July 23, 2009: Morning Session (continued)

Female Voice: Well, some of those items we do not have the data per se, but it seems to me if you are talking to state Medicaid programs, you could also ask their State board of health or medicine or whatever they call it. What do they ask? What are they asking and requiring at the State level? [Cross-talking] I do not have 50 States' worth of data on that.

Female Voice: Yes, as I said, those agencies do not collaborate, so it is not like we get [cross-talking].

Female Voice: Exactly.

Female Voice: It is a whole other phone call. I mean it is a lot more work.

Female Voice: Yes.

Female Voice: And just one thing, [inaudible] Web sites, Medicaid and CHIP, they are not doing a survey of people in all 50 States because [inaudible] response so that is why we have all these other suitable reporting about [inaudible].

Female Voice: I mean the real point here is unless we do something, we will never get out of just ambulatory care. We will never get to the measures of inpatient hospitalizations and EDs, and they are burning, they have quality issues of high cost. And so we have to think out of the box of what Medicaid is, and this goes to Glenn's comment that if we just return the things Medicaid and SCHIPs are already doing, why are we all here?

Female Voice: So I did not know that JCAHO was mandating the use of these other measures around the country.

Female Voice: It has been for a year and a half that they have been requiring them.

Female Voice: I'm sorry, I did not know that. So, if you could send us a piece of information and maybe ask JCAHO to say how they are doing and what their specs are, that will be really [inaudible], enormously helpful.

Male Voice: This is a question for Scot. It has to do with the dilemma around what the definition of integrated care. It seems to me that the most important issues around integration are interfaces between different systems that provide care to kids, and it gets to the interface between ambulatory care, emergency departments, specialty care services, hospital services, and mental health and substance abuse services, and I think it is broader than just assessing medical home. But I'm not exactly sure how you assess that. But I think it is critical in terms of how we minimize cost for individual users of Medicaid and CHIP. I mean because of the definition of things, some people would say, well, integrated care is integration of health and mental health services. I think it is much broader than that, and I would not focus just on that interface.

Scot Sternberg: I think the issue of measurement is a key challenge in there, and I think there are elements in the PPC-PCMH that speak to that, whether it is from the electronic interoperability, or whether it is a component around care coordination. But again, that is not being done by default by the States, and it is done individually. We could make up some composite measures of what the validity and reliability is going to be an issue.

Male Voice: We really like—I mean that is what we are really looking for your feedback on. So for example, there is the National Survey of Child Health and the National Survey of Children's Special Healthcare Needs survey samples of children on a national and State basis. You know, do you get the care you need, and do you feel like your care is coordinated, and that type of question, so that is one way to measure, and if it were done on a large enough sampling basis within plans, you could make those kind of inferences about quality.

The other way is at the organizational level which is sort of what the PC-PCMH measure is, and there again, we are looking for your guidance on this. Those measures go, for example, do you have an agreement in place? Do you have a mechanism for exchanging electronic information in place? So that is kind of a structural/functional measure about whether those systems are in place or not, which again is a different measure than a process measure which is for the last 10 referrals that took place, to what extent was the information available at the time, which would be very hard to measure to be perfectly frank. That is kind of what we are wrestling with, and we would look to your guidance.

I mean the ones that are off the shelf would be thinking of the national surveys and expanding the sampling frame which would have an expense to it but the instruments already exist. Or taking something like—and the closest one that is out there is the PCMH measure—and then requiring States in some measure to apply that and say, just as Scot said earlier, what proportion of Medicaid and CHIP-insured children, for example, are cared for in sites that have level three or level two or level one PCMH? It is a step up from the current measures of to what extent do you have a usual source of care, which I think most of it is a pretty dreadful indicator of quality.

So we are really looking for your input on this point, and to some extent, is anything in this vaguely possible for the core measurement set, or is this all developmental stuff that will be for the 2013? So if you have feedback, that is really what we need.

Female Voice: Cathy?

