pricing approach and how that would actually impact on innovation. >> and i'll just say, we're interested in your answer but briefly. >> well -- [laughter] we know the formula for your response to him, and then i'll -- >> because we're out of time. >> well, i agree with you on the notion of a payer being able to type trait reimbursement and moving away from the sort of binary reimbursement, being able to do things that were developed when you were there or other kinds of schemes like that. i think what we have at least with medicare is a binary approach to reimbursement where in some context everything gets paid for. in another context like diagnostic tests, almost nothing gets paid for. so, you know, that agency, i don't think, is incorporating any kind of evaluation of innovation or where there's, you know, incremental rewards to patients and to how it's paying for things. it's sort of a fixed process where there's different buckets and sometimes you pay for

everything, and sometimes you pay for nothing. if there's other constructs where you can provide pathways for things to get onto the market, get guaranteed reimbursement, i think those make senses. i don't think right now medicare has the capacity to do that measurement, and that's the difficulty. i think it's going to be very hard to build that into the system. >> last word. >> oh, the last word, my favorite. no, i think that everybody is focused on bringing innovation to the market that improves patients' outcomes. and i think we've struggled with the ability to demonstrate that and to have that demonstration accepted by the payers at the other end. so i want to underscore some of the things in terms of reimbursement that rob suggested which is earlier understanding of what would constitute value to patients' outcomes, how big that value has to be, what the appropriate end points are, how it would be measure inside a meaningful way during a randomized written call trial

versus how we would evolve that database or evidence base for that when it got out into the marketplace and where patients were exposed. so earlier and better understanding of what the value add would be, how it would be measured. and then a collaborative real-world platform for us would, at the end of the day, how these things were going to be applied and to design that early enough in the course of drug development so that we were ready to hit the road running and that we didn't have years and years where it wasn't necessary to see a longer term outcome before we ever got any meaningful data on how we had made an impact on patients at the end of the day. >> great. i think we'll hear more about measurement in the third panel today. so i want to thank all our panelists. [applause] and why don't we, i guess the second panel come right up. [applause]

[inaudible conversations] [inaudible conversations]

>> okay. we're going to get started right away. josh, you want to get started now? rob, you're going to have to take your seat. [laughter] your moment is over. [laughter] >> [inaudible] >> okay. so with that introduction from the first panel, what we want to do now is talk about opportunities, specific opportunities to drive innovation with new payment policies. and talk about pilots. and i think, reed, if you could start, united has really been a

pioneer in efforts to measure and pay for value with physicians and hospitals. and more recently with medical products. can you give us some perspectives? >> well, let me just sort of start out by saying how energized i am and we are by this opportunity to talk about rewarding value and aligning it with new innovations and care delivery. this is how we spend our own money. and so we are very clear on how much important it is to invest heavily in what you do to deliver the kinds of outcomes that are necessary. so i don't really see it being a dichotomy at all. i sort of challenge the premise of this question of whether or not there is an alignment between value and innovation. they work together. as a company that spends $100 billion plus in health care expenses every year acting on behalf of our customers, we really want and demand that the innovations and the interventions in health care do deliver value, that they are better outcomes and that they

lower health care costs. i challenge some of the conversation, i hope we can get into it later, some of the perceptions on the first panel because we absolutely have no chance to help the average american, the average small business owner, the average state government, the average large employer. they cannot make it another day. we've spent almost a trillion dollars trying to get access to care for 34 million americans. there is no way that we will be able to sustain that access to care if we don't get control over these costs on the use of technologies and pharmaceuticals. so there is an urgency. this innovation has got to deliver simpler and less complex care. it has got to make care more convenient. it's got to move care from the most expensive places called the hospital to the outpatient, to the community, to the home. it has to disruptively replace more traditional innovation -- interventions. eliminating redundant things that are expensive and then

making sure that the things that it does companion with, that those things become more precise. and so innovation for it own sake has no value. so what we now have to do and what we are seeing is as health care costs are going up because of the way that new innovations are introduced into the system. so i think what with the changes in this reimbursement system is to now be rev hughes their in -- revolutionary in realigning the riff for how new innovations are used and introduced, shifting that risk not only from the people who pay for care, the employers, the average americans, the small business owner, those people who are the ones that are at risk for the cost and moving that risk downstream into the delivery system. so this is the transformative opportunity where now everybody is in the game. and we are aligning the interest for those who have to pay for care with those who make products with those who have to deliver care. we are early in that revolution.

