July 22, 2009: Morning Session (continued)

Female Voice: I would also support what Marina said, and I thought about having your initial small core set that was very easy to implement because implementation is the key here. If you cannot implement, you are not going to get voluntary reporting, and the reality is we will not have the data to look at to do anything with.

Then the second level, the way I would look at the second level also, though, was tying it in with the American Recovery and Reinvestment Act (ARRA) legislation, which says that by 2014, there should be at least a health information exchange on a national basis. And some of the things that we have talked about in the past 24 hours involved significant data extrapolation from the medical record. Those are the things that you put off until you can pull it from the medical record electronically because the reality is nobody -- health plans or States—will be able to pull all of that data from the medical record if all of our metrics include that kind of labor-intensive work.

So I would agree that you come up with a very small set—10, 15, 20 metrics—that is comprehensive enough to meet the initial intent of the law. And that you really think about how we can do that in a way that reflects the overall health and effectiveness associated with CHIPRA over the whole population, not sub-segments of the population.

Female Voice: Two quick things maybe—somewhere in the report you are eventually going to do, would it be possible to add a request or a suggestion that there be an ongoing set of evaluation activities to try to figure out over time whether the measures that one is starting with or the second group perhaps that Marina is talking about in fact do what they are supposed to do? That we see that they are working, they are relatively low cost perhaps to collect, that they are of use back to the States and to providers and plans and so forth and otherwise, just to push on some evaluation of the impact of not only the efforts implied by this legislation but the specific measures—that was one thing.

The other is, and I'm just kind of picking up partly on what you said, I wonder to what extent we want to point to all the work that we will get done in the next 2 years but really actually maybe the next 3 or 4 years using the ARRA Comparative Effectiveness Research? I mean there is going to be a huge amount of information that comes out about effectiveness and comparative effectiveness on all sorts of health care interventions. A lot of it will be pharmaceutical, but some of it will be devices, some of it will be procedures of one sort or another, and some of it will be health care delivery kinds of things. We may have lost track of the fact that within a couple of years between NIH (National Institutes of Health) grants that will be done in a couple of years and the ongoing comparative effectiveness work that AHRQ supports already, there will be a huge evidence base on specific kinds of interventions and their effectiveness and comparative effectiveness that ought to be able to be built into the kinds of measures down the road.

Jeffrey Schiff: We will take Alan's comment, then we are going to I think summarize and finish up this conversation.

Alan Weil: It has been fun listening to the evolution as we go around the room. The conclusion I reached is that in theory, we all want very few measures. And now I look forward to the conversion of that because I—right, so just to make it even worse, since there has been discussion of the—going back to the legislation, I want to challenge somewhat counterintuitively, but I think it makes—I want to challenge the primacy of the importance of actionable on two grounds.

The first is that actionable often means narrow because you have an obvious focal point for doing it, and if we do not want to have very many, then we are going to leave a lot of things that need to change off the table if we focus on actionable. We also know from at least what I have been told of the literature—it is not one that I'm extremely familiar with—that in fact improvement, if you have narrow improvement goals they do not necessarily ripple through. So for example if you say we are going to improve diabetes care and you say, "Well, the best way to do that is through coordination and integration," but then you look at how those same organizations are doing asthma, they are not doing any better so it is not like you have the crossover improvements if you are too narrow.

The second is the statute. A lot of what this is about is the overall picture of quality. It is not just about giving people tools to improve. The core is about helping the Nation understand the status of quality of children's care and that may hopefully lead us to a set of measures that are actionable but, obviously, given the choice between actionable, and not actionable, we would all prefer actionable but I think if we elevate that to too high a principle when we have a limited number of measures, we are not going to achieve some of the important goals.

Jeffrey Schiff: Thanks. Okay.

Rita Mangione-Smith: So we are going to look at all these things that I have put down. I will ask that you briefly look at these before you go to lunch and make sure I captured your burning issues because I do not know if I got them all. And then Jeff and I will work to try to see if some of these are overlapping and summarize them.

