[This Transcript is Unedited]
Renaissance Washington Hotel
999 9th Street, NW
Washington, D.C. 20001
Proceedings By:
CASET Associates, Ltd.
Fairfax, Virginia 22030
(703)352-0091
TABLE OF CONTENTS
DR. STEINWACHS: I'm Doug Steinwachs. I'm chair of the Population Subcommittee of the National Committee on Vital and Health Statistics, and it's a pleasure to welcome you to our day and a half of hearings on the medical home on trying to understand measurement. Both the concept itself, but also the processes and the outcomes associated with the medical home.
I think we have an exciting group of presenters and I know that Dr. Justine Carr, who is chair of the Quality and Data subcommittee, who co-sponsors this session with the Population Subcommittee wants to add her welcome to you.
DR. CARR: Thank you. And thank you to all of the folks who helped set this up today. This is tremendous. It's going to a huge educational opportunity and an opportunity also for a dialogue and exchange. So, I look forward to it and I welcome you and thank you.
DR. STEINWACHS: As I think all you know, the National Committee on Vital and Health Statistics is in its 59th year of advising the Secretary for DHSS and predecessor agencies on both data and health information in the health arena. Hearings are held to gather information to understand what are practices in the community and to understand what are some of the issues that may be relevant when we think about both national health policy and the data that supports that.
Now, these hearings are being broadcast on the web and are being transcribed. So, what you say becomes a matter of record. So, I don't know what that says about the jokes that Harry and I usually say. So, we'll be more careful than usual probably. Before we start, Id' like to go around the table and do introductions and then those people who are sitting around chairs outside also to introduce themselves.
I'm Don Steinwachs. I'm at Johns Hopkins University and a member of the National Committee on Vital and Health Statistics.
DR. CARR: Justine Carr, member of the committee and I'm also from Beth Israel Deaconess Medical Center in Boston.
DR. QUINN: I'm Matt Quinn. I'm with the Agency for Healthcare Research and Quality and I'm working with NCVHS.
MR. REYNOLDS: Harry Reynolds, Blue Cross/Blue Shield of North Carolina and member of the NCVHS Committee.
MR. LAND: I'm Garland Land, Executive Director of the National Association for Public Health Statistics and Information Systems and member of the committee.
DR. FRANCIS: I'm Leslie Francis at the University of Utah and I'm a member of the Committee.
DR. GREEN: Larry Green, at the University of Colorado and a member of the committee.
DR. LINKMAN: I'm Larry Linkman. I'm from the University of Michigan. I'm a practicing family doc and a guest of the committee.
DR. McGEENEY: Terry McGeeney, I'm with the American Academy of Family Physicians.
MR. TUCKER: I'm John Tucker. I'm CEO of the American College of Physicians.
DR. PAWLSON: I'm Greg Pawlson, Executive Vice-President of NCQA.
MS. PARTRIDGE: I'm Lee Partridge, Health Policy Advisor, National Partnership for Women and Families.
MR. NORTON: I'm Chris Norton, from IBM Healthcare and Life Sciences.
MR. COHEN: I'm Jim Cohen with the Center for Medicare and Medicaid Services.
DR. BETHELL: I'm Christina Bethell with the Oregon Health and Sciences University and Director of the Child and Adolescent Health Initiative.
DR. MAXFIELD: I'm Miles Maxfield, I'm with Mathematica.
MS. KANAAN: I'm Susan Kanaan, and I'm a writer for the committee. I'll be writing a report about this hearing.
MS. CROSS: Hi. My name is Doria Cross. I'm an intern with the Agency for Health Care Research and Quality.
MS. WILLIAMS: Hi. Kim Williams. I'm with the National Patient Advocate Foundation.
MS. WHISTLER: Laura Whistler with the Blue Cross/Blue Shield Association.
MS. BUELL: Alison Buell, American Health Information Management Association.
MS. CAPP: Mary Capp, Centers for Medicare and Medicaid Services.
MR. MYERS: David Myers with the Agency for Health Care Research and Quality.
MS. JONES: Catherine Jones, National Center for Health Statistics.
MS. JAMISON: Lucy Jamison, National Center for Health Statistics.
MR. DELMONTE: Good afternoon. My name is Mark Delmonte. I'm from the American Academy of Pediatrics.
DR. STEINWACHS: Welcome to everyone. I'd like to give a little bit of credit before we get started to the hard work of the number of individuals that made this hearing possible. In particular, Dr. Matt Quinn, who is with the AHRQ, and has served as staff to the planning group here. And Larry Green has provided the intellectual leadership for it. And some of the rest of us have tried to join in with enthusiasm as we are here to learn, particularly Justine and myself and a number of others who have worked very hard on it.
Staff has also been very active on it. Cynthia Sydney, who is staffing this, as well as Debbie Jackson, Marjorie Greenberg will be here later, and Susan and others have made this possible. My job, as Matt Quinn explained it to me, was to keep us on time. That may be a challenge, but we do want to try to make sure that in every session that we have chances for questions and dialogue as well as presentation. So, that may be a little bit of a balancing act, but we'll try to see what we can do as well.
The first session is part of an overview, and we've asked three people to provide us with different perspectives on the primary care medical home. You should have in your book an agenda, and we'll be following it with a couple of cases. There may be some changes.
So, let me just quickly introduce the three speakers, each of which has been asked to provide us with 5 to 10 minutes presentation, and that will leave a few minutes for discussion before turning to the second half of that session. It's a great pleasure to have Dr. John Tooker here as Executive Vice-President and Chief Executive Officer of the American College of Physicians, on of the partners in coming together here to define where the key elements of the medical home. And he will be talking to us.
We also have Dr. Gregg Pawlson, from NCQA. They just played an important role in certifying medical homes, and as vice-president of NCQA, and very importantly, it's a great pleasure that we have Lee Partridge here from the National Partnership for Women and Families to give us a perspective from the family consumer side on the medical home at the same time we hear from the others. So, please, Dr. Tooker, if you could lead off.
DR. TOOKER: I'm very pleased to be here today on behalf of the American College of Physicians, but also on behalf of the originators of the national principles for the patient centered medical home. My understanding is that Doug Henley, the EVP of the American Academy of Family Physicians was to be here, but that he was unable to attend today. So, I'm trying to fill in and I talked to Doug this morning as to what he might have said. Then, I will follow that up in the next session with more specifics about who the current actors and stakeholders are in the patient-centered medical home.
So, this will be at a pretty lofty altitude without getting into too much of the specifics. A bit of the history is important and I want to begin by saying that this is not an original idea. And I want to give the credit to the American Academy of Pediatrics for developing a patient centered medical home model for special needs children more than 40 years ago.
This idea was resurrected, and it's been very successful in pediatrics. It was resurrected by other communities, including the American Academy of Family Physicians in their major effort to look at the future of family medicine in about 2004. The American College of Physicians followed in 2006 with a major physician paper called the Advanced Medical Home and together, with the two Academies and with the American Osteopathic Association, the American College of Physicians came together with those three organizations in 2007 to develop a national set of principles around patient-centered care.
I think its important today as we talk about what it is to emphasize that we don't know exactly what it is yet because it is a concept that is now being tested. It stems from the work of others initially in primary care and the Institute of Medicine's Work going way back into the 1970s about what are the major attributes of primary care. And we focused on four of those key attributes, and they are the first patient contact. That is to say that if you are a patient in a patient-centered medical home and you need care, you have a predictable contact that you call, 1-800-MY MEDICAL HOME, and you will have a real entity on the other end who will take care of your needs now.
Two is that in addition to always having a home that you can reach in terms of first contact that once you are a member of that home, you are a member of that home. That is to say that care is continuous now. And it's not episodic care the way that so much of our fee-for-service care is today.
Three, that if you are a patient in that home, you should expect that that home will provide a comprehensive set of services for you. And those services are acute care services. They very much are chronic care services. And they are preventive services now, a full range of services and that that care is going to be coordinated. And I think it is readily apparent now that one of the features of a medical home that patients value the most is that that home will coordinate their care and the more complex the care of patients is, and it doesn't matter if it's isobathic medicine, pediatrics, family medicine or internal medicine, these patients, they need to see a lot of different actors within the healthcare system, and one of their major issues is the ability to A) have their care coordinated with a transfer of information to and from the medical home appropriately now, and when I get into the second part of the talk about what the model actually looked like in graphic form, we will exam those relationships more specifically. But, those four elements of primary care are key attributes of the patient-centered medical home.
And others have done a lot of work, both nationally and internationally, as to how do you define medical home and how do you assess whether the care being given is care consistent with primary care. In addition to that, it became apparent that as patients developed more and more a chronic illness and for a Medicare patient now, on average about 25 percent of Medicare patients have five or more chronic conditions requiring a lot of care coordination.
And this care is best coordinated through team-based care. And team based care is another key element of the patient-centered medical home. A credit for this work, building on the key attributes of primary care, goes to Ed Wagner and his colleagues at the McColl Institute at Group Health at Puget Sound, and they developed the Wagner Chronic Care Model.
And that chronic care model is a model that I think you're probably well aware of, and is really designed to help specifically patients with chronic illness now. Building on the attributes of primary care and taking those attributes into a chronic care model, the four primary care societies that we have talked about they developed these attributes further into the patient-centered medical home to be able to care for, not only for chronic illness, but also for acute illness and to provide preventive care to these patients.
So, I want to emphasize probably repeatedly that what we are calling the patient-centered medical home is built on the key attributes of primary care and on the key attributes of the chronic care model developed by Ed Wagner and his colleagues at Group Health, and I'll show you those attributes in a second, in which I have a model of the chronic care model that is well known today.
So, as of 2007, the key elements of a patient-centered medical home were developed now. And the next major step along the way, and I'm glad that Chris is here, is a very large national effort now to test the patient-centered medical home. Greg and I have been on many programs about this now, and we'll talk about how you know a medical home when you see one.
And I want to emphasize where I heard the term certified medical homes, I want to emphasize that we are not talking about the certification of a medical home. What Greg will be talking about is the qualification of a medical home. That is to say, how do you measure the key attributes of a medical home and know when they are present.
And the reason that I am emphasizing that this is not certification is that there are a lot of quality improvement organizations, including the American Board of Medical Specialties who are very engaged in the certification of physicians, and are increasingly wishing to be engaged in the systems-based practice, and there is a lot of work going on now about how do you certify a system in addition to the individual physician now.
I just want to distinguish the roles and responsibilities of these organizations because they often times are confused. So, now that we have the principles, we now need to implement those principles in pilots and demos and that is being done as we speak, roughly 20 to 25 states now where models are being tested. And they stem anywhere from individual models with individual insurers all the way to large multi-payer, multi-player models with both the government at the state and federal level, as well as with private insurers, and we could not have done this work without a large collaborative called The Patient-Centered Primary Care Collaborative, the PCPCC.
This is a large entity now, and Chris can tell you exactly how many players there are because it changes frequently, but there are more than 128 members of this collaborative, and they are critically important representatives of patients, the consumers. And I'm really pleased to see the national partnership here today.
In addition to the consumers, of course, are state and federal governments, what others call the provider community and we are very interested in making sure that all the members of the team are represented. And that includes the physicians, nurses in the form of advanced practice nurses, and nursing medical assistants, pharmacy, nutritionists, certified-diabetes educators, etc.
Because as patients become more chronically ill, they require a wider range of services to care for them in this comprehensive environment, and it very much includes the employers. And the self-insured employers and the leadership of IBM have been remarkable in their support for this national effort. And they helped a great deal in bringing along the health plans to the table as well.
I'll talk about this more when we get to the current stakeholders and actors, but I want to emphasize that all of the major players are at the table now. So, we are at a stage now where this is a national movement. And it has certainly moved on to the attention of MECPAC, as well as Congress. Right now, this room is a little spooky to me because just a couple of days ago, we were having our national meeting of the American College of Physicians here and we were using this very room, although it was for a cocktail party and not for this.
And just before that, I had been up on Capitol Hill with 500 of our members and all 50 states and the District of Columbia and Puerto Rico and the patient-centered medical home was very much on the minds of the staff, if not the members of Congress right now.
So, the whole business now is getting this implemented. And I want to thank you for the opportunity to testify before you because Health and Human Services, particularly at CMS, where the Medicare Demonstration Project on the patient-centered medical home is a big issue for us right now. These are very important to us. So, I think I'm about at the limits of what my time is expected to be for a general talk. I look forward to returning in the next session for more specifics about what I've talked about so far. And I'm open to any questions at your discretion.
DR. STEINWACHS: John, thank you very much. Why don't we go through the other two panelists and then come back for some questions and do it that way. So, Dr. Pawlson, Greg?
DR. PAWLSON: John is on the board of NCQA, and I didn't want him to have to do my slides for me. That was not an appropriate thing. We will attest that I sort of blanched when you said that because -–
DR. TOOKER: We can see that my main roots are not in the certification/qualification world.
DR. PAWLSON: Well, it gets more confusing because in general, the term recognition of qualification is sort of a relatively loose term of just how do you know the characteristics of something. Whereas, certification has generally had this sort of -- although we use the term on the health plan and disease management side sometimes -- as well as then there's accreditation, which is of facilities and so on. So, it doesn't apply to individuals. So, it's easy to confuse.
What I'm going to just touch on are a couple of concepts that are behind the medical home. First and foremost is that I think there's a strong belief, and this is a model that John Berwick(?) and others have used in the past that professionalism alone, or the marketplace alone, is not really sufficient for healthcare and that what we are stuck with right now is great concerns about both quality and cost. We can't pick up any newspaper or any scientific journal or anything else without having that sort of slap you in the face. And I think it is particularly an international comparisons as well as comparisons within the United States, the variation is much, much greater than any of us are at all comfortable with.
So, this is sort of the pathway, and the medical home in a sense is a pathway that is in the measurement area by being able to measure and look at whether or not practices are functioning in medical homes, we would hope that this information would primarily flow to the provider side and engage them in quality improvement, and it's really important to remember that, that this is first and foremost an attempt to raise all votes. Not to just sort of mark some practices as medical homes and other as never being able to achieve that.
I love this sort of -– that we have a medical sector or healthcare sector. Mike Levitt, the Secretary of HHS, has used that to denote the fact that we really don't have a healthcare systems, and it really is -– the only thing we really do in common is mostly because I think it's CMS imposing some order on this order, as we do have common voting systems.
It's possible to get a bill on just about anything somewhere in the healthcare system. And we really do have a system that I think is sort of the patient family is sort of off to the side, and it's the building system that we are focused on. And one of the real appeals to me as a – my background is in internal medicine, geriatrics and primary care –- is that the patient-centered medical home really is a core piece of trying to create a system of healthcare that is around integration by information systems and organizations. And I think it goes well beyond the early concepts of primary care in that it really builds on, as John has alluded to, the chronic care model, which was all about systems and about information and about organization.
And it also builds on I think, a growing research base that the Picker Institute and others have created around patient-centeredness. We do not have a patient-centered system at this point. Nor, do we have a patient-focused system.
By patient-centered, I'm really referring to what happens in the clinical arena between a patient and a clinician regardless of what kind of clinician. And by patient-focused, I mean what happens in the rest of the system. So, why don't our medical records follow us around, rather than following the billing system around? It's a very interesting sort of concept I think. So, it goes well beyond just the clinician interaction with a patient released as why can't we have a healthcare system that is person focused.
And that really says that the primary care patient-centered medical home is really an important part, but so is coordination by hospitals and by implication, the others laboratories, pharmacy and so on. And also, we've talked about transformation that needs to occur in subspecialty practices as well because many elements of the chronic care model, for example, not primary care, but of the chronic care model, apply to specialty-based practices.
There is no reason why they cannot have information systems. Information exchange, clinical decision support, patient self-management support for patients that are predominantly care for by subspecialties. And this is the last I'm going to talk about in this particular section. And it is really to illustrate, and there is a few other things that I will talk about later on.
There's the original concept of the medical home, which really is the sort of initial formulation for special needs kids that did not really necessarily include all the elements of the chronic care model, nor of primary care particularly. But, this just shows that patient-centered medical home is an amalgam of primary care of Wagner Chronic Care Model and of research on patient-centeredness that really tries to pull all these things together in a common framework around creating this entity that could be the core of the healthcare system, but not all of it obviously, that will allow patients to receive predominantly primary care that also has all of the features that the chronic care model is actually patient- centered.
Now, the devil is in the details of teasing all this stuff out and theorizing out how to really measure it in an accurate, reliable, and not impossible manner, and then in applying the best tools that we have to find out does this really –- do high functioning medical homes really have a positive impact on both quality and cost and certainly on patient experiences and charisma. So, that's where I'll stop.
DR. STEINWACHS: Greg, thank you. Next, we have Ms. Partridge, Lee.
MS. PARTRIDGE: Thank you. I'm please to be here to talk a little bit about the patient-centered medical home from the perspective of the patient and the consumer. Dr. Tooker and Greg have both mentioned, none of us are terribly happy about the quality of care that is available in this country. And the patient-centered medical home concept, especially with this emphasis on readily accessible comprehensive coordinated care and active involvement of patients and families shows real promise of being able to help us improve that system.
And therefore, the partnership has been active for several years and the PCPCC – I can never get all the Ps and Cs straight. We also decided to start talking with some of our consumer advocates in the field all across the country about what they understood the word patient-centered meant.
And from that discussion, which took place about six weeks ago, we have started to develop a set of what we call consumer principles for how we would assess a medical home, partly because there are so many different definitions out there. Greg and Dr. Tooker have been talking about a concept that is a quite elegantly developed concept.
We have concepts floating around across the country that are much simpler. In the interest of full disclosure, I should mention I'm a former Medicaid director and I follow Medicaid very closely across the country. We thought we were establishing medical homes with the PCCM systems 20 years ago, and indeed some of them I think would come very close to meeting that definition today.
I can think of some really fine primary care practices in FH, federally qualified health centers that would probably meet even the complete definition the NCQA is talking about. Anyway, as a result of that conversation, we have kind of a tentative guess of what will be on our consumer principles checklist. And let me go over them briefly.
First, it would include access to a clinician 24/7. I think that's an element in most of the PCCM programs. That certainly is a requirement in the Illinois medical home model that was launched for their Medicaid program last year.
The second is what we call knowing about me, which means the primary care provider should know my medical history, my treatment preferences, my family situation and my support structure. We believe that all four of those are key to helping make proper decisions or treatment course for that patient. If you don't like to take pills, you are probably not going to be terribly interested in the treatment program that relies heavily on medication.
Third, and this comes up in all kinds of other situations as well, treatment with dignity and respect not only from the clinician, but from all the staff. Some of the work that's been done – I think Greg can probably confirm this –- on HMOs is that when you survey the patients, they will say oh I love my HMO doctor, but I hate the plan. And it really means they don't like the front office helping me to see a specialist when needed and coordinating my care with that specialist.
And finally, having public information available that allows me to see how well the practice performs, not only on selected clinical measures, but also patient satisfaction with that particular practice. One other element that may be on our final list -– I think we think it should be on our final list – is the one that talks about helping patients understand enough about their medical condition to be active partners in their clinical care.
Dr. Lynn Mitchell, who is presently the Director of the Oklahoma Medicaid Program is fond of saying that an informed patient is a better patient. Patients and families should not just be passive followers of a clinician's advice, but should know enough about their condition and treatment options to participate actively. The self-management concept that Greg had up on his slides a couple of minutes ago. And this means that clinicians and Allied Health professionals as well as patients may have to learn new techniques to be able to participate fully.
And I was interested to discover last week that in the Vermont Blueprint of Vermont, which is a medical home concept, they have done a fair amount of work at the community level encouraging people to learn how to be good self-managers. I would note that consumer would be weary of any new model that sounds like the latest version of cross-cutting or gate keeping. They immediately translate that into denial of necessary care or, at a minimum increasing the hassle factor. It doesn't mean that we will be necessarily hostile, but if the argument in favor of adopting a patient-centered medical home is to save money, consumers really won't be motivated to choose it.
As we are going to talk more today, we will talk more, I'm sure, about the diversity of models that are out there. They focus on different segments of the population and use different incentives for provide participation and have differing structures for implementation. One may emphasize patient education. Another will emphasize practice redesign.
So, from our perspective, there is a tremendous learning opportunity opening up before us. And that is one reason why in particular, we welcome this committee's interest in the concept. You have at your command here tremendous research tools and I think we need to learn to use them to evaluate what seems to work as this model rolls out, concept rolls out, vision rolls out across the country.
DR. STEINWACHS: Lee, thank you very much. Let's open it up to question or comments. Well, let me start off then. To me, one of the big challenges has been health literacy for all of us. And Lee, you were beginning to talk about it at sort of a level higher than when we usually talk about health literacy, when the doctor talks to me I can understand the words that translates. That's very different it seems to me than in a sense feeling empowered by understanding enough about your care and expectations to have a dialogue with your clinician about choices and pros and cons. Are there any thoughts about how we get there since you had me stimulated?
MS. PARTRIDGE: Christine Bethell, down to my right, is a much better source.
DR. STEINWACHS: So, I should hold it until she speaks then?