but that revolution of fee-for-service transition away from that more towards, now, episodic, care-based, bundled care, cap tated care. as we are moving along this continuum what that now is saying that we are beginning to get the attention of the delivery system. because now how they behave becomes so very important. and so i would say to you that this is probably the key opportunity. now,s has that movement been successful? do we really see a dramatic lowering of health care costs and a dramatic improvement of outcomes and quality? not yet. still early in the game. but we have examples of aggressive first-adopter organizations in health care delivery that are actually decreasing emergency room visits, decreasing preventable hospitalizations, decreasing lengths of stay, increasing the use of generics and decreasing

the use of inappropriate high-technology imaging studies. so those are key. so what do we have to do to get to where we're going? the next panel is going to talk about comparative effectiveness research, so i'll leave that there. but let me just emphasize, we have to have the economic data. we have to know, first, of course, whether the thing works. then does it work better than the existing ways of treating the disease? and then we've got to know is, what are the economics associated with it? real data, not mathematical modeling data. number two, we need specific data on specific products and devices. we've got to get the labeling now of those products that get used so we can track it. is this particular implant better than that one? which one has more complications? which one has more readmissions? we need to have that coding that allows us to track with great specificity what really is going on with individual products. we need to greater specificity on overall coding as discussed

in the last panel. we need to know with greater precision what did the doctor actually do, and what things did they use when they did it? and that means that the icd-10 codes will be important in that, more electronic medical records and, please, let's accelerate meaningful use criteria so that we will be able to have the data that allows us to do this. registries are going to be key. we have to have registries that will give us pooled information so that we can make these assessments not only for the comparative effectiveness research, but also so that we can guide and counsel patients. if you are going to be a rational person to have this intervention, we suggest strong hi you use this product because that is important to your health. we are very interested and engaged in the things that freda and others talked about a moment ago around the health plan working with the device manufacturer or the pharmaceutical company to develop science and evidence early on. you're absolutely right, my

colleague from medco, absolutely right that we have to do that. now, we're doing it. here's the challenge: it takes a lot of internal resources. it costs a lot of money and time to do that company by company, device manufacturer by device manufacturer. so there has to be a prioritization. what's most important, and how do you figure that out? we don't have time in the one minute left to sort of figure that out here, but that's a conversation that has to occur between companies and at a higher level of national discourse among our research, our health service research and our physician community. regarding the physician community, clinical guidance on appropriateness of use. who should get it, what subpopulations, how do you use the thing? that clinical guidance is key. i am terrified that america's medical specialty societies have zero infrastructure to do this at the pace and scale that we

need. therefore, i urge and beg anybody from the accountable care act that can survive the political process needs to be poured into these specialty societies who then need to have an accountability with the nation that they will be turning around this kind of information in a hurry. and lastly, i will say that in this transition period, unfortunately, because of the extraordinary cost escalation which threatens the average american and puts all of them at peril for being able to afford access to anything tomorrow, we're going to have to have very strong medical management protocols by those who have the responsibility of paying the bills. and those conversations will need to be very clear and intense, and the manufacturers of innovation need to be right at the table thinking about that. but how they do academic detailing with their customers is key. because if manufacturer is not

close to the customer, the physicians in the hospitals and affecting their behavior, those companies are going to be in peril. they have to find a way to stop advertising and do better at appropriate use. >> thank you. our next speaker is hervé hoppenot who's president of novartis encology. and maybe you could talk about some specific demonstrations. >> yeah, i can do that. i wanted to make a point on, you know, the subject of the conference which is innovation and reimbursement system. and i think there is one aspect that was not touched. i mean, i agree with some of them, but there is one aspect that was not touched is realizing that the progress we make in medicine on the hiv is a very good example, in fact. the one you were speaking about earlier where there is sort of an asymmetric situation between where you informs and where you save. -- invest and where you

save. most of the investment is viable costs, and most of the saving is fixed costs. i will take the example of cancer. my company, what i'm in charge of at novartis is the encology group, and today in the u.s. the american cancer society numbers show that 17% of the cost of cancer is coming from drugs. and 83% is coming from are hospitalization, everything you can think of. now, every time a new drug is introduced for treating cancer patients in the u.s., many of the comparison in terms of how is it going to impact the drug, but of whoever institution is impacted by aids. and i think it's a similar symmc view that is driving to a very large number of very ineffective behaviors from the communities of providers and payers and everybody else. so i think it's important to

