Arizona Health Care Cost Containment System
Arizona
Conference Room
701 East Jefferson
Phoenix, Arizona
Hortensia Amaro, Ph.D.
Lynette Araki
Kathryn Coltin
Susan
Cypert
Don Detmer
Jason C. Goldwater
Dale C. Hitchcock
Lisa I. Iezzoni, M.D., M.S.
Terri Keagle
Mary Moien
Brent
Ratterree
Alan Schafer
Kathy Valley
George H. Van Amburg
M. Elizabeth Ward
Kari Price and Jan Hart
Terri
Keagle and Brent Ratterree
Susan Cyper and
Alan Schafer
DR. IEZZONI: Thank you for being here, and having us come to visit you. We've been here in Arizona hearing from folks for about a day now, and so you're our last stop, and so we have plenty of questions for you, and we're interested to hear from our last day of hearing from people. So, we're anxious to hear from you.
And what we'd like to do, though, is start by just going around and introducing ourselves, and apologize for Hortensia, who's going to have to leave at 2:00 to catch a flight, but the rest of us will be here through 4:00 o'clock.
I'm Lisa Iezzoni. I'm chairing this session. I'm at Beth Israel Deaconess Medical Center in Boston, and at Harvard Medical School.
MR. VAN AMBURG: I'm George Van Amburg from the Michigan Public Health Institute.
MS. MOIEN: Mary Moien, a contractor, who's going to synthesizing the information from the meetings.
MS. ARAKI: I'm Lynette Araki from the National Center for Health Statistics.
MR. HITCHCOCK: I'm Dale Hitchcock from the Office of the Secretary and HHS. I'm just a staff member on the Committee taking some notes and absorbing a few things.
MS. PRICE: I'm Kari Price. I'm the Health Plan Financial Manager in the Office of Managed Care here at AHCCCS.
MS. HART: And I'm Jan Hart. I'm the ALTCS Manager in the Office of Managed Care.
MS. WARD: I'm Elizabeth Ward from the Washington State Department of Health.
MS. AMARO: I'm Hortensia Amaro, professor at Boston University School of Public Health.
MS. COLTIN: I'm Kathy Coltin. I'm with Hubbard Pilgrim Health Care in Brookline, Massachusetts.
MS. IEZZONI: And, let me just make one comment beforehand. The transcriptionist is indicating that it's a bit difficult to hear us in this room, and we are transcribing this entire transcript, so if people could speak directly into the microphones and try to talk a little bit more loudly. You seem to be right under the flight pattern for the Phoenix Airport.
All right. So, Kari, are you starting out?
MS. PRICE: Well, actually, Jan and I kind of discussed it beforehand, and we don't know how much you already know about AHCCCS, that's why we kind of provided you with an overview --
MS. IEZZONI: Um-hum.
MS. PRICE: -- because we could talk forever on the AHCCCS Program --
MS. IEZZONI: Right. Right.
MS. PRICE: -- and we didn't think that's what you wanted.
MS. IEZZONI: No.
MS. PRICE: And she gave you one specific to ALTCS. And so, what we thought is, it might be better off if you guys tell us what you would want. We looked over your list of questions, and we know generally what you're looking for.
MS. IEZZONI: Um-hum.
MS. PRICE: And we have, you know, some other little handouts that talk about the reporting requirements for our health plans and our program contractors.
MS. IEZZONI: Um-hum.
MS. PRICE: But we thought maybe we'd put it out to you to ask us specific questions you want even related to --
MS. IEZZONI: Are we expecting these additional people to come at various points?
MS. PRICE: Right. And --
MS. IEZZONI: At 2:00 o'clock, there's Terri Keagle, and then a Brent Ratterree, and then at 3:00 o'clock, three additional people.
MS. PRICE: Right.
MS. IEZZONI: I guess one of the concerns that I had in opening up right now with questions, is that a lot of our questions are going to be about information systems, and Terri and Brent might be appropriate people to have in the room to answer those questions.
Frankly, I think we've read a bunch of background material about AHCCCS, as well. And so, I don't personally feel -- do any of the members of the Committee feel as if we need a detailed presentation about the AHCCCS system?
MS. WARD: No.
MS. IEZZONI: No, we don't. So, is there any way to maybe get Terri and Brent here earlier, so they can join you at the table?
MS. VALLEY: Yeah. I think it might be helpful, though, if they want to -- maybe you could talk a little bit about the data that we require from the health plan contractors --
MS. COLTIN: Yes.
MS. IEZZONI: Okay.
MS. VALLEY: -- and just get an overview, because I don't think that's something that --
MS. IEZZONI: Yeah.
MS. COLTIN: That would be helpful.
MS. IEZZONI: Okay. That would be helpful.
MS. VALLEY: -- you probably -- even reading about AHCCCS, you don't get a feel for it. And I'll see if Terri and Brent can come down earlier.
MS. IEZZONI: That would be very helpful. Thank you.
MS. PRICE: This is a list of the deliverables, the reporting requirements of acute contract. So, basically, the acute health plans need to turn in these various items.
MS. IEZZONI: Karen, you can't talk unless you're talking into the microphone. So maybe, Amy, maybe could you help. Okay. We'll just pass it around the table, so -- because we -- unfortunately, since this is a federal committee we have to transcribe everything so it gets put on our Web page and everybody has access to it.
MS. PRICE: And I also just distributed the program contractor's chart of deliverables. This entails probably more information than you would want to know, because it's on every single aspect of our program on dates when certain reports and things are due.
But one of the things I wanted to make you aware, too, is I just this in at the top. We do have a Web page here, and so -- a Web site, so I did go ahead and put the address right on the top so that you would be able to see that.
MS. IEZZONI: We're going to need more copies of this, so --
MR. VAN AMBURG: Yeah.
MS. PRICE: Okay, I've got two more here.
MS. IEZZONI: Okay.
MS. COLTIN: We've got them.
MS. IEZZONI: Do you have them?
MR. VAN AMBURG: Just designate that these are the forms.
MS. ARAKI: Jason, I'm getting --
MR. GOLDWATER: I've got the copies.
MS. IEZZONI: Okay. All right.
MS. PRICE: Basically, all divisions of AHCCCS are collecting information throughout the year from the health plans and the program contractors.
If you're looking on this page, you can see, basically, the financial reporting requirements, and that's what my unit is mostly responsible for, and just the oversight of their viability. And we have a reporting guide that we collect the information from.
We also collect encounter information for all the services they're reporting -- or, providing to the members. That's basically what Terri and Brent will talk about.
We get quarterly grievance reports. OMD collects all types of information on dental, immunization programs, and they're also doing quality indicators, which they'll talk about later on, also, I believe at 3:00 or 4:00.
And then we collect fraud and abuse information, too.
Basically, I mean, if you want to ask specific items, you know, questions about these items that we can collect, we can answer those, or those coming in later can answer them.
MS. HART: And, as Kari said, the ALTCS Program basically mirrors the acute care program, because we do have acute care services, in addition to long-term care services. So, so many of our reporting structures are the same, including, we even have pregnancy termination, even though we're long-term care, and we don't have a lot of pregnancies, we still have to report those, so it's usually zero.
But the only -- you know, a couple of unique things, or one unique thing to the chart of deliverables for the long-term care program, as you can see on Page 3, we have something we just called clinical conditions.
But one of the things we have, like some people have asked me why we have the annual pulmonologist evaluation of vent and dependent members, and what we were doing last year, is our ventilation -- ventilator dependent population has grown so much over the last few years, and one of the things that we felt, is that since that's a small population that we have, once people are ventilator dependent, folks sometimes don't get very aggressive, or don't get creative in treating vent members, and look at the possibility of weaning the vent members, so we started a ventilator dependent weaning program.
And one of the requirements, then, was to have an annual evaluation. So, that's just kind of an unusual thing that's on there, and I've had people ask me why we require that, and it was just to -- that's a requirement in our contract now, to see if there is anyone who can potentially be weaned off of the vent.
MS. IEZZONI: And, have there been?
MS. HART: We just started it this year, so we've been -- we had worked with getting some criteria from Vencor, it's a hospital here in town that specializes in ventilator dependent. And, as a matter of fact, we worked with them before we started the program. But we had to get our policies and procedures in place, and OMD developed policies to go along with that program.
MS. IEZZONI: OMD?
MS. PRICE: Office of the Medical Director.
MS. IEZZONI: Okay. So I guess we should be asking you questions about these two documents, then, is that kind of the plan, for the next --
MS. PRICE: You can ask whatever you want.
MS. IEZZONI: Okay.
MS. PRICE: This was kind of a starting point to let you know what is out there that we collect. You know, we can generate all types of information from the encounter and claims information that we have, and we do that often.
If we need to know age-specific, or eligibility groups, specific information, that's what we do, we go pull it from all the encounter information that we collect. And on an ongoing basis, we're trying to make sure those encounters are complete and coming in on a timely basis, and they're accurate through our data validation studies.
MS. ARAKI: Now, are these data collected here, or are you also -- these are also data that are reported to you from your contractors?
MS. PRICE: It's both. We have claims information here for our small fee-for-service population, but then it's all the encounter information for everybody that's assigned to a managed care plan.
MS. ARAKI: And do you notice a difference in the quality of the data that you're getting from some of your contractors, or are they pretty much along the same comparable --
MS. PRICE: I think there has been some differences in the past. Terri can probably speak to that a little more regarding data validation.
MS. VALLEY: With regards to encounter data, data validation, our 2:00 o'clock speakers --
MS. IEZZONI: Will you -- yeah, okay.
MS. VALLEY: Oh, I'm sorry.
MS. IEZZONI: You're just going to say something about logistics, okay.
MS. VALLEY: Yeah. One of them is off-site. And, they know that the meeting's at 2:00, they'll probably be back a little bit before 2:00. But, as soon as Terri gets back, they'll have him come down, but they may not be here until 1:45 or so.
MS. IEZZONI: All right. Kathy, I noticed that there's some information about pharmacy on here.
MS. COLTIN: Yes.
MS. IEZZONI: Would you like to ask a question?
MS. COLTIN: Yes. I was curious. What is the acute pharmacy report that you ask for, is that an aggregate report, or are you actually getting pharmacy claims information?
MS. PRICE: Unfortunately, I'm going to -- to make sure that's answered correctly, I want them to go ahead and answer that later on. Who do they have coming from OMD? Juman Abujbara at 3:00 o'clock, can answer those specific questions. She's from the Office of the Medical Director.
MR. VAN AMBURG: What are your AIDS HIV reports that are due 90 days at the end of each contract year? What are those?
MS. PRICE: We have a supplemental payment that we pay to our health plans for members who are on protease inhibitors, and so they're -- basically, we leave it to them to report to us those members who are receiving protease inhibitors, and we're sending out a supplemental payment to them.
And we're doing an audit of those members to verify they were actually on protease and eligible during the month that they are reporting them to us.
MS. IEZZONI: We understand that your agency doesn't actually conduct the enrollments, that there's another agency for the State that does the enrollments.
Do you collect information independently on race, ethnicity, other kind of sociodemographic variables, or do you rely on getting that information from the agency that enrolls the person into the program?
MS. PRICE: We basically rely on the agency that's taking in all the information. We don't ever collect any more demographic information once they're enrolled with us.
MS. HART: Now, on the ALTCS side, AHCCCS does perform the eligibility for ALTCS. We have -- there's 15 offices throughout the state, and with the ALTCS Program, in order to be eligible, you first have to go through the financial eligibility portion, it's a two-tiered system. And then the second tier is the medical portion, which is the pre-admission screening testing.
As far as data that's reported, then, from the eligibility process, you'd have to check with our Division of Member Services. That's the division that eligibility is under, so they could answer your questions, if you needed specific data on member race and ethnicity. I'm not sure what they gather on that.
MR. VAN AMBURG: Well, do they give you -- the people that are enrolled in acute care, do they give you the data, the demographics to put in your file, electronically? Like race, ethnicity, birth date?
MS. PRICE: Um-hum. Yeah, because we pull information by race. I'm not going to say how accurate it all is, but, yeah, we have pulled information by race and age groups. Right now we're paying capitation, age, sex specific, so we have demographic information out there that we pull from our files.
MS. IEZZONI: Do you gather information about primary language spoken?
MS. HART: I don't think so.
MS. PRICE: I don't know that.
MS. HART: I don't know.
MS. PRICE: I haven't seen that file -- or that field out there, but I'm not going to say it's not out there, because I know we have a lot of information on demographics.
MS. HART: And I'm not sure, either, on the long-term care site, since all of that's entered into our data. There is a comment screen, so they may, you know, add comments like that because, especially if people do need, you know, interpretation services, that they might have that as a comment. But I'm not sure if that's something that they gather on a routine basis.
MS. WARD: If you were going to go to another state who is putting a data system together, and you were their consultants, from what you've learned and where you are, and what you're learning now, what would you recommend that you like about your system, don't like?
I'm trying to get a sense of -- part of one of our goals is to try to get some recommendations for all the multiple states that are going through their own individual efforts to create databases that can get the best information about Medicaid Managed Care, and what would your consulting advice be?
MS. PRICE: I feel we have an excellent system. I mean, really, we don't have too much that we can't get ahold of, and with really technical people that can pull the information for us over there -- I can't even think of a whole bunch of specific items that we would do differently. I would just make sure -- and I don't know if you're just referring to our system, or this type of reporting. Can you --
MS. WARD: Anything from the world you work in --
MS. PRICE: Okay.
MS. WARD: -- which we're not familiar with.
MS. PRICE: Okay.
MS. WARD: What would your -- and what is it about your system that someone else should replicate that would make it as good as yours?
MS. PRICE: We need our systems people to answer that question --
MS. HART: Yeah.
MS. PRICE: -- we really do, because they're the ones that would be able to provide the technical expertise on that.
MS. HART: Yeah. We're basically users of the system.
MS. PRICE: Um-hum.
MS. HART: And with our --
MS. WARD: What do you -- why do you --
MS. PRICE: But even as users --
MS. WARD: -- why do you like it as a user? What is it that you think users in another state would find really important to have there, as a minimum -- as a recommended system that all people should have? What is it about the system that makes it really good for you?
MS. PRICE: I can't even think. I can't even address it.
MS. HART: Yeah. Because the only thing I could think of is, again, you have to have a system that, you know, allows to pull down data.
I think one of the weaknesses of our systems now that we have is individual users can't go in and -- we have no way to really query the system, ourselves. So that's the only concern that I would have, is that, right now, it's kind of a lengthy process to put in a systems request.
So if we just want a different sort of data, or something like that, it's a little bit more difficult. I know that they're looking at some systems changes now that might help us increase that. I'm used to -- I had come from a smaller organization where we could go in and use a query program, we had a download of our --
MS. PRICE: Kind of like a data warehouse.
MS. HART: Yeah, of our data, and it was really nice to, you know, just say, well, gosh, I wonder what would happen if we just tweaked this a little bit. We could just go in and kind of do a what if, and do several scenarios.
Whereas, here, it's a request that we have to turn into the IS Department, and then it may take them, depending on their priorities, because they have a lot more priorities than maybe just tweaking a report, or something, that it may be several months down the road.
So sometimes there's certain parts of -- some information that sometimes may seem important to me, as a user, but agency-wide, it may not be as important for the entire agency. So, I think if anyone can put together a system, if they can keep that in mind so that they have a system that would have that capability of either a download of a database so people could run queries independently of having to burden their Information Services Department, I think that that would be something to consider.
MS. WARD: Okay.
MS. HART: But I don't know at the time when we were bringing up our system, and whatever, they've been bringing it up, that that was even available at the time, so, you know, that's part of it, too. And you know how it is with systems, just as soon as you get your system in, two years later, you know, there's -- technology's growing so quickly, that some of your processes are antiquated already.
MS. COLTIN: How do you evaluate pharmacy costs? You mentioned for the HIV AIDS population that you get a listing of people who are on protease inhibitors, and then you said you have to go out and validate that. I'm assuming you mean through medical records reviews?
MS. HART: Um-hum.
MS. COLTIN: We had heard that there was decision early on not to require the health plans to report pharmacy claims. Has that decision ever been reviewed? When you think about the data you don't have, that you'd like to have, not just the query capabilities but, in fact, the raw data don't exist, is pharmacy information one of them?
MS. PRICE: Um-hum. It comes up. I'd say that comes up often. I mean, they'd like to have it, they just haven't figured out a way to be able to handle that much information, so -- but, Juman can talk more about what they're doing to pull specific pharmacy information, because I know they are pulling it and taking it from the pharmacy, like, management companies --
MS. HART: Right.
MS. PRICE: -- and quantifying that data and summarizing it by utilization and cost per health plan, and overall, and I don't know if it's both for ALTCS and acute --
MS. HART: Um-hum.
MS. PRICE: -- but I know it's definitely for acute.
MS. HART: Right. Again, I think when you talk to the health plans and the program contractors, the majority of those do have a pharmacy company that basically is giving them their costs. Because I know when you go out and you talk to the health plans and program contractors, you know, normally, they're doing physician provider profiling, and they'll have, you know, physician costs because, you know, I -- and I came from a health plan environment, and I know we looked at that all the time, because sometimes you'd have certain providers that might be an outlier, and then you'd have to think, okay, what kind of, you know, patients, or what kind of patient panel do they have, because maybe that explains it, maybe you've got somebody who over-utilizes, you know.
So, you know, I know the health plans not only have that data, but they also do on-site reviews of the different providers. So, you know, that's a way even for them to validate their own information that they have.
MS. PRICE: The question that you just asked, though, too, you should probably ask that of both groups that come in, because they're working with the data even more than we are in our areas, unfortunately. At least they're coming. So they can probably tell you a little bit more pros and cons of our system.
