[This Transcript is Unedited]

Department of Health and Human Services

National Committee on Vital and Health Statistics

Subcommittee on Standards

November 13, 2012

National Center for Health Statistics
3311 Toledo Road
Hyattsville, MD 20782

Proceedings by:
CASET Associates, Ltd.
Fairfax, Virginia 22030

P R O C E E D I N G S (3:00 p.m.)

DR. SUAREZ: I think we are going to go ahead and get started. It is about 3:02. Welcome to this meeting of the Subcommittee on Standards. My name is Walter Suarez. I am with Kaiser Permanente. I am one of the co-Chairs of the Subcommittee, and I don’t have any conflicts. Let’s just go around the room introducing ourselves.

MS. GREENBERG: Marjorie Greenberg, National Center for Health Statistics, CDC and Executive Secretary to the Committee.

MR. SOONTHORNSIMA: Ob Soonthornsima, Blue Cross Blue Shield Louisiana, co-Chair of the Subcommittee.

DR. SCANLON: Bill Scanlon, National Policy Forum, a member of the Subcommittee, no conflicts.

MS. WILLIAMSON: Michelle Williamson, Staff to the Subcommittee, NCHS.

MR. MCKINLEY: And I am Jim McKinley. I am Blue Cross Blue Shield of Alabama and I co-chair the Assessments Workgroup at HL7, Standards Subcommittee.

MS. BURKE-BEBEE: Suzie Burke-Bebee with ASPE, Staff to the Subcommittee.

MR. DAY: Durwin Day, Health Care Service Corporation.

DR. GREEN: Larry Green, a member of the Committee.

DR. SUAREZ: Anyone else from the Subcommittee? On the phone, anyone from the Subcommittee?

DR. CHANDERRAJ: Raj Chanderraj, a member of the Full Committee and a member of the Subcommittee on Standards.

DR. SUAREZ: And on the floor?

MR. ALFANO: Bill Alfano, Blue Cross Blue Shield Association.

DR. LAZARUS: Steve Lazarus, Boundary Information Group representing CAQH CORE.

MR. RODE: Dan Rode from AHIMA.

DR. JEAN PAUL: Tammara Jean Paul from NCHS. I am also on the Populations Health Subcommittee.

MS. KHAN: Hetty Khan, NCHS.

MS. TUREK: Kelly Turek, America’s Health Insurance Plans.

MS. PLAGER: Lauren Plager, Department of Veterans Affairs.

MS. GARR: Kari Garr, CMS.

DR. SUAREZ: Thank you. The agenda today is two-fold. The first thing we wanted to do was bring back to the Subcommittee the very important topic that we will be addressing throughout the next several months, and that is the topic of attachments.

As you know, under the Affordable Care Act there is a requirement for the Secretary to adopt final regulations on the standards and related operating rules for attachments. And so we wanted to bring you all up to speed on the model that is being used.

The important thing about this message that you will be hearing is how consistent and aligned, and this again, with this theme of convergence that we have been talking about so much here in the Committee, how much this is aligned and consistent with the same standard that is used for meaningful use for electronic health records and for the exchange of clinical information. So that is the first part of what we wanted to do.

And then just spend a few minutes at the end to talk about the Subcommittee plans, basically just go through the materials that we will be going through tomorrow and getting your feedback, and then we will be able to present it tomorrow.

Before getting into that, I just wanted to say a word about Thursday. On Thursday, as you all know, we have a roundtable discussion. It is very, very exciting for a number of reasons. Number one, the topic is sort of the future of information exchanges between health plans and providers in light of and in the context of the many changes that are happening in the health care system.

Really this future of information exchange to support the entire health care transformation that we are going to be experiencing. So there is going to be an opportunity to hear and to talk about, in a very open forum, the forces and the drivers and the possible directions, and since we are the Standards Subcommittee, the elements that will affect the standards that are being used in these information exchanges today.

So that is very exciting. And the second very exciting thing is the number of people that have expressed an interest in this. We are going to have probably a little over 24 people, 25 people in person on Thursday as well as a number of people joining by phone, at least 14 or so people. So it is going to be a very interesting and dynamic discussion.

We wanted to make it a roundtable rather than a hearing, as we explained at the last meeting. So it is very exciting. I think that is going to be a very important topic. And kind of the message that we want to send out is, this is the beginning of this dialog. This roundtable is not intended to complete the process of discussion about this.

Of course, this is a very extensive and important topic. And so this is truly the beginning of a dialog that we see, that is going to happen over the next several months, and that we are very pleased to facilitate. So with that, I am going to turn it to Ob and see if there are any comments from Ob.