MS. PARTRIDGE: I will say that the partnership that we have in looking a little bit at some of the measures of patient activation and some of the work that has been done by Judy Hibbert, for example, and some of the work that has been done by Kate Loring at Stanford. I believe actually its Dr. Loring's model that is being implemented in the Blueprint on Health Structure in Vermont. I think that piece is in its infancy or maybe kindergarten. Greg?
DR. PAWLSON: One of the real struggles, and I think we all would reflect this in sort of defining the current patient-centered medical home model is how much of it is aspirational and how far do you push when you get into certain things that virtually no practices, a couple around the country are doing, versus things that are sort of more foundational that many practices still aren't doing well. So, that's going to be a constant struggle and I hope that there will be some places that will sort of test more advanced concepts, if, and I think really only if, they have some of the other stuff down.
Including, there's different area of the country that are -- like Minnesota, the average practice there, at least on our total scores, in the 50s or 60s. Whereas, we suspect as we've seen in Mississippi or Alabama probably score in the 20s or less.
So, I think there is a lot of room for experimentation, but a lot of these concepts are not horribly well tested and especially in terms of their impact on quality and cost however we measure that.
DR. TOOKER: We have to decide what we mean by health literacy. To some patients, and it's at the level that is so basic that they cannot read, but if you think about what Lee has said about dissatisfaction with the individual physician, but not with the HMO, what typically happens when you are a new patient in a practice?
The first thing that happens is someone hands you a clipboard and asks you to go sit down and write out three pages, double sided, and don't talk to me until you've got that done. And there isn't any way of knowing yet, with the initial exposure to practice, whether that patient is literate with respect to their health needs or not.
Two, they may read very well and that their level of understanding in lots of things is normal, but in terms of health literacy, they find it overwhelming and very confusing. My point is that in order to measure it, we have to develop better tools to assess health literacy of patients and allow them to participate in something like the patient-centered medical home.
In addition to the need to develop tools, we don't know very much about how to interact with patients to be able to measure it because patients can find it very insulting if you translate health literacy into your dumb and that you need to be talked down to in terms of your care.
And you cannot activate a patient in a practice unless they feel, and it's really telling in Ed Wagner's graphic about the chronic care model, that the patient is on exactly the same plain as the practice team is now. That they are viewed as equals. So, how to change the relationship so patients feel that they are equal partners in their care. And while an informed patient is critically important, an activated patient is even more important.
As Lee mentioned, using tools such as June Hibbert has developed to actually measure and assess patient activism in their practice. And also, with respect to measurement, I think it's important for the panel to know that the patient-centered medical home is an accountable entity to all of the parties that the home relates to. And particularly in including patients, it must have tools in place to be able to demonstrate to patients that this is not a gatekeeper model, but rather this is a voluntary home in which they can make a decision whether or not they want to be part of that home or not.
DR. STEINWACH: Larry, you had a question or comment?
DR. GREEN: I do. I actually have two questions -one for Lee and one for John. Lee, if I heard you right, you indicated that you thought that normal people trying to get health care in this country would be wary of anything that sounded like gate keeping or had an attitude that the main reason to do this was to save money?
MS. PARTRIDGE: Right.
DR. GREEN: Can you just keep talking about that now? What is that all about? And I'll ask you to put it in a context that the country has to figure out some way to save some money.
MS. PARTRIDGE: Right. And I certainly think that consumers are equally concerned about holding costs down. I think that still some of the old HMO flavor carries over and there is concern - suppose I'm –- well, I can't use IBM because IBM is doing this and you're doing it well. But, suppose I'm a major employer and I'm trying to encourage you to choose a patient-centered medical home practices as opposed to others? I don't think that the message to your employees can be that the primary reason that we want you to do this is because it's going to hold, even your part of the premium down - although that might be a somewhat effective incentive.
I think it has to be explained to them in a much more positive way. Or else, if you are talking about a voluntary system, you're not talking about a mandatory gate keeping system, and some of the Medicaid programs are mandatory gate keeping systems, but, if you're talking about a non-gatekeeper system, if you're talking about a voluntary system, that can't be your central message.
DR. GREEN: My question for John was a political one. Could you share with us what in the world happened a few years ago when you got the pediatricians, the internists, the family physicians, the osteopaths all to agree about something at the same time?
DR. TOOKER: Well, interestingly, Larry, the pediatricians, family physicians, and general internists, usually agree. But, we have been thrown, particularly family medicine and general internal medicine, into this same very difficult pot right now.
And that is what we referred to in a report to Congress of the collapse of primary care. Primary care is collapsing in this country. And we are trying to ring that bell very loud that we are at risk now of losing primary care in this country. And I think the bell has been heard. MEDPAC has said, in Congress last week staff and some of the members said that we are very worried about primary care at the level of the Senate Finance Committee, and as I said, at MEDPAC.
So, Larry, I think that a galvanizing issue and it's certainly highlighted in the future of family medicine, a project as well as our own paper was that we are very worried. Not just for the profession, but particularly for patients who said that they highly valued having their own primary care physician now. We must change the financing system and the payment system in this country if we are going to be able to provide a primary care physician for patients.
And this model is a physician-led model, but I wanted to emphasize that this is a team-based model that fully incorporates to the level of their training and licensure many other professionals. I also wanted to briefly mention that cost efficiency is a very important part of this concept.
It's not meant to limit the expenditures of payers primarily, but rather for the patient's benefit because patients now are increasingly paying a larger part of their healthcare out of their own pockets now that we have to understand both clinical effectiveness and cost effectiveness of care. And the ACP just released yesterday, published a paper on the need for cost effective care as well as for clinically effective care.
As Lee said, the primary reason cannot be to restrict services or to lower payment, but we know that there are many services today provided that are not necessary. And if you look at the relationship across steps just under Medicare, the amount of money paid per beneficiary from one state to the next varies widely, without any improvement in the quality of care.
So, I think in order to have an activated patient, I think that they have to understand what the cost implications of their care are, as well as the clinical implications. And again, it's not to restrict service, but rather to have an informed patient. They need to know what the consequences are of the cost of their care, as well as the clinical effectiveness.
DR. CARR: Actually, Matt just shared a note with me, and it kind of underscores what I was going to ask as well. You touched a little bit about what it means to testify in front of the NCVHS, and I want to just say a few words about what we saw as the role of this, and also hear from you because there is so much going on nationally, but what are the areas that we could uniquely help with. But, let me just reiterate what Matt has here.
The goals that we identified at the outset were four. One was to bring together stakeholders for shared concepts and definitions. The second was to understand the measures and the dataflow, the standards, and existence, performance and value. And the third is use of datasets for measuring patient-centered medical home, and fourth, means to incorporate the patient-centered medical home requirements into health IT systems.
So, those are things that we thought about, but I think it would be very important to hear from you how should we –- is there something else we should be thinking about, or a way that we should be listening to what you're telling us so that we're certain to get the right message?
DR. TOOKER: If you can do all of what you just said.
MR. BREEN: Can I just ask one question? One thing I didn't really hear much about. You said this would be a voluntary system and I know Lee Partridge, you said that you used to work for Medicaid. So, you know that there are some systems out there. Often times, they are badly under funded or they are set up in a way where. Well, sometimes, even the voluntaryness of it undermines cost-effectiveness of the running of the thing.
So, I'm kind of surprised that you're not talking about funding mechanisms because too me –- well, I'm an economist -- but to me, that's something that really should be considered at the same time as you're thinking about this that maybe we don't want a voluntary system. Maybe we want a real system because people have talked about risk pooling and that voluntary systems tend to undermine risk pooling. They can lead to the kinds of inefficiencies that we have in our system where specialist do an enormous amount of work that generalist could be doing and probably should be doing.
And then a lot of people don't get care at all. And then there's quite a lot of over care. I work at the National Cancer Institute and we're starting to question whether all the screening tests need to be done as frequently as they are. So, there's a lot of inefficiency shall we say within a system that can be created by a voluntary system, which maybe could be smoothed out through thinking about the financial mechanisms as well. So, I wonder if you've been thinking about those and if you're thinking if you put a good system together and some economist will come and help with financing part or how that fits in?
DR. TOOKER: Those are terrific questions, and yes, we have been thinking about them a lot. We are somewhat constrained now by the fee for service model that we have, both in the private sector and the public sector. That is to say that and I want emphasize that for family and internal medicine, we are talking about where most of the care in this country is delivered and these are in small ambulatory practices.
At ACP, roughly 50 percent of our practices are four or less and about 20 percent of our members in practice are still in solo practice. So, we have a huge number of very small entities out there providing the bulk of ambulatory care in this country.
And these are small businesses. So, they have learned to survive by developing their workflow relative to the payment system that they have. And that payment system is that you get paid primarily for the number of face to face visits that come through your office just now.
In order for the patients under medical home model to work, that has to change. While we think that when patients need to be seen at the present time that a fee for service should continue, but when we are faced with the burden of chronic illness and how to manage it, these patients need to be managed in between their office visits. And I would say that when I was in practice that if I saw a patient, if I said, I'll see you in three months, I wouldn't have a clue whether or not I really needed to see this patient in three months or not. But, I did know that that patient with chronic lung disease needed to be managed in-between these visits.
So, what we ware saying is that there still needs to be an additional payment. And that payment is per member per month the care coordination payment. And it's tricky now trying to look at this balance now. Is that a fully capitated payment, which these small practices are going to take risks, or is it not? And at the present time, it is not. But, there is going to be some pressure for practices to be efficient, but we are not advocating capitation at this point, but rather a per member, per month payment to manage the coordination of these patients.
Now, how much does it cost to coordinate the care of a patient? Nobody knows. And for that reason, we have engaged the Urban Institute with a grant from the Commonwealth Fund to do a costing study to see what it really means when we say that we are going to pay for care coordination now.
And as an economist, you will quickly see that this is fairly complex in that for the current business model, for those patients that you would normally see and get paid a fee for service, if you're not seeing them, and then you have to make up that reimbursement by the care coordination now.
And the practice will need to increasingly be able to get information outside of face-to-face visits in order to manage patients. So, that means understanding what remote monitoring is for blood pressure, for blood sugar, that this information comes into the practice and allows for patients to be managed outside of the practice. How do you pay for remote monitoring security mail messaging? And we're beginning to understand that now. So, it's another form of payment.
And there may be payment associated with performance. These are accountable entities in which the practices will be measured. And all of us are heavily engaged from measured development at NCQA and the PCPI through the endorsement measures of the National Quality Forum to the implementation measures through the AQA, HQA and the Quality Alliance Steering Committee, and I won't bore you with all of that except to say that there is a tremendous amount of work going on now to develop increasingly complex measures to understand how that system of care is being developed. And those measures need to be implemented in practice with the information systems available to be able to obtain the data to satisfy those performance measures.
That also includes care coordination, and I think we will get to the point where we get into cost efficiency of a practice now. And what is tricky and we don't know enough about yet is that if you save money, both on the ambulatory side by fewer non-evidenced based referrals to understand now that patients can go wherever they want. We know that there is a big difference in utilization services between South Dakota and Florida in terms of how services are being accessed now.
We can reduce the number of non-evidenced based referrals. Now, by having informed activated patients know that it is not in their best interest to go seek a bunch of referrals. They are not coordinated with their medical home. How do we recognize those savings and benefit from them. And we are looking at shared savings models now between the patients at a medical home and the referral base in which there is a cooperative relationship between the patients at a medical home. Probably though, in response to your question, a bigger source of savings, but I'm out on limb here, is the savings between Part A and Part B money.
For these patients who have carefully coordinated care, and I'm a pulmonologist, so if I have patients with asthma, for example, these patients should not end up in emergency rooms if there care is well taken care of in an ambulatory basis.
So, if they don't end up in emergency rooms and they are Part A savings, how do those savings accrue to the people who are doing the work to keep these patients out of these Part A environments now? And of course the same thing applies to hospitalizations now. So, we have to have a larger national discussion about saving money and the aggregate. It's compounded even further though because you want patients who are not insured to all be enrolled in patient-centered medical homes and the most complex patients need to be really enrolled in patient-centered medical homes and we can't have mechanisms where practices don't want them because they are too hard to take care of now.
You've got to incentivize these practices to be able to care for the most complex patients in these medical homes. And to their credit, the American Hospital Association has had a recent large initiative in which they have talked about the relationship between Part A and Part B and have talked particularly about bundling of payment and have recognized the importance of the patient-centered medical home. So, I'm sure we could have several days of conversation about your question, but I just wanted to let you know at a very high level that we understand the issues and are trying to understand some of the answers to those.
MS. PARTRIDGE: Can I just add a postscript to that? Another thing we might look at as a model is the new special needs plans in Medicare. Because, in addition to the problems of Part A and Part B, with two separate sources of funding, you have Medicare and Medicaid, with two very different sources of funding.
Dr. Tooker and I have talked about this a little bit. The incentives are all wrong. As a Medicaid Director, I am not going to save any money by fewer hospitalizations. My money is on the outpatient side. Ditto, you don't care about what I do in community based care.
These programs are so new, but I think maybe in a couple of years we could – maybe a year – we should be able to dig into the track record of some of them and see what the cost implications have been.
DR. STEINWACHS: Well, thank you very much. This has been a great kickoff to think about the vision of what the patient-centered medical home and the roles that it will play. We were going to turn next to talking about who are the key actors in the patient-centered medical home and stakeholders that are here.
So, we are fortunate to have Mr. Jim Coan from CMS, who is in the Office of Research, Development Information in the Demonstrations Program talk to us from the CMS perspective. And Mr. Chris Nohrden from IBM, and you've already gotten very positive feedback and that's good, isn't it, about the employer side of the stakeholder picture here, which is a very important part in the American healthcare system.
We had hoped to have Susan Edgman Levitan with us, but her health didn't permit. So, Lee, if you end up having some additional comments about stakeholders, we'll bring you back in. And again, we have the pleasure of John Tooker to talk about stakeholders from his perspective. Jim, do you want to lead off?
MR. COAN: Thank you and thanks for having me today. I appreciate it. From a CMS point of view, Medicare demonstration is a research project. Mimicking what has been said a little bit earlier, there is still a lot that isn't known about medical home and how medical homes work. And whether or not there is any benefit to any particular kinds of medical homes. There's a lot of good work going on out there, including the work for IBM. I want to jump on that bandwagon too please.
There is a lot of work going on. And it's very good and it's very informative. But, as far as actors and stakeholders are concerned, as the days go by, we find that we have more and more actors and stakeholders in the area of patient-centered medical homes, and it's for good reason too. The thing to keep in mind with actors and stakeholders is although we might agree in principle on the basics of what medical home care ought to look like or what the medical home ought to be, depending on who you are, it's how you apply the medical home that is going to be the difference. And there are various needs. We've heard a few of them already up to this point.
For example, we hear that there is a clinical quality need for example. It's very important. Clinical quality is very important in the establishment of medical home and medical home type care. They also have a stake in payment, which was also brought out. So, there's a clinical quality principle that has a stake in the payment side of things.
Now, from the CMS point of view, we happen to be a payer who has a stake in the quality aspect of it. So, it's a little bit different. We have to try and find some more of the middle ground. All of the actors, all of the players have to try to establish what exactly we're talking about and how it ought to be compensated.
Depending on what you need, depending on what you're particular agenda item might be, you have a tendency to look at medical home in terms of how can it satisfy that need. We can't prioritize all the needs. We can't put all the needs in some kind of a chart and determine where the priority needs actually are because we'll argue endlessly.
We recently, because of legislation that we got -–- I'll go into that in just a few minutes by the way -– we recently engaged the Relative Value Update Committee from the American Medical Association to establish the relative value units to price the medical home monthly fee for medical demonstration. There were 12 individuals on a workgroup that created recommendations that the overall ruck agreed on just at the end of last month.
And we had several weeks of telephone calls. Twelve individual, 12 very different points of view. Twelve individual stakeholders all deciding on one thing. They pulled it off. They did an excellent job. I was remarkable. But, there was a lot of push and pull and give and take simply to establish what the rules were and what the needs were. Are my needs being met in this group and so on? I thought it was important to start out talking about stakeholders and actors and giving a perspective in terms of the fact that there is going to be certain needs that are not necessarily identical up and down the line and there's going to have to be some understanding that the first thing we need to do, we have to build the house before we furnish it.
So, let's figure out what it is medical home does and doesn't do. And then figure out how we can make it work in a more universal sense. Quickly, I'll go through some slides. It will describe for you what Section 204 of the Tax Relieve and Health Care Act of 2006 tells Medicare, tells CMS what it is we have to do.
We're directed to redesign a healthcare delivery system, providing targeted, accessible continuous coordinated and family center care. Provide care management fees, and provide incentive payments, all directed at a medical home. Medical home is intended in the demonstration to be focused on high need populations, which is described as individuals with multiple chronic illnesses that require regular medical monitoring, advising and treatment.
This is what the legislation tells us. We don't want to build a chronic disease medical home. We want to build a medical home that's good for the sick, that's good for the ill and the seriously chronic ill. We want this to be general fully functional medical home that would apply to anybody. But, our focus for the purposes of the demonstration is that these people be diagnosed with chronic illnesses.
We have taken an interpretation because of certain ambiguities within the language of the Act to make the indivdiual's Medicare beneficiaries who have at least one chronic illness that's determined from a list of chronic illnesses that we are working on right now. That is to make sure that it is important. For example, many Medicare beneficiaries might have essential hypertension and hyperlipidemia, neither of which would lend themselves to very much improvement in medical home type care.
We're really looking for the medical home to be applied to the kinds of chronic illnesses that lend themselves much better to the care coordination. We're not saying anything is less important or more important, but merely for the focus of the demonstration, which I emphasize again is a research project, we want to see how well it works for people who are likely to consume those services in larger quantities. And also consume a greater portion of the healthcare dollar.
Ninety populations, 83 percent we estimate of Medicare beneficiaries have at least one chronic condition. We know now from the list that we aware working on, which is predecisional list, we can't tell you that this is the way it's going to be because we're still waiting on approvals and other such things. But, with our list, 86 percent of Medicare beneficiaries who participate in medical home will be eligible because of the chronic conditions. They will have at least one chronic condition.
We also know that 23 percent is similar to what Dr. Tooker mentioned earlier. Roughly 23 percent have five or more chronic diseases. That's pretty important. Break that down a little bit. They see 13 physicians on an average per year. They fill an average of 50 prescriptions. They are most likely to have preventable hospitalizations, which is an important factor to apair. And also, they have the highest out-of-pocket cost.
Our demonstration is intended to run for a three year period. We'll do it in no more than eight states, which have not been chosen. The demonstration will have to apply to practices that have three or fewer full-time equivalent physicians. And we're emphasizing physicians in large practices as well, particularly, those in rural and underserved areas.
To be a medical home, practices could also be specialties or subspecialties if they are providing a significant amount of primary care to the patient. By definition, we're pretty much tracking along the lines of the joint principles that AAFPACP/AONAP are using because, let's face it, they're right. There's no sense in arguing that.
It will be a physician based practice providing safe and secure technology, promoting access to personal health information, using health assessment tool and so on. This is just very brief, but I think we would probably tell you to look to the joint principles if you wanted know what kind of medical homes we were envisioning.
To qualify as a medical home, it's a voluntary procedure. We're not going to go out and establish medical homes by force. We're going to invite practices in all of the eight areas that we choose to become medical homes. We will use a process very similar to the Physician Practice Connection and patient centered medical home that NCQA has developed.
Once again, it's one of those things where it already exists and there's really nothing that we could do to it to make it improved. So, what we've done is we're working through our contractors to develop a modified version of that that will apply to this demonstration only. And it will not have an affect on NCQA's PPCPCMH tool. It will just very closely resemble it.
So, all the practices will have to meet those standards. And they'll have to agree to maintain or remain in the medical home demonstration. We will pay a monthly fee for the personal physician in the medical home. That would be a PMPM. A per-member-per month monthly fee for each medical beneficiary and a demonstration so long as they maintain their eligibility.
Their expectations are improved coordination that will lead to improved outcomes and improved outcomes that will lead to lower costs. Cost savings could be used for additional incentives. The legislation tells us to do that. And basically, the way we're looking at it is when the patient does better, the practice does better.
So, that's from a stakeholder point of view who happens to be a payer who also has a stake in quality. That's an overview of what the demonstration is probably going to look like. There's no details here obviously became we were in a pre-decisional developmental portion of our demonstration design. So, I can't go into a great deal of detail. And I'll leave it at that. If there are questions later, I'll be happy to answer.
DR. STEINWACHS: Jim, thank you very much. Why don't we go on to Chris? You want to talk about the employer perspective?
MR. NOHRDEN: Thank you. I'm the project manager for IBM's patient center primary care initiative. And I'd like to thank you very much for this opportunity to address you today. The perspective I do want to share with you is that of a large employer, but also a significant buyer of healthcare in the U.S.
We spend currently close to $2 billion annually to provide healthcare insurance for about 600,000 lives across the country. And we've had a continuing emphasis on quality, cost, and access to care for a number of years. And that's helped us maintain our health insurance premiums among some of the lowest in our industry.
