utilization, because to change utilization requires changing how physicians think, and then changing how they behave. and what we've seen in year two, and what we're seeing now in years three and four is that those utilization changes have really started to take hold. so the infrastructure that they're putting in place, for example, to prevent avoidable admissions, avoidable use of the emergency department by doing innovative things like having a nurse practitioner in the emergency room to catch the patient as they come through and triage and figure out is this a patient who really needs emergency care, or does this patient need urgent care? because if they need urgent care, let's take care of them over here where we won't incur the expense of a emergency room visit. we'll take care of the patient's needs. they won't wind up in a bed. because when you have a hospital and isolation and emergency room, sometimes you not only get the emergency room visit, but then you then get an in-patient admission too. so they're putting infrastructure in place to make

some significant changes in utilization that we see in years two and four yielding even deeper savings than they got in year one through the site of service moves. but we all have to realize that changing utilization is that tougher task, because it does involve changing how physicians think and how they act. >> i would like to go back to your data question, because i think it might make you feel a little better about it. we're not at two. we've come a long way. if i were to rank us, i would probably be a six. >> nationwide? i'm talking about nationwide. >> yes. >> six nationwide. >> here is what we just purchased a company that, invita that has a rules engine that we're able to run every one of our 2,200 medicare members through that engine overnight, and it will deliver back to us actionable gaps in care, including issues with drugs not

being filled, been filled and not refilled over a 30-day period. so the first time we ran it, it identified 355,000 actionable gaps in care that we could then turn around and turn over to our teams. and as a result of that, working with the physicians, 31% of the gaps were converted into actions to improve outcomes for the members. so this is a brand-new company that we just purchased. and we can run, like i said, full data through it overnight. >> if i can add to that. one is the sort of data you're referring to which we take advantage of as well has been available in the managed care world for a long time. we can get drug data. we have claims information. we have information from laboratories. i think the point that you were raising relates to interoperability among the emrs where the interest is in richer clinical data. what do the radiology tests show, what does the pathology show, the physical exams show. the way we address that is through company which does it

through brute force. so while we're waiting for standards of inoperability, i would refer you to the 850 hospitals, 200,000 physicians who are linked to this system. a couple specific examples. one is in virginia which is using this capability. and the other is the banner health care system based in arizona, which is actually using it to support a medicare pioneer aco grant. so despite the fact that they have multiple emrs, the ability to couple the traditional data we've always had with interopable data we have now accessed through brute force provides the substrate of information that you were referring to. i would argue that isn't sufficient and a unique abilities to convert that massive amount of information, after all it's quite a bit, usually on complex patients, into activities. so as you think about a patient who is on ten drugs, by definition, they may have ten different diseases. hundreds of different lab

results. how can any conversation notice where the insecurity that drove to the inspiration for the act of health for me as a physician, i couldn't keep up with the literature. i didn't know what other physicians were doing. and the ability to in fact create this composite of data, introduce this interrogation capability with clinical decision support, and then in fact define discreet activities to pursue is really what we're trying to introduce around the country. >> thank you, mr. chairman. and my apologies for being late. too many hearings simultaneously. i know this has been a very good panel. and a i want to start with a question that stems from what i have heard all of you say. not just today, but repeatedly that you're payers. and when you come before the congress and talk to us about issues, it almost always comes back to information which really means data. and you need access to it. and particularly global data.

because you can really only look at what is inside your system. and the fact is that under federal law, you can't really get access to the data. now senator grassley and i want to change that we have a bipartisan bill to open up the medicare database so that it would be possible to look at i think what you call global information, be able to compare what you have in your system to others. and dr. reisman and mr. cardoza, i think you all in particular, i think it's generally true of all five of you, you are really sort of the point persons on this question. dr. reisman, would this be helpful to you, and how would you assess the need for effort legislatively to open up the medicare database so that you really could get access to this kind of information and use it