MR. SOONTHORNSIMA: I am not sure when we talk about the future how far out in the future we are really talking about. Because some of the things we heard today, and some of the things we will be hearing today, I think, the future is already here. And a lot of the entities out there in the health care space, they are already engaging and exchanging data, maybe not in the fashion that we are thinking about.

That is why we need to convene the experts from the industry and get their opinions, and try to identify what does it look like.

DR. SCANLON: Before we hear from our presenters, if I could ask my questions about attachments now as opposed to later, because they may answer them during the course of their presentation. To me there is an arbitrary line between a claim and an attachment, the information that is in the claim versus the information that is in the attachment.

And we have heard before how expensive and problematic attachments are, in terms of the burden that they place on providers, as well as the disruption that they do in terms of the industry. The connection, I think, between improving that situation and this future, which is going to be very different, is that the information for the future needs are going to be very, very different than what we have been working with in the past with respect to claims.

And so my questions have been in some respects, what is the kind of information that we are asking for in attachments, how much of that really should be a part of a claim because we have reduced the burden in terms of being able to provide information in the future.

A great example in my mind are lab values. We have had before this Committee testimony about how the presence of lab values really changes your ability to risk adjust things. And there was the discussion of the riddled diabetics versus the controlled diabetic today. There is the whole issue of how much lab values go beyond having diagnoses.

There is a question in my mind, where should we be drawing the line? And I am thinking that maybe it is different in the future than it was in the past, and in the process we are reducing the burden with respect to attachments, because we have made some things standard as part of a claim. It is not going to be an extra burden because we have different information capacities than we did in the past.

Thank you indulging me. That is where I would like to see us head.

DR. SUAREZ: Let’s go ahead and get started. Thank you.

MR. DAY: I am Durwin Day, Health Care Service Corporation, and I happen to be one of the co-Chairs for the HL7 Attachment Workgroup, along with Jim McKinley from the Blue Cross Blue Shield of Alabama, and Craig Gabron from Blue Cross Blue shield of South Carolina is not with us today.

We are very excited and thankful to have this opportunity to share with you the progress that we have been making with the attachments. Some people in this room as well have been part of the Attachment Workgroup for a long time and have been looking forward to getting this implemented at some point.

So for the next 20, 25 minutes we would like to go over some of these items with you. We want to talk about the definition of attachments, and where it was, and where it is today. And how the scope of the attachments has changed into what we are calling a new model for attachments.

Jim is going to talk about the architectural components of an attachment. Then I am going to talk about the type of attachment needs that we found from the industry. When we last addressed to you back in December of 2011 one of the findings that came out of that was to go out and canvas the industry about what are the current attachment type needs, because they have changed.

So we did that through WEDI surveys. We will talk about that. We will talk about our current proposals that we would like to see as part of a package for attachments. And then Jim will get into some details about the Supplemental Guide and the external LOINC code set that we were proposing.

I will talk a little bit about some potential operating rules that would be helpful, could be helpful for attachments. And then we will end up with a summary and some examples, and then open it up for comments and discussion.

The definition of attachments a decade ago and more was originally to look at automating the process for supporting a claim. So a provider would send in a claim, the payer would bring it into their adjudication system, they would determine that we need the additional information to properly adjudicate it. They would send a letter off back to the provider. The provider would gather that information, send a package back to the payer, and this all takes, as you know, a lot of time, a lot of money

Our original purpose was to automate that. In the process, over the time, we still want to automate the process for the request and the response. But the document types have expanded. They are more than just the administrative type of information. It is clinical and administrative.

A way to automate that is to assign a LOINC code to each of those document types. So the other thing that we found out as time has gone on is that the document types bridge more than just supporting claims, that those same type of attachments that we ask for could also be supportive of prior authorization processes, referrals, and eventually could be useful for care management and clinical decision support as well.

So we still have some structured document types. By structured documents I mean that we have assigned distinctive data elements to each of those types. You will see more about that as we go on. So the document that we are proposing that becomes part of that structured document types, is the Consolidated CDA guide, which you are probably familiar with.

If you are familiar with meaningful use electronic health records, then this is in line with that. We are consistent with EHRs, and we are not making providers, if they have to send a discharge summary to one entity, they don’t have to send a different discharge summary document to a payer. Whichever one they are sending is fine for both.

We are identifying the requests at the document level. And as I said we are using LOINC codes to identify those documents. Now the ones that are part of the Consolidated CDA, there is the eight codified document types which includes the CCD, and we will go through a slide on those parts as well, in a minute.

