July 22, 2009: Afternoon Session (continued)

Jeffrey Schiff: Just a couple of things. We only wanted to look at here the ones that had either already—we did not—some were rated as not feasible but were already sort of voted too low, so we are only looking at ones where the feasibility was questionable, and they were already in the discussion we just had, correct?

Rita Mangione-Smith: Right. So if you did not get the—if there was one here where the feasibility looked a-okay, but then we looked back, and it got a 5 on validity or four on validity, we are not discussing feasibility on it. So we cannot—our sort of screening one, we got to pass validity or at least be questionable on validity. If you make that cut, then we will talk about feasibility and commit that then we are going to talk about it [inaudible].

Jeffrey Schiff: Right. So the ones—these are questionable. Do you have a question?

Female Voice: Yeah. I do not want to just keep bringing it up, but remember that—and it goes to the feasibility—to the extent that the child is not in an insurance plan, you are not going to have this data, okay? Some States collect some data, but generally, HEDIS [Health Plan Employer Data and Information Set] and CAHPS® [Consumer Assessment of Healthcare Providers & Systems] are only for kids that are in an insurance type health plan. So if they are in a fee-for-service or if they are in a primary care case management (PCCM) type model of managed care, you are not going to have it. So you are going to leave a lot of kids out of these counts.

Jeffrey Schiff: So is that true, or is that—?

Rita Mangione-Smith: I'm actually going to make one brief comment.

Female Voice: It is pretty true. [Cross-talking] some States have do have some type of HEDIS-like data they collect like New York does on their fee-for-service, and I know Alabama, but you cannot assume that you will have this for all children.

Jeffrey Schiff: But is that not our goal?

Female Voice: That is why I keep bringing it up to understand HEDIS reporting, CAHPS® reporting are insurance modeled managed care reporting systems.

Jeffrey Schiff: Okay, I see, as opposed to [cross-talking].

Rita Mangione-Smith: Ann, I have a question for you though. Do you think that it is not logistically feasible to expect that States any time in the near future would have access to the types of data they would need to score this on fee for service?

Female Voice: I would say it is not feasible if they are not doing it today.

Rita Mangione-Smith: Okay. So just—I'm going to ask another question. We looked at the URI and pharyngitis HEDIS measure in the State of California in eight plans that were Medicaid managed care, but then we also worked with the electronic data systems (EDS) people, I believe, who were the State fee-for-service databank.

Female Voice: Yeah. And again, some States are doing it today. For example, I know New York shows the equivalent of HEDIS data for fee for service, but that is not universal.

Rita Mangione-Smith: Okay. And how far away do you think we are from that being possible?

Female Voice: I think pretty far away.

Male Voice: Well, the other issue is if the States even want to do that, is there an incentive to do that? That is why you said we should make it more discrete and more targeted than what you want to use with an outcome discussion and focus—the pharyngitis is doable, but is there a reason to do this in our world other than for it to become a [cross-talking]?

Female Voice: I think States today can't afford to reprogram their fiscal agent to do it.

Male Voice: Phyllis?

Phyllis Sloyer: I happen to be from one of those States where we actually do CAHPS® and HEDIS on the PCCM population as well, but I want to qualify that. Where you see these questions, where it says your plan, families do not get that. So those kinds of questions are going to be very problematic. The other issue is we spend—this is resource-intensive. I mean there are additional resources that the States are going to have to invest in if you are going to do this on your fee-for-service or PCCM population. And in some States, that is the only model they have. They do not have risk-based managed care. So if they are not doing it, there is an investment, an annual investment to collect the information, to hire HEDIS auditors frankly to make sure that the information you are collecting is accurate.

Female Voice: And probably they are not going to make that investment right now because they cannot afford to. You are down to four people—I mean States are in very bad shape.