MS. COLTIN: So is your area using the data primarily to look at the financial aspects of program management?
MS. PRICE: Financial capitation.
MS. IEZZONI: Are you thinking about developing a risk adjustment method for setting capitation rates?
MS. PRICE: Well, we've already gone to age, sex-specific, so I think that that -- I don't know if we're planning to go any more specific than that. But I think the health plans were happy to see that, you know, breaking out, you know, zero to one, what is it, zero to one, one to thirteen, fourteen to forty-four, males and females separately, and forty-five and above.
MS. COLTIN: And is that within rating category, like SSI versus --
MS. PRICE: Right. Actually, we just broke out the children in our TANF, Temporary Assistance to Needy Families. And the SSIs have stayed one age group, and the SOPA (phonetic) mothers have stayed one age group, but that just started October 1st, '97.
DR. AMARO: Have you worked on using the data you have on calculating any specific rates for pregnant women who might need intensive services, like substance abuse treatment or mental health services?
MS. PRICE: No.
DR. AMARO: No.
MS. PRICE: It's all covered under the rate that we already have for a SOPA woman.
DR. AMARO: So there's one rate, regardless of risk profiler?
MS. PRICE: Yeah. We haven't done any risk by the type of need that they have, really, besides age and sex.
MS. ARAKI: Now, do you use this to set the capitation rate for AHCCCS here, or also for the contractors who provide services to the Medicaid population?
MS. PRICE: We capitate the health plans, and then they have their own contracts with the actual providers who are providing the care. So we use the utilization information that they submit to us through their encounters, and the cost information and their financial statements that they report to us --
MS. ARAKI: Right.
MS. PRICE: -- to develop our capitation rates.
MS. ARAKI: So you don't use the additional information that they might submit to you with regard to the level of care that they're providing for their different clientele, or as part of the capitation --
MS. PRICE: No. That would be reflected, though, through the financial statements --
MS. HART: And utilization.
MS. PRICE: -- and through any cost and utilization information, if somebody had a higher level of care. Well, level, I guess, wouldn't be necessarily reflected in utilization.
MS. ARAKI: What I'm trying to understand is, does it vary from contractor-to-contractor, the capitation rate, or is it a uniform flat fee contract -- capitation rate?
MS. IEZZONI: Do you mean health plan when you say "contractor"?
MS. ARAKI: Maybe that's what I mean.
MS. WARD: Yeah.
MS. ARAKI: Yeah, because I'm trying to think about the IPA, you know, our presentation that we had this morning --
MS. IEZZONI: And we heard them this morning.
MS. ARAKI: -- where he has their population is --
MS. PRICE: Well, what we do is, we develop an actuarially sound rate range, and the range isn't real large, I won't get into why that is, it's not a published range, and they bid and have to get in that range. If they're under the range, we bring them up to the bottom. If they're over the range, we give them a chance by doing what we call a best and final offer to try and get them into the range, so we have -- so, yes, they have different rates, but they all should be an actuarially sound rate that falls within the range. And that's developed using the encounters and the financial information.
MS. HART: Right. And the ALTCS Program is very similar. We capitate our -- when we use the terms "health plans," when we're talking health plans, we're using talking Acute Care Program. When we say "Program Contractors," then we're talking about ALTCS. That's kind of our way of keeping the two things separate. So you do hear the terms kind of interchanged a lot, and a lot times we'll just say "contractors," because then, you know, that kind of encompasses both program contractors.
And we also have a capitation rate. We have one program contractor per county, because the population just isn't as large, you know, with the long-term care population. However, for our developmentally disabled population, we do have one contractor who covers the entire state, and that's the Department of Economic Securities, Division of Developmental Disabilities. They are the program contractor for that entire population.
And we do -- we have a contract -- a capitation rate with them that basically is a statewide rate; whereas, with the Long-Term Care Program, we have a capitation rate specific to each county, because from county-to-county, sometimes the population can be different.
You know, it's interesting to see the differences. Like, for example, in our Cochise County, and in our Yavapai County, the numbers of the members are pretty much the same, but the populations are so different. In Cochise County, it's a high Hispanic population where there's -- they believe in the extended family and families taking care of their own, and so we have people who are at very high risk staying in their own homes, because that's what they want to do, they want to be with their families; whereas, in Yavapai County, up north, people don't have extended families. We don't need to publish information in Spanish, because we hardly have any Spanish speaking people.
So, when you travel around the state, even, you find some things that can be very, very different, and it may -- you know, even something like that, that sounds like it shouldn't have a big impact on cost, it does, when you've got family members who are helping to take care of the member. So, those are some of the things that you have to think about as you even travel from county-to-county within the state here.
DR. AMARO: That's what I was trying to get at in my question. With respect to populations who may need, you know, different levels of intensity of services, including support services, and I gave you an example of pregnant women who might need, you know, substance abuse treatment services, which, clearly, those women are high end users, and they're going to have a lot of more negative outcomes than women who aren't using, and it's going to impact your cost for neonatal intensive care, you know, et cetera, and infant care later.
So, what motivation is there -- I guess my question, since we're talking about data is, could you use your data to develop those kinds of different approaches to capitation in order to make taking care of those populations doable and attractive to providers?
Because, if you're given the same rate for servicing, you know, a very high end loser who's going to need a lot of specialized services, and translation, and transportation, and substance abuse treatment, I mean, what incentives does that provide for the providers to actually, you know, try to get that market?
MS. HART: Well, and again, all of that data would still come into our databases, and it would be folded into the capitation rate, so a lot of times we don't really identify high risk.
Now, you know, I can give you an example again, in long-term care --
DR. AMARO: But it'll be averaged out, and --
MS. HART: Right.
DR. AMARO: -- providers may have different distributions of those populations?
MS. HART: Um-hum. But then, again, the health plan -- you know, for example, when I was in a health plan, we had a couple of physicians who always looked like outliers on our medical costs, but we also knew that they had certain individuals that they took care of, they ended up with the AIDS population.
So a lot of times you'll find health plans that will actually alter their capitation rate to that physician, so the health plans might be doing some of those adjustments.
DR. AMARO: Okay.
MS. HART: You know, so that's one way that that gets taken into account.
MS. PRICE: The health plans also -- you said something that I just want to address. They aren't going out to get the population, they're assigned the population, so they don't really have a choice on who they take.
MS. HART: Right.
MS. PRICE: So, yeah, there is some outreach, but, specifically, they're coming in through the eligibility offices and they're assigned to the health plans. They don't really have control over taking those members or not taking them because they have too many issues.
MS. HART: And I know that several of the health plans have maternal child health case managers to help, you know, address anybody that are the high risk pregnant members.
DR. AMARO: Okay. Thank you.
MS. PRICE: And, also, going back to your question, where you asked about the kind of acuity level. We also have a reinsurance program. I don't know if you're familiar with -- it's kind of a stop loss program. Once they reach a certain threshold, depending on the size of the health plan, their threshold is different. The smaller health plans have a smaller threshold. The larger health plans have a larger threshold.
DR. AMARO: Um-hum.
MS. PRICE: Once they reach that threshold, we'll pay a percentage of the eligible medical expenses. Typically, they're the inpatient expenses. So there is a stop loss. We aren't, you know -- they aren't responsible for just completely outrageous medical services without us kicking in some.
MS. COLTIN: What proportion of managed care plans that operate in Arizona have a contract with AHCCCS?
MS. HART: Boy, I --
MS. PRICE: Oooh, that's a really good question.
MS. HART: I wouldn't know.
MS. PRICE: I have no idea how many --
MS. HART: I'm not -- I don't even know how many --
MS. PRICE: -- managed care plans there are in Arizona.
MS. COLTIN: I mean, I've been watching television since I've been here, and seen some advertisements for plans that I didn't see listed, for instance, under your --
MS. PRICE: We don't have a private, so --
MS. COLTIN: -- overview, so I was curious --
MS. HART: Yeah, no, there really isn't --
MS. PRICE: Yeah.
MS. HART: Right.
MS. PRICE: -- not very many commercials.
MS. HART: Yeah, the only --
MS. COLTIN: -- whether there were incentives for managed care plans to get involved with the AHCCCS Program, and why some of them may have chosen not to.
MS. HART: You can probably --
MS. PRICE: Well, it's an entirely different population. That's probably why some of them haven't -- I mean, we had a couple commercial plans get into our business about three and a half years ago, or so, and I think they were very shocked at what they found. It was so different than their commercial population, and the requirements were so different with AHCCCS, it's basically really a unique program, and the requirements were very different from anything they'd ever seen commercially.
So the two plans that I'm speaking of, they struggled for a good couple years and, in fact, one of them got out, didn't accept our contract that we had just awarded for 10/1/97. So they stayed in for three years, and lost a significant amount of money and got out.
MS. COLTIN: Now --
MS. PRICE: They didn't have an understanding of what the AHCCCS population was about.
MS. COLTIN: So you think it had more to do with the population and the types of services they need, than some of the administrative requirements that you place? I mean, there's an awful lot of reporting requirements here --
MS. HART: I think it's probably both.
MS. PRICE: Yeah, I'd say it's definitely both.
MS. COLTIN: -- that are expensive.
MS. HART: Um-hum.
MS. PRICE: It's both.
MS. HART: I'd say it's definitely both, and I also think there's a bottom line issue. You know, since we are a taxpayer program, you know, the bottom line isn't going to be as significant as you're going to see in a commercial population.
And they've got stockholders that, in a lot of cases, especially the private-for-profits, that, you know, they have to satisfy their stockholders, and I think, you know, people, they have financial goals, and the bottom lines that are acceptable to us may not be acceptable to the stockholder. So we do have more public entities, or Medicaid only plans that participate in AHCCCS.
MS. ARAKI: Has the Welfare Reform law affected the coverage under AHCCCS at all, or impacted heavily on AHCCCS?
MS. HART: I guess we could -- I was going to say, I'm not the best one to answer it, either. Again, you'd probably, on eligibility enrollment information, and that, if you could -- you may, in the future, want to contact Diane Ross, she's our assistant director for Division of Member Services, because since they do the eligibility for -- since they track all the eligibility and work with Department of Economic Security, who the county -- and the county offices that do eligibility, she could probably answer that more, because she would keep track of that. Because I know when Welfare Reform was first coming about, she was assigned to take a look at that, and --
MS. PRICE: She could tell you if there's been specific impact.
MS. HART: -- track it. Right.
One of the things in your questions that you had asked in your printed questions, talked about any gaps in data collection. And one of the gaps that probably isn't obvious in the beginning is when you've got dually eligible enrolled members with both Medicare and Medicaid, and there's no way that I know of that we can get data, you know, about our members, you know, from the Medicare population, and especially a Medicare risk plan, if they're enrolled.
So that's one area that there is a gap, and I think everyone struggles with that, because I know I've had -- been on the phone a couple of times with people asking how do we gather data differently for that population, and we really don't, because there's no way for us to get ahold of that information.
You know, we do know if people are enrolled in other plans, but a lot of times when they're going on a fee-for-service -- when they're seen on a fee-for-service basis through Medicare, we don't know when they're seeing physicians, you know, we don't know some of the services and some of the costs that they're incurring that, you know, it may be, and I'm sure it is, duplicative, in some cases, that Medicare and Medicaid are both paying for the same things at the same time sometimes with members.
DR. AMARO: Can you tell us a little bit about how you used the data to derive capitated rates for behavioral health care?
MS. HART: Why don't you answer that one.
MS. PRICE: Basically, do you know how our behavioral health system works, how it's a carve-out?
DR. AMARO: We've heard a little bit about it --
MS. PRICE: Okay.
DR. AMARO: -- but it's a carve-out --
MS. PRICE: Okay.
DR. AMARO: -- and we're not sure exactly what's covered in there, but --
MS. PRICE: Okay.
DR. AMARO: -- how extensive it is.
MS. PRICE: The capitation rates used to be a little different. They used to be done based on a per user. So they took the utilization information and the cost information from the Regional Behavioral Health Authorities, we call them RBHAs, and those are contracted kind of like health plans under the Arizona Department of Health Services, that deliver the services to any behavioral health client.
DR. AMARO: When it was under fee-for-service.
MS. PRICE: No.
DR. AMARO: No.
MS. PRICE: No. This is a capitated arrangement.
DR. AMARO: Oh. But I mean before --
MS. PRICE: We were never a fee-for-service.
DR. AMARO: Oh, okay, the behavioral health. Okay.
MS. PRICE: Right. Well, behavioral health was kind of a per user, and we were going back, and we were reconciling to make sure that behavioral health was being delivered, and if it wasn't we were recouping that money.
So, in a way, it was not quite as capitated as it is now. I don't know -- we're doing it now based on the entire population. We used to do it on a per user population. We took all the costs, and now we've spread it out over our whole Medicaid population. So the rate went down significantly, and we pay for -- we say you're responsible for general mental health, seriously mentally ill, and the children's behavioral health population, and pay it out on an entire -- how do I say this -- based on our entire population.
So, we've just used the same utilization information that the RBHAs are reporting through their encounters and through their financial statements to generate a rate, and we spread it based on our entire population, and pay it to the Arizona Department of Health Services, and then they, in turn, pay it out to each one of their contractors who are the Regional Behavioral Health Authorities.
Does that answer your question?
DR. AMARO: Does that cover substance abuse treatment, also?
MS. PRICE: Um-hum. That was all just effective 10/1 of '95, I'd say --
DR. AMARO: Is that like primarily de --
MS. PRICE: -- general mental health, and --
DR. AMARO: -- detox and short-term residential?
MS. PRICE: Counseling.
DR. AMARO: Outpatient, too?
MS. PRICE: It's pretty comprehensive.
But, basically, it's the same as the health plans using encounter information. And we have actuaries that work on the rates, and compared to other states, and --
DR. AMARO: Would it be possible to get a little more information on the behavioral health carve-out piece because, so far, at least when we got --
MS. PRICE: What would you like?
DR. AMARO: -- it, I only got -- just, if you have something equivalent that is a description, a more full description of that.
MS. PRICE: Sure.
DR. AMARO: And, also, you know, what's included in, if there's any contract requirements or, you know, what the contract requirements or language is for those.
MS. PRICE: You want a copy of the contract, maybe?
DR. AMARO: Or, whatever document might give us a sense of what providers are required to --
MS. PRICE: Probably the -- or a policy manual, yeah.
DR. AMARO: Yeah.
MS. HART: I was going to say maybe the RFP would be --
DR. AMARO: -- to provide, and --
MS. PRICE: Right. That's the contract.
MS. HART: Right.
MS. PRICE: Okay.
DR. AMARO: Thank you.
MS. HART: Yeah. And with the long-term care population, the behavioral health is included -- we have no carve-out in the Long-Term Care Program. So, when you were asking about what capitation, again, it's reported through the encounter data.
And, prior to that, they did have -- they were under the RBHA system, and so we did get information from the behavioral health side, and we rolled that into a capitation rate. But now we collect that from our program contractors, as well, through our encounter data system.
MS. COLTIN: So, let me ask a follow-up question so I understand. If you have enrolled someone in the acute program, and that person requires mental health services, the plan makes a referral, but the services get provided under the RBHAs?
MS. PRICE: Right. Right. Except for a small population, 18, 19, and 20-year-olds. We've kept those in the plans, it was kind of a pilot program, I think it started ten years ago, that just has remained, so the health plans are responsible for that small population, otherwise, they're all referred out. They're responsible for maybe the first 72 hours, or some initial part of their behavioral health care, and then they refer them out for ongoing services.
MS. COLTIN: So, is it the health plan that's making the determination to authorize the use of these services, or are they actually just referring someone who then needs to be reviewed under --
MS. PRICE: They're referring them on. They do all their own -- the RBHAs do their own determination --
MS. COLTIN: Determine whether or not the person --
MS. PRICE: -- assessment --
MS. COLTIN: Gets the service.
MS. PRICE: -- of their needs. Right.
MS. COLTIN: Can patients self-refer, as well, as opposed to the health plan having to refer them? Can they approach the RBHA and say I want to see a mental health provider?
MS. PRICE: I believe they can.
MS. HART: Yeah. Yeah. Even in the contract, and that's also written in the member manual. They can self-refer. A lot of times, you know, I don't know that they know the systems that well to know where to refer to --
MS. COLTIN: Um-hum.
MS. HART: -- so I think, primarily, the health plans, and I know especially on -- with the long-term care piece, since every long-term care member has a case manager, more than likely, if a family member, or the individual, themselves, think that they might need the service, they'd more than likely go to the case manager.
I think, in reality, that's what happens, and then the case manager does refer -- you know, in each of the health plans and then the program contractors, they do have a behavioral health coordinator, as well, so everyone is required to have that position in their organization to help assess and get people referred to the right service.
MS. COLTIN: Have you looked at the percent of enrollees who actually receive mental health services? I guess what I'm getting at is, it sounds as if the health plans aren't necessarily controlling AHCCCS, that the beneficiaries could get AHCCCS, themselves, through this program. There's no reason for the health plan to restrict it, because they're not financially --
MS. PRICE: Um-hum.
MS. HART: Um-hum.
MS. COLTIN: -- at risk for these services; and, in fact, if anything, there might be an incentive for them to refer more patients. I know a lot of mental health care gets delivered in primary care, for instance --
MS. HART: Right.
MS. COLTIN: -- in health plans where it's carved in. Not all of it gets referred to mental health, particularly, chronic depression or, you know, acute depression, which can be managed often in primary care.