We've also had a special emphasis on wellness and wellbeing programs and health promotion, which have similarly reaped significant benefits and savings to a tune of about $175 million worth of savings annually to the IBM Corporation.
But, as a large self-insured employer, our health benefits costs are basically costs of doing business. And we require the steady focus on our cost structures to make sure that we can remain competitive in our industry. But, quite frankly, we're really not very happy with the quality and the cost of care that is currently available on our fragmented healthcare system.
The statistics about healthcare quality are particularly disturbing. If you consider the estimate of Dr. Elizabeth McGlynn and others in her 2003 New England Journal of Medicine article that adult patients only receive the proper care for chronic conditions about 50 percent of the time. Our average employee age is approaching 42, and many or them are entering their mid-life decades when they may begin to experience multiple chronic conditions. Our feeling is that these chronic conditions would be much better dealt with in the context of a trusted primary are relationship with a primary care provider. Rather than the later stages of development as is currently the case with our third party disease management identity.
This is not one of the prettiest examples, but we currently have the reimbursement systems in place that will reimburse specialists $20,000 for a diabetic leg amputation, but we don't have the mechanisms in place to pay a primary care provider for the process and procedures and the infrastructure that will allow them to use patient registries and clinical decisions support tools and other types of IT tools at the point of care to get things like annual foot exams and other recommend treatments that will prevent these problems from occurring in the first place.
About a year ago, we began a dialogue with our primary care providers to understand how we could change the paradigm for the type of health care and the quality and the cost factors in particular and discover the concepts of the patient center medical home that was postulated and deposited by the American Academy of Pediatrics and which has now been endorsed by the other three primary care professional associations.
As you're going to hear more I think and through the testimony a patient-centered medical home for primary care requires a long-term relationship. That's one of the primary things that we're looking for, with a trusted primary care provider that's enabled by technology to provide much more meaningful preventative care and wellness coaching.
And this may include elements such as email consultations, electronic health records and clinical decisions support to all sort of link with evidence-based guidelines of care. And currently, outside a number of our health maintenance organizations, we can't buy this type of primary care in the United States. But remarkably, it's a very successful model for the delivery of primary care in a number of other countries like Denmark, New Zealand and Singapore. And I'll speak a little bit about that tomorrow in the context of Denmark.
So, in 2007, we brought together the primary care professional associations, other leading large employers, and the payers to create what John Tooker has already alluded to the Patient Center Primary Care Collaborative. It's a mouthful. We don't have a clever acronym for it, and we're struggling. It is certainly a multi-stakeholder collaboration. It's based here in Washington D.C. It's hosted under the umbrella of the ERISA Industry Committee. And it really provides the form for the policy discussions that we need to have about how we are going to obtain and pay for this better model of care.
The collaborative now represents about 50 million American lives, and it's been signed on to by upwards of 100 leading employers, including those such as Boeing, Caterpillar, Fed Ex, General Motors, among others. Of course, our professional associations, professional provider associations and seven of our national health plans. We had a national summit last October here in Washington. We are honored to have former speaker of the house, Newt Gingrich, speak to us. And one of the most pithy quotes that I took back from that is that technology, prevention and consumer-centered care are the powerful forces necessary to transform healthcare into a 21st century intelligent health system. It's absolutely imperative that we embrace this new model of care which is centered upon the individual in order to save lives and save money for all Americans.
So, I believe that the patient-centered primary care collaborative really sets the stage front and center and models the way for the public policy discussions in particular and the decisions that are needed for widespread adoption of the patient-centered medical home. Central to the patient-centered medical home is the premise that this type of care requires a fundamental shift between our patients and their primary care physicians.
There needs to be a higher level of personalized care coordination, which is beyond acute care episodes and the identification of the key medical and community resources that would meet each patient's needs. The widespread adoption of the information technology for care management and quality improvement, along with adequate payment methods is essential.
So, large employers have been successful in challenging health policy conventions in the past. A couple of key examples that come to mind are Leap Strong and Bridges to Excellence, where these were really grassroots uprisings that were led by employers and have really gotten a lot of traction in the US as far as healthcare reform.
I guess to kind of bring things back full circle and summarize these points, IBM is a large employer. It's a major purchaser of healthcare in the U.S. and we spend quite a bit of money to provide healthcare choices for our employees and their families. We are not satisfied with the current focus of our healthcare system that is on episodic acute care instead of comprehensive coordinated healthcare promotion and disease prevention. And we're also dissatisfied with the quality of care that is currently available.
The patient-centered medical home is a healthcare model that we really would like to buy for our employees to improve the aspects of cost, quality, and access. We're helping to make a difference in the delivery of healthcare by advancing the need for greater adoption and implementation of the patient-centered medical home.
And there is certainly great opportunity for other large employers and stakeholders to come to the table under the umbrella of the Patient-Centered Primary Care Collaborative to make the patient-centered medical home a widespread reality. So, thank you very much.
DR. STEINWACHS: Chris, thank you. I want to apologize. I forgot to introduce you. You introduced yourself, Mr. Chris Nohrden. I appreciate that. The third on the list is John to come back in another role here.
DR. TOOKER: You've already heard a lot and there will be some redundancy here. I'll move through these slides so that we have time for questions. James and Chris have keyed this up very well. And I've heard Susan many times and she's a very articulate spokesperson for the team.
So, the question we have all been asking ourselves is there a better way to organize, coordinate and finance primary and principle care that would lead to better outcomes, more cost and increased interest in primary care. That has been the policy driver at ACP and I think it is fair to say for all of us now.
I want to re-emphasize again in the cases of the key medical home actors that there are strong pieces of primary care model that have been incorporated into the patient-centered medical home. And I really to give credit to the primary care, the Institute of Medicine, and in particular to Barbara Starfield, who I understand will be one of your discussants tomorrow.
Barbara is legendary in primary care, and she has been extremely useful in the sense of can we, at this point really measure within the patient-centered medical home whether all of the tributes of primary care are being fulfilled in that home.
And lots of conversations, I want to emphasize for the panel today, that we are at the patient-centered medical home version one, and this will evolve over time. And it won't be until we have been able to test it in models such as James mentioned about a research model that we'll be able to refine it. But, I think it's key to remember all of these three models of care are now being incorporated.
So, in terms of the key patient-centered medical home actors, I think it's important to step back and think about that we have to start from the pipeline of the beginning medical student all the way through a point where mature physicians are retiring from practice and try to figure out how we are going to be sure that we are addressing in primary care each one of those pieces now.
I think it's really important for those who are particularly concerned about primary care to know that there has been a remarkable fall off and interest in primary care in the last six or seven years. And I mean remarkable. When I talked about the collapse of primary care, in lots of medical centers now, medical schools, less than five percent in some cases, none, of graduating medical students are going into primary care, other than pediatrics now.
The challenges of course are that it takes a long time to restore that pipeline to the point where you have a predictable throughput through the system to be able to provide the workforce that we need. In terms of key actors and training for medical and graduate medical education, they are the AAMC, the Association of American Medical Centers. And a year ago, we petitioned the AAMC to take a position on the patient-centered medical home. And to their credit, they have.
In April, they published a position paper on the patient-centered medical home, which has some key features that are appealing to us. One is that they have agreed that there is imperative to train and incentivize physicians to enter primary care and to train them appropriately.
And this is a challenge in academic medical centers because of this tension between the provision under our current payment system of high volume very expensive services, which are reimbursed much much higher than primary care is and to incentivize so that you are training primary care physicians.
There's the Society of General Internal Medicine, there are the academic societies within family medicine and pediatrics, and the American Osteopathic Association, and one more that is important here –- I just met with them two days ago after this slide was produced –- is the Association of Academic Health Centers.
They are taking a longer position, not just in physician manpower or person power I should say, workforce, but rather, they are taking a point of view of a soon to be released position paper on other medical providers, including those who are part of the team of patient-centered medical care.
Now, you are going to hear a lot more from Terry and others about CME and practice transformation, but it's the professional medical societies that historically have the role of continuing medical education, much more so than academic medical centers now. And in this case, we are not talking so much about traditional CME that is the medical education around the improvement of clinical care, but in practice transformation.
How do you change your practice so that it is able to perform as a patient-centered medical home? And you'll hear much more about that. Collaborative, you've already heard about. And there are smaller entities as well. It's important to know that we don't have the expertise to all of this work and have contracted with consultants -and I know that Mathmatica is here - Mathmatica and the Urban Institute, to help us understand what the patient-centered medical home means. That if we do a pilot of demonstration projects, these must be scientifically rigorous pilots so that we have objective third party information about these pilots that is useful, both in terms of scaling to larger entities, but also to be able to publish our work in peer reviewed publications so that it will stand the test.
And then there are foundations, and I'm only touching the surface here, but the Commonwealth Fund has been extremely helpful about obtaining data in support of the patient-centered medical home. Robert Wood Johnson has also been important. They funded the work of the McCall Institute to develop the chronic care model. And we've already heard from James and also from Lee, about the critical importance of the federal and state government. And I want to point out the leadership that the states, including Medicaid, have provided. In some states now, there is legislation requiring the implementation of a patient-centered medical home. Washington State is a recent example.
Some have gone ahead and implemented because they see the benefit of the model and are able to do so. Tomorrow, you're going to hear from Jim Walker at Geisinger and Geisinger is a remarkable health system in central Pennsylvania that has already implemented patient-centered medical home and have done so successfully.
There are always innovating smaller private practices that have done the same thing. We want to learn from them the challenges having data from these experiences that we can use to test the overall model. Testing is occurring in multiple pilots and demos. There are at least six relatively mature pilots and demos at the state level right now, and James has mentioned the Medicare demo, which we hope to go live – if I'm correct James – in early 2009?
MR. COAN: Yes. Cautiously.
DR. TOOKER: Thank you. I didn't mean to put you on the spot, but you've helped me a great deal. But, that's what the target would be and critically important as James said, was to figure out what the per member, per month is likely to be. The care coordination fee depending on the maturity of the medical home and the work of the rock(?) in doing so.
The policy implications are substantial and, as I indicated earlier, MEDPAC and Congress are now much more involved. And I would say that I have heard a “read my lips” all three presidential candidates mention the medical home now. I don't know if we're talking about exactly the same medical home, but I've been very less than careful to trumpet that I've heard it from all three of their lips now.
And then the role of policy in the professional medical societies. I want to point out in terms of the actors here, that this isn't all kumbaya. There are some real tensions in the medical community about the patient-centered medical home because there are parts of the medical community who really like the current payment model, and that is a volume based model, as Elliott Fisher said, we have to uncouple the volume incentive from payment.
But, they are definitely out there now and because they are out there, we are faced with the 10 percent reduction in Medicare payments as of July 1st because the volume targets have been exceeded now. So, those say parts of our profession who benefit from doing more. Then we have to figure out how to work with them in some sort of a shared model in which the incentives are aligned to improve quality and to reduce cost.
Internal medicine. I think I would be less than candid if I didn't say that those tensions exist within our own subspecialties in internal medicine. And then, what I think is really important and underappreciated is the role of state legislators. State legislators have been more assertive, in my opinion, then the Congress has been and moving faster because of their needs within their states to provide healthcare to their citizens.
And with Greg here, I'm not going to say more about the qualification of the medical home through NCQA. And you'll hear probably tomorrow from Barbara about how well the current PPCA model relative to patients at a medical home can qualify a medical home. Are we really measuring the right things? And that's an important discussion for us to have.
Let me move on quickly to the stakeholders now. And by stakeholders, I'm referring to those who are actively engaged in either providing care or receiving care within the medical home, or a relationship between the two. And I want to emphasize that we are talking about again the relationship with the primary care model to the chronic care model to the patient centered medical home. This is really about patients and families if we are talking about patient-centered care.
And I give great credit to the pediatricians for thinking about the role of families with regard to special needs children, but I also -- and I was smiling when Chris said that the average employee at IBM is approaching 42. That seems quite young to me. As the Medicare population ages, it's the rule of thumb that once a man and twice a child - meaning that the families increasingly become engaged as the family members age now. And that's very difficult to do, to engage the family around patients who are elderly as it takes a lot of work to manage those relationships now.
So, if a patient voluntarily chooses a patient-centered medical home, the patient is at the heart of that patient-centered medical home. They are on a par with the team that is providing care. And the team can and will be constructed relative to the local needs.
And one of the challenges that we have right now in talking about this at the national level, realizing that all care is implemented locally. And there are always a lot of local factors that relate to how care is practiced now. Here are a few of the people who are the professions who might be part of the team: Physicians, nurse practitioners, PAs, certified diabetes educators, licensed social workers, et cetera. And with these small practices, it's a challenge because obviously a small practice can employ and not employ all of the members of its team. So, we need to think about virtual relationships and networks and where the expertise of these professionals is shared among practices in a voluntary arrangement now.
As we mentioned, the role of the patient-centered medical home is not just to provide a primary care and chronic care model, but also to provide preventive care and acute care.
Now, one of the features of the Wagner Care Model is realization that medical care occurs in the context of the community in which patients live, again, the importance of the local relationships at the community level. So that these patients need to move to and from. And I thought James said it extremely well, that when you get to the Medicare populations, the number of interactions that they have with healthcare systems is enormous.
And we may think they all have the same relationships, but you have to stop and think, I do, that each patient has a different set of relationships within the community. The point being it becomes exponential in terms of the number of interactions going on within a community that have to be managed. So, these patients will have a specialty and subspecialty care and we have to manage this relationship and know that the arrows reflect a communication going both ways, and that is one of the things that are most missing now. That the information does not get in the hands of those you are referring patients to and does not get back and leads to a huge duplication of services because the information is not available at the point of care and these are very complex relationships.
My wife has had a set of serious illnesses, and when we get her hospital bill after one of these hospitalizations, it is completely indecipherable from the point of view of the patient. Fifty pages long, item by item by item and we have to change that so patients understand the care that they are getting and what they are being charged for that care and then this relationship to the community.
I'm oftentimes asked what you think about the personal health record. Well, it's a good idea in concept, but when you see the number of interactions with the community that a patient has over the course of the year, you realize how devisly difficult it is to get all that information in the right way within a patient-centered medical home personal health record so that the patients and the practice have the same understanding of what that information is.
When you think you have it right on one day, the next day it is changed again. So, I want us to be really thoughtful as you talk about the technology tomorrow about not just the technology within the patient-centered medical home, but how this relates to the information the patients have if they are going to be activated patients now.
Now, a lot of patients may have a predominant illness. And for those patients, they may choose a – and practices may choose – to have a principle care or a specialized medical home. And that is wonderful because patients should get care where they want it. But, if they view that their cardiologist who's managing their chronic congestive heart failure is the place where they want to get their medical care, as long as that practice is willing to provide the other services that the patient-centered medical home, they should benefit from the payment relative to the work that they do, including this important care coordination now.
So, this isn't all about primary care. And I want to be sure that we understand that there are going to be multiple models of the patient-centered medical home. Not just the primary care model. So, I will stop with that, and I, along with our colleagues, are open to any questions you might have.
DR. STEINWACHS: John, thank you very much. Lee, I'll just ask you if there is anything you wanted to say. Okay. Well, this has really been very useful in talking about the perspectives of payers, employers who have dual roles, and John, you have really walked us through very nicely the key actors and the stakeholders so I'd like to open it up to questions and comments.
DR. FRANCIS: This goes back to some of the points Lee made. First of all, I wasn't clear when you describe the demonstration project as voluntary, whether that was going to be voluntary on the part of the physician or also of the individual patients within the practice. So, that's the first question.
But then, related to that, what are the reasons that people didn't like gatekeepers was that they sometimes liked to shop around if they were unhappy with the care that they were getting. And not to have their first physician know everything that they'd done. I mean, I'm on the privacy subcommittee so this isn't entirely out of –
DR. STEINWACHS: It's good you're asking these questions, Leslie.
DR. FRANCIS: I haven't heard the word privacy and I read the documents that were sent out ahead of time and I didn't see the word privacy anywhere, but I'm just curious about the question of whether there has been any thought given to information flow from the perspective of the patients. So, if a patient with a medical home went somewhere else and didn't want the medical home to know about it for whatever reason, good or bad, and I'm not necessarily saying this is a good thing, how is that being conceptualized? Maybe this isn't the time to ask it, but that's—
DR. STEINWACHS: The first question, I think was to Jim about what voluntary meant, and then we can come to the security question.
MR. COAN: Participation in our demonstration will be voluntary at all levels. The physician practice will choose to participate or not because there is a certain investment in time and effort to become recognized to fulfill the qualifications. Patients will also be invited by the practice to participate. They do not have to and their care will not change as a result of their choice one way or another.
DR. BREEN: Well, won't they get the medical home. I mean, that's an intervention, isn't it? Medical home?
MR. COAN: Yes, only to those that would be eligible because of what the legislation tells us. In our case, it's a fee-for-service demonstration. So, it won't apply to managed care Medicare Advantage patients. The main reason for that is we don't have any data from Medicare Advantage that we could look at for claims purposes to determine whether or not there is any recurring savings as a result of doing the medical home. So, it's not that we don't like Medicare Advantage patients.
The other criteria come from the legislation about having a chronic illness. So, we have to apply that into the demonstration. Basically, a Medicare beneficiary who has both Part A and Part B, but does not have a chronic illness would not be eligible to participate in our demonstration.
DR. FRANCIS: I'm just curious about the thought of information flow.
DR. STEINWACHS: Lee, about the privacy?
MS. PARTRIDGE: You're right about the privacy issue. IT has come up in our early discussions. I think perhaps it might be a little less relevant to the Medicare community than it is to some of the younger people. Christine will probably touch on this or may think about this in connection with teens in particular. You may not want – it's pretty obvious, I think.
Where you strike this balance is very difficult. I think ideally, the medical home environment would be such that you would feel comfortable sharing everything. From the perspective of the clinician, that's certainly desirable. I don't know how you get there and people are certainly going to be very wary.
DR. PAWLSON: A couple of things. One is we're actually looking at trying to build in some communications with the patient about medical home and its properties from the practice itself as part of the basic sort of qualification medical home. Secondly, I think in thinking about this, patients would be – there would have to be some kind of opt-in, opt-out. I'm not talking about the Medicare demonstration because that's something that Jim and his group will be looking at how exactly that's going to be handled, but it is voluntary. It's not a gatekeeper model. You're not forced into it by your health plan. It's a free choice and hopefully, there will be sort of a contract between the practice and the patient saying that I acknowledge you as my medical home. I want you to have information to coordinate my care. But, the flow of information and the information system itself will still have all the safeguards or not that are inherently there. I don't' think that this will particular change it, other than hopefully illustrating to the person that chooses a particular practice as their medical home how important the communications are and if they choose to withhold it, what that means or doesn't mean in terms of their actually having a medical home.
If I don't have information on half the things going on with my patient, it's not going to be much of a home.
DR. STEINWACHS: I have Harry, then Larry, and Justine on the list.
DR. GREEN: I wanted to go back to the Medicare demonstrations with a question about both methods and metrics. What is your current thinking about what measures are that you are particularly interested in? Is that publicly listed and does everyone know what those are? Is that part of the details that we should wait for? And the other one is what is your thinking – well, let me frame the second question just a little bit.
I like research, okay. Guilty as charged. I like stronger methods versus weaker methods. But, I'm also interested in timeliness. And when I look at the needs for getting on with this, waiting for classical randomized control trials to get delimitative very strong methods answers to some of these things makes me think that by the time we get the answers, it will already be decided by other methods and other means.
So, are you thinking about moving to approaches to evaluate different strategies for the medical home, getting mix performance characteristics figured out through methods other than randomized trials, or is the plan that you can't randomize. It's not strong enough. We're not going to play forward. We're not going to do it. What's your thinking about both the metrics and the methods?
MR. COAN: Both good questions. I'm afraid I'm going to sound like I'm copping out on you a little bit. Your first question has a lot to do with how the evaluation of the demonstration will be designed, and it hasn't been designed yet. So, there are concepts at this point. I think it's safe to say though that Medicare is going to be looking very hard at where do the savings come from. The idea in this demonstration is that it be cost neutral. So, that whatever the fees that are paid for the medical home services, which are, incidentally, they're in addition to the usual ENM services that would be covered. So, a doctor could still submit a claim for service that they provide. Medical home fees would be in addition to that.
But, those fees should be offset by savings, probably at the Part A level - at the hospital emergency room or the avoidable hospitalization level. We will be looking at that. Other measures are being considered. They just haven't been brought in at this point.
In terms of the timeliness, in as much as everyone is interested in getting some kind of bottom line results from things. There are some things that you simply can't rush. Not so much is known about medical home and how it's set up. There are pilots. There are demonstrations going on and they are very good, but there are also a lot of differences. However, subtle they might be in their application.
We have to be careful to design a demonstration that first of all fits the legislative language that we use and also put something practical on the street. It's real world to us, it's not theoretical. When we put it out for demonstration purposes, this is how we think it might actually look in the end. Something very close to it, but it has to be something that is replicable. We can't create a demonstration that you can't use throughout the country. I find some very good results, but you can't replicate it anyplace.