Female Voice: And the thing that I wanted to comment on is it depends on the measure, on the data source. If you get into measures that are going to require surveys, medical record review, I can tell you it is not feasible. And this is kind of how I look at it. There are things that I notice in our system that we may not be pooling and reporting now but I can get to. And I think that with other States, there is information within their data systems that would not necessarily require reprogramming, but it is a matter of getting the specifications so that we know what we are pooling. And that is claims data, okay? So that if you look at that information and you determined it, but if you came out now and see it, well, you have to put in a CAHPS® survey to do that, I can tell it is not going to happen next year, it is not going to happen the year after that because we already know what kind of budget problems and issues we are in. And we expect things to get worse, not better, okay?

So when I looked at this, I looked at feasibility being from the standpoint of was it in the system currently. Would it require just some major things of looking at technical specs in order to pull it out of the system, or would it require us to have to go get people and contracts and Requests for Proposal (RFPs) in order to do it? If it was on the other side, then I might not—just really, really low because it is not going to happen.

Jeffrey Schiff: I think this gets to—you know, what Carolyn talked about, too—are we looking at how far do we go; it is a voluntary reporting. I guess I understand what you are saying, Ann. I also feel like it is our job to create a model for measuring the quality of child health care in this country, and so it may be that a lot of things cannot happen now. I think this is part feasibility and part importance as far as how and what we measure because it is voluntary, and I mean I know that people worry voluntary will not stay that way, but I think this is an important issue.

Female Voice: It is. Some people [inaudible].

James Crall: I think Lisa's comment earlier about taking into consideration and being as explicit as we can about there are different levels of burden for different measures, but I would hope we would not constrain ourselves into what can be done in the darkest of times.

Rita Mangione-Smith: Yeah. I'm going to orient us to the whole feasibility piece just for a minute. There are many—fewer measures to look at this time because we are just—I guess it is not true. It is the same one we just went through. So we are talking about measures that had a median feasibility score of 4 to 6, so uncertain feasibility. Four or more subcommittee members ranked the measure in the bottom four. Among the median of 4 to 6, two or fewer subcommittee members put it in the bottom four for feasibility. So again, it is trying to look at those measures with questionable feasibility. These still pass. Interestingly, we did not fail one measure on feasibility, which was stunning to me. But there were some brave souls who put themselves down in that 1 to 3 range, and I think you probably heard from some of them [inaudible]. I know I had some in the 1 to 3 range.

I'm not going to do breastfeeding promotion because, again, remember that is not a real measure, so we are going down to obesity related. You can find that one, the weight assessment and counseling for nutrition and physical activity. Again, in the next round, this will be broken out into three measures; assessment will be one, counseling for nutrition will be a second one, and counseling for physical activity will be a separate one. Here, you can see that the overall feasibility was put at 6, but there are six members of the subcommittee who were in the bottom four categories with feasibility. So this is your opportunity to talk about why specifically this measure may not be feasible or why you think it is feasible.

Rita Mangione-Smith: Cathy?

Cathy Caldwell: We just simply would not have the data. I mean we are all fee for service, and so we are not going to have this information, so we would have to set up some system to collect it.

Rita Mangione-Smith: Okay. And I'm assuming this is—is this medical records-based data? It is a chart review measure, okay.

Female Voice: There are no codes for claims data for any of these three—for weight assessment—or anything for billing purposes, let me put it that way.

Denise Dougherty: According to a HEDIS cheat sheet that you actually do not have—well, you did get it but—it says that the data reporting method is administrative plus medical records.

Female Voice: A hybrid measure.

Cathy Caldwell: And there is probably going to be a diagnosis code related in some situations but any additional detail, maybe not at all.

Rita Mangione-Smith: I assume the assessment part, the BMI part must be chart reviewed, yeah? Is that true?

Jeffrey Schiff: There is an obese—

Rita Mangione-Smith: We have our National Committee for Quality Assurance (NCQA) person in the audience.

Cathy Caldwell: And there is no CPT (current procedural terminology) code for it?

Female Voice: CPT II is not paid for.

Female Voice: I do not believe that there is a typical CPT code. I think we have CPT category II codes that can be used for that. Is that not right, Mary?