But it sounds like, in this system, there might be an incentive not to manage it in primary care, but to make the referral to mental health --
MS. PRICE: Not incur the costs and send it on to the RBHAs who are capitated for that service.
MS. COLTIN: Yeah, exactly. Exactly.
So that's why I was wondering if you saw different rates of people using mental health services under this type of a system, than you might see in states that include mental health in --
MS. PRICE: Um-hum.
MS. COLTIN: -- their contracts with managed care organizations.
MS. PRICE: We could find that out for you and include it in the information, kind of how we compare on the percentage of our AHCCCS population utilizing behavioral health services.
MS. HART: Um-hum.
MS. COLTIN: It's one of the HEDIS measures. I don't know if you get --
MS. PRICE: Yeah, I know they're collecting it.
MS. HART: Um-hum. Yeah, we do collect the data, I just don't know off the top of my head, because we do have a behavioral health unit that looks at that.
But that's one of the things, too, with the -- I don't know if you're aware with the health plans, the program contractors, we do go out and we do reviews of each health plan, and they're very thorough reviews. And I know when people have talked about, you know, a lot of different programs, because I know I was on the provider side where we had Joint Commission come and review us, and they're out there every three years.
The thing that I think is unique about this program, having been on the provider side before, is that, you know, you really know the program, because you're out there every single year seeing these individuals. You know the managers, we meet -- the health plans meet once a week, we meet quarterly with the program contractors, so you really get a sense of their operations. It's not just knowing what our requirements are, you really have an idea.
Like I'll know, you know, down in Cochise that, you know, their grievance coordinator, you know, also has like two other duties because it's a small contractor, you know. And you really get a feel for how things are done. And a lot of times when you're out reviewing the health plans and the program contractors, you're really able to compare how different people perform against each other, because you have so much contact with them on a day-to-day basis.
Even when we get calls in from the legislature, or else even complaints from members sometimes, you'll have an idea of if you think things are a valid complaint, because sometimes you know the operations, and a lot of times you can go, well, I know what you're talking about, because this is how this plan operates. I think that's something that's really unique about the program, is the amount of oversight. And I know with AHCCCS, I noticed you had Nelda McCall.
You know, in the Laguna Report, one of the things they talk about is the AHCCCS -- the administrative oversight being so large. And one of the reasons why it is so large, is there is a lot of oversight. And I think, again, that lends to a lot of protection.
And I know, especially speaking from the long-term care side, I always tell people, I say, you know, to be in this program where you've got a case manager who goes out there and sees you on a regularly scheduled basis, I mean, that's really a great safety net, where the private industry doesn't even have that, you know. So I think that, you know, the administrative oversight lends to the safety and the quality of the program.
MS. IEZZONI: This morning, we actually heard from what sounds like a great plan, the Arizona IPA, and the guy, I had to actually stop him when we were asking him about all the quality reporting that he proposes to you when he comes up for the annual review, because there was so much of it.
But I also got a little bit of a sense that some of the big plans might be trying to crowd out some of the small plans. It was just a sense or a hint that I got.
Is that, in fact, what might be happening here? One of the points that he made was that they spend 13 percent of their per member per month budget on information systems, which he said was $3 a month, per member. And the small plans, you know, the tiny plans probably cannot afford to do that.
Can you just tell us a little bit about that shifting dynamic of whether some of the big plans might -- whether you might be heading towards a situation where you're really dominated by a few big plans, that are great plans. I mean, they've got good quality, but --
MS. PRICE: Well, there's --
MS. HART: There's reality.
MS. PRICE: Yeah. It's kind of just the nature of the beast. When they get that large, they can be so much more efficient than the smaller plans. Of course, we don't want that to happen. We want to have a selection out there for the members.
But, when you're speaking 13 percent, that couldn't be, because our requirement is only, they can only have 10 percent on administrative fees. So, I just had to throw that out there.
MS. COLTIN: I think it's the percent of their administrative --
MS. IEZZONI: A percent of their administrative --
MS. HART: Yeah. That's what I want to --
MR. VAN AMBURG: Administrative --
DR. AMARO: It was 14 percent of their administrative costs.
MS. HART: Oh, okay.
MS. IEZZONI: Right. Thirteen percent of their administrative --
MS. HART: I see now. I see.
MS. PRICE: Um-hum.
MS. IEZZONI: -- but he said it works out to $3 per month per member.
(Whereupon, Dr. Hortensia Amaro, leaves the room.)
MS. PRICE: Um-hum. And we watch that. I mean, if their program requirements that cause our administrative budget to maybe go up, that ratio that we, or that percentage we allow, I mean, it used to be seven percent, our rates came way down, administration was still up, and we were having more and more requirements, and we raised it to ten percent.
So we keep an eye on that, and we make sure that it should be possible for the health plan to function with what we give them for administration.
MS. IEZZONI: So how are you trying to make sure that it remains competitive when, as you said yourself, it's kind of the nature of the evolution that the big plans might swamp? Is there a specific policy that's promulgated from the governor's office through the legislature that --
MS. PRICE: No.
MS. IEZZONI: -- that you want to keep, you know, a certain number of choices for plans, or when you say that, is that just kind of a --
MS. PRICE: Well, we still, in the majority of our counties, we want at least two plans. And then in our larger urban counties, I think we're like four in one of them, and six in another. And then we have a couple plans that are capitated that didn't quite perform as well with the last RP as some of the others did, but we didn't want to transfer their membership, just in the best interest of the state, and in the members' best interest.
Where I was going with that.
MS. HART: Yeah --
MS. PRICE: I think if we make the rates fair and we do reinsurance and we adjust it, based on the size of their plan, and we make sure that these small plans are not treated any differently and are compensated adequately, you know, they'll stand. But, at this point, I don't think there is any legislation --
MS. HART: Um-um.
MS. PRICE: -- or anything that is keeping the big plans from taking over the small ones.
MS. HART: Right. And I think, again, it's an HC philosophy of allowing choice, and it's going to be interesting in the Long-Term Care Program, because we had -- the legislature actually mandated that Pima County and Maricopa County be the program contractor for the Long-Term Program, and then three other counties, Cochise, Yavapai, and Pinal were able to exercise their first right of refusal to be the contractor, and that's going away.
So, it's going to be interesting, because October 1st, 2000, it will be a fully competitive program, and already we're starting to work on issues with that and, you know, that's a real concern for some of the smaller counties, you know, again, can they compete when you've got the economy scale, and the things that a larger plan have.
But one of the things, too, is whenever we go out for RFP, both programs, one of the -- we always ask people, you know, is there something that you would like to see included in the RFP, is there something to consider during this contracting time, because there are some smaller players who kind of get concerned about the fact that they don't have the same economies of scale as the larger program contractors.
But, again, there's so many different areas when we evaluate an RFP, it's not just based on cost, you know, it's program, and network, and staffing issues.
MS. ARAKI: You said it was Pima County, and Coco --
MS. HART: Pima County and Maricopa are mandated, and then Yavapai, Pinal, and Cochise have all exercised what they call the first right of refusal to be the program contractor. But it will be fully competitive in the year 2000. But that's for the Long-Term Care Program.
And there is a map in this other handout that does have the enrollments.
MS. ARAKI: Is this the one that I'm looking at, or is that the different --
MS. HART: Yeah.
MS. ARAKI: Oh.
MS. HART: I don't know. Did that have the enrollments for both?
MS. ARAKI: This one has -- well, this is the health plans by county.
MS. HART: Yeah. And, see, that's acute care health plans, so those numbers are not the same. Because that's one of the things you'll find, like, you know, it would be really difficult for us to have more than one contractor in a county. We've got one county, Greenlee County, that's small, it's got 18 members in it.
MS. ARAKI: Yeah, it says --
MS. HART: So, you know, you really couldn't support --
MS. ARAKI: Well, this 658, that's the acute care, and then -- oh, I see.
MS. HART: Yeah. And then long-term care -- and, this, yeah. So, I mean, the population is so much smaller in the long-term care, it's hard to have the same type of competition. No one's going to go in and try to compete for 17 members.
MS. IEZZONI: Do you have much of a presence --
MS. HART: Actually, 22 now.
MS. IEZZONI: Do you have much of a presence with the Native American population in Arizona? Is AHCCCS involved with them, covering them?
MS. HART: Um-hum.
MS. IEZZONI: Can you tell me a little bit about that?
MS. HART: I was going to say, yeah, neither of us are real experts in that area. We do have, in our office, a policy and coordination that Kathy Valley works in, we have a Tribal coordinator who actually works with the Tribes. I know with long-term care, we do have intergovernmental agreements.
You know, there's 21 Tribes in this state, and I think there's 7 or 8 that we actually have agreements with, that they serve the long-term care population. They do the case management, and basically they subcontract out some of the services, but --
MS. ARAKI: Which ones of these counties are the ones that have small, 500 or less lives that they cover?
MS. HART: Members? In --
MS. ARAKI: I think they're the rural areas, right?
MS. HART: Yeah. In long-term care, it's the rural areas. You know, if you go across the top of the state, almost all of them, you know, all of them do have less than 500.
MS. ARAKI: Well, because one of the materials that I read --
MS. HART: It's almost easier to say which ones do have it.
MS. ARAKI: Yeah. But one of the materials I read, I thought it said it offered the Native American population --
MR. VAN AMBURG: A choice.
MS. ARAKI: -- a choice to join into this, as opposed to the Indian Health Service.
MR. VAN AMBURG: Indian Health Service.
MS. ARAKI: So, would that be --
MS. PRICE: All of them get a choice --
MS. HART: Um-hum.
MS. PRICE: -- if it's available to them.
MS. HART: Right.
MS. ARAKI: That would --
MS. HART: You know, I know in long-term care, if they have a Tribe that has an IGA with AHCCCS, they can either -- you know, then they do have to go with the Tribe. If they're outside of a county that doesn't have the Tribe, then they're with the program contractor. But I don't know with acute care how --
MS. PRICE: I think they're all given a choice.
MS. HART: -- how that happens.
Yeah, I think acute care gets a choice.
MS. PRICE: They all get the choice of going --
MS. HART: Either deal with IHS versus --
MS. PRICE: Right.
MS. HART: -- the plan.
MS. ARAKI: So would that be one of the two choices that's out there that you said --
MS. PRICE: Well, then, in a way, they'd have three, because they'd have two health plans, and then they'd have IHS.
MS. HART: Right. Um-hum.
MS. ARAKI: Oh, okay, so that's a separate choice.
MS. PRICE: Right, right. No, there's always two managed care health plans in each county, at least two.
MS. HART: Um-hum.
MS. IEZZONI: And do you get any sense of whether they're choosing, whether the Tribal members are choosing AHCCCS, versus the --
MS. PRICE: No, I just --
MS. IEZZONI: No, you don't have --
MS. PRICE: -- I don't have any -- I know we have those statistics, I just -- I can't quote them.
MS. HART: Yeah, I don't have them, either.
MS. PRICE: But we can find that out for you. You can leave us a list of questions, if you'd like, and we will research them, gladly, for you.
MS. IEZZONI: Yeah. Because one of the obvious things is kind of cultural sensitivity and --
MS. HART: Right.
MS. IEZZONI: -- and so on, especially to Native Americans who might have certain care practices, as you described your Hispanic population --
MS. HART: Um-hum.
MS. IEZZONI: -- having family issues around certain types of care, so I was just curious about that.
MS. HART: Yeah. As a matter of fact, we do have, in long-term care, for those Tribes that do not have an IGA with us to perform case management, we actually subcontract with the Native American Community Health, and they case manage those members who are not enrolled with the case management of the Tribe.
So, again, that's to address the cultural sensitivity. And it's difficult because, the only thing is, is you can't -- it's not easy to -- there's no way you could have a case manager who's of the same Tribe, even, so there's even some of those issues, too.
MS. ARAKI: Is that part of the Indian Health Service, the case manager for the community -- that's part of the Indian Health Services?
MS. HART: Well, for -- if they have an IGA with AHCCCS to do long-term care, they have to do case management. So on the long-term care side, they do have the case managers. And then if not, we have the Native American Community Health that case manages those Tribes that do not have a specific IGA.
MS. ARAKI: And that would be part of AHCCCS, rather than --
MS. HART: Um-hum.
MS. ARAKI: Okay.
MS. HART: Yeah, that's part of our requirement. And we go out and we review their cases, just like we do any other contractor, to make sure that case management's occurring.
MS. ARAKI: Do you give your information back to -- you know, from your audits and your reviews, to whom do you give this information?
MS. HART: All the -- for both acute and long-term care, a review is written up and it goes back to the health plan or the program contractor, whoever we reviewed, they do get to see the results.
In the long-term care, they have an opportunity -- we actually send a draft report. They get an opportunity to comment back, in case we missed something while we were out there, and they have a chance to correct it then. We then require, on any deficiencies we find, we require corrective action plan. They have a certain amount of time to get that back to us and have it be accepted by us. And then we also forward that information to HCFA. They get a copy of each report that we each --
MS. PRICE: And the legislature, if they request it.
MS. HART: Right.
MS. ARAKI: The state legislature.
MS. HART: Right. Um-hum.
MS. PRICE: If they request it. I don't think it's just automatically sent there.
MS. ARAKI: What about the Department of Health Services? It's the State Department of Health Services. You don't --
MS. HART: They only -- again, with behavioral health, they do reviews of Department of Health Services. Our behavioral health unit does a review of their activities, and they do write up an operational financial review report for those individuals, as well, or for DHS.
MS. IEZZONI: Let me just say that the reason that we chose to come to Arizona, is because we've heard wonderful things about you, and so we wanted to kind of learn from you about what to recommend to the rest of the country, as we're thinking about making recommendations to the Secretary.
For the two of you, are there any final things that you think, from your experience, that you think should be replicated elsewhere that are particularly valuable for you to do the jobs that you do?
MS. HART: I think -- the only thing that I -- when you talk about collection of data, and I'm sure the two that just joined us will say the same thing, is when you start a program, be very specific on how you want things reported, because you start -- I'm sure all of you know, you start looking at data, and then you wonder why sometimes it can look a little bit funny, is because people didn't know the requirement.
MS. PRICE: Define your requirements.
MS. HART: Right. You know, and -- because, even still, we still continue to refine it in one area, where even description of units might, for some person, a unit may be a half hour or an hour, and so you have to be very descriptive, and make sure that everyone's reporting consistently. And if you can get all of that agreement up front before you start generating data, that helps.
MS. PRICE: And just to stick to the requirement.
At first, even, we tend to be more lax, and if you stick with the requirement from the very beginning, and sanction them, or do whatever you have to do to get them in compliance, your information's going to be better. And they'll talk a lot about that in data validation.
MS. IEZZONI: Yes. One last question.
MS. MOIEN: Lisa, can I just ask one question? Kari had mentioned earlier about one health plan that was in for three years and then decided to drop out, and you said that they really hadn't -- I guess, really hadn't estimated what their load was going to be and how much it was going to cost them.
Is there anything like that would come out in the records to indicate whether the patients who were there, or the enrollees who were there, if their care suffered at all, or --
MS. PRICE: We've had just the opposite, actually. The utilization was so extremely high, these members got more care than definitely the average member. So, they weren't keeping -- they weren't controlling their utilization. And, typically, we would see if there was -- it would be brought to our attention fairly quickly if members weren't getting adequate care. I mean, it gets to the news like that, if we have any major issues.
MS. MOIEN: Well, it's only -- I was --
MS. PRICE: And through our grievance process, we would also have heard about it.
MS. MOIEN: Okay. Because I was thinking that, gee, you know, does AHCCCS have any kind of plan for if somebody comes in who really seems like they don't know what's going on, and so they come in with a low bid, but have some plans for you all to do something.
But, if you're saying that in this case they saw everybody maybe too many times, it wouldn't -- that's not even something that you all would have noticed up front.
MS. HART: Well, and even --
MS. PRICE: No, and even doing an operation financial review, you wouldn't really -- well, at financial review we noticed it, that, you know, they were losing money.
But, operationally, in looking at policies and things, what they're actually doing and what's down on paper are definitely two different things.
MS. ARAKI: Was the reason why they were getting too much services is because they were sicker, or it's just that -- I mean, the population that they were dealing with was much --
MS. PRICE: No, their population wasn't really demographically any different than anybody else's population, it's just that they weren't -- they did come in with a low bid wanting to get in with the contract, and then they did not do a good job with their utilization review.
And, can you think of anything else?
MS. KEAGLE: Just, I think, controlling --
MS. PRICE: Management -- their authorization of services.
MS. KEAGLE: Um-hum. Controlling the members, and controlling the claims. They had just a lot of problems.
MS. PRICE: With their claims system.
MS. KEAGLE: Their claims system.
MS. PRICE: Or, maybe even medically reviewing for medical necessity, so they were --
MS. HART: Yeah. That's the other thing is, is when -- if you -- you know, a recommendation would be, for other states that bring something like this up, is we do have what -- we do perform what we call readiness reviews, and that would be for anyone who's new to the business that would get a new offer that might get a contract, and we do go out to see, do they have a claims system, and do they have certain things in place.
And, you know, that's difficult, because if they haven't already been up and running, you basically have to say do they have certain things in place. It's hard to tell, at that point, how well they're going to perform, but at least you find out, do they have some of the basics, the things that you really need to have in place.
And we do have, it's the readiness assessment tool, we call it the RAT, is the acronym for that, so they do have -- I know when they first said that, RAT, that's what the acronym ended up being, the readiness assessment tool, so they do have -- perform those reviews.
MS. IEZZONI: So is that done before they bid, or after they bid?
MS. HART: It's after they bid.
MS. IEZZONI: And you accepted them, at that point, or do they have to go through the RAT before you accept them?
