own ideas on how to increase value and how they are increasing value in the health care system. >> we believe by bringing people together with different perspectives we can improve health care through evidence and collaboration. we have a wonderful audience today, i recognize some of you out there. we want to bring you in to the conversation. so at the end of the program, we've allowed for about 30 minutes of q & a, if you could fill out the cards in your packets somebody will be around to pick them up, and then we'll use them to ask the panelists. we have had a last minute change to the agenda. bob ricemiller was injured

playing hockey. he is not able to join us. mike will cover some of the points he was going to make. why there's a call to action to increase value in our health care system. i'll start first by introducing each panelist and sit down, and then go on. so i'll start with mike, who is one of the nations most highly regarded economists. really what distinguishes mike, not only is he broil yant on the theory and research. probably more than any other economist, he understands how markets work. and he leads much of the health care policy work coming out of harvard medical school, he also sits on harvard's benefit's committee, where he learns firsthand why markets are not always efficient. is that fair to say?

>> he also serves on cbo's panel of health advisers. a member of the massachusetts health connecter board of directors and vice chair of medpac. we have the fortune of having him serve on our advisory board. he's one of the pioneers of value based insurance design. and the go to researcher for medicare programs. please invite me in joining mike to the panel. >> i am thrilled to be here. nicum always does a good job, nancy is too nice in her introductions. it's nice to see you all here. i'm going to talk broadly about payment reform, a little bit of benefit design.

because bob was going to talk about the budget, i want to start with some comments about the budget. the quiz question. i'm not sure boch would have had this, what do these decades have in common. and the answer is, in every one health care spending growth exceeded health care growth. by varying amounts in the '70s, we were 2% faster. it was considered a good year. good decade for health care spending growth. it only exceeded income by 1.6%age points. that's a real fundamental challenge if health care is growing much more quickly than national income. we've had, and i think one reason why it's great to be here today, an ongoing national discussion about what to do in the health care sector, i'm going to largely talk about broad ideas. i want to make an important semantic distinction. in many debates when people talk

about the challenges we face, they're talking about total spending in the country. national health expenditures. there are other challenges. they mean the government is spending too much. those are different issues and different solutions for how you're trying to address the total health care spending in america relative to what the government is spending. i'll make a quick comment about medicare's challenge. in 2015, roughly speaking, medicare was 3 1/2, a little more percent of gdp. which you could take that for what you will. if health care spending growth continued to exceed income growth by two percentage points, which is roughly the average, by

2035, we would be up to 8% of gdp. if we're really successful and able to reduce health care spending growth to one percentage 1 percentage point faster than income medicare spending would be a little more than 6% of gdp. if we were stunningly successful -- maybe i should emphasize stunningly success, and health care growth rose at the rate of gdp, medicare would still rise as a share of overall spending. and the reason is because we have more medicare beneficiaries. so we have fundamentally two problems. the policy solutions for which are different. one of them is there has been rising spending for beneficiary historically in the health care system. and the second is in medicare we're struggling with how to finance a growing portion of our population on medicare.

it should not be surprising that if we have more medicare beneficiaries per worker, the workers will typically have to pay more to fund those beneficiaries. that's not rocket science math. but the implications with how you deal with a rising population is very different with how you deal with rising spending. i think there's been some discussion about medicaid in the statehouses and in d.c. and as medicaid spending grows, it crowds out a whole bunch of other things at the state level. so there's a huge challenge not just about how we finance medicare but how we finance medicaid. to give you some rough numbers, so basically there's other parts of the budget we could think about cutting from. i don't want to go into broad fiscal spending discussions. but if you try to fund spending growth through taxes alone and

you raised everybody's income tax proportionally and health care spending grew at 1% faster than income -- remember the percentage rate is 2% faster, tax rates would have to rise for the highest income group in the neighborhood of 7%. i don't think taxes are going to be 70% in 2060. and it is admittedly true the power of compounding can make numbers seem enormous. as nancy mentioned i still actually chair -- i had the honor of chairing, maybe the misfern, but nevertheless the honor of chairing the harvard

university advisory. and we could see spending growth that put huge pressure on the wages and salaries of people that worked at harvard university. as i talk about the federal spending and state spending problem, this is also a huge private sector issue on how we can control spending growth. what i want to spend the rest of my time on is solutions. just to be clear, i use that word loosely, because i'm not going to have a great magic bullet at the end. if that's what you came for, luckily we have other panelists. one is payment reform and the other what i call consumer strategies or benefit design, things related to engaging consumers through different ways but largely through insurance packages. it turns out one strategy you might use to pay less is to pay less. it scores really well with the cbo. and you've seen an enormous amount that in recent health

reform discussions where an increase in fees to physicians and health care facilities are lower than they have been in the past. there's a huge amount of energy what is known as the alternative payment model, the way of paying differently. and the two big types are episode bundles and a population based payment model. in the federal context that's largely acos. we had a system for paying physicians that was called sustainable growth rate systems which was many things but not sustainable. we now have a rule called macrowhich governs the way physicians are paid, and in it there's a portions called miffs, which is pay for value

component. the thing to understand is the scheduled fee increases for physicians are essentially flat in nominal terms for decades to come. there's some bonuses and stuff that go in there. the key thing to understand is by roughly speaking 2042 sort of the base level of physician fees are scheduled to rise by a little bit less than 10% in nominal terms. slowly over time there would be a reduction in the real value in the fees to physicians. and through the aca there's similar types of adjustments called the productivity adjustments to payment to facilities, where their fees are scheduled to go up less than their input prices. and that's i believe going to bea challenge for provider systems in decades to come in the fee for providers system. what we're trying to do is build

a somewhat different type of payment model where physicians in the delivery system more broadly can capture of the efficiencies. and by pulling those efficiencies out and allowing those provider systems to share in some of those savings, they can perform better financially than if they're just having their fees eroded by inflation over the next 20, 30 years. at least that is the theory how these types of models should work. i was lyening to a very well-known venture captist who said these flexible payment models allow innovators to make changes in ways that aren'tdous torted by needing to create revenue. i'm going to go through those payment models. they have three characteristics. the first one is they tend to transfer risk to providers. that risk means they share some

of those efficiencies. it is always hard when you change sectors business model. often these new models go under the monitor of value based payment. and you see this all over the place. reconciling, prevention and value, from volume to value. the word value is a very commonly used word now in all these things. so the begs the question what do we mean by value? my personal view, may be a little contrary, is the word value is a little bit of the sugar that makes the medicine go down. what we're really trying to do is to shift some of the responsibility and accountability providing high quality care to providers and

then hold them accountable for the financial components shifting risk. and we call it value because it sounds much more appealing than sort of risk-based models. a lot of what is going on here is transferring risk and responsibility down to the delivery system. episode-based payments have gotten a lot of attention. i don't have time to go through an exhaustive literature review, although i would love to do that, actually in my longer talks. in general i think the evidence is pretty clear that they save money. there are examples where they save large amounts of money, or at least have been purported to save large amounts of money. on average they save 5-ish%. at least that's what we've seen. there's a lot of concern there could be an increase in volume of episodes. we're not sure if that's true. we haven't seen a huge impact on

quality. there haven't been big consistent impacts on quality. there's several concerns with episode-based payment. one of them is that the episode models work well for certain types of conditions, hips, knees, maybe some other things. but they probably have in this country a lot of them deal with complicated, interrelated chronic conditions that are much harder to put into episode models. and it's unclear how far you could go capturing the total spend of a nation in episode based framework. and the second one i mentioned was induced use. let me jump to population based models. it used boo to be called capitation. we don't use that word anymore. nevertheless, the delivery system gets a fixed fee to pay for population of patients over time.

all of these or almost all of them are built for a fee for service chassis. that means during the year the fee is flowing in a various way. so, again, i would love to spend an hour and talk about the evidence. it's been controversial. if you follow twitter, at least half the places where i am on twitter is arguing this point. i believe the evidence is pretty clear population based payment models reduce spending by a small amount. many people get disappointed. by a small amount it differs against populations. the private sector clearly does better. and part of the reason is because there's wide variation of prices in the private sector. and in these models it turns out the positions, the organizations that accept this risk, they can

shop better and accept money by lower based settings and providers. it seems clear the results improve over time. and it also seems clear at least in medicare the independent physician organizations do somewhat better. that is a generalization because i'm on academic. so i'm paid to generalize. as a general rule it seems to be the case if you want to save money by keeping out of the hospital, it helps if you're not a hospital. so the other thing that's really important to understand, and i don't know why this is so hard to explain. it seems that i keep saying this and it doesn't sink in. in a shared savings model, the savings get shared. just so you understand how that works. so if you save $100 in less spending, you're not going to keep that whole $100 because you shared it. that's how sharing works. so if you look after a year

you're never going to do as well as you would have done after savings. the shared savings are an incentive to induce the savings. if you didn't share, you wouldn't have any savings to then share. i think we have to worry about trying to create a health care system that is working in 2019. just to be clear, i really want the health care system to work in 2019. but i'm much more interested in us getting on a path are where the health care system will work in 2025 and 2030. and if we spend all of our time trying to get a big win in 2019, we risk abandoning a path that might get us somewhere successful in 2025. so the reason why i have been supportive of these models is not because the evidence suggests huge savings. although i must admit if you're saving a little bit of money and improving quality, in my book that's a win. it might not be a slam dunk win,

but that's a win. and the reason why i think we should keep moving on this path is because if we get all the regulations right, it enables the delivery system to control their destiny and to capture some of the efficiencies, which i think is ultimately where we'll want to go. some quick caveats, the details matter. remember the details of these programs is very different. and the other thing that's important to understand is execution matters. it's not something that puts into aco, it's how it gets implemented, how the organizations that are doing this work accomplishes their task ends up being really important. my summary of this whole thing, we want a mansion, we have a dilapidated house, we know how to build a tiny house. i don't know if anybody watches tiny house, but the point remains it's still a house and

