[This Transcript is Unedited]

Department of Health and Human Services

National Center for Vital and Health Statistics

Subcommittee on Standards

June 11, 2014

Hubert H. Humphrey Building
200 Independence Avenue, SW, Room 705-A
Washington, D.C. 20024

P R O C E E D I N G S (4:42 p.m.)

DR. SUAREZ: Okay, we are going to convene the Standards Subcommittee meeting. We don’t need to introduce ourselves since we all know each other. So the agenda as you have it in front of you says welcome and introductions. So welcome, and we all have been introduced. HIPAA report to congress, finalize any edits. We finalized edits and got approval.

So the next steps, so that you know, we’re going to work with Ob and Terri to finalize the report. I think we’re all going to do a final read also, step back and read it now fresh and probably correct a few other inconsistencies and then finalize it and send it to Debbie for the final publication.

I did notice actually that when we converted, and Marietta I don’t know if you did this, but when we converted the Word document into the PDF and posted it on the web, if you go to page 32 half of the page is missing, and then several pages there is a similar thing, there are sections that are gone. That means basically the PDF generator basically didn’t do a good job of generating the PDF. So once we generate a PDF we’ll have to go through, of course we’ll delete the draft in the background once the report is final, and then generate the one. And then there will be the official formal submission to the appropriate committees and all that process.

MS. JACKSON: It’s a little different process than Letter to the Secretary, because it’s going to the Congress and all of that, so we’ve done that before and we’ll work it out again.

MR. SOONTHORNSIMA: What is the timeline for that? Somebody was asking earlier.

MS. JACKSON: I think we were looking at, depending on when we get it from you and Mary, I know I am going to look it over word by word, by the weekend, and by next week we should have a clean draft, and from there that’s where we’ll start processing it for the congressional —

MR. SOONTHORNSIMA: We talked about not posting it at all until the Secretary —

MS. JACKSON: The Secretary gets the letters, the whole thing is for the recipient to get it.

PARTICIPANT: This is to the Congress, not to the Secretary.

MR. SOONTHORNSIMA: We wouldn’t be posting it on line —

MS. JACKSON: Until the recipient receives it. Just out of protocol, just for that. And then we can talk later about dissemination afterward. We have a nice 10th Report, which I wondered how that could be outdone, and here we get this gorgeous 11th Report. And we’re not sure about printing. The capability that we are in now, we’re in such a digital age and stage, but I will follow up with you as to the best way of disseminating.

As you know, that 10th Report got a lot of traction. There were various meetings, people really enjoyed getting that, it got swooped up. So we’ll make good use out of it.

DR. SUAREZ: So the next topic, which is not actually listed here, before we go to the subcommittee work plan, is the public health letter. So we also have approval of the letter with the caveat of drafting a little introductory statement, this being the first letter to this new secretary, makes sense.

Someone actually, Donna suggested something I thought would also be good, although not necessarily from the committee but from the subcommittee, but these letters are the committee letters, which is actually writing a separate, short, one page letter that welcomes the new secretary and introduces the National Committee to the Secretary. And that serves as sort of a cover to our public health letter. That’s one other way to handle that part. Of course, we would need to draft a letter tonight and present it tomorrow to the Full Committee for approval, something like that.

PARTICIPANT: Or submit a sentence or two.

DR. SUAREZ: Exactly. The two options are we do just one letter with the introduction, welcome, and sort of giving a more detailed background, or we write two letters, one is introduction, welcome —

MS. GOSS: Or maybe there is another option. Maybe what we can do if we want to say hi, this is who we are, to the new secretary, couldn’t that be done between staff and Larry, and then that’s separate from us?

DR. SUAREZ: I suggested it actually when Jim was presenting this morning his update, I mentioned the ability to perhaps meet with the Secretary. And of course they have a whole host of individuals – the whole department is going to begin to introduce themselves and all the people that they work with to the Secretary in different stages. We are one of the hundreds of groups that need to be made aware of to the Secretary.

So my preference would be just to have one single letter as we have it, just an introduction, and extending a little bit more the description of the National Committee, and then that be it. A single letter, the public health letter will have that introduction and that welcome, rather than having two letters. Because the problem is if we have a second letter we have to write it tonight and present it tomorrow.

MS. DEUTSCH: We crafted just what the introduction would be on the preface of having one letter as opposed to having an introduction letter, and it would just start out as; It is a privilege to welcome you as the new Secretary to the Department. We are submitting to you your first letter and recommendations from the National Committee on Vital Health Statistics, NCVHS, and then go right into —

DR. SUAREZ: Excellent. So we’re done. So just to confirm, we’re going to have one single letter, which will be the public health letter with that introduction.

MS. JACKSON: The caveat, we need to get this into the Secretary’s hands. Things like this, as far as background information about the committee – we’ll do communication with her.

