July 23, 2009: Afternoon Session (continued)

Jeffrey Schiff: I want to honor what you're saying, but I think we are in this really weird world of expanding and contracting at the same time because of our timeline, so I think that'll be part of our challenge and hopefully—I think the idea of setting a hard date for the end of August to get everything in that we're going to get in will give us some time to breathe and then do that. Hopefully, some time to breathe.

Denise Dougherty: But I think you're—are you talking about number three up there which is the effort to find good measures in all service categories? That's kind of like representative of a class of problem.

Male Voice: Well, in some sense I think it's sort of the opposite. I mean, I don't know what you meant by all service categories. If you literally mean that, we aren't going to have 10 measures.

Female Voice: No, we're talking inpatient-outpatient mental health. It does [inaudible].

Female Voice: Right.

Male Voice: But then you've got disparities. You have continuity—I mean, we are way past 10. So here's what I—I will tell you what I was thinking, which doesn't need to be what anyone else is thinking. When I think about—I'm thinking how are we doing on identifying developmental delays? How are we dealing with complex populations? How are we dealing with technical quality and safety? How are we dealing with chronic care management? How—I mean, how are we dealing with preventive care? To me, something that represents each of those classes of problems of quality in the system are the only way we're going to get down to a smaller number. That's how I was—that's what was in my head when I said that.

Denise Dougherty: And I think some of those—actually, it's what the legislation had. So there is prevention/health promotion, services to prevent healthy—I mean prevent prematurity, services to aid healthy growth and development. So they include a lot of stuff in here. I can't think of much else.

Male Voice: But I think that your—

Female Voice: That's the legislation piece, so those were the categories captured in the legislation, and that's why we put that up there also.

Female Voice: Right, okay.

Female Voice: So, you know—we will attempt to find good measures that represent that, right? I agree with you—it's going to be incredibly challenging for us to keep this to a limited data set which—that was—there was huge consensus and [inaudible], right? But to still feel like we're covering everything we'd like to cover. I think there's this tension between be comprehensive but be reasonable. I think that that's another—that's a different thing grounded in aspirational, right? Because I think to a certain degree, we could be comprehensive and still have measures that everybody can do, but if it's too many, nobody's going to do it.

Jeffrey Schiff: Right. And I think what we voted for is number one, which was Marina's proposal, in some ways gives us a way of looking up a framework to looking at this as we go forward.

I've been asked by our technical folks to make sure your microphones are turned on and that you are speaking into them so that you can be well heard.

Female Voice: Any other major criteria? One that got raised before was prevalence, which is currently not up there. Severity was one that Denise raised in her document.

Female Voice: No, we did not relate to the [inaudible].

Jeffrey Schiff: I would argue that we should put disparities up there as a separate line even though—

Female Voice: That was going to be my comment. That it's sort of a first among equals in variation. In other words, we're more interested in the disparity variations than perhaps others. And so I'd just throw it out to the group what—are we talking racial-ethnic? Are we talking disease state? I mean children with special health care needs versus not. I mean there is—a number of different ways to cut geography, disparities, rural kids versus non-rural, so I don't know what the answer is, but I think disparities is important.

Jeffrey Schiff: Well, let's—you named a lot of them. I think of the racial-ethnic ones as probably the—probably our first among equals in disparities if its—that's first among equals in variation, but I'm curious about other people's thoughts about that.

Denise Dougherty: What is region or State? And that may overlap with racial and ethnic, or it may not, but we do have some data on that.

Male Voice: Socioeconomics does.

Female Voice: Well, they're all Medicaid or CHIP in the core measure set. [Cross-talk] But there're variations in eligibility category or income level within that. Certainly, there's a lot of difference between TANF (Temporary Assistance for Needy Families) kids and non-TANF kids.

Jeffrey Schiff: And then the other one you raised was children with special health care needs versus—right. Other disparity issues specifically people want to bring up? Okay.

Female Voice: Some of these, there's good data showing the big difference between if you have insurance and don't have insurance now.