hopefully over time we'll be able to put on some additions. comparing episode of population based payments. and they may be able to work together, but some quick summaries. if you save 10% on 5% of care, you haven't saved much money on a population basis. the other models are broader. but the concern is not all geographic areas in the country are yet ready to support population based payment prch for a while we're going to have a mix of these models moving on, and also episode models engage specialists better. other ones tend to be focused on organization of primary care provider. you can take that as better or worse. it's just a distinction. should they continue? i believe they should because in part because of macra's current

law. my general view is when money gets tight, and my hope my beginning slides convinced you money will be tight, the providers are going to want to control the money. and the overarching theme behind these payment models is the mechanism to capture and share some of the savings they create. now i'm going to shift gears. understand one thing in science is to mitigate risk. sometimes we forget that part of insurance, so we're balancing risk spending and incentives in a variety of ways. it's not simply to lower premiums. it is relatively easy to make premiums go down if you make the patient pay more at service. that is shifting from a premium to a premium service which adds

risk, it does not diminish risk. the goal of cost sharing is not to tax the sick. i will just speak for harvard. our goal is not to meet the health care challenges we face by challenging our employees that have cancer or heart disease, even wonderful things like having babies, our goal is not to charge young families to support our health care spending. the reason why we have cost sharing is we're trying to improve the incentatives in two particular ways. one of them is we want to reduce excess utilization. there is waste in the system, and we'd like to improve price shopping. you'd like to use benefit design to encourage some of that price shopping. so a few options. there's high deidable plans, high copays, high cost insurance plans, essentially models that

charge patients more, those tend to be very blunt instruments for a variety of reasons. very sick people face no incentatives for cost sharing. most of the newer more advanced benefit designs try to target a purpose, either price shopping or inducing over utilization. so reference pricing is the idea you pick a price and if an employee goes to a higher priced provider, that steers them -- tiered networks is the same concept. it's for all of your admissions or all of your care. so both reference pricing and tiered networks are about steering patients to lower prices of higher value often. the value based insurance design concept is the idea of trying to

align co'pa-pays with value. my colleague would be frustrated with me if i didn't mention some of that care we get is high value. other care is low value, and we would rather discourage people from using that care. so the basic results of all of these benefit design studies it is thankful to an economist good news that people somebody to cost sharing. financial incentives matter. they shift the site of care, patients in reference pricing in tiered network plans move to lower priced models. not all of them, a relatively small percentage of them, but there's movement associated with that. to give you some idea, there's

potentially large savings in a referenced priced plodal on hips or knees. in a tiered model it represents 5% savings on a population basis. depending on your savings 5% to 14% depending on use. how much emphasis have you put on discouraging use of low value services. so how you design is going to determine the financial implications of the plan. some basic concern, and this is one that's really important and i have to say a little bit disturbing to an economist. i was gloating before, but now i'm a little sheepish. cost sharing does influence a patient's behavior. they reduce their use of high value care and low value care in

about the same proportion. so they reduce the use of preventive service, their use of high value pharmaceuticals. if you charge people more for those things, they actually use les of those high value things. they're not that good at discriminating. that's actually one of the motivations for insurance based design. you can use cost sharing -- i don't want to use the word manipulate buzz it make meez soume sound evil. there's another concern with all these benefit design things that you're transferring risk. if you charge people a lot, you're undoing that. of course one thing we've been worried about is disparities. if you charge people out-of-pocket you really have to worry that's going to impact some populations more than other populations. so a consumer ori wanted strategy always has to grapple

with a disparity concern particularly given people aren't always the best decision makers in terms of what care they should get. so a few final thoughts. the first one is keep it simple. it is really easy to come with dramatically complexfluenc solu. the quality measurement system we have in this country -- and i should say i'm a big fan of quality measurement -- but it has become a big complex entity that creates cost throughout the system. i think as a guiding principle we should do our best to try and keep the system simple to avoid this crushing administrative distraction we could create if we aren't careful. and the last point and maybe the

most important one is don't be so impatient. if you came here today and whatever talks you're going to next week or the week after that, if you're going hoping there's going to be some magic solution and if we get rid of whatever we're doing now and put in this other spectacular thing, that's not the way the system is going to work. by and large my view is that the road to success is always under construction, and we should be happy with small wins and incremental improvements driving in a direction we think we will promote value using inceptatives to the delivery system and patient benefit design and hopefully we'll get to a place that's fisically sustainable. one way or another we are going to control health care spending growth. because in the end math is going to win. but we have to figure a way that's going to do that in a way that maintain quality of care for folks. i think we're on a reasonable

path, and hopefully with continued work we'll keep moving in the right direction. so thank you very much. [ applause ] >> mike, thank you. that was really terrific, very thorough and actually very clear, too, on a very complicated subject. so thank you. now it's my great pleasure to introduce curtis barnett, the president and ceo of arkansas blue cross blue shield, a non-profit mutual insurance company, which is the largest health insurer in the state of arkansas with about 2 million members. curtis is really a champion for improving the health of his fellow citizens in arkansas, the most vulnerable people there, the most vulnerable population. and under his leadership the arkansas blue cross blue shield has instituted a number of programs around opioid

addiction, health literacy and even food scarcity. but he's here to talk about one of his key components to his vigsz of making health care more assessable and affordable. and that's a movement to alternative care models in collaboration with the state and federal government. and they've had a remarkable amount of success in arkansas, and it's really served as a model for cms and for other states. as you can hear today curtis is pretty passionate about building a better health care system, and we're really pleased to have him with us today. thank you, curtis. [ applause ] >> thank you, nancy. ask i think it's very fit fting for me to follow michael in our comments today because michael spent a considerable amount of time in my home state studying

the effects of many of the initiatives i'm going to be talking with you about this afternoon. when i joined arkansas blue cross and blue shield in 1993 i worked with a product called the primary care network. that was a product we made available to large employers and located in rural parts of our state. these were communities with populations as small as 3,000 to as large as 20,000. the communities that had primary care physician clinics and at least one large employer, sometimes more. and these tended to be fortune 500 companies with a plant or a mill located in arkansas. these were gatekeeper type models so that care for the employees and for their family members were directed to the primary care physician. and those physicians treated those patients as well as coordinated their care and then referred them for any additional services. the physicians were under financial risk for how the plans performed. and one of my main jobs was to

work together on a regular basis with employer leadership and physician leadership and bring them together to manage those programs. and we would go through expensive data analysis to do that. and you can imagine in 1993 these were mounds of data and mounds of reports we would work our way through. what we were trying to do in that setting is to look at which primary care physicians were performing the best, which are the most effective and efficient hospitals and specialists for them to refer care to. and we were aligned together to manage those health care costs because we all had an interest in keeping those jobs in those local communities. and we were really trying to work towards two objectives. one was to establish the primary care physician as the care coordinator, and then the second was to identify and reward value. now, fast forward 24 years and while we do not continue to have that particular product in our

state, as an industry we continue to work towards those objectives today. it's what consums us almost on a day in and day out basis. and i would say over that 24 years we have seen progress towards those two oojives. and especially in the last five or six years. and some of that and i think a big part of that has been we've had a new partner involved with us as we've pursued those objective, and that partner has been the public sector. what i want to do today is give you a look at what public and private collaboration has looked like in the state of arkansas. i want to go over the initiatives we pursued, talk about some of the support that's needed to make them work and then just end by talking a little bit about really how all of this work has influenced our view and our approach to driving the health care system towards value going forward. the health care challenges in

arkansas are not too different than those you see in other rural states, especially southern rural states. it's hard to navigate. it's fragmented. there's very limited visibility across the spectrum of care thinking about quality, thinking about cost. health care spending is growing. it's growing at rates that it's severely straining the public and private payers in our states. and the health status tends to be relatively poor. we rank at or near the bottom in a number of key health measures. whether it be obesity, prenatal care or smoking, we tend to rank toward the bottle. but wii working together public and private sectors have begun to transform the health care system in the state of arkansas. this is the framework for how wave been going about that work. it's called the arkansas payment imp provement initiative. and there are two main

components to this. the first is a patient centered medical home model we've had in place with our company going back to 2010, and it ties directly into the comprehensive primary care initiative sponsored by the federal government. and then episodes of care component, which we put in place in 2012. and all of this work was helped tremendously in 2013 when our state received a state innovation model grant from cms. i think it's important to note that arkansas payment imp provement initiative is not just a public/private collaboration. it truly is a multipayer initiative. as you can see all the payers in our state are involved in this program at some level and have been supportive as well as major employer groups who have a commitment towards value based care have been partners in this as well. i want to start by talking just

a little bit about our patient centered medical home approach. these are the objectives of a well-designed patient centered medical home. to achieve these objectives a practice has to go through significant transformation. for example, they must commit to a team approach to care. they must identify their top 10% priority patients, have care plans, use electronic health records, arrange for 24/7 live access to care going forward. so all of these are important activities that they have to undertake in order to transform their practices to be patient centered medical homes. in return, these practices do receive upfront payments, which help fund those transformation activities. in 2010 our company put a stake in the ground, and we said that if we're going to have a

sustainable health care system going forward we've got to have a strong primary care infrastructure. so we embarked upon what turned out to be a two-year pilot program to see if we could transform primary care physician practices into patient centered medical homes in asesting that process. and these are the goals that we sought to achieve. in only two years we did see some very good results. we saw hospital readmission rates for those patients who were attributed to patient centered medical home go down. we saw emergency room visits and emergency room medical costs go down as well. we saw appropriate use of the emergency room go up as well as the generic drug prescribing rates. we also came away from this pilot with some very significant and valuable lessons learned. and probably chief among those is the need for payer alignment when you're pursuing these types

of initiatives. we tend to be the market share leader in the state of arkansas. and we realized very quickly any one private payer would not represent enough volume to get the practices, to make the investments to transform their practices in the way they needed to occur. so we understood that very quickly that without much more volume, these tended to be a one-off effort and would have a very short shelf life. you can also see we had other important lessons learned as well. so in 2012 cms opened up the opportunity for marx to apply for the comprehensive primary care program, we took our lessoned learned, we joined with arkansas medicaid and our states surgeon general and health care and political leaders in our state.