MS. HART: They're accepted tentative on the review.
MS. IEZZONI: Okay.
MS. HART: Yeah.
MS. PRICE: They're offered.
MS. HART: Um-hum. We're also always available for technical assistance.
MS. PRICE: Right. The one plan I spoke about was in for technical assistance weekly, I think, for six months. And we were also going out to them and helping them with whatever they needed in any area of their program, and going out to providers and even trying to figure out all the issues related to claims, and utilization review, and --
MS. ARAKI: Do you share any of the data amongst your different plans to sort of help them get up speed, or -- I mean, you know, when I asked you --
MS. PRICE: They can request, you know, audited financial information, but, typically, we don't share --
MS. ARAKI: Other plans' data, is what I'm --
MS. PRICE: -- utilization information specific to a health plan, no.
MS. ARAKI: Okay.
MS. HART: We do produce, during bid years, we'd produce a data book --
MS. PRICE: Yeah.
MS. HART: -- that has utilization information, so, you know, but it's not by plan.
MS. PRICE: It's by county.
MS. HART: So they would have some idea. You know, they could look at the county-wide data and have some idea of how they performed.
MS. IEZZONI: Okay. Well, thank you for your time.
MS. PRICE: Thanks.
MS. HART: Sure.
MS. IEZZONI: We really very much appreciate your willingness to share with us.
Okay. We have two new folks. I guess you're Terri.
MS. KEAGLE: I'm Terri.
MS. IEZZONI: And you're Brent.
MR. RATTERREE: Brent, right.
MS. IEZZONI: Okay. And we have name plates so we can further identify you.
Because we are a federal committee and all of our work is public, we have to tape record this, and it will be transcribed, and so we will be asking you to speak into the microphone, and the transcriptionist will let us know if there's any problem hearing you. Okay?
MS. KEAGLE: Problem. Okay.
MS. IEZZONI: So, do you have formal presentations for us?
MS. KEAGLE: Yeah.
MS. IEZZONI: You do. Great. Okay.
MS. KEAGLE: I'm Terri Keagle, and I'm with the Office of Managed Care. I'm the research administrator there.
My unit oversees the encounter collection, processing, and the data validation that we do on the encounters, and we also do some rate setting, both the physician, I mean, whole range of -- we have a CAT fee schedule by code. We set that every year. We also do a hospital rate setting, which is a complicated process we go through every year. And we oversee the disproportionate share of payments, the calculations, and the data use for that.
So we kind of do a wide variety of functions. But, I think what you're here today to hear mostly about is the encounter side.
In November, we had a national conference, and we did do a presentation on the encounter, so what I've given you is copies of the slides that I used in my presentation. A couple of them are duplicated, so it's just because I would reference back to them, and that's why it's that way.
And what I plan to do, my presentation was probably over an hour long, what I want to do is just breeze through it real fast, because you wanted like time for questions and answer, right?
MS. IEZZONI: Right. And --
MS. KEAGLE: So I just --
MS. IEZZONI: And there are certain things that we're not going to need to hear, because we already know about them.
MS. KEAGLE: Okay.
MS. IEZZONI: And --
MS. KEAGLE: So, I just figured I'd just touch through the whole thing really fast. It's broken into four areas. You can tell me right now if you don't want to hear about any of those.
The first is just the encounters that we collect, and why we collect them, because of the HCFA requirements.
MS. IEZZONI: Yes, we'd like to hear --
MS. KEAGLE: The second is how we collect them and process them, what fields we collect -- you know, overall, what fields, and that sort of thing.
The third part is on the validation that we do and studying them on the back end, to see how valid they were and what the error rates were, and how we sanctioned the health plans.
And then the fourth area is just what other ways do we rely on encounters, how else do we use it, which will demonstrate how important they are to our program. So that's how --
MS. IEZZONI: Those are all four things we'd like to hear about just quickly --
MS. KEAGLE: Okay.
MS. IEZZONI: -- so we have a chance for our questions.
MS. KEAGLE: Okay.
MS. IEZZONI: And, Brent, are you going to be --
MS. KEAGLE: And Brent -- I'm going to walk through the first two --
MR. RATTERREE: Right.
MS. KEAGLE: -- two sections, and --
MS. IEZZONI: Okay. So you're a tandem team.
MR. RATTERREE: Right.
MS. KEAGLE: Right.
MS. IEZZONI: Okay.
MS. KEAGLE: So Brent's going to go through the second. And he --
MS. VALLEY: Terri, let me just butt in with an operational thing. When we looked like we were -- we thought we were going to be done with the first group early, so I've asked the team who's going to present, who's going to be talking about our utilization and our quality indicators, and that sort of thing, to come in around 2:30, if they can. So, and that may kind of overlap on what these people are talking about.
MS. KEAGLE: Right.
MS. VALLEY: And so, assuming your questions get into that, I just want them to let them know they're coming at 2:30.
MS. KEAGLE: Yeah, because we're more the technical side. We do study the encounters, and we use them, but they're more the actual users of the encounters, and how we measure the care, and measure statistics to make sure that the managed care enrollees are getting access to care, and are meeting certain utilization goals that we have, and that's, you know, with all the HEDIS guidelines, and Brent will touch on that. But, anyway --
MS. IEZZONI: Okay. Great.
MS. KEAGLE: And, okay. Let me just sort on it.
Since our program began in '82, we started up being managed care under an 1115 waiver, so we've collected encounters since the beginning, which is different than a lot of states who traditionally started up fee-for-service and have been converting into managed care. So we've been doing this for a while.
The -- well, encounters, like it says here, that's -- it's not a claim for financial payment, it's just a record of the medical service that the health plan paid for, and it's just reporting it to us so we know what services our members have received. However, we only require encounters to be reported for instances in which the health plan did incur a financial liability. So if the service was denied, they wouldn't report it to us, or for whatever reason they didn't pay for it, it doesn't come into us.
We collect them because HCFA requires us to collect them, and also because they're very important for us to monitor the services that our members receive. Because, in managed care, there's a concern that you're restricting the services they're going to get. They have restricted freedom of choice, unlike you see in fee-for-service programs. So, we just feel it's very important to monitor the care that our members are receiving.
Okay. Let me keep cooking along. I won't talk too much. You can stop me, if you want.
The next slide just shows how it gets to us, and you probably know that. The member receives a service, the provider bills the health plan, the health plan, in turn, submits a claim in the form of an encounter to us, we process it, put it on our database, and then report to HCFA on the encounters that we receive, what we process, what's pended, how we've used it in the validation studies.
But we, in turn, also provide feedback, especially back to the health -- the health plans know, you know, whether or not they got loaded, if they pended, and we report back, too, through the indicator reports, because each health plan is measured against those quality indicator and utilization reports, and they receive feedback.
Okay. We have 400,000 members. We have the three programs: acute; long-term care; and behavioral health. We have 20 MCOs or managed care organizations. And probably throughout my presentation, I'm going to always refer to health plan, but I mean all of them, all three of them.
So we're required to collect a hundred percent of the encounter data where they incurred a financial liability. However, HCFA's allowed us to not collect the pharmacy, Form C's, we call them, encounters from our acute care plans --
MS. IEZZONI: Can we stop there for a second?
MS. KEAGLE: Sure.
MS. IEZZONI: That's interesting to us. We, frankly, think that you might be losing a lot of valuable information by not having the pharmacy claims. A number of states that we're very familiar with use those pharmacy data very productively, for a variety of purposes.
Is this an issue that you all are re-exploring? Can you tell us --
MR. RATTERREE: Yes.
MS. IEZZONI: Can you talk to us a little bit about that?
MS. KEAGLE: We are, and I'm sure when the next group comes in, they're lobbying hard to have us start collecting it. We're not against it. You know, it's a lot of -- a lot more encounters to have to store and process, but we are --
MS. IEZZONI: Computers are bigger now.
MS. KEAGLE: Right. We are exploring it. We're behind the efforts to try to get the agencies to say, yes, we will start collecting it. OMD especially wants the information to better track the types of services that the members are receiving, and better track like certain diagnoses, and how they've treated, and that sort of thing.
And, right now, they do -- we do receive, on the acute care side, summary financial -- or, pharmacy data, but it doesn't have the detail that they want, so --
MS. IEZZONI: Now, is this going to require a legislative fix?
MR. RATTERREE: No.
MS. IEZZONI: No. It's regulatory. How would you be able to do this?
MR. RATTERREE: It's primarily -- as far as collecting the data, it's primarily getting the support to do that, and then --
MS. IEZZONI: From who?
MR. RATTERREE: Both internally and externally. It's an issue where you need to talk to the MCOs, encourage them to -- that it'd be beneficial to collect the data, and then once the data's collected, then we can analyze it and do other things.
MS. IEZZONI: So this would be kind of voluntary, then?
MR. RATTERREE: No, no.
MS. IEZZONI: No?
MR. RATTERREE: At this point, we're away from collecting the pharmacy data for the acute plans. But, it's our view here, at least below executive management -- we're taking this to executive management to collect this data.
In the past, there was an issue of storage cost and some computer issues. We kind of think those are moot now, and we want to proceed and collect the data.
MS. IEZZONI: George, do you have anything? Can you --
MR. VAN AMBURG: Well, I don't think the costs are moot, because there's a lot of claims, pharmacy.
MS. KEAGLE: There are. But the --
MR. VAN AMBURG: And, I was interested in your comment you don't need legislation. Where are you going to get the money to process it, and handle it, and analyze it?
MS. KEAGLE: But that's where we need to get, you know, executive management to support it, so that it can be put into the budget, which, therefore --
MR. VAN AMBURG: Okay. Then you do need legislative approval, essentially?
MS. KEAGLE: Yeah, money-wise.
MR. VAN AMBURG: If they approve the budget, you've got it.
MS. KEAGLE: Right.
MR. RATTERREE: Right.
MS. KEAGLE: And then it would just take out -- right now, it'd just take -- when we -- we're in a five-year acute care contract right now. Every year it goes through a renewal.
What we were hoping for was for the 10-1 renewal to be able to put in there that they're not required, amended to say they're required, and then part of their rate renewal update would include some sort of an adjustment for the costs on their side to submit the encounters to us, the additional encounters to us.
We do receive them on our long-term care and behavioral health plans right now, and there's actually one of our --
MR. RATTERREE: One of the acute ones that submit it now.
MS. KEAGLE: -- health plans submit them right now, but we'd like to get it consistent and start doing it, so it's going to take some time down the road before we can look back and rely upon it and use them, so --
MR. VAN AMBURG: If you have someone submitting them now, you must have a system for processing.
MS. KEAGLE: Yes, we do.
MR. RATTERREE: Oh, yes. The system is in place, it's just a matter of, as far as the computer section, is storage cost.
MS. KEAGLE: It's just the one program has always been exempt from having to submit them, where the large volume is, but we are set up to collect them and process them.
MS. IEZZONI: Sorry to interrupt.
MS. KEAGLE: That's okay.
MS. IEZZONI: It's just been something that's been floating around for the last 24 hours that we've been concerned about.
MS. KEAGLE: Yeah, but it probably will come up with the OMD folks, because they really want to see it happen, so we're pushing.
Okay. So, other than the fees on the acute side, we do collect a hundred percent of the encounter data.
Okay, HCFA requirement is that we provide technical assistance to our health plans, and so my next slide will show you how our unit's set up, but we've got six people that are pretty much dedicated to facilitate the processing of the encounters, and also providing technical assistance to facilitate them, getting the encounters to us, us getting them back to them to load back on their system, and also work through the pended encounters that don't clear our system, because a lot of times they don't know how to fix them or what they need to do, so we've got people on the phone to them all the time trying to work through those pended encounters and provide that assistance.
So let's just flip over right now. I'll come back to that slide.
MR. HITCHCOCK: What page are we on now?
MS. KEAGLE: 3. Top of 3 is the work chart. It's hard to read, I know. We have 13 full-time employees in the actual encounter area. And Brent is the encounter administrator, so he's kind of the top of the encounter side.
It splits into two main areas. One is operations, and one is the validation. We've got an encounter operations manager that oversees the six staff, two being kind of the tape specialist, because we're not doing electronic interchange yet, we're still doing tape processing, so there's two people that do that. And I'll talk about that in a little bit. We hope to move the EDI this year. But there's like four other support positions that just interact with the health plans, track trends and work with them, so that's the operation side. So we've dedicated, you know, quite a few staff to that piece of it.
On the validation side, actually, since I did this presentation, we've gotten one more position. There's actually four validation analysts, so they're actually the people doing the data validation studies. And Brent will talk about how we do those studies, but there are four people there.
And then we also have kind of a programmer-type position. Every month after we receive the encounters and they're processed, ISD, our Information Services Division, dumps a data extract to us of all the encounters that adjudicated. So every month we just continue to accumulate, and we can use that data to do different analysis on the actual encounters that are processing, or being adjudicated. So that's kind of how our unit's set up here at AHCCCS.
And back to the requirements, we're also required to do the data validation studies that Brent will be talking about. And we're also required to sanction the health plans, and we sanction them in a couple different ways. One is on kind of the processing side, where if encounters pend, we sanction them. If they don't correct their pends within a hundred days, we sanction them $5 per pend per month that they're out there not being fixed.
MS. ARAKI: Pend, meaning?
MS. KEAGLE: Pend meaning the encounter was submitted to us and we tried to process it, and for some reason, we couldn't accept it because that provider couldn't perform that type of service, or it's a duplicate, you know, like submitted one that they already submitted, or something like that. And for a wide variety of reasons, it can pend. So it goes out in this pend status, we send it back to them and say you need to fix it. If they don't fix it within a hundred days, that's when those sanctions will start accruing.
And, anyway, and then we also do sanction on the validation side, depending upon the different studies that they're working on, and whether or not they exceed the allowable error rates. There's different sanctions for different types of errors, and Brent will talk about those, too.
As far as -- so those are our requirements that HCFA imposes on us. We also have requirements that we place on the health plans, and most of those are outlined in the RP, or request for proposal, or our contract that we have with them. And they're kind of discussed throughout the walk-through that I'm doing with you.
There are a couple things that I just wanted to touch base on that aren't in the rest of my talk. Jan mentioned that readiness review --
MS. IEZZONI: The RAT.
MS. KEAGLE: -- where we do that site visit for a new contractor that hasn't done business with us yet, once they've been awarded a contract, so we can make sure they're really up and ready to start doing this. So that's something that, on the encounter side, we make sure that they've got the process and the capabilities to, once they pay their claims, they can submit them to us in the form of an encounter, and that they can correct their pends, and that sort of a thing.
And then, okay, then the annual reviews. Every year the health plans go through an operational and financial review. I'm sure they touched base on those, and part of that review is looking at, you know, how have their encounter submissions been, how they've been doing in data validation, do they have policies and procedures in place for their encounter area.
MS. IEZZONI: One of the things that we read was that you have to submit the data to you guys within 240 days.
MS. KEAGLE: Yes.
MS. IEZZONI: That seems like a bit of -- that's generous. Where did that number come from, and is that really the standard that most --
MS. KEAGLE: It's the standard. I mean, about 90 percent of the encounters that we receive, we do receive within the first six months. It's just that -- we're just a -- you know, we've got to allow time for the providers to bill the health plans, and the health plans to work out any issues they have on having to have the providers rebill, or whatever.
Then once they've paid and adjudicated their claims, that's the point when they submit it to us. So that's -- it used to be six months. We used to require six months, and it was found that we needed to allow longer, move to eight months, to ensure that they had enough time to submit a clean encounter to us so we'd reduce the amount of, say, corrections that would have to come in.
But like I said, we do see about 90 percent come in within the first six months.
MR. RATTERREE: There are also some third party issues, where --
MS. KEAGLE: Right.
MR. RATTERREE: -- another party was responsible for the claim. And if, you know, it ended up in medical review at the primary insurer, then that would just delay the claim being paid by the primary and into the health plan, basically.
MS. KEAGLE: Yeah. Because we require them to provide that information on the encounter to us, any third party payments that were made, so --
MS. IEZZONI: But do the people who are using the data for oversight in quality management sometimes feel that they're chasing a train that's way down the track?
MS. KEAGLE: Yeah.
MS. IEZZONI: Yeah.
MS. KEAGLE: So it's like how soon can we pull it, you know.
MR. RATTERREE: Yeah.
MS. KEAGLE: What percentage of the data's in right now. Right now, there's one thing in Kari's area where they used to have the health plan's self-report certain hospital utilization data, and it was done within a month after the end of a quarter, so they could turn around some utilization stats and charts real quickly.
And 10/1 they stopped that requirement, so the health plans don't self-report that information any more, and they want to rely on our encounters. And because we get that big dump every month, we're going to actually generate the reports for them. But, I'm like there's no way you're going to have them a month after the end of a quarter.
So we're actually trying to study at what point after the end of a quarter, so for three months, how many months after that end do you have a certain percent in, so we can at least generate some utilization stats for the health plans that want to see it quickly. Otherwise, they're going to have to go back to the self-reported.
MS. IEZZONI: Yeah.
MS. KEAGLE: So, I think you could find out one of the reasons -- I mean, this is exactly why we've now got HIPPA.
MS. IEZZONI: Right.
MS. KEAGLE: I mean, one of the major problems this country has with paper claims. It has nothing -- I don't -- from my experience, it has nothing to do with this state government or health plans, it's the fact that we are still sending in the mail pounds and pounds of papers that are filled out incorrectly.
MS. WARD: Right.
MS. KEAGLE: Because people are really busy. Everybody knows that once you go to electronic, the accuracy goes up.
MS. WARD: I do, too.