look at it also this way. the second point i wanted to make about trying to find solutions that will be working for every disease and every illness. and we spoke about rare disease where there is a specific need that needs to be addressed. i would compare it to the car his and the truck business, and we tried to put all of that under one roof and say we need a policy for everything that has four wheels. but, in fact, it's not the case. some of the diseases we are dealing with have a very different cycle of development. for example, we spoke about effectiveness and being able to prove prospective, combative cost impacts, etc. we are dealing with products that sometimes, dealing with illnesses where phase ii study is going to drive regulatory improvements. so we don't have compartment studies. we have patients who are no

options. we are either to develop drugs with the academic community that we'd be moving and creating new options for this patient. in that case we don't have data. i have seen payers in some countries around the world who are refusing to pay for clinic. [inaudible] obviously, proof that it was useful, but we didn't have the paperwork that was following what the bureaucrats were asking us to do. it was very odd and very strange. now, the third point is really about the willingness to do risk-sharing programming type of reimbursement. and here i can broaden maybe to other countries outside of the u.s. where it is easier to do this type of deals when the people we are speaking to are payers and providers at the same time. and that's really one of the issue, the complexity of the system we have here in the u.s. where it's really difficult to know exactly who will be the person who will be interested in

a program like, for example, in can kenya and the chronic leukemia where we had a new product that was launched where we were able to in some countries have risk-sharing programs where we would be providing the product and based on academically-defined definition of success, major molecular response, then we would be sharing the risk. if drug doesn't work for that very specific patient,, -- patient, we don't think it should be covering that cost. it is feasible, but it requires a very different set of people around the table to discuss it when the drug is different. >> great, thank you. norma. norma holtz, interim director of pharmacy at health alliance medical plants. please, your thoughts. >> thank you.

currently, we are in a, we have a very unique opportunity as payers to have innovative contracting. risk share. one product that we have collaborated with a pharmaceutical company allowed us to decide if outcome doesn't work, then we're paid a certain amount of money for the outcome. and, basically, we share that with our member. if you fracture and the medication possession ratio has been met and the member has not had any previous diagnosis of a certain disease, then why not provide repayment to the health plan and to the member for the product not working? t just very -- it's just very closely to what jerry states

about electronics. you buy electronic product. if it doesn't work, you take it back. you get your money back. there's something wrong with it. now, medicaid -- medical health, it is a very complex issue so, of course, we have to create very specific parameters for specific drugs or disease states. but you can also look at a very with novartis and the encology and other countries why not have if your product doesn't work, then pay us? but if your product works, then the payer is doing what they're supposed to do, repaying for good outcomes. our member is sick, we have a good outcome. so the barriers for risk-based sharing products or risk-based

contracts, there are a few pairiers. a lot of time there is a delay in claim submission, so you have to pull claims, continue to look and further back into your claims. if you have certain pieces of information you're looking for. a risk-based contract can actually work for diagnostic products, it can work for new technology products, and it can work for pharmaceutical products. you just have to have a collaborative effort between the health plan, the payer and the provider, the maker, the manufacturer. and in doing that you're going to, you know, you need to identify a product. the therapeutic category. identify the member that you are going to be helping. why are we going to spend money on products where, you know, it

is approved, it showed some effectiveness? you have three different products out this competing against each other. you pick the best one, and if it doesn't work, then they pay us back. so there is a huge future out there for risk-based contracting. we have -- i do believe that, you know, we want to have the value of your drug. so with the value of the drug back it up with money. if it doesn't work, reimburse us. so, basically, it's a system that completely works around with each other. there are many -- in biologics there are going to be, there's a lot of potential for this, for this type of contracting in new

biologic medical products. >> thank you. our last panel cyst is dr -- panelist is dr. steve shak at medical health. >> thank you, dana. .. >> chemotherapy, the best available evidence was offered to all women even though it was

shown in those clinical trials that established the standard of care. that only 4 women out of 100 women treated with chemotherapy actually benefited from chemotherapy. thus, we punish the many to benefit the few. as biomedical innovators, we found in gegnomic health in 2000 to establish molecular tests so the patients and the physicians can select the right tests on the underlying biology of their disease and of reliable convincing evidence. we were able to do this now in this century because of the incredible breakthroughs in technology such as the human genome project led by francis collins that at least told us