But there are some that are outside of that as well, that are not part of the Consolidated CDA guide that we want to assign LOINC codes to, to make it easy for them to identify their question and response. So we have taken those other ones that are not defined in the Consolidated CDA guide and made a list and called it an external value set.

And Regenstrief has a panel — they call it a panel — that would be an external code set for that. In addition to that, the other parts of the three legs of this, three parts, would be some metadata around the request — who is sending the request, who is receiving the request, what is the request for, what the document type is.

And probably the most important thing is to re-associate the request for additional information with the original claim. One thing we did not do, and we stayed away from, is talking about transport of this standard. So that is not part of the standards. It is three parts.

I am going to hand this over to Jim and he is going to talk to you about some of the architectural components.

MR. MCKINLEY: As we continue on, within the actual model itself the words solicited and unsolicited refer to the actual act of requesting or submitting the additional information, called an attachment. Solicited is one where it is asked for and responded with. And the unsolicited is one that the provider would already know needs to be submitted at the time their claim or their prior authorization or their referral would go to the payer or the UMO.

So specifically, solicited is in response to an explicit payer request. And the response is to an explicit payer request as well. In the unsolicited model, you may have rules. You may have payer provider, in their provider section within their website they may have rules that identify information to a provider that says when you submit a certain type of surgical procedure with a certain type of modifier on it, I am going to need to have an Op Report to explain further in detail why you had to take those additional steps. That is the example of that.

And as part of this model we are suggesting that no attachment information be exchanged unless it literally conforms to either a solicited or an unsolicited, meaning if it hasn’t been requested either in advance of or at the time of the claim encounter, the referral, et cetera, then it is not to be shipped.

Another piece of the puzzle with regard to solicited and unsolicited addresses the connection with the metadata that we talked about earlier, the metadata that would encapsulate or be attached to the clinical information you are asking for.

Solicited means the payer is assigning that information, or the UMO. So they are giving that information to the provider in their request. In response, they are echoing that same information back, so that we know exactly how to match the information that originated the original need for the attachment information to begin with.

In the unsolicited model, though, since there isn’t an explicit request initiated right away, the provider simply pairs that information into his claim submittal. And there is a place in the existing 837 X12 claim to submit that identifier along with it being a part of the metadata for the attachment, so that the payer gets it, and knows how do I connect the dots, where does this information really need to go.

And Durwin spoke earlier about the structured and unstructured components. Basically a structured document is a clinical, electronic document that has a header on it. It defines discreet data elements below it, which are at the document section and element level, and there are conformance statements that indicate what is required and what is optional.

The unstructured document itself has the identical type of a header. It is, in fact, a structured piece, that information in both the structured and unstructured. But it also in the unstructured document identifies what type of document it is. And it is comprised specifically of things like PDFs, Word documents, TIFs, JPEGs, text, just basic information that you could scan and fax or you could scan and insert into an attachment envelope.

The key is that the unstructured document itself gives the industry an unlimited amount of availability to get an immediate piece of information exchanged within a short time window. If it is not existing now, as a prototype for an attachment, it can be done so in the new model through the external value set within six months, because they update that on a semi-annual basis.

I wanted to give you a visual. I apologize. It is probably very small from the back. But literally that is the header in the document of a structured component. And you can see that the type of metadata that is in the actual header includes things such as patient name, document creator, information source, information maintainer, et cetera. That is common to all of these documents, whether they are structured or unstructured.

Now in the body of this structured document it is going to have the breakdown of this information in one document with multiple sections such as the allergy and the pieces of information,8 18 and CCD I believe you are aware of, that were separate sections of information. And then within each of those sections could be one or more entries, which can be bundled together.

Then the unstructured, this is really the flexibility that this standard allows for. Similar, same, identical header information that is found here, the types of information, but the body is slightly different. They call it a non-XML body because it is really just lacking, it has a minimal amount of structure in XML.

And the rest of it is as if you have just an image, such as a JPEG, TIF, PNG, et cetera, or some type of a word processing document or in narrative format. Those can be PDFs, Microsoft Word, HTML, Plain Text, or Rich Text Flow, RTFs. I will let Durwin come in. We are a tag team on this, so forgive us here.

MR. DAY: So it is not enough just to say the Consolidated CDA is the standard. That is the great, for the format, for the structured format of it. But we still need to know who is sending it and who is receiving it, and that is where this metadata wrapper comes into play.

Here is some of the information that would be typical in that request — who is the payer, who is the receiver, who is the provider of service, what is the patient name and ID. Most importantly is the fifth one here, which is the payer claim control number. That is the re-association key. We associate that attachment to the original claim.