DR. SUAREZ: Since this is for a letter we can reference that actually.

MS. JACKSON: We will find a way. This is I see, just another just so she’ll know the extent, 50 years of work, and then the letter. So don’t feel that the letter has to be the introduction, we need to let her know we’ve been around a bit and doing a lot.

DR. SUAREZ: So that would be separate process.

MR. SOONTHORNSIMA: That letter is complete.

DR. SUAREZ: Subcommittee work plan. Let me show you the work plan that we have been using through different quarters just to remind you of the work that we’ve been doing. This is the work plan that we actually presented at our February meeting, and we talked about it, and I just put a cover, basically I said June, but this is the same slides, I didn’t change anything. So we had in quarter one what we said we were going to do here, NCPDP prior authorization, which we did, Operating Rules for Remaining Transactions and Update. Update on WEDI.

This is all the hearing really, that we had in February. You can see here NCBHA and the ACA review committee, so we started that conversation back in those months, and that’s what is culminating here with what Jim was presenting earlier.

February hearing on operating rules. We held that, we collaborated with Population Health on a possible second hearing, we put a question mark, so we haven’t really finalized or decided to go with a second hearing yet. And then we prepared the letter on public health standards and completion of the HIPAA report. So I think we completed quarter one.

Quarter two, we’re going to continue the e-health vision and roadmap discussion – that we haven’t done. Operating Rules for Remaining Transactions discussion, I think we heard an update and also this hearing yesterday.

Claim Attachment Regulations, very hopefully that they’re going to be possibly out. We heard actually that they’re not going to be out in the near future. I think the whole industry is still determining whether we do have a defined and final set of standards to adopt officially through regulations. So that is now being pushed out.

We do have a deadline with respect to attachments that you should all be aware of, which is January 1, 2016. That’s when the Affordable Care Act Provisions said that the adoption of attachments and implementation of attachments should happen. So we have 18 months, CMS regulators, to draft and prepare any final rules and then have a deadline for compliance with the final rules with respect to claim attachments.

We were going to actually work on a work plan for the review committee just to propose, but I think we didn’t have really much time to do it, and now we are going to be awaiting the official designation by the Secretary to be the review committee so that then we can define a work plan. And then some of the convergence work we highlighted in this work plan.

And of course we had the June hearing yesterday, and we mentioned here completion of the letter for Public Health Data Standards, which we did, completion of the HIPAA Report, and then begin to draft the letter of Recommendations on Administrative Hearing, so yesterday’s hearing, so we’ll need to begin to do that as one of the next steps over the next few months. And then the work group on the review committee of course is pending designation from the Secretary.

Quarter three and quarter four, I don’t think we have it yet to define. Well, this is what we had back in February. So in quarters three-four, we have here the e-health vision and roadmap. The goal here was to try to finalize the discussions. ICD-10 implementation, look at that. How about that? Initial implementation of the review committee work plan.

And we even put HPID enumeration, because that originally was the plan. Now of course the HPID enumeration is scheduled to be happening October 2015 with the intent of implementing it October of 2016. So we’re still a year away, a year and a few months away. So these were the topics we were going to cover. We didn’t really lay out any hearings.

We mentioned a possible follow-up hearing on Public Health Data Standards on certain topics, and that’s something we can talk about again with Population Health and discuss at what point it makes sense to bring back the entire group that we brought in November and have another hearing. I know a lot of those organizations, first of all they were very excited about the letter that we prepared, and they are very excited to know and to be able to engage with the committee.

And so here we have another big important audience. Right now of course, the big audience we have is our HIPAA implementation audience, all the people that have come to our hearings, and yesterday we saw them incredibly showing up, and that is a major audience that we have. Now we opened the door for this new audience, this new group of people. These are all the public health associations – NACCHO, ASTO, NAPHSIS, CSTE, all these groups that have been involved in helping develop and implement Public Health Standards. So they are very much a captive audience for us and we can bring them back to talk more about the standards in specific topics.

We had thought about doing a possible third roundtable on e-health vision and standards roadmap, and that’s still a possibility for Q3 and Q4. So now since we are sort of at a point where we have completed our HIPAA Report, we don’t have to think about another HIPAA report for a few more months or maybe hopefully a year, and then we have to sort of finish up the first round of all these letters. We only have the letters coming from the hearing yesterday. We can consider the second part of this year, so basically Q3 and Q4, to focus on stepping back and looking at the roadmap that we have really been asked to look into, as well as again this Public Health Standard. So those are two possible topics to consider for Q3 and Q4.

So basically that’s it, that’s the work plan, just to finish up basically the expectation between now and September. So whatever we need to do between now and September. As a subcommittee I think we want to continue our meetings as we’ve been doing on an almost monthly basis, to be able to draft and prepare the draft letters that we will be putting together from yesterday’s hearing. And I’m talking about letters, maybe not single letter, it could be a single letter, it could be more than one. And then reviewing them and discussing them so that we can bring them back to the Full Committee in September, that’s one of the big priorities.

