responsibilities there were hospital contracting. And in both Vermont and New Hampshire. So in a letter from Dr. Brumsted to Chairman Mullen, to Chairman Mullen, dated on March 13th, he stated our proposed 0% rate change would create immediate and substantial savings by keeping commercial rates flat. And even as hospital expenses rise due to inflation. So my question to the board is, at a 0% increase at the medical center would essentially result in a reduction in premium. What will that equate to in premium reduction and can you quantify the percent and or value of that reduction? Not without our actual risk of work. If we do the work in free-for-me process. Do you want a direct question to the two insurance companies that are here? Sure, I would be happy to direct the question to each insurer. Yeah, we can't answer that question right now. Sarah, teach out with Blue Cross and Blue Shield. Do you have a number of boxes? Yes, exactly. Is there any other member of the panel that wishes to make a comment or a question? Yes? A series of questions on Mark Stanislaus from the University of Vermont Health Network. Understand the relationship with co-pays and understand the relationship to a risk of plans, but it would be good to put perspective around what your total premiums are as it relates to the employers that buy insurance through either of your entities because those premiums would not change until January 1st, correct? Susan Garkowski for MPP. For exchange members, yes, that's true. However, for large group, it's possible there are some large groups that renew at different times during the year, so something perhaps could be shown for them. Okay, but the normal period is what date that would happen for the majority of your plans? By far, it's January 1. Okay. Okay. So, and then... I thought, go ahead. We're in a very similar position. Yeah, so I thought she was shaking her head there, so. So, then the other question is, and you probably don't have this answer, but it's been clear from some other filings, out of your medical expense base, how much of it is related to the oversight of Green Mountain Care Board as a percentage? I think I've saw some filings that is between 45 and 55 percent. Right. Sarah, teach that again. So ours is about 50 percent? Okay, so up there. Okay, do you have an understanding if the year-over-year cost growth has been growing more quickly on the 50 percent out or the 50 percent in? That's a very important question here to understand the dynamics and what the hospitals are doing. Susan, for MPP, I do not know that. I think it depends on which timeframe you're looking at, so we would have to go back and look at it. Okay. And then, at least from a hospital perspective, and I want to make sure that it referenced the right rates here, as far as the gross. And this is, you know, somewhat in relationship to that. And this is purely for Blue Cross Blue Shield because that's what I pulled off the Green Mountain Care Board website. But, and this comes from your actuarial analysis dated January 11th, 2017. In here, there's a number of factors that you do to understand the rate-setting part of what you're going to charge your employers. And the best I could make out of it that if you look at this, as you said, the trend rate from 2017 to 2018 was 5.4 percent. Okay. There was also an additional trend rate, that 16 actual rate an hour, by 1 percent, 1.1. And when you came in front of the Green Mountain Care Board before, there was 5.3 percent. So if you were to add those two together, that would be an aggregate increase of 6.4 percent. So I just want to talk to the Vermont Hospital system growth from actual to actual. 2014, it was 1.6 percent. 2015, it was 5 percent. 2016, it was 4.4 percent. And in 2017, it was 2.8 percent. And I think the rate approval that was attached to this filing was 9 percent. So my point is, is the hospitals are doing a very good job as a whole managing cost. And what you're seeing as incremental cost increase is the other 50 percent. And that's what we need to work to figure out. Because it is important to the monitors. Because if you continue to take the burden from a premium perspective that the 50 percent that isn't there and put it on the 50 percent of the hospitals which from many benchmarks are doing very well from managing cost and quality, I'm very worrisome just from a financial analytical perspective not from the health network about the financial sustainability of the hospital system in the state of Vermont. Thank you. Sarah teach out with Blue Cross and Blue Shield of Vermont. So there's a lot there to talk about and we'll talk about it again in our rate hearing. But that's a little bit of a simplified view of what's going on. And another component that wasn't mentioned is the cost shift which we also know is a growing piece of the spending here in Vermont. I would like to respond to my cost shift because I did see that in our rating. Just a second. Yeah, that's okay. So those hospital numbers that I referenced those are inclusive of the cost shift from the 50 percent on the hospital side. So once again, if there's cost shift it's from the other 50 percent because these numbers including the cost shift. So one question I would have is and I don't know how the rates are done in the actuarial space and I don't know all the pieces and parts that you guys go through. But if there's an immediate reduction from the provider side or the hospital side why couldn't those rates for all payers for all, you know, Vermonters be adjusted immediately? I don't know what immediately means. So I would love to see when my bill comes to my organization a reduced amount we buy our insurance off the exchange but if it's happening in real time why can't it be real time in both segments of the market? For the QHPs you can adjust the rates mid-year they've already been sold to the consumers. So I don't know about the other lines of business but because of the regulatory structure at the federal level on QHPs that would be difficult for that market. Now that's, you know, 70,000 people as opposed to some of the larger spaces. Do you have anything to add, Sarah or Susan? Okay, other questions or comments from the public? Yes? Carmen Austin again from the VMTEL network. I just also wanted to bring up a point that Todd Keating had brought up during the testimony in one of his 20 years that we do not have different fee schedules for the different payers. So I did take a look at what the impact would be and that we would have to adjust 51 of our agreements here in Vermont. So that would not happen very quickly. It's a very long process to reach agreement and settlement as to what the adjustment may be and the impact and the effect so that I just wanted to bring up that point. Is there any other member of the public who has any comment or questions? I think as we think about a rate increased by Ms. Danislas from the University of Vermont Health Network and the detail that Carmen spoke to and it was filed in a response of what all those individual payer lines, you know, would be. So just note that that is confidential for proprietary information also. But, you know, I think we need to be sensitive to that if there was any change, exactly how it would hit the Vermont payers and any other commercial payers that happen to seek services here that we just don't automatically assume that it carries over to because it gets into the in-migration and out-migration too. So, you know, there's a sensitivity there and that's part of the complexity of some of