MS. IEZZONI: Can I just go back, though, to, Terri, what you were saying? To find out about hospital admissions, is what you were just talking about, do you have a requirement in the state for hospitals to submit discharge abstract data to some state health data repository, or a central data authority?
Like, there are 30 states around the country that do, and is Arizona one of those?
MS. KEAGLE: Does DHS get that? I know that we require certain information, on like newborns.
MR. RATTERREE: Yeah. I don't necessarily think we get that information. There may be perhaps the Office of Vital Statistics that gets that, but I'm not -- I can't say for certain. I don't know.
MS. KEAGLE: I don't know. Yeah.
MS. IEZZONI: Probably, if you don't --
MR. VAN AMBURG: I think that Arizona does.
MS. IEZZONI: Yeah. If -- excuse me?
MR. VAN AMBURG: I think Arizona does.
MS. KEAGLE: We do.
MS. IEZZONI: Well, if they don't know about it, though, it's kind of surprising that you're not tapping into that data source for some of this information, because in Massachusetts now, for example, we have very detailed information on the payor for each hospitalization --
MS. KEAGLE: Oh.
MS. IEZZONI: -- and it's mandated to be reported to the state. And it's done very quickly, so you actually can get data that's pretty recent.
Yeah, Kathy.
MS. KEAGLE: If that's happening, I'd love to know about it.
MS. COLTIN: It isn't always -- the payor code information isn't as good as it should be. You can --
MS. IEZZONI: Right. But --
MS. COLTIN: -- in Massachusetts, at least, you can identify that this claim was incurred by a member of a health plan, but you can't always --
MS. IEZZONI: Well, if they serve Medicaid, you'd be able to --
MS. COLTIN: You can't. In fact, that's the problem.
MR. VAN AMBURG: No. That's the problem.
MS. COLTIN: The problem is that they cannot identify the Medicaid members through the health plan --
MR. VAN AMBURG: Right.
MS. COLTIN: -- and the Medicare members through the health plan. They're pretty good at identifying they're a member of the health plan, but they're supposed to subset them into commercial members, and so forth. And our analysis, because we matched up our claims with the state database.
MS. IEZZONI: Um-hum.
MS. COLTIN: And, in fact, there's at least a 15 percent --
MS. IEZZONI: Right.
MS. COLTIN: -- error rate there. So --
MS. KEAGLE: Well, they're -- yeah. The --
MS. IEZZONI: But, if you knew --
MS. KEAGLE: -- we've had a lot of --
MS. IEZZONI: -- if you had some I.D. number on the patient, though, you'd be able to --
MS. COLTIN: Well, if you do, but you don't.
MS. IEZZONI: -- match by that.
MS. KEAGLE: We've had a lot of -- the hospitals here, I don't think, have been as good as maybe what they wished they would have -- or, I don't think they've tracked it as well as they wished they would have tracked it, as far as our patients.
And even within our population, because we have the Title 19, and then the state only population, because it's effective there, the data, say, that we submit and that Medicare uses for the DHS payments, sometimes they want to argue with Medicare about the Title 19 data.
Well, they can't even go into their systems and identify which ones are Title 19 and non-Title 19, so they -- we have all these consultants always trained to come to us saying we want all the Title 19 member data information for this hospital, and we got -- like a year ago, we just got hit with just a slew of requests, and we didn't produce it.
MS. IEZZONI: Okay.
MS. KEAGLE: It was just going to be an enormous effort. But, anyway, I know that the hospitals do submit data to our Department of Health Services, because with our hospital rate setting, we do use the UAR's at the hospitals --
MS. IEZZONI: Yeah, that makes sense. That would make sense.
MS. KEAGLE: -- submit, but I don't know that it's real timely. I know that we're getting the '96s in right now that we're going to use for our rate setting, and that we used, also, for the DHS calculations. But I don't know if it's got the detailed utilization.
And, also, I know on there when we look, they never fill out the Title 19 or Medicaid area specific, so we can't ever use that to see what percentages.
MS. IEZZONI: Okay.
MS. KEAGLE: So, anyway, I do know that.
MS. IEZZONI: Okay, that's helpful. All right.
MS. MOIEN: Can I just ask a quick question?
MS. KEAGLE: Sure.
MS. MOIEN: Are you saying that the MCOs, they have to pay the providers, they've got to get any third party insurance, and co-insurance, have all of that straightened out, and then they submit the encounter to you?
MS. KEAGLE: Typically. If they do submit it to us before that's all come through, or it comes through subsequently, they're required to submit an adjusted encounter to us so that we do get it. But they're still limited to the 240 days. They still submit it after that, but they'll be sanctioned on the back end for it.
MR. VAN AMBURG: Who codes the diagnosis and procedure, the original provider, or the plan?
MR. RATTERREE: The original provider is supposed to code those.
MR. VAN AMBURG: And if it's rejected, what happens by the plan?
MR. RATTERREE: If the plan rejects it from the provider's office?
MR. VAN AMBURG: Yeah.
MR. RATTERREE: The provider should really find out why it was rejected, and then see if it was correctable. It could be maybe the plan didn't update one of their reference tables in time.
MS. KEAGLE: So they shouldn't submit that encounter to us, because it really wouldn't be an encounter until they paid on it. So I think if they would reject it up front, the provider would typically rebill. And then once it's been paid, then that paid claim comes to us as an encounter.
MS. IEZZONI: Okay. Great.
MS. KEAGLE: Okay. Okay, let's go to Page 4. That's my first break, as far as kind of the intro into why we do this whole thing, and what our requirements are.
I'll move into the collection of the encounters, and the processing that we do next, and then from there is when we'll move into the validation and study of them that Brent will do.
We collect three form types: Form A's or the 1500s that are used to bill the professional services; Form B's or the UBs are the facility services; and the C's are the pharmacy encounters.
We define, for the encounters that come in, the fields that need to be populated. So each claim needs to kind of turn into an encounter. And these are the fields that we want. And we designate that by form type to the health plans.
Then what we do is, we tell them, okay, these are the fields we want, and this is how we want you to lay it out on the tape, and give them exact positions, and that sort of thing. And if you're at all interested in any of that, we've got a manual we can send to you, or give to you today, or whatever.
And the other information I note there, all I wanted to do is make a point there that this doesn't -- the information doesn't change too often. We don't change the fields very often or the layout very often, but there is one change that we're making currently.
And in the past, on the 1500s, we had only been collecting two diagnosis codes, when there's actually room for four, and we had some of the health plans screaming saying I'm not -- you know, because you don't collect all four, sometimes, say, if the well child visit was denoted on maybe the fourth diagnosis, they wouldn't get credit for that encounter, and so they're like why don't you guys collect it.
And it just had never been an issue going way back, and nobody had revisited it, and it's like we should be. And we're making the change, so we've expanded the two to the four, and now the health plans are required to send those in. So, that's just an example of how we made a change.
MR. VAN AMBURG: Do you require the "E" codes?
MS. KEAGLE: Do we require?
MR. VAN AMBURG: "E" codes.
MR. RATTERREE: "E" codes?
MR. VAN AMBURG: Um-hum.
MR. RATTERREE: Yes.
MS. IEZZONI: "E" codes, external cause of injury codes.
MR. RATTERREE: Yes.
MS. KEAGLE: Okay. So, our health plans submit them on cartridge or reel tape, and we know it's real outdated. We're really wanting to get to electronic submission.
On our fee -- we have a limited amount of fee-for-service that we still process claims here for, and we have finally gotten EDI here, so at AHCCCS we process it, but that's a real limited amount of claims that they have to process. So when we hit them with the encounter processing through EDI, it's just -- IS is having to work with us to develop the mechanisms to allow that amount of information to be able to be transmitted to us electronically.
So, we're shooting for later this year. I don't know, have you gotten any more updates on where they're at with it?
MR. RATTERREE: No, I really don't know where they're at. It's a target. Hopefully the target won't move, but --
MR. VAN AMBURG: Are you taking into consideration the proposed to electronic transactions standards under HIPPA?
MS. KEAGLE: The security-type things, or --
MR. RATTERREE: Actually, we're looking to follow a national standard format, and try to make it, you know, as exactly like the national standard format.
MS. KEAGLE: Okay, okay, okay. Okay, when they submit the encounters to us, they submit a new data tape, which means encounters we've never seen before, new ones. And then also pend correct tapes, and those are the pended ones we've sent back to them to correct, so we just get them in two forms. And there's where I talk about that we require them to submit within 240 days.
Then they go through various processing steps, and you can stop me if you want to know more, but this next light bullets out into the next five slides. So first thing is, they go through -- they submit monthly -- or, they submit throughout the month, but we run the cycle once a month.
Dah, dah, dah, they're all -- okay. The first -- once they've submitted the tape to us, we run it through a pre-syntax tape, or test that is just the initial test to say is the tape readable, does it catalog properly within the system, does it have a syntax failure rate of less than two percent, meaning like, are the numbers where they should be and the letters where they should be, and just check -- do an overall check to make sure that the records on that tape aren't more than two percent duplicative against the system.
There's another dup check I'll talk about that, we do later. That's just an initial one, where if it's greater than two percent it kicks out right there and we don't process it any further.
Next, each record on the tape is assigned a unique CRN or claims reference number, and that's how we identify each record on that tape. Each one's assigned a number, and it's 14 digits.
Then, once the monthly deadlines reach, like this month it was yesterday, was the deadline for the health plans to get the tapes into us. And we get them loaded, so then we can start the main cycle running, and that's where, when we start doing the edits and audits test to each record on the tape that we've received.
The edit process checks the data quality of each record that we receive, and the audit process checks each record for potential duplicate status, so if it is, it's going to pend, at that point. The edit checks that we do involve recipient, provider, revenue, diagnosis, procedure edits to your outlier. There's just a whole bunch of them, and I'm sure you guys know about a lot of them. And then the audit, I've talked about.
Once an encounter is passed, you know, first the pre-syntax, and then -- that's the tape, and then once each record's passed these edits and audits, then it's finally approved, and adjudicated, and put on our database. And then also once that cycle's run, and they've made it through all those hoops, we process tapes to go back to the health plans, one being the encounters that were approved and placed onto the database, and then another one being the ones that didn't pass, and they're pended, so --
MR. HITCHCOCK: How long would that take?
MS. KEAGLE: That process?
MR. HITCHCOCK: Yeah, just the very last -- well, the time you get the data, how long's it take before you have to supply something back to the --
MS. KEAGLE: Like, say, the process we're in right now, where we just did -- the deadline was yesterday, so they started running the cycle today. The cycle's been shortened, and it's taking us about --
MR. RATTERREE: About two days.
MS. KEAGLE: -- two days to run the process, then within, what, how many days do we have the tapes back?
MR. RATTERREE: I'd say probably five to seven days we'll have information back to the MCOs.
MS. KEAGLE: Plans. So, but it's sure -- I mean, the process is okay that we do, as far as generating it, it's just certainly nice to not have to have all these couriers in and out of our area bringing these tapes. It's like -- but, soon.
We process about a million encounters every month. And approximately four percent of the encounters that we process pend, and about eight percent of those that pend are over a hundred days and accruing sanctions. And then there's that approximately 90 percent of the encounters that we receive, are received within the first six months.
And this is where we get into data validation. Before we go there, do you have any other questions on the first two areas?
MS. IEZZONI: No, that is good. Yeah.
MR. RATTERREE: As far as the encounter validation methodologies, we've got several different methodologies that we use. One of them is the medical record comparison. That's used primarily for professional services. We also have file comparisons, and we also do like an independent report on newborn births comparison.
The other types that we examine in these studies, we look to see if we received the encounter, for one, was it timely received, and where the coding basically -- is simple coding there, such as is the diagnosis code on there, coded appropriately, the procedure code, revenue code, that sort of thing.
For the professional services, our Acute and Behavioral Program uses the medical record methodology, and we check for all three of these error types. For the ALTCS Program, we simply use a file comparison, and we're just checking for omissions, on that particular one.
MS. IEZZONI: Brent, can I just interrupt you there?
MR. RATTERREE: Sure.
MS. IEZZONI: For procedure codes, do you use CPT, HCPCs (phonetic), what do you use for professional --
MR. RATTERREE: We use primarily CPT, but we'll use them all.
MS. IEZZONI: Okay. Do you use the same convention that Medicare does, whereby hospital-based procedures are ICD9CM coded?
MR. RATTERREE: Yes.
MS. IEZZONI: You do?
MR. RATTERREE: What we're looking at, primarily, though, on the 1500s, is usually --
MS. IEZZONI: That would be the CPT.
MR. RATTERREE: -- medical-related services --
MS. IEZZONI: Right.
MR. RATTERREE: -- CPTs, medical-related services for surgeries.
MS. IEZZONI: And --
MR. RATTERREE: Those are the primary ones that are listed.
MS. IEZZONI: Do you have special codes that you define, yourself, or do you use --
MR. RATTERREE: We have some state only codes for behavioral health, and we do look at some of those in these studies.
On Page 10, for the UBs, the Acute Program uses a file comparison based off of a newborn report. That report, the hospitals will call in to us when a newborn is born, and that's kind of an independent check just to make sure we've got an encounter for that particular one.
MS. IEZZONI: They physically telephone you.
MR. RATTERREE: They telephone in to, I don't remember which department.
MS. IEZZONI: To some 1-800 number, or something? I mean, do they -- they don't actually speak to a person, do they, or --
MR. RATTERREE: I'm not actually familiar with that department. I believe they actually speak to a person.
MS. KEAGLE: Does Susan know?
MS. CYPERT: Twenty-four hours around.
MR. RATTERREE: Twenty-four hours around the clock. They call in and say here's a birth on this particular -- from this particular MCO, and they give some other data about it.
MS. COLTIN: Is that the vehicle by which they then enroll the baby in the MCO, or --
MS. ARAKI: Do they talk to somebody, though?
MS. KEAGLE: They do talk to somebody.
MR. RATTERREE: I believe that has something to do with the way they're enrolled, eventually.
MS. KEAGLE: Oh, we're just talking about how the hospitals call in newborns?
MS. CYPERT: Oh, they don't call it to us, though --
MS. KEAGLE: Do they do it --
MS. CYPERT: -- they call CMS.
MS. IEZZONI: So, can --
MS. KEAGLE: Right, right, right, but I just --
MS. CYPERT: Yes.
MS. IEZZONI: Yeah. Was that about the births, that discussion? Can you speak into the microphone, because this is interesting to us.
MS. KEAGLE: Oh, sorry.
MS. IEZZONI: We're curious about why -- George, do you want to --
MR. VAN AMBURG: Yeah. I assume you're enrolling those children when they're born.
MS. CYPERT: Correct.
MS. KEAGLE: Yes.
MR. RATTERREE: They're enrolled, right.
MR. VAN AMBURG: Right. Immediately. But, yet, you're going to have a State immunization registry that's going to populate that from the electronic birth certificate here. Why are you not using the electronic birth certificate?
MS. CYPERT: I'm not sure, the answer to that. I know that the health plan, the capitation rate depends on when the capitation rate begins payment, depends on how quickly Valplan calls in and notifies us of the birth of the baby. That might be one reason, but I don't -- I'm sorry, I can't really answer that.
MR. RATTERREE: And --
MR. SCHAFER: They also give birth -- they also do birth weight, and a few other things that they track.
MR. RATTERREE: Right. Well, they've been doing this since probably '82, or somewhere around there.
MR. VAN AMBURG: Well, in '82, there were no electronic birth certificates, so --
MS. CYPERT: Yeah.
MR. RATTERREE: Right.
MS. IEZZONI: They're more recently?
MR. VAN AMBURG: Yeah.
MS. CYPERT: That would be a good question to find out. I'm sorry, I don't know the answer.
MS. IEZZONI: So it's the hospital that calls in, though, you said not the managed care organization.
MS. KEAGLE: Well, they both call in.
MS. CYPERT: Both, yeah.
MS. IEZZONI: They both call in.
MS. CYPERT: Yeah.
MS. KEAGLE: Because the health plans have an incentive to call in as soon as possible when that baby's born, because their capitation begins on the day they call us. So, if they wait two days, they're not going to get capitation until the second day, however, they're responsible for the hospital bills, or anything else that that baby's had since birth. So --
MS. IEZZONI: So both the hospital and the managed care company call you.
MS. WARD: That seems crazy.
MS. CYPERT: It isn't necessarily as duplicative as it sounds. I think, for our purposes, our recipient file, we focus on the calls from the health plans, and that sort of thing, for our payment purposes, anyway.
MS. KEAGLE: Right. Yeah. And when that baby's enrolled, probably DMS, the Division of Member Services that wants that baby into the system, because they do track like birth weight, and that sort of thing, so --
MS. CYPERT: Um-hum. And verify on the hospitals on that information.
MS. KEAGLE: But, it's like you said, you've got electronic birth certificates?
MR. VAN AMBURG: How large is this state? How many people in this state? What's your population?
MR. SCHAFER: Four million.
MS. CYPERT: Yeah, probably.
MR. RATTERREE: Four-something?
MS. KEAGLE: Yeah, we've got 400,000 members.
MR. SCHAFER: 4.8 or 4.7.
MR. RATTERREE: Four-something.
MS. CYPERT: The majority of it is in this area here.
MS. KEAGLE: But, like, half the births in Arizona, we paid for.
MR. VAN AMBURG: So you should be having 60 to 70,000 births.
MS. KEAGLE: Yeah, we have 60 or 70,000 births.
MS. CYPERT: And be --
MR. VAN AMBURG: That's a hundred and twenty thousand phone calls.
MS. IEZZONI: That's a hundred twenty thousand. Okay.