But in addition to or separate from that I think we will need to talk about the review committee, pending the designation from the Secretary. We’ll need to talk about the roadmap and how are we going to recapture and reconvene around that. I think we need to talk about any additional work on the Public Health Standards side, in partnership with the Population Health Workgroup.

There could also be some topics to talk about in partnership with the Privacy and Security Workgroup, as we talked about earlier today. So those are maybe four major topics to consider discussion over the next four or five months.

MR. SOONTHORNSIMA: I’d like to entertain an idea that we talked about earlier when Jim teed up the issue around a review committee. I wonder if we couldn’t, once we get the approval and designation, if we couldn’t bring together that roadmap and vision work together with the review committee work. In other words, use a single sort of set of deliverable to drive not only – So when we do the report, it’s not only to look at up and coming offering rules and standards, but also to achieve the strategic vision. Just a thought. That way we truly can converge the two.

MS. KLOSS: Getting a handle up front as to what the metrics and milestones are to be looked at as part of that review would be really helpful. The review isn’t just a list of what got done, but substantively what’s the impact.

MR. SOONTHORNSIMA: We already had the principles. We spent a lot of time crafting the principles.

MS. KLOSS: We don’t have the metrics yet.

MR. SOONTHORNSIMA: Not yet, but the principles should help guide us in our review work.

DR. SUAREZ: Those two are really tied, the review committee and the role, perfect tie-in. So we’ll have to wait and see what happens with the designation really. But in the meantime we can begin to flesh out again during one of our calls some more next steps regarding the roadmap that can again be brought up as part of the review committee work plan. Okay, any other topics?

MR. SOONTHORNSIMA: So review committee and roadmap are the same?


DR. CHANDERRAJ: Data silo are big, and trying to merge everything and come to a convergence. Some of the people are using SNOMED and LOINC, and we’re still going with ICD 10s. We’re creating these huge silos by going with ICD 10s. IS there any dialogue or a talk to merge into something that can be unified so that the data transmission can be seamless?

DR. SUAREZ: Well as far as I know, and the way I see it, we do the three. In fact, we do more vocabularies and terminologies. We code on SNOMED and LOINC, and for public health we use actually CDC coding, and then we also used ICD-10, mostly for billing and administrative purposes, but also for other public health reporting. There are already resources out there that do mapping of the coding systems.

In fact, Kaiser Permanente developed something called the Convergence Medical Terminology, which was donated to the National Library of Medicine and is now on public domain, which one way of describing it is a Rosetta Stone of vocabularies and terminologies. It really serves as a connection between SNOMED, LOINC, ICD, CPT, and other codes. And so if you have an ICD 9 code you can see what the 10 code is, what the SNOWMED terminology or terms are, what any LOINC code is, and what CPT code might be related to it too. And you can extend it to HCPCS and other types of code. So there is already that map in place and available in public in domain.

DR. CHANDERRAJ: The process is one step more to go to the map and do it. It’s much more elaborate and expensive to go to a map and map the ICD10 to a SNOMED code or LOINC code and do that. If you can unify all those three things and eliminate the confusion by going to one code uniform to everybody, that will be easier.

DR. WALKER: I don’t believe ICD 10 is one of the 12 standard terminologies that the standards committee recommended to HHS in November of 2011, SNOMED is.

DR. SUAREZ: Actually ICD10 is.

DR. WALKER: In two, but not three.

DR. SUAREZ: Three hasn’t yet been pushed.

DR. WALKER: The plan was that it would not be in three, it may have changed. To Raj’s point, ICD10, one of the problems with it is that SNOMED is a much better designed terminology.

DR. SUAREZ: For a very specific purpose. You cannot bill using SNOMED. You can bill using ICD and CPT, but you cannot bill – SNOMED doesn’t lend to billing, quoting.

MS. KLOSS: Nor to international statistical measures., I think this discussion is a really good one. Maybe not this year, but I would think that next year we need to balance the focus on ACA with some time on vocabularies and classifications as a standard subcommittee. Certainly there are updates and progress reports on ICD10 maintenance, ICD11 developments, what’s going on with SNOMED.

DR. WALKER: In fact, ICD-10 is not one of the 12 standards that the Standards committee formally recommended in the transmittal letter that Farzad had sent on September 9 2011, and SNOMED is, to Raj’s point.

MS. GOSS: Would that be because they’re focused on what’s used in a clinical document set as opposed to an administrative financial set?

DR. WALKER: It’s what I said, it’s because it’s better designed, it’s better behaved, it is because for all kinds of reasons it is one of the things that’s needed to characterize patient care, and the billing issue is a definitional issue.