our payment schedules because some of those payment schedules are linked between the same plans. And also there's a crossover in payment. Skuggles too to the all payer model too. That's a complexity that we talked about earlier that we just need to be sensitive to when we pull that lever, you know, what's going to happen on the other side. Is there anyone else? No, thank you. And we'll move on to the all payer ACO model implementation update. You can join us up front. Thank you. Last week during a staff discussion of the ACO certification process, there were a couple of questions that were raised pertaining to the ACO budget. And we wanted to provide the best information we have available at this time and also to make clear to you that you'll have final information in the month of April and we'll be providing an in-depth analysis of that final information at that time. But today to answer some of the questions that were raised I wanted to share some key information related to attribution to the ACO. So how many Vermonters by payer are attributed to one care? Plus 18025227307 is now exiting. In the Medicare population, the final attribution to one care for month and this is final attribution for the start of the performance year. It's very important to recognize that this is a high water mark and the attribution will decline as the performance year plays out. The attribution January 1st for the start of the performance year for Medicare to one care of Vermont is 39,702 Vermont Medicare beneficiaries. This differs from the projection in one care's last budget submission or the information we received on December 20th. This differs from that projection which was for 33,474 blacks and is a 19% change. For Medicaid, the final attributed number is 42342 differing by a negative 4% change from the projection in December. For Blue Cross Blue Shield, the final number is 20,838 blacks differing from the projection in the December budget submission by negative 40%. And the self-funded population that is attributed to the ASIO is 9,962 and that is the same as was projected in the budget submission in December. That's the update that I have today. Do you have the total? The total blacks, yes. 638,548,148. 638,000? No. Doesn't he determine what? $100. Excuse me, not with dollars. Excuse me, I'm sorry. 112,844. So about $10,000 less than what we anticipated. That's correct, but as we'll discuss in our more in-depth presentation the actual dollars are more than anticipated because the Medicare population has a higher per member per month amount associated with Medicare and that population grew over what was anticipated. That's all I have for the update. Any questions? Is there any public questions or comments? Yes, Susan. Susan or not? This council, Ina, thank you very much. I'm wondering if this information is publicly available. I couldn't find the facts that you just read off anywhere. I tried to write them down as quickly as you said them but if you could shoot an email with it or post it, that would be great. We'll post the final information that is available and we received this just before this meeting began so it hasn't had an opportunity to be posted yet. Any other questions or comments from the public? Okay, thank you, Ina. Might as well stay there and now we're going to have the certification update so if your colleagues could join you. Thank you Chairman Mullen for the record. Melissa Moss with the Green Mountain Care Board Ina Bacchus and Mike Barber. As we reviewed last week we had went through an in-depth certification process and we were left with two remaining items that One Care Through the Week has responded to us on. You have a slide in your folder describing what we were looking for and what we received. So the staff had asked One Care to describe its process and timeline and criteria for accepting providers into their network for 2019 and in return One Care's Board of Managers met yesterday and endorsed One Care's timeline and 2019 Network Development Expansion Strategy and this includes the provider types from which the ACO is seeking participation. The second item that we were looking for is a policy that allows ACO participants to appeal their decisions and this is eligible participants. They had an existing provider participant appeals policy and they submitted it with an amended version to include eligible network participants. And so these submissions satisfy the items that we still had in process and we also had an open public comment period from March 14th to the 20th and we did not receive any written public comment and so we would like to recommend One Care for a month for their ACO certification based on our reviews. Any other questions? I have one question. The recommendation would be for certification with the monitoring recommendations that you had outlined last week. Is that right? Yes. Are there other questions? Here none. I'll open it up to the public. Dale? I keep having to go through my head before I long to know. You can speak up. I don't think I'm having trouble hearing. Okay. I deal with that all the time. I'll go through my head. What I want to know is, and especially after the conversations this morning, a ratio index. I want to know how many primary care providers per persons in the ACO as a unit of measure. Are they serving 300 people per primary care physician? Are they serving 250? And I want to know that in different locations. I haven't totally thought this through yet, but that can help me to understand what my delivery is, what my timelines would look like, how efficient the delivery would be. And if I break it down into geographical locations, it can help me figure out where I'm not getting good performance and where I am getting good performance. I'm getting a very high vote because I haven't totally thought it through yet. But I'm wondering if that wouldn't be a value. Melissa? Yes. We were also considering that information, but we've been waiting on the final contracts. And so once the Blue Cross contract is final with the final attributed lives, which we just received this week, the final attributed lives, we have considered doing that analysis and could provide at the beginning of April working with OneCare. I would, yeah. Okay. Is there other members of the public who wish to comment or have a question? If not, is there any member of the board who wishes to make a motion? I'll make a motion. I would move that we approve OneCare Vermont's certification with the monitoring recommendations proposed by staff at our meeting last week. Is Maureen on the phone today? This afternoon? Yeah. Oh, she isn't. No, that is a negative. I'm sorry, we must have lost her this morning. So was there a second? I'll second. So it's been moved and seconded to certify with monitoring and is there any discussion by members of the board? So hearing none, I guess since we don't have a member on the phone it's saying to do it by voice vote. All those in favor signify by saying aye. Aye. Any opposed? So let the record note it was 4-0-1 vote. Is there anything else you would wish to discuss with us? We don't have anything else to add on the topic of certification but it's all right with you. I will give a brief introduction to the next agenda item and Rachel Block is here. That would be great. It was just demonstrated by Dale's question to talk about primary care, primary care spending and the role of primary care and healthcare reform is a very hot topic here in our state and nationally and that's why we've asked Rachel Block here from the Milbank Memorial