and we made application and we seized our opportunity at that time. you can see in the period from 2012 to 2016 we were one of seven markets that were selected, we were one of four that were statewide. we had 68 practices selected, 268 providers and five payers participated including arkansas medicaid. and again, that was critical to have the patient volumes that were needed. because now we not had the patient volume but we could offer the incentives necessary gnat would be sufficient enough for these primary care practices to make investment in infrastructure and changes that were needed to transform their clinics. the volumes were absolutely critical to doing that. over the course of the time period that we participated in the cpc and what is now called the cpc classic program, you can see that we did participate throughout that time period.

we saw improvements in patient satisfaction and experience of care during that time period. we saw reductions in hospital admission rates by those patients part of the patient medical home program by 15%, which led the country. we saw improvements in quality metrics and the information that was captured there. and arkansas in 2015 and 2016 did have the net savings as a result of this program. so we came away feeling like it was very much a success. the experience that we've had in cpc classic program led us to want to pursue, to be part of the cpc plus program, which went into effect january 1 of 2017. so we're a part of that as well. we're one of 14 regions across the country that have been

selected for cpc plus. we had 182 practices that are now part of that, and we're up to seven payers who are now supporting this initiative. that's important as well, because we think this will allow us to have a much greater impact going forward. and now i want to turn my attention and talk a little bit about the episodes of tier program. we work very closely with arkansas and medicaid to basically develop and implement the episodes of care program. the purpose of this program was to improve quality and to basically reduce the variation in the types of procedures that were used to treat acute conditions. we felt it was a very important movement in the right direction. for each of the episodes a clinical leader is identified, which is called the principle accountable provider. and so that individual is

identified. there are quality standards that were established. and then there were thresholds also established, which laid out the commendable, acceptable and unacceptable cost levels for each of those episodes. and then we also provide regular recording back to the clinical leads so they can look at the episodes across the entire spectrum of care and look at not only how the quality measures are performing but also the agigation with those different episodes. today we have 22 episodes of care. each of the payers is free to decide which of the episodes they want to implement based upon the population that they serve. for arkansas blue cross we've implemented 14 episodes of care. this is an example of how the payment model works for the episodes of care program. and you can see that at the end of the performance period

average cost per episode is calculated for each of the clinical leads. this is usually at the one-year period of time and that there are commendable levels acceptable and acceptable levels. those who perform and meet the commendable level do identify for game sharing, and those who perform above the unacceptable level have to share the cost of that program. so a portion of their payment is then recouped as a result of that. so putting the program in place we have seen what the intended effects were meant to be, which we've seen variation reduced, wave seen quality improved. so we feel this has been an important movement in our state. these are some high level results from a couple of the episodes we've had in place at least for our plan, the per

natal, you can see quality cost down and also for total hip replacement. i do want to point out if you were to do that assuming that the cost would be growing year over year do the inflationary factors i think the cost reduction would have an even greater impact. and then i did mention this earlier but arkansas was one of the several states who had been selected for a state innovation level grant. our state did receive a grant of $42 million to help fund the arkansas payment improvement initiative, the activities that i've been talking about and helped develop those episodes, help develop our patient centered medical home process. and we think that partnership is part of this is certainly indicative of how public and private organizations should work together. as you can see we've been having

an impact in our state. but i don't think i can overemphasize the importance of data and communications when supporting these types of programs. when we implemented these programs, we rolled them out, we worked very closely with arkansas and medicaid especially on the episodes program. and we held 21 public work group meetings and eight locations around our state, and we received feedback from over 500 providers, patients and other stakeholders who helped really shape how these models would look and how they would work. ongoing communication is absolutely critical as well. we have teams, each of the payers have teams involved in this on a regular basis. we have weekly, monthly, quarterly meetings. and it's not just the payers working together in the program. the arkansas hospital association, the arkansas

medical society as well as the practicing -- the participating physician practices as well. and what we're doing in those sessions, we're sharing challenges, talking about program changes. and as much as anything we're talking about best practices. data is also absolutely critical to the success of these program. we decided early on that if we were to try to build whole new technology tools from scratch they would add tremendous cost to this. so arkansas blue cross made available our advanced information network very early on in this program. it was already being used by 99% of providers in arkansas who had already had a presence in those offices. so we were able to use that as really the way to push information ask data out to all these provider practices and especially all the reporting. we think that was absolutely critical to keep providers from having to log into multiple

systems to get them to feel much more comfortable with the program. we provided a variety of dash boards to our participating practices as part of this initiative, of the drilled down capabilities that they can see exactly how they are performing and where there are potential areas for improvement. one of the things we did early on also is we established a dedicated resource center that providers could contact and with we would also out reach to them to help them understand the reporting and help them understand how to use that information to help drive practice improvement. so i want to end by really going back to my comments at the outset. i feel like we've made in our state very good progress at those objectives i've talked about. and that is to improve care coordination and to identify and reward value. we think a lat of that progress has really occurred as a result

of our collaboration between the public and private sectors, by having arkansas and medicaid, by working with cms, we've been able to bring together a much bigger impact than we could have before. at arkansas blue cross we're convinced we need to continue to move this market forward value based payment. and we think that's absolutely critical to having a sustainable health care system going forward. and we also believe that all the initiatives that we've worked on, that i've spent the last few minutes talking about, some aco-like models we've worked with providers on in our state called collaborative health initiatives, we feel all those different efforts have really helped lay a foundation that we can continue to pursue this going forward, and we're working on those plans today. but we also want to continue to collaborate with public programs. and i encourage public agencies whether it be state or at the

federal level to continue to view the private payers at partner partners invasion. we think that's critical as well. we have seen first-hand in our state that by working better together, by combining our patient volumes, by combining our different perspectives, by combining our different areas of expertise, by combining our resources that we can pursue common strategies that can be scaled to address common challenges. and we believe that by working better together that we can have a much greater impact than any one company or any one agency can have working alone. thank you. [ applause ] >> curtis, thank you so much for your remarks and really for your

collaboration and leadership with the public sector. and we look forward to getting future updates on how you all are continuing to make progress. so now it's my great pleasure to introduce dave around, president and ceo of blue cross blue shield of western new york and blue shield of north eastern new york. dave leads the region's largest health insurer which is community based non-profit plan with one million members. and dave is really a forward thinking intrau pruneurand executive. and he's leading a number of programs to help new yorkers. including an out reach campaign to address the opioid epidemic and a home-based campaign with behavioral and social supports. as you'll hear today dave is committed to strengthsing the primary care to support improved

care, and improved costs. so please join me in welcoming dave to the podium. [ applause ] >> well, good afternoon. plegs do pleasure to be here. i've enjoyed the comments of our prior speakers, and i think the sequencing is good. michael is more of a macro level, and i think curtis is talking specifically to those segments. i want to spend a little time today talking about a very specific market-based, value' based initiative we have brought to that market best practice. it's very regionalized, and it

is also focused on the shortage of primary care that we have in that area of western new york. so i'll give you some context around the population that i'll be talking about and the area. this is us headquartered in western new york. we operate at our headquarters in buffalo. western new york is an area that actually had some shrinkage of population over the past decades, and its population is aging a little bit. so those are factors in the efforts that we have made to transform that population. this is where we're located. again, i'm going to talk specifically about western new york. although we will begin to rollout similar processes in other locations as well. this is the region that we

focused on. it's an eight-county service area. as i mentioned it has static population trends. there are two major hospital systems in that area and controlling about 80% of in-patient care. and there's one distinct cancer specialty hospital is in that area as well. fairly distinct territories that are hospital driven, very doctor centric not patient centric. and as we have evolved into more of a patient entric and consumer oriented segment in health care, we have a need to flip that around. we looked at the area and said how can we begin to transform the provider systems in this specific area to make it more

patient centric and to bring value based contracting to the region. there was a very high degree or concentration of specialists, also not that unusual when you have hospital driven communities. and they were very much a late adopter to any form of reimbursement other than fifa service. our program is called best practice and really had two main objectives. one, to transform into a value based and provide and realign incentives with our provider system as both michael and curtis talked about. and also we have a pretty critical shortage of primary care physicians in western new york. and you can see the numbers there showing our deficit. and they are shrinking all the time, not unusual in a number of communities around the country

because they had not been supported either economically or within their practices in a way that makes primary care their preferred specialty, if you will, in the provider system. as i mentioned we had an aging population, and the result of that the model, the fifa service model called primary care physicians to be reimbursed and their income level as such they have a general incentive to move to specialties. and in a volume based reimbursement model the only way that primary care physicians can progress is to do more volume, which is exactly the opposite of the incentives we wanted to provide. so the current relationship is

broken. much of this you know and you've urd heard. the responsibility of our system in a fiefa service world really falls on the patient. pcps, our priprimary care physicians and specialists do not collaborate as often as they should, often electronic records and other forms of patient records don't follow as concurrently and as accurately as they should. that results in redundant testing and procedures. and there's also in a fifa service environment an inability to reward pcps. essentially they're reimbursed in the same way. and the result we have today is higher cost, lower patient satisfaction scores, very difficult to measure quality scores in that environment. and so we have moved forward towards best practice. this is primary care as we felt it should be.