Jeffrey Schiff: Right. So insurance does which is—

Female Voice: But I'm going to question that a little because I think these measures will all be being done in children who have insurance.

Female Voice: Right. I realize it's a bit of a logic twist. On the other hand, if it's sort of like the overall goal is to look at what difference does insurance make, right? So we know that there're some areas where the kids who are on insurance have big differences from those who don't. So I would think from a legislator's point of view, there's a face validity to that kind of thinking.

Female Voice: Are you thinking continuous coverage versus discontinuous coverage?

Female Voice: That plays into it. I mean—a strong example is the well-child visit, one where we know that uninsured kids have very low rates of not having well-child visits.

Female Voice: Another way to think about it, and I think Christie's data are helpful here, is publicly insured versus privately insured. So if there's a bigger problem in the publicly insured children, that might be reason to make that an important measure. Is that what you're—?

Male Voice: Can I just quickly try to—I think we're talking about variation and disparities, and we're joining the two, and we need to pull them back apart. Variation, even along dimensions like insured-uninsured where socioeconomic status is helpful as a measure of how much of a quality concern might exist in the system. And that's a measure of importance. Disparities is a measure—is a different category. It has to do with, presumably, populations that may have the same kind of coverage or the same—controlling for other things, and we see disparate outcomes. And I think we need to pull these apart and say that variation may be evidence of a quality problem that makes it important, and disparities is both evidence of variation that may indicate quality and may be a quality problem in and of itself, and I'd just—

Female Voice: So are we saying to take these out from under here?

Male Voice: Well, I wouldn't call them disparities. I would call them sources of variation that would be—I think if we define every form of variation instead of disparities, we've eliminated the concept of variation.

Female Voice: Yes, it's the measurement range. If you have a tight confidence interval around something, it may or may not, particularly if it's already looking pretty good, it's not that important an issue. If you have something in a middling range with huge variation, that almost certainly means that it's an important problem. And so it's just strictly—just think of it as point estimates and confidence intervals around them if that helps—rather than talking about variation. And keep disparities, which are critically important, separate because it's not variation in that quantitative or statistical sense.

Jeffrey Schiff: All right, good. Any other comments on this disparities versus variation?

Male Voice: This is another thought—at least when I'm thinking about importance and it's composite measures of these. And it's the importance of a measure's ability to kind of transform the system. We're trying to get people to look at how we look at measurement and how we deliver care in a different way. And to me, a homerun measure or a sweet spot would be, if I were running a Medicaid program, give me a measure: (1) that I can do and (2) that has significant importance to children in my State and has an action piece tied to it that I could implement something within a year to 2 years to improve outcomes and decrease cost. And I think that's putting a lot of these things together, and that it would affect kids across all—it would affect all kids across health disparities or racial disparities or other issues as well.

So it's almost a composite measure because I think if we could do that and it was relevant enough—and the one that comes to mind is asthma. I mean, if you could get 50 States all collecting exactly the same measures around two or three asthma measures and then implementing changes which we know are already out there of how to improve asthma care, and we know it makes a difference in kids' outcomes, but we also know it significantly reduces cost. That's a measure of importance that's kind of a composite measure. But it's the ability for a measure to actually transform what we're doing.

Jeffrey Schiff: Good. Do you have a comment on composite, George, or something like that? Okay.

George Oestreich: And along the same line to Xavier's original comment that gives you a rolled-up opportunity that you could then use for overall program management, unroll it, give you the measurable metrics that you could tie to other actionable items within your program. That's more like a grand slam.

Male Voice: So your comment is it's a grand slam if you're going to hit it at different levels from the provider level to the State level or whatever or not?

George Oestreich: Well, the grand slam is you have multiple opportunities that are rolled together in a single definable area that's relatively universally accepted and documented. You have multiple levels from payer to institution to provider, and you have something that's monetarily actionable as well. It's a pretty good all around opportunity.