Mary McIntyre: That is correct. The thing is, and I was one of the ones that rated this as a 6, and the reason I did it, we do not currently do it, but I was thinking electronic health record (EHR) environment, this because of all the discussion with the pediatricians and the need from the standpoint to be able to capture BMI, so that it is part of the CPT II. The problem with that is it is not statewide.

The other issue is we are looking at the CPT category II codes to try to identify whether or not we can expand into those codes in order to capture stuff that we may not be able to get under that that would really require not major changes when it comes down to system changes. I have already asked the system people what it would require in order to do that, and they see it as something as being doable. It is a matter of getting it on a list and the commission prioritizing it to get it done. So I went ahead to do that because I think it is something that we really need to do, and I'm going to say this, maybe with it being on the list with actually something, it would help those States because I do not see us doing chart reviews, and it never happens. But I think that we can have a way with the stuff that is going on with HIE, and what we have from the standpoint of CPT codes to move in that direction to get it done.

Male Voice: Our current EHR has a page on it that would allow this to be collected. It is voluntary now. We are just starting that part of the process, finding it and finding the relationship to it in the coding would be if it rose to the first four code levels of the billing, it would be highly variable. Of course, a lot of your data falls with that same limitation. So I would support it conceptually, but it is almost something you would almost have to footnote to say [cross-talking].

Rita Mangione-Smith: So maybe it will end up being part of that more enriched set of measures?

Male Voice: I think probably because the other issue was pointed as we move into the HIE, the next obvious opportunity is an interoperable relationship with electronic medical record, then it is infinitely gatherable. But that is in best world several years off.

Rita Mangione-Smith: Okay. I'm going to move on from this one because I do not see anybody else's card tipped. So again, we do now have specific measures for screening for alcohol and illicit drug use in adolescents and screening for depression. Thanks to John Klein [phonetic] who sent us the New York measures. These were two that do make it from a feasibility standpoint but maybe worth some discussion because again, I believe—I do not know how New York State is doing it. Does anybody else—whether it is a chart measure, I'm assuming it has to be because it is about screening. I do not know how you would get that from billing data.

Jeffrey Schiff: Doreen?

Female Voice: It is a survey.

Rita Mangione-Smith: Oh, it is a survey, so they are giving a survey to young adolescents?

Female Voice: Right. It is the YACHS, Young Adult Healthcare Survey.

Rita Mangione-Smith: Okay, so it is survey data.

Female Voice: Yes, it is survey data.

Rita Mangione-Smith: Survey database so nobody has to go into medical records. That is a good thing.

Female Voice: No, they report it separately for Medicaid, so it is part of the New York State—

Female Voice: Okay, because Medicaid is in the public health agency, New York, so—but you may not get that in every State.

Female Voice: They do report it by plan.

Female Voice: Plan again.

Rita Mangione-Smith: Go ahead, Doreen.

Doreen Cavanagh: My understanding is that there are two very new national codes. There is a screening code for substance abuse, and there is also a code for brief intervention. These are Medicaid codes, but the State has to choose to "turn on" the codes, so it is up to the State right now whether they use the codes or not, but the codes do exist.

Rita Mangione-Smith: Good. So there may be more than one way to collect this.

Female Voice: This is just a measurement issue, and maybe it is the way I was trained, and I do not know how the YACHS survey, what the question is, but I'm always troubled when you put in a measure with two very different constructs, so you get alcohol and illicit drug use. And I do not know if that has ever come up before, but I have seen that in a couple of cases where they kind of mix those things.

Female Voice: That is not an issue to the adolescent or adult substance abuse measurement development world. As I said, when we developed the other measure, we accepted all diagnoses for alcohol or illicit drug related abuse or dependence.

Female Voice: I was questioning the survey, the aspect of it, if that was [inaudible].

Female Voice: Okay.

Rita Mangione-Smith: It may be separate items on the survey. I have not seen the survey, but maybe we could get that information. John probably has it. I'm sure he can—John Klein. Okay? Are there any other comments about the feasibility on these two measures? Okay.