So, in terms of the timeliness, it's not something that we're included to want to rush to get it resolved. And whether or not there will be conclusions before we are through doing what we have to do, we'll have to wait and see.
DR. STEINWACHS: Okay. Thank you. Harry?
MR. REYNOLDS: I appreciate all the testimony, and as we listen today and tomorrow, I'm anxious. I just went through a lengthy illness with a family member, much like you mentioned. And watching a specialist come in and each day for 45 seconds a day at a time and not be coordinated is an issue.
But, on the other hand, to think of a primary doctor that would be coordinating all that information also added a significant burden to them quite dramatically. So, any comments on that because that seems to be under normal process, a normal information flow right now. I just don't see how that works. This is changing so fast daily that they are trying to coordinate it with different – the other thing is I keep hearing voluntary, but when there is a payment, voluntary only goes as far as where the voluntary stays together.
So, if a patient signs up to be a voluntary member of a medical home and then decides to go out and get four other services outside the medical home, which means they are immediately not coordinated and if the primary care doctor says things are great today, but I don't think you should go do that other thing, then it started to feel and act a whole lot like a gatekeeper. So, if anybody can help with what you've seen so far how that plays in the real world because this committee has done a lot on seven minute doctor visits and all this other. And then you're thinking e-visits. And if you know anybody who uses a lot of e-mail, e-visits could be the goldmine for them if they're going to be checking with a doctor every 10 minutes.
So, all this stuff wrapped together when you're trying to think about how all this works is what's going to be really interesting to listen to. So, if anybody can make a comment now. But, all these things individually sound good, but when you put them together, I'm still struggling with exact process and how it works.
DR. TOOKER: Maybe I can start it, but I want to emphasize that this is a system of care. And probably hard to overstate how important a system of care is. We're so used to these episodic episodes of care in which they're patients who are not in a system of care. The model is one that is a model of trust in which patients say that if you look at Commonwealth data, they are there. Patients want a personal physician in a medical home.
Now, understandably, no model is perfect and there will be patients who say, I don't know. I need to go see my cardiologist. But, if the model works and as a model of trust, patients will be informed and activated and will be participating in their care. And they will develop just as they do now, a relationship to their practice, and if they trust the practice, then at least they'll say don't you think I ought to go do this, as opposed to not telling their practice that they are seeing other physicians.
The reason that people are doing this is a failure of our current system of care. In order to have a patient-centered model such as Chris was talking about that is well established in other countries, these are places that patients trust. And if they don't trust it, then the model is not working.
In order to be activated, the patients need to understand that if they choose not to tell their primary care practice about other care that they are receiving that they are putting themselves at risk because when it comes time to inform the emergency room about what the conditions are and what medications they are on and by the way, medication reconciliation is one of the most important issues n all of this right now and leacs to considerable harm and expense and the model isn't working. So, as indicated, we're at version one and we need to test it to see what really happens to patients. Fortunately, patients being paid under Medicare, if they are choosing to seek services elsewhere and those services are being billed for, you can detect when they are receiving services elsewhere.
I want to emphasize what I meant by the word voluntary. One is that practices can voluntarily decide if they want to be a medical home or not. And as you pointed out, this is complex, hard work, but a fully trained family physician, internist, or pediatrician operating at their level of competence and license is very capable of running a medical home.
The next piece of that is that there is sense that the doctor has to understand and manage all of this. The doctor is responsible for overseeing the team. And when you get into care management, you are delegating this responsibility to other team members to do the care management that you are responsible for. You don't have to know every detail about every patient in your practice as long as you are measuring your practice and delegating responsibility to those who are also fully qualified to do it.
Second is that the home is voluntary to patients. Now, I agree we may get to a point where particularly with those who are paying the bill saying we want this to be less voluntary if we are going to fully participate, but we've learned the hard political lesson with the gatekeeper model that if you tell patients that they are forced into this decision, they are going to find a way out of it now and it erodes the trust factor in the patient-centered medical home. So, patients that need to know that it's voluntary.
The second feature of it being voluntary is to say you're in a community that is fortunate enough to have more than one political home; patients have a right to choose which medical home they think is best for them based on the evidence that's available. So, to me, these are very testable hypothesis to see whether or not this is going to work and is part of testing the model to see whether or not it does work.
DR. STEINWACHS: Greg?
DR. PAWLSON: A quick addition to that. Pardon me for using a sports analogy here, but I think our current healthcare sector is like paying a basketball team. Very good basketball players, the best you can find to shoot at the basket. Not to make baskets, not to play defense, which I would sort of look at –playing defense is like preventative care, trying to prevent things from happening. And then we wonder why the system isn't working.
I had an experience with my mother very much like maybe you're describing with 14 weeks in ICU and her eventual death. And it was really remarkable how many people managed to make a lot of money off that. I think her total hospital bill was close to $400,000. And doctor bill pretty much the same. Yet, nobody in that whole thing was paid to kind of sit with us and try to figure out what the heck was going on.
And I think this has to happen in the hospital for the patient as well as in the outpatient environment if it's going to really be successful. If you just take an international example, or within the United States, look at the difference in cost and quality between the upper Midwest, Minnesota, Wisconsin and Miami, or the United States and Denmark.
And when you really get into those, what is the difference in those things. What do you have in Miami is small aggregated practices that are all mostly fee-for-service. And everybody is just full out shooting the basket every chance they get. Whereas, in Minnesota –- I don't know whether it's the long winters or what it is, but there's a collaborative model that is much longer there. I think some of it is the influence of the health plans or some of it is large medical groups, but there's a pervasiveness of much more coordination of care, much more focus on primary care. So, it doesn't have to be. We've gotten used to it and then we pay for it that way and we wonder why it keeps going.
DR. STEINWACHS: Maybe we have to pay people to be part of a team too. Next, I have Justine and then Nancy.
DR. CARR: I have a question about whether there are lessons learned from concierge medicine. I'm from Boston and a number of years ago, probably three of our premier internist groups opted to become concierge medicine. And for a few thousand dollars, patients could sign up and they reduced their panel size. And I think many of the concepts of the patient care home were achieved by having more availability.
It wasn't as much of a team. Although, in some ways it was, but the other part that was interesting is that patients clearly opted in because they were the ones that were paying the annual fee. I'm just wondering whether there have been any looks at concierge medicine practices such as this and any observations that could align with the medical home?
DR. TOOKER: Very good question. There are lessons to be learned from the Concierge model which patients pay an upfront fee A) to be able to receive the kind of services that we're talking about, such as availability 24 hours a day, seven days a week. And it is a form of a per member, per month payment to guarantee those services the patients desire.
I think on the patient care side that you will find increased satisfaction of patients. The challenges of course are as a scaleable national model, that one, this is not voluntary in the sense that the practice decides who they want in their practice and who they don't want in their practice and exclude patients from their practice that they don't think are the right patients for their practice. And it can be because of the complexity and the severity of illness. It can be because of the affordability of the concierge fee for the practice. And it then pushes those patients to other sources of care in the community such as community health centers if patients can't afford it.
I don't think it's a scalable model. In order for us to get all patients in which they are not differentiated on the ability to pay. Our sense of the medical home model is that every patient, including those who are uninsured ought to have the same access to care as any other patient. So, how to take care of patients and learn how to manage.
So, I think we can learn how to take a certain amount of money and manage patients well from that. That would probably be the primary lesson learned and what does that generate in terms of patient satisfaction. I don't really mean patient satisfaction, but the ability to measure patient experiences of care. Now, that's probably something to be learned, but I think we have trouble with exclusion of patients on the ability to pay.
DR. CARR: I just want to be really clear. I wasn't proposing that as a model, but simply to move forward on the idea of outcomes and satisfaction.
DR. STEINWACHS: Okay, Nancy, you get the last question before break here. We're running a little bit behind.
DR. BREEN: Okay. I had a couple of detailed questions for James, Dr. Coan, at CMS, and a question again, for Dr. Tooker. When you were talking about the research that you're going to do, the medical home intervention, you did say that you got 12 people around the table and they came to an agreement. And I was curious just in the ballpark approximately what does this seem like it's going to cost?
MR. COAN: It's an excellent question. I have to say for the record, I'm not a doctor, but the committee keeps on trying to give me the credential. I've had to fix that a couple of times. Thank you for paying the compliment anyway. As far as the question, we haven't done the calculations of the RVUs. It has to be done within the Medicare concept. One of our divisions actually puts the amounts together.
So, it's not as easy as looking at the values and saying, oh that's a 2.8 or that's a .4 or something like that. And simply saying, oh that calculates out to about $1500 a patient. It doesn't work that way. And we haven't had those calculations run out yet. So, I can't answer your question.
DR. BREEN: Okay. And the second thing was about patient involvement. I'm cuprous about that because there's quite a lot of literature, especially in the cancer literature, which I would be more familiar with about the extent to which patients wish to be involved. And then whether patients want to be informed in their decision-making, or whether patients want to actually collaborate in their decision-making.
And these are multiple concepts that are out there in the literature now. And I wondered -– you said, encourages patient involvement in their own care -– is this something you've thought about much for this project in terms of how you're going to –- what standards you're going to set?
MR. COAN: I think what we're talking about here, and I'll refer to Dr. Tooker's comments – it's a system of care. Therefore, that would be available to the patient as a service. It would be up to the patient as to whether or not they would want to utilize it or not.
They want to be actively participating in their care or they want to be secondarily interested or active in their care. Again, it's not a mandatory thing for them, but the medical home must have the capability I think from our point of view to provide the patient education, the materials and also the reinforcement should they want to become interested and active in their own self-care.
DR. BREEN: Mr. Coan, I'm starting to worry about how many arms are going to be in your experiment - different variations that are going to be tested.
MR. COAN: Again, we're looking at the system of care. The medical home is going to have to qualify. It will be recognized as having certain core capabilities, which the practice becomes recognized for that purpose. Core capabilities and the practice permit the personal physician that the patient responds to directly. The core capabilities allow that personal physician to provide that level of care to the patient. So, again, it's a systematic sort of a thing.
What we'll be looking for primarily from our point of view are the outcomes. Our endpoints will certainly have a financial aspect of it. It might have some kind of a clinical aspect of it, and we'll certainly have a quality aspect of it as well. But, I don't think we'll be looking at it in terms of did this patient participate in their own care.
DR. BREEN: So, the protocol will be not maybe too specified. It'll be more if you have this situation and it can be delivered in many different ways.
MR. COAN: Yes. I think you just hit the nail on the head in that circumstance.
DR. BREEN: I hope you'll document how it is delivered though. I think that'll be an important part of the research
MR. COAN: If it's possible to do that. I'm not so sure.
DR. BREEN: And then I was going to ask—
DR. STEINWACHS: Is it very short?
DR. BREEN: Well, it's a short question and maybe you can keep it short. I was just going to ask what value added you were gong to do because there is an experiment already going on at CMS on medical homes and what other research questions did you want to address or what did you want to find out with the research that you alluded to in your talk?
DR. TOOKER: The pilots that are going on in the private sector and the public-private partnerships are all different. In the pilots that we started initially, among the research is that if we pick an arbitrary payment of per member per month – by per member I mean the plan in the panel patients that the physician has, is that payment the right amount? Is it too little, the right amount, or too much? And we really don't know the answers to those questions. We would like to be able to measure the impact of changes in patients over the course of the period in question. Patients are not static. At the time that you measure what the severity of illness is, six months later, it may have changed. It may have change for the better. IT may have changed for the worse. So, how do you institute a flexibility within a per member per month to be able t not to manage absolute risk because I want to emphasize at the outset that this is not a strict capitation model. As we just cannot afford to let these small practices fail in this experiment that we've asked them to be a part of.
A key part of this that we want to measure is how do you estimate the balance between the traditional sources of practice revenue, fee for service and the amount of care coordination to achieve a desired outcome. We talked about the ability of patients to engage in their care, we know very little that if you fully engage them, what the impact of that engagement is on their care. And that includes improvement in the quality, but it also means reductions in costs because the less able the patients are able to control their conditions on their own outside of the face to face visits, the more expensive that care is going to be because they are likely going to show up in emergency rooms and other places where care is going to be more expensive.
It's one thing to be able to engage with a patient at the point of care. We have to realize that 99 and nine-tenths percent of the time, patients are not in front of you. They are out where they live and they have to manage their own care from day to day. And those who are very knowledgeable about a serious complex disease realize that everyday there are questions about your disease. And you can't just pick up 1-800 and get the answer to those all the time. You have to learn how to do this. And I'd encourage us to think about an article that Frank Davidoff, who is a former editor of the Annals of Internal Medicine, wrote an article about not how sick people are, but how hard, how much work it is to be sick. And once you get into this realm of severe chronic illness, it is relentless. You can't take a day off everyday. So, how to activate these patients and empower them to be able to feel more responsible and that their illness is more responsive to their interventions rather than relying on the practice all the time is an area that we know almost nothing about. This is unchartered territory for us.
DR. STEINWACHS: John, thank you. I'm going to leave people with that idea of no day off and we are going to break until 3:15. That's about 10 minutes. I'm giving you a little bit of time off because we have to keep going and I want to thank all the speakers for the first two sessions. It's really been very good in setting the stage and we'll move on to measurement issues next. Thank you.
(Break)
DR. STEINWACHS: It's really a great opportunity to get at the measurement question and what's being done currently. Dr. Greg Pawlson will be talking about the NCQA measurement process for qualification. I've now learned the difference between qualification and certification. I'll try and not make this mistake again. Harry could have explained it to me too I know.
And this will be followed by Dr. Christina Bethell talking about measurement in the case of the medical home for children with special needs, which is really sort of the beginning of the medical home concept was really focused on this. And the kinds of data that she has been developing at state level and for national state policy purposes.
So, what I'm hoping, we're going to keep the 45 minute time allotment that each of you will be talking in the 10 to 15 minute range at most because as you can see, there are lots of questions and interest from this group and I want to really be able to tap into your expertise. Greg, please proceed.
Agenda Item: Assessment of the existence of PCMH - What We Should Measure and How Currently
DR. PAWLSON: This has been probably the biggest measurement challenge I've certainly seen since I've been at NCQA because it's sort of like the blind man and the elephant. Everybody's got their perspective on what the medical home should look like and to come up with a set of metrics for an animal that nobody can see except by probing a stick at them has been a major challenge.
So, what I'm going to talk about is how will we know one when we see it and how will we know if it enhances value. This is a very very busy slide, but it's kind of an attempt to pull all of this together and to look at how on the left hand slide individual clinician staff's attitudes and behaviors are going to be important for creating the key –- and that's in the orange –- patient centered ongoing care and how office practice systems are also critically important in that. How that comes together. The boards will do evaluation and the boards have actually adopted some of the same metrics that we're using on the practice side. Evaluate individual clinician attitudes and knowledge about systems. And then how do we evaluate the patient centered medical home? And there are really sort of two branches here.
One is the qualification or the basic characteristics we think ought to be there before you call something a patient-centered medical home, but then there's the question of what does it do and how do you actually evaluate the impact of the patient-centered medical home. And for that, you need two major pierces. One is clinical process and outcome measures, including resource use measures, I do believe, at least at a systems level, not at an individual practice level, but looking at systems that use the medical home as a basis versus systems that don't. There should be a difference in overall resource use there.
We should be able to find a way to measure that. There should also be certainly clinical quality measures that show a difference. And indeed, we've seen at least on a cross-sectional study, differences in practices that score high on the PCP instrument that I'll talk about do have higher aggregate quality scores and diabetes and cardiovascular disease as of right now. And then very very importantly, and some will argue this ought to be part of the basic qualification of the patient-centered medical home is the patient experiences a care tool.
And this is the clinician group capped. We think that it should perhaps be modified and enriched with patient-centeredness kinds of questions. It has some clearly, but there are probably an additional set of questions that ought to be considered for use in the clinician group cap survey, either as part of the qualification of the practice. So you'd actually have to pass a certain level of patient experiences of peer performance or certainly in the evaluation. So, you should see an impact on patient experience of care. You should see an impact on cost. You should see an impact on quality. Both in terms of usual clinical quality as well as some overuse and misuse kinds of measures.
This sort of turns what I just said, into sort of an outline in terms of words. The evaluation of the impact of the PCC I've also sort intimated. And this just goes into a little more detail of what kinds of measures might be available to look at the impact of the patient-centered medical home itself beyond the qualification.
This is a way of just conceptualizing the linkage that John pointed out. The medical home model will not work if we don't change the reimbursement system. Period. End of story. It just isn't going anywhere. And I might add this is what's used in Great Britain. And I think it creates what you really want to do is -– and I'm a little bit hesitant to say this. I think there's an economist in the room – but it's a balance payment system that really encourages the clinician to do what they should do for the patient. Not have entered into that either the negatives of either capitation, pure capitation, pure salary or pure pay for performance or what we have now is pure fee for service in terms of motivations that might, in some cases, overcome what he patient really needs. The qualification tool that we've developed is called the physician practice connection. It's an attempt to create what we hope is a reliable valid and scalable instrument that can be used at the indivdiual's site level.
The site can be anything from a single solo practitioner's site, and includes everybody in that site. So, if they have an office practice nurse and an educator or something like that for some reason, that's site. But it can also include sites that have 50 or 100 physicians.
It's a geographic site with a single medical record. So, it has to have that. In many offices here in Washington D.C., there are eight physicians practicing together in a building and they each have their own medical records. That's not a site. And that they have some personnel policies in terms of their clinical practice that are the same. So, it really requires those things.
It was developed and tested over the last seven years and we've published about five papers now looking at the reliability and validity of the instrument, and it's sort of some basic stuff like do people understand when you ask them do you have a registry, do they know whether they have it not.
And we found out that they don't. But, we've actually incorporated into the instrument a requirement that the practice actually –- and there's a whole bunch of options that the practice can use -– actually demonstrate through attached evidence. They can actually attach right to the instrument whether they have an instrument or not. So, it could be a screenshot, for example, of the registry with the patients' names blanked out.
The tools have been in fairly widespread use already in the Bridges to Excellent program that GE fostered. And it was really based and focused around the chronic care model and gauged the extent to which practices had implemented, not just structured, but actually used the chronic care model. So, rather than saying do you measure quality, it goes on and looks at how the practice actually uses the measurement of quality in terms of feeding it back and using it to improve their systems.
It goes beyond sort of handing out patient education materials to how the patients are engaged in using them. So, it tries to get at use, not just structure. Although, basically, one can term them mostly structure or administrative process measures. In formulating the revised version that is being used in the patient-centered medical home, and we sat down with the four physician groups, plus we have a grant from Commonwealth to sort of revitalize some of the old paper data on patient centeredness and sort of also take the IOM definition and try to come up with tools that measure that. And what we came up with were sort of two things that we think measured patient centeredness. One is the physician practice connection modification that we've done. And the second is some modifications to the questions that are in the patient experience survey.
So, the tools that have come out of this thing, include not only the physician practice connection survey used in the practices, but also some changes we hope to the clinician group cap survey. These are the four, and actually a fifth foundational area is primary care. The first contact coordination continuity comprehensive chronic care model and the elements of the chronic care model. What we've come to define as patient centeredness from the study of the Picker Institute and other formulations of patient centeredness. Cultural competency we're trying to build into this. And then the core medical home stuff that came out of the pediatrics literature in terms of their focus around this.
There is clearly overlap in some of these, but there are areas that thee is no overlap. So, we've tried to create again a tool that's in evolution that captures as much of all this as we can in this practice just in site tool. And then where we feel we can't measure that at the systems site level, lead it into a patient experience survey because there are some things that a patient can tell you whether they are experiencing it or not that you really can't measure anything close to reliably and objectively in the act itself.
This is a slide that you can read a little bit better on the hand out, but it basically shows how we've developed this into nine different areas or what we call standards. There's about 15-20 elements in each of these standards that are trying to sort of find measurable traits in a practice.
Measurable in a reliable and valid way that characterize whether or not all the previous formulations and systems are present or not. It's an evolution. We've relied a lot on some stuff that Ed Wagner has done. There's like 250 published articles on the Wagner chronic care model, including some randomized control trials.
The buttress, the different pieces of the chronic care model. As I've said, we've done some cross-sectional studies correlating the scores on this instrument to quality and we're looking at cost right now in terms of how it may or may not be related to total cost of care. It also is interestingly related to the tool that Steve Shortell has used in group practices across the country to look at, again, the systemness and that instrument has also been correlated with higher quality in certain practices. And finally, this is a very living document.
While we've tried to sort of create something that could be used this year, we're going to be testing 50 new elements related to this testing in some projects this summer to see whether they perform better than some of the existing elements. And as we get information back from having this instrument used in a lot of pilot and demonstrations, they are just starting up right now. We'll be taking pieces probably out of it. Again, looking at how well different concepts actually work and how well, if the pilot and demonstration projects are appropriately evaluated, how these different elements relate to patient experiences of care, clinical quality measures and resource use measures. Because that's ultimately what we want, a refined instrument that takes the best out of each of the areas that we've talked about.