Foundation Foundation or Fund to talk with us about the work that she has done with the fund in looking at how to measure primary care spend. We thought that this would be an important conversation for the board to have both in the context of primary care spending as a percentage of overall healthcare spending in Vermont and more specifically as we look to healthcare reform and the implementation of the ACL model and how we can track over time whether that model contributes to more investment in primary care and with that I'll turn it over to Rachel and thank her again for taking the time to come here today and speak with you about this important work. Well thank you Nina and thanks to the board for inviting us to speak on this topic today. This is something which we've been doing quite a bit of work on for a number of years but specifically the issues around primary care spending for the last year. I'll give you a little bit of a quick overview of the funds involvement in developing the evidence base in support of strong primary care and then talk in more detail about primary care spending measures from our perspective. Why are we interested in these measures? What are we trying to measure and how? What are the results of a recently published study that we commissioned on this topic? The next steps for us and then potentially some opportunities for the State of Vermont to consider as you're going forward. So our mission is to improve population health by connecting leaders to the best evidence and experience. That's a very general mission and the way that we do that is to build evidence through research support. This particular work that I'll be describing is an example of that. We then also disseminate and work with leaders to use that evidence through those reports through a convening of State and other leaders and many policy leaders in the State of Vermont have participated actively in our meetings in the past. Examples of our work specifically relating to primary care the multi-state collaborative. This is a national effort to bring together projects that are participating in national primary care payment reform through the Centers for Medicare and Medicaid Services. Lisa Dulcy Watkins, who I'm sure many of you know leads that project for us. We have published a couple of reports that focus on primary care and behavioral health integration which is one of the elements of primary care transformation and now we have published work on primary care spending measures that we'll be talking about today. Why are we interested in this? Well obviously we do have a track record of general policy interest in primary care but specifically as it relates to primary care spending we believe that we improve what we measure and if we look at the big pie slice this happens to be national health expenditures you could take the Vermont health expenditures as a separate slice as well. What we'd like to know is can we get a separate slice of this pie that we understand is being devoted to primary care? The way we approached this work was to commission research so I want to emphasize as a research project it is not necessarily a policy initiative and we commissioned bailout health purchasing who I think the board and others here are familiar with and which had experience working with this data a link to the report is included in the slides. The basic idea was to undertake a proof of concept study to determine what percentage of commercial medical spending in high performing plans went to primary care and the scope is a very small sample of commercial health insurers from across the US it did not include Medicaid or Medicare participating plans the definition of high performing commercial plans was essentially those that had very high scores in the NCQA accreditation process and the reason we picked high performing plans is we felt that they would be the ones that would be most likely to be spending more in primary care so we were priming the pump a little bit if you will for the purpose of the study results. We also tried to make sure that the sample would be geographically representative. You'll see in terms of the numbers of plans participating after that analysis contacted 29 plans 11 originally agreed to participate and nine were able to produce useful data. I'll come back to some of the logistical issues associated with producing these measures later in my comments. The definition of primary care spending was developed in consultation with other researchers including those in primary care specialty groups and also from a few insurance commissioners and the way that this was done simply was to work with the health plan staff for those plans who agreed to submit the data based on our specifications and they would give us of all the levels the actual dollar amounts as well as percentages of total health plan spending in 2013 and 2014. We looked at two product lines HMO and PPO and the analysis included fee for service payments and also non-fee for service payments those including capitation bonus or other shared saving kinds of arrangements as I'm sure many are aware these are an increasing part of total medical spending so it was important to capture that data. We also included some modest measurement of demographic and comorbidities and I'll speak to that in a moment. The measures of primary care were broken down by specialty by service codes and by age groups and in a nutshell the results are that while much attention is often paid to defining specific types of providers and specialty categories in primary care that there was more difference in spending amounts based on service codes were included as opposed to which specialties were included. To try to make that a little simpler a narrow definition of primary care specialties versus a broader definition of primary care specialties did not yield very significant difference in terms of the amounts and percents but trying to define a narrow range of primary care services by codes as compared to all services provided by a primary care provider many of which would be primary care but some would not that resulted in a more significant difference in the amounts and percentages. The study also found that there was more primary care spending for children less for older adults. So just to give you an idea of some of the numbers the dollar amounts are listed here and you can see that there wasn't a huge difference between the PPO HMO amount to $26 per member per month but part of what was interesting about the results of this study was the range of the participating plans from $14 to $38 per member per month and in terms of percentages we see a similar result the differences between PPO versus HMO not that grid 6.7 versus 7.4 but the range of all the participating plans was 3.4 to 12.5 percent so that is a much more significant difference looking across the full range of plans that were participating. I picked out one of the tables that's in the report this is the age breakout there is no 65 and over because we didn't include Medicare as you can see and it does not include Medicaid so that would have an impact particularly in terms of the children's number for example we could assume would be even higher if we had the Medicaid data in there but you can see the range in terms of going from the younger age groups to the older but also again I kind of highlighted interesting differences there among the plans while one plan would be spending 3% and another plan spending 14% these are interesting things that we'll be looking into in terms