Male Voice: Okay, good. Alright, I was just—I was going to ask you that. Marina?

Marina Weiss: Okay, I'm just sorry I had to be out of the room for a couple of minutes, and so I may have missed this. And if I did, just tell me, and I'll be quiet. But I don't see up there on the list anything that would speak to the issue of the whole spectrum of age that we're supposed to be dealing with vis-à-vis children. I just wouldn't want to wind up with all perinatal measures or all adolescent measures. I'd like for when Denise and Barbara and others go to the Secretary and they say, "This is one heck of a great set of recommendations." I want them to be able to say, "And they meet all of these criteria including having measures that are actionable and are very robust in terms of the validity and whatnot that pertains to different age cohorts of children."

Female Voice: And that sort of speaks of the balanced portfolio kind of concept. Age is one dimension of that. I think earlier we talked about settings of care and those kinds of things—so a balanced portfolio of measures.

Jeffrey Schiff: Right. And to be fair, I think some people's concerns about how we vote for importance now, I think our commitment is that that balanced portfolio has to be developed in September.

Jeffrey Schiff: Okay. We're looking for other comments about the criteria of importance.

Rita Mangione-Smith: I would like to wind this up in the next 5 minutes or so though, so we have time to go through the sheet.

Jeffrey Schiff: You bet.

Rita Mangione-Smith: Okay, are we ready to summarize them? Okay, so we've agreed that our importance criteria as we look at measures and do our Delphi saying, "Yes, they're absolutely important," 7 to 9. "They're of questionable importance," 4 to 6. "They're not important," rank them 1 to 3. Yes?

Female Voice: I guess it's not up there, but I think—I just want to make sure it's implicit. Is that—it responds to Congress.

Rita Mangione-Smith: Responds to Congress?

Female Voice: Right. That's kind of important. But it—it kind of meets their needs to the extent it can. We—given all the limitations.

Female Voice: Stay consistent with the legislative intent.

Jeffrey Schiff: We're at 2:10—

Rita Mangione-Smith: Okay, so I'm going to just quickly summarize, Jeff.

Jeffrey Schiff: Okay.

Rita Mangione-Smith: So we agreed that we will look for the measure being actionable and specifically by that we mean if you're not, if your State doesn't perform well on a particular measure, that there is some action you can take to improve your performance on that measure. Do I have that correct? Because there was a lot of questioning about what we meant by actionable; that the cost of the condition to the health care system is high; that there needs to be accountability of the health care system to fixing whatever problem is identified by that measure. We want to consider the extent of the quality of care problem, and I think implicitly in that is that there is a large problem.

Okay, when you're thinking about extent or degree of the problem, we should consider variation and performance. We should think about are we getting representation of classes of problems that are important for us to include, so maybe Sentinel measures for lack of a better thing to call them. Does the measure address disparities in care—or not address it, but is it important to think about disparities with this measure, and will it help us understand disparities and be able to act to reduce disparities? That if the measure has the ability to transform the health care system—that's a tall order.

Male Voice: That's what we're here for.

Rita Mangione-Smith: Wow, okay. And that it's important as we develop our group of measures to end up with a balanced portfolio of measures and subsumed within that is the need that we cover the entire pediatric age spectrum.

Female Voice: I think that portfolio doesn't need to do everything.

Jeffrey Schiff: No, we don't—I don't think—right.

Rita Mangione-Smith: Exactly.

Female Voice: It might not cover the entire age group. Are you saying it has to?

Rita Mangione-Smith: Well, I think there was a call from Marina that we need some measures that represent each age group.

Female Voice: Okay, yes, basically.

Rita Mangione-Smith: And that it's consistent with legislative intent. We agree on these. Is there any other—Kathy?

Kathleen Lohr: I think—I don't mind having cost in there, but the notion that the operative variable is cost to the health care system I think is a poor one. And because health care systems can be misinterpreted as meaning systems of plans, systems of hospitals, and so forth, there's some—

Rita Mangione-Smith: To society?