Moving on, I want to talk to you about the WEDI survey that we did. One of the findings from our last meeting was to go out and do some current research as to what attachment types are being used today. We sent out two surveys, one to the provider’s community, one to the payer community.

From the provider community it is interesting to note that the top three attachment types that they identified — and there were more, but just the top three — were Op Note, Progress Note, and consent forms. And oddly enough, the payer community looked at it the same way. They said Op Note, Progress Note, but then they said diagnostic images.

It may be skewed because we had quite a few dental providers chime in on the survey as well. And our top priorities then would be to look at Progress Note and Op Note, of course. And if you notice, as you will see, those two are part of the structured CDA Consolidated standard.

Then diagnostic images and consent forms would have to be covered in an unstructured format. So one of the other findings that came out of the survey was asked from the payer’s perspective, what other uses would you find for these attachment types. And that is where it came out that prior to the claim there would be prior authorizations would find use for Progress Notes, history and physical, lab results were also mentioned. So there are other uses beyond just the claim.

One of the other things that came out that was kind of interesting as well is, we asked both providers and payers how would you ask for the additional information, and how would you send it back electronically? And I would say maybe 35 percent of folks said they would use an X12 277, or a 275 as the wrapper around it, but 65 percent had no idea.

And we don’t know why they didn’t have an idea, if it was an educational issue or if they were just going to continue to use paper letters, so on. We don’t know. Next time we will word the questions a little differently and dive into that more.

So here are the eight attachment types that are part of the Consolidated CDA guide. You see that? There is the Continuity of Care Document, of course. But that doesn’t answer all questions clinical documents need to answer. So there is history and physical, discharge summary, and Op Notes. Those were all originally part of our original attachment types a decade ago.

And it is no coincidence that they are there, because the content from our documents, or original AS booklets, were the seed for what became Health Story Documents, which became consolidated in the CDA Consolidated guide. So there is definitely alignment with the content in here.

There are other ones that were not part of this of course. One of them that you will see in here, and this may answer your question as well, is ambulance claims attachment types was one of the original ones. Not necessary any more. That information is now being added into an 837 claim. So we illuminated one of the attachment types there.

Some of the other types of attachments that are outside of that which would require a LOINC code to identify it as an unstructured document. Our interest is pharmacy prior authorization, home health, consent forms, DME invoices, and so on.

So the three things that we are looking to propose as a package is that the Consolidated CDA guide that was balloted for meaningful use would also be used for attachments for the structured part. And just so you know, the HL7 is planning to ballot a newer version in May of 2013. And then they plan to ballot a new one once a year thereafter, to get on a cycle of schedule.

In addition to the Consolidated CDA guide, we would also like to see the Supplemental Guide that we developed at the HL7 Attachment Workgroup also be adapted for, which is going to be balloted in the January balloted HL7. In addition to that there has to be this dynamic set of LOINC codes that is identified as well for the unstructured documents.

So with that I am going to hand it back to Jim, and he is going to go into some details about the Supplemental Guide and the external LOINC code set.

MR. MCKINLEY: Some of the key components, and reasons that we felt we needed to have the supplement was that the Consolidated CDA on its own handled a minimal amount of information for the industry to know about constructing a clinical document in electronic format. Some of the things that we have identified for attachment purposes in this supplement are how to use a LOINC code to identify the request for a document, to codify a request.

Ultimately this would allow a hands free, nobody touching this type of exchange and a request with a LOINC code that says give me electronically this information back. And the response in theory could be programmed on the provider’s end and be able to extract the information and send it back with nobody in the provider’s office having to touch that.

Also in the supplement we depict certain business flows, which help the industry understand how it applies to claims, how it applies to referrals, prior authorizations, et cetera.

Identify the metadata standards that are needed. This is important because the consolidated CDA reflects a document that once it is created out of an EHR system, it is at an integrity with itself. It is a snapshot of whatever you are asking for, the CCD, the discharge summary, the history of the physical. And it has to be created the same way every time.

So we couldn’t insert any specific information that would help connect, back to the payer that was asking for the information. Therefore, that metadata is needed around it to help re-associate the actual clinical content with the need from the payer of the UMO that was asking for it.

This guide also explains a little bit about solicited and unsolicited, as we talked about a minute ago, and structured and unstructured, to give a little better understanding of what those components mean and how they play a role in the attachments.

It also will tell how to access the external LOINC dataset for the unstructured. There will be a little bit of a description in there how to find what the industry has allowed and adopted through regulations. At this point we hope for usage in the unstructured environment.