of future research there are a few limitations to this study we want to be fair about upfront one is obviously it is a small number of plans the data were self generated by those plans the reporting was voluntary and the results were not audited again this was a research study not a regulatory matter that we were interested in and the plans were particularly challenged to provide the data specifications in particular as it related to the non fee-for-service component but we know when it comes to primary care spending that that non fee-for-service component is a very important part of how insurers have been promoting primary care so it's really important for us to be able to get at that data also regardless of the definition we started with the insurers designation of primary care provider so next steps for us we think that this is important foundational work to begin a larger conversation so one of our first steps is to work with states to try to replicate and validate these measures to consider legislation or regulatory measures that would promote further use of these measures Rhode Island and Oregon currently by regulation and statute have primary care spending measurement reporting in place and also both of those states have now prescribed minimum thresholds for health plans to spend and the activities around primary care spending that they need to contribute to so we'd like to at least start with working with more states to replicate these measures and see what the results are when we do that in Oregon they found going back to the range among the small sample of plans that we had they had a range from approximately 5-6% on the commercial side to about 12% for Medicaid and so they subsequently enacted legislation which will require all health plans to spend 12% over a five year period coming up so essentially using Medicaid in their case as the benchmark for primary care spending performance in addition to trying to spread this work among more states we will be working on disseminating these results through professional meetings and articles and we're also hoping to both spur and participate in broader discussions on this topic collaborating with efforts in primary care specialty societies and researchers to continue the dialogue on how to further refine the definitions we are sponsoring additional research phase one to look at Medicare data and how the primary care spending measures might be of interest to the board considering the participation of Medicare in the all payer model we want to connect with others nationally around the development and use of measures and we've had some very encouraging discussions recently with NCQA which is interested in a bigger national study of primary care spend and also the Health Care Cost Institute which included an analysis of primary care spending in their last national health care spending report and finally we intend to continue our support for multi payer models for primary care support and use that platform as an opportunity to continue discussion as far as local regional efforts around primary care payment reform concern so the things that we're discussing with states, I mentioned the opportunity to consider legislation or regulation we have Oregon and Rhode Island as our models today Colorado will be introducing legislation I believe this week and I heard that California may have legislation being introduced as well but we'd also like to explore how primary care spending measures may be aligned with other measures that are being used in statewide initiatives such as the all payer model here or in other states where either ACO or total cost of care measures are being used for projects and clearly there could be applicability here not just at the state level but also at the regional level to the extent that there are significant numbers of regional collaboratives looking at enhanced primary care support generating primary care spending measures using local data and monitoring primary care spending in conjunction with their local advanced payment model and ACO activities so in conclusion we feel the policy and evidence suggests that it is important to measure primary care investment we think primary care is important but we don't have the data to show what we're spending on it or how we're increasing that commitment over time research and state efforts suggest at this time it is feasible to develop and use primary care spending measures there are administrative issues to take into account resources are required and you need to plan for it whether that's on the insurer side if the approach is to ask the insurers to submit this data in a certain format or on the state convener side should a state choose to use its all payer database or any other database that a state may have to try to generate these measures on their own and it's important as with all matters relating to data and measurement to have a transparent process and some vehicle to ensure that there's trust in the data that's being used in terms of policy issues I'm sure this will come up in today's discussion it's important to consider a building to purpose so what was the desired unit of analysis what level of detail or precision is needed as I made clear for us this was initially a research project but as we get these measures in use in both whether it's state driven activities or private activities I'm sure there will be an interest in even further refinement and levels of detail of the information it is important to standardize measures in order to facilitate valid comparisons we would love to see these measures standardize nationally but we need to also grow the usage of these data from a grassroots perspective as well we're interested in looking at establishing or validating the relationship of primary care spending to total cost measures we know some of the national advocates who are supporting increased investment in primary care are concerned that they're not viewed as saying that that investment will also increase total cost of care they want to have increased primary care spending considered in relation to monitoring or holding down increases in total cost of care but we need the spending measures in order to be able to develop that methodology in a systematic way and finally it's important to have these measures in order to evaluate the impact of various all payer models and value based payments on primary care sensitive types of performance measures so perhaps establishing the relationship between spending and some of the utilization and outcome measures that we're interested in associated with a strong primary care system I include this not because it has anything to do with primary care spending but this is the basic components of the CPC plus model of primary care and the reason I include it is simply to illustrate that a primary care spending measure is not an end in itself and it does not capture the full complexity of all of the things that we're interested in in terms of measuring primary care but we still think it is a simple and feasible measure as a piece of this puzzle to be able to better understand how primary care fits in the total spend discussion I've included my contact information as well as Lisa's if you're interested in following up on either our work on primary care spending or total cost of care issues or the multi-state collaborative and I've also included a number of references and links that the board and the public can use to learn more about this topic I'd be happy to answer any questions Thank you, Regent Questions from