Kathleen Lohr:—semantics there, but it's the cost to the Nation that's an issue, not the cost to the health care system. [Group cross-talking]

Rita Mangione-Smith: Wait is that—am I getting dissent, or am I getting—?

Kathleen Lohr: Well, that's my—I don't understand cost to the health care system. That's gameable [sounds like].

Rita Mangione-Smith: Wait a minute. Marina, we were just talking about—

Male Voice: Are you talking about segments?

Female Voice: We were just talking outside about return on investment which is the important—and Marlene has brought up the important—some people in Congress are interested in whether we can save some money, and how it can just—

Female Voice: But I'm not saying we're limited to that—

Rita Mangione-Smith: I don't think you're saying different things.

Kathleen Lohr: Yes, she's not saying that. Including to the system, it's just not limited to—

Rita Mangione-Smith: Correct.

Kathleen Lohr: Right. It's just not correct.

Jeffrey Schiff: But, Cathy, is your comment really directed to that—the issue of gaming within the system? We save inpatient but we increase home care?

Kathleen Lohr: It's just the cost to—of dealing with health care conditions to a health care system does not get at all the problems that might be out there that are never dealt with inside the health care system. So you're really concerned with the costs. If condition there means health conditions and the quality of the care that might be given for those conditions, and you're only concerned with the costs within the health care system, I don't think that's sufficient. Because I think there are downstream costs to the Nation as a whole of not dealing with quality of care problems at early stages. I mean take dealing with prematurity or something.

Jeffrey Schiff: You bet.

Kathleen Lohr: It might be a substantial cost to the health care system, and if you could reduce it, that might be fine, but if you didn't reduce it, 5 years down the road, 15 years down the road, and those kids are alive, there may be a substantial cost to the country of not having dealt with that problem.

Female Voice: You want to call it potential for being the cost curve?

Female Voice: Yes.

Male Voice: Where did I hear that phrase before?

Kathleen Lohr: Actually, where I'm coming from is that if you're concerned with cost-effectiveness the way it's normally thought of by economists, you want to take a societal perspective rather than a health plan or health system perspective and that's where I'm coming from on that.

Jeffrey Schiff: Right. I think let's finish up these comments on importance criteria, if you have any specific like—and then we should—

Rita Mangione-Smith: Yes, we need to move on.

Jeffrey Schiff: We should get the—the sense of the group.

Female Voice: One last thing that doesn't show up here as I re-read the list, we could be talking about adults. And there's nothing here that says, "reflects the unique development—you know, child health perspective of how children are different, they develop" and so on. So I—I mean there's nothing here. You could apply all of those measures to adults, all of those criteria for importance.

Rita Mangione-Smith: But is that—

Female Voice: Except, well, no, because the adults say that they have young adults, they have adults, they have the elderly, they have the frail elderly [cross-talking]

Rita Mangione-Smith: So how would—?

Jeffrey Schiff: So you're saying—

Female Voice: But I'm not sure that's a problem. It's a given, okay.

Jeffrey Schiff: But you want to put developmental impact as an importance criteria?

Female Voice: No, no, no. It was more the—the sort of a whole thing of the "gamishness" [sounds like] of the four D's and all that. If this is the lens we're going to decide on, really cutting down to 10 or 15, I just want to make sure it's in the back of our minds that it really does reflect children's health—

Female Voice: Well, I think there are some things that do transcend and so if you take health care-acquired infections or medical records, there's nothing unique about children in those things.

Female Voice: No. I'm not assuming that every measure is going to meet every criteria, right? Or are we?

Female Voice: No.

Jeffrey Schiff: No.

Female Voice: How about if you just handle it—how about if you just handle it by putting it in the title of whatever this section is?

Female Voice: Okay. If everybody feels—

Jeffrey Schiff: I think we'll have a chance to—

Female Voice: Again, I thought it was implicit, but this is going to be read by lots of people who don't think about kids.