The way that it is used right now in Bridges to Excellence, and we're presuming in most of the demonstration projects, is that there are three levels, actually a fourth level, which is not qualified yet. Some programs are actually going to use that not qualified level as a starting point so that any practice that sort of attempt to engage in this will be included in the demonstration and perhaps in the first year's reimbursement. But, as you grade the reimbursement on a member per month level, higher for the higher levels of practice you will hopefully be encouraging practices to all move up the scale.
It isn't a tournament model. It's everybody can theoretically get to the finish line.
So, in summary, and I've provided a whole bunch of slides that describe our work on this and prior research and stuff in the appendices so you can look at them at your leisure or ask questions about them in the question and answer period.
What we think we've come up with is that the primary care patient-centered medical home is the cornerstone to what we think is a true creation of the health system. Measurements critical for guiding us because it's an improvement process and if you can't measure what you're doing, it's awfully hard to know when you're doing it better. We realized the measures are going to be imperfect. There's a strong disagreement between Barbara Starfield and us at this point.
And I think it's very understandable. Barbara came at this form an almost pure lens of primary care and her research suggests that just having primary care in place, regardless of what sort of some of the systems characteristics of the primary cares does improve care and we agree.
However, we think that these other elements of the chronic care model, patient-centeredness and so on, are just as important as the primary care base in actually getting U.S. systems perhaps to leapfrog or catch up with in some cases like Denmark, catch up with things that if you don't have a registry and you can't tell who your patients are, who are not doing well with diabetes, it's awfully hard to give really superb primary care. And that's sort of where would differ.
And again, we have a start of I think measuring both how to qualify these practices, as well as how do we evaluate the impact, but it's only a beginning. And we've got a long way to go before I think any of us are going to be comfortable with saying that we have a set of tools that really ferrets out the most important aspects of the patient-centered medical home.
DR. STEINWACHS: Greg, thank you very much. Christina.
DR. BETHELL: So, thank you for inviting me here. And I think it's great because what I love about medical home is that it really brings up all the issues of the structural and financing and cultural aspects of health care in America. So, you get it all at one time. So, it's very reach, but it also makes it very difficult to talk about. Especially when you talk about measurements, which is a very concrete thing and we're talking about really a model of care. So, let's wish us all look.
What I'm going to do today is focus on measurement work that's been gong on for about the last 10 years on medical home for children and adolescents, both children with special healthcare needs and without. And I think it has application though. I know it has application. In my previous life, I used to work mostly on adult healthcare and adult chronic illness and care models around adults. So, I know there is a lot of synergy here. So hopefully, we can build off of that.
So, my goals are to briefly give you some background on this research so that you can have it as a resource and go back to it because I'm not going to be able to give you a lot of detail today, although if you ask questions, I hopefully will. And then to review quickly three tools that are in the field now to measure medical home for all children and children with ongoing health conditions for which they require above routine healthcare services, which is the definition I'm using for special healthcare.
I'll give you some current numbers. What do we actually know? We actually have data and I'll give you a resource you can look at to your heart's content in your hotel room. Hopefully, you can go online and scan and find all kinds of things about medical home and what we know about medical home in terms of the data that we have nationally and across states by numerous subgroups of children. And it is compelling.
There's a huge amount of variation and it raises lots of questions about why given standardized measurement and variation, you have something you can actually maybe begin to explore. And then quickly touch on adequacy, which is what I was asked to do about what are the issues we need to consider and continue considering.
So, in terms of background, for about the last 9 to 10 years, I and a team of many other people have been trying to look at how to develop tests and gain consensus on uniform strategies and patient-centered strategies for measuring the presence of the medical home. Especially at the federal state health plan, and hopefully, we'll get there provider level. I think we are closer than we think because that kind of consistency at those different levels allows conversation to happen.
When you have similar metrics being measured nationally and in states and at the system level, you can actually talk about and compare and have some hope that you might be comparing and having a conversation that can get you somewhere.
So, the first chunk of research – and again, I'm just going to tell you what was done – and the resources that you can refer back to and then really go into the review of the measurements itself was done between 1999 and 20002 with the support of the David and Lucille Packard Foundation with some supplemental funding from ARQH to take what was then a nascent child survey for CAHPS that hadn't yet been really fully developed and actually develop it largely around this concept of medical home, but at the time, that wasn't something people were talking about.
So, it turned out that we actually did develop something that speaks to all of that, but it was not necessarily viewed that way at the time. So, we can retrofit and have written a paper that explicates how to retrofit the CAHPS survey that was endorsed by NCQA and NQF as well. So, that's the first chunk.
And the National Survey of Children's Health, where we have two rounds, 2003 and 2007 coming out later this year. Both of which have an explicit set of metrics around medical home that are scored together as an index and numerous subcomponents. We did do a study to do a formal item reduction. We started with 113; went down to 49; and ended up with 13 items that were the most predictive of the qualities that we would expect to see if a medical home were being provided. And this was funded by the Maternal and Child Health Bureau, who really has been the thought leader behind developing the medical home model along with the American Academy of Pediatrics and have a number of state implementation grants around medical home going on in the field. That could be a whole ‘nother presentation for you today.
And then finally, the National Survey of Children with Special Healthcare needs, which we also have two rounds of. The data is available online for you to query, and I'm going to give you a little bookmark to tell you where to go to find out about it.
And in 2001, the measure built off of what we did in the CAHPS Children with Chronic Conditions Survey. And the 2005 version is similar, but it did orient a little bit more to a systems model versus a primary care or personal doctor or nurse model, which I will explicate further as being one of the primary choices I think that needs to be made in measurement.
So, the background process was very typical of a measurement process, starting with expert stakeholders, consumer focus groups and frameworks definitions and so on, as well as criteria for measurement being the standard relevant feasibility and scientific soundness. So, that is my quick tour through the research, but I will leave you with just some key products.
There's a submitters kit that ARHQ has a hold of that goes through the detail of the development of the CAHPS Children with Chronic Conditions Survey. We were one of the maiden voyages on letting external people actually participate in developing CAHPS related measures. We have a number of medical home focus group reports from consumers and families, as well as some literature. And then we just have a manual coming out this week on medical home measurements and how it was done, and the pros and cons and different ways to do it.
And then finally, the data resource center for child and Adolescent Health, which was funded by the Maternal and Child Health Bureau as a free resource for people to access the data that they have created in the two national surveys. And I am going to hand this out because I don't have slides. So, I encourage you to go online and you can click your way to the medical home measure pretty quickly and start querying it by states, nations, subgroups of children, by type of health insurance and so and so forth. So, I will leave you with that.
Let me go on to the measures, the three different measurement tools. So, let's just look at the AP definition. It has seven different definitional components, which Dr. Tooker and Greg have talked about being accessible care, family centered care, continuous, which is very difficult to measure and is not represented in either –- we also reviewed MAPS by the way, which is going to be dealt with tomorrow specifically. So, I'm not dealing with that –- but continuous care is actually not represented, partly because of the way you sample for patient input is they have to have already been enrolled for more than a year usually in a health plan. Continuous care is a conundrum that I think we still need to deal with. And I'll mention that more later. Comprehensive care, coordinated, which could mean many things, comprehensive and culturally effective.
So, all together, there are 37 specific concepts, and to operationlalize all of those 37 would take multiple items through each one. So, obviously that's not feasible. We took all 37 and said, well, which ones relate to all children, which ones relate just to children with special healthcare needs or both. And this is kind of how it fell out.
Twenty-four of the 37 concepts really were relevant for all children, and the remaining having more relevance for children with special healthcare needs. That was important for the CAHPS tool because it allows children to be certified by whether or not. And to compare those two groups is actually a very useful and important thing to do.
So, then looking at those 30 of the 37 were amenable to self-report in terms of being able to imagine that you could collect this information in a meaningful and valid way hopefully from people themselves. So, there are a number of things like registries, which it sounds like providers aren't' even sure that they can report on. But, it is difficult. Obviously, you aren't going to ask patients whether or not they are in a practice that has a registry. So, we don't look at those things in this. And that's where this is complementary to those other systems.
So, you have the structural aspects, which really have to be measured differently to some extent. Not always. So, let's look at the CAHP survey. The first is what are the floor criteria? What's the baseline you have to meet in order to consider to have a medical home? And the unique thing about the CAHPS instrument compared to the other two national surveys is that there isn't a floor criteria. You can actually meet the criteria in a number of different ways. And that distinguishes it from the other two. So, I'll tell you that and say more later.
So, the child has at least one healthcare provider is the first criteria. And again, it's not a floor criteria. You can still meet the medical home model if you didn't. Accessible care and I'm curious about how many people are wondering why that is. There are a number of people who have a usual source, but don't have a person that they think of as their personal doctor or nurse. And there are a number of lower income families that get care in federally qualified health centers and see a different provider, but they still meet a lot of criteria for having a medical home. So, I think this is a key issue of is it a person or is it a place. And if it's a person or a place, that's a different way to measure it.
And there are some indications that having a different person each time actually can raise the information level because they ask similar questions. They don't assume so much. And in other cases, you really need those personal relationships to be consistent. The jury is still out on whether or not you will always want it to be a person versus a place. Accessible care. In the CAHPS survey, there are four times around getting care quickly. Many of you know about these. Family centered and compassionate care is a composite made up of four concepts. The personal doctor or nurse knows the child's health and health history. The doctor communicates well, which is defined by five items. Information that is needed is received. Questions are answered, and shared decision making, which is four, questions the classic that Sherry Kaplan and Shelly Greenfield and others have developed numerous years ago around diabetes.
And then comprehensive care is also made up of a number of components and then finally, coordinated care. And in this case, it's very low bar. If you have more than two providers, does anybody help you coordinate that care? And it's asked in a way with a preamble to hopefully help people understand what that means. Very low bar around care coordination in general in all of these tools. And then finally, culturally sensitive, which is really targeted for those who do have language issues and whether or not they have interpreter access and language barriers.
We have three different scoring approaches. I think you always have to have a flexible scoring approach because zero to 100 continuous can be very important for understanding variation and using it in studies and in regression approaches. Then you also need for communication some kind of yes/no. Do they meet the bar? And for that, there's two main ways, the all or nothing. They meet a minimum threshold on everything that I just outlined or average. On average across all of them, did they get it? Which means they could have gotten zero in one place and 100 somewhere else and maybe work it out still to have a medical home
And the paper that I mentioned earlier has reviewed these approaches as well in showing the different stories that get told. Having different scoring options make sense when you're dealing with something that has a lot of complexity to it and you want to be able to communicate about it in a complex way.
So, what do we know from the original testing that was done across 14 health plans, the threshold on average across all subcomponent with 64.1 percent and for the on every about 44 percent. So, obviously, if you have to meet all the criteria, it's harder to be counted. And then the issue of having a personal doctor or nurse. If you have to have a personal doctor or nurse to meet the criteria, fewer people meet it. So, if you don't need to have a personal doctor or nurse, more people meet it.
In terms of variation, we see about a 25 point variation across the health plans. Twenty-five to 30 point variation across the health plans. And this was actually in Washington State, where they do have a medical home requirement in Medicaid and these were on the plans in Medicaid in Washington S tate as represented here.
So, let me go on to the National Survey of Children with Special Healthcare Needs, which Dr. Blumberg is going to give you a more detailed account of tomorrow. I have the difficult job of reviewing a lot in a little time, but there are many resources to get more detail.
In the case of the National Survey of Children with Special Healthcare Needs, there is a floor criteria. You have to have at least one healthcare provider. And there could be more than one because in the case of children with chronic conditions, they often think of their specialist as the person that they really look to. And so, we learn the hard way that we have to stay at least one and not just one because if we say just one, about 15 percent of people say no because they actually have two. So, we can't leave them out. So, we learn the hard way.
So, at least one person that you consider to be your child's personal doctor or nurse who knows their health and health history. And a usual source for both sick and well care. And this really comes from the Maternal and Child Health Bureau's commitment to have both sick and well care in the same place.
Receives family-centered care. And the way that the scoring goes here as everyone's assumed to have a medical home and then they get knocked off if they don't meet the criteria. So, if you don't have a personal doctor or nurse and so on, then you don't have a medical home.
Receiving family-centered care, which is defined in a certain way. No problems getting referrals and this was a switch from before, which was no problems getting specialist care. It switched to no problems getting referrals because that was considered to be something that a personal doctor or nurse could actually control more, and not whether you actually receive the care. And this is where the systems versus personal doctor or nurse model comes into play the most is around the issue of comprehensiveness and access to specialized services.
And effective care coordination, which was a very very low bar in this case and really targeted to people who needed professional care coordination and not your average person with five different doctors who might need some help.
And then in terms of what we know. There's about 47.1 percent of children with special healthcare needs, which was the focus of this survey, have a medical home and the state range again around 20 points. The variation within states is even larger across different types of children according income, according to type of health insurance, according to their race, and that's where that bookmark I gave you can help you find out more about that if you like.
For the national survey of children's health, there are floor criteria having at least one healthcare provider considered to be a personal doctor or nurse. Having had one ore more preventative visits during the last 12 months, which is key for children because well childcare makes up a core of pediatric care. And if you don't have a visit for well child care, that's considered to really de facto mean that you really don't have a very good medical home that's getting you in for that recommended care.
Receives family-centered care. Again, measured in a similar way to the CAHPS survey. Consistent access to urgent and/or phone advice from the child's personal doctor or nurse, which is very critical for children who have approximately 1.8 acute illnesses per year per child.
So, there's a lot of acute care needs here. And no problem getting specialist care, or if you did, you got help. So, it's a slightly higher bar than before. And then effective care coordination again measured more in terms of if you have numerous providers, does anyone help you. Also, did anyone follow up with you after you sought care outside of a personal doctor or nurse? And that's actually a very interesting measure. Did your personal doctor or nurse follow up with you afterwards was the way that that was operationalized as well.
So, what do we know there? I'm separating here by all children and children who meet criteria for having a special need around 44, 45 percent meeting criteria. The range across the states again, in the 20 to 30 range. A little bit larger of a range for children without special healthcare needs.
So, key issues that were considered for operationalizing. These working definitions are still the issues that we need to continue to consider. First of all, what's the content? What's to include for everyone versus the people who have chronic conditions and what we know is that no person is just a person with one chronic condition. If you've got one, you've got multiple. So, what does it really mean to be condition specific here versus looking at what type of special needs people actually have? What are the service needs that people have? We've done some work with CMS and others on think about that for adults and children.
Also, what are the differences by age groups, especially for children who are developing? There are different needs and so on. Assigning values to responses and weights to components. Is everything created equal? Are all responses created equal? Does never mean the same as always. And equal or unequal weighting. These are all the issues that were considered in the development of these tools that I'm mentioning and continue to be relevant. Setting scoring floor criteria. I mentioned that already. Whether you have to have a place or a person. One person, could it be more than one person. If the person is experiencing medical hominess is that enough or do they have to meet criteria that we think is there.
And finally, maximizing the value of the data for reporting and improvement, which is key to how it's scored. Being able to make sure that there is a person level measure that can be combined with other kinds of level measure that can be combined with other kinds of information like their eligibility and their utilization of care, because then we have a powerful dataset that allows us to really understand what's going on.
There's a lot of scoring models that don't do that. So, we have to make sure that the sampling and everything is done in a way where we have person level data that can then be scored at the aggregate as well as child level.
Other issues which will be in my slides are ongoing about validity, anchoring, breadth versus depth, and so on. And there are opportunities for linkage to administrative and other data is a key issue again because we want to be able to learn from this.
And somebody mentioned, Dr. Green, the end of one I think is something to consider. That there really is a need for multiple models of medical home. And that's not what this slide says, but I'm going to say this now. That if it's really about meeting the needs of people in terms of needs met criteria for medical homes; we know that those needs differ. So, it is difficult to apply a standard criteria. And there are ways to customize care and find out where people are at, both in terms of their need and their preference, both of which should be considered.
So, we know something about looking at patient activation and about where people's preferences are for shared decision making, which does actually impact outcome and costs when that's done. There's a lot of literature on that. But, we really, I think, need to be thinking about a needs met model and different ways of characterizing after we figure out what those floor criteria are, which I think we can.
And the parent report or the patient report was used in these cases. It's most amenable to uniform data collection, which is still not true yet of administrative or medical record data. So, that's one reason to use it. The other is that leverages are already in place.
Data collection tools as Greg mentioned, there's the clinical groups CAHPS, which is in place. So editing it and refining it. The mechanism is already in place to use it. So, that's a really good advantage. And the other is again, the medical home is largely defined in terms of meeting the needs of individual patients. And that's always going to be different. And so we need to think about customizing.
Our next step goals are looking at provider level assessment, which we're kind of doing, but not in a formal research context. We don't have a formal project right now. In continuing to integrate methods, we have a call –- I told Don I could only spend a couple of minutes talking to him after the meeting because I'm doing a memo for a call on Thursday to refine the way we measured medical home in the next version of the National Survey of Children With Special Healthcare Needs. So, we continue to look at opportunities to improve and integrate.
Building off current efforts to understand outcome and costs, and I believe getting information to people about cost could be a really important thing if you want them to be meaningfully involved in their care.
And then further specifying measurement options that are low hanging fruit for already endorsed measures. And I'll just use two for children, such as the promoting healthy development survey and the young adult health care survey, which an adolescent self report. Transition into adulthood is a very important issue. And how do we operationalize medical home for youth. We're close. We could get there quickly and I think we should look at the tools that are already out there and endorsed and have many of the components in them just along the lines of what I've been talking about.
And I do want to acknowledge a number of research partners and advisors, especially Barbara Starfield who will be here tomorrow. And I think she would support registries. She just wouldn't as being the only way to measure it, I would think.
Bonnie Strickland, who is actually up in Boston where I was going to be today, at an evidence-based review meeting looking at the evidence for medical homes. We have a long way to go. I hope we can figure out a way to learn outside of randomized control trials as we practice care, and if we don't – we really will. I'll be gone before we figure out if it makes a difference. At least I think I probably will be and a number of other people. So, thank you for your attention and time and I hope this was useful to you.
DR. STEINWACHS: Thank you very much. Let's open it up for questions, comments. I think you've razzle dazzled the committee, the two of you here. Okay. Larry.
DR. GREEN: I'd like to ask you to say more about two things. You very quickly went by close to the front of your presentation, if I heard you right; you said there was a conundrum of continuous care. Could you just explain what you were thinking about there and what that means? And then, it seems to me that you raised a bunch of important issues, but one of them is, if I follow this correctly, you are usually using a surrogate reporter in behalf of the individual that's getting care in the medical home. And much of your datasets are presumably it's a child of some sort whose parents are designated hitters completing the information on behalf of them, right? Do I have that right?
DR. BETHELL: Yes.
DR. GREEN: Could you just talk about the methods of how you get that done and how you do the translation and how you assign that with an eye toward what it would be like to apply that to Medicare beneficiaries?
DR. BETHELL: Absolutely. And there has been some work on proxy measurement in Medicare as well. So, I think we have a lot to build on. And for children, it's more obvious that it's a family affair getting healthcare, and that the parent's experience of access on behalf of their children.
The provider is meant to involve the child, and there are some questions in the CAP survey that asks about that, whether the provider actually talks to the child and not just the parent. But, the proxy measurement, I think really says –- one of the things that I wanted to actually say is there's a concept in child development called a good enough mother. And I think that there's a concept called the good enough medical home that we may not need to think about. Because when you have proxy measurement, you're always going to have people who are more motivated responding and who are able to respond.
So all the patient reported data or parent reported data is an optimistic view of what's happening out there. Both because of how you sample – and this will get to the continuous care issue that usually sampling is done for people who meet certain baseline criteria of having a relationship already with the system, and so they don't even get into the sample until they've been enrolled for a certain period of time. At least six to 12 months. So, that's already getting you higher up the scale. There are language limitations, and usually, how these surveys are done, English and Spanish, and in Washington state some other languages.
So, in terms of proxy, it is the standard for getting care for children about healthcare quality for children up to about age 11 or 12. And then the young adult healthcare survey, which I mentioned at the end, does start to pick up at that age for children who are entering adolescents.
Where it is important to get care information from them directly about both their health and their healthcare experiences and to begin to establish that private and confidential and personal relationship with the doctor.
So, I don't know if I answered your question about the surrogate piece, but I know Ann Riley from Johns Hopkins does have some ways of getting data directly from children and youth. And if a mechanism were put in place to do that, I'd highly recommend it as a supplement and an important piece.
In terms of the continuous care conundrum, it's just really difficult to figure out what we mean by it. What is continuous care and for different people, does it matter? Or, is it having a normal place where they have your health history? And again, this issue of sampling really puts a wrench in the wheel because you're already starting at a higher level.
DR. STEINWACHS: Greg?
DR. PAWLSON: One of the things that is really critical for this committee to sort of think about is first, as Christina pointed out, I think that the two different methods or approaches are very complementary. I think we also have to consider the clinical measures and the resources use measures if we're really going to look at the impact.