the Board? I have one So I was trying to connect the dots here that where we started out with the NHA expenditure pie and then as we go a few pages later we have that payments equal about 7% of spending so if your work were to be integrated after the methodology in this NHA pie would it slice be about 6% or 7%? We don't know for sure but it's a reasonable guesstimate at this point in terms of the little bit of data so for example I mentioned the health care cost institute report which came out a month or so ago their primary care spend number if I remember correctly came to about 6% that was also only commercial health plan data but it's a big national database so again if you included Medicare and Medicaid the numbers might be a little bit higher or lower so that's as close as we have now to a national measure and as I mentioned we're talking to them about possibly using our measure and creating a separate report that would focus more specifically on primary care spending I noticed in your approach to this that the New England region was more heavily weighted in terms of insurance companies you work with I think after you filtered down to the 9 did any of them because it was agreed that we would not disclose who those plans were again we started with a total number to work with of 29 it largely got little down based on their capacity or ability to provide the analysis that I needed I was curious about what direction Oregon went with their definition we've had two different workers in Vermont work on the issue of defining primary care for two different purposes and so those estimates did vary depending on which sort of purpose it was for but I was curious where Oregon landed I can answer from a process perspective which is Oregon developed their definition in legislation so the definition was part of a legislative process and one of our definitions was also part of a legislative process and that was the broader definition so I can't speak to the details of they decided this one should be in and this one shouldn't the one interesting fact about Oregon is they decided to include psychiatry in their definition in their statutory definition of primary care I think it's part of the reason we wanted people to see what the differences were based on whether a narrower group or a broader group of specialties were included that at least as far as the specialty side of things were concerned it didn't make a huge amount of difference however I think that where there's probably going to be even more discussion when it comes to really the spending piece is that portion of services which you could really attribute to primary care which is a slightly different question than defining the specialties or the providers who are providing and that's why we want to continue our connection to some of the national researchers who are spending a lot of their time also focused on those discussions ok thank you Rachel don't go anywhere yet does any member of the public wish to dail on the Oregon part that is in this could you possibly help me understand better there's been her children but I'm trying to figure out what was their infrastructure did they have a shortage of primary care doctors there's so many things that can affect what they've done versus if we tried to do the same thing can you help me get a better visualization of that I can just tell you a little bit I'm not an expert in every aspect of what's going on in Oregon but I will say this they had a primary care initiative similar to the blueprint for health that was ongoing for a period of time so there was at least a segment of their policy and payer community and provider community come together around the idea of increasing investment in primary care they also had developed for Medicaid their coordinated care organization model which is kind of like an ACO model which applies to Medicaid and their expansion covered population and that model is very comprehensive in terms of care management on a person centered population based kind of model so I mentioned those two things just to say that Oregon has a history of interest in investment in primary care and have committed to a pretty robust care management model at least for their Medicaid population and I would suspect that that is one of the reasons that the primary care spending number for their Medicaid CCOs was almost twice what it was in the commercial sector but again until we have more longitudinal information and then can get into more of the detail which they will look at to see not only how much is being spent but what it's being spent on it's difficult to really draw a lot of other conclusions until we have some more data but that just reinforces our interest in having standardized measures and longitudinal use of those measures so that we'll have a better base of analysis to work from and draw conclusions. Okay, is there anyone else? Yeah, Susan? Hi, Susan Aaron I just wanted to follow up on one thing you just said in your last reply when you described the model in Oregon and you said it was kind of like an ACO model, could you clarify my understanding is that Oregon is not using accountable care organizations you know the corporate model but that they have a different kind of community collaborative that they're using? I didn't say ACO like but yes they call them coordinated care organizations and that is an amalgam they're regionally based and it's an amalgam of payers providers and community based organizations who come together in order to organize and provide care coordination services for almost the whole Medicaid population and almost all Medicaid services so it is a little more comprehensive in terms of the service package that they're responsible for it does include a specific inclusion of community based organizations but it also is only for the Medicaid population Okay anyone else? Thank you very much Rachel Thank you for inviting us and Will to be interested to monitor your work in this area as you go forward Is Sarah here? Nope So you work? Right Getting set up maybe I just wanted to say relevant to our last presentation that I think this is an area that we followed up on from our previous conversations around measuring primary care both in the ACO budget process but also a great review like with that and the stuff so thanks for getting that schedule It actually was interesting timing because I had to ask Sarah just a few weeks ago right to try to figure out what was the primary care standard in the box Just casually mentioned it Hi I'm Amber Health Services Researcher and I'm just here today to give a very high level series of examples about why data can be weird so the specific topic we're going to look at is comparing health expenditure data and we're looking at three examples which I think do a great job of highlighting kind of the trade-offs between different data sources so the first one is affectionately known as the Kaiser data and the first that comes from CMS that comes out about every five years called the State Health Care Expenditure Accounts or SHIA they're actually an extension of something called the National Health Care Expenditure Accounts which they are the official expenditure estimates that the Office of the Actuary at CMS puts out every year this state level stuff only comes out every five years largely due to the fact that it's based on the economic census the second data source we'll talk about is the Vermont Health Care Expenditure Analysis or the SHIA I think of it as the EA and that is something that we have and I'm glad Lori is