who are the molecular players that we needed to study and new diagnostic technologies such as pcr which were developed in the last century in company research labs. our breast cancer test is frequently highlighted now as being a poster child as an innovation that is both improving patient outcomes as well as being cost-savings. so it has been available now since 2004. it has been used by more than 200,000 patients, 60,000 in the past year. and it's now offered and provided to women in more than 60 countries around the world. actually, that's good news. we created jobs in california that are now actively busy since every tissue that is analyzed comes to redwood city in california is analyzed there with the results sent back to physicians and patients all around the world. so what are some of the key

lessons or principles here? what are some of the examples where there was actually innovation, not only technically in the lab but also in terms of clinical research, in terms of payment policy and practice and in terms of communication to patients and physicians? well, i think the one thing i want to highlight is certainly that those innovations actually didn't work -- were envisioned from the very beginning so number one, innovators, scientists, patients, payers and regulators worked together from the very beginning to find the evidence that was needed and that allowed us to do the right studies and have a chance of getting the right evidence. we did some focus groups around 2000. and we asked physicians about this idea, would it be a good idea to have a test to select which patients with breast cancer, new chemotherapy and, of course, the physicians said yes, but when asked the question,

well, when do you expect such breakthroughs to occur, their answer was, perhaps in 10 or 15 years. we also asked patients in 2000 about their view on this advance on whether it was necessary or needed and not surprisingly, the patient said you mean you can't do that already? there's an urgency to bring innovation into clinical practice. so what we're -- again, three or four examples of innovations with regard to payment. first, with regard to the reimbursement of test, i think it's covered more than 95% of lives in the united states are covered. but it required value-based pricing. we are reimbursed at $4,000.75, $4,075 which is the list price and then that list price. up to $2,000 per patient is

saved by using the dx test with regard to the costs -- of direct costs of tests. second of all, there were some specifically payers that were inoratetive and working on their oncologists. highmark and care first went to they're oncologists first if you establish pathways in best practices in breast cancer care, we will reimburse you at a higher rate and, thus, they encouraged best practices in the way that they approached the reimbursement oncatype ds in breast cancer care. third, this is an ovation outside of the united states. the largest health insurer in israel was the first country actually to provide reimbursement outside of the united states. and they did something very simple. and it was commonsense but it

was very, very important and effective, which is that they actually asked all the physicians when they ordered onchtype dx put on a form what were the other standard measures because we had shown in our studies that our test was performed better than standard measures. that you couldn't predict the recurrent score by just looking at the standard measures and they also asked write down, please, what treatment you would give. and then since they were a payer, they could go back and look, what was the actual patient -- the treatment that was given and received. and they could document that treatment decisions were changed more than a third of the time. and they could quickly document from a cost point of view that this made sense within the israeli health care system. by the way, they did this to compete with other payers in israel, not because the central government had already included

o oncotype dx in the payroll. and this was done by united health care and other payers, we will share from the point of view that there was a concern that the test would be ordered and then not used appropriately. that low recurrent scored patients would be given chemotherapy anyway. and so we actually had a arrangements if that wasn't the case, that they would take some of the risk in terms of the cost -- actually, there are now 11 studies that show that test is used appropriately now here in the united states. and in five other countries as well as the united health care population. so we do pay for performance today. too often, though, for us as patients, it is for performance that is not patient oriented. it is for performance that is not best practice. and sometimes it's for performance which is not yet available and so i think it is

really the light bulb that we are and important that we're having this session today to focus on innovation not just in terms of innovative new devices, drugs or technologies but innovations in the system. to conclude, biomedical innovations in the 21st century can improve patient health and be cost-effective. gegnomics, biomarkers and next generation sequencing and we haven't even talked about that -- boy, is that moving much more quickly than i or anyone else had predicted. that promises really to continue to drive personalized care for all patients. however, in addition to next generation consequencing, we clearly also need to have continued innovation in next generation clinical research, next generation reimbursement, policy and practice. next generation dissemination of information and communication.

and it's only with that that we can deliver the care that we as -- that we all urgently need, spending our health care dollars wisely. thank you. >> thank you. >> so in thinking about these performance-based contracts, i think a few points -- norm, if you would take your extreme which is essentially a money-back guarantee if your product or drug doesn't work, then the manufacturer doesn't get reimbursed. there are a few features that make some of these contracts more valuable than others. the first, obviously, is that it should probably be an expensive service if it's cheap, we usually cover it. it's not worth monitoring. but it also needs to be easy to monitor the outcome. so if you need to do a pet scan to figure out whether someone is responding to treatment, it may defeat the purpose of having such a contract just because it's costly to do it.