And I think that we've been struggling along since –- we are also funded by Commonwealth, or have been at different times -– the conundrum of what you can measure at what level, because we've had a terrible time as Christina knows, having worked with us, even getting health plans to the chronic care child. I think it's the lowest care of any of our CAHPS measured. Virtually everyone does the adult CAHPS. Many plans do the children's CAHPS. And many, many fewer, a handful, do the chronic care. And it's really a pressure of who makes them do it. If Medicare or whatever requires it, then they do it. If they don't, then they often don't.
Measuring patient experience of care at the individual site or practice level is a real challenge. I think it can be overcome, but it's right now potentially expensive. You can't just hand out the questionnaires at the practice and say fill it out and send it back.
There has to be some sort of permissible sampling. And we've been working a lot with Dana Safran who was at Tufts and is actually now at Blue Cross/Blue Shield of Massachusetts and has done a lot of work on the adult CAHPS. The predecessor of the clinician group CAHPS called the ACES Survey, to look at things like how big a sample you have to have. I don't know if you are aware of his work?
DR. BETHELL: We worked with NQF on CAHPS.
DR. PAWLSON: How many you have to sample, which questions have the most practice variation at the practice level, not the health plan level and then how does the practice - if this is a quality improvement model, how does that information get fed back to the practice and how does the practice use or not use it in improving their care? There's actually a nice connection in that there's about six or eight points on the physician practice connection practice evaluation tool that relates to whether or not the practice does some kind – and it really allows them to do just about anything right now – some kind of patient experience of evaluation and then whether they actually use that in improving their practice. So, there's a potential loop here, but there's a lot of really major measurement issues that still have to be kind of worked out before we know, I think, how this is going to work on either a regional level, a national level, a site level, or an individual practitioner level.
Since a lot of the thumbprint on some of these questions on the patient experience survey is not site level, they are actually individual clinician level. Does you doctor listen, for example.
DR. STEINWACHS: Matt, I'll let you have the last question here before we go to the next.
DR. QUINN: One of the really fundamental aspects of what I think of as a patient-centered medical home has to deal with care across setting and across care providers, and that unfortunately is also one of the least developed in terms of communication infrastructure. Can you talk about the level to which organizations in your assessment are able to do that and how you're able to measure that?
DR. BETHELL: I can tell you a little bit about the measurement. There are some. NHQ has been working on getting service agreements between primary care and specialty care, and of course, this is the issue around patient safety as well, when we see problems in safety communication across providers is one of the key issues. So, it's very very critical.
In terms of how it's measured in the tools that I mentioned to you, in the two national surveys, which also provide state level data and some regional level data, there are questions asking about whether or not the child's doctor communicates with the school, which is very key for kids. The school-based piece is huge and also other providers and whether there is follow-up when they do see another provider. Like, do they know what the results were or something like that?
And that's what actually drives the measure down. That is the component of the medical home measure that drives the scores far down. I think 39 percent of those really needed their doctors to communicate with the school and other providers and said that that happened in a satisfying way. Most are saying it's either not happening or it's not happening well. So, you're right on the mark about that.
DR. STEINWACHS: Christina, Greg, I want to thank both of you very much. You've given us a lot to absorb. Christina told me that would be the challenge here and Greg did too.
So, you'll probably hear more from us in the future as we try and sort out and understand. I thank you very much.
DR. STEINWACHS: We're now turning to the next session on patient-centered medical home in practice and drawing on the experience that's available. We have three presenters. The first, Terry McGeeney.
He's a clinical assistant professor at the University of Kansas School of Medicine and President and CEO of AAFP Transform Med, which I saw a beautiful model of how all the pieces should come together. And I assume you are going to share some of that with us now. Terry.
DR. McGEENEY: Thank you for the opportunity to be here. Transfer Med is a subsidiary of the AAFP, and I think one of the things hopefully the next three of us can do a little bit is connect some of the dots that have been laid out so far this afternoon.
And what you'll notice is an awful lot of synergy. None of us are operating in a vacuum. We all hear each other's presentations quite frequently and all work together quite frequently, but that's been a very powerful tool in moving the agenda forward. To echo something personally that John Tooker said this morning, I had the pleasure this weekend of a son and daughter-in-law who both graduated this weekend from medical school at the University of Kansas, less than 10 of the class of 176 went into primary care. Pretty scary. That's Kansas. It really is a crisis across this country. This is more than an interesting academic exercise.
The opportunity today, I think, is to discuss kind of the ground level as it were what the patient-centered medical home is with our perspective. The American Academy of Family Physicians, as John mentioned this morning, did a Future Family Medicine Report from 2002 to 2004. At the end of that, it said, as John stated, that primary care was in trouble. It also defined a model of care. Then it was called the new model of care. Now, we call it the patient-centered medical home.
The AAFP decided that a proof of concept national demonstration project needed to be done, which we called a learning lab in a sense, to really flush out what works, what doesn't work and how you implement this model of care. So, the opportunity I have this afternoon is to talk about real world experience. Not just theory.
It was also commented today that earlier this afternoon, the difference between a medical home and a medical house. That kind of fits well with my slide because when I teach physicians, we first talk about a medical house and the building blocks of a house. You have to get it pretty concrete. The other thing to point out is the model care that we are testing and learning how to implement in our demonstration project, which we are just completing, we decided in mid-2005. So, a lot of this was developed by the best thinking of the time based on a lot of work from IHI, Future of Family Medicine Report and others. So, it's been around a while, and thankfully, it fits quite well. But, it is a research project. It is a learning lab. And I don't mean to sit here and say that the model that we have tested is necessarily exactly the right model. Just as Greg Pawlson said a few minutes ago, we still don't have quite figured out how to test the model because everything is in transition. But, it's being done in a very calculated way.
So, as we look at the medical house, there are certain things that are important. Quality and safety, continuity of care services gets to be the collaborative care. Continuity of care, all those things we talked about earlier, point of care services, doing what the patient needs when they needed it.
Information systems. We'll be talking a lot about that tomorrow. We spent a lot of time on that. As was commented earlier, one of the things that we've learned over the last couple of years was how critically important population based, technology based registries are in the chronic care model.
We really didn't understand that fully a couple of years ago, access to care, access to information. That was talked about today. That relates specifically to the patient. Team based care; John talked about that this morning. And keep in mind, we did this back in 2005, being critically important. And another comment from something that was made just a few minutes ago. We made a decision back in 2005 that for the patient-centered medical home for our purposes is a practice.
Every patient in the practice should have a personal physician, but for intents and purposes as we worked to implement the medical home, it was a practice. We then added something a little bit outside of the medical home as a solid foundation and practice management because we fully understood that if you did all these cool things and the practice went broke, it really didn't matter too much.
So, we spent quite a bit of time with practices just working on business 101 as well. And all that is the overarching philosophy of patient-centered care, which we start drifting away from the house and trying to get to the home. And the definition. A continuous relationship with a personal physician caring for the whole person.
As John said this morning, we don't say family physician, we don't say primary care physician, we say they must have a personal physician, but that personal physician must be willing to step up and provide all of the medical home components within his practice environment.
So, wet take all of those things and that's what we now call our Transfer Med Medical Home for purposes of our project, which is the patient-centered medical home. But, then we take that whole thing and then we flip it. And we do that because now we put the patient in the center. It's all the same tenets.
The talk goes on much longer, but each of these things has a lot of things under it, but the point being it's the patient that's in the center with that continuous relationship that cares for the whole person. And that's when we would try to start moving it from a house to a home. The things I want to talk about again at the ground practice community level are today's realities and tomorrow's opportunities.
Maximizing today's realities. Practices have to become viable in today's environment. The project that we were tasked with was not doing a project assuming there was new money, assuming that there was some boatload of funds coming down to reward practices for doing this. It was enticing and working with practices in today's payer environment to become patient-centered medical homes.
They also must provide what the patients demand. They also must provide what the United States healthcare system requires, and there has to be a quality of life focus for physicians, or the pipeline is going to continue to dry up. And the timeline is short. The FM report in 2004 stated that primary care was not tenable in a 10 to 20 year timeframe. And I think there are a lot of us in this room that still believe that.
Tomorrow's opportunities. There are a lot of pilots out there. There are a lot of things going on with demonstration projects. But, practices have to be positioned to be able to participate in them. They have to get a level one NCQA recognition. They have to commit ultimately to being in a level three. And so, it will be a travesty if CMS decides to do a pilot and there aren't any practices that can participate. And we do feel that practices need to be complete medical homes.
Now, we are just winding up a national demonstration project. It was an $8 million project, funded entirely from AAFP reserves. There was no payer money, no pharma money, it was all Academy money. The only disclaimer to that is like everybody else at this table, the Commonwealth Fund did approach our research team and want to expand their work around patient centeredness. Specifically, how does a patient perceive all these changes relative to medical home.
We chose 36 practices from around the country. We had a 17 page application process. Five hundred and fifty practices applied and there was no new money to these practices to do this. The caveat was that the practices had to commit to implement as much of this new model, not parts of it, all of it, as feasible within their environment in a 24 month time period. A Herculean task by any standard.
We then broke the group up into two subgroups. Eighteen, we aggressively facilitated. Those are advanced degree, PhD, MBA types with at least 15 years of management experience working aggressively with these practices. Site visits, conference calls, webinars, collaborative meetings, and then we had another 18 practices that had to try and do the same stuff on their own, with no outside support. The goal was to figure out how much support is really needed to help practices transform into patient-centered medical homes, which is important for a lot of reasons. Not the least of which, I think all of our professional associations need to know going forward how we support our membership.
All practices were studied equally and independently. We had a research team that involved six academic institutions from around the country. It involved PhD statisticians. It involved registered nurses visiting the practices doing chart audits. And that was completely independent from the facilitation component. The facilitation component ends May of this year. We are going to continue to study the practices for another six months with more visits from the research team to see if the trends continue, to see if there is fall back and hopefully get further learning.
This is where the practices were located. They were large. They were small. They were urban. They were rural. We also thought that was important to know what would work in a large practice wouldn't work in a small and vice versa, urban versus rural.
As John pointed out this morning, many primary care physicians are in groups of six or less in this country. And many many of them are away from the population centers. So, it was important to us to study all of those different demographics.
Another thing to just make mention of because it came up this morning as well, we have another demonstration project going on. It's called P4, which is preparing the personal physician – again, we're not saying primary care physician. We're not saying family physician – for practice. It's a residency demonstration project, where we are actually trying to learn new ways to train family medicine residents in how to be able to practice in medical homes when they leave their residency environment. Much of what, I think it was John, mentioned this morning.
To that, we've had two collaborative meetings to date and with a hotbed of research, the University of Oregon is the independent research component of this project. So, we are studying different ways to train residents. That's different than making the family medicine outpatient centers medical homes.
And it's clear to us that family medical residents must also learn and practice in medical homes. So, back to the NDP. But, I think it's important to learn that we're trying to deal with that whole continuum. We do end it May 31st of this year. As a disclaimer, the Academy gave us $8 million. We only spent about seven of it. That made them very happy.
And we have actually – this slide is a couple of weeks old - we actually have several practices now that have implemented all of the components of the new model within a 24 month period of time. It is possible. And we have had the collaborative meeting that brought all of the facilitated and self-directed practices together for the first time. And that was a fascinating experience.
Some of the things that we've learned that I think our relative going forward is we learned how to support primary care in becoming medical homes. Primary care practices are not prepared to change. Most primary care practices are not motivated to change. That goes back to what Greg was saying earlier. It's tough without dollars involved. There are certainly some in our demonstration project that were highly motivated and were willing to do it without any extra money, but when you get out to the rank and file, it could be an issue.
Primary care practices are woefully uninformed. We recently did a survey that only 52 percent of family medicine practices have ever heard of the chronic care model. We have an awful lot of work to do in this arena. And that's something that I think is critically important to understand.
Leadership at the practice level is lacking, particularly around transformation. Communication is a major limiting factor within a practice. Electronic visits were discussed this morning. They are not well accepted by patients. A lot of patients would much rather take off work for a whole day and get paid for it than have to stay at work and do an e-visit. So, the incentives again are very misaligned and we're finding that the practices have embraced them. We put in the technology, but patients aren't wild about them because it's a lot more fun to take off work, unfortunately, and get paid for it.
Access and cost of care are primary importance to patients. They assume quality. EMR and efficiency are backhaul issues. In a lot of my talks, when we talked about electronic health records, I used the analogy when you go to your ATM and stick in your card, you want money back. You don't care if it's the best IBM technology or three short people giving money out of a cash drawer. That's really where patients are with electronic health records and quality. They just assume the practices are going to do it. And we've learned that in spades, we've sent out over 17,000 patient satisfaction surveys. And a lot of the stuff that we're learning is that kind of stuff.
I mentioned earlier that registries are critically important for chronic care, but practices are unwilling or unable to do manual entry of data. Registries must be self populating. They must be associated with the ability to store and transmit data, and we'll talk about that more tomorrow. But, if we are ever going to be successful around chronic disease, we have to be able to get registries to self populate and we have to be able to use that data and move it around between various entities.
And we have actually successfully done that in eight practices with different EMRs in our national demonstration project. So, this is not pie in the sky technology. This is technology where I can sit in Kansas City and look at a practice in Iowa and see how many diabetics they have. Which diabetics are with which doctor? What is the average A1C of a diabetic in the practice? What's the average A1C by doctor, by patient? Which patients haven't had their colonoscopy this year, and on and on and on and on. It's not pie in the sky technology. It's real world stuff that we need to figure out how to move it forward.
The biggest concern about technology implementation is operational, not cost. Loss in productivity, losing data, and all of those kinds of things. Most practices think they are providing quality, but most are not. We all know that to be the case. Safety at the practice level is woefully inadequate. A lot of the work is done at the hospital level. Not a lot of work is done at the practice level about sample safety, immunization, drug errors, patient education around pharmaceuticals, all of those things that are truly practice responsibility.
As mentioned earlier, understanding and expertise of business issues and practices is solely lacking. Practice ownerships, particularly by hospitals, limit medical home implementation. As John mentioned earlier, the hospital world is finally starting to talk about medical home a little bit, but it's been very very slow to come. And in many cases, 54 percent of family physicians in this country are employed by hospitals. And it's important that we bring that whole arena along.
Many providers have lost their skills. Refer too easily and lack confidence in procedures. Much left over from the old HMO days. Advanced access scheduling is poorly understood and therefore, poorly implemented.
Team care is a difficult concept for family physicians to grasp. We were taught captain of the ship. We were taught we know it all. We were taught everything is our responsibility. And that has been a terribly difficult thing to teach. But, it is critically important in the patient-centered medical home concept. The larger the practice, the harder it is to transform.
You hear a lot of conversation around the country that small practices don't have a chance. In reality, what we find out is the smaller the practice, the easier it is as long as you give them the support. The positive. Population based registries do work and are a critical success factor.
Quality outcomes. Metrics do modify behavior. Team concepts really do work. They lead to higher quality, greater productivity and improved job satisfaction by providers and staff. Practices can do well in today's payer environment when operated as a business.
Practice websites are popular with practices and patients. E-visits work, but patients seem to be better motivated and the incentives need to change. Patients and providers both like group visits. Advance access scheduling really works. Eight percent of our practices have increased their capacity without increasing their workday. The entire model of care can be implemented. Point of care. Evidence-based reminders improve quality and provider satisfaction.
Technology will allow that when a patient comes in for a sore throat, technology can remind them that the patient is due for a tetanus shot or colonoscopy or mammogram, whatever. When the provider has to wade back through the record to do that, it often doesn't occur.
The critical success factors for EMR implementation are change management and planning. It does not have to be traumatic. About half of our practices had EMRs at the start of the national demonstration project. Now, all but one does. So, in light of everything else, we did full EMR(?) implementation in most of these practices.
Another thing that is important to keep in mind is that the components of this new model are interdependent. You really can't do successful chronic disease management unless you deal with your access issues. Unless you deal with your patient education issues. Unless you deal with the ability of the patient to access their information. They are very independent. Another thing, and Greg alluded to this a little bit, doing things does not create a patient-centered environment.
One of the things we're all struggling with is that you can check a whole lot of boxes, but you still don't have a patient-centered medical home. So, then it gets to be a real challenge on how you measure that. There's some sole of the practice that is very hard to quantitative. And then interesting enough and it's kind of antidotal, but an inverse relationship between the time the provider spends with a patient and patient satisfaction. They are very concerned about how much time they spend in the practice and who's doing what and the right things are done by the right people at the right time. It doesn't matter that the doctor spends an hour with a patient and asks about Aunt Nilly's rheumatoid arthritis for the fourth time in a row. So, it's really a challenge to us to get into that team concept, right care at the right time, right sizing.
Some bottom line things in practices that are kind of important to just briefly touch on because primary care practices, the reason they are failing and residents aren't choosing them is because the economics aren't there. They can't even service their medical school debt.
But thinking inside the box, but typical business principles are lacking in practices. Primary care practices today, AAFP data, MGMA data, and our own data suggests they only see between two and four patients and hour, but they are very very busy doing all the junk, filling out all the forms. The nursing home calls, the drug company prior authorizations, on and on and on. They're doing all this stuff, including a lot of the CMS stuff and they can't see patients. They need to see at least 3.3 patients an hour to be economically viable in today's environment.
Operational inefficiencies are an issue. Another thing that is important is that practicing good evidence-based medicine can generate more revenue for a practice. What's that mean? It's a cost to the payer. So, one of the challenges that we have going forward is that every patient gets ever test they are supposed to get, costs could go up, which is a challenge that we are facing going forward when we talked about all this budget neutral things.
We've got to be sure to look at everything across as continuum, and it may be that as Chris talked earlier, spending money now on the diabetics may prevent them from being in diabetic ketoasidosis, getting their leg chopped off and going blind 10 years from now. But, the costs today may actually go up, which is something that we're really, really going to have to struggle with and spend some time thinking about as we move forward.
Group visits are not a cash cow, but they do pay for themselves. And almost across the board, we find that in primary care practices, mid-level providers are poorly utilized again, because of the team concept. So, hopefully, I've been able to share a little bit is real world experience, we have done this in 36 practices across the country for the past two years. Medical home can be implemented. It does work, but there's yet an awful lot of work to be done, and thank you.
DR. STEINWACHS: Terry, thank you very much. That's really exciting. Let's move on and then we'll have questions at the end. The next presenter is Dr. Alan Dobson. Allen is vice-president of Carolina's Health Care System and formerly Assistant Secretary and Medicaid Director. So, I guess you've gotten to see the system from different vantage points. Allen reminded me that some people share the same name. So, I expected someone with a different face.
DR. DOBSON: Thank you. It's a pleasure to be here. I'd like to just kind of walk you through North Carolina's experience with the Medicaid program. And as a disclaimer, I want you to understand that we started this project in 1999, and really implemented it in 2001 with the budget crisis. So, we kind of predated down this road before we started talking about medical home. So, as you see what we've developed in North Carolina Medicaid program, I think you'll judge for yourself where we are. In fact, I was talking to Greg about the interest in going back and looking at all our primary care physician offices and actually seeing how many of them really certifies twos and threes after about eight years. It would be very interesting. I know a lot of them have already.
I'm going to run through this pretty quickly, but again, our goal back in 1998-1999 was on the tail end of the gatekeeper model, which had produced access, but not any quality or cost savings, and there are many reasons for that. And we started off with what was a fairly well controlled experiment in 1998-1999, until 2001 hit and we suddenly were faced with budgetary disaster at the state level, at which time the governor said, if this is your model, expand it right now, otherwise we're going to have to slash and burn the Medicaid budget.
So, we went to work and crisis sometimes the best mother of invention. And we rolled out what turned into be our community care network. So, we're now going back and studying what we did. Again, we realized that all of a sudden the money – when you go out and rob banks that is where the money is. Well, if you want to manage the Medicaid program, you go to where the chronic illness is and the people who are sickest in the Medicaid program.
When we went and started looking at our old PCCM model and looking at what we had in North Carolina after HMOs and the managed care exited Medicaid, and we went to the source and we went to communities and talked to the primary care docs and said, what's the problem? And this is what they told us.
They said, we have not care coordination here. We don't feel like we can, with our increasing demands, we are not able really to manage the needy patients in our practice. We really need some help from the state. And what's more, our communities were operating a system where public health, mental health, home and community-based services were not coordinating at all with the medical system. It was very fragmented. And what was worse was that the state made it worse and functioned mainly as a regulator.
And we really wanted to look at kind of s more chronic illness model. And these are things that are happening now that fit us, but our approach was basically North Carolina is a rural state and wasn't rally well conducive to the typical managed cared approaches. We, at that time, had a lot of small practices, loosely organized medical systems, and small two and three doctor practices in rural areas. And our counties remained strong, but not well organized.
So, we really wanted to look at our Medicaid population. Can we control cost and improve quality? And again, our first pilots were around quality efforts, not cost containment. And we wanted to develop some systems that would help us manage our most difficult patients and really develop what was at that time a personal physician, primary care home, but not a medical home.