here it goes back to what the 92 so it's a pretty long standing data source that we've you know it hasn't always been with the Green Mountain Care Board but it's been with whatever energy used to be the Green Mountain Care Board since then very specific data source the Dartmouth Atlas of Health which is part of the Dartmouth Institute in our neighbors to the east so yeah trade-offs, data are all about trade-offs so at a very high level I kind of think of three dimensions for comparing these data sources the first one is how comprehensive is it, is it really kind of a laser-like focus on a certain type of data or does it try to incorporate information across payer sources or different sources of information that may or may not be claims based so both the state health accounts and our expenditure analysis are pretty good for that sort of thing so the Dartmouth Atlas is just limited to Medicare claims for the most part so it's not going to give you as kind of an expansive picture of expenditures as the other two data sources of the two the Vermont Health Care Expenditure Analysis are most comprehensive because it includes things such as administrative expenses and can kind of where as the state health care accounts don't include admin or the net cost of insurance or investments in infrastructure or anything like that the next dimension is whether or not it's derived from a detail data source what I think of is it built from the bottom up or the top down so Shia is first and foremost the top down analysis so they take very high level stuff and try to break it out into pieces whereas both the Vermont Health Care Expenditure Analysis and the Dartmouth Atlas are from the bottom up there are components to the Vermont Health Care Expenditure Analysis that all we have is a kind of a statewide total but that's what it is so I think it's probably the second most granular of these data sources and the Dartmouth Atlas would be the most granular because it comes from detailed planes and finally is the comparability so there's always this tension in data do you want something that's really specific to what you're measuring or do you want something that you can compare in a wide variety of situations kind of the difference between something that's generalizable and specific and for that component the Health Care Expenditure Analysis is really a Vermont Vermont Jam it's hard to compare to other things the Dartmouth Atlas is Medicare data traditional fee for service Medicare data oh great for comparison across states that's probably the best in that dimension and the she is pretty darn good the only tradeoff there is so it goes into the next slide but since the Health Care Expenditure Accounts are based on national survey data it sometimes doesn't do a great job of representing Vermont or other small states so for instance the Economic Census has started sampling smaller businesses and a lot of our physician offices not associated with a hospital may not be included in that sample but the tradeoff there is that it's the same data across all the states so it's national survey data the big ones as I said were the Economic Census they also rely heavily on the American Hospital Association survey that they put out each year that Vermont's hospitals participate in they also use administrative data from like the VA and Medicaid and they do use some information from Medicare claims to help do some adjustments as well as IRS financial data on the other end of the spectrum is our expenditure analysis where we have very detailed information at our hands we've got V-Cures which is our all payer claims database which has information on Vermont residents who are covered by a payer with at least 200 lives commercially and then Medicare and Medicaid we have our hospital budget data which has got a lot of different looks at hospital expenditures than maybe the hospital survey does we also include information from the Vermont hospital health insurance survey which is something we put out periodically that has a lot of really good information specific to Vermont and then again Dartmouth ATLAS is mostly just Medicare claims they do bring in a few other data sources to help age adjust on their age and sex adjust their discharge rates they also get some hospital like geographic information from the hospital association survey too fast too slow good oh good so this is why it's known as the Kaiser data is because Kaiser did a great job of making the Shia information really easy to use and access so you can see that you are able to look at maps or trends and then here you have the different dimensions that it breaks it out to so there's you know hospital care physician and clinical services other professional services prescription etc etc and so here is why I think this came up as Vermont ranks number three in its hospital spend as measured by the Shia again this is just personal health care expenditures it's not going to include admin cost or private health insurance it doesn't include any government activities which is a big component in Vermont it also doesn't include investments and things such as research infrastructure equipment there also is an article in health affairs that kind of I think actually the board member Dr. Holmes circulated that article before the Kaiser data were circulated which kind of talks about the health care expenditure trends and how the real story seems to be and in states that chose to expand their Medicaid programs versus those that did not I'd like to caution people not to get too invested in the sector level estimates because they can be deceptive so for instance if you were just going to look at that to compare Connecticut and Vermont, Vermont ranks third in its hospital care spend and Connecticut rates 23rd and you know there's a 30% difference in there but overall they rank five and six so there's actually only a 3% difference overall in that spend so one of the confounders when you look at care in Vermont is how much of our delivery system is affiliated with our hospital networks so if you look at just the physician and clinical services we actually ranked 31st we look at kind of a good deal if you were to just look at that dimension so it's just something to be aware of the higher level we're trying to do these comparisons with this data source and so if you look at us we're the green line and you can see this is kind of our trend according to this data source this is the per capita spend over time compared to the US so up here is DC so the other thing that you'll notice whenever you're dealing with per capita estimates is they can be a lot funkier for a smaller population so the smaller your denominator is sensitive you can be to kind of noise but you will notice that there is a strong regional effect so that one on the top is the northeast not right below but then they're kind of chunked together here ranging from the midwest to the west and then in the middle there's the US average and whoever else in the south so you can see these curves all pretty much following the same trajectory and again that's really a factor that it's a bottom top down analysis these are built from the national estimates and then needed out which is why it looks so suspiciously similar so another thing so price can be a major confounder here it doesn't adjust for price, age, anything like that so yeah just another note on methodology so the way that they work is they start with their national estimates then they make this they try to figure out where the care was delivered or what we think