Cathy Caldwell: I want to suggest to Denise to send an e-mail to Jeremy Globe [phonetic] or Mark Chassin and just ask them for this stuff. They will fall all over themselves to do this, and I bet if you send an E-mail today, they would at least respond to say, "Sure, we will get to you." And I think moving on Joint Commission stuff would be really important right away. And I'm sure either one of them would be happy to jump.

I wondered about in terms of background information not necessarily for the core measure set, but I have not heard any comments about getting information from the Centers for Disease Control and Prevention (CDC) on what States do with the Behavioral Risk Factor Surveillance System (BRFSS), which is obviously a risk factor outcomes kind of measure. I wondered whether there is anything available from the Administration on Children, Youth and Families because they do not seem to be one of the Federal partners, and so maybe asking them, I do not know.

I also wondered whether the Indian Health Service comes in. Surely they must do some of their own stuff, whether it is through the Health Resources and Services Administration (HRSA) or the Indian Health Service (HIS) directly. So maybe there are some sources here of other State information that is—and in the case of the Indian Health Service, certainly that is a disparity population of considerable interest.

On the long-term basis, and this is sort of like out there beyond 2013, is anybody giving any thought to the use of essentially Web-based, Internet-based personal health care records? And whether down the road as the country may move to that sort of resource at least for some—or activity at least on the part of some kinds of families, there might be some way of getting to that. It is long term, but nonetheless, maybe we will be moving towards Internet-based systems in which families or in my view most likely mothers will keep track of a variety of things on behalf of their kids.

And then the final thing is I'm fully enthusiastic and supportive of the whole medical home thing, but in all candor, I think for our purposes, we ought to be focusing on what might be truly in that core set now and think more about some medical home stuff for 2 or 3 years down the road and try the best we can to actually identify some, if you will, traditional quality measures where we really know there is a link between process and outcome. And I say that because I'm not at all clear what evidence might exist to say that there is a medical home does hardly anything for actual outcomes for kids. And I would think if we are going to have some validity criteria for measures, if that kind of information has really been pulled together yet on an evidence base, then perhaps the medical home ideas and the most integrated notion is something for a couple of years down the road and the stretch or additional measures beyond what we might do with course.

Charles Homer: Thank you on that. The last point, two points, one is we did do an evidence review on medical homes. It was published in Pediatrics, and it showed medium quality evidence, moderate effects. So at least there is some link between processes and outcomes that I think is a sufficient basis. But more to the point, there is this language in the legislation that says States need to measure this most integrated care setting so really, it is a question of interpreting what the heck that means.

Female Voice: Yes.

Charles Homer: And then coming out with the measurements so we can report back to Congress and say we are measuring it. I mean that is at least my sense in trying to figure out if that is what they want, and what is it they want, so we can report it back to—

Female Voice: [Inaudible] Ann?

Ann Clemency Kohler: Yes. My question is really for Lynn and maybe Sarah. [Indiscernible] has really changed their focus on the provision of behavioral health dramatically over the past 10 years. They moved really away from treatment into wellness and recovery and using evidence-based practices for treatment of behavioral illnesses. Have the HEDIS [Health Plan Employer Data and Information Set] measures been adjusted to reflect that change?

Male Voice: I'm not aware of any change in the last 12 years [inaudible].

Lynn Olson: So I think I tried to give an idea of why the measures that we have, what they look at and so it is really a matter of what we have been able to do is based on administrative data. And we actually did a project where we developed a whole series of depression measures that start from screening and assessment and documenting symptoms and looking at outcomes, and that is a set of measure concepts that we have not been able to move to field testing or implementation because it requires a huge change in practice along with reimbursement and coordination across behavioral health care and medical care. So I guess I would say that it is a desire, but it is not something where we have had a lot of support for the measures development or for implementation.

Ann Clemency Kohler: Well, maybe this is something the mental health directors and the Medicaid directors could work on because we do have a technical assistance group, and we are trying to really expand evidence-based practice usage out in the community so maybe we can talk more.

Lynn Olson: We would love to.

Female Voice: Go ahead. Cathy?