MS. CYPERT: Yeah.
MR. RATTERREE: As far as the other programs, behavioral health and ALTCS, that's a file comparison study that we use. Budgets checks for omissions. However, on the behavioral health side, we also check for timeliness there.
The next few deal with the medical record methodology, and unless you really want to get into that, I'm just going to gloss over that and move to the next page.
MS. KEAGLE: That's pretty -- that whole next piece gets pretty detailed in how we actually collect and do the sample and, you know, work with the providers to get the medical records. And, so, anyway, I don't know that you really want to --
MR. VAN AMBURG: But, I guess, you are getting the medical record from the provider, not from the plan.
MS. KEAGLE: Right.
MR. RATTERREE: That's correct. Yeah, that's -- the whole process takes about nine to ten months to turn it around from the beginning of the study, to the end of the study.
MS. IEZZONI: And is it a random sample, is it a stratified random sample --
MR. RATTERREE: It is a random sample.
MS. IEZZONI: -- stratified by anything? I mean --
MR. RATTERREE: It is by health plan or MCO, it's an error-based sampling methodology that's used. So, depending on how well they've done in previous years, will depend how large the sample size.
MS. KEAGLE: Um-hum.
MS. IEZZONI: You over-sample certain types of care?
MR. RATTERREE: Right.
MS. KEAGLE: And, right now, we primarily do it on the 1500s. And we're moving to expand. We need to -- we're in the process of getting approval from HCFA, or asking approval from HCFA.
MR. RATTERREE: We have discussions underway to expand that to the facility side, as well.
MS. IEZZONI: Now, have any of the patients ever questioned their confidentiality? I assume that these aren't blinded records, that the names of the patients aren't masked in any way. Because this is a big issue right now. You might have heard Donna Shelala's talk about privacy, and the fact that the Justice Department is going to be able to do the audits for fraud, and so on. And there are some privacy advocates at the federal level who are concerned about that.
Has that ever been an issue for you in doing your validation --
MR. RATTERREE: That's really never been an issue. The records are in a pretty secure place, file cabinets, office location. There's not a lot of access to those records.
MS. IEZZONI: I guess the concern, federally, is if an auditor finds evidence of criminal behavior, such as intravenous drug use, or something like that, could the auditor give that information to Justice to go after a patient?
But that, it doesn't sound like it's been an issue for you here, yet.
MR. RATTERREE: No.
MS. IEZZONI: No?
MR. VAN AMBURG: Is your sampling frame the eligibility file, or the encounter file?
MR. RATTERREE: It's the eligibility file.
MR. VAN AMBURG: So you are sampling people who had no encounter through the system, perhaps.
MR. RATTERREE: That's correct. And we wanted to do that, specifically, to make sure, you know, if we sample the encounter file, we might actually miss something.
MR. VAN AMBURG: Right.
MR. RATTERREE: On Page 11, the sanction policy. They are allowed a five percent error rate, so when they exceed that five percent error rate, then they are sanctioned, not just on a sample basis, but it's -- there's a weighting that's applied across all of the MCO's encounters.
MS. IEZZONI: Do you weight different types of errors differently? I mean, if it's an error on the date, is that weighted the same as an error on the procedure code?
MR. RATTERREE: All the errors are weighted the same. However, if a particular -- we have one record that may have three errors on it, we will count that whole record as one error, not just three errors.
MR. VAN AMBURG: So it's a record error weight.
MS. IEZZONI: Okay. Yeah. But it's not something like a transposed date, whereas the error on the procedure codes could be gaming or manipulation.
MR. RATTERREE: Right.
MR. VAN AMBURG: Just out of curiosity, do you partition the errors between those made by the provider and those made by the plan?
MR. RATTERREE: No. We make the plan responsible for the provider's errors, as well. Some of them -- there has been some discussion about that from some of the plans, but as long as we reinforce that uniformly, there's no problem.
MR. VAN AMBURG: I was just kind of curious about where most of the errors were coming from.
MR. RATTERREE: I would say most of the errors are coming from the provider.
On Page 12, other data quality efforts that we have, we do some reel time data quality evaluations, such as when we receive the encounter data, this data dump that we receive on a monthly basis, we'll do some like encounter per member month ratios, we'll do some other data element ratios, and frequencies, just do a reasonableness check to see are we getting the data that we think we should be getting.
At the present time, we're kind of building that database. We've only been doing that for about a year. So we want to get a real good feel, historically, what we may have, and what we may be missing.
MS. KEAGLE: And that's something that we feel should have been being done for a lot longer of the time, just to make sure -- well, see, for so long it was like we were receiving these encounters and processing them, but -- and doing the validation, but it doesn't help if you're just doing validation and then you don't do anything in the -- you know, right now to prospectively fix it, you're never going to get them cleaned up.
So, that's something that we would recommend to other people, that whoever -- whatever system you develop, when you start collecting information like this, that you make sure you're looking at it, you have people that know what the data's supposed to look like, and you monitor it, to make sure you're getting complete and accurate data that you can rely upon as soon as possible, so --
MR. RATTERREE: And they have appreciated this, because we have found some errors up front that would have -- they would have been liable for sanctions later on down the road. So many of the MCOs are very appreciative.
MR. VAN AMBURG: What are the types of errors you're finding?
MR. RATTERREE: As they come in?
MR. VAN AMBURG: Yeah. I mean, are you finding wrong diagnoses, wrong procedure codes, or --
MR. RATTERREE: The easiest one to find, really, is perhaps maybe they missed a tape and they omitted something for a month. They omitted -- say, they could omit some 1500s, or UBs, something along those lines. Those are the easiest ones to spot.
The coding deficiencies and problems, that's a little more difficult to find. You've really got to drill down real deep to find those errors. And we have found, basically, some unit problems. And that's -- you just go back to them, and it's kind of a provider education issue.
MS. IEZZONI: Are you going to be following the evaluation and management guidelines that HCFA's promulgating for the Medicare Program for the ENM codes for CBT?
MR. RATTERREE: We try to follow the ENM Guidelines, CBT. We try and follow all those guidelines that are done nationally.
MS. IEZZONI: Because, you know about the ENM Guidelines that are extremely lengthy and detailed that have been delayed being actually finally promulgated, but are supposed to go into effect July 1, '98 for HCFA?
MR. RATTERREE: I'm hoping to, at this point. I'll have to see the file when it comes out, but --
MS. IEZZONI: Well, if it came out a year ago --
MR. RATTERREE: It has come out? Okay.
MS. IEZZONI: -- it's just that they've delayed implementing it because it's so onerous. And educating physicians has been so delayed.
MS. KEAGLE: What's it called?
MS. IEZZONI: Evaluation and Management Coding Guidelines.
MS. KEAGLE: Oh.
MR. RATTERREE: EM Codes.
MS. IEZZONI: It's a big, huge, thick thing that the AMA worked with HCFA to develop for the EMN codes for CBT.
MS. KEAGLE: Okay.
MS. IEZZONI: You might want to look at it --
MS. KEAGLE: Um-hum.
MS. IEZZONI: -- and then make a decision. But, Kathy, you had a comment?
MS. KEAGLE: I'm sorry. Yeah.
MS. COLTIN: Well, I had two questions. One is, do you see any changes in the error rates as a result of feeding information back; and, two is, do you know, if a lot of these areas are, in fact, at the provider level rather than the plan level, are the plans passing the sanctions along to the providers?
MR. RATTERREE: Let me ask the second one -- or, answer the second first. We have heard that some of the plans are passing those errors on to the providers. We do give them provider specific information when we hand them the final reports. If there's a provider on there who's consistently high in errors, I'm sure that's a contract issue that MCO has to deal with.
As far as the second question, would you rephrase that again to me, please?
MS. COLTIN: I was asking whether you've observed any declines in the error rates over time since you've been doing these kinds of checks and feedback.
MR. RATTERREE: Generally, as the MCO has more experience doing this, the error rates do decline. You do see problems, though, when someone changes to a different information system, or you get new staff that's responsible for doing some of these crucial duties. Turnover affects it.
So, if someone has experience, or the MCO has experience, the errors do go down. But, other problems happen, and it bumps up again.
MS. IEZZONI: But they blip up again through the audits, then. Okay.
MR. RATTERREE: As far as our future plans, we've touched on some of those already. To expand some more medical record methodology. Also to move towards an EDI process for our encounters.
Next page, 13. How we use these encounters. Well, we use them to base our capitation fee-for-service rates off of them. Also, the data book relies upon the encounters, and that's -- the data book is the document that is important for contractors doing the bidding.
MS. COLTIN: Would it be possible to get a copy of the data book?
MS. KEAGLE: Sure. You guys want how many copies?
MS. IEZZONI: Just one for staff, yeah.
MS. KEAGLE: Okay. I'll try to grab one when we're done.
MR. RATTERREE: Okay. Also in calculation, disportionate share to do reinsurance, which is our stop loss program for the MCOs, and regulatory oversight. Regulatory oversight involves such as the clinical indicators, which these folks are here to speak about. Also, the utilization reports, which they're also here to speak about.
Our Office of Program Integrity does fraud and abuse monitoring based off encounters. The validation studies are based off of encounters. There's also some OMD studies that are done which uses these encounters. They might, for example, look at some transportation issues, or some other things that may be important at the time, special studies. And I'll just end it there, so you have some more time for questions.
MS. IEZZONI: Well, we interrupted you all along with our questions. So, do we have -- Elizabeth, Kathy?
MR. VAN AMBURG: On this slide, the last slide on Page 13, you've got a regulatory oversight, mammography rates, low birth rates, and cervical cancer screenings. Could you explain a little bit more about that?
MS. IEZZONI: Or will that be the next presentation?
MR. RATTERREE: Actually, the next group in here --
MS. KEAGLE: Right, right. They'll have a handout for you, too, that might be helpful.
MR. RATTERREE: Probably one of the -- probably just to add some of the lessons that we've learned in doing this, is really to keep the communication channels open with your contractors, your plans, MCOs. As long as you have the communication channels open, they know what's expected of them, and you, you know, uniformly apply those expectations, I think things run rather smoothly.
MS. IEZZONI: It doesn't sound as if you've been squeamish about sanctioning plans.
MR. RATTERREE: I haven't.
MS. IEZZONI: No.
MS. KEAGLE: They refer to him as Rambo.
MS. IEZZONI: And it seems as if you believe that that has really had a beneficial effect.
MR. RATTERREE: I think, in some cases, you have to apply both the carrot and the stick. Some of the contractors -- it's just like people, some of the contractors respond more to the carrot aspect, and others respond more to the stick.
MS. ARAKI: What are some of the carrots?
MR. RATTERREE: You know, if they submit all the data into us, it can develop better information for them to utilize. They have a better idea what they're going to bid next time around. There's even been talk of if someone is performing exceptionally well for a period of time, that may end up affecting capitation rate. I mean, that doesn't happen now, but --
MS. KEAGLE: We had -- yeah.
MS. CYPERT: And that they wouldn't have to respond.
MR. RATTERREE: Yeah. And --
MS. ARAKI: That would be a big carrot thrown in.
MR. RATTERREE: And there are some things that, when they do a contract, that they don't have to respond to some issues in the contract if they perform very well.
MS. CYPERT: That's one of the things we were going to touch on, as well, how we're using some of this.
MS. COLTIN: What is disproportionate share?
MS. IEZZONI: That's federal for --
MS. CYPERT: Federal -- yeah.
MR. RATTERREE: Go ahead.
MS. IEZZONI: -- hospitals that care for a lot of indigent patients.
MS. KEAGLE: It's a pain.
MS. CYPERT: Yeah.
MS. COLTIN: So it doesn't have anything to do with the health plans, it's --
MS. KEAGLE: Right.
MS. COLTIN: You use the data to help --
MR. RATTERREE: It's just --
MS. KEAGLE: The way it affects it, is the data that we pull to do the calculations, like some money to the hospitals, a big piece of it is the encounters that the health plan submits. So if the encounters don't -- if the health plans don't get them to us, therefore, we don't have them to use to calculate the DHS payments. The hospitals sue us, and it's like this big circle.
MS. IEZZONI: Kathy, it was part of PPS in 1983 to kind of sweeten the pot for public hospitals.
MS. COLTIN: I was just confused, because it was under the data coming from the health plans.
MS. IEZZONI: Yeah.
MS. COLTIN: I didn't realize it was being used then to feed a different process.
MS. KEAGLE: Right.
MS. COLTIN: A totally different process. Okay.
MS. KEAGLE: Okay. I know you want a data book. Do you want the encounter manual, or not?
MS. ARAKI: Do you have it?
MS. IEZZONI: Yeah.
MS. KEAGLE: I'll go get you one.
MS. COLTIN: I think the most important thing is, we'd like to see what the file layout is, just to know what it is you're asking for, and in what form.
MS. KEAGLE: Okay. And, yeah, whether it's required or optional.
MS. COLTIN: Whether you need everything.
MS. KEAGLE: I'll just run up and get a manual and bring it down. Were any of you part of writing that HCFA publication that the did, a guide for states collecting and analyzing Medicaid Managed Care data?
MS. IEZZONI: No.
MS. KEAGLE: No? Have you seen it?
MR. RATTERREE: He has.
MS. KEAGLE: He has? Well, I didn't know if you had, because I didn't want to bring this and then fault you if you had. But, I thought if you hadn't, this had -- we got a copy of this, and it's like a thick four-inch, five-inch binder from HCFA.
And they did a study on, basically, encounter reporting, and it's -- I found it real helpful just even when I was putting together this presentation to do for our national conference, just to kind of get a bigger picture of, other than our encounters, what's everybody else's take on it.
And they -- I mean, they talk about a lot of different areas from addressing the development and implementing, an encounter data system, to designing one, to implementing it with your MCOs, how and why to measure and improve the accuracy and completeness, like our validation studies, and then an analytic section, which discusses ways to use the data to monitor utilization, access, quality, and program expenditures.
And I just thought it was pretty good. So, anyway, I'll just pass this out in case you guys are interested and you want to order one.
MS. IEZZONI: So, Jason, you --
MR. GOLDWATER: You can get it from me.
MS. IEZZONI: Jason's got it.
MS. KEAGLE: Oh, okay.
MS. IEZZONI: No, no, please pass it out, though. That would be great.
MS. KEAGLE: Okay. And do you want us, Kathy, to stick around, or do you want us to? No?
MS. CYPERT: I don't think so.
MS. KEAGLE: Okay.
MS. CYPERT: I mean, you can scram.
MS. KEAGLE: Okay. I'll just pop in with those documents. Okay. Thank you.
MR. RATTERREE: Thank you.
MS. IEZZONI: Thank you. That was great. Very helpful.
MR. VAN AMBURG: Thank you.
MS. KEAGLE: And our phone numbers are in the front of that, if anyone ever wants any more information from us.
MS. ARAKI: The front of what?
MS. KEAGLE: In the front of our handout that we gave you.
MS. ARAKI: Oh, in your handout.
MS. KEAGLE: It has our business cards on it. Okay. Thank you.
MS. IEZZONI: Okay. The wrap-up crew.
MS. CYPERT: The wrap-up crew, okay. Is that higher expectations, or --
MS. IEZZONI: Well, no -- yeah. Yes, a high expectation, because you're going to talk to us about performance measures, right?
MS. CYPERT: Right. Exactly.
MS. IEZZONI: Great.
MS. CYPERT: I'm going to pass out a copy of our quarterly report that we prepare for HCFA, kind of describing our Quality Management Program, and giving updates on where we are with the development, all of our performance indicators.
We were trying to design our presentation based on some of the questions that we noticed you were asking of the panel.
MS. IEZZONI: Thank you. That's good.
MS. CYPERT: So, hopefully we'll get there, and if we don't, please feel free to ask us.
If you'll start on Page 6, what I wanted to do was kind of describe what AHCCCS has been doing the last several years using encounter administrative data to define quality performance measures, or performance measures.
This initial project was part of our waiver continuation as a demonstration project, beginning in 1994. And what we initially did was worked with our health plans and program contractors to define those kinds of indicators that we were going to use that we thought we reasonably could use encounter data, at least we had some sense of confidence in the validity of the encounter data that we could do that. It's been a long hard road, and one of the things I want to talk about later, some of the limitations that we've discovered as we've gone along. But, if you'll look on Page 8, this gives you a sense of some of the indicators that we're looking at for our Acute Care Plans, and then Alan is going to focus on our long-term care, elderly physically disabled indicators.
A lot of our indicators were initially modeled after, but didn't exactly mirror the HEDIS. We started with HEDIS 2.5, then we had -- I'm getting Quizmic and the original HCFA document mixed up, but then we went to HEDIS for Medicaid. Now the iteration is HEDIS 3.0, which includes a lot of the Medicaid indicators. So that's where we are.
We've had a lot of history with childhood immunization rates. We are -- we have produced our baseline indicators for things like well child visits, low birth weight, annual dental exam, mammography screening, cervical cancer screening, and dental -- I did mention dental visits.
We have discovered, though, as we've gone along, because the encounter system was originally based or designed to address payment issues, financial issues, we have some statutory requirements that allow health plans to submit their encounter data. They have up to 240 days. I think Terri probably addressed that, at some point, which is to their advantage, in terms of payment issues, in allowing them to clean up their encounter data, but it's our definite disadvantage, in terms of doing quality indicators, because we're always reporting on something after the fact. It's a year previous.
And what we're finding is, that that doesn't allow the health plans any real time to make timely interventions, to improve their performance levels, so we had to sort of work around that a little bit. And I've got examples of some of our latest studies, where we have used some encounter information, but we found we had to go out and do chart audits, and a few other things to come up with really valid performance measures.