Our next one is that same measure that we did not like so much because of the 4 months. I'm going to kind of take a chair's prerogative and say let's forget this one because I think we all think there are better dental measures than this one; unless, there is a large objection to that. Okay.

Diabetes care management—I think very early on in the day, this was one of the first ones we talked about this afternoon. We said we really needed to make an effort to collect more specific measures. Again, I do not want to spend a lot of time talking about feasibility because we do not really know what measures we are talking about yet, and I think that will certainly influence our feelings about feasibility, so I think that discussion needs to happen the next time we get together.

So this is back to the HEDIS measure, which again will be presented as a split measure in the next round; initiation and engagement of alcohol and other drug-dependence treatment. You have your cheat sheet there, Denise. Is this a hybrid measure? Is it an administrative measure?

Denise Dougherty: It is an administrative measure.

Rita Mangione-Smith: Okay, an administrative measure so I assume it may be using these codes that Doreen was referring to.

Doreen Cavanagh: No, it would be not [cross-talking].

Rita Mangione-Smith: Oh, looking at visits.

Doreen Cavanagh: Just using the—right, codes in the claims for a visit.

Female Voice: When we spoke about dental, we mentioned a shortage of providers, and the reason I went low on this, there is a huge shortage of services for these populations, and I wonder how that affects any kind of feasibility here or any data that will come back?

Female Voice: Well, I would think it would be a compelling reason to include it because that is exactly what we would want to show. It is that we have an access problem and look at access over time and hope that if this were measured and if it were addressed through technical assistance that that would improve.

Female Voice: Just as a reminder, one of the categories in the legislation that is something that is not typically a quality measure is availability of services, so hopefully at some point we will be able to have some valid measures of availability so that there can be a correspondence between the use measure.

Rita Mangione-Smith: It was available and did not get used or it is not available so that is why—right. Jim?

James Crall: And not about this measure but because dental was mentioned, I mean there is an issue about availability of providers, and there are various levels of that. It is—are there providers and then are there providers who are participating in the plan or the program that is going to provide the services? And those can be two very different things. I think that is why it is important to measure.

Female Voice: If I could just add two things. One is that this is going to be even an more important issue with the implementation of the new Federal Parity Law which takes effect January of 2010, and there will be a need to monitor availability and access to substance abuse treatment for adolescents, very much so. That is number one. Number two, just to make the point, is that this measure was also endorsed by NQF.

Male Voice: We were actually looking at that as a concept in reporting in one of our annual reports, and actually the only way you can really demonstrate that access is available is by doing a survey and a secret shopper survey on top of that. At least that is the only way we could find. So back to the probability of it happening and the relative burden on those reporting, it would be significant in that area.

Female Voice: Could I clarify why you are referring to a survey when we are discussing this measure?

Male Voice: To demonstrate the outcome of whether it was accessible or not, the way to gather that would only be via survey using a secret shopper to see if in fact the provider was available, a part of the program, and also accepting patients.

Female Voice: Okay, that is not what this is measuring.

Rita Mangione-Smith: So in other words, when people who did not get a visit, you would then through the survey determine why that happened.

Male Voice: Well, a yes or no is—is dentist X receiving and taking new patients or a behavioral health provider or a primary care practitioner or whatever?

Rita Mangione-Smith: Linda?

Linda Lindeke: Well, there are other measures, in the emergency room which is where I work, how long these people sit in the emergency room and where they end up going. Sometimes to other States and so forth, so there are measures. They do not get the headlines like the secret shopper, I guess.

Jeffrey Schiff: Yeah, and we should look at those if there—

Female Voice: This brings up a question I have had in the back of my mind. To what degree here are we limited to what States might collect? Because I mean one of the ways of getting at this—it is not perfect—but it is to ask providers if they are accepting Medicaid and CHIP patients.

Male Voice: That is only good for the day and the time you asked them.

