July 23, 2009: Morning Session Resumed

Jeffrey Schiff:—lunch. I'm a little bit nervous now because this is a little bit where we are going off the script, but I think that some of us feel like as we have talked about this, we kind of need to go off script a little bit. We had started this meeting a day and a half ago and said to all of you that we will talk about the size of the measure set and the scope of it next time, and it is really apparent to us that now is that time. So, we I think need to do that because when we start talking about importance, it will be worthless for us to talk about the importance of individual measures without having this whole discussion, and I think I'm looking at some of the wise folks who had been involved a very long time around that.

I think just to frame this conversation—I guess there are a couple of things. There have been a few themes we have heard here, and they run across currents a little bit. One theme is less is better, and we are definitely hearing that, and that we should stick to less, and part of that, a subtheme of that is that an empty chair may be better than a chair that does not move quality forward. The other side of that theme, the other crosscurrent, though, is let's make sure we look at everything that is available in the universe, which I think the staff at AHRQ, we want to keep their blood pressure reasonably low and then healthy. And I think appropriately so, I think that is why Rita said earlier we really have a cutoff, and we need to look at all those measures. So I think that is one thing. I think that is one tension around this right now that we should talk about.

The other tension I think is that because this has not happened before, we all see that what we have now is the trajectory for the future, and I think people are concerned about that. So I feel like we really need to have probably somewhat of a parallel conversation right now, and one is really on the scope of the measurement set, and I think we should talk about scope rather than size. If we pick a random number, we will hang that out there as a goal, but scope may mean things around, for example, how do we include appropriate amounts of dental and mental health measures and inpatient measures and outpatient measures without having 20 of each?

The scope may include things about cost-effectiveness, about whether we do this by survey, chart review, or administrative data, although there are lots of these issues that I think we should vet that now as far as the scope is concerned, and that will hopefully flow into conversation. I'm sort of glad for less public comment for us to have this time into conversation specifically about importance in terms of the important stuff that we already have in here.

So as we are off script, I'm going to now just remember a couple of things. One is what Cindy said yesterday which is that we have—and I have to find it here where we want to be aspirational and grounded. And then the other thing I'll just say, and then I'll be quiet, is what I have said before is around radical incrementalism means that we have to build this bridge over this chasm and not try to make jumps, but we have to actually do steps that actually build this sequentially. So, I'll let Rita add to this and then we will open this up for a conversation.

Rita Mangione-Smith: I have nothing to add because I think that was very well put and summarized. All I'm going to do is a process thing. I'm always most comfortable if I can track what we are talking about, so I am going to try to record our comments and feelings about scope. I think that—oh, I will add one thing that you had mentioned before but you did not mention just now that maybe in terms of scope, we need to think about levels of measurement. Do we want measurement at how many different levels—at the provider level, at the practice level, at the medical group level, the health plan level, and the State level? I think that is important to think about because some of these measures are more amenable to those various different levels than others are. And if we really, at some time in the future, want to be able to look at all those different levels, then we need to have that in mind as we select measures for the core set. Okay. So as usual, Xavier is going to get us started.

Xavier Sevilla: Okay, I'm going to have to put on my pediatrician hat today and talk about this in terms of the practicing pediatrician and in terms of the membership of the American Academy of Pediatrics. As the student committee on quality improvement, we have been asked again and again from the membership: What is the set of pediatric measures that we should be using? That we should be negotiating not just with Medicaid and CHIP but with private insurance companies for pay for performance. So there is a need to have a good comprehensive set of pediatric measures. So my perspective would be that we really need to look a little bit out of just what we have—this has been a recurring theme on the subcommittee—of the existing measures that are being used by Medicaid and CHIP and actually look for the best set that we can get, number one.

The other thing is in terms of how the measures are used, and I think that is a very important point that these are being used by completely different entities, by plans, by States, and down to the solo practitioner or the medical group. I think this may be another way that we can be very transparent in terms of talking about the measures and maybe even have some kind of footnote on each measure that says, "Well, this is a measure that really is much better to assess at a plan level and at a State level as opposed to the practice level." And there are some that are really throughout like immunizations, for example. I mean you can really do it from the practice to the State. So I would advocate that we really look a little bit further and really come up with the best core set of measures that we can for pediatrics. Even if the perfect is the enemy of the good, I will be concerned if we just limit ourselves to what we have available.

Female Voice: Okay. I just want to say one more time, States cannot afford to undertake extensive data collection right now, and I'm concerned that if we do things too complicated or too difficult for States that you will have no one reporting, and we certainly do not want that either. So I think we have to find a middle ground, something that is going to work that we feel comfortable with. We can monitor quality, but that also is doable by the States right now.