Another major issue, and I think someone here raised that question earlier, about what other kinds of access do you have to information. I think when you were asking about electronic birth records. One of the other major problems that we've had is access to things like demographic data that's collected by our Department of Health Services, for a variety of reasons, but primarily because of federal confidentiality restrictions. Our sister agencies are very reluctant to share information like that, so it's been a pretty significant problem.
I'll give you an example, right now. We had been trying to retrieve some lab data from the Department of Health Services. And what we were asking for was to allow the health plans to get aggregate lab data, and not individual specific that would name individual members, and our Department of Health Services absolutely refuses. They said that their statute -- our state statute prohibits them from doing that.
So the only way that we could get aggregate lab data, would be for the health plan to go back to each of their individual providers and ask them to authorize the release of that data, and then they would have to aggregate again. So it's an unwieldy, unworkable situation. And, as a matter of fact, we're going to try and make a change in our state statute this legislative session to try and deal with something like that. Again, vital statistics, the Department of Health Services.
Pharmacy encounters is another big thing, another huge barrier. AHCCCS does not currently collect encounter data for acute pharmacy expenditures, we do for long-term care, primarily because of the enormity of it, I guess, in other words. Just, we don't have the capacity to store that amount of information. So what we're having our health plans do is report that information to us, rather than take it from our encounters. In most of those health plans, you use external entities to do their pharmacy encounter information. So, again, it's kind of unwieldy. We have to go through a lot of iterations and, you know, changing things around.
One of the other difficulties, and it doesn't relate so much to encounter data, but that we've discovered with performance measures, is there are no regional or national bench marks for us to measure ourselves against. So we've always sort of been right out there not knowing exactly where we were. That isn't really a barrier, but is something to consider for other states, or it's something to consider for any organization that's interested in comparing Medicaid states with one another. And I don't think we've gotten very far with that.
I feel like I'm rambling. Is there any specific --
MS. WARD: No, you're doing fine.
MS. CYPERT: -- any specific questions that you have that might help me address something?
MR. VAN AMBURG: In your measures that you have here, like immunization, and low birth weight, and what have you, there's obviously people in the program that started the program, they are then dropped from the program.
MS. CYPERT: Right.
MR. VAN AMBURG: What percentage are you losing in the denominator on these?
MS. CYPERT: Hmm. I don't have that specific number. One of the ways we have to find, for instance, our childhood immunization rates, as part of the definition we're requiring continuous enrollment, and that's -- but, to answer your specific question, I don't know the number. The expert on that is Dr. Abujbara, and she wasn't able to be here today. But we can find out for you.
MR. VAN AMBURG: How frequently do you recertify eligibility?
MR. SCHAFER: How often do we what?
MS. CYPERT: Certify eligibility.
MR. VAN AMBURG: Recertify eligibility.
MR. SCHAFER: I think it's --
MS. CYPERT: Continuous enrollment. Twelve months, I think.
MR. SCHAFER: Some of them are a minimum of six months, and then others -- most everything's annually.
MS. CYPERT: Yeah. Yeah. That's one of the -- that is an advantage that AHCCCS has had, when we were defining these indicators, is that we do have locked in enrollment.
MR. VAN AMBURG: Um-hum.
MS. CYPERT: That made a big difference because, otherwise, you have that in-and-out movement. But the specific percentage that we capture in our indicators, I'm not sure, but I can find out for you, or you can call Dr. Abujbara.
MR. VAN AMBURG: No, I was just curious.
MS. IEZZONI: We heard --
MR. VAN AMBURG: Because this is a problem.
MS. CYPERT: Yes, it is.
MS. IEZZONI: Yeah. We heard this morning that the State is starting an immunization registry. And, is that something that you anticipate using for some of this?
MS. CYPERT: Yes, ultimately, we would.
Let me just talk about my limited knowledge of some of the difficulties with immunizations. But right now, we have a statutory requirement to do annual immunization. We're serving two purposes with this.
MR. VAN AMBURG: Um-hum.
MS. CYPERT: And we have found that primarily because there are so many other places where children can get immunizations, like health fares, or at school, or something like that, that we have had to do chart audits in order to do this immunization. We can't -- we haven't been able to do it through encounters.
But, to get back to your system, it's called ACES (phonetic). I can't remember what the acronym stands for; but, supposedly the Department of Health Services has been working on this for the last good three or four years, and they will ultimately require all providers to send in their immunization information to them. And we do anticipate using it, but we do have some concerns. It's slow getting off the ground, put it that way. But, I think DHS is going to add some incentives to have providers report that information to them.
MS. IEZZONI: Do you an -- I'm sorry.
MS. ARAKI: What about com -- oh, excuse me.
MS. COLTIN: Do you anticipate any data sharing issues around that, like you described on the enrollment information, the demographics, and --
MS. CYPERT: No. This is a little bit of a different ballgame, I guess, one way to say it. Where we run into trouble are things like HIV AIDS reporting, I think where people have more concern about the confidentiality of the member information, but I don't think we're going to have trouble with immunization. We're more concerned about them getting it up and getting going with it, and testing it, to make sure that it's actually validated.
Because one of the things that AHCCCS did differently than HEDIS, or some of the other states is, we validated our own data. We required -- we did not require the health plans to send their encounter data to us. We took the encounter data that we had and validated it internally first, before we released it back out to the health plans. We just wanted to be absolutely sure we were getting what we were thinking we were describing.
MS. COLTIN: There are two different approaches to using the immunization data from immunization rates. I mean, one is just to look at the whole population and say how are we doing at getting children immunized. And the other is to assign accountability to a health plan and say how is this health plan doing. And, clearly, the continuous enrollment is necessary when you're going to do the latter; it's not necessary when you're doing the former.
Would your agency do both of those types of analyses --
MS. CYPERT: Um-hum.
MS. COLTIN: -- or would Health -- oh, okay.
MS. CYPERT: Yeah.
MS. COLTIN: All right.
MS. CYPERT: We do do both. And, as a matter of fact, I brought a copy of the most recently completed study, immunization study.
What we do is, we report the state aggregate, and then the health plan specific rate. We don't release -- this is released to the legislature, but then the health plan specific data is given to them. It's also included in our RFP, our contract requirements, that they need to show a specific percentage of improvement, or we require corrective action plans, that sort of thing.
That's kind of where I was getting. What we've started to do is use these pharmacy measurement results very tentatively to hold the health plans accountable. We wanted to take it slowly, because we wanted to give everybody a chance to make sure that they knew where we were coming from. We're not using this as a heavy stick. This was a cooperative venture, on our part.
But in our last RFP cycle, we did allow some of the health plans who had high immunization rates; for instance, to not have to meet some of the requirements of the RFP. That was sort of a carrot that we could offer them. And that's where we would prefer to go, at least, you know, initially, until everybody gets very used to this.
One of the other limitations that I wanted to bring up particularly pertains to prenatal care. And, again, it relates back to the fact that the encounter systems were originally structured for payment. And we've structured our encounter system so that providers do bundle billing, global packaged.
And so when it came to trying to figure out how many individual visits a woman was getting, or when she initiated prenatal care, it was impossible.
MS. COLTIN: I didn't set you up to say this, but --
MS. CYPERT: And so we had to backtrack and assign specific indicators to prenatal visits, and then we had to train the providers to use those specific modifiers on our indicators, and we had to give them a year practice to make sure they're doing it, then we had to validate that they were, indeed, doing it.
And so, just now, after three years, we're beginning to develop our prenatal indicators.
MS. COLTIN: So you're doing this within the context of the encounter file format?
MS. CYPERT: Right. Yeah. We came up with some specific modifiers pertaining to trimester and entering into care, and then we tested that.
MS. COLTIN: Are these standard fields that, like, would be on a 1500, or are these new fields that you created specifically for this purpose?
MS. CYPERT: I think they're standard fields with modifiers, but I'm not entirely --
MR. SCHAFER: They're AHCCCS specific codes, aren't they?
MS. CYPERT: They're AHCCCS specific, yeah.
MR. SCHAFER: State specific codes, and I can't remember how they're done.
MS. CYPERT: Yeah. But I'll be glad to find out for you. Yeah.
MS. COLTIN: This is a problem we hear all over --
MS. CYPERT: Oh, I'm sure.
MS. COLTIN: -- not just Arizona.
MS. CYPERT: It made sense, from a provider's viewpoint for billing purposes. It was clean, and easy, and familiar, and everything. But, for performance measures --
MS. WARD: Of course, laughing at the size of the handout, the --
MS. CYPERT: Yeah.
MR. SCHAFER: Small data book.
MS. CYPERT: You asked for it.
MS. IEZZONI: You have to carry that back with you. I know.
MR. GOLDWATER: No.
MS. CYPERT: So he needs a truck.
MR. SCHAFER: Kathy will box it up for you.
MS. CYPERT: Yeah.
MS. ARAKI: I have a question.
MS. CYPERT: Yes.
MS. ARAKI: Do you receive grievance reports from -- or, I guess grievance reports, or some consumer satisfaction reports and any grievances from the health plans, and if so, do you use any of that in any of your analyses, feed that back to the health plans for their accountability and improvements of services, or even using it in your next go-around when you do contractor negotiations?
MS. CYPERT: Well, I can't answer that, specifically. Alan, maybe you --
MR. SCHAFER: I can maybe answer a little bit of that. In regards to grievances, when we go out and do what we call our operational and financial reviews, and we do that annually on all the health plans, and the program contractors, the acute care MCOs and the long-term care MCOs, they look at those. They do an area of the grievance and appeal process and look at and making sure they're handling those particular issues appropriately.
We, in the Office of the Medical Director, in the various, the acute care unit and the long-term care unit, we have our own tracking and trending system for quality of care complaints that come in on the health plans and program contractors, and institute, require plans of corrections, systems, and member-specifics, and so forth.
AHCCCS has done a member satisfaction survey. I don't know if you were thinking along those lines, at all, or not.
MS. CYPERT: Were you --
MS. IEZZONI: Well, we would actually like to hear about that, if we could, because one of the things we've heard over the last day, is that the member satisfaction survey submitted by the individual plans differ from plan-to-plan, but that you also, at AHCCCS, have done your own independent survey.
So, is it okay, Lynette, if we talk a little bit about that?
MS. ARAKI: Yeah, sure, that's --
MS. IEZZONI: Yeah.
MS. CYPERT: Can I -- were you talking about grievances pertaining to the quality of care, or grievances with issues with health care coverage? I wasn't clear.
MS. ARAKI: Both.
MS. CYPERT: Both. Yeah. So I think Alan --
MR. SCHAFER: Yeah.
MS. CYPERT: -- kind of addressed where we go with quality of care --
MR. SCHAFER: Well, the grievance and appeals unit, they go out as part of the staff to do the operational reviews. And they do some quarterly, monthly-type reports on grievances that are handled, I think, by the health plans and PCs, as well as what we handle at the AHCCCS administration level in the Officer of Grievance and Appeals, but you'd have to --
MS. CYPERT: I don't think they used it in the last RFP cycle, though, any results --
MR. SCHAFER: No.
MS. CYPERT: -- specific to that, so --
If you'll turn to Page 13 in the handout -- oh, I'm sorry.
MS. COLTIN: I have another --
MS. IEZZONI: Yeah, we just wanted to follow up on the satisfaction survey.
MS. COLTIN: Before you move off of this onto the satisfaction, though, I have a question --
MS. IEZZONI: Okay.
MS. COLTIN: -- on what we've just talked about.
You made a statement about the desirability of having bench mark information on some of these indicators, and not being able to get it. Is that because other states are not collecting these data, because other states are changing the definitions, rather than using a standard HEDIS definition, or is it because the eligibility categories are different from state-to-state, and you feel that the measures aren't comparable, or is it another reason?
MS. CYPERT: No, it's sort of all of the above. AHCCCS is fortunate that we have the greater length of history with dealing with encounters. Most other Medicaid states are moving from fee-for-service, to a managed care type of situation, and don't have that ability or history of collecting encounter data. And, so, we're comparing apples to oranges.
A lot of other states don't necessarily have the locked-in enrollment issues. They may have bits and pieces in managed care for some parts of their population, but not overall. So, you know, this is something that NCQA, and HCFA, and the various other Medicaid agencies are struggling with.
MR. SCHAFER: And we do --
MS. CYPERT: And that's particularly true for Alan, so.
MR. SCHAFER: -- we do have an ALTCS side of the program that's even more of a problem for us. We, number one, have limited indicators that we actually collect off the system. We do a number of -- we collect data a number of ways. We have the QM departments of the program contractors collect data, we've used our medical eligibility pass assessors, collecting data off the system -- not off the system, when they go out and -- at the time they do the eligibility review.
And, so, we've looked at ways to go out and collect this data, but also to minimize the additional costs for labor, it's very labor intensive to do a lot of these things. And then, we just -- there are no -- there really are not limited standards out there, in regards to that at-risk nursing facility population of what to compare back to. But we're forging ahead, and just really just starting the second year of collecting data in our long-term indicators.
And, so, we don't know if they're valid. We did the same process as the acute folks did in gathering good participation from financial people, program contractors, QM. It was just a good mix of different levels of experience and tried to come up with the best approach that we thought would give us some reasonable information to look at, and then from there, decide if it was something valid that we could affect some good changes with, or to just see where we're at, in general, from a practice outcome level.
MS. IEZZONI: So are those what's on page --
MR. SCHAFER: Ours are on the -- the ALTCS ones are on Page 10.
MS. IEZZONI: 10?
MR. SCHAFER: Start at the bottom of Page 10.
MS. IEZZONI: George had a follow-up question to Kathy's --
MR. VAN AMBURG: Yeah. Can you give me some examples of bench marks you'd like to have, that you don't have?
MS. CYPERT: I know what -- where we started the process. We used healthy people 2000 goals, and that seems to be the most commonly accepted across states.
I don't know what to say about what we would like to have, because there's so many different structures in different states. I don't know what they could produce. But, I mean, obviously, things like immunization rates would be a critical forum, and that's where we get closest to having a bench mark.
I don't know. Prenatal care is a big issue for us, you know, out west. I think Juman probably would have better ideas about that. Would you --
MR. VAN AMBURG: I was just kind of curious whether you were using healthy people 2000, and --
MS. CYPERT: Yeah.
MR. VAN AMBURG: -- of course, city-to-city has promulgated what they think should be for immunization rates.
MS. CYPERT: Right. And that's --
MR. VAN AMBURG: And, have you thought about comparing to some of your state data, and perhaps risk adjusting some of it?
MS. CYPERT: Now, that's a whole different ballgame. Because we didn't really get away, we kind of shied away from risk adjusting when we initially defined these, but it's certainly becoming more and more evident that that's something we're going to have to address.
And it's hard for people to figure out exactly how to do that, you know, particularly given the limitations to sharing of data. I mean, if we can't get population descriptions, and things like that, it's going to be very difficult to do any risk adjustment. But I know that is something --
MR. SCHAFER: On the ALTCS side, we have a little bit -- we're a little bit better --
MS. CYPERT: Yeah.
MR. SCHAFER: -- when it comes to getting demographics, because everyone has to have a pre-admission screening to get into the program, so you have all -- you have the ethnicity, the --
MS. CYPERT: Um-hum.
MR. SCHAFER: -- and everything that you really want, you have a lot more on those people, so you can sit there, and do they live in a home or do they live in a nursing facility, do they live in an alternative residential setting. So, sometimes our indicators get beyond just looking at one thing. We start -- we're able to at least analyze a lot more from things other than just sex and age ranges.
MS. CYPERT: Yeah.
MS. IEZZONI: So, for nursing home residents, do you rely on the minimum data sets that HCFA promulgates?
MR. SCHAFER: We will be, once it comes up automated in the state of Arizona. We're expecting sometime in June, whenever the final -- those become -- are supposed to be implemented. ADHS licensure has all the computer equipment in, they've been playing around with it a little bit, and they're getting ready to do a statewide training to all the nursing facilities. So, we're really looking forward to the availability of that information.
Then the next thing we want is the standardized data for home health agencies, but that's going to be limited because, we were talking about this yesterday, is that home health -- not all of our members get services from home health agencies. They'll get services through unlicensed personal care, attending care agency, housekeeping, family members, a brother is the attending care provider for this member.
Well, we're not going to have that baseline data. But we still have the path data on it, but -- so there are some limits on the home based side, and that's the program side that we're trying to grow. And we're at about 39 percent now in our program. We started less than ten. And so every year we get it increased, but we want to make sure -- we have to make sure we can continue to provide the safe care in that area, and monitoring that becomes real critical. And so what data are we looking at to make sure it's happening.
MS. CYPERT: Yeah.
MS. IEZZONI: Are you watching what the federal government is doing with home health to -- for the Medicare Program, promulgate certain data collection processes? Is that something that you're following?
MR. SCHAFER: Well, I'm not real close. I know -- I mean, the only thing I'm aware of, really, is the Oasis is the main thing, if that's what you're speaking about.
MS. CYPERT: Is that that Quizmic? Were you talking --
MS. IEZZONI: Yeah. Yeah. Now, the Oasis, is that going to be something that you'll find useful?
MR. SCHAFER: I'm not sure. I haven't seen enough of it to know. But if it's similar to the MDS, yes, it could be. But, again, when I look -- what's the other one, the SF --
MS. IEZZONI: SF36 is a functional status member that --
MR. SCHAFER: That doesn't do any -- some people have asked us to look at that. That does no good for an at risk population. And it's amazing how some people have commented and said use this, it's really great. And I've looked at it, and I'm at a loss of how I could effectively try to incorporate it into the process. It's --
MS. COLTIN: It's people in Mateo.