Jeffrey Schiff: Yes, so we are going to try to make sure they are unique categories. We have not talked about this, so this is improvisation or [cross-talking].

Rita Mangione-Smith: We are doing this on the fly.

Jeffrey Schiff: But I would be happy to have a couple of folks who have been interested in this process, not everybody, to help us over lunch if you are interested. We are going to have five votes; you can vote all five for one core thing or—?

Rita Mangione-Smith: Right. Or you can spread them out if you want.

Jeffrey Schiff: We will do that immediately after lunch, and I think that will help the focus. I think there has been some seeming consensus around this group—around us. I think it will be good. This is actually a great time to have lunch so people can lobby their points, and then we will see whether consensus is real or not when we come back and see where we go with it because I think this has been an important conversation.

Female Voice: One quick question—somewhere before the end of the day, could you remind us what the process is for August and September, and what happens after we meet in September, and what is facing you in October and some of that? Perhaps we can offer to be helpful somewhere along the line, but I do not have the schedule quite in my head. [Cross-talking] yeah, but even so. And the other question that I had is what happens after—I assume this is a 60-day or 90-day public comment on the report and the proposed measures or something? I do not have a clue what happens after that.

Rita Mangione-Smith: It is not scheduled out of the legislation, so we are [cross-talking].

Female Voice: Right. We are still addressing that, but the way we have understood it to this point is that the Secretary is going to be publishing this core set as of January 1st that will be for public comment, and then that information will be used in terms of the development of the quality measures program. So that is how we are interpreting it right now, but we are actually still awaiting some input from our Office of the General Counsel, but that is the strategy right now.

Female Voice: Jeff, I did not quite understand what the job of your lunch volunteers will be.

Jeffrey Schiff: We need to—I think what we need to do is to synthesize. We have—I do not know. I have not counted them.

Rita Mangione-Smith: It is eight.

Jeffrey Schiff: If there are two things that have been written that are similar, if we do not synthesize these, we will waste our votes and we will be looking at this going, "This is not fair." Actually, I think our job—and I'm looking for volunteers to help because our brains are kind of full, too—is to actually help make sure we synthesize these so we will get down to the right number, so when we vote, we are not voting on overlaps and diluting—so that is what we are looking for volunteers to do.

Female Voice: I'll volunteer.

Jeffrey Schiff: You will volunteer? Great.

Rita Mangione-Smith: How about we say like three volunteers.

Jeffrey Schiff: I could tell you we could take a couple more.

Rita Mangione-Smith: We could get a couple more people, thank you. You will be one more?

Jeffrey Schiff: Great, okay. Good. With that—

Female Voice: I have a question though. These are all sort of theoretical—

Jeffrey Schiff: Right.

Female Voice: How do you vote your top five? Like, keep contacts in mind—I would waste a vote on that because—

Jeffrey Schiff: I think it is a matter—

Rita Mangione-Smith: So some of those might not end up in our synthesized list, and it is not a cut to anybody who said certain things, but if they are not really a voteable thing so we are not going to have some on it. If there are things that you consider voteable that are not up there, that I did not capture, you can add them.

Jeffrey Schiff: I think this is, again, we are advisory to the chair and all that, and this is our chance to sort of see whether that—we are going to just try to confirm the consensus or lack thereof of the conversation we just had so that when we move to talking about specific things, about importance for specific topics, we have some sense of where this group feels like it should go.

Rita Mangione-Smith: If we truly as a group are in consensus that we want a small group of measures, then we have to be ready to take up [cross-talking]. That list is long, and it is going to get longer because we are going to do additional Delphi on additional measures, so if that is what the group consensus ends up being, just keep that in mind when we go to that importance conversation. Thank you. So if there is something you want to add, add it, and then go to lunch.

Jeffrey Schiff: Right.

Lunch break.

Return to Contents
Proceed to Next Section