Jeffrey Schiff: Can I—this is sort of a counterpoint in some ways, so I would like to hear from other folks. Before we just go off to another point, let's spend a little time on this because I think this is really a little bit of where we are at. Do we create—are we aspirational, or do we make things that can be done easily, or some combination of that? I think that is—I'm hearing both sides on that, and Cathy, you are—do you want to address this point, Cathy?

Cathy Caldwell: Well, I was just going to say that much is—my heart would be with you. I think I'm on the side of let's try to find a much narrower set of measures that are expected to make a difference in the outcomes for children and make some recommendations about what might be regarded as obsolete or not terribly helpful such that Medicaid directors and CHIP directors and so forth could know what they could abandon in order to pick up on something that we regard as better conceptually and with evidence behind it. I actually do not have any trouble with saying, although scope should be ambulatory, inpatient, and mental health certainly, that the scope might be done that way, but we put a ceiling on the number we would recommend. I mean 10 might be too few but 25 might be just fine, and then if we could really focus in on the best within that number, even if it seems arbitrary, it might put some parameters around what our task is and your task is come September.

Jeffrey Schiff: I'm wondering if this [cross-talking]

Cathy Caldwell: Even if we do something that is a stretch beyond health care reform perhaps but certainly beyond 2010 and into 2013 and beyond.

Jeffrey Schiff: So part of this is how we will present this undoubtedly, but part of it is also sort of how we try to move it forward. I think your point, though, about abandoning measures is certainly valid as well. Okay, so, I guess this is sort of the essence of the whole conversation. So, you are on.

Marina Weiss: Okay. Well, I'm sort of in the same place I think as Cathy. My heart is with you, and I think that that should be the goal but I'm looking at this from the standpoint of making it comfortable for the States and making it work and so, I guess on that spectrum from aspirational to grounded, I would be closer to the grounded end of the spectrum at this point. However, I do think that we need to signal the States that there is more to come, and so I think I would—if I were trying to diagram this for you, I would think in three columns.

One, are maybe the 10 to 25 measures we would like for you to move on as soon as possible, and I would make this really comfortable based on what most States are doing, not a lot of new resources, just begin to think in a uniform way across the country. But then I would have a second column which is maybe moving toward the aspirational end of the spectrum in which, frankly, we are signaling the States these are good measures. The specs are there. They are ready to go. There are a lot your peers and colleagues out there already using them and listen up, we think this may look like phase two. And then, the third column would be for the Secretary by and large and that is, and for the folks at the Centers for Medicare & Medicaid Services (CMS) and AHRQ, we think that some work needs to be done in development of measures in these areas because there are significant gaps that are incredibly important to fill in the not terribly distant future. That would also serve as a signal to the research community about where they need to be focusing their attention.

Jeffrey Schiff: I think that is kind of a well said division. Wow. There are not too many whose cards are not up.

Female Voice: [inaudible]

Jeffrey Schiff: But I think your perspective is very good. Let's go down here.

Female Voice: Well, I would say thank you for the conversation. I think actually what I have heard over the past day and a half is there are really remarkable consistencies and views on these things, I think. And that is I think we all recognize for January of 2010 there is a very pragmatic need out there right now. We have gone from 20 to 4 people at the State; I think we really heard that and understand that. And at the same time, what Xavier was saying—I'll expand on behalf of pediatricians—I think there are similar things going on in primary care, and that is that they are being asked to report on so many things in so many different ways and they are also really begging or searching for some consistency and do not want to waste time on unnecessary data collection where they are not seeing payoff from it.

So I think the degree to which we can move toward consistencies across people who are asking for data is also what we are looking for. I think the thing that people are fearing, recognizing the pragmatic for January of 2010, is that that does not get us into path dependence so that—because we have started there, that is the way we have to keep going. And what we want is we want to hear—where is the opportunity to create the better plan? So I think in doing that, we do not start with, "Okay, give us a list of measures that are there now." Instead, all you smart people would start with, "What are the questions we would want to be able to answer at the end of the day, and what are the most efficient means of getting those data?" So I do not think those things have to be in conflict. I think if people feel that there is a way of having both of those goals over time, I think there is a lot of consistency in views.