it's not perfect. and we did not want to go from a to z all in one year. i don't think that would be possible, but i think it's a really good step in the right direction. first of all, pcp coordinates care and would be compensated accordingly. and i'll get into that in just a second. and to have a pcp assignment does not have to be an hmo type plan where it's chosen. we use a tool to attribute all of our members to certain pcps such that they have a full panel. we focus on the total health, which i think should become apparent in a minute. not just treating illnesses. pcp becomes the source of referrals. and we provide support -- we have changed our provider support model such that the pcp

has concurrent data to share with specialists. the net result of the financial model is that we'll reward outcomes. we have already seen higher patient satisfaction scores. and we are early, this program that i'll display here in a second started january 1st. but we are already beginning to see primarily because of changes in referral patterns we are beginning to see cost decline. so functionally -- and i'll apologize for some of the granilarity of some of this, but i think it's important you know functionally what graduation based contracting can mean. it's a term we use a lot. but to bring it right down to a relationship between the health plan, the member and the provider i think is important to understand. we launched ours in this

eight-county area on january 1st. we actually started the april before and began educating the pcps in our region for nine months prior to implementation. we now have 1,000 pcps participating, and we have about 400,000 of our members attributed to those pcps. one of the disadvantages although conceptually aco and aco-like arrangements or even patient-centered medical home arrangements, one of the down sizes is they generally work on smaller populations. they're specifically about a named population. it might be a segment base that might be very small regional basis or whatever that particular provider group can care for. we wanted a bigger effort than that. we wanted to transform the membership in an eight-county area across the board.

so one of the things we have been able to do here is we're including medicaid. we're including medicare, and we're including our employer population, whether it's self-funded or fully insured. so it has a broader impact across the population. what we've done is we've created a combination of fifa service, which we have just been saying is a bad thing, and i know we've been saying that all morning. but there are aspects where fifa service works. as well as a monthly payment we refer to as a care management fee. also because michael indicated we don't want to use state capitation anymore, so we refer to it as a state management fee. but i have put capitation back into this explanation on purpose because i think it makes the point, although it is not capitation in the way that most of us remember it.

so here's generally how it works. what we did is we took those 400,000 members, we looked back over a two-year period of time. we started in '14. we have that claims data. on a per member per month basis, we created a base rate and said the cost of that population historically on those claims under that traditional model is a certain amount per member per month. that's our base rate. we then used a form of capitation adjustment, which has historically been age and gender related. that's nothing new. but what we then did is we used the makesen risk scoring tool which can then predict what cost

based on an individual score. and we applied those factors for everyone of those numbers. so for example a primary care physician has 100 of our members. he actually has a risk score that is different for everyone of those hundred members based on whether they have chronic illness, based on they're very well, based on their age. so they are all different. we then take that population and we look to the providers. we use our heat of scoring methodology, and we create ten guidelines within our report and we apply those against the factor on a per patient basis. now, we could debate hetus. it's not perfect, however, we've find we've used the risk system

as well, and we found the results of those two analysis to be similar. so we're fairly confident using our scores was an adequate measure of quality. and also on our work like most health plans, we're having to do the scoring methodology and do the code checking as well. so that seemed to make sense. and we've had reasonable acerousy with that. and then on the efficiency model we used the mikesenrisk scoring tool with that, and we combined those two and that applied towards every member's risk score. what that creates, and i have the word capitation there, but that's our monthly service fee. what that creates is an actual capitation or monthly service fee to the primary care

physician, which is unique to every one of the patients that is attributed to that pcp. and as i said, there are a number of things particularly in the preventive and wellness areas that we want to have happen and we continue to pay them on fee for service. so that creates the reimbursement model for the primary care physician. we believe everybody wins. i won't read everything on the chart. certainly available if you'd like to have it. but one of the things it does is it creates a baseline level of cash flow for a primary care physician to be able to build their practices and to know they have a certain income level attributed to their patients that they can count on. and there are factors within that base level of compensation that then allows because there's

revenue there every month for a physician, it allows them to manage their patients the way they think it's best. and a good example is under the traditional model of fee for service, in order for them to receive any compensation for consultation the patient has to come into the office. it nay not be necessary. this way if they want to use telephonic or videoconferencing to communicate with their patients around conditions that are not as severe, there's actually a piece of that service fee that contemn plates that, and they feel they have adequate compensation in order to adjust their practice accordingly. they also have access to cost and quality data that they traditionally had not had. and that cost data is important in the referral pattern process. most times what we have found is the primary care physician

making a referral to a specialist at one place or another had no idea of what the cost, quality and efficiency measures were for the specialist they were referring a work in p. but we are already seeing a difference in the referral patterns against this population from the primary care to the specialists because that, as wellness and health of that population improves for those patients that are attributed to that primary care doctor, then he begins to receive more money because of the way that the monthly service fee is calculated. and his reimbursements go up. he's rewarded for the higher health status of his population. so he wants them to go. he wants his patients to go to the most effective, highest quality environments they can, to receive their care. historically, he did not know where that was.

the member, they're dedicated to a pcp. generally, we find our members, although choice is always important, want to have a pcp that helps them guide them through the system. the care is not limited to a standard office visit, as we mentioned. it's up to the physician and the patient to determine how that care, primary care would be administ administered. it's better coordinated. so when they leave their primary care on referral, their records are largely supported, and they're very current. and that allows the member to have shared decision making with their pcp. on our side, on the payer's side, it is aligned with our hedus measured, and affects our risk scoring, can affect our risk scoring or risk adjustment revenue. as well as star ratings.

there's a greater focus on population health. and as i mentioned, one of the things we have seen that we have underestimated a little bit, there is a change in referral patterns to more cost effective specialty environments. so it's a major step from fee for service. western new york was very traditional. i arrived there about four and a half years ago. there was essentially no value-based contract in the market at all. so it was a territory that had lagged a little bit. and as i mentioned, we now have 90% of our pcps. and essentially all of our attributed population on that basis of that contract. i mentioned we spent nine months educating in advance. one of the critical things i heard curtis say the same thing,

that it is essential in these environments that you change your provider support model. they need data in a different way and in different timing in order to make the right decisions over what they have had in the past. we have added about 25 provider service representatives that do nothing but administer these contracts with those thousand pcps. so outcome. as mentioned, 90% are working with us now. we were told directly by cms that we were awarded cpc plus contract as one of the 14 pilots for 2017 specifically because of our best practice program. i had hoped i would have a little runway in this process, but my largest competitor has copied us almost already. and i guess that's a way of

saying that we feel we made some of the right decisions. and then after one year, we're not quite done with the year yet. what do we expect to see? we would have a higher degree of cost transparency at the pcp level. that's also helping them make appropriate referrals. we have, as i mentioned, a couple times, a change in the referral patterns as a result. we have a very different level of provider engagement. and our value-based literacy among pcps, which they didn't know even two to three years ago, and we have seen hedus score improvement greater than traditional pay for performance programs. which has helped us in risk adjustment processes as well as our star rating. as far as actual savings, what we're looking at, we're pretty aspirational. we think probably on an overall

basis, around 2% in 2017. we don't know that yet. it looks like that's about what we're tracking. and again, i think maybe -- i agree with the comment michael made. that may not seem like a lot. it seems like a lot of work and a transformation, entire provider system for 2%. but this is a multi-year arrangement. and we think in year one, that would really be excellent. and remember that in a compounding environment, that 2% savings resets the baseline. so as you move that forward, that 2% becomes very significant over time once that baseline continually is reset. so that's best practice. we've -- we'll hopefully have a chance to come back and report towards the beginning of next year after we have a full year in. but hopefully, that was helpful

to give you some granularity around how some of these concepts are applied at a market level. thanks for having me. appreciate it. [ applause ] >> so, dave, thank you so much for that granular look at how these contracts really work, and it's very impressive that you already have 90% of your primary care physicians under contract with it. so now we'll hear from avik roy. always interesting. avik is the cofounder and president of the foundation for research on equal opportunity. a nonpartisan, nonprofit think tank that conducts original research on the impact of public policies on americans with incomes or wealth below the u.s. median. so it's a very topical issue that you're addressing in your new foundation. you know, you will know him. he's a leading conservative

change agent. he's advised three republican presidential candidates and influences informed the policy debate in his role as opinion editor at "forbes." he also has experience in medicine and finance as a former medical student and jobs early in his career at baen capital and jpmorgan. we're also honored to have avik as a member of the advisory board, and he's a senior adviser to the bipartisan policy center. he's a fierce advocate for the free market and a very effective advocate for the free market and patient centered reforms to lower the cost of health services and prescription drugs. we're just delighted to have him here today. thank you, avik. >> thanks, nancy. it's great to be with you all and share the stage with what i thought were really interesting presentations from some very

impressive people. as nancy mentioned, our new think tank, the foundation for research on equal opportunity is focused on trying to find ways to achieve the goals and the principles of both progressives and conservatives, to say let's find solutions. let's find reform ideas that can move the needle for the people who have the least economic opportunity in america today. i think all of you understand that prescription drug prices is one area where affordability and health care are particularly impactful to low-income communities, because we all go to the pharmacy. and pay those co-pays when we pick up our prescriptions. so it's an area where, of course, it's not alone in terms of being a place where health care is expensive, as we have heard today, but it's an area where in particular, patients are exposed to those costs. we published in may a white paper called the competition prescription. and the point of this paper is to try to break through the log jam we have had in prescription

drug pricing. as you all know, the dynamic we have had for the last several decades has been, well, the democratic side, there's been a lot of concern about high prescription drug prices. advocacy of price controls and variations of price controls to try to regulate the problem. and on the conservative side or the republican side, the view has been well, we don't like price controls. so we're basically going to change the subject. i think the argument here is to say, actually, the situation we have today where we have such high prices for prescription drugs, it's not the result of a free market. it's the result of specific policy decisions that congress and the fda has made that have allowed prices or incentivized prices to go up the way we have. and the solution to that problem is actually more competition. competition has been the biggest driver of reducing pharmaceutical prices in the united states. so we have successes and failures. and the goal here is to hopefully learn from the successes to address the failures. so you all know this story.