Female Voice: True, but I mean it is a measure where you can look over time and you can compare States. It is a relatively cheap thing to do. I mean it is cheap compared to claims data or chart abstractions.

Female Voice: You may get [cross-talking].

Male Voice:—quarter million dollars through the survey, too.

Female Voice: Yeah, and you may get a real undercounting because a lot of times, providers are willing to take a Medicaid client, but they do not want it advertised, so they may say no. [Cross-talking] I mean we have done these kinds of surveys of pediatricians on the State-specific level. So I mean I just bring it up as it is possible to do, and I'm not saying that it would necessarily be the State's burden, but could it be something that would be at the Federal level?

Jeffrey Schiff: I think there are challenges about what that means as far as outcomes, I guess, at that survey. Phyllis and then Jim, I think.

Phyllis Sloyer: I do not read this measure to be a measure of availability of providers in its entirety. There are reasons beyond availability of providers why youth do not initiate treatment within a certain time period. Some of this may be helping them understand the importance of it, navigating them to the treatment. So this measure, I do not think of as an availability measure, and I do not want to mix it. I mean I think there is another one that deals with that, and you certainly want to look at the combination of those, but I define this a little bit differently.

Female Voice: I'm not suggesting this is an availability measure. In fact, we need availability measures.

Rita Mangione-Smith: To understand the results of this measure, right.

Female Voice: And actually we are having a paper done under contract to look at various availability measures.

Male Voice: Jim, did you have a comment?

James Crall: Well, again, mixing the dental piece back in, but my understanding based on a very brief conversation with Conan Davis at the Centers for Medicare & Medicaid Services (CMS) is that there is some active discussion about hosting a list of providers who are either current providers or willing—

Female Voice:—who we know. It is in the law.

James Crall: Yeah. So I mean the whole notion about whether or not it has to be done through a secret shopper or whether or not there might be some other sort of mechanism that would drive gathering in that sort of information.

Female Voice: I think what is going to be reported is the providers that are on the fiscal agent site or in the plan directory. It does not mean they are going to take new clients.

Female Voice: That is right.

James Crall: But I think there is some discussion about—

Female Voice: And there are some significant concerns.

James Crall:—whether or not it needs to go [inaudible].

Female Voice: There is significant concern among the Medicaid directors that we share with CMS that this is going to result in dentists dropping out. And CMS is committed to us. We are actually keeping track of counts so that we can demonstrate if indeed they do drop out, which is clearly an unintended consequence of the CHIP law, we are going to try and get it repealed for that reason.

Female Voice: I would just like to bring us back to this measure though because I think we got far afield. It is not an availability measure; it is a measure of the provider's ability to attract—address your issues about getting there, using what we would call motivational interviewing or other motivational skills. This is a clinical skill-based measure of are you able to help the child initiate treatment and keep the child in treatment, which is the engagement. And those are clinical skills that we are measuring through this particular measure.

Jeffrey Schiff: Let's just take Linda's comment on this and let's try to—

Rita Mangione-Smith: We need to move on.

Linda Lindeke: Well, now, I'm really confused about what we are going to rate because to me, this is feasibility of measurement, not feasibility of provider skills. No?

Female Voice: This discussion is feasibility of using this measure, and it is feasible because it is claims-based. And there is a bill. For every service, there is a bill. And what we are looking at here is, is there a—once the child has been assessed as having a problem, and that claim comes in, and there is an assessment, and it shows that there is a substance abuse diagnosis, then does the child get to another visit within 14 days so that you initiate the treatment? And then, a second measure—does the child get engaged in treatment? And that is measured by two more visits within 30 days, and that is easy to measure because you have either a claims-based or you have encounter data in which you can follow those bills and those claims in the database.

Jeffrey Schiff: Paul, a quick comment.

Paul Melinkovich: One question related to this. Is this linkage from a primary care diagnosis of substance abuse disorder to treatment with a substance abuse treatment code, or is it diagnosis in the substance abuse setting?

Female Voice: It is the latter, yeah.