but i didn't hear anyone mention one other feature of these contracts that i'd like to ask but, which is as rob pointed out in the last panel, we have a problem with adherence. when you move outside the clinical trials, patients often drop off their therapy. and if these contracts are structured so that it says to payers as well as patients you can get your money back but only if you comply with the regiment, is there any hope actually that these performance-based contracts could actually stimulate an improvement in adherence? >> well, i think it's important that you raise this because there is in addition to the -- we have the simultaneous development of at-risk contracting for the delivery system. we also have now the development of new benefit program and design for consumers that also gives them incentives for

aligning their choices and behavior. so i think there's no question that that is going to occur. you are going to inevitably a rapid escalation in the use of narrow physician and hospital networks. narrow networks are going to be based upon the ability of the delivery system to prove that it has a distinguishing abilities and to have care that is at a lower cost. financial benefit plans offered by the purchasers of health care will increasingly drive patients to align through those narrower networks and give them incentives and in some cases disincentives not to do that. combining that then with their own behavior and their choices. well, all of these things. so you really are starting to see a landscape that has a

transparency around performance and behavior by the manufacturer, the delivery system, and the individual -- all of that being much more transparent and having the financial incentives across-the-board starting to line up. this is the new frontier that we are on. >> we'll take questions in a second but one more point. reed, you also noted in your introductory remarks that you've had modest success with these performance-based contracts. if you look at the experience of medicare, it's been very modest incentives. and the question is, if we had much -- these risk-based contracts that we see in the pharmaceutical side are rather extreme. you don't get paid if you don't -- if you don't see a response. what if we did the same thing on the physician side? would you see larger effects? >> the only reason why i'm being cautious in the use of my words is because we are talking about science. we're talking about innovation.

and so the studies that are analyzing the actual real cost demonstration savings was, of course, always starting with the quality improvement first. those studies are still a little bit away. the data is being collected and analyzed. we're probably a year out from having the kind of scalable level of data to prove the point more definitively. so the only reason why i'm being cautious in my terminology is because the studies are underway. if you look at the california experience, where so much of this capitated design -- we have much more history of that, we're seeing certainly some slight effects but are we seeing dramatic decreases in the overall cost curve? not really. so what that says is that this new extra tool that we're putting in the toolbox called performance-based contracting, where we're actually now really much more specifically aligning the contract with the outcomes is really, i think, starting to be the catalyst that will move us forward. i'm being cautious only because

we're at the beginning of the revolution. >> not 'cause of the television? [laughter] >> steve? >> i would add another example where there are drivers other than money that influence physicians and can make -- and change practice. where i've been the inventor after it had been approved we started a registerry that allowed to us collect data. we opened it all to centers and cf patients in the united states and it ultimately enrolled, i think, 70 to 80% of all patients in the united states. we collected data actually on every outcome and i give the cf foundation a lot of credit really spearheading this effort. but the interesting thing that we did from a medical policy and delivery point of view is since we had information on every encounter, in cystic fibrosis, there's a metric that tells you

quality of care. one key metric which is how normal or healthy are kids' lungs as a function of age. and we took 150 centers and could evaluate them in terms of the quality of care. that indeed, some of the centers were in the top 10. some were in the bottom 10. of course, every center thought they were in the top 10. initially, we fed this information back to each center privately. it greatly increased their attention to what are the practice patterns that deliver the best health outcomes to our cystic fibrosis patients. the registry also allowed to us correlate which practice patterns were associated with better outcomes at the best centers. we can drive that kind of analysis across the health care system if we invest in it and if

we continue to develop the people and the expertise to do that right. so i think that's another example of a practice that indeed allows physicians to do what they want to do, which is why they went into medicine, which is to give patients the best available care. >> the other reason why it's important in terms of the registries, in addition -- 'cause we've talked a lot in terms of -- this question you asked about consumers as being sort of a financial hammer and stick or carrot or incentive. that's important but, you know, that's not really what's fun. what's really fun are the tools that are now available to help consumers make better choices and decisions. the new information systems, coaching, and counseling -- the registry data and information is going to be key because now what you'll be able to sort of say to the person is, hey, look, we actually have some real information about this particular intervention that can help make a rational, prudent person make a rationally and personally appropriate choice. >> by the way, the cf foundation got a letter a couple of years