of as a provider based analysis so whenever you're doing healthcare research there's a big decision you have to think about do you care about where the care was delivered or do you care about to whom it was delivered so the care delivered in Vermont is not necessarily going to look the same as the care delivered to Vermonters they're very different questions and ways to look at the problem so the way that Chia starts is it starts with where the care was delivered how much care was delivered in places with the Vermont zip code and then they adjust those to try and account for where people live so they basically make this 51 by 51 grid one the columns are where the person lives the rows are where the care was delivered and then they build up this ratio of what would happen to residents over where the care was delivered and they call this the net patient flow and Vermont is considered an exporter meaning that the amount of care delivered on behalf of Vermonters is greater than that delivered in Vermont so that we are actually at 112% we're a bit of an outlier the only state with a greater exporting ratio is Wyoming at 124% so again you might your first question I say is does that may be affected by sample size but essentially what they'll do is take the care delivered in Vermont for all the Medicare service sectors and multiply it by 112% and that's how they decide how much was delivered to Vermonters they just consider all Medicaid spend before Vermonters they don't do any adjustment there but for the private or all other coverage they incorporate H-COP which is a national database of discharges just for the case mix for the inpatient services and they do something similar for the physician claims but unfortunately the data source they use to adjust that information is something called market scam produced by Truvin and the blues don't submit to it so our major payer does not have information in there so any weighting they do might be biased they also are pretty upfront in their methodology that it's not the SHIA is not great for adjusting for seasonal migration so snowbirds get dinged they just associate it with the primary residents now I'm not as familiar with NCH the data source for Medicare information that they're using so I know in vcures it's always the mailing address on the first of the month but I'm not sure how exactly they're counting residency I couldn't find that in their documentation Can you explain the snowbird ding? Oh, yep so they're saying they're saying that since there's no way to account for more than one residency in their database if you go to Florida for three months but say don't change your mailing address all that spend is going to come back to Vermont even though you technically may have been a resident of Florida depending on your definition like tax they love talking about definitions of residency so what about all the people that come here to ski how are they treated? we would get the reverse effect for that so yeah yeah I wouldn't know I wouldn't hazard to guess the magnitude of that effect my hunch is that since one is largely Medicare I would guess that the snowbirds would ding us more than the skiers but that's just thinking off the top of my head don't move me sorry so the expenditure analysis again it's something that we do here and I think that the next one will be ready in the spring here coming up within the next month or so next couple of months I should say but this is just an excerpt you can see the most recent presentation here there's the URL but we do make sure to compare our per capita expenses to the NHEAs per capita expenses over time we also look at the percentage of domestic products and do those kind of comparisons it's a very rich data source I really think that there's a lot of really great stuff in there and Lori and her team work super hard to get it prepared every year and one thing that the data team would like to assist with and is currently trying to figure out implementation strategy is to make it more Kaiser format more of a kind of friendly dashboard to make it more accessible to people I think that will be something that we can add to it but if you look at the trends between our expenditure analysis and the CMS expenditure analysis you know I would say that's pretty similar trend I mean I know there's a gap here but if you look at the slopes of these lines the rise over run you know the estimated increase over time is only about 100 bucks which I know adds up over time but you know I would say that they're not measuring the same thing but they're measuring something that's correlated so you know it would make sense to me that you know given the bottom up and top down nature that I'd be more alarmed if they were exactly the same frankly this is the total spend yeah this is the per capita spend actually so the per capita total spend comparing those two data sources um can you break that off my hospital service area we certainly can break out components of it by hospital service area for the expenditure analysis and that's one of the enhancements that we're actively thinking about how to build in again the question is going to be are you interested in the service area of where the care was delivered or where people live because again that really makes a difference and so we would want to do some thinking about what and why we're looking for that breakout for example this morning we had a hospital say they were the third least expensive hospital in the state and I'm just wondering what data source that was brought yeah I can follow up on that I probably missed that part of the presentation but yeah and the catchment areas are a whole another animal and that's like what I've been talking to other states about this where one of care patterns don't respect state lines which is okay with v-cures but that's not going to tell the whole story either so you know there are exciting challenges that will be happy to help address I'm glad you're excited Sarah we can give you lots of excitement and finally just a word about the Dartmouth Atlas for Health this is a really incredible tool I don't know if you've had time to hang out on their website DartmouthAtlas.org but they do a really fantastic job of talking about different types of service they do some chronic conditions work and this is really provider basis so what they do is they look at what is called hospital referral regions and what they say is okay so for this Burlington hospital referral region what's this is you kind of think of it as a catchment area but it's like where do people go for the majority of their neurology and I think it's cardiac complicated procedures and then they kind of build these boundaries that way so it's much different than a hospital or primary care service area and then they adjust it for age sex and if you choose price this is just adjusted for age and sex and you can see that on the scale here Burlington's HSA is one of the better deals in the region compared to you know things get you know red is more in this scale so and you'll see Vermont really is mostly between two hospital referral regions can anyone guess the other one but we have a little bit Albany and a little bit of Springfield, Massachusetts so we really kind of have four referral regions that touch the region but there's two major players and so one of the things you can look at is trends over time so this is just again Medicare but it's the Medicare reimbursement so the top line is Springfield Mass Albany, New York is the blue line Burlington, Vermont is the red line