to drive improved quality and hold down costs? >> so i would suggest that there were two elements of this discussion of data. one is retrospective analysis of aggregate data to identify trends, to support comparative effective research to understand what really works best. and we think that's enormously important. in fact, we're working with the administration and todd park at the cto office in order to in fact take advantage of those data capabilities. and one of the capabilities that we bring to that is the ability to apply our analytics to ask some of these important questions. the other element which is related, but i just want to define it as being separate is the notion of availability of real-time data at the point of care to support the physician in regard to taking care of the patient who is sitting in front of him. so there should be in fact the record locator that would allow me to identify data about you, analyze those data, and make sure that what i'm doing for you

is, again, consistent with the best clinical evidence and is not contraindicated relative to other activities that other doctors are pursuing with you. thing is the aggregate and the real-time. but in any case, the availability of information that residence within the medicare database would be enormously important for a number of reasons. >> i would second that. i think what there is to be encouraged about on this topic that we're talking about it, there is consensus that data matters and sharing it among cl clinicians matter. it wasn't long ago you couldn't have that data. physicians are fiercely protective of their medical records. we've seen a sea change in the last three years on working with physicians on this topic. they're coming to understand, as i said in my opening comments, medicine is really complicated now. and it takes a team. it's not an individual walking into his office in the morning and back out at night. and he is all by himself and he is taking care of the patient. those days are past. so they know they need to

interact with other physicians. they understand the importance of sharing data. and by law, patients have access to their records. and why wouldn't other physicians involved in the care, team care of that patient not have access to them as well? we are at the advent of this. but i think it's going to move fairly quickly. >> why don't i bring the other three of our valuable witnesses in to the second topic i wanted to ask. and if any of you would like to elaborate on the question medicare database, certainly we can do that either in writing or as you respond to this. but the second question i wanted to ask all of you since you come from the private sector, and you watch the federal government. and obviously the federal government to all of you sometimes looks like it's moving very slow, and slow to change and slow to adopt and slow to evolve. and traditional medicare, even as we talk today is still in the sort of demonstration project

kind of stage. what would be your recommendations for speeding all of this up? and particularly you've got the chairman, ranking minority member here. we're in a position to look at ways to speed up and accelerate these changes so they get out of the demonstration kind of project area. and can be sped up. so why don't we take our other three witnesses who didn't get a crack at that one and your counsel on how to speed up changes and reforms for the second one. dr. safran, why don't you start. >> sure. i'd say that over the last couple of years, what we've seen actually is quite impressive speed with respect to the uptake of the accountable care organizations. >> right. >> and that i would leverage that. because as we were talking about before, the key is going to be for medicare to be able to move away from a model of payment that deals with individual

actors and yet holds them accountable for the behaviors of every other doctor across the country. and if others are using too much, my rates are going to go down next year, to a model where i have group -- a group of peers that i have accept issed accountability with. and we're working together to manage total medical expense, quality, and outcomes. so the fact that you have stood up 32 pioneer acos in such a short period of time and that medicare shared savings program is getting under way, i think sort of sets out the beginning of a continuum that to me actually reminds me very much of the way we waded in to the aqc model that i talked about today. when we launched the aqc in 2009, we hoped that by the end of that year or possibly the following year we might have 10 or 15% of our network accepting that broad accountability for total medical expense, quality,

and outcomes. by the end of year one we had a quarter of our network contracted that way. and at this time we have close to 80% of our network across the state contracted in that way. why did it happen? why do we have that fast uptake? i think there are lessons to be learned for the federal government. and a big part of it was it was voluntary to begin. we weren't forcing anybody in. we said if you believe this is a better way and you can see that you can earn well under this model by making care better and by contributing to affordability over the long-term, then come on in to this contract. and then what i think led to the rapid acceleration was a couple of things. one, organizations started to see that the initial pioneers, no pun intended, in our aqc model were succeeding, both at improving quality and at managing to their budgets. second, they saw that the fee-for-service system was starting to look pretty

unattractive, was starting to look like low or no payment increases, no real opportunities to advance. and that created some acceleration. they started to understand that the kind of support they were getting from us as a payer and that i think the federal government will have to work out similar models to help them as they transitioned from a volume-based system to a value-based system. >> take that last point, because i think that's the ball game. >> yeah. >> i think that is the ball game, and that of course is what we started essentially almost three decades ago in our part of the world, whether it's group health up in seattle or providence or other kinds of plans in our area. what could the federal government do to accelerate that transition. >> right. >> beyond fee for service? >> well, i think there it goes back to your earlier question about the data sets, right?