ago saying, there is this information on different health care centers. can you make it public? 'cause before it had all been private. i give the foundation credit. they actually several years ago made that public, so every patient and their family could indeed look to see what was -- what was the quality of my center, how is it doing? and that is very powerful. >> i don't think there's any question that one of the themes started with mark's introductory remarks has been registries and data are going to help and, in fact, are necessary in order to have performance-based contracts and to measure value. so i'd like to have questions -- did you have a comment? >> i had a quick comment on the issue of data information and how we bring it together because if we think the technical revolution we are seeing, which is personalized medicine is really is going to happen then you don't have data to merit

because you don't speak of millions and by individuals. i'm worried about looking at the analytics as a solution of everything, what you end up is missing, you know, the individualization of treatments revolution that is just happening as we speak. just a point of caution there. >> so dana, cathy bhutto from johnson & johnson. when you talk about -- i think several of you mentioned that performance-based or risk-sharing agreements are kind of money-back guarantees, and they're not. we've entered into probably a dozen different arrangements around the world. and actually they're more like what herve just described. i don't know if you heard his description. they are based on certain metrics of what would be considered success in the way a drug is being used in a population. money-back guarantee is not a risk-sharing agreement.

it's shifting risk to the manufacturer 100% for 100% efficacy. they really depend on their being a very good diagnostic to see whether therapeutic benefits has actually occurred or not so they're very hard to execute. so i want to be clear that people shouldn't be thinking about themselves too simplistically. and my question is really back, i think, herve's point just now which is, i thought the panel was going to address more directly this issue of how payment policy can actually stimulate innovation. and one area that i know companies are concerned about is personalized medicine because we're investing in companion diagnostics, biomarkers that's going to mean a smaller population, not the end of one, necessarily. how are the economics going to work? and if they don't work, will that investment slow down or even really just, you know, really dry up?

so i'd really ask the panel to address that issue of how the economics are going to work as we drive to what we can do through science to come to a more personalized solution. >> yeah, it's a good question. i'll hope it up to the panelists. but i would just say -- i mean, a good example is schizophrenia where you have differential response to atypical antipsychotics and one of the responses you see in benefit design may be to take the one that is the average that works best in the population, but that means there's a lot of patients potentially who wouldn't respond so let me open it up, norma? >> dana, first, i'd like to say i didn't mean to make it very simplistic because risk-based contracting is not as simplistic as does it work or does it not? we do have to look at metrics was the patient compliant with their medication? was -- was there any particular,

again, diagnosis previously to them being on this medication that does not work well with the medication? so it's not as simplistic as -- as it seems. >> one point about the u.s. standard, i think, is very striking when you compare to 50, 60 countries where we are operating is a complex city of a number of people you have to speak to get anything done. and it's very -- and still, i've lived in this country for 20 years so it's not about the country itself, but it's about really how the system works. and also the fact that there are some rules that are really inherited from the time when the pharma industry was chemistry. so like paper or kilogram of this and kilogram of that. when a patient has to increase a dose of one of their drugs that they are using, maybe because of

metabolism or anything, they will have more and the dose is higher because in terms of value, in terms of what the drug is doing. in fact, it isn't changing. if you take 800 or 400 milligrams of drug, it's doing what it does for two different patients. and the result of that, for example, is that injectable product and our product are treated and you have tradeoffs in terms of using injectables but it's not based on the value but it's based on the way the system works. i mean, speaking of revolution from early this morning, i think there's really something to be thought about of sort of trying to put that on the roof that take the value to patients as a route the way the system is built instead of all of the processes that have been historically built. >> and actually, there's an interesting lesson for manufacturers. the company that developed evaston and when it came time to use it for eye disease where you

needed vastly smaller quantities, the price went from several thousand dollars in the cancer context to about $50 for the eye. and it led to the development of lucentis. >> reed, i'll give you the last word. >> i think cathy asked a very complex and very appropriate question. and by the way, as you knew when you asked it, there's no easy solution. we have to deconstruct this problem in its pieces. first and foremost, i think that what is important is that we are able to have forum for a collegially and informed discourse about these tough challenges across these industries. and i think that institutions like the percori, to compared effective research venue, those places are going to be hugely important as a place to have the dialog. second, i think that the reaction that people on my side

of the fence anyway have to deal with is the absolute need to control what the chaos that exist das. there's just no way people can continue like they are acting. and my comments were directed to how do we get control over what's happening now? it's incompatible with the ability of the average american to afford it. third, once you get at that, then we start getting at least having some ground rules, some fundamentals around how this innovation gets identified as appropriate and how you introduce it into the system? that's why this oncotype dx is such a good example. getting the manufacturer engaged directly as to how their customers use the product and making sure we start to get at this waste, this inappropriately which is morally unacceptable. that's got to get done.