Denise Dougherty: It is in the legislation that this is the first step, and the next step is developing a set of priorities for advancing and improving pediatric quality measures for all payers' programs, which CMS—this is probably back in the States, a $60-million program for 4 years is what we have been talking about that has to be in place by January 2011 and hopefully, sooner than that. So there is the legislation—it does not say it. It does not have a logic model in the legislation, but it does lay out that this is the plan and then this January 2010, it is going up for public comment. So I think this conversation has been very rich in having that request for public comment, be very specific about what we want to know from States, including are you going be able to voluntarily measure and report on this?

So there is a plan that it all works together; it is just we are trying to start somewhere within—and do a good job which is why we brought all you experts together. We have this very methodical approach to doing it so that we can be credible with the Secretary, and I certainly like Marina's ideas for the three columns and somebody said some—so yes, that is a long-term plan. It is in the legislation. There is the money for it. The specs have not been written yet. I could not tell you if they were because this little [indiscernible] on the street but there is a long-term plan. So we will get to improve measures. Whether $60 million is enough, that is a different issue but there is other legislation. There is health reform that is looking at quality, and I think we need to make sure that the needs that we may not be able to meet under this CHIPRA bill get into the health reform pieces.

Female Voice: And if I could just add one more point, of course, all of this work in terms of the quality measures and the development, there is noted in the legislation that we are to make recommendations to the Secretary if we need additional legislation in order to promote this, so that opportunity is there.

Male Voice: Great. George?

George Oestreich: Just a couple of observations. I think Marina is right that we have silos of methodology to follow in the development. I do not think we have the resources however to go beyond the first silo right now.

With respect to Xavier's comment, with regard to pay-for-performance issues, I certainly support that within our program and generally; however, those metrics, the specificity of those are generally as we have used them and as I have seen them at a level of much higher detail than what I think we would want. So the challenge I think for us is to use this platform to develop the opportunity for inclusions that would allow us to enroll into that level of metric as we move forward, but I fear if we were to try to do that in great detail at this point, we would struggle mightily in the timeframe that is allowable.

Female Voice: I'll say four quick points. I understand the issue about trying to reach outside but also not tax the States, and I really believe that the sweet spot is right in front of us by looking outside existing measures in Medicare at what States are already doing and what national efforts are doing. The infrastructure is there; it means making some connections between maybe the State Board of Medicine and the State Medicaid folks, and so there is not a lot of duplication. That is so clear at least to me in terms of that is our first step because it is doable.

The second is I would advocate strongly for a very limited set, and I think you said 10 to 15 or so, I would even say 5 to 10—someone said that. I mean I think we need good measures and a small number, and we realize it is a starting point. That being said, I would make sure that as we pick that set, that there are two things. We have talked all about the evidence behind it and the feasibility and validity. What that all comes down to me is, if you tell me my rate, do you know what to tell me to do? Is there a package that if I say, "Here is your rate, now here is what you do." We need that layer on here, and if the answer is, "I do not know what to tell you," then maybe that should be off this first pass, but if there is a solution of how to change your rate, then that should come to the top of the heap.

The last is that I think particularly after listening to Obama last night, this first set I think needs to be clearly focused on saving costs and showing a big bang for the buck so that we can say for the people that have voted for SCHIP and use their capital to get a pass that it was a good investment, and they can keep arguing for it. So those are my four points.

Jeffrey Schiff: Thanks.

Male Voice: Many of the comments I was going to make have already been said, so I will just add on to those. I do think it needs to be a limited set. I think it has to be a limited set for which there is some evidence that you can make the change. I think that is critically important.

I think with regard to scope, we have to be clear that there may not be in all areas of child health, measures that are worthwhile or are evidence-based or meet the gold standard that we can actually do something. I would say that, therefore, we may not be able to cover all areas. Not that I do not think its scope is important, but I think that unless there is a good measure in all areas, we should not just say that in order to be equitable, we are going to spread them out.

The last comment I would make is with regard to why I think it is so important to have a limited scope, and that is I believe that this will not only drive how we measure Medicaid and CHIP, but it will have ripples in other parts of our Federal Government. A good example I think would be in HRSA and how HRSA sets the Uniform Data System (UDS) report. My belief is that if we came out with—the only thing that is measured in the UDS report right now for children—and this is for all community health centers around the country and world health centers and migrant health centers—is pediatric immunizations at 2 years of age.

My belief is if there was another measure that came out of this that HRSA believed was good and reliable and valid and feasible, they would add that. So I think that we should not just be thinking of how this can be used for Medicaid and SCHIP programs, I think it could be used in other parts of federally supported programs.

Mary McIntyre: I just want to kind of second, third and fourth the limited number. From the standpoint of we have actually tried to take this on within the State, and one of the things we looked at was the need to try to keep the numbers low of the measures that we were trying to do because of resources. And the other thing on top of that as far as staffing is the ability to be able to pool the data with all of the other needs that are within the agency for data. The requests coming from all kind of directions, for information related to this specific survey and measures that require a small group of people to pool that data and the information.