Cathy Caldwell: I just wanted to say thank you to many of the authors and many of the groups that they represent. A special thank you to NASHP. In the absence of a CHIP director's association, they have really played that role for us since day one and helped CHIP programs immensely, but also specifically to the Urban Institute and HMA, they offer States a huge amount of data and data analysis. We do respond to a lot of surveys, but when I see them from these groups, I really put other things aside and try to provide my input because the analysis is so useful to us.

I also look forward to some technical assistance specifically from Jenny around how to come up with these measures or how to measure continuity of enrollment. I mean that is a huge issue for us even in our program. Much less, looking at data in the Medicaid program and then the privately insured and the uninsured, so helping us think through how to get meaningful data from that perspective. I'm looking forward to that.

And then one question to Patricia. It is a Centers for Medicare & Medicaid Services (CMS) note. I'm not asking for any new requirements but CHIP programs or standalone CHIP programs do not contribute data to the database that you talked about. If that is a useful database, and there could be some meaningful measures gotten from that, meaningful information from that on a national level, what kind of burden would we be looking at for State standalone CHIP programs to contribute data to that database?

Patricia MacTaggart: And I would say on the eligibility, the CHIP eligibility data is in there. There is a code for that, so putting that aside, what you are talking about are services. Most State standalones are through managed care plans. They do not collect encounter data for most of their managed care plans, so that would quite frankly be a huge policy discussion on whether States that are doing standalone are willing to get encounter data.

From a format and a technical perspective, since States are already transferring that information in, it would be an issue going beyond the Medicaid agency to the CHIP standalone, but seeing the bigger issue because it is collecting it from the entity, that current contract probably does not have encounter data. Now that I do with a caveat because some States even when they are standalone, the managed care contract is getting encounter data. So for those States, it is a very different question than for a State that does a traditional managed care commercial plan and does not get encounter data; that is a very big discussion.

Cathy Caldwell: We are a 100 percent fee for service, and I have claims data for 12 years so—

Patricia MacTaggart: You could do it tomorrow.

Cathy Caldwell: Oh, okay, or even if one of our researchers had our claims data, would they have access to this database so that maybe one of our researchers could merge all the Alabama data?

Patricia MacTaggart: Well, one of the issues that is in my paper that actually the CMS does not do right now, which is a big issue for States, is feeding back that data to be able to allow States to do their comparison. But the database is there, so you could quite frankly, technically, you could feed in the CHIP data, and it could become a part of that. There is no reason it could not be, it is just not a requirement right now. But in addition to that, if you did not and wanted to do the comparison, the specs, the guidelines and everything are on their Web site. They have updated it as of February '09, so it is very current. It even deals with ICD-10 transition stuff, so there is a lot of the stuff there.

How it is then used, and again for the staffing resources issue at CMS, I also do not want to minimize it, it takes somebody to do that. But technically, that is one of the things that is very, I use the word doable, feasible like tomorrow.

Rita Mangione-Smith: Go ahead, Doreen.

Doreen Cavanagh: Just a little postscript to the discussion of the integrated medical home. It would be just a request that whenever you say mental health, you say substance abuse treatment. We found out in substance abuse if you do not mention it, it is very easily overlooked. I would just suggest that in the field, there is a tremendous amount of work that has to be done just in integrating mental health and substance abuse so that we have integrated treatment in that area and that might be a subset that you would want to look at in your work, too.

Rita Mangione-Smith: Okay, we are on to Marina. No, I think everybody here is done, right?

Marina Weiss: Mine is just a real quick comment. I'm really intrigued by Trisha's notion that we look at the claims data mechanism as a format within which maybe to help drive all of this forward. I love the timeliness, the accuracy, and so on. There are those of us with groups here in town who if we can make a good case for this, we would be more than delighted to go to Congress and seek the resources that would be needed to make it happen. But the issue is what could we pick up from the claims database and the enrollment data that would link up with what States are already doing through HEDIS and CAHPS? And is there an overlap there that we could maybe look at on a longer term of basis as a place to try to drive the system overall seamlessly?