the story of the fact that americans pay a lot more for prescription drugs than everybody else. right? so per capita, prescription drug spending in the united states is basically double the rest of the wealthy countries in the world. at $1327 per person in i believe 2014 is what this data comes from. so that's a lot. right? what you might not know is that, you know, it's particularly thematic in the context of the value and volume debate. we're actually the best country in the world at putting people on cheap, low-cost, unbranded generic drugs that cost less to manufacture than a bottled water. so you know, we talk about the high prices in america. but those high prices are driven in particular by the fraction that's not that 82%. it's the 18%.

the 18% of drugs that aren't generic that are branded drugs, and particularly the subset of those are like the patented brand of drugs, that's where the high prices are. the vast majority of prescriptions in the united states are for cheap, generic drugs. there we're doing well, thanks to one of the far sided laws the congress has ever passed, hatch/waxman of 1984, which did a lot to stimulate the growth in prescribing generic drugs. compared to the european median of 21%, we're actually four times better in terms of making sure that where there's a cheap generic alternative available, people have access to it, they're using it, and the clinical effectiveness of the generics is as good as the old patented competitor. we're doing a great job, actually, at engineering and encouraging competition when a lot of drugs go off patents. the problem is that for the branded drugs, the prices are

going up. you'll hear people say, well, you know, growth in pharmaceutical spending is growing by x percent. what often you don't see in those charts or those slides or those tables is that actually, if you just look at branded drugs, the inflation is even higher. right? so in 2012, we're spending $228 billion on branded drugs. in 2016, spending was up 50%, 5-0 percent on branded drugs. spending on generic drugs and this is nominal data, not inflational data. in 2012, unbranded generics were spending $52 billion. and 2016, we were spending $50 billion. so there was actually a real decline, a nominal decline in the dollar value of our spending on generic drugs even though the prescription share of prescription drugs were going up. generic prices were going down. that's the untold story about drug spending in america.

is that where cheap, generic drugs are available, costs are actually declining. but for those branded patented drugs, costs are skyrocketing. one thing you hear the pharmaceutical industry point out is that, well, you can't just go by those prices you see in the newspaper. the list prices. the list prices aren't actually what we receive in revenue. you have to account for all of the discounts that come out of that list price before it actually gets to the consumer. and they do have a point. but i think it's also overstated to say that the net price that pharmaceutical industry reports is actually the true price that's paid by the consumer. that's not true either. and i kind of walk through it here. there's the list price. that's the price that almost nobody pays. but it says -- the sticker price, you could say. then there are wholesalers and distributors. the three best known are amerisource, cardinal health, and mckesson.

they actually buy these drugs in bulk from the pharmaceutical manufacturer and then sell them to pharmaciepharmacies. and so they take a discount for buying in bulk. and they have calculated that for branded drugs it's on average about 16%. you discount the sticker price by about 16% to get what imf called the invoice price. then to get the net price, you subtract out a few other things. so, for example, what pharmacy benefit managers do, they get rebates from the pharmaceutical administrator to get a more favorable position on a formulary. sometimes if you have a lot of cholesterol lowering drugs, some are generic and some are branded. the branded ones have to give rebates to the pbms which are contracted with insurers to incentivize the pbms to stimulate more utilization of the jubranded drugs when the

generic would take over. that's money coming out of the pharmaceutical industry to get their drugs to consumers, but it's actually a mixed bag in terms of what's happening with consumers. in some cases, they're using a costlier drug, which drives up health insurance premiums. the cost to the consumer actually goes up. another example is co-pay assistance. a lot of pharmaceutical companies will say we know our drug cost is over $100,000 per patient per year, but for low-income patients who can't afford that, we'll pay their co-pay and deductible so they don't have out of pocket costs. it's a win/win, right? because they get to say that they're, you know, being very charitable to low-income populations, and they get a lot more utilization of their drug. but the higher utilization of a high cost drug means what? spending goes up and premiums go up. so the end result for the consumer is not the net price. the net price is the end result maybe to the pharma company, but the price to the consumer is actually higher than the net

price because they're paying for these higher premiums as a result of higher utilization of expensive drugs. i try to illustrate that in this next chart. if we look at the other wealthy countries in the blue there, that's the average pharma spend per capita for them. and then we say, okay, what is it in the united states? and what are the different components? so $899 is the net price, if you add all this up, it's like $1600. that's like the list price, the sticker price. the real impact to consumers is basically this piece plus this piece. it's about $1150. so you know, yes, we should factor in the fact that the price that consumers pay is lower than the sticker price. that should be part of your thinking about high drug prices, but even if you take that into account, we're talking about prices that are more than double the rest of the wealthy economies in the world. so it's still a huge policy problem.

so i alluded to in the beginning, why are we in this situation? a lot of people say it's the free market's fault. the free market has lowed the high prices to emerge. if we had more price control, we wouldn't have higher prices. i'm not dismissing the fact if you regulated prices, you could in theory have lower prices than we have. that's true, but it's also true that we don't have a free market for drugs, let alone anything else in health care. and that's also a big driver. in fact, a primary driver of why drug prices are so high in the united states. so here, i have a bunch of bullets on many of the ways in which we through, again, congressional policy and fda policy, have made it easier for drug companies to charge higher prices, because they're insulated from competitive pressure. so the most important is that not only do we not pay for drugs directly for the most part if we have insurance, we don't actually shop for the insurance ourselves. we have third-party payment of

drugs and then we have third-party payment for the insurance that pays for the drugs with a third party. there's really ninth party payments for drugs and everything else in health care, and no wonder that patients are divorced from the clinical value and the cost of the drugs that they're using. right? they say, hey, i have insurance. my plan should cover this drug. but they don't always have a real sense of how much that drug costs the system because they're so indirectly exposed to those prices. there's also the fact that there are legal monopolies, constitutionally sanctioned monopolies. if you have a patent, that patent lasts 20 years and ten of those years might actually still be active when the drug gets through the fda approval process. and we don't want to get rid of that, right? we want to reward innovation. we want people to develop innovative drugs and capture an award for that, but that's not a market. when you have a monopoly, it's not exactly a market. there's also the fact that federal health care programs

like medicare, like obamacare, and medicaid and v.a., in different ways all basically mandate that drugs be covered. so medicare, basically every drug that's approved by the fda has to be covered by medicare. so guess what. that gives companies, pharmaceutical and biotech companies incentive to charge high prices knowing their drugs have to be paid for by the taxpayer, regardless of what that price actually is. so this is particularly a problem in oncology. in cancer, who gets cancer? mostly old people who get cancer. medicare is the health insurance for old people. so that's why oncology drugs in particular have high prices. in those areas of medicine that primarily affect people below -- in the workforce, pre-elderly individuals, there's a lot more leverage that insurance companies can use to fight that off and say hey, we're not going to reimburse for this or pit you against a generic drug or what have you. in particularly diseases that affect the elderly, that's where the leverage has been the lowest

because of the situation where medicare is effectively required to pay for everything. it's true in obamacare too. this is more of a cms regulation, but cms basically the regulation in the last administration that for every therapeutic class, there has to be one branded drug reimbursed by a plan participating in the obamacare exchanges. that means the companies have leverage to charge higher prices. there's also the fact that there isn't a level playing field for most situations where insurers and drug companies are fighting against each other. right? so the drug company if it's a branded company has monopoly, but the insurers are competing with one another. they're fragmented. there's a monopoly on the other side, and the end result is that the drug company has more leverage to say hey, if you're not going to reimburse for this drug but your other competitors are, you're the bad guy. you're the one not paying for the life saving drug, and the patients because they're insulated from all the price signals complain to the insurance company, not to the drug company, for charging the

high price in the first place. there's also the fact that the cost of r & d has gone up considerably over the last several decades. that, of course, means that venture capitalists, pharma companies, biotech companies try to recoup their costs through higher prices. it's very important to say you hear a lot of people say, well, we have to charge high prices because of the cost of innovation. there's basically no correlation, and the paper that the study is based on, we actually have gone through a lot of this data. there is almost no correlation between the cost of developing a specific drug and the price that the manufacturer of that drug actually charges. that's more driven by market power. if you've got a monopoly situation where you know insurers have to pay for the drug, you're going to charge a high price. if you're developing the tenth cholesterol lowering drug, you charge a lower price. market power is driving a lot of prices. a great example is the ultra rare disease area. you might have a drug for the disease where the disease only

affects about 4,000 people in the u.s., and you're charging $300,000 or $500,000 per drug. the cost of developing the drug is quite low because the cost of r & d is directly correlated to the size of the clinical trials you have to do. if you're doing a 40-patient clinical trial, because there's only 400,000 people in the country with that disease, your costs are low. if you're doing a test with 20,000 people for diabetes, the costs are in the billions of dollars for developing the drug. but diabetes drugs can't charge high prices because there's 20 other diabetes drugs out there. so it's really market power, not the cost of innovation, that drives prices. whenever you hear somebody say, well, we have to charge this high price because the cost of innovation, roll your eyes. the fda, and i'm going to highlight this in the next slide, the fda has actually in certain ways created artificial monopolies for drugs that are off patent. the famous example is the martin shkreli drug in the news. that was actually a really old