And then I wanted to comment on even though limited that I'm really—I guess I stand beyond Medicaid—saying limiting it to Medicaid measures already in use, and the reason I'm saying that I would not just do that is because we found out we started on that road and realized quickly that if we really wanted to get anywhere, that we needed to go beyond what Medicaid was doing. And we quickly kind of revamped even though we started internally, and we pooled external stakeholders in from the State standpoint. And we do talk to the medical association. We have people involved; we know what they are doing.

So that one of the first things we did was to find out from a health care quality improvement standpoint, what initiatives they had going on? Had everybody presented it to the group as a whole so that we can all get on one accord as far as what each other was doing because the reality was we had no idea? And in doing that, the whole discussion then moves towards, well, okay, do we recreate something when there is already something in progress that is showing improved health outcomes? So that instead of going and putting in something that we already have, then the focus moved to what was already out there that we can just kind of jump on a bandwagon and kind of follow what they are doing.

So I think in limiting it to measures that are just currently Medicaid then we are going to miss it because we involved State employees insurance, Blue Cross Blue Shield, who is our major private payer. We have the Quality Improvement Organizations (QIO) involved. We have assistance in health with the Blue Cross Blue Shield, with our Medicare group because there is the involvement also with them in addition to private insurance. So I think it is important to keep that in mind—limited numbers, yes, and maybe even a phased approach where you identify so many to start up with the first year and additional still core measurement set but give people time to get used to doing it before you add on additional measures.

Female Voice: It was very helpful to me to be referred back to the State law or to the Federal law that we are working under and also to be reminded of the mental health bill that was passed that this is the reality. We are not here to develop measures to save children and to move to number one in the world, nice as that would be. We are here to demonstrate CHIPRA's effectiveness, and so that is why I really do think keeping the eye on that, realizing we have all these other organizations including our own professional groups there, speaking for children's health from an epidemiologic perspective.

It disturbed me yesterday that I could not see breastfeeding, I could not see hypertension, adolescent STDs, and then I'm reminded this is about health insurance. And so, the continuity of getting kicked off, picked up, not knowing your provider, all those are the things to keep uppermost for this particular round of measures. I'm going to have to explain that to my colleagues in my organization because they are not going to like it when they do not see their special measure there.

Female Voice: At the risk of pushing Jeff and Rita and Denise over the barricades ahead of me, I would like to suggest—I want to second what Mary said. Forget this business of "in use currently" as if that is the constraint, and if that is what it says in the legislation, then basically say the legislation was written wrong. And the better conceptual thing is to reach out as broadly as possible and not sort of get in this Promethean bed of only what might be in use by Medicaid and SCHIP, so I'm seconding Mary and some other people on that.

I also had a question. The legislation is written -- and I may be over-interpreting what the legislation said—but it does seem as if we started out with this real straightjacket that we ought to be able to get out of. The other question that I had is although it says it is voluntary reporting at the moment, is there any particular reason that we assume that over the longer run, with the use of at least matching Federal tax dollars, some reporting does not become mandatory? So that in principle at least or in theory, some funding behind it whether it is Federal tax dollars or State tax dollars to begin to make clear that yes, the States, even if it is just for Medicaid and SCHIP, do need to be reporting on, say, at least these core measures, and it is no longer voluntary at least down the road.

Female Voice: And just a quick comment, I'm going to go back to the same thing I have been saying. Again, that is another opportunity in terms of making recommendations to the Secretary that that is a legislative change that is recommended. That in order for States to succeed in this and learn from it and be able to demonstrate, they need to be funded in order to do that, so that is an opportunity.

Jeffrey Schiff: And I have one other comment about that, too. I think I have been in positions where I have watched reporting become mandatory, and if it is not done with the appropriate emphasis on quality improvement and transparency in measures, what it becomes is garbage.

Female Voice: Absolutely.

Jeffrey Schiff: And so, I think that we have to be careful to make any recommendation like that that we build something that will be meaningful going forward.

Female Voice: But then do you not come back to Marlene's point about the measures that we might mention first or pick first are ones in which you are able to say here is what you do with those data? You are getting them back, and now here are the next steps to improve, here are the benchmarks and so forth, and that is part of what eventually might go into your report.

Female Voice: There is also under 401A, there is reference to technical assistance and specifically, it is saying that technical assistance should be provided to States through the State planning process, so there is another area that if the appropriate guidance is given to States in terms of how to do this, it will be part of their State plan when they are submitting amendments or just that whole plan process.