And how to acquire a new attachment type, because once I believe these go live or near live, we are going to see payers and providers come forward and say here are some other things we would like to see included. So how do you actually go through the process of working through the Attachments Workgroup at HL7 and coordinating with OSS and any other entities from a regulatory standpoint to make this happen?

It also addresses some of the gaps that we had found in C-CDA. For instance, there are multiple LOINC codes that are used for a discharge summary or for any one of those structured documents. Consolidated gives us a preferred code, and then a series of additional codes that are valid. So we need to lend clarity to the industry. What code do you use and under what circumstances do you use it so that we don’t have a free for all?

We also found that in the guide itself it referenced EHR to EHR exchange, for which a payer has no stake. They don’t have an EHR. So we had to clarify what that really meant in a payer environment. And also, the meta data, as we spoke about a minute ago, specific to attachments.

Now a little bit about the external value set through the LOINC database. We are currently limiting this to unstructured, although there are discussion about expanding that to include all codes that could be in structured as well.

Basically this would say their industry, if I need to know what clinical documents can be exchanged under regulation, we are proposing to have them in an external value set so that you can go as a payer and be prepared for them, and a provider to be prepared for them, to be able to be exchanged. That is in the unstructured world right now. Again, we are talking about including them in the structured environment possibly.

There is a tabbed display. The way the registry will display it, it will be clearly obvious that these are codes used for regulation for this attachment regulation. This again is updated on a semi-annual basis. So in theory if the industry needed a new attachment type on January, they could come to us and by July in theory they could have an unstructured capability to exchange that, under the regulation, through this standard.

Then the additional attachment implementation supplement would also provide guidance on the LOINC database usage itself. The process for obtaining this is still very rough, and not yet formalized. But in essence, entities that would need a new attachment type must initiate it with our workgroup at HL7.

The reason for that is, it will always start out as an unstructured request because we haven’t built a structure for it. But we will assess it, and assuming that it meets the criteria, we then would start the process of building a project at HL7 to develop the structured equivalent of what they come to us asking about.

Now that enables electronic exchange without having to have providers touch it. In theory, it enables an ability for that to happen in a structured environment. We would see including at least OESS and any other entity that it would make sense, because we are technically including, every time we do an update to this, we are including new attachment times that would be imposed under regulation to be exchanged, using the standard.

Again, we see them initially starting as unstructured. And we would then roll them out into a structured format, because there may be some that aren’t candidates for structure. And an example of that could be a DMA receipt. Would it be beneficial to try to structure something like that or not? And then again, if it is the structured candidate, our workgroup at HL7 would develop a structured component for that.

MR. DAY: We just tried to put out some food for thought for some potential uses for operating rules. And if the X12 guides are used, of course then the core infrastructure rules would apply, batch acknowledgements, companion guide templates, and so on, for that.

But the most important thing, I think their use would be in identifying the transport alternatives and issues, whether it is a traditional EDI channel where they will be sent through, or an HIE type of environment. So we just try to put those things out there, to seed things.

MR. MCKINLEY: After what we have said so far, there are basically a handful of advantages that we see to this new model that really stand out. First, it does leverage the exact standard that is being used for EHR incentive, meaningful use stages one and two. We expect to see that for stage three based on what we are hearing preliminarily.

It relies on document level representations. This was important to the clinicians. As we talked to them, they didn’t want any single piece of information by itself. It needed to be taken in context. So if we just extracted a piece of information, it was not informative as much to them, we were told, as having the entire document for it to be framed in context.

Conformance statements, again, indicate what is required and optional. And the unstructured document can accommodate any attachment need. External value set, we felt like this was just a key to the industry. You don’t have to wait on a new publication of a standard, and/or potentially a new regulation, to add new attachment types. This is very responsive to the industry itself. When they need them, they can get them within six to nine months.

And it offers the providers and payers the flexibility of adapting to, no matter what level of technical expertise you have. This was a point in the slide where if I had had my own laptop with some software on it that would allow you to visually see this, this was where I was going to show — I will just try to describe it — you can represent the XML and you can see this lengthy XML statement.

And then using a style sheet and a common browser, you can take that exact XML, open it with the product, and you will see a complete, easy to see, human readable version of that same XML data. So it is a common CDA model similar to what HTML does with the web browsers.

We have some examples. I will just hurry through these, just to give you a sense. The claim attachment example in a solicited model is pretty straightforward, where arrow number one depicts the health care claim or encounter being submitted from the provider to the payer. The payer, upon review, determines they need additional information, so arrow two sends a request back to that provider for that information, and arrow three reflects the return of that information.