because from washington, d.c. or baltimore, it will be hard to partner with the provider organizations that have the courage to sign up for these new models. in the ways we've seen have been critical to their success in our market. but imagine that if those who sign up for it are able to partner with their private pairs who are also paying in that model, and if those private payers and the providers who have come into it have all the data to work with, if we could be doing the same rich analytics for the providers in our market that are aqc organizations and also medicare pioneers, this would be enormous assistance to them. if we could then take those analytics and be helping them with the performance improvement guidance that we give them on the commercial side, giving them that same guidance on the medicare side, i think you'd start to see more rapid uptake

across the country because fear is one of the rate limiters right now. i think folks think i wouldn't know the first thing about how to transition from a system that pays me for every unit i produce to a system that is now going to ask me to have accountability for overall spending and quality. so you have to help them. >> mr. chairman, when a witness says that they support the effort along the lines of what senator grassley and i are talking about to expand access to medicare data and they want to promote transition beyond fee for service, i usually think i ought to quit while i'm ahead. so i really -- >> you're doing just great. >> i thank you for the time. >> to follow upon your first point, and you mentioned it, mr. cardoza, that is earlier physician resistance to access to medicare data. i assume some positions are, you know, proud of their building practices, wondering whether their billing practices will be

questioned. maybe there is some medical liability issues there. and i'm just wondering if you could help us figure how to bridge that gap. because i do think it makes sense for that data to be available. but we should do it in a way that is sensitive to legitimate physician concerns. >> well, it's a journey. and in our setting, we've been doing this for 25 years. and our 3,000 plus physicians are in an accountable structure, and they know they're being watched. so in areas of the country where there is no transparency at all and they just walk into their silo in the morning and out at night, yeah, there is going to be some trepidation, just that somebody else is going to be looking. but you have to go there. and in response to the question of how do we accelerate, how do we get there, put the money where you want the system to go, and it will go there. so payment reform has to precede

delivery system reform. it has to enable delivery system reform. so the more we can create population management -- i mean population-based reimbursement methodologies along the lines being espoused in massachusetts, the faster we will get there. >> so do you suggest modified payment reform under medicare? >> yes. >> and what would it be? >> well, you just went above my pay grade. i'm much more eloquent describing the problem. i do think the underlying principles as we've all been talking about are to put in place policies that give physicians reason to group up, to get connected to organizations so that they have -- because i'm telling you, the physicians on their own can't do this. it's not what they were trained to do. it's not what they signed up for. >> right. >> the expectations of them now are very different from what they thought they were signing up for. they're okay with it. they're willing to sign up for it because they understand it's the right way to go.

they just lack the skills and wherewithal to do it. so organizations like ours, the kinds of organizations in massachusetts that has been described are enabling structures for them to do what they would like to do if they could. at the granular level, they have no chance. >> could i -- i'm sorry, could i just suggest we link the data question to some of the incentive questions too. if in fact we had access to these data and the purpose was so say you're a bad guy and you're a good guy, obviously physicians are concerned about. but suppose we shifted the incentives and we're talking about managing real populations and we said gee, there is a population in an adjacent county where in fact the number of coronary angiograms done is half as many as you do. and by the way, the incidents of obstructive coronary disease is three times higher, suggesting that you're doing angiograms on people who in fact don't need them. in fact, we could ascertain that as well. if we went to that community and said gee, we're actually changing the payment structure from fee for service,