Jeffrey Schiff: It is a lot of work to write that. Go ahead, Jim.

James Crall: Alright. Being sensitive of the time, I really like Marina's suggestion about the—I had thought about it on two levels, sort of the grounded, sort of what really could you do now, and the aspirational, but I think that middle tier, and with the signal of the States that this is something that is not going to take a lot of invention. I mean it is information that is at least collected by significant numbers of entities right now and perhaps with a little tweaking could be relatively easily implemented. That signal piece is really nice and I think that is what Mary was saying as well because nobody else is going to say anything about dental—I'll tip my hand a little bit about the dental piece.

I mean from my standpoint, right now, it is one measure with perhaps a minor modification. Now you are going to tell me it is a cluster of measures, but it is basically what is being done in 50 States across this country. In the District of Columbia, to report CMS-416 measures, it is as good a starting place I think to get that into CHIP and to be able to do comparisons in different segments of the population and different ways in terms of those plans are implemented within States. It would be a terrific starting point, and even the research community is starting to be able to use those measures and to get the deeper questions about kinds of providers and adequacy of services so that is my sort of pitch for that.

The dental CAHPS® [Consumer Assessment of Healthcare Providers & Systems] piece, we have decided collectively that it is feasible, and it is valid, and the final vote will be whatever the final vote is, but I think even letting States know that there is such an instrument out there, that it has been used and is being used for at least 10 percent of the kids covered by CHIP in the country, given the state of the budget situation in California from one day to the next and how many kids we dropped off CHIP, but it is already out there, and it has been used, and you can see how it actually has been used. It is something that is already right there.

Now the aspirational pieces, I think everybody has their ideas about that but I think by putting that in that third tier, and mine will be around the continuity and care, medical homes which are derivatives to dental home and adheres to professional guidelines is a piece that there are data elements out there now. Crude as they are, it could start to really shed some insights there so those are the thoughts. I think if we look at that framework, we will not be too narrow in our thinking, and it will give us maybe an iterative process for the group to sort of work through and sort things out among those categories.

Female Voice: Yes, I support what Marina said and what Jim just said, and I think keeping us abreast of the context and the fact that we are in a 5-year process to me is really important, so that I'm feeling that if I do not do it right now, there are still opportunities to make change. And to that end, I would suggest that we consider the input from this group in terms of what will be looked at in 2011 in the program that you mentioned and have everyone here feel that what we recommend in terms of what I would call measure refinement or measure development will get very high priority. It would make it easier to not include your particular issue or your particular need in the first core measure set so that is just a recommendation that I have that we put that forward.

Also, just to remind us of yesterday's discussion about the fact that the legislation requires a measure of duration and the measures of availability, and we have not gotten back to availability, we are pretty much stuck on effectiveness, but just a question about when we are going to get to those.

Denise Dougherty: That is a paper being written by an author who is going to help us look at what the measures are out there and what is reliable and valid and so forth. She could not be here today.

Male Voice: I want to make a call for us while we are being innovative to also stay faithful and grounded in the actual legislation. And there are a number of areas in the legislation that we actually really have not gone down the road enough and one of them, Doreen just mentioned, it is one of the first things in the legislation on page 1, "The presence and duration of health insurance coverage over time"—I really think we need to tackle that in an informed way, and if there are no existing standards, then we develop our own. Lynn had a New England Journal of Medicine paper that got a lot of attention about sporadically insured children; we cannot leave them out of our quality measures.

As you turn to the next page, and it is more than one who stated that racial and ethnic disparities must be part of our standards. In addition, it mentioned socioeconomic status and children with special health care needs, but the fact that they mentioned racial and ethnic disparities twice means to me I think it is important that we think about ways to either integrate that into individual measures or have it over writing but that is part of what we develop.

On the next page, they talk about mental and developmental conditions. We have touched upon that a little bit, but again I'm not sure that we have come up with some standards that we are all comfortable with and, again, if we limit our set, let's at least come up with some that we think are reasonable. And then this really has not been talked about at all, but there is a section toward the end about the denials of eligibility and redeterminations of eligibility. I think those are essential quality indicators, and they are also ways that the system can be gained, and if we leave those out, we may be overlooking again what the legislation is after but also an important quality aspect. Again, it has been said both from a dental perspective and then Lisa mentioned overall, but there is a section on CAHPS® both in terms of quality and satisfaction. I think it is important we keep that part of the standards that we develop as well.

Return to Contents
Proceed to Next Section