It is different from the unsolicited model in that there is no request. In the unsolicited example, arrow one is the claim itself or the encounter, and arrow two is the simultaneous or near simultaneous submittal of that additional information or that attachment information, with the claim itself. That is, again, one of those rules based things that has been set up in advance.

Then there is a prior auth example for solicited. It is very similar, where the provider submits to the payer, the UMO, the service authorization request. The payer, the UMO, says hey, we need more information about this before approving it. So arrow two, they send the request back to the provider, and the provider returns the attachment information. Very simple, straightforward flow.

And again, the difference in this for the unsolicited, prior authorization, is that a provider submits at the time that they are submitting the service authorization request. They submit the additional information or the attachment at that time to the payer or the UMO.

We rushed through a lot of information. We will be happy to fill in anything we can.

DR. SUAREZ: Thank you so much. I think this has been probably one of the clearest descriptions of this complex issue, claim attachments and attachments in general. So I really appreciate the presentation. We are opening it for comments.

DR. SCANLON: This was very, very helpful. As Walter said, I think it gives us a lot more clarity about attachments. To go back to where I was — I am looking for things that simplify things. And your comment about structured attachments and how it takes the provider’s hands out of the picture, the things can be sort of done automatically, machine to machine, in some respects.

The issue of unsolicited, I guess the question there is how much standardization we can have on unsolicited. I think when you went through slide six, you mentioned that a provider could see which plan is asking if this is a surgery with this modifier, what the plan is asking for. What is the feasibility of getting that not to be plan-specific, but just that this is the surgery, this is the modifier, then this is always going to be the unsolicited attachment that is going to go with it?

In my mind that achieves in some respects what I am talking about in terms of modifying a claim. Again, I am not looking for a claim, for all claims to be encyclopedic in terms of the amount of information that flows. It is conditional. It is like for this diagnosis, this procedure, here is the information that needs to go with it. Can we achieve that to any significant extent? Or reduce what are regarded as the hassle of attachments?

MR. DAY: I think there are some opportunities for that. One of the popular ones now is when a claim is sent with the modifier 6222 on an operative claim, then out notes are expected. And I think several plans have that as a rule for the unsolicited now. So maybe it would be universal. That is something that we would have to do outreach to everybody on. But there is an opportunity for that.

MR. SOONTHORNSIMA: I think to answer your question, you are talking about prior auth specifically, right? That is a good example.

DR. SCANLON: Or a claim. Either way it comes back.

MR. SOONTHORNSIMA: There already are circumstances where prior auth happened automatically if the provider answers certain questions. A lot of these are happening on line today. So this would be, I assume, another vehicle. It comes directly from, whether it is EMR or whatever the system they use at the hospital to request prior auth, as long as they pass all that information through.

So there wouldn’t be any 275 at all for those circumstances. But for other circumstances where additional information would be needed, then that would be a —

DR. SCANLON: What I am thinking is, if I am dealing with 10 plans, it is conceivable that if they have different rules in terms of unsolicited, that I can program my system to deal with those rules. But then when I get the 11th plan that I just signed up with, suddenly I have to re-program in order to deal with their rules if they are different.

So the question is, can we come to an agreement that the plans are going to say, we accept this as the standard.

MR. SOONTHORNSIMA: That is really more of a medical policy issue, I think, than the transport. Correct?

DR. SCANLON: Oh, it is definitely a medical policy issue, yes. And that is where it becomes a more difficult question. Some of these other things we have been talking about are technical, and I understand that one of the things that distinguish plans is their medical policies.

DR. SUAREZ: It is a major question, I think. And it goes really beyond the standard, per se, to try to create some consistency among the industry, particularly the help lines, on the expectations of things like unsolicited attachments. So we have the standard, this is the standard to send it.

The rule for when it is required to be sent or not, and the way in which it is required to send, like unsolicited or solicited, becomes part of a larger decision. And it is part of the question of could that be something that the operating rules can at some point get into? The operating rules are exactly the kind of rules that are set to help eliminate some of the variability or discretionary kind of decisions among the industry.

So there could be clearly an opportunity to elevate that to an operating rule decision that says as an industry, we all agree that if you are going to send a claim with these following characteristics, and Durwin introduced the technical terms of the codes. But if we send it, we know that every time you are going to have to submit an attachment. I think that is the operating rule.

DR. SCANLON: How we get there is not clear to me. I think in terms of trying to improve efficiency in this process, it is a key question and we need to struggle with it. And I think in some respects, where we are with respect to the technology is, we can err on the side of having more information.

It doesn’t all have to be used when the plan gets it, in terms of their medical policy. We can transmit more information than a plan might think is totally necessary. And that will reduce the overall burden sort of both to the provider and the plan, and then medical policy can say here is how I am going to use that information, and that is up to the plan.