drug that had been off patent for a while, but because of an artificial monopoly the fda helped create, that drug was no longer a monopoly and shkreli could raise the price. there are a number of ways in which the fda regulations have created prices and there should be more competition. biosimilars are a very important emerging area for more competition. as many of you know, but maybe not everyone, there's a difference in fda regulation between small molecules which are pills you could synthesize in a high school chemistry lab, like aspirin, like tlipitor, versus other big proteins which come out of the dna revolution of the last several decades. those larger proteins have to be manufactured much more carefully and much more specialized ways. and they're not so easy to replicate. and as a result, the fda has been much tougher on and has created much higher regulatory barriers for generic competition

for those biologic drugs which has meant that the original manufacturers of the original innovative drug have had much longer runways in terms of having their monopolies than they would have had otherwise. we'll get into that more later on as well. there's also a kind of intellectual and cultural bias at the fda and elsewhere against so-called me-to drugs. we have to have innovative drugs not me-too drugs. me-too drugs are bad. that's not true. if you have me-too drugs that are similar in profile and structure to an existing drug, guess what that means? if you have two or three or four drugs competing for that group of patients, prices go down. so it has this bias against me-too drugs because intellectually, we say that's not as innovative, but economically, from a public health standpoint, it's very important to encourage the development of me-too drugs because me-too drugs mean lower prices for every american. let's focus on these three areas

in particular. before we get to that, let me talk about one of these ideas, which is orphan drugs. this is an example of where congress has created artificial monopoly. in 1984, congress passed this law called the orphan drug act. basically, in order to stimulate r & d for diseases with less than 200,000 patients in the united states. the idea being that because those markets were relatively small, they had fewer patients, the drug companies and biotech companies didn't have an incentive to develop drugs in those areas. basically what the law did is said okay, even if the molecule you're testing has no patentability, it's an old drug, around 100 years, if you develop it for a disease that's relatively rare, we'll give you seven years of monopoly exclusivity, as if you had a patent. what's happened effectively because of that law, which has a lot of good intent and good policy results behind it, is congress kind of overshot. i shouldn't say kind of, it really overshot. so the end result is what

companies can do now, okay, we can develop a drug and study it in this very small patient population that has a particular m muitation for lung cancer, and when it runs out, we can study another cancer and get seven years of exclusivity, for a drug that has no patents because it's been around for several decades. so some manufacturers have taken advantage of that legal opportunity, completely legal opportunity, to have long monopolies for drugs far contrary to the intent of the designers of that law. so it's gotten to a point now where orphan drugs are taking over our pharmaceutical stem, as i show in the slide. in 2015, so-called rare diseases are actually a quarter of all pharmaceutical spending in the united states. by 2020, it's going to be a third. $176 billion for diseases that affect or for drugs that try to treat very rare diseases.

i'm not saying those aren't important. if you have one of those rare diseases, you really care. but diseases like diabetes and high blood pressure matter too. alzheimer's, right, a disease that affects millions of americans, creates an incentive for drug companies to move away from studying those big disease populations because there's so much more of an economic incentive to study rare diseases. so again, let's focus on some of these areas where public policy has created artificial monopolies and what we can do to change it. we talked about the stat seven-year monopolies for orphan drugs. there's also something the fda started about ten years ago called the unapproved drugged initiatives where there are a bunch of drugs that have been on the market for long time before the fda exists. because they pre-dated the fda, they have been on the market legally for many years. the fda a few years back said we don't like that. we're going to try to get the unapproved drugs off the market.

here's how we're going to do it. we're going to encourage manufacturers to do clinical trials for those drugs, and the first one to do kind of a traditional set of clinical trials, we'll give them a monopoly for a period of time in order to reward them for the cost of doing the clinical trial. the end result has been a lot of this martin shkreli type action where okay, a company does those clinical trials. they get a monopoly, and then they jack up the price by 6,000 percent. so again, the fda had certain good intentions. we want to make sure that these drugs that have been around for hundreds of years have a sort of regulatory and safety profile we can certify, but they didn't really think ahead and say okay, we're creating this incentive for exploitative pricing. that's one area where the fda has kind of maybe again, not thought ahead about cost. another area that fda commissioner scott gottlieb has written a lot about in his pre-fda life when he was a colleague of mine at forbes and has talked a lot about in his current role is the issue of generic drugs where there's a delivery, a specialized delivery

technology associated with it. so epi-pen is an example. epinephrine, the drug in the epi-pen, has been around for over 100 years. but the injector that the company mylan used to administer epinephrine, that has a patent around it. what stymied competition in this area is that if you want to be a generic manufacturer of an epi-pen, you have to prove to the fda that the exact physical kinetics of the way your epinephrine enters the body are the same as mylan's. but you can't do that without running into mylan's intellectual property around their proprietary injector. so the fda has not had the toolbox to say, you know what. it actually doesn't matter if it's exactly the same as epi-pen. we just have to make sure we have the same clinical effect. so the patient who needs the epinephrine injection actually can get treated. right? and so there are things that the

fda can do on its own to try to address that. but commissioner gottlieb, again, in his previous work, has long argued that really, congress, if congress helped create as an amendment to the food, drug, and cosmetic act, a new pathway for those kinds of medicines so that if there was some delivery technology around it, we could create a new pathway to make sure the fda could show that they were clinically equivalent, these two different injectors. even if they weren't physically equivalent. that would do a lot to stimulate competition in this area. another area that you may have heard of is risk evaluation mitigation strategy. the famous example is actually one of the reasons the fda was created in the first place, a drug that was marketed in the '50s for leprousy. it turned out it was causing birth defects in pregnant women, and so it was clear that we have to have an agency that's keeping an eye out for this. so the drug has been around for

a very long time. a company based in jersey discovered that actually there was a lot of utility for the drug in multiple myeloma, so they did clinical trials to get it approved for that. and then they had a little bit of exclusivity because they did the trials, but the real monopoly opportunity for it was, and the fda said well, because of this risk of birth defects, we have to have a risk management strategy to make sure this drug doesn't get into pregnant women. they developed that pathway. they patented it so no one could create a generic vision because there was a patent arnt it, and the fda went through a spate of creating more and more of these protocols that people could patent, therefore blocking competition. there's, again, a way to standardize maybe rems protocol so they don't have intellectual

patents around them. the last bucket is the biosimilars piece. biosimilars have been accelerated in the marketplace as a result of a kind of a provision within the aca called the biologics price and competition innovation act of 2009. bpci. that bill tried to create some measures that are like hatch/waxman so biosimilars can get to the market in a more standardized pathway, but the problem is it's not exactly the same. so as i note here, if you have an off-patent drug that's a small molecule like an aspirin and do clinical trials, you can get five years of exclusivity, and then there's generic competition. for reasons that basically have nothing to do with economics or policy but have everything to do with politics and lobbying, that's 12 years for biologics. there's no reason why it should be 12 years for biologics and five years for small molecules. it should be basically the same because if you don't have a patent, there should be the opportunity to actually develop competitive drugs in that area.

if you have a patent, fine, have that patent and have that legal constitutional monopoly for an innovative medication. if it's not innovate frb, there should be a harmonization between a small molecule and a large molecule. another key area, one of the reasons the chart i showed in the beginning about the 82% uptake of generic drugs is because at the pharmacy level, walgreens or duane reade or cvs, can substitute. your doctor prescribes lipitor, a cholesterol lowering drug, the pharmacist can substitute generic drugs without asking your permission. here's your bottle, without asking the doctor's permission. that's one of the key elements of hatch/waxman that's led to higher uptake in this country compared to other countries of generic drugs. the bpci is not as strong on that point. so a lot of states at the behest of local companies have created barriers where you have to ask the doctor's permission or the patient's permission or both or go through a bunch of other

hoops before you can get the biosimilar prescription. in that way, competition has been inhibited at the state and local level for biosimilar drugs. so we have alluded to a number of the ideas that we talked about in this paper for how you can actually increase competition in prescription drugs and therefore lower prices. some of the things i haven't highlighted yet that i would encourage you to think about is i mentioned that we have this bias towards me-too drugs. one way in which the bias manifests in fda policy is fda has fast track innovation. the fda will give you certain more accelerated review times and sort of an easier process to get through the fda. when we could easily have something similar for areas where there isn't enough competition. we could have a fast track for areas where there's a monopoly or one or two drugs for a particular disease area but we really need more because of that public health problem that comes from monopolies and lack of

affordability. another thing we could think about is to emulate the success of medicare part d and leveraging generic drugs into medicare part a and b. we have basically three different parts of medicare that in which the pay for prescription drugs. part a is for the hospital based drugs. part b is for drugs administered in a physician office, and then the retail farmalpharmacy drugs part d. we could do a lot of things innovate to allow the cost savings to be leveraged throughout those other medicare sections or medicare parts. another area we should think about is maybe creating a safe harbor for insurers to jointly negotiate with the drug company in a particular state for the prices of that prescription drug. so instead of having this unlevel playing field where the drug company has a monopoly and the insurers are fragmented, maybe the insurers could actually jointly negotiate for

that drug price and therefore not fear if one insurer doesn't cover it and get beaten over the head, they might have a stronger ability to negotiate and get the prices down. a nice advantage is it would diminish the incentive for insurer cull sawnidation, which is a problem we have been dealing with. at the end of the day, the best solution for making sure that we have a more competitive system with drugs is to make sure that every american or as many americans as possible are choosing the health insurance plan that covers them. that's something the aca accelerated in a certain way. that's something that pre-market conservatives want to accelerate. the more people choosing their insurance, that's going to do the most to stimulate and incentivize lower cost prescription drugs and llalization of the low cost options we have available today. with that, you have a copy of this paper. the full paper in the folder, so if you picked one up outside, if you're watching this online or on c-span, you can download the paper from our website,