Patricia MacTaggart: Claims data which is actually the source of some of the HEDIS non-hybrid and traditional administrative data is claims data. So if you are looking at some of the measures you were talking about for dental, those are strictly claims data. Did they get in and get a service during the year? You could take that to way beyond dental, but we do not even look at some of the special needs things because of the small numbers. When you are looking at claims data across the entire Medicaid of children, you can look at some of the things for cystic fibrosis. You can look at some of the things for things that would be traditionally small numbers for a plan but rolled up to a State and there is information by diagnosis and core claims data elements that will allow you to answer those questions. You could go to any one of them, but those are the basic concepts. And it would allow you to cover everything from the baby's birth.

The limitation on the fee-for-service data right now is an identification issue which was somewhat alluded to yesterday. Some States and some plans in that first month running into the mom's number, which is allowed by Federal law, does not work in the world of managed care. Anyway, so most States are trying to retrofit the eligibility to make sure the eligibility of the baby happens the month they were born. But I admit that is another technical issue we do need to work out but we need to fix that for a lot of these.

Marina Weiss: Just a real quick comment on that going back to both Cathy's comments and also Lynn's. Again, it just seems to me that we need to be thinking long term. I know our tenure as part of this project terminates on the 31st of December. but really, this is a long-term project, and the objective as I understand it is to create a structure within which these two programs can go forward and where we can begin to build the capacity and the skill sets necessary in order to be able to standardize this data collection to clean out some things that we ought not be asking the States to repeatedly talk to us about it in different formats and what not and certainly to be able to do some comparative analyses across States and across regions. So going forward to our two chairs, if you would just indulge me and keep this concept on the table, I would be very appreciative.

Rita Mangione-Smith: George?

George Oestreich: I guess the one thing that continues to frustrate me a little bit as we look at the potential measures are certainly the number of measures as a Medicaid unit that we are already reporting. But then as we listen to the work that is being done on the surveys, I'm reminded of the fact that were you to survey any of several of the guidelines that we have discussed, you would need to actually go to three different departments within our State. We talk in various levels with the other two departments, sometimes meaningful, most times not. You have the quality of the survey instrument that you are using. You have the quality of the respondent to the survey, that is the variable. And then I'm thinking, "Okay, if we take all of these and try to find the ones that fit and then try to accentuate those, have we really gained anything?" We have gained maybe not doing the same thing twice, but we still have replicated perhaps the non-validity, if you will—if I want to use that as a word—of the data that we are trying to submit.

So I'm wondering if we need to take a step back after we do this and say, "Okay, what do we really want for measurable tools from this project?" Why do we not set a new goal of getting usable, replicable data that we can go and actually do comparisons from State to State from Nikki's comment? And the only other thing that the personal health record (PHR), the personal health information will be using PHRs within the Medicaid department, but the personal health records are going to be the person's record, and we are not going to be able to extract any data from that; however, we could from the electronic health record. So I guess that is more of a regurgitation of frustration, but hopefully we will [cross-talking].

Female Voice: George, I mean what you are saying is resonating so strongly for me, and I think it is something that Jeff said the first night we were here as we were preparing was, "Let's really think hard about is it better to leave the chair empty at the table or to fill it with whatever we have?" And I think we need as a group moving forward to really keep that in mind that a couple of times it has been raised. It is in the legislation. We have to have measures that are covering X. If there are no valid feasible measures in that area, then we have to go back to the Secretary and say, "Here is where the development piece is. There just is not anything worth putting up in the core right now for this." So I think, as a group, we really need to think hard about that this afternoon and as we move forward.

Are there any other questions? Oh, Lisa, I'm sorry.

Lisa Simpson: One quick thing. I think the whole group really needs to look at Section 402. It is on the second to the last page of the legislation. Because it is clear and given our conversation yesterday that there is a requirement, not an option, for the core set, a requirement for States to report on CAHPS.

Female Voice: It says in there.