Rita Mangione-Smith: Okay, I want to move us on to the next feasibility issue which is one of the dental measures. It is the one about members who receive any dental service over a 1-, 2-, and/or 3-year period, and it applies to members continuously enrolled for 1, 2, or 3 years. For children enrolled in multiple years, calculation is based on the longest period the child was enrolled in the plan. So we have five people putting it in the lowest four categories. So it is one that got tagged for discussion. I wanted to know why those five people thought it was not feasible to collect this.

Male Voice: That seemed like to me to be a very basic paid claim data scheme. I thought that was one of the more easily—

Rita Mangione-Smith: Like an easier measure to you?

Male Voice: Yeah.

Rita Mangione-Smith: Okay. Some people must have thought it was not, though, in the way they ranked feasibility.

Cathy Caldwell: I think I ranked it pretty high, but to answer you concern, we would have a really difficult time segmenting out the individuals that have been enrolled those 3 consecutive years if that is what this is saying, so that would be a limitation and actually the denominator.

Male Voice: Yeah, it could be a limitation. We generally use 3 rolling years of data. We were using 2. So I know some States that of all of the things you would want to change, changing from 2 to 3 years would be relatively painless.

Cathy Caldwell: But even with 3 years of data, I think this is saying that the denominator would be only of those who had those 3 years of continuous enrolment.

Rita Mangione-Smith: No, it can be for 1, 2 or 3.

Cathy Caldwell: Okay.

Male Voice: And you just write an algorithm when you did the ad hoc to [cross-talking].

Rita Mangione-Smith: Calculation is based on the longest period the child was enrolled of 1, 2, or 3 years. Does that make sense? Are there any other comments on this one?

Family experiences with care, patient satisfaction with care. It makes it from a feasibility standpoint. It made it from the validity standpoint, but there are four people in the 1 to 3 range; very unusual for feasibility for this group to be in that very low range. So I would like to hear from people what your issues might be about feasibility for collecting this data.

Female Voice: I thought you said a 5 rating for feasibility, so it is [cross-talking].

Rita Mangione-Smith: No, it passes feasibility and validity, but there are a fair number of people down in the lower end.

Male Voice: Right.

Female Voice: Yeah, most States do not do CAHPS® surveys for the fee-for-service world. New York I think does, but that is a really unusual thing so again, CAHPS® is more when you are in an insurance product because there are lots of requirements from CMS on what States have to do when they put somebody into an at-risk insurance program. So, if you are in fee for service, you do not do CAHPS, and so you can't get this from any State that is in fee-for-service. And as Mary pointed out, some States are all fee for service, and some are mixed. There are rural areas with fee for service, and there are urban areas on managed care. I think you will be missing a lot of data.

Rita Mangione-Smith: Denise, I just have a quick question. Why is this separated out from the HEDIS/CAHPS® survey measures?

Denise Dougherty: This is one of those that was on the CMS survey list, and it was not clear whether it was for kids or not, and it is labeled "patient satisfaction with care." And the CAHPS® people are emphatic [cross-talking].

Rita Mangione-Smith: About not calling it a satisfaction measure.

Denise Dougherty: They are not calling it a satisfaction measure so [cross-talking].

Rita Mangione-Smith: Okay, I just was trying to clarify whether it was CAHPS. I'm assuming it is.

Female Voice: Actually, I believe this is from our managed care summary reporting, basically what that is, it was a general question to States, so it could have been actually captured in any number of ways, so it is an eclectic measure that does not have a specific set of specifications.

Rita Mangione-Smith: Okay.

Female Voice: Oh, just that there will be a great deal of missing data, and the burden of capturing this is very high.

Male Voice: Okay.

Rita Mangione-Smith: Any other comments? I think that might—well, no, we are going to skip that because we rated it incorrectly the first time around. We want to rate it as a composite the next time around. Okay? So I think we are done, and it is 5 o'clock.

Jeffrey Schiff: Wow. Thanks Rita.

Return to Contents
Proceed to Next Section