DR. SUAREZ: I don’t know if we are saying different things necessarily. Meaningful use does require EHRs to be capable of supporting ICD10.

DR. WALKER: ICD10 is acceptable, but not required.

DR. SUAREZ: There’s no argument that for clinical purposes SNOMED is the most appropriate terminology, there’s no argument about that. But there’s no argument either that when you are converting a clinical encounter into an administrative transaction like a claim, you have to change from SNOMED to something, and that something right now is ICD9, and it’s expected to be ICD10

DR. WALKER: But there is substantial controversy whether 10 makes more sense than 11. Raj has a fair point. The committee has stated that it thinks going to 10 is the right thing to do, and that’s also a fact, but Raj’s point is fair that that is not an uncontroversial position. We wrote an editorial, Halamka and Overidge, and whoever it was, Ruschel and whoever and I wrote an editorial several years ago, that proposed we just wait and go to 11. So it’s not nuts, it’s just not the committee’s position.

DR. SUAREZ: But the argument of waiting and going to 11 still creates two different codes. So there’s no difference in Raj’s argument.

DR. WALKER: The other thing to Raj’s point, if you want to get technical, is you need about 5000 ICD-something codes if you decide to use ICD for billing, and you don’t need the rest. In a rational system you could say we’re going to code everything in SNOMED, and then whenever we need to draw up a bill we’ll know how to translate the SNOMED into one of the 5000 or so you would need to actually draw up bills.

DR. SUAREZ: That is exactly what is done.

DR. CHANDERRAJ: The other point is SNOMED conversion to administrative reasons for (?. Why don’t we put the onus on the administrative aspects to change their code for billing purposes rather than asking the physicians to change now to ICD-10 and move on to something else in the future, it’s so cost prohibitive for physicians.

DR. SUAREZ: Move to what else?

DR. CHANDERRAJ: ICD-11 is going to come.

DR. SUAREZ: By the time we move to ICD-11 probably all of us will be retired and who knows where else. To be fair, I hope you’re not saying, because this is a critical part, I hope Raj you are not saying that we should only use SNOMED for clinical and billing purposes, just to be clear.

DR. CHANDERRAJ: For uniform systems so that the physicians don’t have to bear the cost of converting from one system to another to another.

DR. SUAREZ: In order to generate a bill and to process a bill and a claim, you need a coding system like ICD. You cannot do bill processing with SNOMED. That’s just simple, you cannot. You have to create so much intelligence – SNOMED is a system of terms, it’s a system of terms that includes some codes and includes some other basic clinical terms, but it’s not a billing system. The reality is we need a coded system for billing purposes, and certainly for clinical coding there is already a defined coding system called SNOMED that is required. So I don’t think there is any issue with that.

I think what you are saying is ideally behind the scenes the physician doesn’t have to do, okay for clinical I have to go with SNOMED, here’s the code, and now me, the physician, I’m generating the bill, so now I have to change and do the ICD. The system behind the scenes is intelligent enough to generate by mapping SNOMED into ICD to generate the bill and send it out, the physician doesn’t even need to do anything, they just code in SNOMED. So the silo concept is not in my view, a silo concept of SNOMED, you can do the mapping and there are already tools to do the map.

DR. STEAD: We had a very interesting conversation at lunch. Sometime we really do need to put some substantive discussion into this, because I at least believe that we have proceeded down a very slippery slope of incrementally adding things to things we transmit with bills, and expanding billing codes, because we have them and we can do that and we run the transactions through everything, and it has been the simplest thing to do.

I’m in the group that although I support what this committee has said believe that ICD10 went over the hump of what was actually useful to try to actually include in the billing system. The reason for that is that the granularity of information we need to characterize patients from either a treatment point of view or a risk point of view is much greater than can currently be documented I believe in any of those systems.

And my example is to go to cancers. We’ve classified cancers as lung cancers and different types of lung cancer and colon cancer, and pancreas cancer. We now know that the protein profiles and genotypes of the tumors, not the patients, are in fact the distinguishing characteristics both in terms of response and in terms of treatment. So I suspect that the label, I think we will discover that lung cancer is a useless label, and categorizations that are built on it are useless.

And so as you think about a coding structure that changes very rapidly as we’ve experienced, and our understanding of biology that’s changing a lot more rapidly, we’re going to have to get it where we have a way to access, even for purposes, if we are trying to risk adjust payment for example, we are going to have to be able to access that rich clinical information, and I think it’s going to somehow be accessed in some means that is similar to some of the things that we’re talking about in the interoperability stuff, or even back to PCAST 2010. Because it’s going to be too rich to be able to send through a transaction processing system.

The example I used at lunch was it’s like people who send 26 MB digital pictures through their email instead of putting it on a site and sharing them. I think we’re getting to that breakpoint, and one of the things we could do in our vision leading is actually show people that we need to put different techniques side by side in a way that let us really cut back on the complexity a little bit. We should be able to cut way back on the complexity of what we have to send through billing transactions.