notwithstanding the sgr issues to one where you in fact will receive a case rate or a global rate for your community. and by the way, by looking at this cms data, we in fact can assure you that by reducing utilization and being a little more thoughtful about a your use of angioography, you in fact could put yourself in a position where you could responsibly assume risk, financial risk without compromising the care of your population. so i think for a lot of these issues, we need to think about companion solutions and actually collect and combine some of the issues that we've been talking about. >> don't underestimate the power of peer pressure. if we can profile these practices and create the data and make that data available to people, we had two large cardiology groups in adjacent counties. and the utilization practices in one of those counties was egregious. we went to those cardiologists and we showed them their data compared to the next county. and if we had just sent that out to them and not engaged them, they would have thought, well, i

guess that means we're diagnose a better job. so instead we were able to engage them, hold their feet to the fire. and now two years later their utilization practices are exactly what the other county is. driving toward the mean. >> it is working in mcallen, texas? >> i don't know. >> and the incentives -- but the incentives aren't there. >> i'm talking about peer pressure. >> not alone. >> pass on mcallen. it was written several years ago that is going to guide us. >> my understanding is there is some movement there since they have been exposed. >> we're finding peer pressure to the point, agree with the point. among the 300 panels we formed of small groups of primaries, there is peer pressure within the panel. and then there is peer pressure across panels, how. adoing relative to others. you could have two physicians in a panel of ten that are high, wide, and handsome. and the other eight have their incentives based on how the

total panel does, and they start to police themselves. >> doctors hate being an outlier. they just hate it. >> but theys will like making money. i would suggest peer pressure plus financial incentives create a synergier? plus financial incentives create a synergier synergistic model. >> what is the role of medical schools here? >> i think there is a considerable role. based on my experience recently, these issues aren't being addressed particularly at all. there is a little bit more of a focus on primary care. there is a need to further acknowledge the contribution that other types of practitioners can make there is a lot of anxiety of course about compensation around primary care. but we're not doing nearly enough to in fact introduce these issues to the curriculum in medical school. >> senator baucus, i want to go back to a point i made earlier.

we're not training physicians today to enter into a health care delivery system. we're training them to be medical rescuers. we're training high-tech, giving them lots of tools, and that's where the money, and that's where the glamour is. it's a real problem. so medical schools are not doing what is needed today. but i -- you know, so is a lot of the system. i'm not going to demonize them. i think if we start setting this out there and challenging them, i think they can move in this direction. but they're not there now. they're training medical rescue. >> this tension between specialists and primary care docs. it's my understanding that a lot of the medicare reimbursement schedules is contracted out like to ama and weighted towards specialists to the disadvantage of primary care physicians. i don't know if that description is accurate. but just your thoughts on how we

can deal with this difference in reimbursement between special -- i don't want to take anything away from the specialists. but your point triggered my thought. everybody is trained in the high glamour stuff and technology, and that's where the money is and so on and so forth, and probably be a bit solid as well. i don't know. but i'm trying to figure out how we get a little more focus on primary care physicians here. >> i think that the models that you have heard us discuss today, while we haven't explicitly said it, each of them is primary care centered. so i'll speak for our model. the only requirement we have of an aqc organization from the perspective of what that organization has to look like is it must have primary care at the center. beyond that, if they want to have specialists in their contract, if they want a hospital as a partner in their contract, they may, but they don't have to. they still have to be accountable for that whole care across the continuum.

well, that coupled with the fact that the quality incentives are so largely primary care-based has really changed the dynamic of power and resources within these organizations. because these organizations understand that they cannot succeed at managing total medical expense and at improving quality and outcomes if they aren't investing in primary care at the core. and so we are seeing them looking to hire more primary care clinicians, physicians as well as nurse practitioners and medical assistants, investing in the infrastructure in primary care practices, rewarding those practices for the success that the organization is having at managing to their budget and improving quality, and it's interestingly changed the dynamic with specialists in a very important way. specialists are sitting forward saying a couple of things. one, gee, how can we be helpful in this new model?