Jeffrey Schiff: Right. I think when we get to writing to our core set, we will have that—

Female Voice: Somewhere.

Jeffrey Schiff:—conversation, you know? So I'm not trying to put it off for everybody. We'll get there. Okay, other comments about importance criteria? Okay, then—it's 2:20—

Female Voice: We have till 3:30 to get through all of these.

Rita Mangione-Smith: We discussed the ones we all want to discuss, so if they're ones you're totally comfortable with—if you have comments to make in relation to importance on any of these, okay? And Doreen just said some of these are not measures. I want you to pull out your measures sheet if you have it because these got abbreviated because we couldn't fit the whole thing on there, okay? So they do correspond to measures that are in front of you.

Jeffrey Schiff: Right.

Rita Mangione-Smith: Our idea was you would look at each of these. If you had comments you wanted to make, either saying that, "I really think this is important and here's why," or, "I really think this is not important and here is why." This is your chance to influence the group before everybody puts their votes in. Does that make sense?

Female Voice: [Inaudible] also has been told with the question marks [inaudible].

Jeffrey Schiff: Okay. I think there's a lot of—Let's do 5 minutes of study hall; people will look the list. Is that okay before we start discussing them?

Female Voice: Yes, just one quick question. I'm looking at the validity thing. Just remind me. If it has a question mark anywhere near the name on the validity score sheet, did that mean it was not a real measure that we know of? Is that—that piece is hard for me to remember what was real and what's not real.

Rita Mangione-Smith: To be clarified in the next round, I promise. In some cases—I think in most cases where there's a question mark, it was either questioning whether the measure was applied to children in the States that reported they were measuring.

Female Voice: So the specs were specific for kids?

Rita Mangione-Smith: Right, we don't know. So there's like diabetes, okay, that has a question mark, so we weren't sure whether it was measured in kids or not. There are other ones though, like under coverage duration, where no measure is identified as being in use.

Jeffrey Schiff: Okay, I think the other thing I'll say is we want to talk about importance now and not about validity and feasibility. So we'll try to [cross-talking] it'll spill over but—

Rita Mangione-Smith: [Inaudible] sure. Why don't we, as a group, just vote for [inaudible] would like to say about a particular measure? Okay. Is that all right? So frequency and ongoing prenatal care.

Female Voice: [Inaudible]

Rita Mangione-Smith: No, no, no. We're just throwing in what everybody has to say. That means for [cross-talking].

Female Voice: Today or later? [Cross-talking]

Jeffrey Schiff: Right. And the scoring will be the 1 to 9. Okay.

Rita Mangione-Smith: Just like you did the rest of the Delphi, okay?

Female Voice: Instead of the higher it may go.

Jeffrey Schiff: No, four or higher to make it—

Rita Mangione-Smith: [Cross-talking] 7 to 9 is very important, absolutely important. A score of 4 to 6 is unclear on importance, and 1 to 3 is definitely not important, okay? All right. So frequency of ongoing prenatal care is the first one; 69 percent statewide average with some of the information. We have the range across plans; it's 53 to 86 percent.

Female Voice: And one we also discussed putting back on the list for the next round of voting might be adequacy of prenatal care, the Kotelchuck Index.

Female Voice: We'll bring that back later.

Female Voice: Go back to make sure.

Female Voice: So I'm just mentioning that now in terms of thinking about—

Rita Mangione-Smith: Okay, any other comments [inaudible] to move on. Okay. Smoking cessation and prevention. Okay, we have one study that showed that 56 percent were counted during prenatal care. I don't know if this was—we have the question mark because we weren't sure who was being measured.

Jeffrey Schiff: Right.

Female Voice: And we went over that yesterday as the measures that were actually broken out where there was smoking cessation, the type of visit, and then there was one per child in this.

Rita Mangione-Smith: Right.

Female Voice: So which one are we working on?

Rita Mangione-Smith: So the measure was—I think those were examples. That's one of those cases where Denise gave us examples where studies have been done that showed an effect of counseling in those groups.