DR. SUAREZ: You’re saying we need to cut back to simplify – But in reality you actually say in order to understand the complexity of a condition in a patient you need even more information than what systems or coding systems like ICD can transmit.

DR. STEAD: I don’t think they are going to be transmitted in any coding system. I don’t think they’re going to be represented in a coding system, I think they’re going to be a complex digital signature of a set of attributes, and maybe those attributes can be represented in coding systems, but if so they’re going to be a lot more like gene oncology than SNOMED or ICD.

MR. SOONTHORNSIMA: So the idea is, for today’s purposes of getting the claims paid, ICD10. Going forward ICD10 will be it, that’s a position. However, in the future, the point is if we have to do risk adjustment it would be more and more complicated. ICD10, the billing will continue to exist, but to augment that information may not need to come through the ICD code. For example, you can grab conditions perhaps more quickly through third party as opposed to having those conditions be submitted. That’s too simplistic, but that’s the idea, instead of having to code everything through ICD10. Is that right?

DR. STEAD: I like international ICD10, I think American ICD10 oversteps what needs to be done. I understand it’s what we’re going to do, no problem with that, but it is a level of complexity that we certainly don’t want to add to. And I think we can subtract from it.

DR. CHANDERRAJ: I think the language in ICD10s and ICD9s is so complex for ordinary physicians to comprehend. They don’t know what to code, and they’re limited to the billers to code it, and the billers select the code that pays the highest. It’s totally irrelevant to what the actual patient is having. That’s the problem with ICD10. It doesn’t correspond with the language that the physicians are used to, that’s what’s happening.

MS. KLOSS: I think as we look ahead we need to be driving toward a day when all of this is behind the scenes. I don’t think the physicians should have to pick any of this, that the mappings should be rule driven, and certainly by the time we get to ICD11 – But we don’t have the mechanisms for changing versions and managing vocabularies, those just aren’t capabilities that we have the toolset for yet. But we should be using other classifications for oncology. There are so many other ways of taking the same data and recasting it depending on the analytic purpose.

DR. SUAREZ: I think physicians are not coding ICD10s to send bills, they use coders and billers, coders basically internally to handle the billing, and they code and encounter using more of their natural medical language like SNOMED and other things, and then behind the scenes either the system is intelligent to translate that or will be intelligent to translate that into a coding mechanism that is appropriate for billing, or there are coders specialized in reading the notes and converting it into code.

I think it’s a very important discussion what I think we’re capturing here in terms of something that we need to have a more in depth talk about. Part of the reimbursement of the future, because this ties a lot about reimbursement and risk adjustment and other things, where do we need to see change in the future. Because yes, on the one end SNOMED is way too granular and it will be impossible to handle for billing purposes, on the other hand ICd10 might still be too granular, and in some cases might not be granular enough to determine what is an appropriate risk adjusted —

DR. WALKER: I think part of what Bill is saying is right now it’s diabetes type II, and maybe there’s comorbidity. Well, we know that type II diabetes is really what 40 or some fairly large number of entities, all of which have a fairly similar final common pathway.

Well, when we get that worked out and we start saying you should use Metformin for these 12 versions of what we used to call diabetes type II, and you should start with a sulfonylurea for these 24, then where are we going to be with ICD-whatever? We’re going to have to pull that whole thing out and put a new thing in. I think that’s the problem, I don’t think we have quite conceived how dynamic our understanding of diagnostic entities is going to be in the next decade or two. ICD-anything, maybe 11, but not 9 or 10, is just not at all designed to be able to manage that kind of dynamism.

DR. SUAREZ: My sense is in the future it doesn’t matter what is done behind the scenes, you can plug in ICd-9, 10, 11, 12, whatever in the front end, the provider is going to be using SNOMED.

DR. WALKER: It will all be hidden from the provider, you’re right. Right now it already is in good systems. That is not the point the point is that I think it’s like sending a 22 MB picture through your email. You can do it, it may wreck your recipient’s reception, but it’s a dumb way to do it and at some point you’re going to run into enough volume or enough something else that it’s just going to break on you.

DR. SUAREZ: So Larry, you have the last question before we go to our next talk.

DR. GREEN: Maybe we should go to the speaker.

DR. SUAREZ: We had actually in the agenda some post-hearing discussion and next steps. We’re not going to do that right now, we have a lot of things to assemble and analyze based on the hearing yesterday. We will be doing that and then sending out to the subcommittee for our first call in the next few weeks. We haven’t scheduled those, we’ll schedule those, but we’ll send you a high level summary of the themes that we’ve heard.

I want to of course introduce our invited guest to the subcommittee, Margaret Weiker. She’ll introduce herself, but I just wanted to mention that this topic is something that we want to talk about, because it relates to the possible adoption of a set of standards. So I’ll leave it to you, Margaret.