DR. SUAREZ: Now that we have a named authoring entity for all the transactions, including claim attachment, I think this is something that is to be elevated to that.

MR. SOONTHORNSIMA: Bill raised a really good point. In order to achieve efficiency, there are different levers. And that is just one of the levers that is outside of this attachment. But I think that is a lever that can enable efficiency down the line. We just have to understand, how much can that really affect efficiency.

Can I ask a different question? To make this happen, right now I guess HL7 is really waiting to get this thing approved and moving forward — what does it take at high level to turn this on so to speak on the provider side, to make it happen? How difficult is it, because this is not a common practice today.

DR. DAY: I think that is going to bring up the question about the transport as well. Because if the providers are already exchanging discharge summaries, Op Notes in that respect through their electronic health record systems, they have the means to do that. But traditionally with payers, the transport has been through an EDI channel.

So there are some questions there as to what would be the transport method used between the providers and the payers, and getting both sides to use a common transport. So I think that would be probably the key to exchanging that information.

I think the information and documents are there for the provider to pull. We are not asking them to do anything different than what they already have. If they have the documents stored as a discharge summary, send that discharge summary to us. It’s just what channel to send it on is the question.

DR. SUAREZ: From my perspective, I am thinking of a large provider system. The good thing is that this is building on the EHR standard that is used by EHRs to pull out data and generate the data. But inevitably there is going to be new workflows that have to be created in order to fulfill the actual extraction from the EHR system into a C-CDA with appropriate characteristics. And then packaging it and transporting it in the right way.

But I think the sense is that once you create that route, because the standard is set to go, I think the number of times you have to create new workflows is minimized significantly because now we have a standard.

Now the transport is a big question. If I have 10 different transports it goes 10 different payers deal with 10 different ways of me sending in that data, then I have to have one standard to package the information with C-CDA and then develop this specific transport mechanism. And I have to create it.

MR. MCKINLEY: An interesting point, to know that right now in the early phases of meaningful use, a CCD creation and discharge summary creation, are pretty discrete. You kind of know a picture of everything at this point in time about create it and ship it, or the discharge associated with the patient’s recent admission.

But the triggers for progress notes, procedure notes, operative notes, things like that, have really not been developed for EHR meaningful use. So maybe we are at the right timing to develop the triggers for both attachments and those kinds of things at the same time.

DR. CHANDERRAJ: I think I agree, that we need to develop specific requirements for every plan, to submit or not submit specified reports. I think if there is for meaningful certification you need to have certain standards. There are no standards for billing systems, physician practice billing systems.

I think this should become a part of, we should try to develop the certification process for the billing companies to meet certain requirements. And one of the requirements will be to submit a certain number of informations, that they are required to pay the process.

And I think most of the insurance companies — this is my personal opinion — I think they are using this as a delay tactic, not to pay the money. They are wanting more information. And by the time a service is provided and the bill is paid, it is almost more than 90 days.

You go to a hairdresser and you say I will pay you in 90 days, he wouldn’t accept that. Whereas a doctor, who wants $30 per visit, he has to submit innumerable reports to different insurance companies requiring different insurance. And this is a delay tactic. And I think if we quickly develop some standards for certification for billing practices, then I think this will be smoother.

I have some other questions. Why do payers want diagnostic images? Wouldn’t a diagnostic report of an image be sufficient?

MR. DAY: Yes. It is called a diagnostic image report. Yes, a big difference.

DR. CHANDERRAJ: Second thing is what if sometimes we get the prior authorization but still the claim is denied saying we don’t want to pay for this? How do we address that in the claims attachment?

MR. MCKINLEY: I don’t know that we have a standard created to address that in the model. To me that is a business practice.

DR. SUAREZ: That might be a business issue not related to the standard itself but some other —

DR. CHANDERRAJ: If a standard is being created for prior authorization submission —

DR. SUAREZ: If the standard were to be applied to prior authorization, there is still the opportunity to deny something, unless you are talking about it being denied because you are sending it by paper and not —

DR. CHANDERRAJ: No. There is no explanation or anything. The procedure is denied but it has been approved before.

MR. SOONTHORNSIMA: That might be an outlier. I think what we are trying to do, what was Bill’s point earlier, is there a sweet spot for us to make huge efficiency leaps? And my question was, what would it take?

MS. BURKE-BEBEE: I am wondering if the ambulance example that they gave, that was folded into the X12, was — these are policy issues, not technical issues. Was there something that you learned about the ambulance attachment going away because it folded into the 837? Was there a consensus? What was it?