Lisa Simpson: Okay? So unless I'm misreading, it is Section 402. It says, "Information required for inclusion in State annual report." And it says, "Data regarding access to primary and specialty services, access to networks of care, and care coordination provided under the State child health plan in using quality care and consumer satisfaction measures included in CAHPS." It is pretty specific, so when you are rating in the next Delphi, this is required, guys.

Rita Mangione-Smith: Well, so that kind of means if it is up to the Secretary, and that is her read and the legislation, it will go in probably regardless of how we rank it.

Lisa Simpson: Well, it is not in the core set. The point here is this is legislation that is requirements under outreach and enrollment.

Rita Mangione-Smith: Right.

Lisa Simpson: Okay, so we have been framing our conversations very much around the core set. So again, I'll defer to legislative analysts who read all this stuff but I think this is a requirement. And it is interesting that it says quality under CAHPS, not just satisfaction.

Female Voice: Well, again, I think in terms of congressional intent, the thing to pay attention to is that there is a core set that is just to kick off the effort. This whole quality section was contemplated as a longer term project.

Jeffrey Schiff: Okay. We have the public comment sheet up here, and we have two people on the public comment sheet. If there is anyone else who wanted to make public comments, can you please come up?

Rita Mangione-Smith: The first person we have for public comment who signed up is Henry Ireys from Mathematica.

Return to Contents

Public Comments 

Henry Ireys: I'm Henry Ireys, a senior fellow over at Mathematica Policy Research. Thank you very much for allowing us to make public comments. First of all, I'll just say I appreciate the magnitude of the task that you all have in front of you. I wish you well in doing all that work in a very short period of time. I guess that really relates to my first comment which I really would urge the committee to, in its final report and recommendations, put this considerable weight that you all have in terms of your credibility and expertise behind, recommending a few strong measures rather than trying to be fully comprehensive and nuanced. I think in the long run, the Nation will be well served by being able to hang its hat on a few measures for which there is great consensus even if we are sacrificing some comprehensiveness as a result of that.

I know as a researcher, certainly, there is a tendency to want to be comprehensive, but I think the opportunity that you have in front of you, I would just vote for make a pitch for really pushing hard on a few good measures for which there is considerable consensus rather than diluting the effort in terms of trying to be too comprehensive when around the margins we really may not have the data or knowledge or consensus for those quality measures.

The second comment I would like to make is that I believe the resources for organizing and analyzing information are seriously lagging behind the technologies for collecting information. We have really at our disposal and soon as electronic medical records begin to dominate private practices, eventually they will, we will be able to have an access to an enormous amount of information, but we have not yet begun to plan or establish the frameworks for analyzing that information. Those are tough. It has to do with lots of different difficult issues like privacy, and who has access to information, but it seems to me that you will not have an opportunity to begin laying the foundation for that. It is not really an issue any longer, I do not believe, about the availability of information. That is on its way. The real next challenge is figuring out how to analyze and organize that information.

The final comment I would just like to make is seconding many other comments, especially Trish MacTaggart's comments, about the utility of MSIS and its derivative, the Medicaid Analytic eXtract file, or MAX as it is known. That really does represent an unexplored and very valuable opportunity to develop child health quality measures that really would be able to tap into a big portion of the Medicaid population. So I do hope that there will be an opportunity to really push that as a resource for looking at child health quality measures. Thank you very much.

Jeffrey Schiff: Well said. Charlie?

Charles Homer: My name is Charlie Homer. I'm the CEO of the National Initiative For Children's Healthcare Quality, and the main reason I wanted to get up was just to express my excitement, which I have not had the chance to do previously in the meeting, and my congratulations to the committee for its work and appreciation to AHRQ and CMS for convening this. I think this has been a long time coming, and I just primarily wanted to let you know how much both me personally and our organization are committed to this task and very excited about the work.