when you finally get those sort of things carved out, now we start getting down to whether or not -- excuse me, and the last piece of that is going to be unit cost pricing. we love comparative effective research. we like coverage with evidence development. we get all excited with coverage evidence development and get in touch with pharmaceuticals to do that. the problem always start when you start to get at that, it becomes a floor for unit price negotiation and the floor is almost always give us what's been given now as a point of departure because the people are worried -- so it doesn't take costs down. it starts out with, okay, this is just -- the basis of my point it really gets ugly in terms of unit cost and negotiating, which you're trying to represent the interests of the american people and affordibility and it's a tough conversation. so then you take that out and you start -- now you get down to, okay, if you have a smaller epidemiological base that is appropriate for your intervention, what does that do to your pricing?

well, that's a conversation we all need to have. and you're not going to solve it in a room like this. but all i'm saying -- and i don't want you -- i'm sad that you're frustrated with us because if you don't clear out the other cobwebs, you can't get there. but if you start where you were and you don't deal with all this other stuff, we'll be bankrupt before we get there. >> that's all the time we have so thank you -- i want to thank our panelists. [applause] [inaudible conversations]

[inaudible conversations] >> and this is a panel focusing on issues in measuring innovation and value. while everyone up here is getting mic'd up, i want to make just a few introductory remarks to connect what we're doing on this panel to what you've been hearing about so far this morning. you've already heard a lot about the importance of measuring performance of medical care and performance of the key technologies involved in delivering better health to patients. i'm going to start out with newt gingrich's remarks this morning about pushing towards a mechanism for learning much more and much more quickly from the actual delivery of care and from a transformed approach to the development of medical technologies. you heard people on the first panel from scott gottlieb that if we are moving this direction of payments tied more to the

results of medical technologies, that good measures of the outcome associated with those technologies and the impact of those technologies. good outcomes are needed especially for long-term outcomes. you've heard on the last panel about how important it is to have this information in a way that is collected both efficiently and includes relevant measures both of outcomes and impacts on costs. that's something that reed was very passionate about. herve on the last panel emphasized the point about making sure that we're paying attention to individualization of these treatments and their use as well. so that creates a lot of challenges for measurement in bringing all of these kinds of reforms to bear as well as developing the evidence that can guide treatment decisions using a vast array of more personalized technologies more effectively. and that's what we want to focus

on in this panel. how are we going about measuring the value of innovative technology, particularly on more individualized application. how is our current infrastructure doing in supporting those efforts, our electronic data systems, their lack thereof, our methods of collecting information on medical technologies, both before and after they're approved and out in use? and how can we make more progress in these areas? we've got another very diverse and distinguished panel this morning to help us discuss these issues. our presenters include uday kumar who's the founder and chief medical officer of irhythm technology, vicki seyfert-margolis who's the senior advisor for science innovation and policy for the food and drug administration. lou garrison who's a professor in pharmaceutical outcomes research and policy in the department of pharmacy and an adjunct professor in the

departments of global health and health services at the university of washington. and sean tunis who's the founder and director for the medical technology policy. so as we've done in the last panel, i'll ask people with me on stage to make a few opening comments and then we're going to turn this into a discussion about these issues related to measuring value in medical innovation. so let me start with uday. when you founded irhythm technology, uday, i know your explicit goal was to develop medical technologies that not just improved health but also saved money and the explicit goal in process to be able to measure how that was occurring. so maybe you can tell us about that experience and the implications for what we're talking about today. >> great, thanks for having me. i think as a representative of startup, as a founder of startup i think the challenges from startup companies are more challenges than larger companies. so if you're a patient who might feel light headed or