Our key visions for the state were that if we were going to have to do this, we needed more than a regulator; we really needed to be a proactive healthcare manager. And that's a real difference for a state Medicaid agency to go out and partner with communities and providers in that sense.
And so, in its essence, community care is really a clinical program. It's not a government or a financing program. So, we created public private partnerships between the state and providers and the medical home, a place where usual care for our Medicaid recipients was key.
And we also wanted it to be physician led and community based. We wanted it to be quality and system oriented, and we wanted to economize by raising the quality and improving the care, rather than going in and typically lowering fees.
So, basically our operating principles are we had to engage the physicians in the community to make this work. But, beyond that, had to engage the hospitals and health departments because it really is about the hand off. It's really about coordinating patient's move through the system. And the ownership of the process needed to be vested in those who really did the work, which is at the community level.
So, we had to get people working together and focus on quality and we need to provide the information because information is key to changing behavior. So, community care is basically making sure people have a medical home, and we create community networks that connect those medical homes to the rest of the local communities.
So, if you've seen one of our networks, you've seen one our networks because every community's resource are really the same. This is really a function-based system we have created in North Carolina between hospitals, health departments, social services and a lot of other stakeholders now locally, including schools, and mental health agencies and other social supports in the communities.
So, this was in 1998. When we envisioned some pilots, this is what it looked like. And they were just a collection, whether it was out of a community health center or health department or a collection of pediatric practices or health departments, we were testing multiple models. But, now in 2008 we have 14 networks, 3,500 primary care physicians operating 1200 medical homes. We manage actually 800,000 now Medicaid enrollees. And because of the success in the budget, this legislature has asked us to manage the ACE fund and S-chip as well. This is what the map looks like.
Again, networks of medical homes and community providers all across North Carolina and they self select themselves. If you've been in government, you don't draw maps if you want to keep your job. So, very political.
So, what are our networks? Well, they are not for profit. They are all inclusive of all providers, including the safety net. We asked them to have a medical management committee, something that we can start doing practice change. And we invest in those local networks by providing a PMPM payment for the Medicaid patients. But, we also pay the practices to participate. It's the spin-off of our old PCCM access fee, but now it's to participate in quality and what we're doing.
And because we're rural, when we wanted case managers, rather than fund the practices directly, we send the money through these networks and then some practices get reimbursed for their staff if they are big enough to have it. Otherwise, it becomes a virtual network for resources that are shared among multiple practices. Another thing that makes this work is we pay fairly well as a Medicaid program.
So, as we add more functions, we're gradually increasing our PMPM payment to the communities, including the aged, blind, disabled, that number has gone up to about $6 or $7 PMPM because of the additional services we're providing. Each network has a medical director, care coordinator, and we hired case managers, care coordinators. It's a basket of functions they perform. And some small practices share them, large practices have their own.
We now actually put clinical pharmacists in the networks and we have a pilot going where some of the practice actually has a practicing pharmacist in their practice. And I'll tell you why in a minute. Well, I'll go ahead and tell you.
When we looked at the reasons for admissions for congestive heart failure and we were looking at telemedicine and things, the real reality is only a third of the patients with congestive heart failure actually took their medicines. So, it's a real adherence problem in actually knowing what the patients are on as they pass between the hospital and the practices and specialists. So, clinical pharmacists, we felt could work within the team that would help that. So, there's more to come on that.
Key attributes of our medical home is just the entry level medical home - access, arranging for hospital, coordinating care. But I think our key innovation is the organization of our rudimentary medical home within a community and creating some physician leadership around that because that's really what makes it work. How do you get practices to change? It's really not a new program that we pay for in Raleigh. It's really about someone going and sitting down in the practices. It's less about the physicians then it is the physician's office staff. And another thing that we are getting ready to do that I'll mention is related to that - providing that practice health.
Another thing is including the safety net providers particularly for publicly funded patients. Having the safety net providers at that table is extremely important as well. So, we're really about building local systems of care. The most exiting thing is what really happens in our local communities. We have practices who get together and then locally, their local management committee gets together and we can drive change in Raleigh by saying we need you to work on this. And then the local practices and the local community care networks will pilot things of their own invention. And the best practices will become state initiatives. These are some of the things we have done statewide.
Again, congestive heart failure, interestingly, when we did the pilots, we discovered the biggest problem was really on medication adherence and case management. We've done interesting things like dental screening by pediatricians and family physicians because we don't have dental access.
But, each network also does their own quality improvement. We also are moving into new things; depression screening, e-prescribing, total location of mental health providers. Again, trying to create some sort of system of care. And what we've found is that by participating in some of these layering -- Robert Wood Johnson funded improving performance and practice -- we applied that to some of our networks and what we found was that practices needed help. So, we've decided that we will fund our AHIC system to actually provide the practice improvement specialist in all the communities.
It's in the governor's budge this year, as well as also the community care networks will be funding those practice improvement specialists to go into the practices to hopefully improve their performance.
Also, we have a new governor's quality initiative, which is a multi-payer initiative around improving practices performance. Particularly in our ambulatory primary are practices.
So, our basic results, we monitor quality. We do claims and chart extractions, but it is pretty low tech. We go in and measure our networks and practices. That's as close as we can get. But, we are moving Medicaid quality to a fairly respectable range compared to commercial payers.
Now, it's not respectable for best practice, I don't think. But, we're actually moving the bar. In our early work, when we really did have some randomized control trials, we did lower hospitalization ED rates and it improved our performance, and then we started moving it out. So, we are kind of the researcher's nightmare. But, the big question is does it save money. And the answer, at least in the Medicaid program is yes. According to Mercer, for that investment in the PMPM, we've saved money from the Medicaid budget every year compared to ourselves last year.
What these numbers mean is that on the left, this is the same counties that were participating reduction in what we anticipated they would spend by new initiatives for the following year. So, what we are doing is flattening the rate of growth. So, it's not cutting budget. It's lowering the rate of growth.
So, in our women and children, we dropped our inflationary increases in the Medicaid budget to four and a half percent per year, which is very respectable. Over on the aged, blind, disabled, it was going double digits, which is why the state mandated us to actually start managing them as well.
The right column is compared if we had done nothing. So, it's fairly sizable savings. And I can tell you, for the first year, our legislature has actually booked savings on their budget in anticipation of changes. That's how much confidence they have that this clinical model will work. And I used to be able to throw a little change out and the legislators walked around at the end of budget time, but they've actually budgeted this.
So, let's look at aged, blind, disabled. This probably mimics the Medicare population a little bit better than a typical Medicaid population. But, we did a pilot of 10 networks, 30,000 patients and as you would expect a lot of prescriptions. A lot of them have co-morbid conditions. A lot of chronic disease. Surprisingly, or maybe not surprisingly, almost 8,000 of the 30,000 had a co-existing mental health condition, which I don't think we've even really started to look at. And then again, we've got suggestive heart failure, which we've found was mainly a medication adherence problem. We'll find out when we fix it if we fix it if we actually were to reduce hospitalizations. But, one year of savings for that 30,000 patients was $53 million. So, you extrapolate that to the rest of our aged, blind and disabled population in North Carolina, that's $250 million of potential savings by improved care.
So, can you save money and reinvest? The answer is yes. You just have to marry the savings, the quality and reinvest in resource and not just let it role back into the budget. That's where the future funding will come.
Our next step is to strengthen the medical home. Again, going back and doing the work. Since we were in such a hurry to roll this out statewide, it's going back and really investing and trying to raise the rest of the -- get a little bit better penetration locally and improve the functioning of all the medical homes in each community.
Also to enhance the practices of the networks' ability to support patient self management. So, we're doing group visits and some other things. Also, integrate some of the specialists expertise. This was built on a primary care system, and as we get into more complicated diseases, particularly in the children with special needs population, you have pediatric neurologists, you have a lot of hand-offs that are extremely important. So, we've invested in care coordinators at some of the medical centers to enhance the handoffs back to the local communities as far as communication.
Also information systems and a big one are Medicaid directors. How do you do the duals because again, you've got two different programs and how do you really manage them to benefit because when you do a good job with the duals, most of the money accrues to the Medicare side out of part A. Or, if you do some of the special needs plans, you end up spending more on home-based services, which helps with hospitalization. So, we have really different kind of initiatives. We've at least postulated and applied for a 646 demo, which would let us manage it clinically in a third of the state and see what happens on a total patient cost savings.
And our projections through Mercer were $1.4 billion over the five year demonstration just on the Medicaid side of the house. So, it's significant savings by improving care. It's just in different buckets. After eight or nine years, I just want to talk about limitations and reflect on something Larry said earlier is that if you're trying to create a model and do it in the next decade, I don't think you can do double-blind studies. What we have found is that the limitations of studying this are hard because of the multiple initiatives operating concurrently in a community. Which one had the effect?
The other part is it was--I sometimes quip, our control group was last year. That's about all we can get to. It's usually last year's budget, not really anything else because it's really difficult to hold a control group if you're operating in a community. And really what was hard is physicians get excited about really making a difference. And it's hard to prevent sharing a beset practice even though you might want to.
We've had a lot of trouble staring this so long ago having really solid pre-intervention numbers from a population. And because of the pressure on budgetary control, which we're probably going to start feeling again with the downturn of the economy, there's a pressure on being a gunslinger and just rolling out programs and not be as thoughtful. And each community really starts from and practice starts from a different position. So there's a lot of variability and penetration of these efforts. I think we need small, but valid sets of measures that we can carry over time to look at the entire population because too many measures will actually make it more difficult for practices to appear and to keep up with it.
I hear our docs hear I've got ADD because of so many things coming at me. From a state perspective, we need evaluation help. States have limited resources and it's like a limited resources and quite frankly limited patients for really doing good research and the resources for study are usually low priority having walked through the legislature.
And the long-term success will require some on the ground system to reside. I do not think that you can throw something out there and the system will reform itself. I think there needs to be some more directed and more help, which is I think what North Carolina was able to do through its Department of Health and Human Services in creating this nice public-private partnership.
I guess my take home message would be there are key attributes of the community care that I think are applicable to other states even though we are a constellation of stars of what happened in North Carolina is not really duplicatable, but there are some key messages that our journey through eight years of developing community care, I think that the government should take a more proactive role in defining what the system should look like, which is the importance of this work around medical home and defining the outcome we don't see.
It's not all about being a single payer. It's all about being a single system because as a family physician in a small town, we are the healthcare system in my town of 1000, along with the local pharmacist. That is the resources. And you'll know that very clearly because we are an all payer system. We don't turn people away. I think there are some key principles as we've talked about. IBM taking the lead. This is not about who pays.
I just think the government can take a more proactive role and understand that this is a function based system. These are functions that need to be provided for patients. I think they are best provided through a medical home and a primary care physician, but it could be provided in a relevant sense as we've done in many of our communities. Because our principles of community care do allow for local variability. We have a website and we try to post all our data. And I'll say that the Commonwealth Fund and the Urban Institute, we've got a lot of help right now in trying to go back and look at seeing what are the core elements of success that we've had?
Again, ours is high level. It's about managing the budget and getting better care. And we know, at least on a global level, that we've done that, but we don't know all the moving parts and what are the most important steps to do that.
DR. STEINWACHS: Allen, thank you very very much.
The third presentation is from Dr. Myles Maxfield, who is vice-president for Health at Mathematica. Mathematica has the CMS contract for the evaluation of the medical home demonstration. And just as you've heard previously from Jim, it's early in the process. So, Myles is going to try and share with us what he could share with us about what the thinking is, I guess, and the directions you're pursuing.
DR. MAXFIELD: Thank you for letting us shares some thoughts about what some of the implications of the medical home model are for data and data systems. As Don said, my disclosure here is that we are teamed with NCQA to design for CMS its medical home demonstration.
I want to be clear about the perspective of my remarks today. I'm going to be talking from the perspective of an operating Medicare medical home program within a fee for service Medicare environment. A couple of key points under the implications of talking about an operating program. I'm gong to be focusing, as I say, on data implications. I'm going to be -- most of the databases and data systems that I'll talk about have to do with all of the participating providers. They are not a sample of the providers. I'm not talking about a research study here.
Another key point is that many of the needed data systems that I'll be talking about that seem to be either required for implementing a medical home model or that would at least facilitate a medical home model, do not exist in the typical practice and will have to be developed. They are going to be new.
Second, a broad point is that I am taking a broad view of data, which is to say most of the databases and flows of data and data systems required to implement a medical home are outside of any federal surveys program. And this is one of the themes of my talk. Moreover, much of the data flowing among all the parties to a practice and operating medical home, much of the data will never flow to a federal agency. Obviously some of it will and we'll talk about which flows through the federal agency and which doesn't, but I will be focusing on even the systems and the databases that do not flow to a federal agency and I'll talk about why that is.
A third perspective is I will be using NCQA's PPC tool as a template to structure many of my remarks. So, many of the speakers today and tomorrow have talked about what a medical home is and what it's components are, why it is important. I am not going to talk about that. I am also not going to talk about the design of the CMS medical home demonstration. There are two reasons for that. The first is that it changes everyday. The second is that CMS has not made it public yet. And therefore, I am speaking on my own behalf. Not on CMS's behalf.
So, here's my list of what a medical home is, and the reason I put it up here is that these are the functions that will drive the need for data and data systems. Maintaining long-term -- care must be coordinated in two dimensions months over time. Secondly, at a cross-sectional sense across all providers and all care settings.
Care has to be delivered pertaining to all types of problems faced by the patient acute care, chronic care and so on, including mental illness, substance abuse and lifestyle issues. Obviously, there's a focus on evidence-based care. And lastly, we've heard from several speakers about access to care. I will be drilling down and access to care and what the data implications of that are.
So, this is a useful triumvirate that helps me organize my thoughts about data and data systems. What do we need data for? We need data to qualify medical homes, which involves data flowing from the provider to -- I'm now talking about operating a CMS program, so flowing to CMS, CMS is likely to develop a feedback report, which flows back to the provider.
The second big function is operating a medical home. So, the medical home provider communicates and exchanges data with a patient. He or she also exchanges information with other providers as part of the coordination function. The third major function is assessing the performance of the medical home program as a home and assessing the performance of the provider, the medical home and the medical home specific. Here, there will be data exchanged between the provider of the medical home and CMS and importantly, there will be data exchange from the patient to CMS in the form of a CAHPS like survey of some sort.
And that would be folded into any feedback report that would go back to the provider. So, let me take these three major functions of data one at a time beginning with qualification. The initial qualification will require, and let me just continue to use the NCQA's PPT tool as my case study here. It will require ht provider to fill out that instrument, provide that information to CMS. That process is likely to be much more efficient if it is a web-based application. So, here' I've got my first data system implication. One implication that's specific to a federal program is that sense--and several speakers have touched on this--let me back up and give a little background here. Sometimes it's easy for us to think that oh my goodness, we're cutting edge and medical homes are the next big thing.
Well, medical homes are in fact, at least in some provider communities, very fashionable right now. They are all the buzz. That's good and that's bad. The bad part of that is that it is easy to say that you're a medical home and it's hard to document that you're a medical home. This poses a challenge to a federal program, which is to say ensuring the accuracy of the qualification data calls for a lot of documentation. A lot of verification. Well, it's sufficiently onerous to document and verify that every box you've checked is in fact in practice and in use that there's a burden problem that CMS implementing program or demonstration has a burden budget and it doesn't want to blow it's entire burden budget on this exercise.
So, we need to find the right spot in that tradeoff. There will also in a non-going program be a need for periodic requalification - perhaps annual. This can either done by asking the participating practices to fill the thing out a gain and perhaps document themselves again, but that would be burdensome.
And alternatively, CMS could adopt an audit approach, auditing a sample of these medical homes. And/or adopting a performance measurement approach, which I will come to in this next slide. Okay. So, what we just did was qualification, and now I've jumped to performance. The answer to the question what do we need data for.
So, this is the way I think of performance measurement in medical homes. Beginning with clinical quality of care, and here, I'd like to focus on an ongoing program of CMS, the physician quality reporting initiative, which does now capture something like $119.00 different measures, most of which are clinical process measures. CMS is considering developing some outcome measures and expanding the program to include those.
CMS could require participating medical homes in a Medicare medical home program to participate in PQRI, it would be a convenient way to collect this information. Cost savings. Any Medicare program is going to want to measure cost performance. In the demonstration, there will be some sort of experimental or quasi experimental design. IN an operating program, that won't be feasible. An alternative that is not equivalent, but an alternative approach is to use efficiency measures. And CMS is currently experimenting and testing efficiency measures for Medicare.
The third category here is practice structure and structure isn't quite the right word, but what I want to do is link this category of performance back to the qualifications question, which is to say do you have all the necessary capabilities and are you using these capabilities required to qualify as a medical home. It could be possible to expand PQRI to include structural measures designed around the medical home model. They're not there now, but that would be possible.
And last, is the patient experience dimension, which a number of speakers have - I think we have to be very careful never to forget it. I've called it ambulatory CAHPS. Greg called it the physician group CAHPS. But, I'm referring to the same thing.
So, it would be possible for CMS to require participating practices to administer or to hire someone to administer such a survey probably based on a sample of their participating patients. And then this would be folded into any performance measurement system and into feedback reports.
Okay. The third major answer to what do we need data for is operating a medical home. And this is one I really want to focus on. The key points here are operating medical home will require many databases and many data systems. Require or be facilitated by. I don't want to be too heavy handed about this. Much of this operational data may never flow to the federal government. And so you ask what is the issue for the federal government? The issue is is there a responsibility in the federal government for setting standards for these data systems? This will become more clear.
So, it's somewhat arbitrarily -- this is a repeat of the slide of what is a medical home. And I've highlighted the access function of being a medical home. And this is completely arbitrary. I just picked this function to illustrate how many data systems are implied by the model.
So, what's implied here? Well, many of the access related functions covered in NCQA's PPC are facilitated by having a fairly fancy appointment scheduling system. A system that will triage the needs of the patient in determining how urgently to make the appointment. A system that will schedule as much stuff to happen in one visit as possible. So as to minimize the number of times a patient has to come back and so on. A fairly fancy system.
Another function, jumping down to the bottom, having to do with access is providing off-hours access. This can include an interactive website where some questions can be answered over the web and some actions can be taken over the web in off hours. Secure e-mail consultation. Twenty-four hour telephone access. All of these things require data systems to be implemented.
This is a list of other operational systems that clearly have a data system implication. And I'm not going to go through them all because it's late and we're tired, but the point here is there are a lot of them, and the data resulting from these will not necessarily flow to the federal agency. And yet, for the federal agency to verify that a practice is functioning and operating as a true medical home there may be a need to set standards and possibly even certify some of these.
Now, I'm not saying certifying the practice as a medical home. I'm saying certifying that the e-prescribing system meets some minimum standard of functionality. And this is not quite as radical as it sounds because CMS is currently considering certifying electronic medical records systems for its electronic medical records demonstration.
It's also considering certifying disease registries as sources of data flowing into the PQRI program. So, this is not -- they haven't made a decision yet. I'm not reflecting that, but certifying software is something that is on CMS's agenda.
So, to wrap up, adopting a Medicare medical home program perspective. The model requires lots of databases, lots of systems. Most of which will never see a federal agency. I think PQRI could be expanded to handle most of the performance measurement issues.
And lastly, that the federal government should consider a role in setting standards for at least some of the really critical data systems in operating a medical home. Than you.
DR. STEINWACHS: Myles, thank you very very much. Let's open it up for questions and comments.
MR. REYNOLDS: Myles, on your presentation you mentioned -- why is it key whether this data would or would not flow to the government?
DR. MAXFIELD: I was making an assumption about the interest and purview of the committee as being really focused on federal data systems. And the point I'm making is that even data systems that are not federal data systems should be of interest to this committee.
MR. REYNOLDS: Right. And they are.
DR. GREEN: Terry and Dr. Dobson, could you react to an assertion that I heard woven into all three of your presentations, how less is more around this measurement of this stuff about how much data collection will the traffic bear? How much can you pull out? While you're taking care of all the folks, how much data can you collect? You talked about self-populating aspects of it. Terry, you've talked about how you can't get everything you want, but you've got to get something.
And then Myles, I listened to you talk and you look at your four categories. If you guys had to tell now, the ten most important data elements that you've just go to get into this to be able to get going, what would they be and if you don't have that list firmly in your mind, do you have an idea about creating a process to find out?
DR. DOBSON: Well, I think the first thing is your claims data. At least in state Medicaid agencies, we pay the bills. We provide the data for our community care programs, which they can then use to at least get some rudimentary population measures. When we started doing disease management, we picked several key measures, rather than a whole arm's list because we had to go into the practice and ask them to collect certain things we couldn't get off claims data.
So, again, it's about administrative burden I think more than anything else. I think the biggest thing the federal government could do is to get us the Medicare claims.
DR. McGEENEY: And I think you can integrate the claims data with -- there are quite a few measures you can get out of a practice if they have an EMR, integrating a lot of the information on that. Even if they don't, with some simple manual practice entry and integrating the lab systems and the billing systems, to get some pretty good data and technology exists is not terribly expensive. And so, I think if you could integrate some payer data and some things that can be gotten pretty simply, if it's a labor burden for the practice, they are not going to do it. They are just not flat out going to do it.