MS. WEIKER: I am Margaret Weiker, I am chair of the ASC X12 Insurance Subcommittee, and the Insurance Subcommittee at X12 is responsible for the X12 standards that have been adopted under HIPAA. So this kind of melds the DSMO request and the DSMO report that I gave yesterday with a letter that we sent to Walter via Terry on April 30. And this has to do with the DSMO Request 1186, as well as with the letter. And that is regard to XML schema.

Several years ago we approached the committee, some of the members, about adoption of XML schema. And at that time it was a fairly new thing that we had developed at X12. And we thought it was a good idea, but many more higher priorities, so you know what happens, it kind of gets pushed to the bottom. Well, since that time what we’ve seen is a lot of companies developing proprietary XML. A lot of this is done for their internal systems, but then some also use it to exchange the data as well.

But anytime you have proprietary things, sometimes they’re not coded correctly. So you may have proprietary XML that you think meets those HIPAA TR3 guides, but in actuality you may have coded something wrong. So there’s no clean way to do that when you start developing your own proprietary XML. As time has gone on we have found more and more people doing that.

At the same time, what X12 did was develop a guideline that allows us to create XML based upon the TR3s. So we have a database that actually houses all the rules and structures in regard to those TR3s. And what we’ve done is using that same database we’ve created XML. So sometimes we may refer to those XMLs as TR3 schema, you may even have heard it called way back when isomorphic XML. But typically we call them either in-guide schema, or TR3 schema. They are WC3 compliant in their structure and their layout.

We have had many companies purchase the XML for use. Some of those asked , well if I use this, is it HIPAA compliant? And some have asked before they bought, some have asked after they bought in regard to that. But we said well they have not been adopted under HIPAA, so technically you cannot say you’re HIPAA compliant if you use them, because they have not been adopted. However, they’re generated from the same database that the TR3 is generated from.

So as time has gone on more and more people asked, we have shared the schema with OESS. And basically they’re like well there’s not a lot of demand for this, so we’re just going to kind of put it over here too. Well, several months later the DSMO request was submitted. X12 did not prompt anyone to go out and submit this DSMO request, this was done on their own, where they said basically we need some standardized schemas for the X12 transactions. So it went through the DSMO process with the other organizations opting in, discussing it in their group. We are not proposing to throw the baby out with the bathwater in regard to the EDI standards.

We see two possible paths to move forward with the XML schema, and these are outlined in the letter. One of these is NCVHS to recommend that OESS published an official statement around this, basically be an FAQ type of thing that they had done in the past, where you get a question about the schema and the answer in regard to the schema. They could also put it in a notice. They’re going to have to create a notice for the HPID. There’s also the NPRN that’s being created by them for the NCPDP changes, although I don’t believe this requires an NPRM, nor does the HPID, but it may be an avenue to announce it so to speak.

So one of the options we believe is through an FAQ type of thing, where they have done that in the past around what’s valid HIPAA, what’s not, is this compliant, is that not. And then, along the line of the NPRM, but we recommended it be an interim final rule, because that tends to be quicker than an NPRM in the life cycle. That basically says these are equivalent of those. So trading partners, we’re not saying you must support this now. It’s if you use these schemas, then they are compliant. We’re not saying that EDI throw it away, we’re not saying that you have to support both. This is between willing trading partners whether they want to do this or not. But what we’d like to have done is for NCVHS to make a recommendation that we go ahead and have these adopted as being HIPAA compliant.

DR. SUAREZ: Who is this going to affect in the sense that you mentioned there are a number of companies out there that have developed their own proprietary schemas? If this becomes sort of the officially designated or named or identified schema, what’s going to happen with those proprietary schemas?

MS. WEIKER: Depending on how it’s written, I don’t think, if you have your proprietary schema, these adoption pieces, if you use these TR3 schemas you are HIPAA compliant. So they run the risk of still remaining to be not HIPAA compliant if they coded it wrong. If they coded it right it works, they don’t need to change at this point. But what we’re saying is if they use the X12 generated schema then you know that you’re a HIPAA valid schema and you don’t have to worry about disagreement between trading partners and all of that.

MS. GOSS: Margaret, to what degree is there a concern on the liability and the real risk that organizations don’t want to take in using this XML to sort of support other innovative products they want to put out there related to HIPAA transactions? Is actually not having this FAQ, and maybe even reinforced through an interim final rule kind of notification, stifling innovation?

MS. WEIKER: Potentially.

MS. GOSS: And if we have the XML schema in a HIPAA compliant format, how does that then translate towards the work that we want to do related to EHR integration and convergence? Is there a better ability for that HL7 and X12 worlds to mesh?