freeopp.org. i look forward to your questions. thanks for your time. >> well, that was terrific. you covered a lot of ground there. and i think all of us learned some new things. so at this time, i would like to open up the floor for questions. so if you have a question, please complete your card and the team will be going around gathering them. i'm going to take this opportunity, the fact that we have two ceos that are major players on their state exchanges, to ask with open enrollment coming up next week, we're all wondering what you're expecting and what you're expecting for 2018. we know there's been a tremendous amount of uncertainty, especially around

csrs. curt curtis, would you like to go first and share a few thoughts on what you're expecting? >> thank you, nancy. and i think from our perspective, this has been an issue, obviously, that we have been watching very closely for quite some time. and as we stay close to the market and we do that through feedback we get through our call center, we have retail stores throughout our state where we interact with those in the individual market on a regular basis, and also with our agents and brokers, which are very critical to working with these individuals. you know, we've been working with them to try to get an idea of what we might see with the open enrollment period. i think as we go into that, all of the confusion, all of the things that have occurred over the last several weeks, we're expecting a slight decline as we move toward that. i don't think that should be a surprise to anyone as we move in that direction. i think one thing that we have seen that maybe has surprised us

a little bit is all this confusion has bled over into some of our other customer segments. and that's principally been our senior market. so as news reports come out about increases in premium rates and are csrs funded or not funded? we're receiving a lot of questions from the senior population who are going through their own enrollment period with questions about how that affects them. we're spending a lot of time at our company, a lot of resources to outreach to the individual market and the senior markets and other who might be going through an enrollment period to communicate exactly what this does mean to them. >> dave? >> i would say first of all, in new york, we have a state-based exchange which was developed by the state, which was one of their options. and i think fortunately for many of the members in new york, it has run on a fairly stable basis.

however, i would echo curtis' comment about the instability that has occurred. we -- our activity is already up substantially, meaning people calling in, people going into our stores, asking what's going on. because there isn't a clear delineation of what the cost sharing reductions apply to and the average consumer doesn't understand that level of funding in those programs. so they feel that their own coverage is in jeopardy. and they're concerned, not only about losing their coverage, but i think that the -- our members feel like they have done their part. they came out of the uninsured, the ranks of the uninsured and made the effort to enroll in health insurance coverage, and now they don't know what its status is. and they don't know what is going to happen to that coverage, whether they're going to lose it, whether it's going to be changed, how it's going to be funded going forward.

and that level of instability is, i think it's going to cause a number of our members drop back out of the system. >> great. great. well, that was very interesting. we'll get back on point with the value-based payments. and i'm going to start with a question from u.s. news and world report that was submitted ahead of time. what does the future hold for value based payment if the affordable care act is repealed or overhauled in the future? is there an agreement that a shift to a value based payment is needed for cost control? any of the panelists? mike, you want to start? maybe. >> well, as i said, i think that the challenge here is much bigger than what you might think is going on in the affordable care act, and frankly, there discussions around the affordable care act haven't really centered on this aspect of the affordable care act. i think my expectation is that

there's going to be some version of alternative payment models moving forward, not because we really have figured out how to make them work perfectly or have perfected the regulations, although there's a lot of work going on in that area, but because the alternatives are very challenging for people in a variety of ways. so i expect that many -- well, many of the models going on aren't directly related to aspects of the affordable care act, and there's a lot going on in the private sector, and you're increasingly seeing the private sector continue to lead in developing models. the one thing you might not have realized in listening to curtis and dave talk is they both developed different models that worked in their markets, and i think that level of flexibility you see in the private sector is such that there will be a demand from the delivery system to maintain this time of momentum. >> one thing i might add is that a big part of the reason we haven't had a value-based health

care model to begin with is because of 70 years of federal policy that has created incentives to go in the opposite direction. the reason we have a value-based iphone model and a value-based car model and a value-based everything else model is because the consumers directly benefit if companies and manufacturers deliver higher-value, lower-cost, higher-quality products. that's not necessarily true in health care because of all the incentives we have put in the other direction. really, we talk about value, i think it's very important to understand that value at the end of the day is determined by patients. patients should be the ones, not experts, who determine what's valuable and the best way to insure that patients get to participate in that conversation, what value is, to make sure they're controlling the health care dollars that are spent on their behalf. >> curting or dave, you want to weigh in? >> i would add, i thought michael did a great job answering it. i think the alternative payment models or value, moving toward

value-based care, i think it's here to stay. i think it's taken hold in the marketplace. i think we as private insurers, as we hear from our customers, i think the expectation is we will continue to move down that path. it's desperately needed, a big part of our responsibility to our customers is to provide affordable coverage for health care services, and we think moving toward a more value-based system is a critical way of getting there. >> dave? >> i would agree. i think that one of our challenges in going forward with the models, even though they are -- i wouldn't say they're relatively new, but i think in a lot of markets like ours, they're relatively new, is that the definition of value to a health care consumer is changing. and we will be challenged to make sure that the model changes with that. >> great. here's a question from cmf. and avik, it's right up your

alley. getting rid of pay for delay and patent laws that allow for evergreening seem to be low-hanging fruit to help lower drug prices. why aren't we seeing any movement on these issues to help with drug pricing? >> i'm actually a skeptic on the so-called pay for delay issue. for those of you who don't know what that is, basically what happens is you have a branded drug where the patent is about to expire, but nobody knows when the drug will go off patent because there are three or four pat nlts that matter, and they litigate to try to invalidate some of the patents. because those are going through the court, nobody knows will the drug expire, will it go off patent in 2022 or 2027? we don't know, and no one knows until the judge decides. and so what often happens in those cases is the branded innovator and the generic competitor settle. and they say, you know what. in exchange for us being able to go first and have a certain

amount of exclusivity as being the exclusive competitor instead of having ten other generics in the market, let us launch maybe not in 2022, but in 2025. so not exactly when the last one expires but sort of in between. so some have criticized these, again, called pay for delay and say what's happening here is branded companies are gaming the system to extend the life of their patents longer than they should be. i don't think that's accurate. what's happened in these cases is that both sides are basically shedding risk. they're kind of averaging out. so the pharma company is saying okay, instead of going off patent in 2027, we'll start having competition in 2025, and the generic company is saying instead of getting to start in 2022, we're starting in 2025. so both sides half win and half lose. it's a settlement. the reason this happens is nobody is certain who is going to win in court. if you know you're going to win in court, if the generic company has a really strong case, they can say, no, i'm not going to settle because i think we're

going to be in the market in 2022, go to hell, pfizer. so that's why the settlements happen. i don't think it's a gaming of the system. in fact, the ftc a few years ago tried to litigate because they thought it was anti-competitive and they didn't get anywhere because of the fact that these were legitimate settlements in my view. i think there are a lot of things we can do to encourage competition. i think that one is more of a red herring. >> anyone else want to comment or shall i move on? okay. here's a question. oftentimes clinical quality measures change because the science underlying them change. how do the changes impact how we assess the quality of care and use these metrics for pay for performance? so talking about the underlying measures of quality and how when the science gets better, the metrics change. >> so i think there's two parts to that answer. the first one is sort of the process answer is there's a general system, it actually works differently in different

companies, but overall, what i would simplify it as saying is members get vetted through processes like mqf to figure out what measures they'll have and how they want to work, and an enormous number of private insurers, cms is others to figure out how the measures are going to be incorporated into a range of contracts. and so i mentioned earlier that i was concerned about the administrative costs of all of this. my general sense is we have a process that is not ideal, and it's in many ways administratively burdensome, but it's important we maintain some version of the process, and i think where we're going to go now is more streamlining. we're never going to be in a situation where our quality metrics capture all dimensions of quality and are updated as timely as you would like, because the simple administrative costs of being able to gather the data and implement them become overwhelming. i think we need to begin to

think through how do we take the basic current framework we have and simplify it to a meaningful and administratively efficient approach. and that's going to be a challenge, i think there will be other areas like patient inported outcomes and other types of quality measures we're going to go to, but i would discourage you from trying to say we could come up with the exact best scientific evidence, and we need to get that in front of our contracts, you know, as soon as we absolutely possibly can. there's so many ways we can complain about the quality of the measurement system now that i think that might be an updating all the measures might be lower on my list of things we should do. >> all right. i'm going to move on because we have so many questions. unless somebody wants to jump in. dave, did you want to jump in? >> well, i wanted to emphasize the mission around quality. i think there has been a debate on how quality should be

measured and what it is inside of health care for 30 or 40 years. what you need to do in order to take positive steps forward is you have to determine your methodology and definition of quality. and that is probably not going to be identical to the next person. and i agree with michael. but that's okay. i think what we need to do, particularly in this value-based contracting process of putting quality in that consideration, is that it needs to be directional. if we're waiting for it to be perfect, it will never get done. that's why it hasn't gotten done any quicker than it has. at the same time, it has to have science behind it. but if we're going to wait for some uniform definition and

process to determine quality that is going to be applied in every circumstance, nothing will get done. >> great. all right, question from the national business group on health. how are the employers engaging in this discussion? maybe curtis. >> i'll take that one. and you know, i talked a bit earlier with the arkansas payment improvement, and we have had strong support from our employer customers. and that's not happened by accident. you know, we spent a lot of time talking about the individual market, but the group market is very, very important to us. and it's very important that as we move in this direction, we take them into account as we do that. so we made, especially the episode of care program, we made that central to our network strategy. it became part of our network reimbursement. before we did that, we created opportunities to visit with all of our large employers in

particular, and we went out there and explained these programs to them, how they worked and how it would benefit them and got their feedback on the front end so we could build it into our program design from the very beginning. i spent a lot of my time out with our customers. invariably, the issue of value-based payment comes up in those conversations. that's an opportunity for them to share with me as well, and we're going to continue down that path as we refine these programs going forward, we're going to continue to seek the input of our group customers. >> great. >> i'll speak briefly sort of from the perspective of harvard and what i know more broadly. there's this subset, i would say a relatively small subset of leading large employers, curting probably spends a lot of time with walmart, for example, boeing, caterpillar, a bunch of others you can name that are doing a range of innovative things, sometimes in the space of payment reform. for most employers, it's very hard for you to engage directly with the delivery system because