how can we be helpful at managing total medical expense? and aren't there any measures for us? aren't there any good quality measures for us? that's a welcome question because our -- the available measures that are nationally endorsed really at this point very much primary care focus and being able to have accountability for quality of care in specialty environment is very important. >> i agree with dr. safran. i think the pcp is the quarterback. >> say it again? >> the pcp is the quarterback of the team and they need to funnel the care to the efficient specialists that they have. our most efficient model that we have is where the specialists are capitated and the pcp is driving the care to the most efficient specialist that he has in the network. >> i think we again need to get focused on this team-based focused orientation. suppose i'm a specialist doing bariatric surgery and aetna offers a case rate.

you're the surgeon. typically you might ignore the role of the primary care physician with with regard to the patient who has bariatric surgery. but as surgeon i'm at risk if the patient is readmitted with metabolic problems, automatic the things we've been talking about. suddenly the primary care physician is my best friend to the extent i don't want to have to see this patient again. as we realign incent issives and create dependencies if you will on the specialist on the primary care physician in much the same way where the primary care physician refers to the specialists, i think we can restructure the relationships and restructure the reimbursements so the primary care physician is more generously reimbursed. >> we're going to -- i mean i'm not going to ask you to do this. i think it's a bit much. but an earlier panel consisteded of former cms directors and the subject was sgr.

and at the end i decided why not? so i tasked them to come back to us with recommendations how to reform sgr. those recommendations are due tomorrow. i'm trying to think it's too bad we weren't all together here to talk about this. but anyway, so what should we be looking for when they give their recommendations? what are some of the key points that you think are most important in order to help us advance the ball here to get reform of sgr in a way that you think makes sense given your experience and how you compensate physicians, looking forward toward collaborative patient-centered approach and delivery system reforms, et cetera? what should we be focusing, looking at when we get those recommendations? >> where is the value-based

component to the compensation that goes to the physicians. if we're going to continue to just pay for fees for the services they provide, then they're going to just keep providing services. so i would look for that as one thing along with the other themes we talked about. >> i would echo that, that it's got to be a global measure for the outcome of a defined population of patients and a structure of accountability, principally through the primary care physicians. it's not a price movement that you're looking for, it's an overall cost of care. and the only way you can improve that is have that accountability and incentives to get better outcome. >> how do you measure quality? >> largely on outcome. >> how do you measure outcome? >> we look at it principally as reductions in the evidence of the fragmentation of the health care system, having fewer readmissions, fewer er visits, fewer drug interacts, not so many gaps in care. we're looking at outcome measures that show the patient has been stabilized or their

risks mitigated. >> okay. >> i think you back out of things that are most fearful with regard to particular disease state. so the bad thing about being a diabetic isn't your sugar, it's not that you're going to have a heart attack or a stroke or end up on dialysis. so the outcome isn't did you test this or test that or get your sugar to that level, the outcome is did the people end up in fact having strokes, heart attacks or end up on dialysis or blind or any other dreaded complications associated with diabetes. >> that depends on some kind of follow-up records. >> yes. which is a lot of what we're talking about. so you need this longitudinal effort to see how things have turned out in relation to some of the questions senator wyden was asking before. with regard to the sgr question, i would hope that the answer would be more expansive and immediate reaction to sgr in isolation. i would hope there would be companion solutions that are suggested. and to the extent that those

companion solutions are in their recommendations, a realistic assessment about whether or not many practices have the infrastructure, the technological capabilities, even the financial wherewithal to manage this new approach. >> thank you. >> i would echo. a couple of things for physicians. i think you got to get a transition from peace work system that they're in today. they're in a piecework system to one that more appropriately rewards their ability to coordinate care and performance. to me that's one of the biggest things we need to do with the system. >> dr. safran? >> i'd look for five things. i'd look for them to give you a model that moves from a focus on individual actors to a model that focused on organizations. i'd look to them to give you a model that moves from the focus on individual services and the fees for those services to the global view of total medical expense and quality and