DR. DOBSON: I think expanding once you've created the system is easier than giving people a laundry list of a hundred things. It becomes daunting and again, you're trying to change the landscape of practices across the country.
DR. MAXFIELD: If I may, the laundry list of performance metrics that I put on the slide, that's in a sense -- I'm being a little bit rough and ready, but that's not the burdensome particularly - the burdensome part that I was referring to is the qualification step where you may -- one would be tempted to ask for a lot of documentation to verify that you're actually functioning as a home and not a house. And we've heard a variety of estimates of how much time it takes to provide all the required documentation to support a PPC instrument completed. It takes about an hour and a half to two hours to fill the instrument up. Okay. That's the eay particular. But the estimates we've heard range from 20 hours to 80 hours to supply the documentation for every piece of that instrument.
DR. CARR: Thank you for great presentations. Really very, very interesting and fascinating actually, the interplay. It seems like one of the things we've heard repeatedly today is alignment of incentives. Especially-- well, not especially, every kind of incentive that has come about. And I think that is a profound message. What impressed me very much in North Carolina, the medical management teams. I feel like that is the one place where the incentives come together because the data you have to collect suddenly becomes important to what you are trying to do. And I also think that it comes very close to what the clinicians see themselves as what they want to do.
I realize I'm being provider-centric. I know I've been accused of before. I'll get onto the patient in a second, but as we're struggling with what are we going to do? What are we going to measure? Who's going to measure it? How much time is it going to take? Who's going to fill it out? I'm struck by in that setting of sharing best practices, people come to the table more readily because there's a tangible outcome. And so, to not be provider-centered, it's informed and directed by the person-centric dimension that I think the whole medical model does seem to be changing. It's clearly changing slowly, but we are getting more person centric. But, I thank you all for your very interesting insights and your guidance on how we navigate these very challenging and conflicting waters.
DR. DOBSON: I would just add that not all incentives have to be economic. Having help certify is an incentive, but may not be a direct economic incentive. Providing practice incentive is an incentive. Providing additional resources to a practice or a patient. Again, if you mandate patients into this, it will never work. It has to be a value-added proposition, where people think this is the better way to go and choose to participate. Patient and physicians.
DR. STEINWACHS: Nancy, I saw you raise your hand.
DR. BREEN: I was going to initially ask this question just of Allen Dobson, but I think I'd like to hear Dr. McGeeney's response to it, too. I was really struck as you were talking about North Carolina. I think this is a Medicaid program, but I think you've leveraged a lot of funds. And maybe assistance, technical assistance and help from other sources as well in order to get the job done or the incentive structure. You've put this all together in a really creative way. Could you talk a little bit more about that? About what kinds of monies are coming together because it's not all Medicaid money, is it? And then you've provided assistance and the hours needed to be paid for, or may they are volunteers, but all of that work needed to get done in an organized fashion and timely fashion.
DR. DOBSON: I think if you're building a community system or your' transforming a healthcare system, you've got to get away from even the solid thinking of the funder. We did build it on Medicaid because that's what we had at the time, but it was a confluence of private foundations and some public foundations that actually put some seed money in to get this going.
Medicaid was the first step, but we also had in-kind help from local communities and hospitals as we tried to build this. What we found though is if you go into a practice and change it for one payer, they don't change their practice behaviors just for Medicaid asthmatic patients. You really are talking about a practice change at the local level, which is why some of our networks got into doing the uninsurer. If I could just have a care manager, I would take care of the uninsured patients for free if I could have someone help me get them through the system because it shouldn't take me more time to take care of someone for free than it does for someone I'm getting paid for.
DR. BREEN: So, where'd you get the care manager? I mean, is this just your personal charisma? How did you pull all this together?
DR. DOBSON: Well, again, I think that we had private foundations who helped fund, and in particular --
DR. BREEN: Did they come to you or did you go to them?
DR. DOBSON: Both. We have a unique alignment of people wanting to help and that's why -- our AHQ system is a partner. Rather than going and hiring more staff to do chart audits for our communality care audits, we went to a resource that was already in our state with our area health education centers in all states and said we would like you to do our chart audits. We paid the AHICs to do chart audits. It was a natural next step to say if we want help transforming practices and help giving the extra help to set up registries to train their staff, that we would use existing resources.
So, it's why you see one network being one network or one set of practices. It's because we try not to create something new. We try to leverage existing competence resources. So, one community's health department may be a provider and another one, it provides the case managers.
DR. BREEN: And Terry McGeeney, how did you do it because you pulled off something similar?
DR. MCGEENEY: I'll just echo a little bit first what he said. The care manager's concept really does work, but the key is that those care managers be connected back to the medical home. That's why a lot of case management today hasn't worked that well because it fragments the system. And I think that's what we are going to have to guard very carefully against. And that's where, again technology and information will really support that. But, the care management model does work. And it doesn't have to be a physician doing all that, but it must somehow be created to the medical home. The second comment from what Allen made. All incentive and support doesn't have to be financial. Helping practices collect data. Help practices store data in a language that can be moved. Help practices get population-based registries is an extremely valuable tool - probably more than money in many environments.
Back to the economic issue for us. We had a unique situation because the American Academy of Family Physicians gave us a big bucket of money and we were totally funded by them.
DR. STEINWACHS: Debbie, you were next, was this on the same thing? Then Harry and then Larry.
MS. JACKSON: That was the same question I had. I did hear from you Dr. Dobson a statement less about the physician as about the physician staff. Was that something you actually heard?
DR. DOBSON: It's a statement of this is team sport, and it really involves around what dos the patient need? Because a patient, particularly one that has medical illnesses or needs continuing care interacts with the healthcare system in multiple places. And it really is about making sure the staff and the medical home work correctly. You can't rely totally on the physician to make this happen. So, if you really want a system to change in a physician's office, you have to have the nurse, the office manager, the front office folks. If you want a community system change, you have to engage more than just one agency, one hospital. You really have to get everyone around the table and refine how that patient experience is a system. Just take your own experiences with relatives or yourself and walk through the system and understand how that works.
I heard someone say their experience in the hospital. Not that often, but I still have hospital privileges and we still operate a hospital service. And I'll tell you that our job is translation. It's translating and making sure that none of the balls get dropped. And then it's the handoff back home. That's where all the action is.
DR. STEINWACHS: Harry and then Larry.
MR. REYNOLDS: Nancy, playing off of your point. Because of what Dr. Dobson and they are doing in North Carolina and because of their relationship with physicians and because of what they've done. And I think all of us on this committee have struggled with adoption of anything in a doctor's office. And who is their advocate and how do you group them up. And we've done this. We're all in the middle right now. We'll be hearing Wednesday about the national provider ID, which is still struggling to happen one doctor at a time, one site at a time.
But, for example, in North Carolina, our company is advocating e-prescribing for every doctor in North Carolina. They are the conduit to deliver it. So, you talk about funding sources. If we believe that e-prescribing and we want to raise the boats for everybody and they are the conduit who has already set up the medical home, already work in the process. Already knows how to deal with offices. Already knows what they can do. We're in those discussions right now. Not pre-announcing anything. Just talking about the fact because now we've got someone to talk to that's got 3500 doctors.
We could go out and talk to 3500 one at a time and you're dealing with 3500 boats. Some of them never leave the dock. And so, once this gets going and once you show these relationship, then there are other funding sources, other capabilities that become multi-payer that raises the boat for everybody and start making a difference and that makes it incredibly more attractive to anybody who wants to get in and make a difference because we are dealing with one set of people for 3500 doctors, not 3500 doctors one at a time trying to go meet their office staff and do everything else.
And oh, by the way, that person--once you do e-prescribing or something--that person is going to actually watch what's happening. So, then you get a synergy where it has nothing to do with who's who. It has to do with how do we make a difference in North Carolina. That's a dramatic next step.
DR. DOBSON: Leveraging the pharmacists who work for our networks to go get the other pharmacists who are not currently using e-prescribing so that you can move a community.
DR. GREEN: I wanted to ask Chris Nohrden if he wouldn't mind reacting to what he heard from these three gentlemen.
MR. NOHRDEN: Some of it I did plan on talking about tomorrow and it addresses the whole issue of data integration and particularly in the smaller practices where it is really becoming an issue with a lot of the state and community pilots. If I can hold the question until then and give me overnight to think about it.
DR. GREEN: Let me narrow it substantially. Did you find this encouraging or discouraging on a multiple choice test?
MR. NOHRDEN: I find it encouraging.
DR. STEINWACHS: Leslie.
DR. FRANCIS: Can I follow up with a version of that. I heard the first presentation as the glass half empty. That's Terry's. And the second is the glass is half full. That's Allen's. And I want to ask is part of that was that you were, in some respects, looking at different levels? You were looking more system.
DR. DOBSON: I was glass half full because I'm not the Medicaid director right now. I don't have to worry about the budget.
DR. FRANCIS: I wanted to ask whether that seemed accurate to other people or to the two of you. And if so, if you had any thoughts about why it might be so?
DR. McGEENEY: I structured it that way because I wanted to drive home the point how hard this stuff is. And it kind of goes back to even how hard it is to measure it. It is really really hard to do and it's easy for us to sit around and talk about, but it's very hard to do. So, I think that was my theme. On the other hand, having had practices that have gone through it, they are saying the yare finally enjoying medicine again. I'm finally taking care of patients - which is what I went into primary care for.
So, when you ask the providers that have actually done it, it's more than half full. For us that are out there trying to move the needle, it's probably still half empty.
MR. LAND: I heard community resources being brought to the table, but I didn't hear anything about public health. And I'm just wondering if public health has a history of registries? Has a history of ensuring that patients are taking their medications and so forth. So, I was wondering if public health was one of the resources brought into it?
DR. DOBSON: They are required in our networks as a partner. So, that was early on that we understood that, particularly for the Medicaid program, but really overall having the Public Health Agency as one of the required partners in any of these not for profits. So, I would say that in our networks, they carry multiple roles from provider. Some of them actually provide the care coordination services to the entire network. Some of them are the lead agency in our networks.
Others take a more passive role and have gone to being the provider of patient information and are dealing with some of the literacy issues. So, there are some interesting things the health departments are doing, but they are a key player.
DR. McGEENEY: We spend a lot of time focusing on collaboration of care and teamwork. And one of the questions we always ask our community resources. And one of the things we always work with is practices to identify what the community resources are. And then we ask them how they are engaging them. Having said that, one of the biggest disconnects is between community resources and private practices. You never know what immunizations a patient got and when they got them if you are in private practice and so, moving that information is a real challenge.
Dr. STEINWACHS: Any other questions or comments? Well, I really want to thank Terry, Allen and Myles. It's been a great sessions. I thought it was at least two-thirds full by the time I was done. You can see we're into metrics. Very good. So, thank you. Thank you very much.
Next on the schedule, we do have some time if there is anyone for public testimony or comment. Those who aren't around the table or those who are at the table who are not members of the committee or scheduled to present.
So, if anyone would like to, we certainly welcome that.
MS. PARTRIDGE: Before we lose Dr. Dobson, I want to set the record straight. I think I said in my testimony earlier this afternoon that you are still a gatekeeper program. And he tells me that you are not a gatekeeper program anymore. Is that right? It is more akin to the concept that CMS is talking about with a voluntary--
DR. DOBSON: We grew out of the gatekeeper concept from a federal -- I don't know if we are or not. We don't operate as one.
MS. PARTRIDGE: IN other words, you're primary care provider doesn't have to authorize access to the specialist?
DR. DOBSON: The specialist only has to have the provider's number to get paid. They all have them. So, we don't enforce that part of it. And patients can change their primary care doctor at will. For patient access, again, we found that for women and children we left that program in tact. Aged, Blind, Disabled are obviously not a mandated program. So, we operate it as a non-gatekeeper model.
MS. PARTRIDGE: The reason I thought that was important is that your cost savings projects are based on what you're doing right now. In other words, a gatekeeper is not a prerequisite?
DR. DOBSON: No, in most of our cost saving-driven, and you can see that in our report for decreased hospitalizations and ER visits, advanced access, it's good policy. If you want people to have access after hours, what we did is we said anything after five and on weekends is considered an after hours visit and we pay an extra fee to incent the docs to stay open. To see work-ins in Medicaid and the pediatricians do. So people can walk in and be seen in a lot of areas of the state. And we're seeing our ER rates drop for that.
DR. STEINWACHS: Is there anyone who would like to say something that hasn't had a chance?
DR. BREEN: Well, I'll move to North Carolina for healthcare.
DR. BETHELL: Don, is it okay if I -- there were a couple of things that I wanted to make sure got taken away. I think it's really common to think of patient input as satisfaction information. I think it's really important to recognize that there aren't really very many -- none of the tools that I've presented have any questions about satisfaction on them. And it is true that people can be very satisfied and get bad care and be unsatisfied and get good care. And that's the main reason. So, it's just important for you to know that the tools that are out there are asking about concrete experiences related to clinically recommended aspects of care, including things like self-management and shared decision making and access. So, I think that's important to recognize and also the issue about children, especially children with special needs.
As much as we care about them, they are still a really small proportion of dollars even for Medicaid. And so, to keep on the docket children and youth may be hard because of that. And it's an economically driven world. But I would just like to put a plug in for that because I think the potential benefits in the long run are well worth it.
DR. DOBSON: I agree with you, but if you take the larger view, children with special health care needs belong to families who also are citizens and members in the community and the impact on the family and the community goes beyond taking a single line item in the budget. And that's really true. There's a lot of other budgets where the social supports and stuff. So, I think that's something to remember when you look at any particular population. Not just to consider the single siloed budgetary item.
DR. BETHELL: And it's about one and four families that qualify on a pretty stringent criteria for having a child with a special healthcare need. It's not a small proportion.
DR. BREEN: One in four families with children or one and four families?
DR. BETHELL: One in four families with children. Excuse me. In lower income Medicaid and it's about 21, 22 percent in the higher income. So, it's a good chunk of families. And that's a pretty high bar for quicker identifying such children.
DR. DOBSON: One thing I didn't realize until I was a Medicaid director was if you look at what is a disabled Medicaid recipient before they are a disabled Medicare recipient, they are almost always uninsured. And what is a dually eligible Medicare/Medicaid patient before they ere dually eligible, they are usually elderly and have gotten poor enough from their healthcare to become dual. So, we have to remember that we think in silos, but for the patient, it really is much different. I just want to say for special needs kids, you have to think about the family and the community they live in. The impact is much beyond that single budget.
DR. STEINWACHS: Thank you Allen, and thank you Christina. We have a last -- Leslie, yes?
DR. FRANCIS: I actually had a question for Christina that I didn't ask before, which is whether when you've been thinking about adolescence, whether the design looks really different. And I had the confidentiality question behind that, and I just wanted to know if you had anything to say about that quickly?
DR. BETHELL: Yes. First of all, one thing I didn't say was that the medical home measures I presented are very correlated with many other aspects of the quality that we also measure, including health outcomes in the school and many other things. So, there's much to be mined in there.
And when it comes to adolescence, private and confidential care is absolutely essential, and there are some aspects of medical home that there are really not suited to share about, but there are other aspects that they are really really better suited to report on than parents. And we have a pretty good -- I would say we are about 60 to 70 percent there wit the current measures we have. And it's just a matter of putting them together with another round of testing and development. The private and confidential is absolutely key for adolescents.
Agenda Item: Committee Discussion
DR. STEINWACHS: Any other questions, comments? We put a few last minutes here for sort of end of the day conversation for the committee about what are the take aways or things we ought to summarize or think about from today's session. Is everyone worn out? Justine? Harry?
MR. REYNONLDS: No, again, you started off the conversation by thanking everybody. This has been a great panel. Incredible learning in a half a day and we still have a whole day to work with it. And I think it is very very thoughtfully put together. And many differing perspectives and I think it's really going to move everybody forward. I wish more members of the full committee had the opportunity to hear a lot of this because I think regardless of whether it turns out to be exactly the medical home, it sure does open all the discussions about where everything might end up. And I think that's the thinking that we are going to be facing over the next couple of years, figuring out how to move the ball forward. So, thank you. Thank you to everybody that's done it.
DR. CARR: Yes, and I would echo that again. I also think that, as I said, just a kind of preliminary themes, one is the incentives and another is the tempo. We're hearing about some very exciting things and we're hearing about some very thoughtful prospective analytics. And trying to strike the balance between achieving the benefits while doing it in a thoughtful, durable, understandable fashion, I think is a very formidable challenge but one we might want to think more about.
MS. GREENBERG: In that regard, I guess Myles' assumption that we kind of disabused him of that if it wasn't federal data, we weren't interested in it.
DR. STEINWACHS: We love all data.
MS. GREENBERG: Yes, we'll have to send him the 50 year anniversary. We have to educate him a bit about the NCVHS and how we poked our head in whether we were asked to or not. No, I mean, I thought his presentation was outstanding and they were all excellent and I want to echo the thanks to leadership here and to Matt. I mean, I have been involved in planning a lot of hearings over my day. Mom Greenberg here, but it was as good a process as I've ever seen.
And then you always say well the proof will be in the pudding, but so far the pudding has been really exciting. But, I think regarding the whole standards issue, which of course, is one of my strong interests, is that one way, of course, to do this would be to set standards and then certification standards and then go the whole thing would be to have this whole medical home concept be a use case for the AHIC or the post AHIC, and of the health information technology standards policy, but I think -- I wasn't there, but I was told when they revisited this at their last meeting, they felt that there was not enough -- that this was too new, too developing, et cetera. Too much in motion to really develop a use case and standards for it. But, it's something that I'm sure they will revisit. I mean, there was a lot of interest apparently. I don't know. Debbie, you were there. Do you want to say anything more about that?
MS. JACKSON: (Responds off microphone.)
MS. GREENBERG: So, it's something for us to think in terms of -- I mean, there is a process there. It's just that it wasn't considered ready yet, but it could be in the future.
DR. QUINN: One set of thoughts that I had is the PPCPCMH assessment tool and probably the one that CMS is going to adopt rates medical homes on level one through three. One being sort of the good enough medical home and three being that high functioning one that has all of the bells and whistles that Myles listed at the end of his presentation.
Accordingly, a level one medical home would probably receive one level of additional reimbursement. Two, another level and three, more. And ultimately, I think one of the key questions to decide here is whether one is good enough and how you set the -- you know, you get what you pay for I guess. And how much better, as you measure it is a level three or high functioning medical home versus one. And are we happy as those who pay for it. And figure these things out to have organizations stuck at one. And then say divide it up at some level. Or is there a three and maybe is there a four in the future?
That incremental value or benefit measured and progressing along I think is really that key thing. And that will lead to a lot of complicated questions about, if a registry is good enough, we can figure out a data standard for a registry. Or, if you have to add in an exchange between organizations and exchange of blood pressure data with consumers in their homes, then you're talking a whole ‘nother ball of wax.
DR. STEINWACHS: Larry, I saw you tip your head this way. Did you want to say something or not?
DR. GREEN: No, I'm listening.
DR. STEINWACHS: Anything else you'd like to share at this time? It's been a great day. I agree.
DR. GREEN: In anticipation of tomorrow, something that happened for me this afternoon that I did not anticipate was I think that this set of presentations the way they happened to come together and be presented really did provide the committee with essential background for being able to think about the data model. And the fact that we've got some folks coming tomorrow to talk about the data model is something that excites me. And I think I'd like to forecast that that's going to be interesting, and it might even be important. And I want to ask all of you to listen hard and help think through what the implications of the patient-centered medical home are for a data model. And then in reverse, once you get to thinking about that data model, what are the implications for measuring these characteristics that Myles laid out so well about the different measures.
That strikes me as being close to the pay dirt for the population subcommittee of NCVHS. If we can get our mind around that just a bit tomorrow, it seems to me that perhaps it would have been worth the effort.
MR. REYNOLDS: And one other thing I would add to that is as we listen tomorrow and we talk about data models and the whole flow is how we would want to possibly engage other subcommittees on this subject of medical home. So, for example, when he mentioned the data model, well, with the current standards that we have right now out there don't capture the data that's needed, then it isn't going to magically appear. Leslie has brought us some discussions on the privacy. We've looked at privacy through many many filters and not necessarily the filter of the medical home, which may have that doctor going further than we would have normally assumed that a doctor. We'd have gone with a one doctor, one patient relationship. Not the doctor being in charge of the overall care.
So, those are the things that I would also like to make sure that we extrapolate as we think about it because populations will have the ball on this subject, but on the other hand, if we don't bring this stuff along through some of the other opportunities that we have, the same thing with quality and data that we're going to be looking at and so on. If we don't bring it forward to make this picture actually play out, then I think that's another thing on a subject this big that we need to figure out how we are going to adjudicated it in general. And possibly, that's populations leading it or however we do it. So, thanks.
DR. STEINWACHS: Excellent point. We start at 9:00 tomorrow morning. Nine makes it a little bit easier. I understand.
(Whereupon, the meeting adjourned at 5:50 p.m.)