MR. DAY: One of the decision points a long time ago was, what belongs in the claim versus what belongs in an attachment. And one of the closely rated ones was ambulance, because the 837 accommodated maybe half of those elements but didn’t include the run report and that kind of stuff.

What we did decide to do was that there was so much so closely related that why have this debate all the time of what belongs in a claim versus what belongs in an attachment? If the elements are there closely related, and we only need to do a couple of changes to the X12 837 to accommodate the attachment, we don’t need to create another document for it. So that was the decision.

MS. BURKE-BEBEE: So there were people at the table at HL7 that went with the X12 and decided at the Consensus that it could be done?

MR. DAY: Yes.

MR. MCKINLEY: But not confident that moving from 4010 to 5010 had anything to do with it, but with the move to 5010 it may have expanded the fields that were available as to making that step.

MR. SOONTHORNSIMA: If we were to stretch this out a little bit, think about five years from now, the type of information might be different. We might get a lot more clinical data. So what are some of the use cases you can see in the future? Jim, you and I kind of broached this topic a little bit. And how does this work with an HIE, or instead of using HIE as a means to do that?

MR. MCKINLEY: Putting on my payer hat?

MR. SOONTHORNSIMA: And provider. We are trying to make it easy.

MR. MCKINLEY: From the perspective as a payer we have an extensive health management, case management, disease management effort. And when you transition care from one point to the next, it is very helpful to have a staff nurse from our company hold the hand of the patient to be sure that medication reconciliation occurs, make sure they understand that the transition of your care from place one to place two is addressed in the most favorable way to reduce the discharge.

And by helping them, we are finding that we are reducing that. We see it in the pilots that they are running now, longitudinal care, working to improve, I would say, the advocacy on behalf of our members, which are ultimately the patients. To have the greatest outcome for them would be our goal from the payer perspective.

MR. SOONTHORNSIMA: So the care coordination is what are you talking about specifically?

MS. BURKE-BEBEE: Is that what you were describing?

MR. MCKINLEY: Well Ob was asking, I believe, some of the down the road use case examples for it, clinical information exchange between providers and —

MR. SOONTHORNSIMA: Using this as a vehicle? Using Attachment 275 specifically?

MR. MCKINLEY: Very possibly.

DR. SUAREZ: I think we are just about running out of time. We are not going to be able to cover the other item, which is the Subcommittee plan, but we have the chance to present to the full committee tomorrow and at that point we will be going over the plans that you all have seen, because it is the plans that we have been sharing over the last several months about the short-term, the intermediate things that we need to do, and then some of the long-term activity.

I want to recognize my goal. We have a public comment period and I know Steve wanted to jump in, and any other public members that want to join.

MR. LAZARUS: Thank you, Walter. I am Steve Lazarus from Bounty Information Group, representing CAQH CORE today. Wendell and Lowes may be on the phone as well. I just want to briefly comment on this issue of data flows and workflows for attachments, whether it be from an EHR, another EHR and the provider, or through an HIE or any other kind of exchange.

There is a research initiative that CAQH CORE launched earlier this month regarding attachments of all types among all different trade partners in this arena so that we come up by February, or before February, with hopefully testimony before the Subcommittee.

We will have done the groundwork research with more than 20 structured interviews and building a consensus about what the best practices are today and what are the barriers, and how we need to deal with operating rules to eliminate some of those barriers. So we are in the process of doing that research right now and you will have a much deeper dive and answer some of these questions by that time.

DR. SUAREZ: Excellent. Thank you so much. Michelle?

MS. WILLIAMSON: His comment is right on track with what I was thinking, because when Bill mentioned about the differences from one attachment to another, my concern is that we could end up in the industry like we did with HIPAA with the companion guides. So there were companion guides. We had the HIPAA standard but we had hundreds of companion guides that needed to support them.

Are we going to end up in the same situation with attachments, where we have an attachment, but for this payer it needs to be this.

DR. SUAREZ: That is the kind of thing we want to avoid, certainly. I think the standard, we are going to have attachments for specific purposes. But I don’t think there are going to be attachments for specific purposes, for specific payers. I think that last part we are not going to get to, because the standard defines the content of the attachment for that specific purpose, like an ambulance run or an operative note.

Any other comments from the public? Anybody on the phone that wants to make a comment? Anyone from the public? All right, well Jim and Durwin, thank you so much again. This was terrific, and I look forward to the remaining work. I think our February meeting is going to include a full day, or at least a full day or half day hearing on this topic again. So thank you again.

I think we are adjourned. We are going to transition to our next meeting.

(Whereupon, the meeting was adjourned.)