MS. IEZZONI: Um-hum. Yeah.
MS. COLTIN: It doesn't discriminate.
MS. IEZZONI: Yeah. You'll not get much disagreement around here.
MS. COLTIN: No.
MR. SCHAFER: Well, that's good. I'm glad to hear that, because I didn't know if I was just being close-minded about it, or what it was --
MS. CYPERT: You were being stubborn.
MS. IEZZONI: No.
MR. SCHAFER: -- because I've had some people that are dealing with it on a research basis who really are promoting it.
MS. IEZZONI: Research is different.
Okay. Susan, can you tell us a little bit about the satisfaction --
MS. CYPERT: Sure. It starts -- the description starts --
MS. VALLEY: Let me just mention, I went ahead and got a survey --
MS. CYPERT: Oh, great.
MS. VALLEY: -- and summarized it really nicely, so one document you can take with is a summary of the actual survey.
MS. IEZZONI: Okay, thank you.
MS. CYPERT: Right. When we did the member satisfaction survey, that wasn't actually our first one. We focused on prenatal care, satisfaction with prenatal care, sort of a little mini to test what our effort was going to need to be to do the member satisfaction survey.
We did complete that survey in April of '96. We surveyed more than 14,000 members. It turned out to be exceedingly expensive to do. We were fortunate to get it partially funded through a grant from the Robert Wood Johnson Foundation. We ended up using a telephone survey. And, as Kathy said, she has copies of the survey questions.
We used a telephone survey because so many of our members -- you know, mail-in surveys would not work with the Medicaid population at all. And we worked through Arizona State University. They have a telephone research center there. They have a great deal of experience working with Medicaid populations. We did focus groups, to begin with, before we started the actual survey to kind of test what questions they would respond to.
And, I think we looked at household, I think that's how we ended up tabulating the data, rather than individual members. But, overall, as you'll see in here, the results were quite good, and that was very gratifying, you know, for us. We did run into some significant data gathering issues. It took a lot longer to do it than we thought, and a whole lot more money, so --
MS. IEZZONI: How does this make you feel about the satisfaction survey that each of your plans are required to submit to you every year, your own experience --
MS. CYPERT: I --
MS. IEZZONI: -- how does it make you feel about what the plans submit?
MS. CYPERT: Again, I think Dr. Abujbara would be the better person to respond to this, since she is our acute care manager.
MS. IEZZONI: Oh, um-hum.
MS. CYPERT: She would see those actual surveys. I don't see them.
MR. SCHAFER: On the ALTCS side, we look at those, but they can be pretty on the simplistic side.
MS. CYPERT: Yeah.
MR. SCHAFER: I mean, we do a member satisfaction survey, we go out and do annual case management service reviews, we do random samples on each of the program contractors. We ask some real basic questions, and we're generally finding satisfaction above 90 percent with providers, types of services, amount of services, and case management, but we do find issues.
But, again, it's trying to do something that you can do in a short amount of time, and doesn't get overly burdensome on the limited, you know, staff hours available to do certain processes, and to put it all together.
MS. COLTIN: Are you familiar with a consumer assessment of health plan survey for Medicaid that was developed by the agency for health care policy research consortium?
MS. CYPERT: I'm not, personally.
MS. COLTIN: There are six states that have already implemented it, and a number of them are going to require it this coming year. And when you had mentioned bench marks earlier, this is an area where, you know, using a standardized survey would allow you to compare with other states.
And there are lots of differences in how states are implementing it, actually. Some are being implemented by the state, and they're paying for it. Others are passing the costs actually onto the health plans and saying you must use an independent vendor, and you must do it, but you must use this survey instrument and follow this protocol, you know.
MS. CYPERT: Now that you mention it --
MS. COLTIN: And then the plans have to pay for it. But then it --
MS. CYPERT: Yeah.
MS. COLTIN: -- it'll end up coming back to you, probably, in the capitation, but, still --
MS. CYPERT: We worked with Dr. Jean McGee to design the survey questions, and the last time she was here, she did mention that.
MS. COLTIN: And she's been on that, yeah.
MS. CYPERT: But I didn't have a chance to explore it any further with her.
MS. COLTIN: She has worked on that project, too.
MS. CYPERT: That's what I thought.
MS. COLTIN: Yeah.
MS. CYPERT: Could I get the name of that again from you?
MS. IEZZONI: CAHPS.
MS. COLTIN: The Consumer Assessment of Health Plan Survey, C-A-H-P-S, CAHPS. And there is a version for the Medicaid population. And I believe it's in Spanish, as well as English.
MS. CYPERT: We are now embarking on a provider satisfaction survey, and we'll again be working with Jean McGee on this. We're just now starting the discussion whether it should stay focused on acute, and acute and long-term care, and so we're just now in the planning stages for it, but --
MS. COLTIN: Just for your information, their users group conference is being held today and tomorrow here in Phoenix at the Sheraton Crescent.
MS. CYPERT: Oh.
MS. COLTIN: And there are some states that are reporting on their experience using this with the Medicaid population. Probably missed it.
MS. IEZZONI: Kathy's going to be going to that meeting tomorrow morning.
MS. COLTIN: Tomorrow. Just putting in a plug.
MS. CYPERT: It didn't ring a bell at first, but now, when you were describing it, I remember her talking about it.
MS. COLTIN: I mean, I know Massachusetts is going to require it for us to do.
MS. IEZZONI: Are there any other things that -- Alan, we didn't really give you a formal opportunity to present. You've been chiming in.
MR. SCHAFER: Well, I think just from the perspective of your looking at how do you get data that's useful across all states, I think ALTCS probably has -- long-term care, I think, is a little bit more challenging than what's out there, although, with the MDS now, it's going to help with that population.
But now you have states moving away from nursing facility placements, and expanding in the home community base, so that's going to dwindle down to percentages of 50, 60 percent of the long-term care populations will only be residing in nursing facilities, and so what are we going to do in those other areas, and how are we going to track that. And those are -- that's very time consuming to track that information, if you're going to do something --
MS. IEZZONI: Do you feel that your legislature is giving you adequate funding for doing this, for thinking about it? Do you have a staff that you need to be able to think about this?
MR. SCHAFER: I think in a lot of ways, you know, I'll say for my unit, I think we do very good with staffing. There's always more that you want to do, but I can't complain about the staffing that we have.
MS. CYPERT: He can't, because I'm his boss, so he can't say anything about it.
MR. SCHAFER: We're able to do a lot, and it's something that Brent said that's so key, no matter how you cut it and what you're looking at. If everything can't be uniform, it's the State Medicaid agencies working very closely, and being able to work very closely with the managed care organizations.
And we go out there and do a lot of monitoring and work very closely with them, have very good relationships, and can effect change when it's necessary and, to me, that's just so key to the whole process. And having other states comes in to speak to us here about our program, and they talk about how they only have "X" amount of staff, I do not know how they can effectively implement Medicaid Managed Care.
If you don't have the oversight, you don't have any idea what's going on out there, you don't set the standards, you don't set what's expected, and someone has to be that standard bearer out there, and it has to be the State Medicaid agencies.
MS. CYPERT: That is absolutely true, because in talking with other states, they always ask, well, how many staff people do you have, or how many administrative staff people, how big is your ISD Division; 500 people. I mean, that is --
MS. ARAKI: Five hundred people, is that all it is?
MS. CYPERT: Yes.
MR. SCHAFER: ISD, no.
MS. CYPERT: Well, no, ISD is not 500.
MR. SCHAFER: No.
MS. CYPERT: Division of Member Services is 500. ISD is several hundred, right?
MR. SCHAFER: A couple hundred, a hundred and eighty, a hundred and fifty.
MS. CYPERT: Yeah. So, I mean, it's an interesting comparison.
MS. IEZZONI: And are they all operational in Phoenix, or are they spread throughout the state? Are there offices elsewhere? Because, it's a big state.
MS. CYPERT: Information Services.
MS. IEZZONI: Yeah.
MS. CYPERT: It is, yeah.
MS. IEZZONI: And so, do you just have a big travel budget so you --
MR. SCHAFER: Well, Information Services is here --
MS. CYPERT: Is here, yeah.
MR. SCHAFER: -- but, basically, Division of Member Services, they're the -- they have the -- the reason they're the largest, is they do the medical eligibility and financial eligibility for the program, so they have, I think, 13 to 15 local offices throughout the state.
So they're placed out there, and they do a lot of traveling, obviously, because we go to the member's home, or where they're living to do the medical, and sometimes they'll even go out to the home, or hospital, nursing facility for the financial, it just depends upon the situation. And so, in that way, there's a lot of travel.
But, I mean, our staff are out there quite often, sometimes a week, week and a half, a month out in the field doing audits on program contractors, and sometimes we do spot quality of care, work with nursing facilities on particular issues when they seem to be getting out of hand.
MR. VAN AMBURG: What's your ratio of staff, compared to other states in Medicaid? It sounds to me like it's very large, compared to other states.
MR. SCHAFER: The state of Arizona has over, what, 1100 --
MS. CYPERT: Um-hum.
MR. SCHAFER: -- employees.
MS. ARAKI: In Medicaid?
MR. SCHAFER: In Medicaid, but that includes people who do the -- ALTCS long-term care eligibility do a little bit of acute care eligibility, but there's a lot of verification staff on the -- that do a lot of quality reviews related to the acute care eligibilities where there's staff that go out and do that.
MS. CYPERT: In our office alone, we have about 54 people, and that's the Office of the Medical Director. That's not our only world, is performance measures, and things like that, but it's certainly a big piece of what we do.
MR. VAN AMBURG: You're doing almost everything internally, except for perhaps your survey, which requires some other expertise.
MS. CYPERT: Right.
MR. VAN AMBURG: Other states that I know about are looking at going from fee-for-service to managed care to reduce their costs on the medical care side, but also reduce their costs on the state personnel side.
And what you're saying, is that really isn't going to happen, if they're going to do it right.
MR. SCHAFER: I don't know how you --
MS. CYPERT: That would be my advice.
MR. SCHAFER: -- set there and have an independent group go out there and say go do an audit here, and go do an audit there. They don't work with those people in the situations day-to-day, and you can't -- if you don't know what the problems are, the potential issues are, the new programs that you've implemented, when you go out and do auditing, that's when you perk your ears up and you start saying, well, next year I better add this to the process, because we're missing it here, and no independent group is going to pick that up unless they happen to be the same people you're using review after review after review, and then you might as well have them in-house. I think we get a lot more accomplished and a lot more done.
We do use --
MS. CYPERT: We do use --
MR. SCHAFER: -- I mean, we use for medical audits and stuff we will --
MS. CYPERT: Yeah.
MR. SCHAFER: -- but we're very key to that process as it goes along
MS. CYPERT: Right. We do use our pro organization to do -- we still have a 60,000 fee-for-service population that we use them to do concurrent review and utilization management for us. And then as Alan said, we use them for specific audits. They do the data collection, data validation.
We also use them for quality of care issues, where we have peer review concerns, or that sort of thing. It's not cheap. I mean, it is going to cost us.
MS. IEZZONI: Now, if you're going to make a comment, you have to come to --
MS. VALLEY: I'm doing administrative stuff again. Earlier, with the other group, there were some questions that dealt with information systems questions. And I do have someone available in our Information Systems area, if there's time and you want to ask questions specifically to an ISD-type person. I didn't know how you were doing on quality indicators --
MS. IEZZONI: Okay.
MS. VALLEY: -- and if you think you were going to be busy until 4:00. Do you want me to call in this person?
MS. IEZZONI: No. I think we're fine.
MS. WARD: I don't have any questions that I --
MS. COLTIN: Yeah.
MS. IEZZONI: Yeah. I think we're fine here. Yeah, this is it.
MS. VALLEY: Okay. I just wanted to make that available.
MS. IEZZONI: This has been very good. Thank you very much for offering that to us.
Yeah.
MR. VAN AMBURG: Completely different topic. Everybody that we've talked to has talked about the encounter data system, and I have to assume you're building a longitudinal record for each of your clients. Are you doing that?
MR. SCHAFER: Um-hum. We know it's -- yeah, we see them all the time.
MR. VAN AMBURG: So, Johnny Smith has been in five years, you've got all five years or the records tagged in a longitudinal file.
Do the patients have access to that information and an opportunity to correct it, if they think there are problems.
MR. SCHAFER: I don't believe there's any outward -- I mean, process in place that we actually go out and do that. I --
MR. VAN AMBURG: No, no. I mean, but if --
MS. CYPERT: But through the health plans --
MR. SCHAFER: Well, I think if they --
MS. CYPERT: I would imagine they would.
MR. SCHAFER: -- I think if they identified a problem, or an issue, or concern, yes, I think --
MS. CYPERT: Yeah.
MR. SCHAFER: -- they're open to their input on many, many things. We get lots of phone calls from members.
MS. CYPERT: Yes. Yeah, and again, through their PCP or, you know, in long-term care, if they have specific concerns, I think, by statute, they are able to see their medical record. Now, that's not the same, necessarily, as their encounter record. So, I'm not real clear where your concern was.
MR. VAN AMBURG: No, I was more interested in their total longitudinal record.
MS. CYPERT: Oh.
MS. ARAKI: Where is the total longitudinal record, is it here, or is it at the provider?
MR. SCHAFER: Well, a combination of it. We have all the encounter, the medical eligibility, financial eligibility, service plan -- and on the ALTCS side, we also have service plan authorizations on the system, and then you'd have the actual file out at the PCP that would contain a lot of information.
And then if it's an ALTCS member, nursing facility, home health agency. So the record is spread all out. But on the ALTCS side we have a case -- every member has a case manager and they should encapsulate, basically, the primary activities and issues going on with that person.
MS. ARAKI: So conceivably, the member could get the information, or ask to see the record with the case manager.
MS. CYPERT: That stat information, right.
MR. SCHAFER: Right, um-hum. For an ALTCS member, your case management record is where you'd be best to start off with; or, if they're a nursing facility resident, that nursing facility.
MS. MOIEN: What unique identifier is used to follow them across? Suppose they change plans, they move from Phoenix to --
MS. CYPERT: They have a specific AHCCCS I.D. number that's assigned to them when they're enrolled.
MS. MOIEN: Okay.
MS. ARAKI: Is that their Social Security number?
MS. CYPERT: Yeah. No, it's a --
MR. SCHAFER: It used to be, and now it's an AHCCCS specific number.
MS. CYPERT: Yeah. We do run into problems like, you know, in the Hispanic culture there are a lot of very similar names, like Rodriguez. I mean, we must have 10 million Rodriguezes, but that's -- is that one of the reasons they switched from Social Security, or --
MR. SCHAFER: I'm not sure why they switched --
MS. CYPERT: Okay.
MR. SCHAFER: -- from that, except that they did one year.
MS. CYPERT: But it's a unique number. Yeah, I don't know very much about longitudinal records, in terms of our encounters. I do know we have the large system, and then we have lots of subsystems, and we use the AHCCCS I.D., you know, to create the record, eligibility, we have all kinds of eligibility codes so that we can track what happens to an individual over time, you know, when they come on, when they come off, if they change eligibility status to a different kind of eligibility, like SSI.
MR. SCHAFER: In the ALTCS Program, we even know if they've been in a home setting, alternative residential setting, nursing facility over time, so we have a little bit more specific tracking --
MS. CYPERT: Yeah.
MR. SCHAFER: -- than you would on the acute side.
MS. CYPERT: Yeah.
MS. COLTIN: How many years of encounter data do you keep on-line at any one time?
MR. SCHAFER: Keeping --
MS. CYPERT: I don't know.
MR. SCHAFER: I'm trying to remember what that is. They've gone to taking some of that data off the system and just keeping the essential information. I think for on the long-term care side, they are starting to take off, I think, '92 or '93, so maybe it's five years, is what they're keeping on the system. And then --
MS. CYPERT: But it is archived, right?
MR. SCHAFER: -- but it's archived --
MS. CYPERT: Yeah.
MR. SCHAFER: -- and we can get at that data when we need to get at it. But, usually -- I've rarely run into a situation where you really need to go back further than that, really.
MS. IEZZONI: I think I notice a petering out of our Committee.
MS. ARAKI: Just kind of end of day, huh?
MS. IEZZONI: Yeah. And yesterday was very productive, as well. Are there any other questions from the subcommittee, or others?
This has been extremely helpful. You know, again, we'd come to Arizona because we hear that you guys are among the best, and so we wanted to learn from you, and we have. So thank you very much for sharing time with us.
MS. CYPERT: It was a pleasure.
MR. SCHAFER: Yes.
MS. CYPERT: And I took most of his time.
MS. IEZZONI: Thank you.
MR. SCHAFER: We do things pretty much the same.
MS. IEZZONI: Yeah. Okay.
MS. CYPERT: Yeah. I'd like to leave you with these copies of our latest reports. I think they're fairly self-explanatory.
MS. ARAKI: They're very informative.
MS. CYPERT: Oh, good. Thanks.
MS. IEZZONI: Thank you.
(Whereupon, the proceedings were concluded at 3:47 p.m.)
CERTIFICATE OF REPORTER
STATE OF ARIZONA )
) ss.
COUNTY OF MARICOPA )
BE IT KNOWN that I took the foregoing tape recorded proceedings; that I was then and there a Notary Public in and for the County of Maricopa, State of Arizona; that this is a true and accurate recording and transcription of the proceedings consisting of 148 pages, done to the best of my skill and ability.
WITNESS my hand and seal of office this
day of February, 1998.
Angela T. Thornton Notary Public
My Commission Expires: