July 9, 1997 Public Forum: Coding and Classification


DR. MOORE: It is time again. I think we can finish this before five. There was a question asked during the break about the meeting that is coming up in August that was referenced at the last presentation. That meeting is an ANSE X12 meeting that is taking place in Rochester on the 10th through the 14th of August. That is the meeting where these implementation guides will be discussed and further refined. So those of you who are interested in implementation guides should go to that meeting.

As this implementation effort, we are not holding a special meeting on implementation guides. The only other meeting that we have referenced in this session that is taking place in August is one that is being held by the National Committee on Vital and Health Statistics that Dr. Detmer referred to earlier, and I think that is the 5th and the 6th.

The next team for a presentation is the one on coding and classification. This is the medical coding and classification, not all the other codes that might be necessary in some of these transactions.

We have three people who have been co-chairing this effort: Betsy Humphreys from the National Library of Medicine, Pat Brooks from the Health Care Financing Administration, and Donna Pickett from the National Center for Health Statistics. I think Betsy is going to be making the presentation, and I'm sure that we will have a lot of questions when that is over. Thank you. If we can speed it along, we will be out of here early.

DR. HUMPHREYS: As Bob said, this is the coding and classification implementation team. Just to remind ourselves about what we are talking about here, we are talking about codes and classifications for diseases, injuries, impairments or other health problems, or as we might say, whatever is wrong with the patient, the causes of these conditions, which are represented in some of the administrative transactions, such as first report of injury and so forth, and actions taken to prevent, diagnose, treat or manage these conditions, aka procedures, and also we are including obviously the substances, read drugs, and others, equipment and supplies used in the procedures for diagnosing and treating the conditions. So these are the major areas that are covered by this team.

In setting up the charge for this team, the department has looked at this team as one that obviously has immediate responsibilities for the initial administrative standards, but because of the large number of vocabulary issues that relate to the longer term agenda of providing recommendations under Kennedy-Kassebaum to the Congress related to electronic medical records, this team has some responsibilities that relate to that as well.

Our first, and the one that we have been focusing on to date, given the timetables here, is to identify and select the standards for codes and classification for the administrative transactions that are covered by the act, and to insure that there are appropriate mechanisms for distribution and maintenance for the selected standards.

We also see as a longer term effort recommending the set of health vocabularies that are useful for full electronic health records, and looking at the issues related to appropriate mechanisms for distribution and maintenance of those. Then we see also as a responsibility that the department should have to insure that there is some sort of a reasonable map between the more granular vocabularies that will in the future be used in the full patient record, and the administrative codes and classifications which we assume will continue to be necessary for statistical and administrative purposes.

However, having said that, that is the whole charge. But now what we are going to focus on is the first part of it, which is what we have been working on up to date, which is the selection of the initial standards for administrative transactions.

You will not be too surprised by these recommendations. Let me tell you where they came from. There was a combination of inputs here, including the input from ANSE in terms of the inventory of existing standards, information that was known to many of us within the department who have a long history with the various codes and classification systems for other projects, and then primarily we took advantage of, and actually assisted in planning and setting up questions and so forth for the National Committee on Vital and Health Statistics hearings on this particular topic, codes and classifications, which was held April 15 and 16.

So in essence, what we are recommending is based largely on what we heard during those hearings. The overview is that everyone said that for the year 2000, it would be -- I don't want to say everyone, but the great preponderance of testimony said that for the year 2000, we could not deviate from the status quo of what is generally used in health care transactions now. The recommendations therefore for the initial standards follow that.

That means for the diseases, injuries, et cetera, we are dealing with ICD-9CM. As you can imagine, one of the big issues that was addressed at the hearing was whether we should recommend for the year 2000 9CM or 10CM. After evaluating all the testimony that was received on this point, although there were some proponents of moving ahead to 10CM, there were in essence a larger body of opinion that ICD-10CM could not be reasonably implemented by the year 2000, and that therefore we should go with 9CM as the initial standard.

In terms of procedures, we are again with the status quo. THErefore, although there was a great amount of testimony in favor of moving to an integrated single procedure framework or system, where there was a closer relationship at least between any different levels of systems that were needed for different purposes, everyone was pretty well unanimously telling us that for the year 2000 we had to stick with what was in use now, which is ICD-9CM, the third volume, CPT for physician controlled procedures and out-patient procedures, CDT for dental procedures, and HCPCS, which is the HCFA common procedure coding system, which as those of you who use it know, encompasses CPT and CDT.

For drugs, for most administrative transactions, we are dealing with HCPCS. It has a relatively coarse level of granularity for drug coding, and for pharmacy transactions, where the choice is the NDC code, the national drug code. When we get to the device area, again we are dealing with HCPCS.

So we are not talking about revolutionary change for the year 2000, but when we go out with our statements about the year 2000, there will also be some additional warnings and issues and heads-up that life will not remain o the status quo forever.

One of the main issues that came up in the hearings is that we will be mandating or proposing the use of the official implementation guides. In general, most of these systems that I mentioned have official implementation guides, but there is uneven implementation of them, that is, they have to be followed perhaps for certain sets of transactions that come to HCFA, but then people may or may not rely on the same implementation guides in other environments, and people say that this is a problem, so there will -- of course, since we are dealing with national standards, there is going to be an implementation guide that will cover everything.

I think we heard in very strong terms that whether a particular recipient is willing to pay for a particular item at a particular level of specificity, if the code that is sent to them is at least valid and part of the code set, the claim should not be rejected totally, that is, not be able to be processed if it has official codes in it. So there was a lot of discussion about that.

I think we feel very strongly that there is major likelihood of changes in some of these standards in the year 2001 and beyond. One major example here is that it is highly likely that it will time to move to ICD-10CM in the year 2001, and there is also a strong move to have something done on the procedures side. We also have ICD-10 procedure coding system, a new system that HCFA has sponsored the development of, which is in testing.

Let me just say to you that as part of the process of determining what happens after the year 2000, we are going to be very interested in input and testing and test data and reaction, both to the version of ICD-10CM, which will be released for testing in November, and you will be able to find information about this on the NCHS Web page, and ICD-10 PCS is already available for testing, and the information about that is obtainable from the HCFA Web page.

Those of you who are interested in looking at the underlying relational files for the PCS system, which the National Library of Medicine is also interested in, those are not available yet, but they will be available shortly, and when they are available, that information will also be announced via the HCFA Web page.

If you anticipate, as we all do, that there will be changes in the year 2001 and beyond, then everyone who processes these administrative transactions had better start their plans now, because we can guarantee that at some point in the year 2001 or shortly thereafter, you will get codes for these things that are longer than the five digits that are currently being used.

So in terms of other implementation issues, we heard a lot about concern about the openness of the update process for privately owned and maintained systems, and of course, there are a few maintained here in what we are recommending. There were concerns about what would be cost to use restrictions for these systems. There have also been concerns, not so much expressed at the hearings, but we have heard them from the informatics community via other routes, that they are concerned that all of the standards be available in electronic formats that are suitable for the full range of users.

Let me just translate this a bit. Some of the electronic formats that are available for some of these systems may be useful for the individual who is looking up codes and selecting them. They may not be quite as friendly for the people who are building them into large systems, because they are not fully specified database versions of the things that are easier for people that are dealing with large systems.

So having made what we think is the correct initial selection based on all the input, we are now going to be looking at any of the issues that I have mentioned here: what exactly is the situation now with the systems that are have recommenced, and what if any changes have to be negotiated or discussed with the privately maintained systems between now and the year 2000 when these become national standards.

I think that is what I have to say. So we are all available here for questions.

DR. OWENS: I have two questions. The first is relative to supplies. You had it in your earlier slide, but you had no standard listed. Do you have a standard in mind for the supplies?

DR. HUMPHREYS: The alpha numeric HCPCS -- and I'm sorry, we should have expanded this to say devices and supplies. Those of you who are familiar with the system know that this is at a certain level of granularity. I think that in the future, we may want a more granular system, and there is discussion about that.

But again, these initial -- for example, there has been discussion about moving forward to various types of product codes, and in the case of the device area to more granular device systems such as the universal medical device nomenclature and so forth. But again, what we were strongly hearing from the people was that they would not be ready for a move to these systems in the year 2000.

DR. OWENS: Then the second question is relative to the drugs. You had mentioned you recommended NDC. There are three different formulas within the NDC as far as a numbering format. Then there is a fourth that is used by a lot of systems. Are you recommending the three different levels, or the fourth, that is actually an 11-digit NDC that is used by systems? Or do you have a preference?

DR. HUMPHREYS: I think that we have heard that we want to use NDC codes for pharmacy transactions, but we really have to get down to the specifics of that. So any specific input that you have on that issue, we would be delighted to have it.

DR. OWENS: From a systems standpoint, both the major databases and pharmacy actually add a leading zero in either the fourth, the first, the middle or the end section of the number, making it a constant 11-position field.

DR. HUMPHREYS: I think that -- we have heard your comment. We will look at that.

DR. ROBINSON: I have a question regarding the procedure codes. You said get ready for it, it is going to be greater than five digits on the procedure code.

DR. HUMPHREYS: I just said in general on the codes. I would expect it to happen in the procedure coding system as we move ahead. But for example, the version of -- I believe I'm correct that the version of ICD-10CM that is going out for testing -- granted, NCHS may receive comment on this issue -- also have six-digit codes.

I think that the major message here -- and if we were to move to NDC codes across the board for drugs, which many people feel is the way to go in the future, as the gentleman just said, whichever version we use here, we are using something that is much longer than a five-digit code. So I would anticipate that this would probably be true for procedures, and certainly the PCS system that is available for evaluation has a seven-digit code. But it is a general comment. I think people should be thinking about the fact that we are going beyond the point in electronic transactions where they can count on from year to year that we will never change the length of a code.

DR. ROBINSON: Then if it is just a general comment, you probably won't be able to answer my question, which was going to be, have you taken into consideration if you are going to be crosswalking the current five-digit procedure codes to a new number?

Also, the other part of that question is, what is the maximum length that you are thinking?

DR. HUMPHREYS: Let me answer the first question, because we did heard a lot of testimony on that issue. Most everybody believes that to the extent that it is possible to have a reasonable crosswalk between what becomes a new administrative or statistical code for procedures and what was an old one, that such a crosswalk will definitely have to be created.

In some cases, if you change the semantics and the basis of the system to a certain extent, you can't get a perfect one. But something will have to be done, because we will continue to be concerned obviously in terms of health services research and quality and whatever, to be able to compare from year to year. So something like that will have to be done.

Then the issue is, will there be a clean change, so that anyone could guarantee that on January 1, they would receive no data under the old system. Highly unlikely.

In terms of the maximum length of a code, I won't touch that one now.

DR. ZUBELDIA: To what extend are you going to regulate implementation issues? Let me give you some examples. For instance, some payers want providers to use surgery codes for anesthesiology, because it gives them better detail on what the service was. Some payers will not take punctuation in the ICD-9. They only hold five positions, and you have to drop the period. Most payers will not accept E codes in the ICD-9. Some payers will accept three-digit codes when there is a five-digit code available, but they will still accept a three-digit code.

All of these are implementation issues. Are we going to have regulation on this, some sort of standardization on the implementation itself, or just the codes?

Then, after you answer that, I have another one.

DR. HUMPHREYS: I think that what we have heard from the people testifying to the NCVHS is, the desire is to have a standard implementation guide as we have for the other standards, and have that apply across the board, which is what I meant when I said before, that just saying, hey, we are going with the status quo, maybe it won't be exactly the status quo if you had chosen certain options to deal with this in your particular environment.

So I think that there is going to be a lot of discussion on that issue. But I believe the intent here across the board is to have an implementation guide and say, everyone is implementing these standards in these ways. If they are not, then we don't have a standard, do we?

DR. ZUBELDIA: Then will Medicaid programs be allowed to just pay the fine and continue creating their own codes? They are paying with our tax dollars.

DR. HUMPHREYS: Would somebody from HCFA like to respond to that? I guess my feeling is, the law has certain fines, and I don't know whether allowing people to pay the fine is exactly the way to say it.

DR. MOORE: Are we talking about Medicare or Medicaid? Medicare will pay the fine probably, if we don't do it.

On the Medicaid side, there is a burning interest back at the office about, are we going to make funding available to the states to do this. That has been the issue from the states.

One of the things that was mentioned earlier about this, some people have said, this is going to take a great deal of resources, I'm going to have to completely revamp my system in order to implement these transactions. I don't believe that. We already process these transactions in many ways. It is going to require more standardization of the transactions as they come into the system. We are going to have to -- and I think HCFA is guilty of this -- do less customized programming, if you will, where we have more opportunity for failure, and work with translators, work with the tools that are being developed in the industry.

These standards, X12 and others, are being used for manufacturing, for ordering supplies, for doing a whole lot of business that is already moving in an electronic way. One of the things that we are going to have to do here, and this incudes the state agencies and Medicaid, and I know I don't have the authority to speak for governors, but they are going to have to start complying and seeing what they are going to have to do as well.

DR. ZUBELDIA: Bob, the reason for the question is that some Medicaid programs say that to comply with state specific requirements, some legislation in their state, they have to issue new codes that are not HCPCS codes. They are required in their state only.

DR. MOORE: This is the issue that I think was mentioned by Betsy, where some states are saying that the level of specificity that is contained in the HCPCS is not sufficient for their needs. They want a greater level of specificity because of pricing and other things of that nature.

Choosing the HCPCS does not address that issue. Where HCPCS has a -- for a catheter it is a general code, and it can have multiple types of catheters, and there are different prices, and the state wants to narrow that pricing mechanism, so they are pricing more equitably. That is going to be an open issue that I don't think we can address.

The only way that we can do that is to open HCPCS up for more codes. And if that is the case, then we may -- if we select this as a standard, that may be the burden that HCFA has to pick up and add to it that says, we will do a better job more timely of doing this to meet the needs of the other payers in the states, in order to make this a standard code. Otherwise, I don't see how we can make that call.

DR. BROOKS: I'll just mention one point. They are attempting over the past couple of years to increase user friendliness with the alpha numeric HCPCS process, and bring more codes in and do that quicker. They are not perfect, either. But I think that is a long-range goal that we have to work on, is a better coding system, not before the year 2000, to get more specificity for everybody.

DR. EISENBERG: Betsy, Barry Eisenberg from the AMA. I have two questions also, if I could. Yo really emphasized the idea of the status quo versus what might take place after the year 2000, 2001 and so on. Stating that I think is going to create a lot of interest and some anxiety amongst a number of people.

Is it your intention at this point or in the NPRM to lay out your thinking about the timetable and process that you are going to use to address those longer-term issues?

DR. HUMPHREYS: We have a couple of issues here, in terms of what happens in 2001 and beyond. We have the two major codes and classification system efforts within the department that have not been fully reviewed and tested by the field, which are ICD-10CM and the PCS system.

The current timetable calls for us to have had a fair amount of testing and response from the field on both of those systems by about March of next year. I think I've got that about right. So that is going to give us some additional data on which to work.

There is a strong interest as expressed -- perhaps that was mentioned this morning; I came in in the middle of Don Detmer's presentation. The NCHS has made their recommendations to the department, and they have strongly suggested that the department should set in motion a procedure, a process for coming up with quote, a more unified procedure system for implementation. I believe the dates they indicated in their recommendation was like in the year 2002 or 2003. That of course is the NCVHS recommendation. The department has to take their recommendations pretty seriously, in terms of looking at those issues.

So we feel that once we get out the NPRM that relates to the year 2000, that is one thing which is a big thing to do, but we have these other issues to address, in terms of what is going on.

One of the things that I should mention is that there will definitely be additional hearings as we had said in April on the longer term issue of what we see as being required as a full vocabulary that is needed for the full patient record, which also relates to this whole issue of an integrated procedure coding system and everything else. Those opportunities, meetings related to that, are being in essence scheduled now, but are likely to come up in the fall, in the November time frame. They will be heavily advertised, so people will have an opportunity to provide input on those issues as well.

I don't know if I have answered your question. We have to get this done and out. Having identified that we can't make some of these changes by the year 2000, then we have to immediately get on how do we set in motion the procedure to deal with them. But we will definitely be using as one set of information to help us grapple with these issues the information that comes to us from the field on the testing of ICD-10CM and ICD-10PCS.

DR. EISENBERG: Thank you. My second question is related, but goes also to timing. That has to do with the slide just previous to this, where you enumerated a number of concerns. I think you really did a very fair job of enumerating the concerns which have been brought out in the NCVHS hearings.

I take it from the comments that you made that these are also a group of issues which are going to be addressed in terms of the 2001 and beyond issues.

DR. HUMPHREYS: I think they certainly are going to be involved in 2001 and beyond. But in this particular case, we also have to look at -- we have to do an assessment of what is the reality of the situation for the year 2000, because there are systems in use now, and they are distributed under reasonable -- what are considered reasonable conditions related to HCFA's requirements and so forth. The issue is, are those also reasonable if we now are imposing this as a national standard on every type of situation for every type of transaction.

So we are going to be looking at these things. I don't know the answer to that question, you understand, but it is something that has come up as an issue even for the designation of quote the status quo as official standards for the year 2000.

DR. EISENBERG: Thank you very much.

PARTICIPANT: Hi. Do you foresee HCFA reviewing new codes more frequently during the year? For example, at our company, we are having problems rejecting particular the HCPCS codes without a description on electronic claims. I wasn't sure with new drugs and devices coming on the market after the review process to accept new codes, it would be more frequent during the year.

DR. BROOKS: That is one of the things our committee is supposed to work on for implementation. For the short term, no, there are no planned changes, just because that is what industry expects in the systems. It works better to change them once a year. But I think our group will be discussing, is this the best mechanism, or do we need to do it more frequently or less frequently.

DR. HUMPHREYS: These issues were expressed to us as they are enumerated here. People were more concerned about these issues as they related to the privately maintained systems. But I'll tell you that internally, we have also looked at those that are currently maintained by the federal government; are we also in a position to distribute these and to -- we need to look at the procedures across the board, not just for the privately maintained systems. But we have to make sure that we have got something reasonable in place for all of the systems.

Any other questions? If not, thank you.