MS. WEIKER: HL7, a lot of what they’re doing is more XML based. When you look at version three it’s all XML based. So going from XML to XML is pretty easy. There are tools that are fairly cheap on the market to do those kind of things, and some open source code even, if you trust open source. But it would need some of the translation so to speak if you had to go back and forth.

MS. GOSS: It seems like sort of a no brainer. Why not do this? But I don’t know that an FAQ is going to be enough for some people in the compliance world. Maybe it could be both.

MS. WEIKER: Maybe it is both. An FAQ which I would hope could be done a lot quicker, while then the notice that would be done in regard to HPID and that type of thing, maybe it’s put in that notice kind of thing.

MR. SOONTHORNSIMA: I think we need to clarify the XML versus X12, they’re two separate, right?

MS. WEIKER: There’s the EDI. All in the same database, and from that database you can produce the TR3, and that’s what I’m going to call your legacy, that’s your typical EDI that you’d use today for the most part. The XML is generated from that same database, but instead of it being in the traditional EDI format it’s in an XML format.

DR. SUAREZ: Just to understand, because I know people come up with XML and X12 and all these Xs around. So X12 is the body, the company, the organization that developed the standard. The standard for example is the 837. The 837 is the claim, and the implementation guide for the 837 for HIPAA is the one that defines how the data elements on a claim will be put together in an electronic structure.

There is a way to create a claim based on the X12 standard in a simple ASCI string, and then there is a way to create that same claim using the X12 standard using XML, the eXtendable Markup Language XML in websites and all sorts of places. The standard for the string is everybody uses it, and that’s the X12 implementation guide. The problem is with the XML there is a number of companies that have converted the implementation guide into an XML language, and each one has done it differently, and there are concerns that people are not really doing it correctly. And so X12 the company developed a standard XML, the language, XML template, the best way to say it. So the recommendation is to make sure that when people are doing XML coding of an X12 transaction that they use the appropriate template to do it. Someone needs to name the X12 XML template as the standard template for developing XML. Is that correct?

PARTICIPANT: XML format is not officially recognized.

MS. KLOSS: Does that mean that there’s some urgency because it’s getting more fragmented and proprietary and divergent as time goes on?

MS. GOSS: This is going to complicate the enforcement processes for the interpretation of the X12 transactions. Having managed that process for several years for OESS, I can only imagine what’s going to happen when they get their first XML compliance related issue for the proprietary version if it’s not tracked back.

MS. DEUTSCH: I have several parts to my question. The first one is have you spoken to OESS to see that this would be able to be handled through an FAQ.

MS. WEIKER: We have approached them, Gladys Wheeler is the person we worked with in regard to an FAQ as well as an interim final rule, and it’s what we proposed. At the time when we gave the XML for them to evaluate and talked to them, they said that’s a possibility, but until they said yes we’re really going to do this, it was set there so to speak.

MS. DEUTSCH: And the other part that I’m concerned about is if you have proprietary fixes, who is going to determine that the fixes, that they are being created by different entities, are correct and meet the guidelines? Because what happens then is you’re going to have multiple ways of doing something, and that goes against the whole administrative simplification.

MS. WEIKER: Which is what is happening today. You have proprietary XML out in the field today. I know many companies developed their own, I helped develops some of that for some of those companies. So you have some that’s coded correctly, and you have some that is not coded correctly. And in many instances there may be small impact, or it could be huge impact, especially when you start converging into coordination and benefit type of things, and data getting dropped and those kind of things.

So we’re not saying okay, proprietary XML, how do we fix that, because we can’t, because we don’t know the whole universe that’s doing it that have to be validated, that’s a lot of work. But what we’re saying is hey, instead of you maintaining and keeping that proprietary XML out there, here’s an option to say yes, if you do this then you are compliant, and you don’t have to worry about somebody calling you and saying you’re not compliant, or you dropped this data element, or you didn’t convert it right, or those type of things. I’m giving an alternative.

MR. SOONTHORNSIMA: You have to stop the bleeding, that’s it.

DR. CHANDERRAJ: When you say compliance, are you meaning compliance to a mandate, or external as the body?

MS. WEIKER: It is compliant to the TR3, which has been adopted.

DR. SUAREZ: So it is compliant with the implementation guide.

MR. SOONTHORNSIMA: Which basically states here’s the type of data that you need, here are the elements that you need.

MS. WEIKER: Here is the situation when you send this data element in this format.

MR. SOONTHORNSIMA: The traditional approach was basically ASCII delimited files, now XML which is extendable.

DR. WALKER: This sounds like the right thing to do. From a business standpoint, is this giving X12 a monopoly they could exploit, or are they a public service organization?

MS. WEIKER: Well there is a lot of public service X12 does. X12 as an organization relies on volunteers to do all of its work. So companies donate people to do it, sometime they donate a lot of their time, sometimes they donate a little bit of their time. We have mandates where we say we at least cover our cost. I’m not sitting there on millions of dollars.