palpitation, you have a heart fluttering you'll go about your day it's nothing. you may go to your primary care doctor who say well, you look fine now and let's watch what happens. if you go to this a emergency room, let's say, they may say you look fine and the symptoms may be gone. come back. and you'll go through this process multiple times and potentially finally get to your primary care to a specialist like myself a cardiologist. i would order a test and i get paid for a test but those tests may monitor you a day or two and you have your symptoms a week or two and i don't want anything and you have tests that are longer like the microphone i'm wearing a pager with wires. it's very hard to comply with. you want to wear it for several weeks, you don't. so we don't find what happens. i does physician gets paid for that test so this process continues and continues and finally you'll finally eventually get a test that may prove or not prove that you have an abnormal heart rhythm from

your heart system. you may found it's a crazy system from a physician standpoint. what do patients want? they want to be diagnosed when they're first seen. we realize there's a big gap how you can make the system more efficient by focusing on value as actually a requirement in developing a new technology. so what we developed was really a simple little patch. all it does is stick to your body for up to 14 days while you sleep and shower with it. it's disposable. it can be put on an emergency room or by the primary care doctor. the whole idea it gives you enough time and it's highly compliant and patients can use it and get an answer. so we wanted to figure out can we obviat multiple tests of medical costs as well as reduced costs to get to the same outcome? so that's what we do? in a nutshell we realize the value and innovation could be things that we could use to promote ourselves and was actually a benefit to what we're doing. what did i learn? two or three things from a small company entrepreneurial

standpoint and again, i'm happy to say, you know, small company innovator we have created 330 jobs we're being sold all over the country but we have to underwrite what we do. while we can comply with some existing codes there's no thing for exactly what we do right now. so what's important is that companies and payers expect companies to raise tens of millions, hundreds of millions of dollars to fund the usage of their products. once through the fda. while big companies may have the luxury of doing that, the funding environment is much more challenging these days for funders, venture capitalists even corporate to put in that money. so that is a real stifler of innovation. one of the first things i've learned is little companies have to raise lots of money to actually prove that they're a benefit. second thing i've learned is, payers need to be potentially involved. i mean, we would love to go to payers early and say, look, we can model what we're doing but you have to accept that modeling may be all we can do. we can't wait three years to prove what we do 'cause we'll be out of business by then so you can say we'll pay for you but you're not paying for anything since we don't exist anymore. that doesn't work.

since little companies are the drivers of a lot of innovations that really big companies also buy up, this is creator of u.s. jobs, it is important to think about payers having a stake in the game if we can come to some model of what the reasonable approach so modeling what we do, that, okay, let's all risk share, risk sharing is more like i've gone to the fda i have an approved product and help me pay for it and if i keep proving my outcomes, to get to a regulatory approval why can't there be milestones for achieving certain clinical outcomes that can translate the cost? the third thing is, payers have data. they say they want data from companies. i as a little company say i want your data. you know the historical context and the cost of what it costs you to pay for a disease state. over time. so payers need to understand that cost is not measured on an annual beneficial cycle. and it may be different for dying nottics and therapeutics. if i could have access to di

nothing -- diagnostics and if i hit something and you pay a little more and hit something pay a little more so that we're all going in the same direction. but i don't have the money to pay for, you know, a pharmaceutical, pay lots of money to consultants to look at medicare databases so that's difficult. the final thing would be, how do we put this in a framework so i'm trying to think about and some of the other efforts that i do how do we simplify this process so others can think similarly? there's four things that most things fit into acute dying diagnostics and there's glucose monitoring every day and acute therapeutic an implant of a stint or an ied or taking a drug every day or getting diall list every day. four different bug buckets that reasonably cover most things and if we can understand what are the outcomes for each over the course of a patient's lifetime

for that disease state based on historical data that payers have on the cost of that historical data, maybe there could be some alignment of what i need to do from a pricing perspective. payers will be willing to pay early on to support getting these innovations that ultimately benefit them. and then finally, one of the other hats i wear at stanford is teaching innovation in a global context in other countries, india, singapore. now, what i've seen is those countries cannot only create products just as good as ours but they do and have done so thinking about costs in a different society. so if we want to remain globally competitive it's not creating value for our own citizens but we're -- med tech is one of the biggest exporters outside of the u.s. we will fast become an importer if we don't try to figure out ways to reduce the barriers for innovators, entrepreneurs, small companies to be able to survive a tough funding environment. it's just not feasible to expect millions and millions of dollars to be raised to prove something

that everyone can agree is intuitively potentially obvious but how can we come to some