Just give me a second, if I can, to just put on the record who NICHQ is. The National Initiative for Children's Healthcare Quality is an independent not-for-profit organization whose vision is a world in which all children receive the health care that they need and whose mission is to improve child health by improving the quality of children's health care. So, what we are doing here is completely aligned with that. Our principal strategies for doing that include building capacity for quality improvement, which I think is part of what we are talking about here, advancing best practices increasingly through public advocacy, which we are thrilled to be a part of, both in CHIPRA and then through the health reform activities.

In our health reform proposals which we put materials out there, I simply wanted to articulate several of our priorities. Before doing that, saying our primary focus here is in three areas, which we have touched on, has included perinatal care, and we were thankful to the March of Dimes' efforts and our partnership with CMS in that, and I should point out for those who are collecting data from States, there are a number of States that have important demonstration projects in the perinatal area and are creating measures reflecting that, such as the proportion of children with planned late pre-term birth, for example, would be something that could be measured.

We are also working extensively in childhood obesity, and I commend you for the BMI measurement activities I think as that is the most pressing health problem confronting children. I think it is critical that you have recognized that and included that measurement in your group. The third area relates broadly to medical home and more specifically improving care for children with special health care needs. We have been pleased to partner with HRSA and the Maternal and Child Health Bureau in that work.

Our policy agenda includes that children be included in all efforts around quality and in all efforts around health reform. Clearly, that is reflected in the CHIPRA bill, but we hope that is reflected in the rest of it. The second is the creation of a national and regional infrastructure to support quality improvement and measurement, so that includes the kind of technical assistance we have heard about from the CMS level as well as what I hope will be extension service kind of activities at State levels, both in support of quality generally such as the kind of work Marlene has been talking about for hospital-based safety and other work, as well as the primary care extension service model, which we think will be essential if we are going to promote the kind of reorientation of our health care system towards primary care.

Health IT is critical. I think it is critically important that everyone in this room articulate that as meaningful use gets defined, and as we extend the work that Henry talked about. We make sure that the health IT highway that is built includes an on-ramp or whatever the right metaphor is for children and families. Clearly, the expansion of measurement is critically important. I would urge us that although we clearly have and should have this Medicaid and CHIP focused, this is the CHIPRA bill, but as Lynn talked about kids and providers think about—while families and providers think about their children, they do not think about their specific plan at any given point in time. So we need to move towards an all-payer measurement system for children even if do not have the single payer that someone else mentioned. We need to be able to see how are kids doing regardless of or as a result of the rotating on and off different systems.

And last and just to emphasize, we believe that the family role in both the transformation process and the improvement process is critically important and urge us not only to measure that but to think about how we can involve families as critical partners in the selection of these measures. For example, so do we have adequate representation in the room of family stakeholders and groups, and how do we build the capacity of family groups and consumer groups to articulate for and use this kind of measures as an important strategy?

So with that, again, I know that was a little long-winded, but I particularly want to express my appreciation to the Agency, to the committee, to the Secretary, to CMS, and everyone else for getting us to this. The [inaudible] as we would say in Hebrew, getting us to this happy occasion. So thank you.

Jeffrey Schiff: Thanks.

Charles Homer: One last comment. I cannot resist it. I think it might have been to the National Center for Health Services Research (NCHSR) that I submitted one of my very first grants on how to use Medicaid claims data to create quality measures. To be perfectly frank, I do not think it was ever funded because the people are like—we submitted it like 46 times. It was I think 1989 that I wrote that first grant, so it just shows how old you are and how long it takes, but I'm really glad to see that it is happening.

Denise Dougherty: No wonder everybody wants to get everything done today or tomorrow because—

Charles Homer: If I have it, it is one like of those floppy disks that is like this thing—

Female Voice: I think I can top that, Charlie. I did my dissertation on the use of Medicaid data from the [inaudible] and NCHSR to improve quality in the Medicaid program looking to antibiotics and other medications. So, it goes back a long time.

Jeffrey Schiff: We should have a history of pediatric quality improvement. Are there any other public comments? Okay. Seeing none, we actually have 45 minutes before dinner and—

Female Voice: Lunch.

[End of Public Comments]

Return to Contents
Proceed to Next Section