MS. GOSS: This is more for me about having the source of truth being consistent with the same source of truth that generates the HIPAA adopted guides that it is. But along the same questioning of what Jim just asked, I wanted to understand what negatives came out of the DSMO related discussions on this topic, if there was anything that we needed to understand was why not to do this and what were the concerns from some of the other organizations and their vetting process.

MS. WEIKER: There really weren’t any negatives. We did have discussion on would it be mandated for a health plan or a payer per se to support it. That was more the question of is this going to be mandated that I now support this, because maybe I don’t want to support XML, for whatever reason. So we said no, that’s not what we’re doing, we’re not saying throw the baby out with the bathwater, because there has been too much investment in the legacy system in those ASCII strings. But what we’re saying is for those entities that want to use XML, that want to develop products, that want to easily convert from HL7 version three to X12, here is an option to do that.

MS. KLOSS: Are you asking us to make a decision today?

MS. WEIKER: Not today. I would like to have one made today, but I’m not demanding one be made today.

DR. SUAREZ: We have to have a discussion and we’ll have to have a drafting of any possible -– But it will be done in the next few weeks.

MS. WEIKER: The letter can be shared that we sent to Ob and Walter. You can look at the actual DSMO request and the response and that kind of thing. Because I know you all are tired, you’ve just been fed by a water hose for the last couple of days.

MS. KLOSS: I have got to get educated.

DR. SUAREZ: We will have processing of this request.

PARTICIPANT: XML is spelled X, M –


DR. SUAREZ: Larry, you have the last chance here.

DR. GREEN: I want to re-ask Jim’s question. Also, can you just offer in plain language the market pressure that leads to the desire for documents in XML format? What’s going on that leads to this development and this situation evolving? What’s pushing it?

MS. WEIKER: I can speak from my experience in regard to this. The company I’m speaking about, I won’t name names, but it had many customers from many different perspectives. Government, both Medicare and Medicaid, DOD, VA, commercial payers. So we had a system that can adjudicate claims. Receive the claim, pay the claim. But we’d have one way to come in, which would be the EDI, but then to go to these multiple systems I’d have to translate the EDI into another format. Well, this one wanted this, and this, and this. So what we did was standardize the intake of those systems into XML. Because it’s easy to code, so to speak, there are commercial validators, open source, a lot of people can accept it, you can go into explorer and actually open XML and read it. So what we did was make a common front end to all of these different adjudication systems to say now we’re going to do one translation, it’s going to be in XML, we’re sending it to you, so then that system could pick out what they needed based upon the type of adjudication they were doing.

DR. GREEN: What was the answer to Jim’s question about what advantage does this create for X12?

MS. WEIKER: We are producing the TR3s today, and we create the TR3s today, and yes we charge for the TR3s, because we have development costs, we have to pay a vendor for tools, et cetera. We have to pay for meeting space. So yes, there is a cost to the TR3 schema. But we’re not saying if you use XML you must only use our XML, what we’re doing is saying if you use our XML then you know that it is valid HIPAA XML.

DR. WALKER: You have been producing the TR3s for some years, and have not abused that effective monopoly.

MS. WEIKER: I wouldn’t call it a monopoly.

MS. GOSS: Having been the former chair of the committee, they have financial issues because they haven’t had a good revenue stream. Because at best what this looks to me like it’s going to do is give them a little bit of extra revenue stream that will help with the overall standards development process, and it probably won’t even be that much revenue.

MS. WEIKER: We are not making millions off of this. Like I said, the most we make is we try to cover our cost.

MR. SOONTHORNSIMA: From the industry perspective, people who have to actually use these things, we are already way ahead. We’re already using XML, I know exactly what you’re talking about. We’re already exchanging XML today with entities. We’re trying to get caught up here with TR3.

DR. WALKER: From your perspective X12 is a public service?

MR. SOONTHORNSIMA: Yes, absolutely.

DR. SUAREZ: X12 was not an independent organization until just recently when it became a non-profit with a board of directors, all of which are members of X12, and members of X12 are basically providers, payers in industry —

DR. WALKER: So our blessing of this in some way would be roughly analogous to the Standards Committee blessing SNOMED.

DR. SUAREZ: Or X12, or HL7, or any of the other ones. So just to finish up, because I know we’re a little over our time. We will be summarizing this, we will be bringing it back to the subcommittee. We’ll have some conversations with OESS about the appropriate processes and all that, and then we’ll present it back as a recommendation and bring it back to the Full Committee.

How’s that? Any public comment before we close off the meeting? Anyone in the public wants to make any comment? Any final comment from any members of the committee? All right, we consider the meeting adjourned. Thank you very much, we’ll see you tomorrow then.

(Whereupon the meeting adjourned at 5:46)