and then Lebanon New Hampshire is the other one so this is kind of the reimbursement is per and rolly adjusted again adjusting for age sex and race at the HRR level so yeah it's hard to know what they have a lot of really wonderful information and stuff that I would love to replicate on an all-payer basis you know as we're gearing up our analytical efforts here so yeah just to kind of summarize no data source is perfect which is why I'm always so excited and the best way to answer a question it depends on the question or questions that we're asking and you know that's kind of what we're here for is to help you answer questions as best we can and then we'll give you a long list of boring caveats that you can stick in your footnote but I just think that it's really one takeaway I would love for people to always keep in mind is whether you're looking on who the care was delivered to or where it was delivered totally different worlds so that's why like when you talk about ACO budgets versus hospital budgets it's completely kind of different lenses when you're trying to think about this stuff so yeah any questions, comments, concerns that really is a follow up of what you just said in terms of the question does matter and I think one of the areas that we as a board have been talking about for a while and I personally have been talking about probably since the day I got here is being able to crosswalk across the different regulatory processes so I think I would love for you to think about that as you're considering sort of recommendations about different analytics and different ways we should be using data if ACO budgets are different than hospital budgets but we need to be able to crosswalk between the two more exciting any time Sarah any time you need excitement let me know other questions or comments from the board now we'll open it up to the public any member of the public have a question or a comment Mark Mark thank you this has been very helpful and particularly on the Shia data and I think getting to Robin's point is the hospitals really want to understand what part of the cost curve they are driving versus the assumption that anything in health care spend is driven by the hospital okay so just as an example on the Shia data their expense base for the hospital spend understanding that she cautioned against looking at sectors was 2.0 yeah so is it is 2.9 billion dollars okay our green mountain care board budget for all of the hospitals for that same time period was 2.1 billion dollars that included physician services that included some net patient some skilled nursing facilities and just by my rough math if you took those out that would knock it down to about 1.7 billion that would be left and then obviously you know the population number hopefully there's no differences I think the number they use was 627,000 so that difference is very important and to understand where those numbers are coming from if we are serious about lowering the cost curve because the focus needs to be what's driving more of the cost just the assumption then that it's the hospitals that are driving that cost and what makes me feel that there might be an imbalance in that relationship just when I take a look at the University of Vermont Medical Center and compare it to its academic peers through double AMC they are in the top quartile of cost and quality taking a look at all of Vermont and taking a look at the Commonwealth report Vermont is ranked number one overall number one over access and affordability number one in prevention and treatment number 12 in avoidable hospital use and cost number five in healthy life number two in equity so it's just on one hand we're hearing the hospitals are adding more and more cost to the system are driving up the premiums and we go to these other you know resources and that 112% makes some sense now because how do you regulate the care that's going outside of your state one of the goals of the all payer model that's counter protective to the patient service revenue point okay and and then just a very specific number for the University of Vermont Medical Center in the most recent cost survey which is through these AMC all academic medical centers there's about 105 and there was 92 you know respondents were there were there well the wage index and case mix index adjusted cost per adjusted discharge was 86 46 that was in the top quartile the most favorable cost they're 50% of the system so if you have 50% of the system in the top quartile of the other academic medical centers it's hard to make these correlations so anything we can do to draw better correlations to it exactly what's driving the cost where the focus needs to be to drive the cost down and once again I'm going to throw my worry out there that if all of that is put on the burden of the hospitals to make up this difference and they are not the driving factor of this I worry about the financial sustainability of the whole hospital system so I think it's important to understand all of these differences in the correlation to the all-payer model but I mean very important thank you I would just say Mark, the rule of substance you've got to go where the money is and the hospitals aren't where the largest share of the money is well there's some exporting there I think we need to get an answer on that we are one of the highest export states in the nation so okay, other questions if I go back to the slide that shows that one I remember a conversation around I have to have a certain size population to even deliver some services thinking that way when I look at this map I have the difference in cost from region to region but what could this possibly tell me in terms of the University of Vermont Medical Center that it needs that region at least that region to even have the cost that it has or can you break it down to look at that and that way because if you're going to talk about driving costs doesn't that become part of the equation and the data my cost is also associated with how many I'm actually serving do you have any sense to me yeah, I think this is a little bit different way to look at it but I think what you're saying is that maybe we should build a brain surgery center in Newport well yeah, but you also mentioned the cross the cross boundary issue those are different cultures those are different right and I think that it kind of speaks to access as well because if you notice, there's not even enough information up here so the distance that people have to travel to get these services might kind of confound the hospital for other reasons okay, alright, thank you okay, any other member of the problem not, thank you very much Sarah is there any old business to come before the board if you did not would you like to make a motion? are you going to move on the minutes today? these are the minutes from last meeting that we didn't they were not done yet, they had to be corrected this morning I believe they're all correct right now correct everybody's review doesn't have to yep so I move that we approve the minutes on March 21st I second so it's been moved and seconded to approve the minutes of Monday, March 21st without any corrections additions or deletions is there any discussion? if not all those in favor signify by saying aye aye, any opposed those minutes are approved so hearing no old business is there any new business to come before the board hearing none, is there a motion to adjourn? I move to adjourn I don't want to adjourn I move to adjourn all those in favor signify by saying aye aye, any opposed