Female Voice: This just gives [inaudible].

Rita Mangione-Smith: Right. So it gets a U.S. Preventive test versus a grade "A" as—

Female Voice: Really good, yes.

Rita Mangione-Smith:—effective during counseling. The measure itself when we gathered information from the States just said smoking cessation and prevention. So we know it's effective for women in prenatal care. We know it's effective for adolescents and—

Female Voice: For adults?

Rita Mangione-Smith: Oh for adults, not [cross-talking]. So, insufficient for adolescents. Okay, so it's an "I" rating for adolescents; it's an "A" rating for adults and women who are pregnant.

Female Voice: And just on the importance thing, this is one of the single most preventable causes of low birth weight, and in Ohio and Northern Kentucky, we have 40 percent smoking during pregnancy in our Appalachian population, so.

Rita Mangione-Smith: So it measures ability to transform the system.

Phyllis Sloyer: So I'm actually worrying in this one that there may be treated with measures, just the concept. I guess that's the question I'm asking [inaudible].

Rita Mangione-Smith: We're trying very hard, Phyllis, between now and—

Phyllis Sloyer: It can happen then.

Rita Mangione-Smith:—the next round in [cross-talking].

Phyllis Sloyer: Some of these should actually happen [inaudible].

Jeffrey Schiff: Right, exactly.

Rita Mangione-Smith: Exactly. We tried really, really hard.

Denise Dougherty: We went to war with the army we had. We scheduled a meeting, and then we got as much information as we could.

Jeffrey Schiff: It's a good way of putting—

Denise Dougherty: And later on, we'll talk about how we really got the information that would be really useful for the final discussion.

Jeffrey Schiff: Right.

Rita Mangione-Smith: Okay, any other comments on that one?

Female Voice: For what? I don't know if this is helpful for you to know or not, but the bills that are moving in the House and Senate both address this issue in the prevention arena. It's huge, and they're nudging States in the direction of doing both counseling and pharmacotherapy for pregnant women.

Female Voice: And low birth weight costs a lot of money.

Jeffrey Schiff: Yes.

Rita Mangione-Smith: Any other comments on smoking cessation? Okay, I'm going to move us on to screening. This isn't an actual measure. We have actual specs. It's an [inaudible] measure. I'm sure many of you are very familiar with that.

Male Voice: Just a quick comment. The Society for Adolescent Medicine is promoting—they have a—they say a valid QI effort that's showing results with what does it cost in a children's hospital, and they're promoting—trying to develop an electronic tool for assessing Chlamydia screening nationwide.

Rita Mangione-Smith: Do we have any prevalence information on this?

Female Voice: We do.

Jeffrey Schiff: There is prevalence.

Denise Dougherty: It is 64.1 percent [inaudible]; 33 percent in variation of [cross-talking]

Rita Mangione-Smith: So the prevalence [cross-talking].

Female Voice: Oh yes.

Rita Mangione-Smith: And then for disease itself, not performance.

Jeffrey Schiff: There are—

Rita Mangione-Smith: There is clearly variation in performance.

Jeffrey Schiff: There are some pretty significant screening data from some—just random screening in the ER that's—I don't know if I want to—that's very high as far as the prevalence, so—

Rita Mangione-Smith: Okay, so we're moving on to number of children diagnosed with rubella per 1,000. I'll draw your attention to the fact that there were 12 cases reported to the CDC in 2007.

Female Voice: I don't know if this is legit so Jeff, shut me down if I'm asking inappropriately, but why did we drop—why did we drop lead screening considering so many States are doing it and—?

Jeffrey Schiff: I don't—

Rita Mangione-Smith: We didn't drop it. It's in the round two Delphi.

Female Voice: It's in a different place, okay.

Jeffrey Schiff: It had a—yeah, it was—it had a different prevalence. These are the only things that first passed validity and feasibility.

Rita Mangione-Smith: And feasibility, and that was on the cost.

Return to Contents
Proceed to Next Section