your carrier insurer is between you and the delivery system. so most of the employers that i'm familiar with that are doing things are focusing on aspects of benefit design because that's the much, much easier lever for an employer to pull. there are some, i think, hopeful movements forward, things like the health transformation alliance where large employers are trying to come together in ways to push the system forward, but i think by and large it's much more difficult for employers to engage in a payment reform movement than it is to adopt novel benefits. >> did you want to weigh in too? >> yeah, you know, i often get asked by particularly large employers, you know, we're the ones who they say, we're the ones who have the most to gain from lower health care costs so why doesn't washington, why don't experts talk to us and learn from what we do to try to lower costs? and what i often respond to them when asked is, you guys have

actually been the biggest drivers of health care inflation for the last 70 years. it's because, again, the huge tax break we give to employer-based coverage that makes workers incensensitive toe price and value of the care and coverage they're consuming that has led to all the health care inflation. medicare was built on that system as well. and actually, yes, employers do have an incentive to keep their health care costs down. but they also have an incentive not to make their workers mad by tweaking too much their health benefits. michael dealt with this a lot at harvard where they tried to tweak the benefits to make them more cost effective and the faculty went absolutely nuts and complained. right? that's maybe not the typical company, because it's harvard facul faculty, but you know, the same dynamics apply everywhere. a lot of large employers are afraid to do a lot of things that all of us at the table would be cost effective and appropriate ways to refine your health insurance coverage. one thing that i'm excited about or intrigued by in the last

couple weeks has been president trump's executive order which of course has been the subject of a lot of noise and commentary. there's a piece in there about expanding the flexibility of employers to use health reimbursement accounts, which are kind of like hsas, not exactly the same, but that could be a really interesting tool for putting workers and therefore patients more in charge of the health care dollars spent on their behalf. we could see depending on what statutory authority they find they have, a lot of innovation in health care delivery that comes out of those moves. >> great. >> i agree with much of what avik said about the role of employers and his related analysis. i want to point out one thing about competition in health care that does relate to the role of employers in a lot of the current policy discussion. again, as an economist, i'm a relatively free market economist. i'm relatively pro-competition

in the grand scheme of things but health care is not asparagus, and one of the key challenges we face is this pooling component of health care that doesn't occur for many of the other consumer products that we talk about. we're pooling sicker people with healthier people, older people with younger people, and a part of that involves just managing people in different risk profiles, and part of it is systematic pooling of people that, you know, don't necessarily want to be pooled with each other. and one of the important roles of the employer-based system provides is a way of pooling that is not inherently risk based, and it reason health care has been so challenging is because if left fully on their own where consumers have total autonomy to choose the plan that works best for them, they will do that, which again, chalk one up for economists, but on the other hand, that leaves other people in a situation where they might not have the same options they had in the past. and i don't have any great answers to how to deal with that

amount of pooling or whether the aca did too much or too little. that's a debate i will leave for other folks, but i do think when you do an analysis of what's going on and think about these kinds of things in health care, you cannot get to somewhere close to the right answer if you aren't really cognizant of how the pool is held together or allowed to separate. >> great. all right. we've got a question from the american medical association. one way to bend the cost curve is to shift the focus of american medicine from treatment to prevention. can a sufficient emphasis on primary prevention and preventative medicine be achieved through free market solutions? what is the best approach? >> i would, with all due respect to the ama, i would dispute the premise of the question. prevention is very important for public health, but we all have to die of something. so if we're not going to die of cancer, we're going to die of alzheimer's.

if we cure alzheimer, we're going to die from falling down it stairs or something else. we're always going to die. if we're always going to die, there's always going to be some expensive care at the end of our lives. so while prevention is very important for public health, obviously, if you diet and exercise and don't eat a lot of barbecue like i do idaustin, you're probably going to live longer and that's good, and you won't have diabetes and all those other complications and those are all things we should try to achieve from a public health standpoint, but we shouldn't be kinlsed that's going to reduce cost. a lot of times, prevention can increase cost. if every woman in america got a mammogram, the mammogram test cost money. we would spend a lot more money on mammograms and have some incremental benefit in terms of early detection of breast cancer, but the reason why a lot of independent boards and medical boards say it's only women of a certain age or risk profile that should have mammograms is because we want to make sure the right people are getting the mammograms so we're not massively inflating the cost of running all those tests. so prevention is good, but

prevention is not an unalloyed good. it has clinical risks and economic risks that we also have to take into account. >> and mike then has to also -- you have such a great handle here. >> i'll try to be quick. >> i love it. >> i agree again with avik. i have two fitness trackers. i'm a big fan of wellness and prevention. that just says ocd, but the issue, if you could snap your finger and make everybody healthealth y until they were 105 and then they died by falling down the stairs, that would be probably a better world than we have more. the problem is most of the evidence suggests most prevention programs actually are very cost effective. i might say at high value. cost effective is not the same as cost saving. and oddly enough, when we promote prevention, we're often promoting primary prevention, which is giving healthy people things and the most important type of prevention would be secondary prevention, where you are trying to find people who have chronic conditions,

diabetes, and prevent exacerbations or bad things happening, and things like the affordable care act where there's an exemption, that focus only on the on the primary preventi prevention. if we expanded to allow secondary prevention to be treated as prevention for the purposes of hsas and other types of things like that. >> this is a question for you from todd at one of the federal agencies. caring for patients and multiple payers to transform to your models. how do you get the pcps to do your best practices when -- why you're large portion of their business you're not the only payer that they are dealing with. >> well, it's interesting. we initially thought that we would force this issue on our pcps because we wouldn't know.

as you saw from the report that i gave we're at well over 90%. we never had to get to that point. doing so for a variety of reasons but one they are getting better data that allows them to be more accurate and efficient at the point of service that they historically have been frustrated with. if they should improvement in the quality stores and health scores. i don't think i mentioned it but for those performing at the top end of those quality scores, those pcps. they are reimbursements go up by 10% in 2017. we'll go to 20% next year.

it's a pretty significant difference. they have been very enthusiastic about adopting it. our primary competitors is very close follower. i think it's going to become the standard of reimbursement in the region we're operating. >> do you want to weigh in too? you face some of the same challenges in arkansas about one of the days you dealt with it was combining with medicaid. >> we did. that was 2010. when i talked about earlier when we started our pilot and the results we experienced probably a bit different then than what they experienced today. had we done this in the same time frame as what david and his plan have done, we may have seen something different. we think as we talked about earlier, local markets, local care, the dynamics, all those

things are important. you have to be open minded as a plan that ordseers to solve problems and offer up the best solutions you have to adapt to your particular market. that's what drove us to the approach we found. we thought it was a great opportunity to collaborate. >> mike, there's a couple of questions about medpac and ma macra. maybe there are soon to be or current comments on it. >> i'm not on medpac now and i do not speak for them. when i was on medpac, i did not speak for medpac.

i believe that i'm sure they work very closely at a whole range of things and how it's working. they have been critical of aspects of the program for a bunch of reasons, i think. we'll see what they come out with in terms of the recommen recommendations. not all physicians will be above average. i say that with some confidence. that creates a political dynamic that will be hard to really move forward over long period of time with the model. i'll just give a shout out to what a food job they do in that analysis. i really earn courage yncourage through the materials. >> here's an interesting question and kind of relates to a conversation we had last night at dinner.

this is from somebody at national organization. it's a two part question. what are your thoughts on amazon entering the pharma business and see more increase in marketing as a result and express scripts? >> based on what we know about amazon's strengths and advantages and the evidences in terms of what kind of licenses they have applied for, there's more it will be on the d distribution side. they bought whole foods.

they now have these brick and mortar facilities that they could use to build their own network of pharmacies. i wouldn't be surprised to see them try to attack that as well. they are a big retail organization. pharmacies are retail organizations. that's another area where their strengths are. i think that's what's driving amazon's activity based on what we know publicly. i think the cvs aetna situation is a lot about what's going on. >> anyone educationlse want to ?

>> we're still evaluating that. think about that on the surface. there's many health plans that build our own pbms. we happen to have not done that. there's a lot of health plans that do that. you bring that in house and operate as an integrated system. cvs and aetna are ending up in the same spot. i believe that controlling your pbm services through at risk health plan type populations adds stability to the pbm marketplace. i think that's what they are looking for. i think they will start on the distribution side.

amazon has completely transformed the consumer model in every business they've touched. i think they're very aware that the consumer model in health care hasn't been very good. there's been a lot of dissatisfaction with the actual users of health care services and i think it's my estimate that they have an intention to change that model. >> with that, i think we should wrap it up and federal thank you all so much for joining us today for this great discussion. i'd like to thank senator ron for sponsoring our event and i'd like to thank my terrific team.

let's give a final round of applaud for our great panelists. [ applause ] they're slides are on our website already and within about a week there will be a video on our website.

representatives from twitter, facebook and google testify before congress on tuesday and wednesday. as part of the investigation into russia's influence on social media and the 2016 election. tomorrow at 2:30 p.m. eastern. live coverage on c-span3. on wednesday, two hearings live

on c-span3. the house select intelligence committee, collin stretch of facebook and kate walker for google. watch all three hearings on c-span3 online at cspan.org or listen live on the free c-span radio app. that announcement came the same day as a house oversight sub committee hearing that looked at the regulation of political advertising on social media and other platforms. witnesses discussion certain diz closure rules for online advertisements and the implications of russia purchasing facebook ads during the 2016 election. this is just under two